uva-dare (digital academic repository) degradation and ... · proefschrift – degradation and...
Post on 25-Jun-2020
3 Views
Preview:
TRANSCRIPT
UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Degradation and analysis of synthetic polymeric materials for biomedical applications
Ghaffar, A.
Link to publication
Citation for published version (APA):Ghaffar, A. (2011). Degradation and analysis of synthetic polymeric materials for biomedical applications.
General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.
Download date: 08 Jul 2020
Degradation and Analysis of Synthetic Polymeric Materials for Biomedical Applications
Abdul Ghaffar
The front cover image (courtesy of Bozhi Tian) reprinted with permission (published in T.
Dvir et al., Nature Nanotechnology 6, 13-22 (2011). The foreground of this image represents
polymeric fibres (purple). The background of the image shows a scanning electron
micrograph of an electrospun polymeric fibre mesh. The cells are shown in light blue.
Proefschrift – Degradation and Analysis of Synthetic Polymeric Materials for Biomedical Applications by Abdul Ghaffar ISBN: 978-90-5776-231-4 90-5776-231-5
This research was conducted at Analytical Chemistry Group (former Polymer-Analysis Group), Van’t Hoff Institute for Molecular Sciences, FNWI, University of Amsterdam.
and was supported by
Higher Education Commission of Pakistan
Degradation and Analysis of Synthetic Polymeric Materials for Biomedical Applications
ACADEMISCH PROEFSCHRIFT
ter verkrijging van de graad van doctor
aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
prof. dr. D. C. van den Boom
ten overstaan van een door het college voor promoties
ingestelde commissie,
in het openbaar te verdedigen in de Agnietenkapel
op donderdag 27 oktober 2011, te 14:00 uur
door
Abdul Ghaffar
geboren te Lahore, Pakistan
Promotiecommissie:
Promotor: Prof. Dr. S. van der Wal
Co-promotor: Prof. Dr. Ir. P.J. Schoenmakers
Overige leden: Prof. Dr. Ir. W. E. Hennink
Prof. Dr. D. W. Grijpma
Prof. Dr. Ir. J. G. M. Janssen
Prof. Dr. C. G. de Koster
Dr. W. Th. Kok
Dr. A. A. Dias
Faculteit der Natuurwetenschappen, Wiskunde en Informatica
Dedicated to my beautiful children NAVERA and SARIM
Table of Contents Chapter 1 ................................................................................................................................1
1. Methods for the chemical analysis of degradable synthetic polymeric biomaterials*......1 1 Introduction .................................................................................................................2 2 Degradable biomaterials ..............................................................................................3 3 Analytical strategies.....................................................................................................5 4 Degradation methods ...................................................................................................7
4.1 Degradation under non-physiological conditions ..................................................7 4.2 Degradation under physiological conditions .........................................................8
5 Chromatographic methods for degradable polymeric biomaterials ............................10 5.1 Size-exclusion chromatography ..........................................................................11 5.2 Adsorption liquid chromatography .....................................................................14 5.3 Liquid chromatography at critical conditions......................................................17 5.4 Two-dimensional liquid chromatography ...........................................................18
6 Gas chromatography..................................................................................................20 7 Direct mass-spectrometric analyses ...........................................................................22 8 Nuclear-magnetic-resonance spectroscopy ................................................................25 9 Conclusions ...............................................................................................................26 10 Scope of the thesis ...................................................................................................27 11 References ...............................................................................................................28
Chapter 2 ..............................................................................................................................33 2. Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials: structure elucidation, separation, and quantification of degradation products...................33
1 Introduction ...............................................................................................................34 2 Experimental..............................................................................................................36
2.1 Materials .............................................................................................................36 2.2 Procedure of hydrolysis ......................................................................................39 2.3 1H NMR spectroscopy of hydrolysate.................................................................40 2.4 Size-exclusion chromatography (SEC) analysis..................................................40 2.5 HPLC-ESI-ToF-MS analysis of hydrolysate.......................................................42
3 Results and discussion ...............................................................................................43 3.1 Optimization of hydrolysis method.....................................................................43 3.2 Product identification..........................................................................................44 3.3 Molar mass characterization and quantification of PMAA .................................48 3.4 Quantification of monomeric products by HPLC-ToF-MS.................................50
4 Conclusions ...............................................................................................................56 5 References .................................................................................................................57
Chapter 3 ..............................................................................................................................59 3. Monitoring the in vitro enzyme-mediated degradation of degradable poly(ester amide) for controlled drug delivery by LC-ToF-MS ....................................................................59
1 Introduction ...............................................................................................................60 2 Materials and methods ...............................................................................................62
2.1 Materials .............................................................................................................62 2.2 Solubility ............................................................................................................62 2.3 Enzyme activity ..................................................................................................62 2.4 In vitro enzyme-mediated degradation................................................................63 2.5 Size-exclusion chromatography ..........................................................................64 2.6 LC-ToF-MS study...............................................................................................65
1
3 Results and discussion ...............................................................................................66 3.1 Solubility ............................................................................................................66 3.2 Overall effectiveness of in vitro enzyme-mediated degradation.........................66 3.3 Molecular-weight of remaining material.............................................................68 3.4 LC-ToF-MS analysis following enzymatic degradation......................................68 3.5 Factors affecting enzyme activities .....................................................................71
4 Conclusions ...............................................................................................................75 5 References .................................................................................................................76 6 Supporting information..............................................................................................77
6.1 Two-dimensional H,C-correlated spectrum (HSQC) of PEA..............................77 6.2 ESI-ToF-MS spectra of the identified peaks – enzymatic degradation................77 6.3 Chemical Degradation – optimization of the LC-ToF-MS method .....................81
Chapter 4 ..............................................................................................................................85 4. A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s*............................................................................................................................85
1 Introduction ...............................................................................................................86 2 Experimental..............................................................................................................88
2.1 Materials .............................................................................................................88 2.2 Dynamic coating of stainless-steel capillaries.....................................................89 2.3 On-line LC-ToF-MS analysis .............................................................................90
3 Results and discussion ...............................................................................................92 3.1 Continuous-feed mode ........................................................................................92 3.2 Pulse-feed mode ..................................................................................................98 3.3 Comparison of pulse-feed mode and continuous-feed mode..............................101 3.4 Application to tri-block PEA coatings ..............................................................102
4 Conclusions .............................................................................................................104 5 References ...............................................................................................................105 6 Supporting information............................................................................................106
6.1 Solubility ..........................................................................................................106 6.2 Molecular weight (Mw) and dispersity of tri-block PEA ...................................106 6.3 NMR experiments.............................................................................................108 6.4 ESI-ToF-MS spectrum of α-chymotrypsin........................................................110
Summary ........................................................................................................................111 Samenvatting ..................................................................................................................113 Acknowledgments ..........................................................................................................115 Bibliography...................................................................................................................118
*A. Ghaffar, P. J. Schoenmakers, Sj. Van der Wal, to be submitted.
Chapter 1
1. Methods for the chemical analysis of degradable synthetic
polymeric biomaterials*
The performance of biodegradable polymeric systems strongly depends on their physical, as
well as on their chemical properties. Therefore, the detailed chemical analysis of such
systems is essential. Enzymatic and chemical hydrolysis are the primary biodegradation
mechanisms for these materials. This review provides an overview of the strategies and
analytical methods used for the structural and compositional chemical analysis of non-
degraded, partially degraded and fully degraded synthetic polymeric biomaterials with an
emphasis on modern solution-based techniques that yield large amounts of information. The
degradation methods that facilitate the study of polymeric networks are also described.
Chapter 1
2
1 Introduction
A biomaterial is a substance that has been engineered to take a form which, alone or as part
of a complex system, is used to direct, by control of interactions with components of living
systems, the course of any therapeutic or diagnostic procedure, in human or veterinary
medicine [1]. Synthetic polymeric biomaterials are of great importance in the medical field
due to an aging population and because of their potential to improve the quality of life [2].
There is a clear trend to replace non-degradable by degradable materials [3]. Biodegradable
polymeric implants are intended to degrade gradually and their degradation products are
meant to be excreted benignly by the body, so that they do not need to be surgically removed
after their functional role (e.g. as drug-delivery carrier) has expired [4]. This causes
increasingly strict demands on the design and synthesis of biodegradable polymeric
materials for applications in drug-delivery devices, gene transfer, regenerative medicine,
scaffolds for tissue engineering, and surgical implants, such as rods, sutures, pins and screws
for fixation devices [5,6]. Degradation of biomaterials has many biological, physical, and
chemical facets. Biological assessment involves cell tests or implantation. Morphology is
important to understand degradation behaviour. For example, crystallinity plays a crucial
role in the degradation of poly(lactic acid) [7].
This review is limited to the in vitro chemical analysis of synthetic polymeric biomaterials.
The suitability of synthetic polymeric biomaterials for medical devices can be inferred from
their chemical structure, mechanical properties, degradation kinetics, and the
biocompatibility (tissue response) of the polymers and their degradation products [8]. The
molecular weight, hydrophilic or hydrophobic nature, fractional composition sequence and
(stereo-) regularity of the monomers in multi-block co-polymers, length of kinetic chains in
photo-polymerized networks, nature and concentration of additives, shape and morphology
of the specimen, and incubating media can all influence the degradation rate and mechanism
in terms of surface erosion or bulk degradation. The biodegradation mechanism for such
materials primarily involves enzymatic and chemical hydrolysis. Highly reactive species,
such as peroxides, are produced in reaction of the human body to the biomaterial (foreign-
body response). Such species may also degrade the polymer chain and contribute to the
overall degradation of biomaterial [9].
Various reviews have been published on the synthesis and application of synthetic
biomaterials. However, despite their increasing use in the biomedical industry, very few
articles have reviewed selective characterization techniques. No review has been published
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
3
that summarizes in detail the analysis methods that lead to the identification and structural
analysis of degradable polymeric biomaterials. Therefore, we set out to review different
analytical methods used for the analysis of degradable polymeric biomaterials as a starting
material, after partial hydrolysis under physiological conditions, and after complete
hydrolysis. Methods involving chromatographic separation followed by spectroscopic or
mass-spectrometric detection are discussed here. Degradation methods required to bring the
complex copolymer and insoluble networks within the realm of chromatographic techniques
and direct mass-spectrometric analysis are also emphasized.
2 Degradable biomaterials
Synthetic degradable polymeric biomaterials contain one or more functional groups, such as
an ester, ether, amide, imide, thioester, anhydride, etc., in their chemical structure. This
enables such materials to degrade gradually, either through chemical stress or through
biological processes. The polymeric chains in a degradable polymer can differ in terms of
their length, chemical structure, architecture, etc. On the basis of their chemical composition,
they can be divided in homopolymers and copolymers. The sequence of the different
monomers in polymeric chains further differentiates copolymers into block copolymers,
alternating copolymers, random copolymers, graft copolymer, etc. The architecture of
polymer molecules can be linear, branched, hyperbranched, or dendrimers. Polymers may
also form three-dimensional chemically or physically cross-linked network. The arrangement
of different fragments in polymeric chains not only determines their configuration (stereo-
heterogeneity, such as isotactic, syndiotactic and actactic), but also their ability to rotate
around a single bond (so-called conformational heterogeneity). All these parameters may
directly (e.g. through chemical stability) or indirectly (e.g. through the crystallinity)
influence the rate of degradation of biomaterials. Hence, all of them need to be investigated.
Poly(2-hydroxyethyl methacrylate) (pHEMA), poly(glycolic acid) (PGA), poly(lactic acid)
(PLA), poly (lactide-co-glycolide) (PLGA), polycaprolactone (PCL), and functionalized
cross-linked polyacrylates are the most extensively studied polyesters for biomaterials [10].
Polyurethanes have been investigated in the biomedical industry and their properties have
been tailored by incorporation of ester and ether components to generate poly(ester
urethane)s or poly(ether urethane)s. Poly(ester amides)s (PEAs), prefereably with natural
amino acids, are attractive for biomedical applications such as drug-eluting stent coating
[11]. Ulery et al. recently published a comprehensive review describing in detail the
Chapter 1
4
biomedical applications of synthetic and natural biomaterials [12]. A few examples of
different functionalities of synthetic polymeric biomaterials are tabulated in Table 1.
Table 1 Selected synthetic biodegradable polymers and copolymers.
Polymer types
Structure Reference
Polyester
[13]
Polyether
[14]
Polyamide
[15]
Polyimide
[16]
Polyurea
NH
NH
O
R2R1NH
NH
O
n
[17]
Polyurethane
[18]
Polyanhydride
[19]
Polythioester
[20]
Polyphosphoesters
[21]
Polysiloxane
Si
R1
O
R2n
[22]
Poly (ester amide)
NH
R2
OR1O
O
n
[11,23]
Poly (ester urethane)
NH
O
O
R2R1O
O
n
[10]
Poly (ester urea)
[24]
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
5
Poly (ester ether)
[14,25]
Poly (ether urethane)
[26]
Polycarbonates
[12]
3 Analytical strategies
Many procedures and techniques can be applied to study the properties and degradation of
biomaterials. Water-uptake (swelling-ratio) measurements provide useful information on the
hydrophilic or hydrophobic nature of the materials. The results can be related to the degree
of crystallinity of the structure. Monitoring the changes in the pH of media as a function of
degradation indicates the acidic or basic nature of the released degradation products and their
ultimate effect on the surrounding environment (cells, tissues, etc.). Weight-loss studies are
almost universally performed to estimate any change in the mass of biomaterials during
degradation. Changes in the specimen dimensions and surface morphology, such as crack, or
micro channels, and changes inside the material can be highlighted by microscopy
techniques, such as scanning electron microscopy (SEM), transmission electron microscopy
(TEM) or atomic-force microscopy (AFM). The surface chemistry of the biomaterial may
alter or influence proteins and cells and may affect biocompatibility. The common methods
to characterize the surface chemistry include contact-angle measurements, Fourier-transform
infrared – attenuated-total-reflectance (FTIR-ATR) spectroscopy, X-ray photo-electron
spectroscopy (XPS) and secondary-ion mass spectrometry (SIMS) [27]. Differential
scanning calorimetry (DSC) and wide-angle x-ray diffraction (WAXS) are commonly used
techniques to estimate changes in the crystallinity of a biomaterial during the degradation
[28].
However, to tailor the properties of biomaterials, including physically or chemical cross-
linked networks for a specific biomedical application and to estimate the compatibility of
their degradation products with the surrounding biological environment, an in-depth
knowledge of their chemical structure is mandatory [29]. This includes characterization of
the starting material and the degradation products. Chromatographic separations, mass
Chapter 1
6
spectrometry (MS), and FTIR and NMR spectroscopy can provide more insight in the nature
and chemical structure of the degradation products.
There are three fundamentally different approaches to the structural characterization of
biomaterials. The first approach is the analysis of biomaterials without degradation. If the
biomaterials are soluble their identity and average molecular weights can be determined by
NMR spectroscopy and and by size-exclusion chromatography (SEC), respectively. The
compositional analysis of oligomers and low-molecular-weight polymers can be achieved by
mass-spectrometric techniques, such as liquid chromatography coupled through an
electrospray-ionization interface to a tandem mass-spectrometer (LC-ESI-MS/MS or LC-
ESI-MSn) or to a time-of-flight mass spectrometer (LC-ESI-ToF-MS), or matrix-assisted
laser-desorption/ionization (MALDI) ToF-MS. However, all MS techniques have their
limitations for high-molecular-weight synthetic polymers. When using ESI multiply charged
ions swamp the spectrum, amplifying the number of different ions arising from the
molecular-weight distribution. In MALDI both statistics and charge affinity may cause low-
molecular-weight oligomers to dominate the spectrum.
In the second approach the polymer can be degraded at harsh conditions, such as high
temperature or extreme pH, to complete degradation. This approach is suitable for the
characterization of networks that lack solubility and thus cannot be subjected directly to
chromatographic analysis [10]. When a polymer is being hydrolysed the degree of
degradation can be monitored by NMR spectroscopy. The degradation products can be
separated and quantified by, for example, LC with MS or UV-vis detection. The kinetic
chain length of poly-addition backbones (-C-C-) can be determined by SEC.
The third approach to study prospective biomaterials involves a chemical or a specific
enzymatic degradation under physiological conditions. This allows one to study the kinetics
of degradation. First degradation under physiologically relevant conditions is performed,
resulting in partially degraded material, the constituents of which may be identified [11].
Then complete and fast degradation of the products of the first step (oligomers, intermediates
and other products) is performed, followed by quantitative analysis [30].
The collected information is helpful (i) to ascertain the composition of the original networks,
(ii) to evaluate and optimize the synthesis of functional materials, (iii) to evaluate the
toxicological of the degradation products at an early stage, (iv) to determine the rate of
hydrolysis at different sites prone to attack, and (iv) for the rational design of new materials.
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
7
Figure 1 Schematics for the degradation and analysis of synthetic polymeric biomaterials. This review concerns the green part of this scheme.
4 Degradation methods
Degradable polymeric materials contain moieties that are prone to chemical or enzymatic
degradation (cf. Table 1). The degradation of such materials can be divided as follows.
4.1 Degradation under non-physiological conditions
This type of degradation involves harsh conditions, such as extreme pH values (both acidic
and alkaline) and/or elevated temperatures. Such degradation methods can be used (i) to
measure the kinetic chain length in photo-crosslinked polymeric networks, which lack
solubility [31], (ii) to reduce the size of polymeric chains in complex multi-block
copolymers to allow chromatographic separations, followed by sequence analysis with mass
spectrometry [32], (iii) to estimate the composition of complex polymeric networks by
quantifying each completely hydrolyzed building block [30], or (iv) to investigate the
Chapter 1
8
stability of different chemical bonds in a copolymer under extreme conditions. Haken et al.
reviewed the importance of vigorous chemical degradation of condensation polymers prior
to their chromatographic analysis [33]. Matsubara et al. designed a novel set-up to study
degradation, using supercritical methanol at 300oC and 8.1 MPa in a stainless-steel autoclave
(placed in a GC oven (Figure 2a). They could selectively decompose the ester linkages in
UV-cured acrylic esters [31]. Later, the developed set-up was successfully applied for the
characterization of the network structure of radiation-cured resins of poly-functional acrylic
ester and N-vinylpyrrolidone [34].
Peters et al. investigated the hydrolytic degradation of poly(D,L-lactide-co-glycolide 50:50)-
di-acrylate network coatings. They followed a two-step degradation process, first degrading
the coatings in PBS at 37oC and then further hydrolyzing the released products at 90oC in
10-M sodium hydroxide [30].
To tailor the performance of degradable synthetic polymeric biomaterials, it is important to
understand their structure. High-molecular-weight polymers or complex copolymers are
difficult to analyze by routine SEC or HPLC methods. The analyses usually involve time-
consuming sample-preparation steps especially in case of networks. To assess the structure
of such synthetic polymers, thermal degradation or pyrolysis can be a useful tool. In
pyrolysis, the polymer samples (introduced in the form of a solution or as a solid) break
down into small fragments (e.g. monomers or oligomers) by supplying thermal energy in an
inert atmosphere or vacuum [35]. The small fragments can be separated and analyzed by
chromatographic techniques such as GC or GC-MS [36].
4.2 Degradation under physiological conditions
The degradation of biomaterials under physiological conditions is studied to estimate their
degradation rate and to investigate phenomena involving surface degradation or bulk
degradation. Various other aspects, such as pH changes, degree of swelling, weight loss,
surface chemistry and morphology, and toxicity of the released products can also be studied.
These kinds of degradations are performed at 37oC, with different incubation media, such as
phosphate-buffered-saline (PBS) solution, enzyme-containing buffer, serum, or simulated
body fluids (SBF) at a suitable pH [28]. In conventional batch-mode analysis, the
biodegradable polymers (films, coatings, 3D scaffolds, etc.) are immersed in the respective
media followed by incubation at 37oC [40]. In vitro degradation conditions cannot mimic
real physiological conditions. However, the selection of appropriate enzymes, incubation
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
9
media, ratio’s of surface-to-mass of the specimen and surface-to volume of the medium,
duration of the experiment, and dynamic or static conditions during the degradation with
respect to the site where the biomaterial is implanted may help to find conditions closer to
the physiological ones [28].
Figure 2 (a) Schematic diagram of fast-degradation apparatus used for supercritical methanolysis (reprinted with permission from ref. [31]), (b) Schematic diagram of the reaction vessel used in the dynamic encrustation of urinary-tract devices based on polyurethanes, percuflex and silicone (reprinted with permission from ref. [37]), (c) Schematic showing the apparatus for studying the degradation of biodegradable scaffolds under dynamic conditions. Using a peristaltic pump, the scaffolds were subjected to a continuous flow (250 µl/min) of phosphate-buffered-saline (PBS) solution (pH 7.4) at 37oC (reprinted with permission from ref. [38]), (d) Schematic of dynamic flow simulation system used to study the effect of fluid flow on the degradation of poly(lactide-co-glycolide acid) (PLGA) for in vitro degradation of PLGA/b-TCP composite scaffolds (reprinted with permission from ref. [39]).
At the anatomical sites where there is minimal fluid flow, such as articular cartilage tissues,
the mass-to-surface ratio may strongly influence the degradation kinetics [41]. The level and
type of agitation (rotation, vibration, flow) during degradation may not only affect the
Chapter 1
10
release of the degradation products from the bulk or the surface of the material to the
surrounding media but also influence the contact between soluble reactants (e.g. enzyme)
and the insoluble substrate [28].
Agrawal et al. demonstrated the effect of static and dynamic conditions on the degradation
of scaffolds, fabricated from a copolymer of poly(lactic acid) and poly(glycolic acid), in PBS
at 37oC for up to six weeks. Figure 2d illustrates the apparatus used by the authors to achieve
dynamic conditions. They found that fluid flow decreased the degradation rate significantly
[38]. Gorman et al. investigated the encrustation of urinary-tract devices based on
polyurethanes, Percuflex and silicone in artificial urine under dynamic conditions. The same
level of encrustation was observed under static and dynamic conditions and significantly
higher levels of calcium and magnesium were found under static conditions [37]. In another
study, the effect of fluid flow on the in vitro degradation of poly(L-lactic acid)/β-tricalcium
phosphate (PLLA/ β-TCP) composite in PBS was investigated. Significantly faster
degradation was observed with a dynamic flow-simulation system [39].
Hooper et al. investigated the effects of SBF and PBS on the degradation of tyrosine-derived
polymers. They noticed a good similarity between the in vitro degradation kinetics of the
polymers in PBS and SBF and their in vivo results [42].
5 Chromatographic methods for degradable polymeric biomaterials
Novel degradable polymeric biomedical devices are developed using more-complex
polymers, i.e. random, block and graft copolymers or polymer blends. The characterization
of such polymers requires the use of chromatography. This involves the determination of the
molar-mass distribution, which reflects the length distribution (dispersity) of the polymeric
chain. Another important application of chromatographic systems is the separation of
polymers on the basis of their chemical heterogeneity, functionality type and sequence
lengths [43]. The size and the chemical nature of the degradation products determine the
adoptability of degradable polymers by the in vivo environment. To estimate the
toxicological nature of the degradation products sensitive and selective chromatographic
methods are required [29]. Biomaterials that are soluble in water or common organic
solvents can be analysed with common liquid-chromatographic (LC) methods. Some (but
not all) degradable polymeric materials designed for biomedical applications are of very high
molecular weight or based on insoluble polymeric networks. Such polymeric systems need
to be degraded prior to their chromatographic analysis. For structural analysis, the
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
11
degradation methods involve chemical hydrolysis at harsh conditions, methanolysis, or
partial degradation under mild conditions [10,30]. However, to estimate the degradation rate
and release of degradation products, degradation experiments are carried out under
physiological conditions, such as in PBS or enzyme-containing buffer at 37oC (see section
4.2).
5.1 Size-exclusion chromatography
During the polymerization process in which biodegradable polymers are formed, a large
number of chains are grown. The length of the resulting chains may vary. Therefore, it is
important to determine the molar-mass distribution (MMD; or molecular-weight distribution,
MWD). Size-exclusion chromatography (SEC) (also called gel-permeation chromatography,
GPC), is a popular analytical technique to separate polymer chains based on their size
(hydrodynamic volume). Unlike other LC methods, entropic effects are dominant in SEC
(T∆S >> ∆H) [43]. Mobile phases and packing materials are selected that minimize the
enthalpic interactions of the polymeric chains, so that the partition equilibrium is essentially
governed by the conformational entropy differences among the polymeric chains in the two
phases [44]. The information related to peak-average, number-average, weight-average, and
z-average molar masses (Mp, Mn, Mw, and Mz respectively) can be deducted from the position
and shape of the peak. Differential refractometry (dRI), UV-visible spectrometry, and – to a
lesser extent – evaporative laser-light scattering (ELSD) are concentration-sensitive
detection methods that are widely used in SEC experiments. In such experiments the MMD
and molar masses are typically calculated from a calibration curve, constructed using a set of
narrowly dispersed polymer standards. Light-scattering detection methods, such as multi-
angle laser-light scattering (MALLS) may provide useful information on the molecular size
of polymers, as well as on chain branching, conformation, and aggregation [45]. A change in
the shape and size of polymer molecules in solution influences the viscosity. Therefore,
viscometric detection methods are also used to determine the MMD of polymers [43].
“Triple detection-methods” (typically dRI, light-scattering and viscometry) are used to
determine “absolute” (i.e. accurate) molecular weights of branched and star-shaped
polymers. Absorbance or fluorescence detection and MS may – often in combination with
dRI detection – provide useful information on the distribution of specific fragments within
the chains or end groups in a polymer [46].
Chapter 1
12
Burdick et al. used aqueous SEC-dRI to characterize kinetic-chain-length distributions of
poly(methacrylic acid) (PMAA) in the hydrolysates of highly cross-linked systems based on
methacrylated sebacic acid, designed for orthopaedic applications [47]. The authors
investigated the relationship between kinetic chain length and the structural evolution of the
network. Themistou et al. determined the molecular weights and the molecular-weight
distributions of the hydrolysis products and precursors of cross-linked star polymer model
networks (CSPMNs) [48]. The linear and star polymers and their extractables were
determined by SEC-dRI with tetrahydrofuran (THF) as an eluent. The CSPMSs studied were
based on methyl methacrylate and the diacetal-based dimethacrylate cross-linker bis[(2-
methacryloyloxy)ethoxymethyl] ether and designed for biomedical applications [48].
Mojsiewicz-Piénkowska et al. reviewed the applications of SEC-ELSD for determining the
molecular weights of linear polydimethylsiloxanes (PDMSs). These authors also highlighted
the experimental conditions, such as calibration curve, mobile phase, flow rate and columns
used to characterize the PDMSs and the precision and accuracy of the developed methods
[22].
Peters et al. calculated the kinetic chain length of poly(acrylic acid) (PAA) backbone and the
average lengths of chains between cross-links in UV-cured networks prepared from mixtures
of di-functional (polyethylene–glycol di-acrylate) and mono-functional (2-ethylhexyl
acrylate) acrylates after hydrolysis. They used aqueous SEC coupled on-line to reversed-
phase LC with dRI and mass-spectrometric detection. The results were used to express the
chemical network structure for the different UV-cured acrylate polymers in network
parameters, such as the degree of cross-linking, the number of PAA units which were cross-
linked and the network density [49]. In another study, the same group of authors used
aqueous SEC-dRI to monitor the release of PAA chains during the hydrolytic degradation of
cross-linked poly-(D,L-lactide-coglycolide 50:50)-di-acrylate film. An increase in the
molecular weight with degradation time indicated that the release of these polyacrylate
chains was controlled by the number and type of ester groups that had to be degraded
hydrolytically to dissolve the chains [30].
Lin et al. used SEC with triple detection in chloroform to determine absolute molecular
weights. They confirmed that the star architecture in their biodegradable star polymers
consisted of hydrophilic hyperbranched poly-(ester amide) as core and hydrophobic PCL as
shell [50].
SEC is often used in off-line combinations with information-rich detectors, such as MS, or
FTIR or NMR spectroscopy. Rizzarelli et al. used matrix-assisted laser-desorption/ionization
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
13
time-of-flight mass spectrometry (MALDI-ToF-MS) as an off-line detection method for the
detailed structural characterization of complex polydisperse copolyesters, such as
poly[(R,S)-3-hydroxybutyrate-co-L-lactic acid] and poly[(R,S)-3-hydroxybutyrate-co--
caprolactone]. The results of compositional analysis were in good agreement with NMR
results [51]. Montaudo et al. demonstrated the use of NMR as an off-line detection method
for the compositional analysis of random copolymers with units of methyl methacrylate,
styrene, butyl acrylate and maleic-anhydride. They calculated the polydispersity index of the
copolymers by off-line MALDI-MS of the SEC fractions [52]. Nielen et al. explored the use
of electrospray-ionization – time-of-flight – mass spectrometry (ESI-ToF-MS) as a potential
detector for SEC analysis of polyesters. The absolute mass calibration of the SEC system
based on the polymer itself and determination of monomers and end groups from the mass
spectra were achieved [53].
BA
Figure 3 (A) On-line SEC-1H NMR traces obtained by monitoring the methoxy proton resonance at 3.59 ppm (a) and the α-methyl proton resonances at 0.86 ppm (….........) and 1.20 ppm (———) due to rr- and mm-triads, respectively (b); NMR signals due to α-methyl protons of the PMMA eluted in the elution periods F1 (c), F2 (d) and F3 (e) are also shown (reprinted with permission from ref. [55] ). (B) On-line SEC-NMR analysis of PMMA-block-poly(n-BuMA) prepared with t-C4H9MgBr in toluene at -60oC (reprinted with permission from ref. [56]).
Chapter 1
14
The stereo-regularity of polymeric chains and the chemical composition of copolymers may
affect their degradation rate and processing. On-line coupling of SEC with NMR
(continuous-flow NMR spectroscopy) makes it possible to study directly the chemical
composition and stereochemistry (isotactic, syndiotactic, atactic, etc.) of complex
copolymers separated according to their molecular size [54]. Hatada et al. studied the
tacticity of PMMA with on-line SEC-NMR. The results (Figure 3) showed a higher
concentration of rr-triads (syndiotactic) in fraction F1 of the higher-molecular-weight range
of the SEC chromatogram and a higher concentration of mm-triads (isotactic) in the lower-
molecular-weight fraction F3 [55]. In another study, they investigated the chemical
composition of block (PMMA-block-poly(n-BuMA)) and random (poly(MMA-ran-n-
BuMA)) copolymers of methyl and butyl methacrylates as a function of their MMD by on-
line SEC-NMR [56].
5.2 Adsorption liquid chromatography
Complex degradable polymeric systems are synthesized from (“telechelic”) oligomers and
polymers possessing terminal functional groups. The nature and the number of functional
groups on a chain may vary. Precursors for polymer synthesis, intermediate products, the
produced polymer, and the degradation products after hydrolytic or enzymatic degradation
can be separated based on different numbers of the same functionality or different
functionalities in a polymeric chain by analytical technique, such as adsorption liquid
chromatography (LC). The presence of side-reaction products and chiral impurities in the
degradable synthetic polymer can strongly influence their degradation rate and
biocompatibility [57]. Chromatographic methods with multiple detection methods are
needed for the separation and characterization of such impurities. Adsorption LC involves
enthalpic interactions between the stationary and mobile phases and the analyte molecules
[58,59]. Interactions between flexible polymeric chains in solution and the surface of the
stationary phase depend on the magnitude of the adsorption energy. The higher the
adsorption enthalpy (∆H) the stronger is the adsorption to the packing materials [43].
Adsorption LC is used in the normal-phase (NP) mode (using a polar stationary phase) or in
the reversed-phase (RP) mode (using a non-polar stationary phase).
Vu et al. reported on the use of LC with UV detection at 210 nm for determining the
oligomeric distribution of concentrated lactic-acid solutions [60]. Ding et al. developed an
LC method for the separation and quantification of water-soluble impurities and degradation
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
15
products in PLGA, to estimate changes in the polymer “micro-climate” (e.g. in pH). The
released products containing ester groups were derivatized with a common chromophore to
produce bromophenacyl esters prior to their gradient elution from a C18 column with UV-
vis detection at 254 nm [61]. Al Samman et al. investigated the influence of the degree of
branching on the retention behaviour of linear and branched aromatic polyesters in LC with
UV and ELSD detection. The branched polyesters showed a stronger adsorptive interaction
with the stationary phase than the corresponding linear molecules [58].
MS has been extensively exploited as an on-line detection method for the identification of
oligomers and low-molecular-weight degradable polymers [18,25,62]. Elliott et al. isolated
the degradation products of L-phenylalanine-based segmented polyester urethane ureas
degraded with chymotrypsin on a solid-phase-extraction cartridge for subsequent LC
separation and identification with LC-MS/MS. They observed the cleavage of urea, ester and
urethane bonds [18]. In an interesting study, Tang et al. investigated the enzyme-mediated
degradation of radio-labeled polycarbonate-polyurethanes (PCNUs). The water-soluble
degradation products were separated by LC with diode-array UV detection. The radioactivity
of the collected fractions was measured by a multi-purpose scintillation counter. The
products were identified by LC-MS/MS. The profile of the released degradation products
was in agreement with the structural analysis of synthesized polymers [62]. Deschamps et al.
simulated the in vivo degradation of segmented poly(ether ester)s block copolymers based on
poly(polyethylene glycol) and poly(butylene terephthalate) by their accelerated in vitro
degradation in PBS. They demonstrated the potential of LC-UV-MS for the detailed analysis
of the soluble degradation products. The results showed high amounts of the PEO fraction in
the soluble degradation products, while a PEOT/PBT fraction was found to be insoluble. The
results were confirmed with NMR [25].
Rizzarelli et al. found evidence for selective hydrolysis of aliphatic copolyesters, such as
poly(butylene succinate-co-butylene adipate), P(BS-co-BA), and poly(butylene succinate-co-
butylene sebacate), P(BS-co-BSe) induced by lipase. The water-soluble products, including
co-oligomers with identical molecular weights, but different sequences, were separated and
identified by on-line LC-ESI-MS/MS). The results showed a preferential cleavage of sebacic
ester bonds in P(BS-co-BSe) and succinic ester bonds in P(BS-co-BA) [32]. Carstens et al.
investigated the in vitro chemical and enzymatic degradation of monodisperse oligo(-
caprolactone) (OCL) and its block copolymer with methoxy poly(ethylene glycol) (mPEG-
b-OCL) by monitoring the water-soluble degradation products with LC-MS. The slow
degradation of OCL ester micelles in phosphate buffer at pH 7.4 was accelerated by lipase
Chapter 1
16
[63]. Pulkkinen et al. reported a fast analysis of the soluble degradation products of 2,2′-
bis(2-oxazoline)-linked poly-ε-caprolactone (PCL-O), degraded in simulated intestinal fluid
by LC-ESI-MS/MS. The polymer degraded primarily by ester hydrolysis, while amide bonds
showed greater stability [23].
Peters et al. demonstrated the use of LC-MS to identify and quantify the various water-
soluble oligomeric and polymeric degradation products released during the hydrolytic
degradation of poly(D,L-lactide-co-glycolide 50:50)-di-acrylate networks. The products
were analyzed directly after release and also after complete hydrolysis of the soluble fraction.
They found a rapid release of residual photo-initiator followed by a gradual release of
lactide/di-ethyleneglycol/glycolide oligomers with varying chain length and composition
[64].
Figure 4 (1) Liquid chromatographic separation of PEG 1000 (chromatograms A and B) and soluble products derived from PEOT:PBT (71:29) copolymer during hydrolytic degradation at 100oC (chromatograms C and D) with UV detection at 251 nm (chromatograms A and C) and mass-spectrometric detection applying atmospheric-pressure chemical ionization in the positive-ion mode (APCI(+)) conditions (chromatograms B and D) recorded in scan mode (m/z=200–1600) (reprinted with permission from ref. [25]). (2) UV-absorbance and radioactivity chromatograms for the degradation products from radio-labeled polycarbonate-polyurethanes: (a) buffer incubation and (b) cholesterol esterase incubation (reprinted with permission from ref. [62]).
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
17
The hyphenation of the most powerful spectroscopic technique i.e. NMR (containing solvent
suppression features) with HPLC is recently getting more attention for the online chemical
composition and molar mass determination of oligomers and low MW polymers separated
on reversed phase HPLC column according to their chemical structure. Pasch et al.
investigated the chemical structure, molar mass and end group analysis of poly(ethylene
oxide) by an online HPLC-NMR setup [65].
5.3 Liquid chromatography at critical conditions
Liquid chromatography at critical conditions (LCCC) is receiving increased attention for the
separation of complex polymers. LCCC separates the polymers at the so-called critical
conditions, i.e. the chromatographic conditions where the enthalpy and entropy effects
compensate each other (∆H = T∆S) [66]. Under these conditions the retention of polymeric
species becomes independent of their molecular weight [66,67]. LCCC allows the separation
of polymers on the basis of their functionality type distributions (FTDs). It has been applied
for the separation of functional polymers, block copolymers, branched polymers, and
polymer blends and to assess their stereo-regularity [15,58,59,66,68]. The incorporation of
hydrophilic and hydrophobic components and stereo-regularity of degradable copolymers for
biomedical application control their degradation behaviour. Lee et al. resolved oligomeric
PLLA block species of poly(ethylene oxide)-block-poly(L-lactide), (PEO-b-PLLA) by
RPLC at the critical conditions of PEO. They confirmed the composition of each species by
off-line MALDI-MS analysis [67]. In another study, they fractioned the LLA units in tri-
block PLLA-b-PEO-b-PLLA copolymer by RPLC at the critical conditions of PEO and
confirmed the results by off-line MALDI-MS. In tri-block copolymer, unlike in di-block
(PEO-b-PLLA) copolymer, they observed a splitting of the eluted peaks containing the same
number of LLA units. They assigned this peak splitting to the different length distributions
of PLLA blocks at the two ends of the PEO block [14]. Mengerink et al. developed a method
for the separation of linear and cyclic oligomers of polyamide-6 by LCCC-ELSD. ESI-MS
did not allow discrimination between the linear and cyclic products.[15]. Peters et al.
reported the FTD of functional PMMA, obtained by LCCC-ELSD. The mono- and
bifunctional PMMA peaks were identified by ESI- MS [68].
Philips et al. discussed novel developments in water-based LCCC. They varied the buffer
concentration and the proportion of organic modifier in the mobile phase to approach the
critical condition for two polymer systems, viz. poly(styrene sulfonate) and poly(acrylic
Chapter 1
18
acid). The critical condition of poly-(acrylic acid) was then used to study the retention
characteristics of a copolymer containing both acrylic acid and n-vinyl pyrrolidinone [69].
De Geus et al. utilized LCCC, with UV and ELSD detection, to separate PCL polymer
samples with different end groups, in order to gain insight in the initiation process of
enzymatic ring-opening polymerization. PCL chains with three different end groups were
separated, i.e. (i) linear carboxylic-acid end-functionalized species, (ii) linear hydroxyl-ester
species, and (iii) cyclic species. The identity of each peak was confirmed by offline MALDI-
MS [70].
LCCC with on-line NMR analysis can provide detailed information on the end groups and
chemical composition of polymeric chains. Hiller et al. demonstrated the use of on-line
NMR detection for the analysis of complex mixtures of fatty alcohol ethoxylates (FAEs) by
LCCC. The peaks were detected using an ELSD detector [71]. In another study, they
investigated the separation of block copolymers of PS-b-PMMA and blends of PS and
PMMA at the size-exclusion conditions for PS and critical conditions of PMMA with on-line 1H NMR detection [72]. Unfortunately, as the critical conditions are strongly dependent on
the mobile-phase composition, but also on temperature and pressure [73], especially for
high-molecular-weight polymers, LCCC is only rarely applied successfully to polymer
systems with molecular weights exceeding 100 kDa.
5.4 Two-dimensional liquid chromatography
Complex polymers, including degradable synthetic polymers, exhibit several simultaneous
distributions. For example, all functionalized polymers exhibit an MMD and a functionality-
type distribution (FTD) and all copolymers exhibit an MMD and a chemical-composition
distribution (CCD) [43,74]. Moreover, the different distributions in complex polymers tend
to be mutually dependent [74]. SEC or HPLC by themselves may not reveal correct
information on the MMD or the molecular heterogeneity of the polymeric chains [46]. To
gain insight in multiple, mutually dependent distributions analytical techniques such as
multi-dimensional separations are indispensable [74].
Kilz et al. provided a detailed description of the two-dimensional chromatographic
techniques for polymers [75]. Pasch et al. reported on the two-dimensional separation of
PEO-b-PPO block copolymers. In the first dimension, they separated copolymers with
respect to the length of the PEO block by LCCC. The collected fractions were further
separated in the second dimension, either by supercritical-fluid chromatography (SFC) or by
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
19
SEC based on the length the PEO blocks [76]. Early two-dimensional LC techniques were
based on “heart-cut” or fractionation methods and they were very specialized or time-
consuming [74]. With the advent of modern technology, two-dimensional LC methods are
becoming faster and more comprehensive.
Van der Horst et al. wrote a highly useful review on the advantages of comprehensive
LC×LC for polymers over “heart-cutting” LC-LC [59]. Kok et al. reported on FTIR as an
on-line detection method in comprehensive LC×SEC for the characterization of copolymers
based on styrene and methacrylates [77]. The results were confirmed by UV detection. The
generated functional-group contour plots showed a distinction between UV-active and non-
UV-active groups of the polymer.
Figure 5 LC×LC contour plots of (A) poly(2-ethylhexyl acrylate) P2EHA macro(RAFT agent), (B) copolymer (2-ethylhexyl acrylate and methyl acrylate)-1h, (C) copolymer-2h, (D) copolymer-4h, and (E) copolymer-8h. (F) is a rotation of 90o and inclination of 35o of the LC×LC chromatogram of the sample of copolymer-8h. 1st dimension: gradient LC with 0 to 70% THF in methanol in 200 min at 0.05 mL/min on PLRP-S 5 μm (Y-axis). 2nd dimension: SEC with THF at 1.5 mL/min on PL HTS-C (X-axis). Calibration: PMMA. Detection: ELSD (reprinted with permission from ref. [79]).
Chapter 1
20
The hydrodynamic volume of a branched polymer of certain molecular weight may be
identical to that of a linear polymer of lower molecular weight. Based on this principle Edam
et al. demonstrated the use of comprehensive two-dimensional molecular-topology
fractionation (MTF) × SEC for the separation of branched polymers based on their topology
[78]. Raust et al. performed two-dimensional LC separations with a combination of LCCC
and SEC to gain insight in the polymerization process of copolymers based on 2-ethylhexyl
acrylate and methyl acrylate, (P2EHA-b-PMA), produced by reversible addition-
fragmentation chain transfer (RAFT)-mediated polymerization in organic dispersion (Figure
5). The LC×SEC chromatograms revealed a certain heterogeneity of the polymer and
allowed the precise characterization of the MA block length in the copolymer. For
compositional analysis the results were confirmed by LC-1H NMR [79]. In summary,
LC×LC methods can be useful for the characterization of complex degradable polymers in
terms of several distributions (MMD, CCD, FTD, etc.) simultaneously. LC×LC can provide
an efficient, reliable and comprehensive characterization of biodegradable polymers.
6 Gas chromatography
Gas chromatography (GC) is another powerful analytical tool for the identification and
quantification of impurities, additives, and degradation products of degradable polymeric
biomaterials. GC is most often applied in combination with flame-ionization detection (FID)
and MS for the analysis of oligomers and low-molecular-weight polymers [36]. Barlow et al.
reviewed the applications of GC in combination with pyrolysis for the analysis and
characterization of polymer degradation [80].
Hakkarainen et al. investigated the nature of low-molecular-weight degradation products of
PLA, PGA, and their copolymers. They reported a convenient and rapid solid-phase-
extraction (SPE) – derivatization technique to improve the qualitative and quantitative GC-
MS analysis of hydroxy acids released by the degradation of PLA and PGA in buffer
solution [81-83]. The GC-MS analyses showed a difference in the patterns of degradation
products released in biotic and abiotic media. In another study this group utilized GC-FID to
explore single-drop micro-extraction (SDME) in combination with multiple-headspace
(MHS) extraction for the quantitative determination of lactide in thermally oxidized
polylactide [84]. During a study of the esterification reaction between lactic acid and
different fatty acids, Torres et al. utilized GC-MS to estimate the degree of polymerization in
polymerized fractions of commercial LA [85]. Vu et al. characterized the oligomeric
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
21
distribution of lactic acid in aqueous media by GC-MS [60]. Urakami et al. reported a rapid,
precise and accurate compositional analysis of co-poly(DL-lactic/glycolic acid) (PLGA).
This was performed by pyrolysis – gas chromatography – mass spectrometry (Py-GC/MS)
combined with one-step hydrolysis and methylation in the presence of
tetramethylammonium hydroxide (TMAH). They found good agreement of the analytical
results with 1H NMR data [86]. Plikk et al. studied the chemical changes in porous scaffolds
based on various L,L-lactide (LLA), 1,5-dioxepane-2-one (DXO) and -caprolactone (CL)
copolymers after sterilization with electron beam and gamma irradiation [87]. The formation
of low-molecular-weight degradation products was studied by GC-MS. Burford et al.
described the rapid qualitative and quantitative analysis of polyester-based polyurethane
elastomers by GC-FID after polymer cleavage into the corresponding glycol, dicarboxylic
acid and diamine fragments by molten alkali fusion at high temperature [88]. All the
caroboxylic-acid products were reacted to dimethyl ester derivatives prior to their GC
analysis [88]. Mallepally et al. investigated the enzymatic degradation of hyperbranched
polyesters (HBPEs). The release of free fatty acids was studied using GC [89].
Figure 6 Pyrograms of UV-curable resins based on bifunctional poly(ethylene glycol)-diacrylate (PEDA). Pyrolysis at 400°C in the presence of TMAH. (a) prepolymer; (b) UV-cured resin. (reprinted with permission from ref. [91]).
Chapter 1
22
Eldsäter et al. studied the degradation of poly (ester amide) and poly(buty1ene adipate-co-
caproamide) in aqueous environment at 37oC, 60oC, and 80oC. GC-MS with SPE was used
to investigate the nature of degradation products at different degradation conditions. Changes
in the polymer composition were investigated by Py-GC-MS [90]. Matsubara et al. used
pyrolysis GC (Py-GC) in the presence of tetramethylammonium hydroxide (TMAH) to
characterize the network structure in UV-cured bifunctional poly(ethylene glycol)-diacrylate
(PEDA) [91].
Kaal et al. developed a fully automated on-line SEC-Py-GC-MS method [92]. The polymer
samples were separated based on molecular size and fractions were transferred on-line. The
SEC solvent was evaporated in a programmable-temperature-vaporizer (PTV) injector prior
to pyrolysis and GC-MS analysis. The scope of the method was extended to include aqueous
SEC and RPLC by introducing a sintered liner, filled with sintered glass beads (60-100 µm)
to approximately half of the cross sectional area [93]. The developed systems provided a
great deal of quantitative insight in the composition of the on-line collected LC or SEC
fractions. Recently, Chojnacka et al. investigated the effect of monomeric ratio of N-vinyl-2-
pyrrolidone (VP) and vinyl acetate (VA) on the dissolution behaviour of their copolymers in
water using Py-GC-MS. The compositional analysis of the fractions, collected at different
time intervals during the dissolution study, revealed that copolymers with higher contents of
VA dissolve considerably slower than the other copolymers [94].
7 Direct mass-spectrometric analyses
Mass spectrometry has emerged as a powerful analytical tool for the characterization of
synthetic polymers and copolymers. A time-of-flight (ToF) mass spectrometer offers high
sensitivity for multi-ion detection, a large mass range, and good mass resolving power.
Therefore, ToF-MS is most commonly used as a mass analyzer for the characterization of
polymers [95]. A ToF-MS can be conveniently combined with ESI or with MALDI.
However, when using ESI multiply charged ions are usually formed, which complicates the
interpretation of the sprectra. In MALDI both statistics and “charge-ability” (a combination
of several parameters, including affinity to charge and efficiency of transfer from solid to
vacuum) may cause low-molecular-weight oligomers to dominate the spectrum. Several
books and reviews have been published that describe the developments in the field of mass
spectrometry of synthetic polymers [95-99].
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
23
ESI is a soft ionization technique, which produces multiply charged ions and little
fragmentation. The analyte solution emerging from the column is nebulized, ionized and
transferred to the vacuum of the mass analyzer and the MS detector [98]. Andersson et al.
compared the degradation stability of stereo-complex poly(L-lactic acid)/ poly(D-lactic acid)
(PLLA/PDLA) with plain PLLA. The composition of degradation products was estimated
semi-quantitatively by direct ESI-MS. The results showed a shorter hydrolysis time for
PLLA/PDLA and more acidic degradation products [13]. Höglund et al. studied the effect of
plasticizer (acetyl tributyl citrate) on the degradation of PLA. They investigated the water-
soluble products and the plasticizer by ESI-MS [100]. In another study, this group studied
the effect of surface modification on the hydrolytic degradation and investigated the
degradation of PLA grafted with poly(acrylic acid) (PAA). The water-soluble degradation
products were analyzed by ESI-MS [101]. Recently, Rizzarelli et al. developed a convenient
direct ESI-MS method to determine concentrations of sebacic-acid (SA) and terephthalic-
acid (TA) residues in biodegradable copolymers. The obtained results were in agreement
with LC-UV data. The assay was proposed as a fast and sensitive alternative to currently
employed methods for acid quantification [102].
MALDI is also a soft ionization technique. It allows the detection of large, non-volatile and
labile molecules. The compounds of interest are desorbed and ionized by the combined
influence of a laser beam and a chemical matrix, usually under vacuum. The resulting
(predominantly singly charged) ions are directed to a (ToF) mass spectrometer by a
continuous high voltage [99]. Burkoth et al. used MALDI-ToF-MS to characterize the
molecular-weight distribution of (mostly) linear poly(methacrylic acid) degradation products
as a function of the network evolution (i.e. double-bond conversion), rate of initiation, and
monomer size during the degradation of cross-linked polymers based on PMA and sebacic
acid [40]. Rizzarelli et al. employed MALDI with tandem mass spectrometry (MSn) to
investigate the fragmentation pathways of poly(butylene adipate) (PBAd) oligomers [103].
In another study, they applied post-source-decay (PSD) MALDI-ToF-MSn for the sequence
determination of aliphatic poly(ester amide)s synthesized from dimethyl sebacate or sebacic
acid and 2-aminoethanol or 4-amino-1-butanol [104]. Luo et al. found a symmetric
distribution in low-molecular-weight star polymers prepared by grafting poly(ethylene
glycol) (PEG) arms onto a cholic acid core via anionic polymerization [105]. Weidner et al.
performed fragmentation analysis by means of MALDI with collision-induced dissociation
(CID) and MSn to determine sequences and end groups of complex copolyesters based on
hexanediol-neopentylglycol-adipic acid copolyesters [106].
Chapter 1
24
During the course of the degradation process, the surface chemistry of the degradable
polymeric device may influence the biological environment. Thus, it plays an important role
in determining its biocompatibility [27]. Therefore, techniques for characterizing the surface
of the biomaterials are gaining attention. Secondary ion mass spectrometry (SIMS) in
combination with ToF can be used for surface characterization and (quantitative) analysis of
synthetic copolymers and polymeric blends [96,107]. Belu et al. reviewed the application of
ToF-SIMS for the structural characterization of biomaterial surfaces. They described the
technique as a flexible and powerful surface-characterization tool [108]. Chen et al. used
ToF-SIMS to study the in vitro hydrolytic degradation at the surface of different
biodegradable polymers, including PLA, PGA, PLGA, poly(sebacic acid) (PSA), and two
random copolymers of poly(fumaric-co-sebacic) acid (PFS) of different compositions. It was
reported that useful information on reaction kinetics can be obtained from the ToF-SIMS
spectra by analyzing the intensities of the molecular ions in the distribution [107].
Figure 7 (A) The 2 m drift-tube IMS-MS instrument design and operationa and (B) typical output for ions separated in the gas-phase detected by MS in different modes ofoperation (cf. details in ref. [110]). (reprinted with permission from ref. [110]).
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
25
Recently, the combination of ion-mobility spectroscopy (IMS) through CID with MS has
been gaining recognition for the structural characterization of synthetic polymers [97]. In ion
mobility, ions are separated on the basis of their conformational state (size and shape), as
they drift through a gas (e.g. He, N2) under the influence of an electric field [109]. A major
benefit of including IMS as an intermediate stage in LC with MS detection is the reduction
of “chemical noise” due to the additional selectivity of IMS. This is especially important
when determining trace amounts of compounds in complex mixtures such as body fluids.
Trimpin et al. reported on the use of a multi-dimensional IMS-MS methodology that
rendered a detailed view of molecular components in complex mixtures, based on the
combined analysis of the three-dimensional geometries and masses of polymeric components
adducted with metal cations in the gas phase [110]. The structure of PEGs with different
functionalities, PPG, poly(tetramethylene glycol) (PTMEG), and several poly(alkyl
methacrylate)s (PAMA)s (with alkyl = methyl, ethyl, butyl, etc.) was investigated using
IMS-MS.
8 Nuclear-magnetic-resonance spectroscopy
Nuclear-magnetic-resonance (NMR) spectroscopy is an extensively used analytical
technique in the field of synthetic polymers. The microstructure, region-isomerism,
stereochemical isomerism, geometric isomeric, branches and end groups, copolymer
composition, number-average molecular weight (Mn), chain conformation, and
intermolecular association of the polymers are among the parameters that can be investigated
by high-resolution NMR and 2D NMR experiments [111]. LeMaster et al. studied the effect
of T1 and T2 relaxation on the 2D 1H-13C correlation spectra of linear commercial polymers.
The results were used to estimate the concentration of end groups in polyester urethanes (Mn
40 kDa). They estimated an uncertainty in Mn of 6-7% (r.s.d.) [112]. Two-dimensional
homo-nuclear correlation spectroscopy (1H-1H COSY) was used to confirm the formation of
poly(α-peptide) in the protease-catalyzed polymerization of L-glutamic acid diethyl ester
hydrochloride [113]. Pergal et al. synthesized polyurethanes-siloxane copolymers containing
high contents of PCL-PDMS-PCL segments [114]. The structure of copolymers, the lengths
of hard and soft segments, and the connectivities between homonuclear or heteronuclear
atoms with single or multiple bond were investigated by 1H, 13C NMR and 2D NMR
experiments, such as 1H-1H COSY, HSQC (heteronuclear single quantum coherence), and
HMBC (heteronuclear multiple-bond correlation). In an interesting study, the generation of
Chapter 1
26
hyper-branched poly(amine-ester)s was confirmed by 13C-, DEPT-135 NMR and 2D NMR
techniques [115]. Shaver et al. attached six arms of poly(lactic acid) to dipentaerythritol
cores. 1H NMR experiments provided useful information on the tacticity of the synthesized
star polymer [116]. To investigate branching, Cooper et al. performed 1H-, 13C-, COSY, and
HSQC NMR experiments and SEC to determine the number of end groups and repeating
units in the backbone of poly(lactic acid)-polyurethane functionalized with pendent
carboxylic-acid groups [117].
9 Conclusions
Separation methods based on chromatography are essential analytical tools to estimate the
FTD, CCD, and MMD of complex degradable polymeric systems. The analysis of the
chemical nature of the degradation products not only highlights the stability of different
degradable bonds, but also reflects the toxicological nature and the biocompatibility of
biomaterials. SEC with dRI, UV-vis or ELSD detectors yields molecular-weight
distributions. Light-scattering or viscometry may provide additional information, such as
molecular size and absolute molar masses. LC separates polymers on the basis of their
functionality and chemical composition. LCCC is a method of choice to separate low-
molecular-weight functional polymers, copolymers and polymer blends at critical conditions.
NMR spectroscopy as an on-line detector for SEC, LC, or LCCC provides useful
information about the functionality, chemical composition, and tacticity of the polymeric
chains along their MMD. Coupling SEC on-line with MS also broadens its scope.
Comprehensive two-dimensional LC techniques, such as LC×LC, LC×SEC, LCCC×SEC,
MTF×SEC, etc. are developing into promising analytical tools for the detailed analysis of,
for example copolymers, branched polymers, and polymer blends. Gas chromatography is
extensively used to separate and identify degradation products. For complex polymers and
networks, Py-GC-MS provides more insight in the chemical composition by pyrolyzing the
liquid or solid samples. Direct mass spectrometric methods, such as ESI-ToF-MS, MALDI-
ToF-MS, and SIMS can provide rapid analysis of the chemical composition of oligomers or
low-molecular-weight polymers. MALDI and SIMS allow studying the surface chemistry
before and after degradation. IMS-MS promises to contribute to utilize the 3D structure of
the polymers for additional selectivity. Despite the limitations associated with each
analytical technique, a combination of selective and sensitive methods can usually be
devised for the analysis for different classes of polymeric biomaterials.
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
27
10 Scope of the thesis
The objective of this thesis is to develop new analytical methods by exploring different
analytical techniques. The thesis deals with the degradation and analysis of synthetic
polymeric biomaterials. The performance of degradable polymeric biomaterials depends on
their chemical structure and on the chemical nature of their degradation products. Therefore,
Chapter 1 reviews the different strategies and analytical techniques for the chemical analysis
of degradable polymeric biomaterials, in particular those based on chromatographic
separations.
In Chapter 2 of this thesis, the quantitative structural analysis of polymeric networks that are
insoluble under normal physiological conditions is described. An in vitro method is
developed that is well suited for the quick and complete hydrolytic degradation of poly(2-
hydroxyethyl methacrylate) (pHEMA), poly(lactide-coglycolide50:50)1550-diol
(PLGA(50:50)1550-diol) and polyester-urethane-acrylate-based networks, using a microwave
set-up. The microwave glass vials are coated internally with Teflon (PTFE) to avoid contact
of the alkaline medium with the inner surface of the glass vessel and thus prevent the
formation of residues. The degree of hydrolysis is monitored by NMR spectroscopy. The
hydrolyzed components can be separated by liquid chromatography and quantified by mass
spectrometry or UV-vis detection. The kinetic chain length of poly-addition backbones (-C-
C-) can be determined by SEC.
Chapter 3 describes an in vitro enzymatic degradation of multi-block PEA with α-
chymotrypsin and proteinase-K at 37oC. The release of different monomeric and oligomeric
products is monitored by LC-ESI-ToF-MS. Semi-quantitative analysis of water-soluble
degradation products reveals the protease and/or amidase activity and provides indications of
the relative fragment (bond) stabilities. The polymer does not degrade by chemical
degradation under physiological conditions. The structure of the polymer is characterized by
NMR spectroscopy.
In Chapter 4 the development of a miniaturized and automated system is reported that can be
used for the fast, on-line investigation of the the in-vitro enzymatic degradation of PEA
coatings. The system can be used under both static and dynamic (flow) conditions and
includes on-line LC-ToF-MS analysis of the hydrolysate (containing enzyme and
degradation products). The system is investigated with respect to different injection volumes
(pulses) of an α-chymotrypsin (α-CT) solution, flow rates of injected α-CT band or α-CT
solution through the coated capillary, concentration of α-CT, and lengths of coated capillary.
Chapter 1
28
The versatility of the system makes it easy to follow the course of degradation and to
differentiate between primary and secondary degradation products.
11 References
[1] D.F. Williams, Biomaterials 30 (2009) 5897. [2] G. Wnek, G.L. Bowlin, Encyclopedia of Biomaterials and Biomedical Engineering,
Informa Health Care, 2004. [3] G. Mabilleau, M.F. Moreau, R. Filmon, M.F. Baslé, D. Chappard, Biomaterials 25
(2004) 5155. [4] Aylvin A. Dias, M. Hendriks, Drug Delivery Technology 10 (2010) 20. [5] P.J.A. in 't Veld, P.J. Dijkstra, J. Feijen, Clinical Materials 13 (1993) 143. [6] H. Sun, F. Meng, A.A. Dias, M. Hendriks, J. Feijen, Z. Zhong, Biomacromolecules
12 (2011) 1937. [7] M. Hakkarainen, in, Springer Berlin / Heidelberg, 2002, p. 113. [8] K.D.F. Vlugt-Wensink, X. Jiang, G. Schotman, G. Kruijtzer, A. Vredenberg, J.T.
Chung, Z. Zhang, C. Versluis, D. Ramos, R. Verrijk, W. Jiskoot, D.J.A. Crommelin, W.E. Hennink, Biomacromolecules 7 (2006) 2983.
[9] R.S. Labow, Y. Tang, C.B. McCloskey, J.P. Santerre, Journal of Biomaterials Science, Polymer Edition 13 (2002) 651.
[10] A. Ghaffar, P.G. Verschuren, J.A.J. Geenevasen, T. Handels, J. Berard, B. Plum, A.A. Dias, P.J. Schoenmakers, S. van der Wal, Journal of Chromatography A 1218 (2011) 449.
[11] A. Ghaffar, G.J.J. Draaisma, G. Mihov, A.A. Dias, P.J. Schoenmakers, S. van der Wal, Biomacromolecules 12 (2011) 3243.
[12] B.D. Ulery, L.S. Nair, C.T. Laurencin, Journal of Polymer Science Part B: Polymer Physics 49 (2011) 832.
[13] S.R. Andersson, M. Hakkarainen, S. Inkinen, A. Södergård, A.-C. Albertsson, Biomacromolecules 11 (2010) 1067.
[14] H. Lee, T. Chang, D. Lee, M.S. Shim, H. Ji, W.K. Nonidez, J.W. Mays, Analytical Chemistry 73 (2001) 1726.
[15] Y. Mengerink, R. Peters, C.G. deKoster, S. van der Wal, H.A. Claessens, C.A. Cramers, Journal of Chromatography A 914 (2001) 131.
[16] B. Rubehn, T. Stieglitz, Biomaterials 31 (2010) 3449. [17] E. Wisse, R.A.E. Renken, J.R. Roosma, A.R.A. Palmans, E.W. Meijer,
Biomacromolecules 8 (2007) 2739. [18] S.L. Elliott, J.D. Fromstein, P.S. J, K.A. Woodhouse, Journal of Biomaterials
Science, Polymer Edition 13 (2002) 691. [19] K.E. Uhrich, A. Gupta, T.T. Thomas, C.T. Laurencin, R. Langer, Macromolecules 28
(1995) 2184. [20] M. Kato, K. Toshima, S. Matsumura, Biomacromolecules 6 (2005) 2275. [21] Z. Zhao, J. Wang, H.-Q. Mao, K.W. Leong, Advanced Drug Delivery Reviews 55
(2003) 483. [22] K. Mojsiewicz-Pienkowska, Journal of Chromatography B 865 (2008) 7. [23] M. Pulkkinen, J.J. Palmgrén, S. Auriola, M. Malin, J. Seppälä, K. Järvinen, Rapid
Communications in Mass Spectrometry 22 (2008) 121. [24] J. Zimmermann, T. Loontjens, B.J.R. Scholtens, R. Mülhaupt, Biomaterials 25
(2004) 2713.
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
29
[25] A.A. Deschamps, A.A. van Apeldoorn, H. Hayen, J.D. de Bruijn, U. Karst, D.W. Grijpma, J. Feijen, Biomaterials 25 (2004) 247.
[26] E.M. Christenson, S. Patel, J.M. Anderson, A. Hiltner, Biomaterials 27 (2006) 3920. [27] B.D. Ratner, A. Chilkoti, D.G. Castner, Clinical Materials 11 (1992) 25. [28] R.L. Reis, J.S. Román, Biodegradable Systems in Tissue Engineering and
Regenerative Medicine, CRC Press, 2004. [29] A.-C. Albertsson, M. Hakkarainen, in Chromatography for Sustainable Polymeric
Materials -Renewable, Degradable and Recyclable ; Advances in Polymer Science 211, Springer, 2008, p. 85.
[30] R. Peters, J. Lebouille, B. Plum, P. Schoenmakers, S. van der Wal, Polymer Degradation and Stability In Press, Accepted Manuscript (2011).
[31] H. Matsubara, S.-i. Hata, Y. Kondo, Y. Ishida, H. Takigawa, H. Ohtani, Analytical Sciences 22 (2006) 1403.
[32] P. Rizzarelli, G. Impallomeni, G. Montaudo, Biomacromolecules 5 (2004) 433. [33] J.K. Haken, TrAC Trends in Analytical Chemistry 9 (1990) 14. [34] H. Matsubara, H. Kataoka, H. Ohtani, Polymer Chemistry 1 (2010) 1056. [35] W.J. Irwin, Analytical Pyrolysis: A Comprehensive Guide, Marcel Dekker Inc., New
York, 1982. [36] F.C.-Y. Wang, A.D. Burleson, Journal of Chromatography A 833 (1999) 111. [37] S.P. Gorman, C.P. Garvin, F. Quigley, D.S. Jones, Journal of Pharmacy and
Pharmacology 55 (2003) 461. [38] C.M. Agrawal, J.S. McKinney, D. Lanctot, K.A. Athanasiou, Biomaterials 21 (2000)
2443. [39] Y. Yang, Y. Zhao, G. Tang, H. Li, X. Yuan, Y. Fan, Polymer Degradation and
Stability 93 (2008) 1838. [40] A.K. Burkoth, K.S. Anseth, Macromolecules 32 (1999) 1438. [41] A.M.A. Ambrosio, H.R. Allcock, D.S. Katti, C.T. Laurencin, Biomaterials 23 (2002)
1667. [42] K.A. Hooper, N.D. Macon, J. Kohn, Journal of Biomedical Materials Research 41
(1998) 443. [43] H. Pasch, B. Trathnigg, HPLC of Polymers, Springer-verlag Berlin Heidelberg, 1998. [44] T. Chang, Journal of Polymer Science Part B: Polymer Physics 43 (2005) 1591. [45] A. Streigel, W.W. Yau, J.J. Kirkland, D.D. Bly, Modern Size-Exclusion Liquid
Chromatography: Practice of Gel Permeation and Gel Filtration Chromatography, John Wiley & Sons Inc., Hoboken, New Jersey., 2009.
[46] T. Provder, H.G. Barth, M.W. Urban, Chromatographic Characterization of Polymers: Hyphenated and Multidimensional Techniques, American Chemical society, 1995.
[47] J.A. Burdick, T.M. Lovestead, K.S. Anseth, Biomacromolecules 4 (2003) 149. [48] E. Themistou, C.S. Patrickios, Macromolecular Chemistry and Physics 209 (2008)
1021. [49] R. Peters, V.M. Litvinov, P. Steeman, A.A. Dias, Y. Mengerink, R. van Benthem,
C.G. de Koster, S. van der Wal, P. Schoenmakers, Journal of Chromatography A 1156 (2007) 111.
[50] Y. Lin, X. Liu, Z. Dong, B. Li, X. Chen, Y.-S. Li, Biomacromolecules 9 (2008) 2629.
[51] G. Adamus, P. Rizzarelli, M.S. Montaudo, M. Kowalczuk, G. Montaudo, Rapid Communications in Mass Spectrometry 20 (2006) 804.
[52] M.S. Montaudo, Polymer 43 (2002) 1587. [53] M.W.F. Nielen, F.A. Buijtenhuijs, Analytical Chemistry 71 (1999) 1809.
Chapter 1
30
[54] K. Albert, On-Line LC–NMR and Related Techniques, John Wiley & Sons Ltd, 2002.
[55] K. Hatada, K. Ute, T. Kitayama, T. Nishimura, M. Kashiyama, N. Fujimoto, Polymer Bulletin 23 (1990) 549.
[56] K. Hatada, K. Ute, T. Kitayama, M. Yamamoto, T. Nishimura, M. Kashiyama, Polymer Bulletin 21 (1989) 489.
[57] S. Inkinen, M. Hakkarainen, A.-C. Albertsson, A. Södergård, Biomacromolecules 12 (2011) 523.
[58] M. Al Samman, W. Radke, A. Khalyavina, A. Lederer, Macromolecules 43 (2010) 3215.
[59] A. van der Horst, P.J. Schoenmakers, Journal of Chromatography A 1000 (2003) 693.
[60] D.T. Vu, A.K. Kolah, N.S. Asthana, L. Peereboom, C.T. Lira, D.J. Miller, Fluid Phase Equilibria 236 (2005) 125.
[61] A.G. Ding, S.P. Schwendeman, Journal of Pharmaceutical Sciences 93 (2004) 322. [62] Y.W. Tang, R.S. Labow, J.P. Santerre, Biomaterials 24 (2003) 2805. [63] M.G. Carstens, C.F. van Nostrum, R. Verrijk, L.G.J. de Leede, D.J.A. Crommelin,
W.E. Hennink, Journal of Pharmaceutical Sciences 97 (2008) 506. [64] R. Peters, J. Lebouille, B. Plum, P. Schoenmakers, S. van der Wal, Polymer
Degradation and Stability In Press, Corrected Proof (2011). [65] H. Pasch, W. Hiller, Macromolecules 29 (1996) 6556. [66] S.G.Entelis, V.V.Evreinov, A.V.Gorshkov, Advances in Polymer Science 76 (1986)
129. [67] H. Lee, W. Lee, T. Chang, S. Choi, D. Lee, H. Ji, W.K. Nonidez, J.W. Mays,
Macromolecules 32 (1999) 4143. [68] R. Peters, Y. Mengerink, S. Langereis, M. Frederix, H. Linssen, J. van Hest, S. van
der Wal, Journal of Chromatography A 949 (2002) 327. [69] S.L. Phillips, L. Ding, M. Stegemiller, S.V. Olesik, Analytical Chemistry 75 (2003)
5539. [70] M. de Geus, R. Peters, C.E. Koning, A. Heise, Biomacromolecules 9 (2008) 752. [71] W. Hiller, A. Brüll, D. Argyropoulos, E. Hoffmann, H. Pasch, Magnetic Resonance
in Chemistry 43 (2005) 729. [72] W. Hiller, P. Sinha, H. Pasch, Macromolecular Chemistry and Physics 210 (2009)
605. [73] Y. Mengerink, R. Peters, S. van der Wal, H.A. Claessens, C.A. Cramers, Journal of
Chromatography A 949 (2002) 337. [74] X. Jiang, A. van der Horst, V. Lima, P.J. Schoenmakers, Journal of Chromatography
A 1076 (2005) 51. [75] P. Kilz, Chromatographia 59 (2004) 3. [76] H. Pasch, C. Brinkmann, H. Much, U. Just, Journal of Chromatography A 623 (1992)
315. [77] S.J. Kok, T. Hankemeier, P.J. Schoenmakers, Journal of Chromatography A 1098
(2005) 104. [78] R. Edam, D.M. Meunier, E.P.C. Mes, F.A. Van Damme, P.J. Schoenmakers, Journal
of Chromatography A 1201 (2008) 208. [79] J.-A. Raust, L. Houillot, M. Save, B. Charleux, C. Moire, C.l. Farcet, H. Pasch,
Macromolecules 43 (2010) 8755. [80] A. Barlow, R.S. Lehrle, J.C. Robb, Polymer 2 (1961) 27. [81] M. Hakkarainen, A.-C. Albertsson, S. Karlsson, Polymer Degradation and Stability
52 (1996) 283.
Methods for the chemical analysis of degradable synthetic polymeric biomaterials
31
[82] M. Hakkarainen, S. Karlsson, A.-C. Albertsson, Journal of Applied Polymer Science 76 (2000) 228.
[83] M. Hakkarainen, S. Karlsson, A.C. Albertsson, Polymer 41 (2000) 2331. [84] M. Hakkarainen, Polymer Degradation and Stability 95 (2010) 270. [85] C. Torres, C. Otero, Enzyme and Microbial Technology 29 (2001) 3. [86] K. Urakami, A. Higashi, K. Umemoto, M. Godo, C. Watanabe, K. Hashimoto, Chem.
Pharm. Bull. 49 (2001) 203. [87] P. Plikk, K. Odelius, M. Hakkarainen, A.C. Albertsson, Biomaterials 27 (2006) 5335. [88] R.P. Burford, J.K. Haken, P.A.D.T. Vimalasiri, Journal of Chromatography A 321
(1985) 295. [89] R.R. Mallepally, I. Smirnova, W. Arlt, M. Seiler, S.K. Klee-Laquai, G. Hills, Journal
of Applied Polymer Science 112 (2009) 1873. [90] C. Eldsäter, A.-C. Albertsson, S. Karlsson, International Journal of Polymer Analysis
and Characterization 5 (2000) 415. [91] H. Matsubara, A. Yoshida, Y. Kondo, S. Tsuge, H. Ohtani, Macromolecules 36
(2003) 4750. [92] E.R. Kaal, G. Alkema, M. Kurano, M. Geiler, H.-G. Janssen, Journal of
Chromatography A 1147 (2007) 182. [93] E.R. Kaal, M. Kurano, M. Geiler, H.-G. Janssen, Journal of Chromatography A 1186
(2008) 222. [94] A. Chojnacka, A. Ghaffar, A. Feilden, K. Treacher, H.-G. Janssen, P. Schoenmakers,
Analytica Chimica Acta (2011). [95] L. Li, MALDI mass spectrometry for synthetic polymer analysis, John Wiley &
Sons, Inc. Hoboken, New Jersey., 2010. [96] C.N. McEwen, P.M. Peacock, Analytical Chemistry 74 (2002) 2743. [97] S.M. Weidner, S. Trimpin, Analytical Chemistry 82 (2010) 4811. [98] M.C. McMaster, in, John Wiley & Sons, Inc., Hoboken, New Jersey, 2005. [99] F. Hillenkamp, J. Peter-Katalinić, MALDI MS: A practical guide to instrumentation,
method and applications, WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, 2007.
[100] A. Höglund, M. Hakkarainen, A.-C. Albertsson, Biomacromolecules 11 (2009) 277. [101] A. Höglund, M. Hakkarainen, U. Edlund, A.-C. Albertsson, Langmuir 26 (2010) 378. [102] P. Rizzareli, D. Zampino, L. Ferreri, G. Impallomeni, Anal. Chem. 83 (2011) 654. [103] P. Rizzarelli, C. Puglisi, G. Montaudo, Rapid Communications in Mass Spectrometry
20 (2006) 1683. [104] P. Rizzarelli, C. Puglisi, G. Montaudo, Rapid Communications in Mass Spectrometry
19 (2005) 2407. [105] J. Luo, G. Giguère, X.X. Zhu, Biomacromolecules 10 (2009) 900. [106] S.M. Weidner, J. Falkenhagen, K. Knop, A. Thünemann, Rapid Communications in
Mass Spectrometry 23 (2009) 2768. [107] J. Chen, J.A. Gardella, Macromolecules 32 (1999) 7380. [108] A.M. Belu, D.J. Graham, D.G. Castner, Biomaterials 24 (2003) 3635. [109] T. Gruendling, S. Weidner, J. Falkenhagen, C. Barner-Kowollik, Polymer Chemistry
1 (2010) 599. [110] S. Trimpin, D.E. Clemmer, Analytical Chemistry 80 (2008) 9073. [111] F.A. Bovey, P.A. Mirau, NMR of Polymers, Academic Press, 1996. [112] D.M. LeMaster, G. Hernández, Macromolecules 33 (2000) 3569. [113] H. Uyama, T. Fukuoka, I. Komatsu, T. Watanabe, S. Kobayashi, Biomacromolecules
3 (2002) 318.
Chapter 1
32
[114] M.V. Pergal, V.V. Antić, M.N. Govedarica, D. Goäevac, S. Ostojić, J. Djonlagić, Journal of Applied Polymer Science (2011) n/a.
[115] Y. Pang, Q. Zhu, J. Liu, J. Wu, R. Wang, S. Chen, X. Zhu, D. Yan, W. Huang, B. Zhu, Biomacromolecules 11 (2010) 575.
[116] M.P. Shaver, D.J.A. Cameron, Biomacromolecules 11 (2010) 3673. [117] T.R. Cooper, R.F. Storey, Macromolecules 41 (2008) 655.
A. Ghaffar, P.G. Verschuren, J.A.J. Geenevasen, T. Handels, J. Berard, B. Plum, A.A. Dias, P.J. Schoenmakers, Sj. van der Wal, Journal of Chromatography A, 1218 (2011) 449-458.
Chapter 2
2. Fast in vitro hydrolytic degradation of polyester urethane acrylate
biomaterials: structure elucidation, separation, and quantification of
degradation products
Synthetic biomaterials have evoked extensive interest for applications in the field of health
care. Prior to administration to the body a quantitative study is necessary to evaluate their
composition. An in vitro method was developed for the quick hydrolytic degradation of
poly(2-hydroxyethyl methacrylate) (pHEMA), poly(lactide-co-glycolide50:50)1550-diol
(PLGA(50:50)1550-diol), PLGA(50:50)1550-diol(HEMA)2 and PLGA(50:50)1550-diol(etLDI-
HEMA)2 containing ethyl ester of lysine diisocyanate (etLDI) linkers using a microwave
instrument. Hydrolysis time and temperature were optimized while monitoring the degree of
hydrolysis by 1H NMR spectroscopy. Complete hydrolytic degradation was achieved at
120°C and 3 bar pressure after 24 h. Chemical structure elucidations of the degradation
products were carried out using 1H and 13C NMR spectroscopy. The molecular weight (Mw)
of the polymethacrylic backbone was estimated via size-exclusion chromatography coupled
to refractive index detection (SEC-dRI). A bimodal Mw distribution was found
experimentally, also in the pHEMA starting material. The number average molecular
weights (Mn) of the PLGA-links (PLGA(50:50)1550-diol) were calculated by high pressure
liquid chromatography - time-of-flight mass spectrometry (HPLC-ToF-MS) and 1H NMR.
The amounts of the high and low Mw degradation products were determined by SEC-dRI
and, HPLC-ToF-MS, respectively. The main hydrolysis products poly(methacrylic acid)
(PMAA), ethylene glycol (EG), diethylene glycol (DEG), lactic acid (LA), glycolic acid
(GA), and lysine were recovered almost quantitatively.
The current method leads to the complete hydrolytic degradation of these materials and will
be helpful to study the degradation behavior of these novel cross-linked polymeric
biomaterials.
Chapter 2
34
1 Introduction
Synthetic polymeric biomaterials are of high importance in the medical field due to an aging
population and their potential to improve the quality of life [1]. There is a gradual trend to
replace non-degradable materials with degradable materials mainly because of the need to
avoid reinterventions when complications arise with non-degradable materials [2]. This is
most vividly seen with the move in the stent coating area where stable drug eluting coatings
are being replaced with biodegradable coatings [3]. Such kinds of materials have their
potential use as joint and limb replacements [4], artificial arteries [5], and skin [6], contact
lenses [7], dental implants [8], catheters [9], in tissue engineering [10], and as systems for
controlled delivery of drugs [11] etc. An important class of degradable biomaterials are
chemically cross-linked polymeric networks predominantly based on pHEMA and PLGA
[12,13]. Since its birth in 1936 [14] and first reported application for contact lenses in 1960
[15], pHEMA is one of the most extensively studied polymeric biomaterials in biomedical
applications [16] because of its biocompatibility, hydrophilicity, softness, high water content
and permeability [17], but it has poor mechanical properties [18]. However, numerous
studies reported the modification of the hydroxyl group with poly(ε-caprolactone) (PCL) [3],
poly(3-hydroxybutyrate-co-3-hydroxyvalerate) [19], dextran [11], poly(2-
(dimethylamino)ethyl methacrylate) [20], poly(ethylene oxide) [12],, poly(tetrahydrofurfuryl
methacrylate) [21], poly(ethylene glycol)-methacrylate [22], poly(dimethylsiloxane) [23],
sulfopropyl methacrylate [24], and cross-linker to tune the biomechanical properties of the
pHEMA.
PLGA is an FDA-approved biodegradable and biocompatible polymeric biomaterial [25].
PLGA is widely used as a drug delivery matrix using numerous forms such as microspheres
[26], nanoparticles [27], scaffold [28], microfibers [29], tablets [30], in the field of control
release delivery devices, and tissue engineering. Currently, the focus on synthesis of
copolymers of PLGA with other polymers has been increased such as PLGA-PCL-PLGA
[31], MeO-PEG-PLGA-PEG-OMe [32], PLGA-PEG [33], and PLGA-grafted dextran [34].
Chemical and enzymatic hydrolysis are the primary biodegradation mechanisms for such
materials. Phagocyte-derived oxidants, produced as a result of foreign body response, may
also contribute to the in vivo degradation of aliphatic ether groups in these networks [35].
The suitability of the polymeric biomaterials for medical devices can be inferred from their
chemical structure, the degradation time and the biocompatibility of the polymers and their
degradation products [11]. Swelling ratios (water contents) of the hydrogels [10,12], weight
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
35
loss [10,23], pH of the medium [36], kinetic chain length [37], and so on are the most
common parameters used to assess the in vitro degradation of material. These parameters
may be insensitive in the early stages of degradation and are not very informative on
toxicology. Chromatographic methods that can give more insight into the structure of these
networks and can be used to predict their properties more accurately are desired. However,
networks lack solubility, a prerequisite for such analysis. This requires a very sensitive
method of analysis, or at least an accelerated in vitro chemical hydrolysis of the novel
biomaterials at extreme pH values or high temperature, possibly avoiding the formation of
any insoluble product, followed by the structural analysis and quantification of their
degradation products. The collected information will be helpful not only (i) to ascertain the
composition of the original networks, but also (ii) to evaluate the biocompatibility of these
polymeric networks and their degradation products and (iii) to modify the existing and to
design new biomaterials for specific applications. Recently, H. Matsubara et al. reported a
supercritical methanolysis to achieve the selective decomposition at ester linkages in a UV-
cured acrylic ester resin to characterize the cross-linking structures, but no quantification of
the decomposition products was done to assess the degree of methanolysis [38].
A more detailed second approach to study these prospective biomaterials is a chemical or a
specific enzymatic degradation during physiological conditions, allowing one to study the
kinetics of degradation. Again, specific and sensitive chromatographic methods will be
needed to draw sound conclusions. In particular a method is needed as the second stage in a
two-step procedure and is reported here. First degradation under physiologically relevant
conditions is performed, resulting in partially degraded material of which the constituents
may be identified. Then complete and fast degradation of the products of the first step
(oligomers, intermediates and other products) is executed for quantification.
In the present study polymeric biomaterials based on pHEMA (backbone) and
PLGA(50:50)1550-diol (PLGA-links) were subjected to fast hydrolytic degradation. One
reason to select these samples is that pHEMA, frequently formed as an intermediate
hydrolysis product in polymeric network biomaterials, is only partially hydrolyzed under
physiologically relevant conditions [11] and no detailed study on the complete hydrolytic
degradation and direct analysis of its degradation products has yet been published to our
knowledge.
In this paper first the development and optimization of a method for the microwave-assisted
in vitro hydrolytic degradation is reported of pHEMA, PLGA(50:50)1550-diol and the photo-
crosslinked polymeric biomaterials such as PLGA(50:50)1550(HEMA)2 and
Chapter 2
36
PLGA(50:50)1550-diol(etLDI-HEMA)2. The hydrolysis of polymeric biomaterials and their
model building blocks, pHEMA (backbone) and PLGA(50:50)1550-diol were performed at up
to 120oC, for different periods of time. The hydrolysis time and the temperature were
optimized while monitoring the degree of hydrolysis of the starting material with 1H NMR
spectroscopy. Then the structure elucidations of the degradation products (Figure 1) were
carried out using 1H and 13C NMR spectroscopy and quantification of high Mw hydrolyzed
polymethacrylic acid backbone by SEC-dRI and LA, GA, EG, DEG, and lysine by HPLC-
ToF-MS in the hydrolyzed sample are reported. The Mw distribution of the hydrolyzed
backbone was estimated via SEC-dRI. The Mn of the PLGA-links was measured by HPLC-
ToF-MS and 1H NMR.
2 Experimental
2.1 Materials
DL-lactide and Glycolide were purchased from PURAC (CSM Biochemicals, Gorinchem,
The Netherlands), ethyl ester of lysine diisocyanate from Kyowa Hakko Europe GmbH
(Dusseldorf, Germany), caprolactone from Solvay, methacryloyl chloride via Fluka. Irganox
1035 was obtained from Ciba-Geigy (Basel, Switzerland). pHEMA [Mv = 300 kDa (192,066)
or 20 kDa (529,265), solvent and temperature conditions of Mv (viscosity average MW)
determination are not known] and all other chemicals were purchased from Sigma- Aldrich
(St Louis, MO, USA). The chemicals were used as such unless otherwise stated. In all the
experiments deionized water was used.
The experimental batches of PLGA(50:50)1550-diol, PLGA(50:50)1550-diol(HEMA)2 and
PLGA(50:50)1550-diol(etLDI-HEMA)2 were synthesized at DSM Biomedical, Geleen,
Netherlands, according to the following procedure:
Preparation of PLGA(50:50)1550-diol : DL-Lactide (51.6 g, 0.358 mol), glycolide (41.5 g,
0.358 mol) and diethyleneglycol (6.85 g, 6.45 mmol) were weighed in the glovebox and
melted at 150oC under nitrogen conditions. 1 mL of a stock solution Tin(II)-ethylhexanoate
(290 mg in 10 mL n-hexane) was added as a catalyst. The reaction was allowed to proceed
for 18 h upon which the reaction mixture was cooled to room temperature to obtain
poly(lactide-co-glycolide50:50)1550-diol [39].
Preparation of PLGA(50:50)1550-diol(HEMA)2: poly(lactide-co-glycolide50:50)1550-diol (100
g, 65 mmol), 200 mg Irganox 1035 and triethylamine (13.05 g, 0.129 mol) were dissolved in
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
37
150 ml dry tetrahydrofuran (THF). Methacryloylchloride (13.49 g, 0.129 mol) was added
drop wise to the solution at controlled temperature (<5°C). Immediately a white precipitate
was visible (triethylamine.HCl salt). The dropping funnel was rinsed with THF (50 ml). The
reaction mixture was stirred at room temperature for 18 h. The reaction mixture cooled till
5°C and filtered to remove the triethylamine.HCl salt. The THF was removed via
evaporation with a rotavapor. The remainder was dissolved in 200 ml ethyl acetate. The clear
solution was extracted once with 300 ml 0.1 HCl solution, once with 300 ml 5% NaCl-
solution and 300 mL water. The resulting solution was dried with Na3SO4 and evaporated to
dryness. Poly(lactide-co-glycolide50:50)1550-dimethacrylate was obtained as a slightly
coloured yellow oil. 30.49 g poly(lactide-co-Glycolide50:50)1550-dimethacrylate, 13.1 g
HEMA and 0.86 g Darocur 1173 were mixed in a clear formulation [39].
Micro-particles preparation of PLGA(50:50)1550-diol(HEMA)2: 10.52 g of this formulation
was mixed with 39.88 g PEG 35k (40% m/m in water), 30 g water and 5 g aceton. This
mixture was stirred mechanically for 10 min at 800 rpm before polymerization. The
polymerization was allowed to proceed for 60 min under UV light (Macam Flexicure
controller, D-bulb, 200 mW/s/cm2, Livingston, U.K.). After polymerization, the micro-
particles were filtered through a 0.8 µm filter (Supor-800, Gelman Sciences, Ann Arbor, MI,
USA) under vacuum and rinsed with 250 ml water. The morphology was checked with light
microscopy. The methacrylate conversion was >96%. The micro-particles were sieved
afterwards using ethanol as a solvent (Retsch sieves, aperture 63, 125, and 250 µm, Haan,
Germany). The micro-particles were dried via freeze drying [40].
Preparation of PLGA(50:50)1550-diol(etLDI-HEMA)2: Hydroxymethylacrylate (HEMA, 26
g, 0.20 mol) was added drop wise to a solution of the ethyl ester of Lysine diisocyanoate
(etLDI) (45.25 g, 0.2 mol), Tin-(II)-ethylhexanoate (0.080 g, 0.186 mmol), Irganox 1035
(0.260 g) and dry air at controlled temperature (<5°C). Subsequently the reaction mixture
was stirred overnight at 40°C. The etLDI-HEMA was obtained as a slightly yellow oil. The
reaction was monitored with gel-permeation chromatograpgy (GPC). Poly(lactide-co-
glycolide50:50)1550-diol (100 g, 0.064 mmol) was dissolved in 150 ml dry THF. etLDI-
HEMA (46.05 g, 0.129 mol) was added to the reaction mixture at room temperature.
Subsequently the reaction mixture was stirred overnight at 40°C. In the morning the reaction
mixture was analysed with IR (no NCO peak ν = 2260 cm-1 visible). The reaction was
complete, based on IR spectroscopy when all the THF was evaporated. The poly(lactide-co-
glycolide50:50)1550-(etLDI-HEMA)2 was obtained as a yellowish oil [40].
Chapter 2
38
1M
KO
H
2
*
*
OO
H
CH
31
n
xy
5
OH
5
OH
OH
9
OH
O
CH
38
NH
2
15
14
13
12
11
OH
O
NH
2
OH
10
OH
O
OH
6
7
O
7
6
OH
PM
AA
EG
DE
GLA
GA
Lysine
++
++
+
OO
OO
OO
O
O
O
O
O
NH
OO
CH
3
O
NH
O
O
NH
OO
CH
3
O
NH
O
O
CH
3C
H3
O
O**
CH
3
O
O**
CH
3
xy
25
24
O
24
25
OO
21
O23
O23
O21
O
O
O
O
OC
H3
20
CH
320
22
O19
OH
O
O
CH
318
19
O22
OH
O
OC
H3
18
xy
OO
OO
OO
O
O
O
O
OC
H3
CH
3
O
O**
CH
3
O
O**
CH
3
OH
O
3
4
O
**
CH
3
( a)
( b)
( c)
( d)
+O
H
16
CH
3+
CO
2
1M
KO
H
2
*
*
OO
H
CH
31
n
xy
5
OH
5
OH
OH
9
OH
O
CH
38
NH
2
15
14
13
12
11
OH
O
NH
2
OH
10
OH
O
OH
6
7
O
7
6
OH
PM
AA
EG
DE
GLA
GA
Lysine
++
++
+
OO
OO
OO
O
O
O
O
O
NH
OO
CH
3
O
NH
O
O
NH
OO
CH
3
O
NH
O
O
CH
3C
H3
O
O**
CH
3
O
O**
CH
3
1M
KO
H
2
*
*
OO
H
CH
31
n
xy
5
OH
5
OH
OH
9
OH
O
CH
38
NH
2
15
14
13
12
11
OH
O
NH
2
OH
10
OH
O
OH
6
7
O
7
6
OH
PM
AA
EG
DE
GLA
GA
Lysine
++
++
+
OO
OO
OO
O
O
O
O
O
NH
OO
CH
3
O
NH
O
O
NH
OO
CH
3
O
NH
O
O
CH
3C
H3
O
O**
CH
3
O
O**
CH
3
xy
25
24
O
24
25
OO
21
O23
O23
O21
O
O
O
O
OC
H3
20
CH
320
22
O19
OH
O
O
CH
318
19
O22
OH
O
OC
H3
18
xy
25
24
O
24
25
OO
21
O23
O23
O21
O
O
O
O
OC
H3
20
CH
320
22
O19
OH
O
O
CH
318
19
O22
OH
O
OC
H3
18
xy
OO
OO
OO
O
O
O
O
OC
H3
CH
3
O
O**
CH
3
O
O**
CH
3
OH
O
3
4
O
**
CH
3
( a)
( b)
( c)
( d)
+O
H
16
CH
3+
CO
2
Figure
1. Proposed reaction schem
e for the hydrolytic degradation of (a) pHE
MA
(b) PL
GA
(50:50)1550 -diol (c) PL
GA
(50:50)1550 -diol(HE
MA
)2 . and (d) P
LG
A(50:50)1550 -diol(etL
DI-H
EM
A)2 . P
MA
A represents poly(m
ethacrylicacid); E
G, ethylene glycol; D
EG
, diethyleneglycol; L
A,
lactic acid and GA
, glycolic acid. The num
bering corresponds to NM
R peak assignm
ents in figure 6.
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
39
Micro-particles preparation of PLGA(50:50)1550-diol(etLDI-HEMA)2: 15.58 g PLGA1550-
diol(etLDI-HEMA)2 and 0.31 g Darocure 1173 were mixed together mechanically at 100
rpm in a 250 ml beaker at 50°C, then 62 g PEG35K (40% m/m in deionized water) and 58 g
deionized water were added. This was stirred mechanically for 30 min at 900 rpm. The
polymerization was allowed to proceed for 60 min at 70°C and 900 rpm under UV light
(Macam Flexicure controller, D-bulb, 200 mW/s cm-2). The particles were wet-sieved with
deionized water over a sieving tower (Retsch test sieve Aperture 250, 125, 63, and 45 µm)
and dried under vacuum at room temperature for 18 h. Afterwards methacrylate conversion
was checked: >98% (FT-IR, 1640 cm-1 and 815 cm-1) [40].
pressure sensor
sensortemperature
microwaveenergy
coolingmedium
PTFElining
pressure sensor
sensortemperature
microwaveenergy
coolingmedium
PTFElining
Figure 2 Schematic diagram of CEM Discover microwave apparatus used in this work, with additional PTFE lining of 1 mm thickness.
2.2 Procedure of hydrolysis
20 or 40 mg of each sample was dissolved in 2 mL of 1 M KOH (Merck, Darmstadt,
Germany) in a 10 mL pressurized glass vial (CEM Corporation, NC, USA) using a magnetic
stirrer. The 10 mL pressurized glass vial (i.d. = 12 mm) was internally lined with a PTFE
Chapter 2
40
tube of 1 mm thickness and i.d. = 11 mm (locally made at the mechanical workshop of the
University of Amsterdam, Figure 2). The homogeneous mixture in the glass vessel was
placed in the microwave instrument (Discover BenchMate, CEM) and hydrolysis to PMAA
and EG was carried out at 120oC, 3 bar and for 24, 20, 15, 10, and 5 h. Similarly,
PLGA(50:50)1550-diol , PLGA(50:50)1550-diol(HEMA)2 and PLGA(50:50)1550-diol(etLDI-
HEMA)2 were hydrolyzed at 120oC for 24 h at 3 bar. The mixture was weighed before and
after each hydrolysis.
2.3 1H NMR spectroscopy of hydrolysate
0.5 mL of each hydrolysis solution was acidified by adding carefully a few drops of 37%
HCl with vigorous stirring at 90ºC. The PMAA precipitates and along with supernatant
(containing ethylene glycol (EG), diethylene glycol (DEG), lactic acid (LA), glycolic acid
(GA), lysine etc.) were dried overnight at 40oC with an air flush.
The dried mixtures of the hydrolysates were re-dissolved in d4-methanol (Euriso-top,
France). Samples of un-hydrolyzed pHEMA and PLGA(50:50)1550-diol were also prepared
in d4-methanol. 1H NMR spectra were recorded on a Varian Inova 500 MHz NMR (Varian
Inc., USA) equipped with Probe: 500 5 mm 13C/31P/1H GS. Pulse repetition time: 25 sec,
Pulse: 3.6 µsec, Scans: 63 and temperature: 25oC were used to record 1H NMR spectra.
2.4 Size-exclusion chromatography (SEC) analysis
pH neutralized ( 0.2 mL) hydrolysis solutions were diluted with 0.2 mL aqueous SEC mobile
phase. The SEC experiments were performed on an HPLC system equipped with in-line
degasser, Model 600 pump, 717 plus TRI-SEC auto-sampler and Model 410 differential
refractive index detector (all Waters, Milford, MA, USA). Data were recorded and
chromatographic peaks were treated using Empower 2 software (Waters, Milford, MA,
USA). Calculations for molar mass distribution (MMD) on the chromatographic peaks were
executed using software written in-house in Excel 2003 (Microsoft).
All aqueous SEC separations were performed on the following set of columns used in series:
PL Aquagel-OH Guard (8 µm, 50 mm × 7.5mm i.d.), PL Aquagel-OH 50, 30, and 10 (each 8
µm, 300 mm × 7.5 mm i.d.) columns (Polymer Laboratories, U.K.). For 20 kDa pHEMA
hydrolysates, same set of columns was used except PL Aquagel-OH 50. The mobile phase
was (0.2 M NaNO3, 0.01 M NaH2PO4, pH ≈ 7) pumped at a flow rate of 1 mL min-1.
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
41
Poly(methacrylic acid) sodium salt (PMA-Na) standards (Table 1) were used to calibrate the
SEC-dRI system. The calibration curves for MMD of PMAA in (300 and 20 kDa) pHEMA
hydrolysates are given by cubic relations of logM and retention time, x: log(M) = –
0.00915x3 + 0.51649x2 – 9.98592x + 70.73912, R2 = 0.999 and log(M) = –0.032x3 + 1.218x2
– 15.62x + 73.11, R2 = 0.998, respectively. To quantify the concentration of hydrolyzed
backbone as PMA-Na in hydrolysates the calibration lines were recorded using PMA-Na
standards with Mp 65.8 kDa (at six concentrations 0.2–2 mg mL-1) and 22.5 kDa (at five
concentrations 1–5 mg mL-1). Highly pure water for mobile phase preparation was obtained
by means of an Arium® 611 Ultrapure (18.2 MΩ*cm) Water System (Sartorius AG,
Goettingen, Germany).
Table 1 Peak molecular weight (Mp), weight average molecular weight (Mw), number average molecular (Mn) and dispersity (PDI) of the Poly (methacrylic acid) sodium salt standards. Data as specified by the supplier.
Standard Mp (D) Mw (D) Mn (D) PDI Supplier
PMA-Na-1 1220 1250 1040 1.197 PSS
PMA-Na-2 1670 1700 1520 1.120 PSS
PMA-Na-3 3180 3150 2700 1.169 PSS
PMA-Na-4 7830 7750 7220 1.073 Fluka
PMA-Na-5 8210 8280 7480 1.108 PSS
PMA-Na-6 22,500 22,100 21,100 1.047 PSS
PMA-Na-7 31,500 31,100 30,400 1.023 Fluka
PMA-Na-8 65,800 62,500 60,600 1.031 PSS
PMA-Na-9 78,300 75,100 73,300 1.025 Fluka
PMA-Na-10 201,000 192,000 186,000 1.029 PSS
PMA-Na-11 480,000 421,000 380,000 1.108 PSS
PMA-Na-12 549,000 483,000 429,000 1.126 Fluka
Size-exclusion chromatography of pHEMA (300 and 20 kDa) was performed on two PL gel
MIXED-C (5 µm, 300 mm × 7.5 mm i.d.) columns with DMF (Acros Organics, NJ, USA)
containing 0.02 M lithium chloride (Acros Organics) as a mobile phase pumped at a flow
rate of 1 mL min-1 via an LC-10AD solvent delivery module coupled with a RID-10A dRI
detector (Shimadzu Corporation, Kyoto, Japan). A Rheodyne 7120 manual injector
(Rheodyne Europe GmbH, Alsbach, Germany) with 20 µL loop was used as an injection
system.
The resolving power of a SEC system can be visualized by an integrity plot, which gives the
integrity index (IIsec) as a function of sample Mw and (Mw/Mn-1) [41]. (IIsec) indicates the
fraction of dispersion of the experimental peak variance that is caused by the polydispersity
of the sample itself and not by dispersion due to the column or extra-column band
Chapter 2
42
broadening [42]. The integrity plot for the used SEC system was constructed using the
polymer standards listed in Table 1 and clearly demonstrates its suitability even for narrowly
distributed polymers in the range of 2–200 kDa (cf. Figure 3).
Figure 3 Experimental SEC-integrity plot as a function of the sample (horizontal axis;Mw/Mn-1) proportional to log(PDI-1) and molecular weight (vertical axis; log M)). System: PL aquagel-OH Guard (8 µm, 50 mm × 7.5 mm i.d.), PL aquagel-OH 50, 30 and 10 ( each 8 µm, 300 mm × 7.5 mm i.d.) columns; mobile phase: 0.2 M NaNO3, 0.01 M NaH2PO4, pH ≈ 7) pumped at a flow rate of 1 mL min-1.
2.5 HPLC-ESI-ToF-MS analysis of hydrolysate
0.1 mL of the hydrolysate was mixed with 4 mL of deionized water and then pH neutralized.
The mixture was filtered with a 0.2 µm pore size PTFE filter (Grace Davison discovery
science, IL, USA), prior to injection. Stock solutions of LA (Fluka), GA (Fluka), EG
(Aldrich), DEG (Fluka) and D-lysine (Sigma) were prepared by dissolving in 1 M KOH
solution and quantification was done with a standard addition method in order to correct for
signal suppression of target analytes by co-eluting compounds. The chromatographic
separations were performed on a Prevail C18 column (250 × 4.6 mm i.d., 5 µm particle size,
Alltech Discovery Sciences, IL, USA) at a temperature of 35ºC. The injection system
consisted of a Rheodyne 7010 manual injector (with 5 or 20 µL loops). The aqueous mobile
phase containing 0.1% (v/v) formic acid (Fluka), 0.03% (w/v) sodium iodide (NaI) Aldrich)
and 1% (v/v) acetonitrile (Biosolve) was pumped via Shimadzu LC-20AD solvent delivery
module at 2 mL min-1 and was split between a waste reservoir and electrospray ionization
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
43
(ESI) interface by means of a zero dead volume T-piece to enssure a flow of approximately
0.2 mL min-1 into the ESI interface.
In order to evaluate the recovery of the procedure of the method about 10 mg each of PMA-
Na standard (Mp = 78.3 kDa), DEG, EG and D-lysine and approximately 50 mg of LA and
GA standards were dissolved in 5 mL (5.2 g) of 1 M KOH to make control solutions. 2 mL
of the mixture was heated at 120oC and 3 bar in the microwave for 24 h. The concentration
of each analyte was determined before and after heating to calculate the percentage recovery
of the method for each analyte.
The LC system was hyphenated with an Agilent 6210 series ToF-MS (Agilent Technologies,
Waldbronn, Germany) via an ESI interface. The conditions of the ESI-ToF-MS were as
follows: drying gas was nitrogen (N2) at 8 L min-1; and at 300oC; 30 psig of N2 ; capillary
voltage, 3500 V; fragmenter, 140 V; skimmer voltage, 60 V; octopole dc1, 33 V; octopole
radio frequency, 250 V. The data were acquired in the scan mode from m/z 50 to 500 D with
0.88 scans/sec. An Agilent MassHunter Workstation A.02.01 and AnalystTM QS 1.1 software
(Applied Biosystems) were used for data acquisition and data analysis, respectively.
3 Results and discussion
3.1 Optimization of hydrolysis method
Initially the hydrolysis method was optimized by degrading the pHEMA (300 and 20 kDa)
in a 10 mL pressurized glass vial specially designed for the CEM microwave instrument.
After hydrolysis the glass vessel contained the hydrolysis solution and white material, stuck
on the wall of the glass vessel. We were not able to dissolve this material, for further
analysis by NMR, SEC or HPLC, except at very low pH. The residues were originally
considered to be silicates. For further analysis, these residues were washed three times with 5
mL of water, methanol and DMF to wash out possible impurities of hydrolyzed and un-
hydrolyzed pHEMA and dried overnight at 210oC in an oven. The XRF (X-ray fluorescence)
spectrum (Eagle-III Spectrometer, EDAX Inc., Mahwah, NJ, USA) of these residues
confirmed the presence of silicates primarily originating from the glass vessel in alkaline
conditions at high temperature (Figure 4). However, the CHN elemental analysis (Truspec,
Leco, Germany) also revealed the presence of carbon contents in this material. Based on the
percentage of these carbon contents, it can be concluded that up to 35% of the starting
material (pHEMA) is lost by inclusion in the white residue from the hydrolysis solution. It
Chapter 2
44
may be assumed that at high temperature the highly reactive silanol groups present on
silicates react with hydroxyl groups of pHEMA [43].
The formation of white residues during hydrolysis will lead to wrong quantification of the
relative percentage of starting material in the hydrolysis solution, so to avoid contact of the
alkaline solution with the inner surface of the glass vessel (to prevent the formation of white
residues) it was internally lined with PTFE (Figure 2). The hydrolysis was performed
repeatedly after this modification and no formation of white residues was observed. The
hydrolysis time was optimized while monitoring the cleavage of ester linkages in pHEMA
with 1H NMR spectroscopy. Then the hydrolysis of PLGA(50:50)1550-diol,
PLGA(50:50)1550-diol(HEMA)2, and PLGA(50:50)1550-diol(etLDI-HEMA)2 was conducted
at 120oC for 24 h.
Figure 4 XRF spectrum of white residues without lining the pressurized glass vessel with PTFE.
3.2 Product identification
The overlay of 1H NMR spectra of pHEMA hydrolyzed for different times (Figure 5) show
clearly the cleavage of ester groups of pHEMA i.e. the scission of side chains from the
backbone chain and the formation of free ethylene glycol (peak 5 at δ 3.6 ppm) and a small
quantity of diethylene glycol (peak 7 at δ 3.56 ppm and peak 6 at δ 3.68 ppm). The peaks of
free ethylene glycol and diethylene glycol were confirmed by taking the 1H NMR spectrum
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
45
of sample spiked with EG and DEG standards in d4-methanol. The signals at δ 3.75 ppm
(peak 4) and δ 4.21 ppm (peak 3) corresponding to the side chain of pHEMA almost
vanished after 24 h, indicating the degree of hydrolysis of pHEMA. In the starting material,
before hydrolysis the positions of peaks 3 and 4 were observed at δ 4.08 and δ 3.80 ppm,
respectively. DEG observed in the 1H NMR spectra is present as an impurity in pHEMA.
The 1H-1H gCOSY (two-dimenssional homonuclear H, H gradient-correlated spectroscopy)-
correlated NMR experiment indicates the proton connectivity between signals at δ 3.80 and δ
4.08 ppm in the starting material and at δ 3.75 and δ 4.21 ppm (peaks 3 and 4) for the
hydrolysate affirming the presence of the side chain in both the starting material and the
hydrolysate. Also, the proton connectivity in the signals (peaks 6 and 7) of diethylene glycol
was confirmed. The 13C NMR spectra in DEPT135 mode were recorded to assign the
methyl, methane and methylene group and quaternary carbon in both the starting material
and the hydrolyzed products. The single bond connectivities between 1H and 13C were also
determined by the two-dimensional 13C, 1H-correlated HSQC (heteronuclear single quantum
coherence) NMR experiments.
*
CH3
HO
OCH2
*
HO
H2C
CH2
OH
*
CH3
O
CH2
OCH2
*
H2C
OH
+
n
n
H2C
CH2
OHHOCH2
H2C
O
+
1 2
7
5
6
3
1
2
7
5
6
4
2
3 76
1
5
4
ab
ced
*
CH3
HO
OCH2
*
HO
H2C
CH2
OH
*
CH3
O
CH2
OCH2
*
H2C
OH
+
n
n
H2C
CH2
OHHOCH2
H2C
O
+
1 2
7
5
6
3
1
2
7
5
6
4
2
3 76
1
5
4
ab
ced
Figure 5 An overlay of 1H NMR spectra (CD3OD, 25 oC, 500 MHz) of pHEMA (300 kDa) hydrolyzed after different times of hydrolytic degradation (a) 5 h (b) 10 h (c) 15 h (d) 20 h (e) 24 h at 120 oC in the microwave instrument.
The relative percentage of non-hydrolyzed pHEMA in the hydrolysis solution of pHEMA
(300 kDa) and pHEMA (20 kDa) was determined using the following equation.
Chapter 2
46
100 4
5pHEMA of % Relative
0.20.1
3.47.3
A
A (1)
In which Ax is the peak area for the response at the shift of x ppm.
The relative percentage of un-hydrolyzed pHEMA in the hydrolysate of 300 kDa pHEMA
decreases from 6% at 5 h to less than 0.2% at 24 h. pHEMA with Mv = 20 kDa degraded
much faster .
The 1H NMR spectra of PLGA(50:50)1550-diol before (Figure 6A) and after hydrolysis
(Figure 6B) show the cleavage of ester bonds in PLGA(50:50)1550-diol chains and the
formation of free LA (CH3 = 1.33 ppm and CH = 4 ppm), GA (CH2 = 3.88 ppm) and DEG
(HO−CH2− = 3.7 ppm and −O−CH2− = 3.58 ppm). More acidic hydroxyl groups are formed
after hydrolysis and this leads to more hydrogen bonding. Consequently, the hydroxyl group
signal shift towards higher frequency after hydrolysis. The multiple signals for the
methylene groups (number 22) indicate sequential or tacticity effects (starting material is
DL-lactide). Unfortunately, the signal at 1.15 ppm is an isopropanol impurity (an
experimental artifact).
The NMR of PLGA(50:50)1550-diol(HEMA)2 and PLGA(50:50)1550-diol(etLDI-HEMA)2
polymeric biomaterials were not possible because of their lack of solubility in any of the
available NMR solvents. Upon hydrolysis it was possible to dissolve the acidified and non-
acidified degraded backbone (PMAA/PMA-NA) in CD3OD and D2O, respectively.
PLGA(50:50)1550-diol(HEMA)2 is a 3D ladder like polymer network consisting of backbones
of pHEMA interconnected via cross-links of PLGA(50:50)1550-diol. If all the ester bonds are
cleaved, the polymer network should result in polymethacrylate backbone, EG, DEG, LA,
and GA. The 1H NMR of hydrolyzed PLGA(50:50)1550-diol(HEMA)2 (Figure 6C) shows that
the CH3− signal (peak 1) that belongs to the poly methacylate backone is overlapped by the
CH3− signal of LA (peak 8). This was confirmed by 1H-1H gCOSY NMR experiments. The
peak 2 belongs to the −CH2− group of the backbone. Rest of the signals corresponds to the
hydrolyzed PLGA(50:50)1550-diol 1H NMR spectrum (Figure 6B). The peak that appears at
2.72 ppm is unknown.
The PLGA(50:50)1550-diol(etLDI-HEMA)2 is similar in chemical structure to
PLGA(50:50)1550-diol(HEMA)2 except that the backbones of pHEMA and PLGA-links
(PLGA(50:50)1550-diol) are interconnected via ethyl ester of lysine diisocyanate (etLDI)
linkers. So upon complete hydrolytic degradation all the products should be the same as in
case of PLGA1550-diol(HEMA)2 except for lysine and ethanol. The signals 11, 12, 13, 14,
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
47
and 15 (Figure 6D) belong to lysine (Figure 1). The signal bond connectivity between lysine
signals were established by H-H gCOSY. The peak 16, a triplet, belongs to a −CH2− group
of ethanol. Peak 17 belongs to PEG present as an impurity in the biomaterial. Mn and the
molar ratios of LA and GA in the PLGA(50:50)1550-diol were calculated by integrating the
CH group of LA, −CH2− group of GA and -CH2- groups of DEG (Table 2).
8 25
11
10
6 7
9
-OH
8 2510
6 7
9
12
-OH
-OH
810
6 79
13
14
15 16
-OH CD3OD
CD3OD
CD3OD
CD3OD
17
6 7
A
B
C
D
18 20
19
21
22
23 24
25
?
1
1
8 25
11
10
6 7
9
-OH
8 2510
6 7
9
12
-OH
-OH
810
6 79
13
14
15 16
-OH CD3OD
CD3OD
CD3OD
CD3OD
17
6 7
A
B
C
D
18 20
19
21
22
23 24
25
?
1
1
Figure 6 The 1H NMR spectrum of (A) PLGA(50:50)1550-diol before hydrolysis and (B) PLGA(50:50)1550-diol (C) PLGA(50:50)1550-diol(HEMA)2 (D) PLGA(50:50)1550-diol(etLDI-HEMA)2 after hydrolysis at 120oC for 24 h. The numbering of peaks corresponds to the numbering in Figure 1.
HPLC-ToF-MS is much more sensitive than NMR, but it is an indirect method:
determination of LA and GA occurs after hydrolysis. Still, the Mn of the PLGA-links
measured by HPLC-ToF-MS is considered to be more accurate, because in NMR accurate
integration of the peak area is difficult as the signals of EG and DEG slightly overlap. This
results in apparently higher Mn values by NMR except in case of PLGA(50:50)1550-diol links
(Table 2).
Chapter 2
48
Table 2 Mn of PLGA-links (PLGA(50:50)1550-diol) and molar ratios of LA and GA calculated by 1H NMR and HPLC-ToF-MS.
1H NMR HPLC-ToF-MS Sample
Mn
(Da)
% RSD
(n=3)
Molar ratio
(LA:GA)
Mn
(Da)
% RSD
(n=3)
Molar ratio
(LA:GA)
PLGA(50:50)1550-diol 1317 5 51 : 49 1371 2 49 : 51
PLGA(50:50)1550-diol(HEMA)2 1549 2 50 : 50 1001 3 48 : 52
PLGA(50:50)1550-diol(etLDI-HEMA)2 1791 2 50 : 50 1028 4 49 : 51
3.3 Molar mass characterization and quantification of PMAA
Figure 7 is showing the aqueous SEC separation for the hydrolyzed backbone (in the form of
PMA-Na) for the samples subjected to hydrolysis: pHEMA 20 kDa (a), pHEMA 300 kDa
(b), PLGA(50:50)1550-diol(HEMA)2 (c) and PLGA(50:50)1550-diol(etLDI-HEMA)2 (d). All
the peaks show a non-Gaussian distribution. One potential cause of this bimodal MMD is
that two chains are cross-linked due to the esterification reaction between alcohol in one
chain and carboxylic acid in the other chain. There is no evidence of cross-linking found in 1H NMR spectra, so the extent of cross-linking is very small and the signals may not appear
in the spectra. Zainnuddin et al. suggested that the presence of ions in the solute can promote
the formation of physical crosslinks between two neighboring hydrophilic and hydrophobic
groups of the polymer chains [44]. However, it is unlikely that these physical cross-links
stay intact when subjecting the sample to SEC.
In order to see the effect of hydrolysis on bimodality of the hydrolysate product, PMA-Na
standard with Mp 549 kDa, which showed a uni-modal MMD without hydrolysis, was
hydrolyzed in 1 M KOH with and without EG at 120oC and 3 bar for 24 h. The hydrolysates
were subjected to SEC and no influence of the hydrolysis was observed on the uni-modality
of the MMD. The bimodal distribution of PMA-Na was independent of injection volume.
pHEMA with low a Mw is soluble in water but as the Mw increases, its solubility in water
decreases [45].
Therefore, to further investigate the origin of bimodality in the hydrolyzed backbone, SEC in
DMF of pHEMA (300 and 20 kDa) was performed. The MMD of 300 kDa pHEMA renders
a non-Gaussian behaviour (Figure 8), so the bimodal distribution appears already present in
the starting material.
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
49
Table 3 SEC-dRI data for the hydrolyzed backbone (PMA-Na)
Sample Mn
(kDa)
% RSD
(n=3)
Mw
(kDa)
% RSD
(n=3)
PDI
% RSD
(n=3)
300 kDa 52 8 108 6 2.1 7
20 kDa - - 16 1 - -
PLGA(50:50)1550-diol(HEMA)2 46 4 223 3 4.7 4
PLGA(50:50)1550-diol(etLDI-HEMA)2 34 8 152 6 4.5 9
The molecular weight data are PMA-Na-relative.
-0.2
0.2
0.6
1
1.4
1.8
0 4 8 12 16 20 24 28 32 36 40
Ret. volume (mL)
dRI
sign
als
c
d
b
-0.2
0.2
0.6
1
1.4
1.8
0 4 8 12 16 20 24 28 32 36 40
Ret. volume (mL)
dRI
sign
als
a
-0.2
0.2
0.6
1
1.4
1.8
0 4 8 12 16 20 24 28 32 36 40
Ret. volume (mL)
dRI
sign
als
c
d
b
-0.2
0.2
0.6
1
1.4
1.8
0 4 8 12 16 20 24 28 32 36 40
Ret. volume (mL)
dRI
sign
als
a
-0.2
0.2
0.6
1
1.4
1.8
0 4 8 12 16 20 24 28 32 36 40
Ret. volume (mL)
dRI
sign
als
a
Figure 7 SEC-dRI chromatograms of (a) pHEMA 20 kDa, (b) pHEMA 300 kDa, (c) PLGA(50:50)1550-diol(HEMA)2, (d) PLGA(50:50)1550-diol(etLDI-HEMA)2 after hydrolysis for 24 h at 120oC and 3 bar pressure.
In 20 kDa pHEMA hydrolysate, the PMAA elutes from 11 mL to 16.2 mL and shows a
small shoulder on the low molecular weight side. Figure 7a indicates that very low molecular
Chapter 2
50
weight (< 1000 Da) PMAA is eluting together with the high concentration of salt. (This
could be solved by application of columns with a resolving range at low molecular weights).
Therefore, the Mn value for 20 kDa pHEMA hydrolysate is not reported in Table 3. The Mn,
Mw and PDI of PMA-Na in the samples are listed in Table 3. The hydrolyzed backbones in
case of PLGA(50:50)1550-diol(HEMA)2 and PLGA(50:50)1550-diol(etLDI-HEMA)2 are more
polydisperse than those of the pHEMA (20 and 300 kDa) standards. The monomeric
products of hydrolysis such as LA, GA, EG, DEG, and lysine elute after the permeation
limit. The quantitative results of hydrolyzed backbone in different biomaterials are tabulated
in Table 5.
-0.05
0.05
0.15
0.25
0.35
0.45
0 5 10 15 20 25
Retentional volume (mL)
dRI
sign
als
a
b
-0.05
0.05
0.15
0.25
0.35
0.45
0 5 10 15 20 25
Retentional volume (mL)
dRI
sign
als
a
b
Figure 8 SEC-dRI chromatograms of starting materials (a) 300 kDa, (b) 20 kDa pHEMA.
3.4 Quantification of monomeric products by HPLC-ToF-MS
The low molecular weight products were analyzed by HPLC-ESI-ToF-MS. The TIC
chromatogram of the PLGA(50:50)1550-diol(etLDI-HEMA)2 hydrolysate is shown in Figure
9. The peaks at 1.59, 2.26, and 4.15 min correspond to lysine, EG, and DEG, respectively.
The peak at 1.68 min is the aggregate of sodium formate clusters. In case of pHEMA (300 or
20 kDa) EG is the major product of hydrolysis as compared to DEG and TEG (triethylene
glycol), present as an impurity in the starting material. TEG elutes at 8.40 min. The peak
intensities of EG are much lower than that of DEG and TEG (Table 4), because alcohols are
not easily charged either in positive or negative mode while DEG can be charged easily due
to the presence of its ether group. The three diols form proton [M+H]+, ammonium
[M+NH4]+, sodium [M+Na]+, and potassium [M+K]+ adducts. It was attempted to promote
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
51
the formation of protonated molecule or the ammonium and sodium adducts by adding 0.1%
(v/v) formic acid, 0.02% (v/v) ammonia solution (25%) or 0.03% (w/v) NaI to the mobile
phase. The peak intensity increased with NaI addition (cf. Table 4). Therefore, NaI was
selected to make sodium adducts for quantification. Lysine preferably makes [M+H]+ adduct
even in the presence of NaI. Lysine remains un-retained even at very low concentration of
organic modifier. GA and LA in PLGA(50:50)1550-diol, PLGA(50:50)1550-diol(HEMA)2 and
PLGA(50:50)1550-diol(etLDI-HEMA)2 were quantified in negative ESI mode as [M–H] ¯ ion
without NaI in the mobile phase and are eluted at 2.19 and 3.06 min (Figure 10). The
standards prepared in water showed peaks of linear dimer (m/z 161.0432) of LA at 7 and 8
min. To avoid this, all the solutions of standards were first prepared in 1 M KOH to convert
the dimers into monomers and then diluted up to a dilution factor of 20 with deionized water
and pH neutralized with hydrochloric acid.
Table 4 Relative peak intensities of EG, DEG and TEG in pHEMA hydrolysate with different combinations of formic acid (FA), ammonia (NH3) and sodium iodide (NaI) in the aqueous mobile phase for HPLC-ToF-MS containing 1% organic modifier
Intensity (Mcps) Mobile phase
containing
Selected ion
EG DEG TEG
0.1% FA [M+H]+ 0.06 2.60 0.43
0.1% FA + 0.02% NH3 [M+NH4]+ 0.08 1.40 0.40
0.1% FA + 0.03% NaI [M+Na]+ 0.51 4.00 0.56
When quantifying target components in samples one has to take care to avoid matrix
overloading. The undetected co-eluting matrix components may reduce the ionization
efficiency of the analytes and cause poor reproducibility and accuracy [46,47]. As the lysine
elutes at t0, significant signal suppression occurred and the apparent yield (relative to the
assumed structure of the starting materials as in Figure 1) was 25%. To compensate for this
matrix signal suppression, the quantification of components was performed by standard
addition giving a yield of lysine up to 68%. This indicates that either not all of the etLDI is
converted to lysine, may be due to the presence of lysine-diacrylate cross-links (Figure 11B)
or some of the cross-link chains are deficient with etLDI. The summary in Table 5 shows
that all the hydrolyzed samples were recovered quantitatively with respect to the total
amount of sample subjected to hydrolysis, except for pHEMA (20 kDa), which can be
explained by incomplete separation (cf. Figure 7a) and for PLGA(50:50)1550-diol(etLDI-
Chapter 2
52
HEMA)2, because PEG impurities, ethanol and carbon dioxide, which were produced during
the hydrolysis of the last material were not quantified.
In addition to overall recovery, the recovery of separate hydrolysis products when treating a
mixture of the hydrolysis products by the same hydrolysis and analysis procedure was
determined. These recoveries are also indicated in Table 5. Although the recovery of GA is
significantly less than 100%, the yields are acceptable, considering the small sample size.
Figure 9 TIC and XIC chromatogram of PLGA(50:50)1550-diol(etLDI-HEMA)2 hydrolysate. Conditions: positive ESI mode with isocratic elution with water containing 0.1% (v/v) formic acid, 0.03% (w/v) NaI and 1% (v/v) acetonitrile, flow rate 1.5 mL min-1, column C18 Alltech Prevail (250 mm × 4.6 mm i.d., 5 µm). peaks at 1.59, 2.26 and 4.15 min are traces of m/z = 147.12, 85.03, and 129.05 respectively and correspond to [lysine+H]+, [EG+Na]+ and [DEG+Na]+ respectively.
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
53
Table 6 shows the theoretical and the experimental molar ratios among different components
of pHEMA, PLGA(50:50)1550-diol(HEMA)2, and PLGA(50:50)1550-diol(etLDI-HEMA)2.
The theoretical ratios were estimated from the Figure 1 and the experimental ratios are based
on the quantitative data presented in Table 5. The results indicate that both pHEMA with 300
kDa Mv and PLGA(50:50)1550-diol(HEMA)2 are more deficient in EG than pHEMA with 20
kDa Mv. This suggests either the presence of cross-linking between two neighboring PMA
backbones via esterification and the formation of ethyl diacrylates or the pHEMA as a
starting material is partially hydrolyzed. In case of PLGA(50:50)1550-diol(HEMA)2, the
decrease in the amount of DEG with respect to PMA and EG may either be attributed to the
missing cross-links or to the presence of dangling chains without DEG. For
PLGA(50:50)1550-diol(etLDI-HEMA)2 the molar ratios between different building blocks are
close to those of the ideal structure except the lower amount of lysine compared to DEG.
Figure 10 TIC and XIC chromatogram of PLGA(50:50)1550-diol(HEMA)2 hydrolysate. Conditions: negative ESI mode with isocratic elution with water containing 0.1% (v/v) formic acid and 1% (v/v) acetonitrile, flow rate 1.5 mL min-1, column C18 Alltech Prevail (250 mm × 4.6 mm i.d., 5 µm). Peaks at 2.19 and 3.06 min are traces of m/z = 75.01 and 89.02 respectively and correspond to [GA-H]¯ and [LA-H]¯ respectively.
Chapter 2
54
Based on these experimental ratios between different components, the following structures
could be suggested for PLGA(50:50)1550-diol(HEMA)2 and PLGA(50:50)1550-diol(etLDI-
HEMA)2 (Figure 11).
B
A
BB
AA
Figure 11 Proposed chemical structure of (A) PLGA(50:50)1550-diol(HEMA)2 (B) PLGA(50:50)1550-diol(etLDI-HEMA)2 based on quantitative hydrolysis results (Table 1 and 2). Polymethacrylic acid ( ), ethylene glycol ( ), ethyl ester of lysine diisocyanate ( ), PLGA ( ), and diethylene glycol ( ).
However, it should be stipulated that this aposteriori analysis of the monomers after
hydrolysis only determines averages and cannot discriminate between different distributions,
e.g. of lysines in side chains, which could have implications for degradation of the
material. This stresses that it is imperative to involve analyses in each step of manufacturing,
assessing the starting materials and intermediates as well as the final product. The presented
method was not set up to determine the structure of the biomaterial network, but as a
balancing check accounting for all the resulting degradation products, since response factors
of oligomers are not precisely known, unlike those of the composing monomers.
In degradation studies this method allows quantitative analysis of oligomeric and other
intermediates that constitute the majority of degradation products under physiological
conditions, in the second stage of a two-step procedure: the first step uses physiologically
relevant conditions, while in the second step a fast and complete degradation is executed for
quantification of the final products.
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
55
Table 5 Sum
mary of the quantitative results of degradation products of biom
aterials performed w
ith HP
LC
-TO
F-MS.
292
4105
595
398
4108
Average %
yield3
20.0
20.05
20.2
20.3
39.8
Average m
ass of sam
ple
41.80
‐‐‐‐
‐‐‐‐
‐‐‐‐
‐‐‐‐
96 ±
4Lysin
e
25.76
47.10
‐‐‐‐
‐‐‐‐
422.43
92 ±
2GA
47.08
49.04
‐‐‐‐
‐‐‐‐
327.18
94 ±
4LA
41.24
61.57
20.13
50.20
43.09
97 ±
2DEG
14
1.39
12.41
77.87
36.27
‐‐‐‐
100 ±3
EG
72.43
25.26
116.81
118.97
‐‐‐‐
99 ±
1PMA‐Na1
% RSD
(n=3)
Amount
(mg)
% RSD
(n=2)
Amount
(mg)
% RSD
(n=2)
Amount
(mg)
% RSD
(n=2)
Amount
(mg)
% RSD
(n=2)
Amount
(mg)
PLG
A(50:50)1550 ‐
diol(etLD
I‐HEM
A)2
PLG
A(50:50)1550 ‐
diol(H
EMA)2
pHEM
A(20 kD
a)pHEM
A(300 kD
a)PLG
A(50:50)1550 ‐d
iol
Reco
very2
+/‐CV (%
)Hydrolysate
samples
292
4105
595
398
4108
Average %
yield3
20.0
20.05
20.2
20.3
39.8
Average m
ass of sam
ple
41.80
‐‐‐‐
‐‐‐‐
‐‐‐‐
‐‐‐‐
96 ±
4Lysin
e
25.76
47.10
‐‐‐‐
‐‐‐‐
422.43
92 ±
2GA
47.08
49.04
‐‐‐‐
‐‐‐‐
327.18
94 ±
4LA
41.24
61.57
20.13
50.20
43.09
97 ±
2DEG
14
1.39
12.41
77.87
36.27
‐‐‐‐
100 ±3
EG
72.43
25.26
116.81
118.97
‐‐‐‐
99 ±
1PMA‐Na1
% RSD
(n=3)
Amount
(mg)
% RSD
(n=2)
Amount
(mg)
% RSD
(n=2)
Amount
(mg)
% RSD
(n=2)
Amount
(mg)
% RSD
(n=2)
Amount
(mg)
PLG
A(50:50)1550 ‐
diol(etLD
I‐HEM
A)2
PLG
A(50:50)1550 ‐
diol(H
EMA)2
pHEM
A(20 kD
a)pHEM
A(300 kD
a)PLG
A(50:50)1550 ‐d
iol
Reco
very2
+/‐CV (%
)Hydrolysate
samples
1Obtained from
SEC
-dRI. 2R
ecoveries of each analytebased on control solution. 3T
he average % yield contains the data corrected w
ith mass of the
buildingblocks in the structure.
Chapter 2
56
Table 6 The theoretical and experimental ratio between different components of biomaterials
Ratio between PMA : EG EG : DEG PMA : DEG Lys: DEG
(Theoretical ratio) (1:1) (2:1) (2:1) (2:1)
PLGA(50:50)1550-diol(HEMA)2 1.0 : 0.74 2.0 : 0.77 2.0 : 0.61 --
PLGA(50:50)1550-diol(etLDI-HEMA)2 1.0 : 1.0 2.0 : 1.05 2.0 : 1.05 2.0 : 1.85
pHEMA (300 kDa) 1.0 : 0.56 - -- --
pHEMA (20 kDa) 1.0 : 0.83 - -- --
4 Conclusions
The current method leads to complete hydrolysis of pHEMA (both high and low Mw),
PLGA(50:50)1550-diol, PLGA(50:50)1550-diol(HEMA)2, and PLGA(50:50))1550-diol(etLDI-
HEMA)2 in 24 h at 120oC. The Teflon lined microwave vial was helpful to avoid contact of
reaction medium with the glass vial. This leads to the complete degradation of biomaterial
without the formation of insoluble residues under harsh conditions (high pH, temperature
and pressure). NMR proved to be a good analytical technique to monitor the cleavage of
bonds in these biomaterials. HPLC-ToF-MS can be utilized to quantify the monomers in the
hydrolysis mixture. The origin of bimodality in the MMD of PMA-Na can be inferred from
the non-Gaussian distribution of the starting material. This study will be helpful to
investigate the hydrolytic degradation and for the compositional analysis of novel polymeric
networks including pHEMA as an intermediate product.
Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials
57
5 References
[1] G. Wnek, G.L. Bowlin, Encyclopedia of Biomaterials and Biomedical Engineering, Informa Health Care, 2004.
[2] G. Mabilleau, M.F. Moreau, R. Filmon, M.F. Baslé, D. Chappard, Biomaterials 25 (2004) 5155.
[3] B.D. Ratner, A.S. Hoffman, F.J. Schoen, J.E. Lemons, Biomaterials Science, Second Edition: An Introduction to Materials in Medicine Academic Press, 2004.
[4] J.B. Park, J.D. Bronzino, Biomaterials: Principles and Applications, CRC Press, 2003.
[5] J. Stitzel, J. Liu, S.J. Lee, M. Komura, J. Berry, S. Soker, G. Lim, M. Van Dyke, R. Czerw, J.J. Yoo, A. Atala, Biomaterials 27 (2006) 1088.
[6] S. Suzuki, K. Matsuda, T. Maruguchi, Y. Nishimura, Y. Ikada, British Journal of Plastic Surgery 48 (1995) 222.
[7] S.H. Kim, C. Marmo, G.A. Somorjai, Biomaterials 22 (2001) 3285. [8] M. Schuler, G.R. Owen, D.W. Hamilton, M. de Wild, M. Textor, D.M. Brunette,
S.G.P. Tosatti, Biomaterials 27 (2006) 4003. [9] P. Klement, Y.J. Du, L.R. Berry, P. Tressel, A.K.C. Chan, Biomaterials 27 (2006)
5107. [10] S. Atzet, S. Curtin, P. Trinh, S. Bryant, B. Ratner, Biomacromolecules 9 (2008)
3370. [11] K.D.F. Vlugt-Wensink, X. Jiang, G. Schotman, G. Kruijtzer, A. Vredenberg, J.T.
Chung, Z. Zhang, C. Versluis, D. Ramos, R. Verrijk, W. Jiskoot, D.J.A. Crommelin, W.E. Hennink, Biomacromolecules 7 (2006) 2983.
[12] N.S. Khelfallah, G. Decher, P.J. Mesini, Biointerphases 2 (2007) 131. [13] S.C. Baker, G. Rohman, J. Southgate, N.R. Cameron, Biomaterials 30 (2009) 1321. [14] E.I.d.P.d. Nemours, Ind Eng Chem 28 (1936) 1160. [15] O. Wichterle, D. Lim, Nature 185 (1960) 117. [16] S. Abraham, S. Brahim, K. Ishihara, A. Guiseppi-Elie, Biomaterials 26 (2005) 4767. [17] J.-P. Montheard, M. Chatzopoulos, D. Chappard, Rev Macromol Chem Phys C32
(1992) 1 [18] F.O. Eschbach, S.J. Huang, J.A. Cameron, J Bioact Compatible Polym 9 (1994) 210. [19] J. Gracida, J. Alba, J. Cardoso, F. Perez-Guevara, Polym Degrad Stab 83 (2004) 247. [20] X. Jiang, M.C. Lok, W.E. Hennink, Bioconjugate Chem 18 (2007) 2077. [21] P.Y. Ghi, D.J.T. Hill, A.K. Whittaker, Biomacromolecules 3 (2002) 991. [22] M.Y. Arıca, D. Tuğlu, M.M. Başar, D. Kılıç, G. Bayramoğlu, E. Batislam, J Biomed
Mater Res B Appl Biomater 86B (2008) 18. [23] D. Horák, A. Adamyan, O. Golubeva, N. Skuba, T. Vinokurova, J Mater Sci Mater
Med 17 (2006) 123. [24] J. Lukáš, K. Smetana, P. Petrovický, V. Palečková, J. Vacik, B. Dvořánková, L.
Brož, D. Pospíšilová, Z. Holíková, J. Bartůňková, J Mater Sci Mater Med 12 (2001) 639.
[25] M. Houchin, E. Topp, J Pharm Sci 97 (2008) 2395. [26] L. Sun, S. Zhou, W. Wang, X. Li, J. Wang, J. Weng, Colloids Surf, A 345 (2009)
173. [27] T. Ishihara, T. Kubota, T. Choi, M. Takahashi, E. Ayano, H. Kanazawa, M. Higaki,
Int J Pharm 375 (2009) 148. [28] T. Yoshioka, N. Kawazoe, T. Tateishi, G. Chen, Biomaterials 29 (2008) 3438.
Chapter 2
58
[29] C. Hwang, Y. Park, J. Park, K. Lee, K. Sun, A. Khademhosseini, S. Lee, Biomedical Microdevices 11 (2009) 739.
[30] M. Takahashi, H. Onishi, Y. Machida, J Controlled Release 100 (2004) 63. [31] S.H. Choi, T.G. Park, J Biomater Sci Polym Ed 13 (2002) 1163. [32] Y. Duan, J. Xu, Y. Lin, H. Yu, T. Gong, Y. Li, Z. Zhang, J Biomed Mater Res A 87A
(2008) 515. [33] D.-D. Lu, J.-C. Yuan, L.-G. You, Z.-Q. Lei, J Macromol Sci, Part A: Pure Appl
Chem 45 (2008) 705 [34] S.-W. Jung, Y.-I. Jeong, Y.-H. Kim, K.-C. Choi, S.-H. Kim, J Microencapsulation 22
(2005) 901. [35] K. Sutherland, J.R. Mahoney, A.J. Coury, J.W. Eaton, J Clin Invest 92 (1993) 2360. [36] G.T. Chao, Z.Y. Qian, M.J. Huang, B. Kan, Y.C. Gu, C.Y. Gong, J.L. Yang, K.
Wang, M. Dai, X.Y. Li, M.L. Gou, M.J. Tu, Y.Q. Wei, J Biomed Mater Res A 85A (2008) 36.
[37] J.B. Hutchison, A.S. Lindquist, K.S. Anseth, Macromolecules 37 (2004) 3823. [38] H. Matsubara, S.-i. Hata, Y. Kondo, Y. Ishida, H. Takigawa, H. Ohtani, Anal Sci 22
(2006) 1403. [39] R. Peters, Ph.D thesis, Characterization of polymeric network structures, University
of Amsterdam, 2009. [40] A. A. Dias, A. Petit, in European Patent: EP1996640 (B1), 2009. [41] S.T. Popovici, P.J. Schoenmakers, J Chromatogr A 1099 (2005) 92. [42] S.-T. Popovici, W.T. Kok, P.J. Schoenmakers, J Chromatogr A 1060 (2004) 237. [43] D. Bikiaris, V. Karavelidis, G. Karayannidis, Macromol Rapid Commun 27 (2006)
1199. [44] Zainuddin, D. Hill, A. Whittaker, L. Lambert, T. Chirila, J Mater Sci Mater Med 18
(2007) 1141. [45] J.V.M. Weaver, I. Bannister, K.L. Robinson, X. Bories-Azeau, S.P. Armes, M.
Smallridge, P. McKenna, Macromolecules 37 (2004) 2395. [46] S. Ito, K. Tsukada, J Chromatogr A 943 (2001) 39. [47] M. Villagrasa, M. Guillamón, E. Eljarrat, D. Barceló, J Chromatogr A 1157 (2007)
108.
A. Ghaffar, G. J. J. Draaisma, G. Mihov, A. A. Dias, P.J. Schoenmakers, Sj. van der Wal, Biomacromolecules 12 (2011) 3243-3251.
Chapter 3
3. Monitoring the in vitro enzyme-mediated degradation of
degradable poly(ester amide) for controlled drug delivery by LC-
ToF-MS
To scrutinize materials for specific biomedical applications, we need sensitive and selective
analytical methods that can give more insight into the process of their biodegradation. In the
present study, the enzymatic degradation of multi-block poly(ester amide) based on natural
amino acids, such as lysine and leucine, was performed with serine proteases (α-
chymotrypsin (α-CT) and proteinase K (PK)) in phosphate-buffered saline solution at 37oC
for 4 weeks. Fully and partially degraded water-soluble products were analyzed by liquid
chromatography hyphenated with time-of-flight mass spectrometry using an electrospray
interface (LC-ESI-ToF-MS). Tracking the release of monomeric and oligomeric products
into the enzyme media during the course of enzymatic degradation revealed the preferences
of R-CT and PK toward ester and amide bonds: both α-CT and PK showed esterase and
amidase activity. Although within the experimental time frame up to 30 and 15% weight loss
was observed in case of α-CT and PK, respectively, analysis by size-exclusion
chromatography showed no change in the characteristic molecular-weight averages of the
remaining polymer. This suggests that the enzymatic degradation occurs at the surface of this
biomaterial. A sustained and linear degradation over a period of 4 weeks supports the
potential of this class of poly(ester amide)s for drug delivery applications.
Chapter 3
60
1 Introduction
The demand for design and synthesis of biodegradable polymeric materials for application in
drug-delivery devices, scaffolds for tissue engineering, and surgical implants, such as
sutures, pins, rods, and screws for fixation devices, is increasing [1,2]. From this perspective,
degradable poly(ester amide)s (PEAs), preferably those containing natural amino acids, have
gained much attention. These materials have been studied extensively in terms of their
biocompatibility and degradation and have been tested in vivo as drug-eluting stent coatings
in clinical trials such as the NOBLESSE trial [3,4]. PEAs contain ester bonds that are
susceptible to hydrolytic degradation. The inclusion of α-amino acids results in improved
mechanical and thermal properties through hydrogen bond interactions [5]. In general,
degradation of biomaterials may take place either through bulk degradation or via surface
erosion [6]. It is important to differentiate between these two mechanisms for a given
biomaterial if one is to control the drug release rate by changing the ratios of ester, amide,
and methylene groups. This requires knowledge of the relative rates of hydrolysis at different
sites within the polymeric chains and determination of the structure of degradation products,
which may critically affect the biocompatibility and rate of clearance from the body [7].
The sequence of the monomers in a multi-block polymer affects its crystalline or amorphous
nature, which in turn influences the degradation rate [8,9]. Therefore, it is important to
separate and identify the low-molecular-weight water-soluble products during
biodegradation in order to correlate the structural parameters with rates of degradation [10].
The collected information will be helpful (i) to estimate overall structure and optimize the
synthetic approach towards a functional material, (ii) to evaluate the toxicity of the
degradation products at an early stage, (iii) to determine the rate of hydrolysis at different
sites (e.g., specificity of enzyme for ester or amide bonds) and (iv) for the rational design of
new materials.
Several studies report the use of liquid chromatography for the separation of water-soluble
degradation products of polyesters and their identification with UV-Vis using standards [11],
FAB-MS [12], ESI-MS [13], and so on or their offline analysis with NMR [14]. Rizzarelli et
al. utilized the combination of high-performance liquid chromatography with electrospray-
ionization mass spectrometry (LC-ESI-MS) or tandem mass spectrometry (LC-ESI-MS/MS).
They noticed selective ester hydrolysis in aliphatic copolyesters catalyzed by lipases [15].
Mass spectrometry has emerged as a powerful analytical tool for the characterization of
natural and synthetic macromolecules [7,16], but the technique has limitations for high-
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
61
molecular-weight synthetic polymers. This is true for both ionization techniques that are
commonly applied for high-molecular-weight analytes, that is, ESI and matrix-assisted laser
desorption/ionization (MALDI). When using ESI, multiply charged ions swamp the
spectrum, amplifying the number of ions arising from the molecular-weight distribution
(MWD). In MALDI, both statistics and charge affinity may cause low-molecular-weight
oligomers to dominate the spectrum.
In the current chapter, stainless steel disks coated with a class of PEAs (Scheme 1)
composed of sebacic acid, 1,6-hexanediol, lysine, and leucine with two different sequences
of repeating units were subjected to in vitro enzymatic degradation with α-chymotrypsin (α-
CT) and proteinase K (PK). The weight-average molecular weight (Mw) and dispersity of the
remaining un-dissolved material were analyzed by SEC before and after degradation. The
water-soluble degradation products (monomers and oligomers) were separated on an LC
column and their structure was characterized by time-of-flight (ToF) MS.
The objectives of the current study were (i) to determine the effectiveness of α-CT and PK as
model enzymatic systems to degrade the present PEA by measuring weight losses, (ii) to
differentiate between surface and bulk degradation by estimating the average molecular
weights and dispersity of the polymer before and after degradation, and (iii) to assess the
specific activity (qualitative assessment of preferences and reaction rates) of the enzymes
with the present PEA by (semi-quantitatively) determining the monomeric and oligomeric
products released during biodegradation.
The implication of this research is that a comparative degradation study of several PEAs by
model enzymes may guide the development of PEA biomaterials towards specific properties,
such as degradation rate, controlled drug-delivery rate, targeted release of active
pharmaceutical molecules, and so on.
Scheme 1 Structure of poly (ester amide)s subjected to enzymatic degradation. Where x = 3 and y = 4. The solid and dashed arrows represent the possible cleavage of ester and amide bonds, respectively.
HN
HN
OO O
n
O
NH
O
O
O
OHN
O
O
y x ym
Chapter 3
62
2 Materials and methods
2.1 Materials
The polymer was synthesized at DSM Ahead following a published procedure [17]. The
chemical structure is shown in Scheme 1. The actual values are m/n = 3:1, x = 3, and y = 4.
Figure 1 shows the 1H NMR spectrum of the PEA recorded in d6-ethanol (Euriso-top, Gif sur
Yvette, France) on a Varian Inova 500 MHz NMR (Varian, Walnut Creek, CA, USA)
equipped with a 5 mm 13C/31P/1H GS Probe 500. A pulse-repetition time of 25 s, a pulse
duration of 3.6 µs, 63 scans and a temperature of 25oC were used to record 1H spectra. The
integration of characteristic protons signals for 1,6-hexanediol (15) and lysine (5) confirmed
the molar ratio of 3:1 between m and n block of the polymer. There is 1:1 ratio between
lysine (5) and benzyl group (18). Lysine (5) and sebacic acid (6) possess 1:1.13 ratio in the n
block of the polymer. The 1,6-hexanediol (15) and sebacic acid (6 + 9) signal ratio also
supports the m/n = 3:1 molar ratio. Accurate integration was difficult in case of overlapping
signals. The HSQC NMR experiments were done to confirm the characteristic signals (see
Figure S1 in the supporting information). Glass transition temperature (Tg) of the analyzed
polymer is 33oC, determined via DSC. The test sample, 5 mg, was vacuum-dried at 65 ± 5º C
and placed in a crucible pan. Next the sample was analyzed on a Mettler Toledo DSC 822e
instrument and Tg was derived from the second heating curve.
2.2 Solubility
The solubility of the biomaterial in water and in a number of common organic solvents was
assessed by combining 2 to 3 mg of the PEA with 1 mL of the respective solvents (Table1)
at room temperature (25oC). The PEA was considered to be soluble when the solutions
became completely transparent.
2.3 Enzyme activity
N-Suc-Ala-Ala-Pro-Phe-pNA (Bachem, Bubenhof, Switzerland) was used as a chromogenic
substrate to determine the activity of α-CT from bovine pancreas (Fluka, Steinheim,
Germany, pr.no. 27270, > 68 units/mg protein). The amount of p-nitrophenyl anilide
released was determined by recording the absorbance at 410 nm and 25°C using a
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
63
spectrophotometer. The amount of enzyme activity releasing 1 µmol of chromophore per
min is defined as 1 unit [18].
The activity of the PK from Tritirachium album (Sigma Aldrich, Steinheim, Germany, pr.
no. P2308, >30 units/mg protein) was measured by colorimetric analysis according to the
quality-control procedure described by the supplier (Enzymatic Assay of Proteinase K, EC
3.4.21.64). One unit will hydrolyze urea-denatured hemoglobin to produce a colour
equivalent to 1 µmol (181 µg) of tyrosine per min at pH 7.5 at 37 °C [19].
Figure 1 1H NMR spectrum of the PEA in d6-ethanol.
2.4 In vitro enzyme-mediated degradation
Approximately 20 mg of PEA were drop-cast on one side of stainless-steel round disks
(diameter 13 mm, thickness 80 µm, surface area 133 mm2). The coatings were applied in
three layers by pipetting 70 µL of polymer solution prepared in ethanol (0.1 g/mL, filtered
through a 0.45 μm filter). Each layer was allowed to air-dry for at least 2 h at ambient
temperature before the next layer was applied. After the final layer was applied, the coated
disks were air-dried overnight at room temperature, followed by drying at 40oC under
reduced pressure to a constant weight.
Chapter 3
64
The coated dried disks (in triplicate) were immersed in 1.5 mL of phosphate-buffered saline
(PBS) buffer (0.2 g KCl, 0.2 g KH2PO4, 1.15 g Na2HPO4, 8 g NaCl in 1 L, containing
sodium azide (0.5 g/L, to inhibit bacterial growth) with α-CT (17 U/mL; pH 8) or with PK
(5 U/mL; pH 7.4) in 15 mL polypropylene conical tubes (BD, Franklin Lakes, NJ, USA)
and incubated at 37oC and 120 rpm (Innova 44 Incubator Shaker Series, New Brunswick
Scientific, Edison, NJ, USA). The enzyme solutions were refreshed every 48 h and stored at
–20oC for LC-ToF-MS analysis after their pH was monitored. The remaining polymer
samples (on disks, each in triplicate) were collected for gravimetric analysis randomly on
day 7, 14, 21, and 28. The solutions were aspirated and the disks were rinsed three times for
5 min with distilled water. The samples of remaining polymer were then dried under vacuum
at 65oC for 48 h. The samples were dried for additional 24 h at 65oC to enssure constant
weight. The percent weight loss was calculated using the following formula:
100)disk theofweight ()sample of weight initial(
)disk theofweight () sample theofweight (1(%) lossWeight
disk
disk
Enzyme blank and polymer blank samples were also incubated and analyzed to correct the
data. Under the current experimental setup, the enzyme media was replaced with fresh
enzyme after every 48 h. Therefore, to estimate the decrease in the activity of enzymes, we
collected samples from the enzyme media (both α-CT and PK) incubated with coated and
non-coated disks after 8, 24, 32, and 48 h for LC-ToF-MS analysis.
2.5 Size-exclusion chromatography
The SEC experiments were performed on an LC system equipped with a LC-10AD solvent
delivery module, a CTO-6A column oven, a SIL-9A auto injector, an SPD-10AVvp UV-vis
detector, and an RID-10A refractive-index detector (all from Shimadzu, Kyoto, Japan). The
SEC analyses were performed on three PLgel MIXED-B columns (10 µm, 300 × 7.6 mm
i.d.; Polymer Laboratories, Church Stretton, U.K.) connected in series. THF stabilized with
butylated hydroxytoluene (BHT) (BioSolve, Valkenswaard, The Netherlands) was pumped
at a flow rate of 1 mL min-1. The injection volume was 50 µL, and the column oven
temperature was set at 50oC. Polystyrene standards (Polymer Laboratories, Shropshire, U.K.)
were used to calibrate the SEC-UV-dRI system. Data were recorded and chromatographic
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
65
peaks were treated using Class-VP 7.4 software (Shimadzu). Molar-mass distributions
(MMDs) were calculated from the chromatograms using software written in-house in Excel
2003 (Microsoft).
2.6 LC-ToF-MS study
We injected 5 μL of the enzyme solutions collected after every 48 h without dilution after
filtration through a 0.2 µm pore size PTFE filter (Grace Davison). The stock solutions of
1,6-hexanediol (Aldrich), sebacic acid (Fluka) and benzyl alcohol (Fluka), were prepared in
PBS, and quantification was based on a standard-addition method in positive ESI mode for
1,6-hexanediol and in negative ESI mode for sebacic acid. The concentration of benzyl
alcohol was determined by LC-UV at 254 nm.
The chromatographic separations of soluble degraded products were performed on a Prevail
C18 Column (250 × 4.6 mm i.d., 5 µm particle size) (Grace Davison, Deerfield, IL, USA)
connected to an Agilent 1100-series LC system consisting of a degasser, a gradient pump,
and an auto-sampler (all from Agilent Technologies, Waldbronn, Germany). The column
oven (Temperature Control Module from Waters, Milford, MA, USA) was set at 40oC.
Mobile phase A was 0.1% (v/v) aqueous formic acid (Fluka, Steinheim, Germany) and B
was acetonitrile (Biosolve). The gradient was started at t = 0 min with 5% (v/v) B, reaching
60% (v/v) B in 25 min, held constant for 2 min, and then back to 5% (v/v) at 30 min
(tend = 35 min) at a flow rate of 1.5 mL min-1 and was split between a waste reservoir and the
ESI interface by means of a zero-dead-volume T-piece to enssure a flow of approximately
~0.2 mL min-1 into the electrospray. Organic solvents used for the LC mobile phase were of
LC grade. Highly pure water for mobile-phase preparation was obtained from an Arium 611
Ultrapure (18.2 MΩ*cm) Water System (Sartorius, Goettingen, Germany).
The LC system was hyphenated with a 6210 series time-of-flight mass spectrometer
(Agilent) via an ESI interface. The conditions of the ESI-ToF-MS were as follows. Drying
gas was nitrogen at 8 L min-1, 300oC and 200 kPa. The capillary voltage was 3500 V; the
fragmenter was set at 140 V and the skimmer at 60 V. Octopole dc1 was set at 33 V;
octopole radio frequency at 250 V. The data were acquired in the scan mode from m/z 50 to
3000 with 0.88 scans/s. An Agilent MassHunter Workstation A.02.01 and Analyst QS 1.1
software (Applied Biosystems, Carlsbad, California, USA) were used for data acquisition
and data analysis, respectively.
Chapter 3
66
The LC-ToF-MS separation and identification of the water-soluble degradation products was
optimized by monomers and oligomers, generated by means of chemical degradation. (see
the supporting information).
3 Results and discussion
3.1 Solubility
The solubility of the PEA in a number of organic solvents at room temperature (25oC) is
summarized in Table 1. The PEA is completely soluble in most of the polar (protic and
aprotic) solvents, except in water, acetonitrile, and ethyl acetate.
Table 1 Solubility of the PEA used in the present study in common organic solvents evaluated at 25oC.
solvents Solubility solvents Solubility solvents Solubility
n-Hexane - Dichloromethane + H2O -
Cyclohexane - Tetrahydrofuran + Methanol +
Chloroform + Ethyl acetate - Ethanol +
Toluene - Dimethylformamide + Isopropanol +
N,N`-dimethyl acetamide + Acetonitrile - Acetic acid +
Dimethyl sulfoxide + Acetone +
Soluble +, insoluble -
3.2 Overall effectiveness of in vitro enzyme-mediated degradation
The results in Figure 2 show a greater weight loss (30%) in case of α-CT than when using
PK (15%) after 4 weeks under the current experimental conditions. When incubating with
only PBS buffer, no noticeable weight loss was observed. The biomaterial showed a
sustained and near-linear degradation over a period of 4 weeks (Figure 2). This weight loss
the polymer indicates constant erosion of the surface by the enzymes. No solid particles were
observed in the reaction media. No decrease in the pH of the degradation media was
observed during the course of the enzyme-mediated degradation, implying that the erosion
does not result in an accumulation of acidic degradation products.
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
67
-5
0
5
10
15
20
25
30
35
0 5 10 15 20 25 30
Days
Gra
vim
etri
c w
eig
ht
loss
(%
)
Figure 2 Percentage of polymer weight loss as a function of the degradation time (days) in α-Chymotrypsin (α-CT, 17 U/mL) and proteinase K (PK, 5 U/mL) in PBS. The “blank” experiment in PBS is shown for comparison. Lines with triangles, squares and diamonds correspond to blank, PK and α-CT experimental data, respectively.
0
1
2
3
4
3 3,5 4 4,5 5 5,5 6
log M
dR
I re
spo
nse
0
7
14
21
28
Figure 3 The molar-mass distribution of the PEA before and after enzymatic degradation with α-CT. The side bar on the right side represents the number of days of treatment.
Chapter 3
68
3.3 Molecular-weight of remaining material
The Mn and dispersity of the starting material were 31 kDa and 2.1, respectively. The
molecular weight and molecular-weight distribution (MWD) of the sample prior to
degradation and of the remaining un-dissolved samples after incubation with α-CT were
determined. The SEC-dRI system was calibrated with polystyrene standards in THF. No
significant differences were observed between the molecular weights of the samples prior to
and after enzymatic degradation (Figure 3). This maintenance of molecular weight is in line
with the enzymatic degradation being a surface-erosion phenomenon. Similar results were
reported by Fan et al. [20,21] and by Guo et al. [22].
3.4 LC-ToF-MS analysis following enzymatic degradation
The chromatographic separation and identification of the monomers and oligomers librated
as a result of enzymatic degradation was monitored using LC-ToF-MS with electro-spray
ionization in the positive and negative modes. Formic acid was used to make proton adducts.
Sodium adducts were also observed because of the presence of sodium ions in the incubating
medium. The presence of multiple charge ions were identified by the differences of m/z
values between the isotopic peaks of respective masses. The negative mode spectra showed a
difference of 2 and 24 Da for proton and sodium adducts, respectively, as compared with
positive mode spectra. Benzyl alcohol showed the [M-OH]+ adduct only in positive mode.
The m/z and the retention time of benzyl alcohol in the enzyme solutions were confirmed by
the benzyl alcohol standard. The ESI-ToF-MS spectra of the identified peaks (Table 1) are
presented in the supporting information. The copolymer consists of five components; leucine
(L), lysine (K), 1,6-hexanediol (C6), sebacic acid (C10), and benzyl alcohol (BnOH). All the
LC10 and KC10 combinations contained an amide bond. In case of LC6, the carboxylic
group of L is connected to C6 via an ester bond. The carboxylic group of the lysine is end-
capped with benzyl alcohol and forms an ester bond. Scheme 1 shows the possible
fragmentation pattern in case of both ester- and amide-bond cleavage. If only the ester bonds
are hydrolyzed, degradation products will be C6 (peak number 4), benzyl alcohol (6), LC10L
(11), LC10LC6 (17), and the oligomers LC10(KC10)nL originating from n block of the
polymer. If amide-bond cleavage is simultaneously taking place along with ester hydrolysis,
then the additional products that may appear are K (peak number 1), L (3), LC6, C10 (8),
LC10 (9), C10K (5), LC10K(7), and LC10KC10 (10).
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
69
Figure 4 displays the separation of the different degradation products at various intervals of
time during enzymatic degradation of PEA with α-CT. Mass-spectrometric analysis shows
the release of L (peak number 3), C6 (4), C10K (5), benzyl alcohol (6), LC10K (7), C10 (8)
LC10 (9) LC10KC10 (10), LC10L (11). Peaks 12–14 represent oligomeric blocks of
(LC10[KC10]nL) with n number of repeating unites ranging from 1 to 4. The sensitivity of
ESI-ToF-MS decreases significantly with increasing n. Therefore, no direct conclusion can
be drawn from the peak areas in the chromatograms. The products generated at different
stages of the enzymatic degradation (moments in time) indicate that ester hydrolysis has
dominated during enzymatic degradation up to day 18; after that, an unexpected change was
observed and α-CT acted mainly as a protease cleaving predominately the amide bonds.
Figure 4 Selected TIC chromatograms of the degradation products obtained at different time intervals during α-chymotrypsin degradation at 37°C. Val and Phe represent valine and phenylalanine, respectively.
In case of PK, the appearance of L (peak number 3), C6 (4), benzyl alcohol (6), LC10K (7),
LC10 (9), LC10KC10 (12), and so on indicated the cleavage of both ester and amide bonds.
The enzymes in both cases showed preference towards the amide bond between L and C10
compared with the one between K and C10 because no release of K was observed (Figure 5).
Table 2 shows the structural assignments of different peaks separated in Figures 4 and 5
based on the m/z ratios from their ESI-ToF mass spectra.
Chapter 3
70
Figure 5 Selected TIC chromatograms of the degradation products obtained after different time intervals during enzymatic degradation of the PEA with proteinase K at 37oC and 120 rpm.
Figure 6 Cumulative increase (y-axis) in the concentration (µg.mL-1) of 1,6-hexanediol (◊ and ♦ symbols), sebacic acid ( and symbols) and benzyl alcohol (∆ and symbols) during the course of enzymatic degradation of the PEA with α-chymotrypsin (solid lines) and proteinase K (dashed lines). The concentration of benzyl alcohol was determined by LC with UV-absorbance detection at 254 nm.
Because C6 is connected on both sides with L through ester bonds and C10 is bonded via
amide bonds either to two K or to ne L and one K, the emerging C6 and C10 can be used to
monitor the ultimate degradation due to ester hydrolysis or amide bond-cleavage. Figure 6
demonstrates a nearly linear increase in the cumulative concentration of benzyl alcohol, C6
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
71
and C10 in the enzyme media collected after every 48 h during the enzymatic degradation
with α-CT and PK. It is obvious from the quantitative data that the release of C10 greatly
increased after the 18th day in α-CT degraded hydrolysates, whereas in case of PK, very
little C10 was librated. An interesting observation is that in case of the PK-mediated
degradation the detected C6 in solution corresponds to 100% of the content of this moiety in
the hydrolyzed polymer (estimation is based on the observed weight loss). However, more
than half of the potential C6 appeared in fragments other than C6 itself when using α-CT.
Although it is clear (cf. peaks nos. 3, 7 and 9) that PK exhibits amidase activity, the amount
of C10 was negligible in the medium. This stresses the need for quantitative analysis of the
intermediate degradation products.
Unfortunately, we can only analyze the oligomers in a semi-quantitative manner so far. The
contribution of K, L, valine (Val) and phenylalanine (Phe) from the enzymes to the reaction
media was confirmed by LC-ToF-MS analysis of the sample blank used in the study and also
by the hydrolytic degradation of the enzymes.
3.5 Factors affecting enzyme activities
The effects of equilibrium or saturation of the degradation products on the activity of the
enzymes were estimated by analyzing the reaction media incubated at 37oC and 120 rpm
after 8, 24, 32, and 48 h by LC-ToF-MS. Figure 7 shows the peak areas for peaks from 11 to
15. In case of α-CT, the heights of these peaks are almost approaching almost constant
levels. However, the peaks heights at 48 h are decreasing gradually as the molecular weights
of the oligomers are increasing. This suggests further degradation of the water-soluble
oligomers in the enzyme media at these prolonged incubating times. In case of PK, this
situation has not yet been reached after 48 h.
The decrease in enzyme activity could be caused either by self-digestion or by denaturation
of the enzyme. A gel-electrophoresis experiment was performed to compare a freshly
prepared enzyme solution with a solution which had been stored at 37ºC for 3 days [23].
This experiment clearly showed several small enzyme fragments in the stored sample.
The change in the activity of the α-CT during the enzymatic degradation was measured after
6, 18, 24, 30, and 48 h. At 0 h, the activity measurement involved only enzyme solution and
the standard substrate (N-Suc-Ala-Ala-Pro-Phe-pNA). But after 6, 18, 24, 30, and 48 h, the
enzyme media collected from the incubated coated disks also contained water-soluble
degradation products from PEA. These products in the presence of standard substrate do not
Chapter 3
72
appear to inhibit the activity of enzyme. Nevertheless, we found a decrease of >90% in the
activity of enzyme after 48 h under the current experimental conditions.
Figure 7 TIC LC-ToF-MS chromatograms of the reaction media injected after 8 (∆), 24 (), 32 (), and 48 (◊) hrs of enzymatic degradation. (a) α-chymotrypsin (b) proteinase K.
In summary, the PEA degraded primarily via ester hydrolysis during in vitro enzyme-
mediated degradation (with α-CT and PK at 37oC). The survival of KC10-containing
fragments under the applied enzymatic conditions suggested that the amide bonds between K
and C10 are more stable than those between L and C10. The analysis shows only fragments,
which theoretically could be derived directly from the anticipated polymer structure. This
suggests a polymerization process without side reactions. A significant acceleration of the
degradation may possibly be obtained if the enzyme solution is more frequently or even
continuously refreshed. An on-stream-analysis system to study and explore these synergistic
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
73
effects is currently being developed in our laboratory. The rate-determining factor should
still be the accessible surface area of the biomaterial, but the conditions will be more
favourable for degradation than those used here.
Table 2 Structural assignments for the fragments arising from different monomers and oligomers, separated as shown in Figures 4 and 5 following the enzyme-mediated degradation of the PEA in α-Chymotrypsin and Proteinase K. The symbols K, L, C6 and C10 represent lysine, leucine, 1,6-hexanediol, and sebacic acid, respectively.
Peak no.
Symbol Structure Number of amide bond broken
m/z [M+H]+
1 K
2 147.1129
2 formic acid
3 L
1 132.1033
4 C6
0 119.11099
5 C10K
2 331.2211
6 benzyl alcohol
0 91.0554
[M-OH]+
7 LC10K
1 444.3106
8 C10
2 201.1154
[M-H]-
9 LC10
1 316.2143
Chapter 3
74
10 LC10KC10
1 628.4277
11 LC10L
0 429.3072
12 LC10(KC10)1L
0 741.5221
13 LC10(KC10)2L
0 1053.7141
14 LC10(KC10)3L
HO
(CH3)2HC
O
NH
O
O
HN
HN
4 2
HO OO
O
NH
OH
CH(CH3)2
O4
3
0 1365.9289
15 LC10(KC10)4L
0 1678.1283
16 C10LC6
1 416.2977
17 LC10LC6
HO
(CH3)2HC
O
NH
O
O
HN
4
CH(CH3)2
O
OOH3
0 529.3803
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
75
4 Conclusions
Poly(ester amide) polymer was subjected to enzymatic degradation conditions. Hyphenation
of liquid chromatography with electrospray – time-of-flight – mass spectrometry proved to
be a powerful analytical tool for the chromatographic separation and identification of the
water-soluble degradation products after enzymatic degradation. The technique allowed the
identification of fully and partially degraded polymer fragments, thus providing information
on the polymerization process and on the intrinsic polymer structure. The polymer was found
to degrade at a steady rate with both enzymes during this study. A lack of significant
changes in the average molecular weight of the remaining polymer strongly suggests that
surface erosion occurred during the enzyme-mediated degradation. Furthermore, no
accumulation of acidic byproducts was observed during the course of the experiment. In this
respect, the polymer performs better than conventional polyesters. The experiments
confirmed that this class of (polyester amide)s shows a remarkable hydrolytic stability in the
absence of enzymes.
Moreover, to avoid the further degradation of the water-soluble degradation products in the
enzyme media due to long incubation time and to enhance the degradation rate by
continuously refreshing the enzyme solution, the development of a system for on-stream
analysis of degradation products is in progress.
Chapter 3
76
5 References
[1] in 't Veld, P. J. A.; Dijkstra, P. J.; Feijen, J. Clin. Mater. 1993, 13, (1-4), 143-147. [2] Mihov, G.; Draaisma, G.; Dias, A.; Turnell, B.; Gomurashvili, Z. Abstract book: 11th
European Symposium on Controlled Drug Delivery 2010, 340-342. [3] Sarkar, D.; Lopina, S. T. Polym. Degrad. Stab. 2007, 92, (11), 1994-2004. [4] M.DeFife, K.; Grako, K.; Cruz-Aranda, G.; Price, S.; Chantung, R.; Macpherson, K.;
Khoshabeh, R.; Gopalan, S.; Turnell, W. G. J. Biomater. Sci., Polym. Ed. 2009, 20, 1495-1511.
[5] Paredes, N.; Rodriguez-Galán, A.; Puiggalí, J.; Peraire, C. J. Appl. Polym. Sci. 1998, 69, (8), 1537-1549.
[6] Burkersroda, F. v.; Schedl, L.; Göpferich, A. Biomaterials 2002, 23, (21), 4221-4231.
[7] Adamus, G.; Hakkarainen, M.; Höglund, A.; Kowalczuk, M.; Albertsson, A.-C. Biomacromolecules 2009, 10, (6), 1540-1546.
[8] Montaudo, G.; Rizzarelli, P. Polym. Degrad. Stab. 2000, 70, (2), 305-314. [9] Gan, Z.; Abe, H.; Doi, Y. Biomacromolecules 2001, 2, (1), 313-321. [10] Rizzarelli, P.; Puglisi, C.; Montaudo, G. Rapid Commun. Mass Spectrom. 2005, 19,
(17), 2407-2418. [11] Scherer, T. M.; Fuller, R. C.; Lenz, R. W.; Goodwin, S. Polym. Degrad. Stab. 1999,
64, (2), 267-275. [12] Kitakuni, E.; Yoshikawa, K.; Nakano, K.; Sasuga, J.; Nobiki, M.; Naoi, H.; Yokota,
Y.; Ishioka, R.; Yakabe, Y. Environ. Toxicol. Chem. 2001, 20, (5), 941-946. [13] Scandola, M.; Focarete, M. L.; Adamus, G.; Sikorska, W.; Baranowska, I.;
Świerczek, S.; Gnatowski, M.; Kowalczuk, M.; Jedliński, Z. Macromolecules 1997, 30, (9), 2568-2574.
[14] Abe, H.; Doi, Y.; Aoki, H.; Akehata, T.; Hori, Y.; Yamaguchi, A. Macromolecules 1995, 28, (23), 7630-7637.
[15] Rizzarelli, P.; Impallomeni, G.; Montaudo, G. Biomacromolecules 2004, 5, (2), 433-444.
[16] Wang, N.; Li, L. J. Am. Soc. Mass Spectrom. 2010, 21, (9), 1573-1587. [17] Chu, C.-C.; Katsarava, R. US Patent # 7304122 B2, 2007. [18] Graham, L. D.; Haggett, K. D.; Jennings, P. A.; Le Brocque, D. S.; Whittaker, R. G.;
Schober, P. A. Biochemistry 1993, 32, (24), 6250-6258. [19] Folin, O.; Ciocalteu, V. J. Biol. Chem. 1927, 73, (2), 627-650. [20] Fan, Y.; Kobayashi, M.; Kise, H. J. Polym. Sci., Part A: Polym Chem. 2001, 39, (9),
1318-1328. [21] Fan, Y.; Kobayashi, M.; Kise, H. J. Polym. Sci., Part A: Polym Chem. 2002, 40, (3),
385-392. [22] Guo, K.; Chu, C. C. Biomacromolecules 2007, 8, (9), 2851-2861. [23] Castro, G. R. Enzyme Microb. Technol. 2000, 27, (1-2), 143-150.
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
77
6 Supporting information
Two-dimensional H,C-correlated NMR spectrum of the starting PEA. Mass spectra of the
peaks identified during enzymatic degradation of PEA with α-CT and PK. Details of LC-
ToF-MS method optimization for the separation and identification of the water soluble
degradation products generated by means of chemical degradation of PEA. This material is
available free of charge via the Internet at http://pubs.acs.org/.
6.1 Two-dimensional H,C-correlated spectrum (HSQC) of PEA
Figure S1 Two-dimensional H,C-correlated spectrum (HSQC) of the starting PEA. in d6-ethanol recorded on a Bruker Avance 400 MHz NMR spectrometer. symbol ‡ are assigned to NH groups of amide bonds. The numbering to each signal corresponds to the numbering of protons and carbon present in Figure 1 in the Chapter 3.
6.2 ESI-ToF-MS spectra of the identified peaks – enzymatic degradation
The chromatographic peaks for the respective monomers and oligomers in Figure 4 and 5
were identified by their ESI-ToF-MS spectra (Table 2). Figure S2-S5 show the m/z values.
Chapter 3
78
All the spectra showed protonated adducts. Sodium adducts were also observed due to the
presence of sodium ions in the incubating buffer.
Figure S2 The LC-ToF-MS spectra of peaks 1, 3, 4, 5, and 7 in positive mode. The spectrum for peak 8 was recorded in the negative mode.
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
79
Figure S3 The LC-ToF-MS spectra of peaks nos. 9-13 in positive mode.
Chapter 3
80
Figure S4 The LC-ToF-MS spectra of peaks nos. 14-17 in positive mode.
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
81
Figure S5 The LC-ToF-MS spectra of peak 6 (Table 2) in positive mode. (A) Benzyl alcohol spectra in the hydrolysate (B) XIC-chromatogram of benzyl alcohol in the hydrolysate (C) spectrum of Benzyl alcohol standard (D) XIC chromatogram of Benzyl alcohol standard.
6.3 Chemical Degradation – optimization of the LC-ToF-MS method
6.3.1 Chemical degradation
Approximately 20 mg of PEA were hydrolyzed at 120oC and 300 kPa pressure for 24 h
while stirring in a 10 mL glass vessels with 2 mL of a mixture of 1 M NaOH and ethanol
(75:25) in a microwave instrument (CEM Corporation, Methews, NC, USA)33. Similarly,
approximately 20 mg of PEA were mixed with 2 mL of a mixture of 3 M HCl and ethanol
(75:25) and hydrolyzed at 90oC and 300 kPa for 24 h in the same apparatus as described
above. Both hydrolysates were twofold diluted with LC-grade water and neutralized (pH ≈
7) with HCl or NaOH prior to their injection into the LC-ToF-MS. Similarly, PEA in both
mixtures were also hydrolyzed at room temperature.
Chapter 3
82
6.3.2 Chemical degradation: LC-ToF-MS analysis
Figure S6a displays the TIC chromatograms for the partially hydrolyzed products in 1 M
aqueous NaOH solution at room temperature. The peaks are labeled according the structural
assignments based on their ESI-ToF mass spectra (Table S1). The formation of leucine (L)
and 1,6-hexanediol (C6) indicates that the polymer is degraded primarily via ester
hydrolysis. The amide bond between L and sebacic acid (C10) appears to be more
susceptible to cleavage than the amide bonds between lysine (K) and C10. Peaks from 20.2
to 22.5 min represent LC10L and oligomeric blocks of (LC10(KC10)nL) with the number of
repeating units (n) ranging from 1 to 5 in order of increasing retention. Because ethanol was
added to enhance the solubility of the polymer in 1 M NaOH, ethylation may occur. This
was observed for various peaks. For example, peak EtLC10K is the ethylated product of
peak LC10K. Peaks from 22.5 to 26 min indicate the presence of ethylated products of peaks
LC10, LC10KC10,LC10L, and oligomeric series LC10(KC10)1-3 (eluting in reverse order,
as hydrophobicity increases upon ethylation). The ESI-ToF-MS spectra of peaks from 22.5
to 26 min show the co-elution of several singly ethylated products originating from LC10,
LC10KC10, and LC10(KC10)1-3L (Table S1).
The polymer degraded nearly completely to its monomers in 1 M NaOH at 120oC in the
microwave instrument (peaks 1, 3, 4 and 9) except for four fragments, namely C10K,
LC10K, LC10, and LC10L. No ethylated products were observed (Figure S6b).
In case of acid hydrolysis (3 M HCl, at room temperature) the oligomers were separated
according to the degree of ethylation i.e. without ethylation, once ethylated and twice
ethylated oligomers, and once ethylated oligomers containing C6 at the other side (Figure
S7a). Peaks EtLC10KEt, EtLC10(KC10)2Let, and EtLC10(KC10)1LEt reflect twice
ethylated oligomers originating from the compounds associated with peaks LC10K,
LC10(KC10)2L, and LC10(KC10)1L , respectively. Peaks EtLC10(KC10)2LC6,
EtLC10(KC10)1LC6, and EtLC10LC6 contain additionally one ethyl group and C6 in
comparison with the analytes of peaks LC10(KC10)2L, LC10(KC10)1L, and LC10L,
respectively. Thus, the degradation products form a complex mixture of oligomers
possessing different end-group functionalities (amine, alcohol, and carboxylic acid or ethyl
esters). Based on hydrophobicity each class is eluted in a retention order with specific
retention increments.
Monitoring the in vitro enzyme-mediated degradation of degradable PEA by LC-ToF-MS
83
K
L
C6C6L
C10
LC10
LC10L
L
Ile
C6BnOH
LC10K
EtLC10K
LC10
LC10KC10
LC10KC10K
LC10L
LC10(KC10)1LLC10(KC10)2L
LC10(KC10)3LLC10(KC10)4L
LC10(KC10)5L
(7) EtLC10(KC10)1-3L
(8) EtLC10(KC10)1-3L(9) EtLC10L
a
b
(6) EtLC10, EtLC10KC10
67 8
9
K
K
L
C6C6L
C10
LC10
LC10L
L
Ile
C6BnOH
LC10K
EtLC10K
LC10
LC10KC10
LC10KC10K
LC10L
LC10(KC10)1LLC10(KC10)2L
LC10(KC10)3LLC10(KC10)4L
LC10(KC10)5L
(7) EtLC10(KC10)1-3L
(8) EtLC10(KC10)1-3L(9) EtLC10L
a
b
(6) EtLC10, EtLC10KC10
67 8
9
K
Figure S6 TIC chromatograms of the PEA degraded in 1 M NaOH aqueous solution at (a) room temperature and (b) 120oC and 3 bar in a Microwave instrument.
Figure S7b shows the TIC chromatogram of the polymer degradation products after nearly
complete degradation in 3M HCl:ethanol (75:25) at 90°C. At high temperature and under
acidic conditions, esterification of the carboxylic-acid groups in L and C10 was observed
(peaks EtL and EtC10). However, only one carboxylic acid was esterified in case of C10.
Similarly, only one hydroxyl group of the C6 was etherified with ethanol to generate an ether
group (peak EtC6). It has been reported in literature that the esterification of carboxylic acid
with alcohols to produce esters is very slow at room temperature, but that direct microwave
heating of the materials greatly speeds up the reaction [Pipus, G.; Plazl, I.; Koloini, T.
Chemical Engineering Journal 2000, 76, (3), 239-245]. Lysine remained un-retained in all
the chromatographic separations under the applied conditions and no esterification of its
carboxylic-acid group was observed. The analysis shows only fragments which theoretically
could be derived from the polymer structure, which suggests a polymerization process
without side-reactions.
Chapter 3
84
L
L
C6EtL
C6L LC10KEtLC10K
LuC10KC10K
EtLC8KEt
LC8
LC10KC10
C6 EtC6
EtL
C10 EtC10
LC8LC6
K
LC10L
LC10(KC10)1-5L
1
23
4
5
EtLC10(KC10)2LC6
EtLC10(KC10)2LEt
EtLC10(KC10)1LC6
EtLC10(KC10)1LEt
EtLC10LC6
a
b
67 8
9
K
L
L
C6EtL
C6L LC10KEtLC10K
LuC10KC10K
EtLC8KEt
LC8
LC10KC10
C6 EtC6
EtL
C10 EtC10
LC8LC6
K
LC10L
LC10(KC10)1-5L
1
23
4
5
EtLC10(KC10)2LC6
EtLC10(KC10)2LEt
EtLC10(KC10)1LC6
EtLC10(KC10)1LEt
EtLC10LC6
a
b
67 8
9
K
Figure S7 TIC chromatograms of the PEA degraded in 3 M HCl aqueous solution at (a) room temperature and (b) 90oC and 3 bar in a Microwave instrument.
A. Ghaffar, G. J. J. Draaisma, G. Mihov, P.J. Schoenmakers, Sj. van der Wal, to be submitted.
Chapter 4
4. A versatile system for studying the enzymatic degradation of
multi-block poly(ester amide)s*
The suitability of biomaterials for specific biomedical applications can be investigated
through their in-vitro biodegradability with selected enzymes and through their degradation
kinetics. A system was developed for studying the enzymatic degradation of poly(ester
amide) (PEA) coatings under sink conditions, with on-stream analysis of degradation
products by liquid chromatography coupled to time-of-flight mass spectrometry (LC-ToF-
MS). A coated capillary was treated by an enzyme solution in pulses (pulse-feed mode) or
continuously (continuous-feed mode) with different flow rates. The water-soluble products
resulting from the interaction of enzyme with the PEA coating were deposited on-line on a
reversed-phase LC column, separated by gradient-elution LC, ionized by electrospray-
ionization (ESI), and identified based on ToF-MS data.
The experiments underline the benefits of the experimental set-up, which requires only small
amounts of coating and enzyme and produces detailed results rapidly. The system was
investigated using different injection volumes (pulses) of an α-Chymotrypsin solution in
varying concentrations, different flow rates, and different lengths of coated capillary. The
versatility of the system makes it easy to follow the course of degradation and to
differentiate between primary and secondary degradation products. The system was applied
to study the degradation of a di-block and a tri-block PEA. Specific degradation products
showed different time profiles than the (more gradual) overall weight loss. Continuous-feed
mode analysis allowed the convenient determination of highly stable amide-bond-containing
fragments, while pulse-feed mode analysis revealed benzyl ester-containing products as
primary degradation products.
Chapter 4
86
1 Introduction
Biodegradable polymeric implants that degrade gradually and that yield degradation
products that are excreted benignly by the body prevent surgical re-interventions to remove
them after their role (e.g. as drug-delivery carrier) has subsided [1, 2]. However, the
suitability of such materials for biomedical applications requires an extensive evaluation in
term of their biocompatibility (tissues response), mechanical properties and, most
importantly, their degradation behaviour [3]. Chemical and enzymatic hydrolysis are the
primary biodegradation mechanisms for such materials. The highly reactive species
produced during the foreign-body response may degrade the polymer chain and contribute to
the overall degradation of biomaterials [2, 4]. The hydrophilic or hydrophobic nature of the
polymeric chains can affect the degradation rate and mechanism. Bulk erosion and surface
erosion are typically distinguished [5]. Poly(ester amide)s (PEAs) have attracted the
attention of the biomedical field for temporary implants due to their good physical
properties, biocompatibility and controlled degradability [6, 7]. PEAs contain ester bonds
susceptible to both hydrolytic and enzymatic degradation, while the amide bonds are prone
to degradation by enzymes.
To optimize the use of such biomaterials for specific biomedical applications and to assess
the potential risks of intermediate and final degradation products, it is crucial to understand
their degradation kinetics. The amounts and toxicological nature of the degradation products
define the acceptability of a biodegradable device by the biological environment [8]. The
kinetics of the enzymatic degradation of biomaterials are conventionally studied first under
in vitro conditions by incubating the specimen in a medium containing enzyme at 37oC with
or without agitation. Such a batch-mode study does not closely mimic the in vivo conditions.
Therefore, it does not allow rigorous modelling of the kinetics. However, the selection of
appropriate enzymes, incubation media, surface-to-volume ratios and duration of the
experiment may help to approach the physiological conditions where the biomaterial is
implanted, [3]. The degradation of degradable synthetic polymeric devices is also influenced
by the static or dynamic conditions of the media [9]. Agrawal et al. proposed the schematics
of an apparatus, which they used to study the degradation of biodegradable scaffolds based
on copolymers of polylactic acid (PLA) and polyglycolic acid (PGA) under dynamic (fluid-
flow) conditions. They reported a decrease in degradation rate under dynamic as compared
to static conditions [9]. Gorman et. al. studied the encrustation of urinary-tract devices based
on polyurethanes, Percuflex® and silicone under static and dynamic conditions with
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
87
artificial urine in a reaction vessel. Significantly higher levels of calcium and magnesium
were observed in the static mode, but the level of encrustation was the same in both cases
[10]. In another study, a dynamic simulated system was developed to study the effect of fluid
flow on the degradation of poly(lactide-co-glycolide acid) (PLGA) in Hank’s simulated body
fluid (SBF) for 30 days. Slower degradation was observed in case of dynamic conditions
[11]. On the other hand, porous scaffolds prepared from poly(L-lactic acid)/β-tricalcium
phosphate (PLLA/ β-TCP) composite, when degraded with phosphate-buffered-saline (PBS)
solution in a dynamic loading set-up, showed significantly faster degradation than in static
incubation [12].
During static studies high concentrations of degradation products may affect the activity of
an enzyme and oligomeric products released as a result of surface erosion are prone to
further degradation in the enzyme-containing media due to long incubation times [7].
Therefore, we decided to develop an on-stream-analysis system to study and explore these
synergistic effects. We constructed a small, flexible experimental set-up to (i) quickly
establish the effectiveness of a particular enzyme towards biodegradable polymer coatings
using small amounts of enzyme, (ii) assess the preference of a selected enzyme for ester or
amide bonds, (iii) allow on-line mass-spectrometric analysis of the intermediate and final
degradation products, with the aim of estimating their toxicological nature.
In the developed on-line system the enzyme solution passes through a capillary coated with
PEA either in continuous-feed mode or in pulse-feed mode. The introduced enzyme
degrades the surface of the polymer under flow conditions and the degradation products are
collected along with enzyme on a reversed-phase octadecyl-silane (C18) column. Attaching
this column to a gradient pump by switching a valve allows the chromatographic separation
of the degradation products and their identification with ESI-ToF-MS. A class of multi-block
PEAs, composed of sebacic acid, 1,6-hexanediol, lysine-benzyl ester and leucine with two
different sequences of repeat units (Scheme 1), was used in this study. The experimental
system was optimized by varying the concentration of enzyme, injection (pulse) volume,
switching time, flow rate of enzyme through the coated capillary, and length of the coated
capillary. The generated data were used to assess the effects of these parameters on the rate
of degradation of PEA and the generation of specific degradation products. The applicability
of the system was ultimately demonstrated for the degradation of a tri-block PEA containing
additionally isosorbide (1,4-dianhydrosorbitol (DAS)) under both continuous-feed and pulse-
feed conditions.
Chapter 4
88
Scheme 1 Structures of (A) di-block poly (ester amide) based on sebacic acid (i.e. y=4), 1,6-hexanediol (x=3), lysine-benzyl ester and leucine. For di-block PEA, m ≈ 3×n. (B) tri-block poly(ester amide), additionally containing isosorbide (1,4-dianhydrosorbitol (DAS)). The solid and dashed arrows represent the possible cleavage of ester and amide bonds upon enzymatic degradation with α-CT, respectively. In case of tri-block PEA, m:n:o = 6:5:9.
The developed system provides a convenient approach and a miniaturized system to study
the kinetics under sink conditions. The low enzyme activities required to use this on-line
system widen the scope of degradation studies, allowing a broader range of appropriate (but
possibly expensive or with a limited availability) enzymes to be tested.
2 Experimental
2.1 Materials
Both the di- and tri-block PEAs (Scheme 1) were synthesized at DSM following an open
literature procedure [13]. The chemical structure is shown in Scheme 1, where x = 3, y = 4.
The structure and the molar ratios between different structural components were confirmed
by 1H NMR and HSQC (heteronuclear single quantum coherence) NMR experiments [7] (for
details see the supporting information). For di-block PEA, m 3n; for tri-block PEA, m:n:o
6:5:9. Solubility in a range of common solvents was assessed (see the supporting
information). The molar-mass distributions were determined by size-exclusion
chromatography with differential-refractive-index detection (SEC-dRI) in tetrahydofuran
(THF). The molecular weight and dispersity of the tri-block PEA are 56 kDa and 2,
respectively (see the supporting information). The enzymatic degradation of coated
capillaries was carried out by protease α-Chymotrypsin (α-CT) from bovine pancreas (Fluka,
A
B
HN
HN
OO O
n
O
NH
O
OO
OHN
O
O
y x ym
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
89
Steinheim, Germany, pr. C4129, > 40 units/mg protein). Different concentrations of α-CT
were prepared in PBS buffer (0.2 g KCl, 0.2 g KH2PO4, 1.15 g Na2HPO4, 8 g NaCl in 1 L
demineralized water, containing 0.5 g/L sodium azide to inhibit bacterial growth) of pH ≈ 8.
2.2 Dynamic coating of stainless-steel capillaries
The PEA coatings were applied to pre-weighed, dried and cleaned stainless-steel capillaries
(i.d. = 1.15 or 0.98 mm, length = 40, 60, 80 or 100 mm) by drawing the polymer
formulations (0.1 g/mL solution in ethanol) through the capillary at a flow rate of 5 μl/min
with a syringe pump using withdrawal/injection mode. After passing four capillary volumes
back and forth the coated capillaries were placed on a rolling bar mixer (Stuart SRT9D,
Keison Products, Chelmsford, Essex, U.K.) at 40 rmp for 1 h, to homogenize the thickness
of the coating. Subsequently, the coated capillaries were dried at 40oC while flushing with a
very gentle stream of nitrogen gas (approximately one bubble/s, tested in water) for 24 h.
The coating procedure was optimized by cutting the coated capillary (with known amount of
coating) in four pieces of equal lengths. The amounts of coating were calculated in each part
by weighing before and after the coatings were removed. Approximately the same amount of
coating was obtained from each part with a relative standard deviation (r.s.d) of 10%.
Katsarava [14, 15] reported that polymers containing amino acids with fatty lateral
substituents (e.g. those possessing –CH2-CH(CH3)2 groups, such as PEAs based on leucine)
showed contraction of the film upon exposure to PBS during biodegradation. Therefore,
film-thickness measurements could be misleading and these were not taken into account.
The results obtained by coating capillaries of different lengths are given in Table 1. The
coating weight was always about 20 µg/mm2, except for the shortest capillary.
Table 1 Data obtained upon coating capillaries of different length and diameter with PEA.
Number L of capillary
(mm)
ID of capillary
(mm)
Surface area of
capillary (mm2)*
PEA coating
(mg)
coating
(g/mm2)
n RSD
(%)
1 40 0.98 125.7 1.61 13 6 13.7
2 60 0.98 188.6 3.53 19 4 11.4
3 80 0.98 251.4 5.03 20 4 8.2
4 100 0.98 314.3 6.45 21 4 11.1
5 107 1.15 386.7 8.09 21 4 7.7
*Calculated.
Chapter 4
90
2.3 On-line LC-ToF-MS analysis
An Agilent 1100 series HPLC system consisting of a degasser, an isocratic pump, a gradient
pump, an auto-sampler (all from Agilent Technologies, Waldbronn, Germany) and a column
oven (Waters Temperature Control Module, Milford, MA, USA) set at 37oC was used for the
on-line analyses. The system was configured by connecting the isocratic pump and a PEAs
coated capillary in series to port 6 of a 2/6 micro-switching valve (Agilent, Waldbronn,
Germany). The loop of the valve, connected at port 1 and 4, contained a Zorbax XDB C18
column (150 × 4.6 mm i.d., 5 µm particle size) (Agilent Technologies, Wilmington, DE,
USA). Port 2 and 3 were connected to the gradient pump and to a 6210 series ESI-ToF-MS
(Agilent Technologies, Waldbronn, Germany), respectively. Figure 1 describes the system
for the two positions of the valve. Port 5 was used for waste in either position.
The system thus configured was used for an on-line study of the enzymatic degradation of
PEA coatings in two different modes. (i) Continuous-feed mode involves continuous
pumping of enzyme solution through the coated capillary at different flow rates. The release
of degradation products is monitored by collecting these through switching the valve at
given time intervals during the course of the degradation. (ii) Pulse-feed mode, in which the
isocratic pump is pumping water at different flow rates through the coated capillary. Enzyme
solutions are injected using sandwich injection [16], comprising of 10 µl PBS plugs before
and after the enzyme plug in valve position A. The contact time of the enzyme in the coated
capillary (at 37oC) was varied by changing the flow rate of the isocratic pump from 10 to
100 µL min-1. The enzymatic degradation products were collected at the top of the C18
column. Subsequently, the valve was switched from position A to B for gradient elution of
the degradation products.
The separations of the adsorbed degradation products on the C18 column were carried out by
running a gradient from 5% (v/v) B at t = 0 min to 60% (v/v) B at t = 25 min, held until t =
27 min and then back to 5% (v/v) B at t = 30 min (tend = 35 min after a 5-min final hold). The
flow rate was 1.5 mL min-1 and the flow was split post-column between the waste reservoir
(approximately 1.3 mL min-1) and the electrospray-ionization (ESI) interface (approximately
0.2 mL min-1) by means of a zero-dead-volume T-piece. 0.1% (v/v) aqueous formic acid
(Fluka) (mobile phase A) and acetonitrile (HPLC grade, Biosolve) (mobile phase B) were
used to constitute the gradient. Highly pure water for mobile-phase preparation was obtained
by means of an Arium® 611 Ultrapure (18.2 MΩ*cm) Water System (Sartorius AG,
Goettingen, Germany). The sensitivity of ESI-ToF-MS proved different from day to day.
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
91
Therefore, for each set of experiments a leucine standard was injected during the run to
adjust for differences in sensitivity. Blank experiments without coating were performed to
determine the enzyme-related background.
1
6
5 4
3
2
37oC
Waste
A B
Internally coated stainless steel tube
TOF-MS
C18 Column 37oC
Grad-Pump
Water
Iso-Pump
α-CT
Auto-sampler
A
1
6
5 4
3
2
37oC
Waste
A B
Internally coated stainless steel tube
TOF-MS
C18 Column 37oC
Grad-Pump
Water
Iso-Pump
α-CT
Auto-sampler
A
1
6
5 4
3
2
37oC
Waste
A B
Internally coated stainless steel tube
Iso-Pump
TOF-MS
Grad-Pump
α-CT
Auto-sampler
B
C18 Column 37oC
Water
1
6
5 4
3
2
37oC
Waste
A B
Internally coated stainless steel tube
Iso-Pump
TOF-MS
Grad-Pump
α-CT
Auto-sampler
B
C18 Column 37oCC18 Column 37oC
Water
Figure 1 Schematic diagram of the on-line testing for the enzymatic degradation of PEAs coatings. (A) Position A; injected enzyme (in PBS) comes in contact with the coatings in the capillary; the degradation products are adsorbed on the C18 column (present in the loop); (B) Position B; Gradient is run to separate the degradation products and to identify them by ESI-ToF-MS. Iso-pump and Grad-pump represents isocratic and gradient-elution pumps, respectively. The auto-sampler is used to inject pulses of enzyme-containing solution in case of pulse-feed mode. In case of continuous-feed mode the cooled enzyme solution in PBS was pumped by the iso-pump.
The HPLC system was hyphenated with a 6210 series Time-of-Flight Mass Spectrometer
(Agilent Technologies, Waldbronn, Germany) via an ESI interface. The conditions of the
ESI-ToF-MS were as follows: drying gas was nitrogen (N2) at 8 L min-1, 300oC and 200kPa.
The capillary voltage, fragmenter and skimmer, were set at 3500 V, 140 V and 60 V,
respectively. The octopole dc1 and octopole radio frequency were set at 33 V and 250 V,
respectively. The data were acquired in the scan mode from m/z 50 to 3000 with 0.88
Chapter 4
92
scans/sec. An Agilent MassHunter Workstation A.02.01 and AnalystTM QS 1.1 software
(Applied Biosystems) were used for data acquisition and data analysis, respectively.
Figure 2 Selected TIC chromatograms the degradation products of (A) di-block PEA and (B) tri-block PEA obtained in continuous-feed mode with 2 min switching time (20 μL) with α-CT (0.025 mg/mL, 1 U/mL in PBS); iso-pump flow rate 10 μL/min. The peak numbers correspond to the structural assignments based on m/z values in Table 2 and Schemes 2 and 3. Peaks with * represent the α-CT peak in each chromatogram (see the supporting information for ESI-ToF-MS spectrum). Symbol § represents the switching time from position A to B (Figure 1).
3 Results and discussion
3.1 Continuous-feed mode
Figure 2a shows the degradation products of di-block PEA (Scheme 1) generated as a result
of continuous-feed enzymatic degradation. Table 2 lists the m/z values and the structures
assigned to each peak. All combinations of LC10 and KC10 contained an amide bond
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
93
between respective amino acid and C10. In case of K(Bn), the carboxylic acid group of the K
is end-capped with benzyl alcohol and forms a lysine-benzyl ester bond. The carboxylic acid
group of L is connected to C6 via ester bond in all LC6 combinations. For the optimization
of the continuous-feed mode peaks 8, 9, 10, 14, and 18 (Scheme 2) were used. In this mode
the enzyme was dissolved in PBS and the iso-pump (Figure 1) was continuously pumping
the enzyme solution through the coated capillary. After running for an hour at the constant
flow-rate used, the valve was switched from position B to A and then back to position B
(Figure 1) to direct a specified volume of the effluent (containing water-soluble degradation
products as a result of enzymatic degradation in the coated capillary) to the C18 column. By
changing the switching time different volumes of the solution of degradation products were
loaded on the C18 column before analysis by gradient-elution LC (Figure 3a).
0
5000000
10000000
15000000
20000000
25000000
30000000
35000000
40000000
45000000
50000000
0 5 10 15 20 25 30 35 40 45
Inj. Vol. (μL)
Pea
l are
a (c
ps
)
0
200000
400000
600000
800000
1000000
1200000
1400000
1600000
Pea
k are
a (c
ps
)
0
2000000
4000000
6000000
8000000
10000000
12000000
0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0 4.4
Reaction time (min)
Pe
al a
rea
(cp
s)
0
50000
100000
150000
200000
250000
Pea
k are
a (c
ps)
A
B
0
5000000
10000000
15000000
20000000
25000000
30000000
35000000
40000000
45000000
50000000
0 5 10 15 20 25 30 35 40 45
Inj. Vol. (μL)
Pea
l are
a (c
ps
)
0
200000
400000
600000
800000
1000000
1200000
1400000
1600000
Pea
k are
a (c
ps
)
0
2000000
4000000
6000000
8000000
10000000
12000000
0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0 4.4
Reaction time (min)
Pe
al a
rea
(cp
s)
0
50000
100000
150000
200000
250000
Pea
k are
a (c
ps)
A
B
Figure 3 (A) Peak areas of peak 8 (♦) and peak 9 () (left-hand-axis) and peak 14 () and peak 18 () (righ-hand axis) vs. injection volume (calculated from switching times) of α-CT solution (0.025 mg/mL, 1 U/mL) passing at a flow rate of 10 μL/min through the coated capillary. (B) Peak areas of peak 8 (◊), peak 9 (∆), peak 14 (), and peak 18 () as a function of flow rate for passing 10 µL of α-CT solution (0.025 mg/mL, 1 U/mL) through the coated capillary. Capillary dimensions are L = 40 mm, i.d. = 1 mm.
Chapter 4
94
The results indicated that the linearity of the ESI-MS detection was acceptable. Similarly,
identical injection volumes (10 µL) were loaded with different flow rates of α-CT solution
(0.025 mg/mL, 1 U/mL) (Figure 3b). The contact (direct reaction) time decreased with
increasing flow rate. Contact times up to 4 min led to a proportional increase of the
concentration of peaks 8 and 9 at the selected enzyme concentration. Peaks 14 and 18
showed opposite release profiles as the contact time increased. This indicates further
degradation at longer contact times of C6LC10L (ester bonds between L and C6 in peak 14)
and LC10[K(Bn)C10]1L (ester bond between Bn and K in peak 18), respectively. Previous
results showed that the ester bond in the structure that gives rise to peak 14 is more stable
than that of peak 18, since no benzyl ester peak was observed in case of batch-mode analysis
[7]. An injection volume of 10 µL and a flow rate of 10 µL/min were selected for further
experiments.
To study the effect of enzyme concentration, the degradation of the PEA coating was
monitored with α-CT solutions with concentrations of 0.5, 1, 2, and 5 U/mL for 24 h at a
flow rate of 10 µL/min. After every 30 min, 10 µL of the effluent of the PEA coated
capillary were loaded on the C18 column and characterized by gradient-elution LC-MS.
Figure 4a shows a decrease in the peak area of peak 8 (Figure 2) over a period of 24 h of
32% at 5 U/mL and 22% at 0.5 U/mL. The decrease in peak area is due to a decrease in
enzyme activity, the faster decrease at higher enzyme concentration likely indicates that the
binding sites of the enzyme saturate the surface area of the coating [17]. Figure 4b and 4c
show the rate of formation of different degradation products. Again, peak 14 and 18 showed
a non-monotonous, contrasting profile due to further hydrolysis.
After 24 h degradation each capillary was dried in an oven at 40oC while flushing with
nitrogen, to calculate the percentage weight loss for different enzyme concentrations. A non-
linear increase in the percentage weight loss was observed with an increasing enzyme
concentration. The on-line system in its current mode of operation appears to be very
efficient for determining normalized relative degradation rates (cf. Figure 4d). A very small
amount of enzyme (here 0.2 U α-CT) is required, because of the low mass detection limits of
LC-ESI-ToF-MS.
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
95
0100000200000300000400000500000600000700000800000900000
1000000
0 1 2 3 4 5 6
[Eo] U/mL
Ra
teo
ffo
rma
tion
(cp
s/h
ou
rs)
0
5000000
10000000
15000000
20000000
0.5 3.0 5.5 8.0 10.5 13.0 15.5 18.0 20.5 23.0
Time (hours)
Pe
ak
are
a (
cps)
0
5000000
10000000
15000000
20000000
25000000
30000000
35000000
40000000
45000000
0 1 2 3 4 5 6
[Eo] U/mL
Rat
e of
for
mat
ion
(cps
/hou
rs)
0
100
200
300
400
500
600
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
[E] (U/ml)
loss
(ng/
day/
U/m
m2 ) flow
batch
A
BD
C
0100000200000300000400000500000600000700000800000900000
1000000
0 1 2 3 4 5 6
[Eo] U/mL
Ra
teo
ffo
rma
tion
(cp
s/h
ou
rs)
0
5000000
10000000
15000000
20000000
0.5 3.0 5.5 8.0 10.5 13.0 15.5 18.0 20.5 23.0
Time (hours)
Pe
ak
are
a (
cps)
0
5000000
10000000
15000000
20000000
25000000
30000000
35000000
40000000
45000000
0 1 2 3 4 5 6
[Eo] U/mL
Rat
e of
for
mat
ion
(cps
/hou
rs)
0
100
200
300
400
500
600
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
[E] (U/ml)
loss
(ng/
day/
U/m
m2 ) flow
batch
A
BD
C
Figure 4 (A) Decrease in the peak area of peak 8 for 10 μL α-CT injected at different intervals of time for a coated capillary (L = 40 mm, ID = 1 mm, surface-to-volume ratio 4.1 mm2/μL) flushed at a flow rate of 10 µL/min with four different α-CT concentrations 0.5 (), 1 (♦), 2 (), and 5 () U/mL. (B and C) Rate of formation with respect to peak 8 (◊), peak 9 (), peak 10 (), peak 14 (∆) and peak 18 (). (D) normalized weight loss during 24 h as a function of enzyme concentration with continuous-feed mode (♦) and batch-mode analysis () [7].
It is clear that the specific degradation (Figure 4b and 4c, peaks 8, 9, 10, 14, 18) and the
overall degradation (figure 4d, weight loss) are rendering different information. At high
enzyme concentrations the normalized weight loss per unit enzyme slows down twice as fast
as the formation of degradation product LC10L (peak 8) (and at different contact times).
Thus, these two effects are not directly correlated. The results may be compared with the
approximately 1% weight loss of PEA coating with 17 U/mL of enzyme in 24 h observed in
a batch-mode study [7].
The effect of the surface area on the rate of degradation was studied by delivering α-CT (1
U/mL) at a flow rate of 10 µL/min to capillaries with different lengths (L = 40 mm, 60 mm,
80 mm, 100 mm) but constant internal diameter (1 mm) for 24 h. After every 30 min, 10 µL
of the effluent was loaded on the C18 column for gradient-elution LC.
Chapter 4
96
0
5000000
10000000
15000000
20000000
25000000
30000000
35000000
40000000
0.0 4.0 8.0 12.0 16.0 20.0 24.0
Time (hours)
Pe
ak
are
a(c
ps)
0
500000
1000000
1500000
2000000
2500000
3 4 5 6 7 8 9 10 11
Capillary Lenght (cm)
Rat
eof
form
atio
n(c
ps/
hou
rs)
490
500
510
520
530
540
550
560
570
580
100 120 140 160 180 200 220 240 260 280 300 320 340
Surface area (mm²)
Loss
(n
g/m
m²/U
/24hr)
0
10000000
20000000
30000000
40000000
50000000
60000000
70000000
3 4 5 6 7 8 9 10 11
Capillary Lenght (cm)
Rat
e of
for
mat
ion
(cps
/hou
rs)
A
BD
C
0
5000000
10000000
15000000
20000000
25000000
30000000
35000000
40000000
0.0 4.0 8.0 12.0 16.0 20.0 24.0
Time (hours)
Pe
ak
are
a(c
ps)
0
500000
1000000
1500000
2000000
2500000
3 4 5 6 7 8 9 10 11
Capillary Lenght (cm)
Rat
eof
form
atio
n(c
ps/
hou
rs)
490
500
510
520
530
540
550
560
570
580
100 120 140 160 180 200 220 240 260 280 300 320 340
Surface area (mm²)
Loss
(n
g/m
m²/U
/24hr)
0
10000000
20000000
30000000
40000000
50000000
60000000
70000000
3 4 5 6 7 8 9 10 11
Capillary Lenght (cm)
Rat
e of
for
mat
ion
(cps
/hou
rs)
A
BD
C
Figure 5 (A) Decrease in the peak area of peak 8 for 10 μL α-CT injected at different intervals of time for different lengths of coated capillary 40 (♦), 60 (), 80 (), and 100 () mm (each i.d. = 1 mm) flushed at flow rate of 10 µL/min with α-CT solution (1 U/mL). (B, C) The rate of formation with respect to peak 8 (◊), peak 9 (), peak 10 (), peak 14 (∆) and peak 18 (). (D) The normalized hydrolysis rate during 24 h as a function of capillary area.
Figure 5a shows a gradual decrease in the peak area of peak 8 (Figure 2) for each coated
capillary with time. The rate of formation of peak 8, 9 and 10 increases almost linearly with
an increase in capillary length (Figure 5b). The weight loss of PEA coating as a function of
surface area (varied through the capillary length) is shown in Figure 5d. There is a slight
gradual decrease in the weight-loss per unit area as the length of the capillary increases.
However, not all peak areas increase regularly with capillary length. This is most obvious
from the release profiles of peaks 14 (C6LC10L) and 18 (LC10[K(Bn)C10]1L) in Figures
3b, 4c, and 5c. It can be concluded that these instable products showed deviant behaviour
when varying the contact time, enzyme concentration, and capillary length, respectively.
Therefore, the rate of formation of different degradation products cannot be equated with the
overall weight loss of the polymer.
The coatings were also flushed with PBS for 24 h to study its contribution to the degradation
of the PEA coating. No water-soluble products were observed in these latter experiments.
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
97
O
(CH3)2HC
ONH
O
O
HN
4HO 3OH
CH(H3C)2
O
HO
(CH3)2HC
ONH
O
O
HN
4 OH
CH(H3C)2
O
H2N NH
HO O
2
O
O
OH4
H2NOH
CH(H3C)2
O
OH
H2NHN
2
HO OO
O
NH
OH
CH(CH3)2
O4 HO
O
O
NH
OH
CH(CH3)2
O4
O
(CH3)2HC
ONH
O
O
OH4HO 3
OHHO 3
H2N NH2
HO O
2
O
(CH3)2HC
ONH
O
O
HN
HN
4 2
O O
*
m
O 3
HN
CH(CH3)2
O
*
O
O4 n
3 24
56713
8
14
9 10
11
12
15
16
17
18 18
19
20
21
22
23 24
24
24
n = 1n = 2 n = 3
n = 4
O
(CH3)2HC
ONH
O
O
HN
4HO 3OH
CH(H3C)2
O
HO
(CH3)2HC
ONH
O
O
HN
4 OH
CH(H3C)2
O
H2N NH
HO O
2
O
O
OH4
H2NOH
CH(H3C)2
O
OH
H2NHN
2
HO OO
O
NH
OH
CH(CH3)2
O4 HO
O
O
NH
OH
CH(CH3)2
O4
O
(CH3)2HC
ONH
O
O
OH4HO 3
OHHO 3
H2N NH2
HO O
2
O
(CH3)2HC
ONH
O
O
HN
HN
4 2
O O
*
m
O 3
HN
CH(CH3)2
O
*
O
O4 n
3 24
56713
8
14
9 10
11
12
15
16
17
18 18
19
20
21
22
23 24
24
24
n = 1n = 2 n = 3
n = 4
Scheme 2 Degradation pattern of di-block poly(ester amide) based on sebacic acid, 1,6-hexanediol, lysine-benzyl ester and leucine, subjected to enzymatic degradation (see text for experimental conditions). The structures proposed are numbered according to their separation in Figures 2a and 6a. The m/z values are listed in Table 2.
Chapter 4
98
3.2 Pulse-feed mode
Figure 6a shows the separation of water-soluble products (mainly monomers and oligomers)
released as a result of enzymatic degradation of di-block PEA (Scheme 1A) during pulse-
feed analysis. The numbering of each peak corresponds to the list of structures deduced from
the m/z values (Table 2).
The switching times of the valve were optimized by recording the area of the α-CT peak (1
mg/mL) at different switching times using no capillary, an un-coated capillary (L = 107 mm,
i.d. = 1.15 mm) or a coated capillary of the same size, with a 13 µl capillary loop instead of
an LC column installed.
Figure 6 TIC chromatograms of the degradation products of (A) di-block PEA and (B) tri-block PEA obtained in pulse-feed mode with 20 μL injection volume of α-CT (0.125 mg/mL, 5 U/mL in PBS) at 100 μL/min iso-pump flow rate. The peak numbers correspond to the structural assignments based on m/z values in Table 2 and structures in Schemes 2 and 3. Peak marked * represents the α-CT peak in each chromatogram.
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
99
Figure 7a and 7b show the resulting profiles as a function of flow rate and configuration. The
40 µl injection broadens into a much wider band. This is monitored by sampling via the 13
µl capillary loop at different switching times. The (coated) capillary causes most of the band
broadening, while the system adds approximately 0.1 mL band width to the 40 µL plug of α-
CT injected. This increased band broadening (and accompanying dilution) is almost constant
across the range of flow rates studied and it is not affected by the coating on the capillary
wall. Despite the variation in concentrations, mutual comparisons can be made based on the
total enzyme activity, because the entire resulting plug is analyzed by LC. The peak areas of
oligomers were observed to increase linearly with the injection volume of α-CT.
0
500000
1000000
1500000
2000000
0 1 2 3 4 5 6 7 8 9 10
Switching time (min)
CT
-Pe
ak
are
a
05000000
10000000150000002000000025000000300000003500000040000000
0 5 10 15 20 25 30 35 40 45 50 55 60Time (hours)
Pe
ak
Are
a (
cps)
0
500000
1000000
1500000
2000000
0 2 4 6 8 10 12 14 16 18 20 22
Switching time (min)
CT
-Pe
ak
are
a
A
B
C
0
500000
1000000
1500000
2000000
0 1 2 3 4 5 6 7 8 9 10
Switching time (min)
CT
-Pe
ak
are
a
05000000
10000000150000002000000025000000300000003500000040000000
0 5 10 15 20 25 30 35 40 45 50 55 60Time (hours)
Pe
ak
Are
a (
cps)
0
500000
1000000
1500000
2000000
0 2 4 6 8 10 12 14 16 18 20 22
Switching time (min)
CT
-Pe
ak
are
a
A
B
C
Figure 7 (A) Peak areas of α-CT (40 μL, 1 mg/mL, 40 U/mL, injected with no (un)coated capillary installed) as a function of the switching time. Flow rates 100(),50() and 25() µL/min. (B) Band profile without capillary (), with uncoated () and coated () capillary. Flow rate 25 µl/min (C) areas of peaks 8 and 9 for 10 μL α-CT injected as a function of time. Capillary dimensions L = 107 mm, ID = 1.15 mm.
When 10 μL of α-CT was injected on the coated capillary after different time periods, a
decrease in the peak areas of water-soluble oligomers was observed. Figure 7c shows a
decrease in the areas of peaks 8 (LC10L) and 9 (LC10[KC10]1L) with increasing time. This
can be attributed to a decrease in enzyme activity in the supply vessel at room temperature.
The flow rate of the iso-pump to transport the enzyme plug and the length of the plug
determine the contact time of the enzyme with the coating in the capillary. Figure 8 shows
the peak areas of selected oligomers generated at different reaction times, when 20 µL of α-
Chapter 4
100
CT solutions of different activities were injected at different flow rates. As expected, the
formation rate for different types of degradation products is clearly different and strongly
dependent on enzyme concentration and reaction time. It appears that ester-containing
reaction products require a lower optimal enzyme concentration than other products (e.g.
diacids). Two effects are thought to contribute to this observation, viz. the preference of α-
CT for ester bonds and subsequent hydrolysis of the primary degradation products.
0
10000000
20000000
30000000
40000000
50000000
60000000
0 3 6 9 12 15
Reaction time (min)
Pea
k ar
ea (
cps)
0
2000000
4000000
6000000
8000000
10000000
0 3 6 9 12 15
Reaction time (min)
Pea
k ar
ea (
cps)
0
5000000
10000000
15000000
20000000
25000000
0 3 6 9 12 15
Reaction (min)
Pea
k ar
ea (
cps)
0
10000000
20000000
30000000
40000000
50000000
0 3 6 9 12 15
Reaction time (min)
Pea
k ar
ea (
cps)
B
C D
A
0
10000000
20000000
30000000
40000000
50000000
60000000
0 3 6 9 12 15
Reaction time (min)
Pea
k ar
ea (
cps)
0
2000000
4000000
6000000
8000000
10000000
0 3 6 9 12 15
Reaction time (min)
Pea
k ar
ea (
cps)
0
5000000
10000000
15000000
20000000
25000000
0 3 6 9 12 15
Reaction (min)
Pea
k ar
ea (
cps)
0
10000000
20000000
30000000
40000000
50000000
0 3 6 9 12 15
Reaction time (min)
Pea
k ar
ea (
cps)
B
C D
A
Figure 8 Change in the areas of peaks 8 (A) 14 (B) 9 (C) 21 (D) with concentration of α-CT enzyme. 1 U/mL (♦), 5 U/mL (), 10 U/mL (). Reaction time was varied by changing the flow rate of iso-pump from 10 to 100 µL/min.
The effect of PBS on the degradation of PEA coating was also tested in pulse mode by
injecting 40 µL of PBS solution after every 30 min for 24 h. No water-soluble products were
observed. When different α-CT concentrations (1U, 5U, and 10 U/mL) were injected at
different flow rates (10, 25, 50, 75, and 100 μL/min), the peak areas of amide-bond-
containing fragments (peaks 8 to 12) increased with a decrease in flow rate. However, the
peak areas of the diol- and benzyl ester-containing fragments (peaks 14 and 21, respectively)
showed an optimum reaction time (flow rate). This suggests that at lower flow rates (longer
contact times) products such as those reflected in peaks 14 and 21 convert to products such
as those of peaks 8 and 9 due to further ester hydrolysis (Figure 8).
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
101
3.3 Comparison of pulse-feed mode and continuous-feed mode
The on-line system was tested in both the pulse-feed and continuous-feed modes for the
enzymatic degradation of PEA, coated in capillaries, with α-CT as the enzymatic model. In
continuous-feed mode, the coated capillary was continuously flushed with enzyme solution
and a certain volume of post-capillary degradation solution was injected in the LC column
by switching from position B to A and then back to position B (Figure 1). Under the
dynamic sink conditions used, the benzyl ester containing degradation products released
from the surface of the coating to the flowing solution were degraded further. Peaks 8 to 12,
the oligomers containing amide bonds originating from the n block of the PEA (Scheme 1),
were stable under these dynamic conditions. This continuous-feed mode of analysis can be
used to study the weight loss of PEA as a result of enzymatic degradation under flow
conditions with on-line detection of water-soluble degradation products. The results obtained
in this mode showed a faster normalized degradation rate of PEA as compared to the batch-
mode degradation experiments performed before [7].
In pulse-feed mode only a band of α-CT passes through the coated capillary. The moment
this band is injected, it starts diffusing in the surrounding PBS. Therefore its reaction time
and concentration can only be approximated and semi-quantitative results (for mutual
comparison) are obtained. A 40 μL plug is diluted about five-fold during passage through the
coated capillary. The separation of degradation products formed in this mode shows benzyl-
containing products (Figure 2a). Such products were neither observed in case of batch-mode
analysis of this PEA [7], nor in case of continuous-feed mode analysis, except for peak 18.
This shows that pulse-feed analysis makes it possible to determine the primary degradation
products. With this mode of analysis a number of enzymes can be tested in very small
amounts and in a short time for their ability to degrade a certain PEA coating and their
selectivity towards ester and/or amide bonds. Delivery of microliter volumes of active
enzyme can be realized from the cooled tray of an auto-sampler. In addition, the effect of
different peroxide-containing solutions (2 < pH < 9) and simulated body fluids on the
degradation of coatings can be studied. The information provided by the pulse-feed analysis
is ideally suitable to design long-term degradation experiments with selected enzymes.
Chapter 4
102
3.4 Application to tri-block PEA coatings
The system was applied to test the degradation of tri-block PEA coatings with α-CT in both
the continuous-feed and pulse-feed modes. Figure 2b shows the water-soluble products
released during continuous-feed mode analysis with α-CT (0.025 mg/mL, 1U/mL in PBS) at
a flow rate of 10 µL/min with 2 min switching time (20 µL loaded on C18 column). Peaks 8
to 12 represent the oligomeric series from the n block of the polymer. Benzyl ester and DAS-
containing peaks were not observed. The chromatogram differs from that obtained from the
di-block PEA (Figure 2a) in that peak 18 is absent in Figure 2b. The degradation of tri-block
PEA in the pulse-feed mode (Figure 8b) exhibits benzyl ester (peaks 18, 22, and 23), diol
(peaks 14, 15, and 21), and DAS (peaks 25, 26, and 27) containing fragments in addition to
stable amide-containing (peaks 8, 9, 10, 11) n-block oligomers. The ester bond associated
with DAS (o-block) also exhibits degradability under the current experimental conditions
with α-CT.
3 2
25
8
14
910
11
12
18
18
2122
23
n = 1 n = 2n = 3
n = 4
4
26
27
o = 1
3 2
25
8
14
910
11
12
18
18
2122
23
n = 1 n = 2n = 3
n = 4
4
26
27
o = 1
Scheme 3 Degradation pattern of tri-block poly(ester amide) based on sebacic acid, 1,6-hexanediol, lysine-benzyl ester, leucine and isosorbide (1,4-dianhydrosorbitol (DAS)) subjected to enzymatic degradation (experimental conditions see text). The structures proposed are numbered according to their separation in Figures 2b and 6b. The m/z values are listed in Table 2.
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
103
The major degradation patterns of the di- and tri-block PEA are given in Schemes 2 and 3,
respectively. The oligomers in Scheme 2 that are not mentioned in Scheme 3 are present in
low concentrations. The schemes differ through the presence of isosorbide-containing
oligomers in Scheme 3, originating from the third block of the tri-block PEA. The m:n block
ratio is clearly reflected in the concentration ratios of the respective oligomers in Figures 6a
and 6b.
Table 2 Structural assignments for the water soluble fragments (monomers and oligomers), separated as shown in Figure 2a and 2b as a results of enzymatic degradation during pulse feed mode and continuous feed mode analysis, respectively. K, K(Bn), L, C6 and C10 represent lysine, lysine-benzyl ester, leucine, 1,6-hexanediol and sebacic acid, respectively.
Peak No.
Symbol No. of Amide bond broken m/z [M+H]+
1 formic acid
2 L 1 132.11
3 C6 0 119.11
4 Benzyl alcohol 0 91.055 [M-OH]+
5 KC10 1 331.23
6 LC10 1 316.21
7 C10LC6 1 416.31
8 LC10L 0 429.31
9 LC10[KC10]1L 0 741.51
10 LC10[KC10]2L 0 1053.71
11 LC10[KC10]3L 0 1365.92
12 LC10[KC10]4L 0 1679.12
13 KC10L 1 443.36
14 LC10LC6 1 529.39
15 C6LC10[KC10]1L 0 841.59
16 LC10[K(Bn)C10]1 1 718.46
17 C6LC10K(Bn)C10K 1 946.58
18 LC10[K(Bn)C10]1L 0 831.54
18 LC10[K(Bn)C10]-[KC10]L 0 1143.78
19 C6LC10[KC10]2 0 1039.77
20 LC10[K(Bn)C10]-[KC10]LC6 0 1243.86
Chapter 4
104
21 LC10[K(Bn)C10]1LC6 0 931.65
22 LC10[K(Bn)C10]2-[KC10]L 0 1546.03
23 LC10[K(Bn)C10]2L 0 1233.81
24 LC10[KC10]3 1 1251.87
24 LC10[KC10]2-[K(Bn)C10] 1 1341.86
24 LC10[K(Bn)C10]2-[KC10]LC6 0 1646.10
25 (DAS)LC10L 0 557.34
26 LC10K(Bn)C10L(DAS) 0 959.59
27 LC10L(DAS)LC10L 0 967.62
4 Conclusions
On-line analysis by LC-ToF-MS is useful to separate and identify the enzymatic degradation
products of PEAs coatings generated under dynamic sink conditions. Low flow rates of the
enzyme solution and longer coated capillaries enhance the extent of degradation. The
percentage weight loss increases gradually with an increase in α-CT activity, but it is hardly
affected by the capillary length. Normalized formation rates of degradation products depend
on enzyme activity and contact time and are highly compound specific, so that specific-
product profiles may be quite different from the overall weight-loss trend. Pulse-feed
analysis generated benzyl ester-containing degradation products, which were not observed in
continuous-feed analysis. The developed system allows testing of highly expensive enzymes
for their ability to degrade a particular novel PEA-based biomaterial. The biodegradation of
two types of PEA with a selected enzyme was investigated.
The use of very small amounts of enzymes, no sample preparation steps, conditions closer to
those of physiological degradation, and shorter analysis times are significant advantages of
the proposed system to study the degradation of biomedical materials and to identify the
nature of their degradation products. This is expected to result in better insight in the
underlying mechanisms.
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
105
5 References
[1] Aylvin A. Dias, M. Hendriks, Drug Delivery Technology 10 (2010) 20. [2] A. Ghaffar, P.G. Verschuren, J.A.J. Geenevasen, T. Handels, J. Berard, B. Plum,
A.A. Dias, P.J. Schoenmakers, S. van der Wal, Journal of Chromatography A 1218 (2011) 449.
[3] R.L. Reis, J.S. Román, Biodegradable Systems in Tissue Engineering and Regenerative Medicine, CRC Press, 2004.
[4] R.S. Labow, Y. Tang, C.B. McCloskey, J.P. Santerre, Journal of Biomaterials Science, Polymer Edition 13 (2002) 651.
[5] A.K. Burkoth, J. Burdick, K.S. Anseth, Journal of Biomedical Materials Research 51 (2000) 352.
[6] L. Castaldo, P. Corbo, G. Maglio, R. Palumbo, Polymer Bulletin 28 (1992) 301. [7] A. Ghaffar, G.J.J. Draaisma, G. Mihov, A.A. Dias, P.J. Schoenmakers, S. van der
Wal, Biomacromolecules 12 (2011) 3243. [8] A.-C. Albertsson, M. Hakkarainen, in Chromatography for Sustainable Polymeric
Materials -Renewable, Degradable and Recyclable ; Advances in Polymer Science 211, Springer, 2008, p. 85.
[9] C.M. Agrawal, J.S. McKinney, D. Lanctot, K.A. Athanasiou, Biomaterials 21 (2000) 2443.
[10] S.P. Gorman, C.P. Garvin, F. Quigley, D.S. Jones, Journal of Pharmacy and Pharmacology 55 (2003) 461.
[11] Y.-y. Huang, M. Qi, M. Zhang, H.-z. Liu, D.-z. Yang, Transactions of Nonferrous Metals Society of China 16 (2006) s293.
[12] Y. Yang, Y. Zhao, G. Tang, H. Li, X. Yuan, Y. Fan, Polymer Degradation and Stability 93 (2008) 1838.
[13] C.-C. Chu, R. Katsarava, in U.P. B2 (Editor), US Patent 7304122 B2, 2007. [14] G. Tsitlanadze, M. Machaidze, T. Kviria, N. Djavakhishvili, C.C. Chu, R. Katsarava,
Journal of Biomaterials Science, Polymer Edition 15 (2004) 1. [15] R. Katsarava, V. Beridze, N. Arabuli, D. Kharadze, C.C. Chu, C.Y. Won, Journal of
Polymer Science Part A: Polymer Chemistry 37 (1999) 391. [16] Y. Mengerink, R. Peters, M. Kerkhoff, J. Hellenbrand, H. Omloo, J. Andrien, M.
Vestjens, S. van der Wal, Journal of Chromatography A 876 (2000) 37. [17] K. Mukai, K. Yamada, Y. Doi, International Journal of Biological Macromolecules
15 (1993) 361.
Chapter 4
106
6 Supporting information
6.1 Solubility
The solubility of the tri-block PEA in water and in a number of common organic solvents
was assessed by combining 10 to 30 mg of the polymer with 0.5 mL of the respective
solvents (Table S1) at room temperature (20oC). The samples were vortexed and allowed to
dissolve overnight. The PEA was separated on a 250 x 4.6 mm Zorbax Eclipse XDB-C18
column from the solvent, using THF as a mobile phase and sandwich injection in the solvent
to prevent precipitation in the injection system.
Table S1 Solubility of the tri-block PEA used in the present study in common organic solvents evaluated at 25oC
solvents Solubility solvents Solubility solvents Solubility
Toluene - Tetrahydrofuran + Methanol +
Dichloromethane + Ethyl acetate - Acetic acid +
Isopropanol + Dimethylformamide + Formic acid +
Chloroform + Acetone -
N,N`-dimethyl acetamide
+ Acetonitrile - water -
Dimethyl sulfoxide + Ethanol +
Soluble +, insoluble -
6.2 Molecular weight (Mw) and dispersity of tri-block PEA
The SEC experiments were performed on an LC system equipped with a Waters 2590
Alliance separation module (Waters, Milford, MA, USA) and an RID-10A refractive-index
detector (Shimadzu, Kyoto, Japan). The SEC analyses were performed on three PLgel
MIXED-B columns (10 µm, 300 × 7.6 mm i.d.) (Polymer Laboratories, Church Stretton,
U.K.) connected in series. THF stabilized with butylated hydroxytoluene (BHT), (BioSolve,
Valkenswaard, The Netherlands) was pumped at a flow rate of 1 mL min-1. The injection
volume was 50 µL and the column-oven temperature was set at 50oC. Polystyrene standards
(Polymer Laboratories) were used to calibrate the SEC-UV-dRI system. Data were recorded
and chromatographic peaks were processed using Class-VP 7.4 software (Shimadzu). Molar-
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
107
mass distributions (MMD) were calculated from the chromatograms using software written
in-house in Excel 2003 (Microsoft).
Table S2 Peak molecular weight (Mp), weight-average molecular weight (Mw), number-average molecular weight (Mn), and dispersity (PDI) of the polystyrene standards. Data specified by the supplier (Polymer Laboratories).
Polystyrene Standards
Mp Mw Mn (PDI)
PS1 580 640 555 1.16 PS2 1310 1300 1220 1.07 PS3 2100 2100 2010 1.05 PS4 4920 4890 4740 1.03 PS5 9920 9910 9700 1.02 PS6 19880 19680 19220 1.02 PS7 30230 30500 29600 1.01 PS8 52400 51950 51050 1.02 PS9 70950 69200 67350 1.03
PS10 96000 94650 92350 1.03 PS11 126700 124100 121200 1.03 PS12 197300 201500 196800 1.02
PS13 299400 297100 292200 1.02
y = -0.00248x3 + 0.15813x2 - 3.71023x + 35.55551
R2 = 0.99985
0
5
10
15
20
25
30
35
40
45
50
17 18.5 20 21.5 23 24.5 26
Ret. Vol. (mL)
dR
I R
esp
on
se
0.000
1.000
2.000
3.000
4.000
5.000
6.000
Lo
g M
PS-1 PS-2 PS-3 PS-4 PS-5 PS-6 PS-7 PS-8
PS-9 PS-10 PS-11 PS-12 PS-13 PEA3
Figure S1 Size-exclusion separation of tri-block PEA (PEA3) and polystyrene standards (overlayed chromatograms).
Chapter 4
108
Table S3 Mw, Mn, and dispersity of tri-block PEA (Scheme1B)
Tri-block PEA Mw Mn PDI
I 55445 27100 2.05
II 56233 28360 1.98
III 56707 28128 2.02
Average 56128 27863 2
% RSD (n=3) 1 3 2
6.3 NMR experiments
Aproximately 20 mg of the sample was dissolved in 1 mL of d6-ethanol and 1H, 13C, 1H-1H
gCOSY (two-dimenssional homonuclear H, H gradient-correlated spectroscopy) and HSQC
(heteronuclear single quantum coherence) NMR experiments were recorded on a Varian
Inova 500-MHz NMR (Varian, Palo Alto, CA, USA) equipped with a 500 5mm 13C/31P/1H
GS probe.
2.2113.43
2.00
1.61
4.62
1.54
5.51
5.02
13,14
*
8
17
34
2
16, 712, 11
6
9
5
15
1, 1022, 23
*
*
21
18,20,23,
19, 24
25
6
7
7
9 8
8
8
89 8
8
8
87
7
9
13
1412
10
11
1
2
3
4
5
15
1516
17
17
16
19
20
21
22
23
24
18
25
2.2113.43
2.00
1.61
4.62
1.54
5.51
5.02
13,14
*
8
17
34
2
16, 712, 11
6
9
5
15
1, 1022, 23
*
*
21
18,20,23,
19, 24
25
6
7
7
9 8
8
8
89 8
8
8
87
7
9
13
1412
10
11
1
2
3
4
5
15
1516
17
17
16
19
20
21
22
23
24
18
25
Figure S2 1H NMR spectrum of tri-block PEA in d6-ethanol.
A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s
109
The integration of characteristic protons signals for 1,6-hexanediol (15), lysine (5),
isosorbide (21) confirmed the molar composition of the tri-block polymer (m:n:o =
0.31:0.27:0.42; see Scheme 1), which is very close to the intended ratio 0.3:0.25:0.45.
Lysine (5) and benzyl (25) groups show a 1:1 ratio. Lysine (5) and sebacic acid (6) exhibit a
1:1.1 ratio in the n-block of the polymer. Accurate integration was difficult in the case of
overlapping signals. The HSQC experiment was performed to positively identify the
characteristic signals.
13,14
8
17
3426
9
515
21
18,20,23
* **
*Solvent
16, 712, 11
19, 241, 1022, 23
13,14
8
17
3426
9
515
21
18,20,23
* **
*Solvent
16, 712, 11
19, 241, 1022, 23
Figure S3 HSQC spectrum of tri-block PEA in d6-ethanol.
Chapter 4
110
6.4 ESI-ToF-MS spectrum of α-chymotrypsin
In Figure 2 and 6, the peaks labelled with the symbol (*) are identified as α-CT enzyme
based on the ESI-ToF-MS spectrum of the peak (Figure S4). The molecular weight
calculated from this spectrum is 25 kDa, which is in agreement with the α-CT supplier’s data
(Sigma, C4129, 25 kDa).
+10+11
+12+13+14
+15
+16
+17
+18
+19
+20+21
+22
+23
+24
+25
+26
+27+28+9
+10+11
+12+13+14
+15
+16
+17
+18
+19
+20+21
+22
+23
+24
+25
+26
+27+28+9
Figure S4 ESI-ToF-MS spectrum of α-CT for the peaks labeled with symbol (*) in Figure 2 and 6.
111
Summary
In recent years, the demand for synthetic degradable polymeric biomaterials has been
growing continuously for temporary applications in the field of drug-delivery devices, for
tissue engineering, in scaffolds, or in surgical implants. The information obtained from their
chemical analysis can be used to modify existing biomaterials and for rational design of new
materials. The suitability of such materials for specific applications strongly depends on their
rate of degradation and their biocompatibility. Therefore, the objective of the present study
was to develop methods for the analysis of biomaterials by degrading them under non-
physiological and physiological conditions, followed by the analysis of their degradation
products.
The thesis starts with a review of different degradation methods and analytical strategies,
which may be applied to gain in-depth knowledge on the chemical structure of degradable
polymers and the toxicological nature of their degradation products (chapter 1). Methods for
the degradation and analysis of degradable materials using chromatographic separations and
spectroscopic or spectrometric detection methods provide useful information. Each
analytical technique has its limitations, but combination of techniques may provide powerful
tools to aid in optimizing the performance of degradable polymers.
In chapter 2 an approach is described to analyse the structure of different polyesters and
networks based on polyester urethane acrylates, based on completely degrading the polymers
at harsh conditions, such as elevated temperature and extreme pH. Degradation was
performed in alkaline media in a microwave instrument. Residues in the glass vessels were
prevented by incorporating an internal PTFE (“Teflon”) liner. Hydrolysis until completion
was monitored using NMR spectroscopy. The amount and the kinetic chain length of
poly(methacrylic acid) backbone were determined by size-exclusion chromatography (SEC).
The monomeric products were separated and quantified by liquid chromatography
hyphenated to mass spectrometry (LC-MS). The study provided useful insights in the
composition of these novel polymeric networks.
LC-MS was exploited to analyse the products of multi-block poly(ester amide) (PEA),
partially degraded at extreme pH or high temperature (chapter 3). This analysis provided the
basis to monitor the release of different monomers and oligomers generated by the enzyme-
mediated degradation of PEA with α-chymotrypsin and proteinase K under physiological
conditions. The compositional analysis by LC-MS yielded useful information on the
Summary
112
polymerization process. Quantitative analysis of several degradation products revealed the
different esterase and amidase activities of both enzymes. The potential of the studied
polymer for drug-delivery applications was supported by the steady degradation of the
polymer under physiological conditions, as was evident from the observed weight loss and
from the average molecular weight of the remaining polymer the degradation, which was
preserved according to SEC measurements. This confirmed a surface-degradation
mechanism. The molar ratios between blocks of the copolymer and different components of
the polymer were ascertained by 1D and 2D NMR measurements.
In chapter 4 a miniaturized and automated system is described to study the kinetics of
degradation and the products formed in an efficient way. Further degradation of the
generated products in the enzyme media could be avoided. The system allowed tracking the
degradation products of PEA, coated internally in stainless-steel tubing, under dynamic sink
and static conditions. The system was investigated using different reaction times, enzyme
concentrations, and capillary lengths. LC-MS proved to be good analytical technique for the
on-stream separation and identification of the released degradation products. Analysis in
continuous-feed and pulse-feed modes allowed to differentiate between primary and
secondary degradation products. Weight loss curves and concentration profiles of specific
degradation products often did not coincide. Shorter analysis times, no sample preparation
and more favourable conditions make the developed system a versatile analytical tool to
study the degradation of polymers. The minimal requirements on sample quantities and
enzyme activities highlight the efficient application of this system.
This thesis resulted in the quantitative analysis of complex polyacrylate networks and an
innovative system for fast assessment of degradable polymeric coatings.
113
Samenvatting
De laatste jaren laten een gestage groei zien van de vraag naar degradeerbare biomaterialen
op basis van synthetische polymeren. Deze worden onder andere toegepast op de gebieden
van gecontroleerde medicijnafgifte, in weefselkweek, als botschraag en als implantaat.
De analyse van de chemische structuur van biomaterialen levert informatie die kan worden
gebruikt voor verbetering van de huidige en het rationeel ontwerpen van nieuwe materialen.
De geschiktheid van zulke materialen voor specifieke doeleinden is sterk afhankelijk van
hun afbraaksnelheid en hun biocompatibiliteit. De doelstelling van de onderhavige studie is
dan ook om methoden te ontwikkelen voor de analyse van biomaterialen door ze (deels) af
te breken onder al dan niet fysiologische omstandigheden, gevolgd door de analyse van hun
afbraakproducten.
Het proefschrift begint met een overzicht van verschillende afbraakmethoden en analytische
strategieën die toegepast zouden kunnen worden om diepgaand inzicht te verkrijgen in de
structuur van degradeerbare polymeren en de toxicologische aard van hun afbraakproducten
(hoofdstuk 1). Methoden voor afbraak en analyse van afbreekbare materialen met behulp
van chromatografische scheidingen en spectroscopische of spectrometrische
detectiemethoden leveren nuttige informatie omtrent de structuur van de materialen en hun
afbraakproducten. Elke afzonderlijke analytische techniek heeft zijn beperkingen, maar
combinaties van technieken kunnen krachtige hulpmiddelen zijn ter ondersteuning van de
optimalisering van de eigenschappen van afbreekbare polymeren.
In hoofdstuk 2 wordt de analyse beschreven van de structuren van verschillende polyesters
en netwerken op basis van polyester urethaan acrylaten door de polymeren geheel af te
breken bij extreme omstandigheden zoals hoge temperatuur en pH. Een PTFE (“teflon”)
voering was nodig om neerslag te voorkomen in de monsterbuis van de magnetron. Met
NMR werd de hydrolyse tot het eindpunt gevolgd. De hoeveelheid en ketenlengte van het
gevormde poly(methacrylzuur) werden bepaald met size exclusion chromatografie (SEC).
De overige (monomere) producten werden gescheiden en gekwantificeerd met behulp van
vloeistofchromatografie gekoppeld met massaspectrometrie (LC-MS). Dit onderzoek
verschafte nuttige inzichten met betrekking tot de samenstelling van deze nieuwe polymere
netwerken.
LC-MS werd gebruikt om de producten van gedeeltelijk bij extreme pH of hoge
temperatuur afgebroken multi-blok poly(esteramide) (PEA) te analyseren (hoofdstuk 3).
Samenvatting
114
Deze analyse maakte het mogelijk om het vrijkomen van verschillende monomeren en
oligomeren te volgen, die worden gevormd door de enzymatische afbraak van PEA onder
invloed van α-chymotrypsine of proteinase K onder fysiologische omstandigheden. Analyse
van de samenstelling met LC-MS leverde informatie over het polymerisatieproces.
Kwantitatieve analyse van verscheidene afbraakproducten bracht ook de verschillende
esterase en amidase activiteiten van beide enzymen aan het licht. De goede vooruitzichten
van dit polymeer in toepassingen voor medicijnafgifte werden ondersteund door de
geleidelijke afbraak onder fysiologische omstandigheden, zoals bleek uit het
gewichtsverlies en het gelijk blijven van het gemiddeld molucuulgewicht van het
overblijvend polymeer, dat met SEC tijdens de afbraak werd onderzocht. Dit bevestigde dat
deze afbraak een oppervlakteverschijnsel betrof. Tenslotte werden met 1D- en 2D-NMR
technieken de molaire verhoudingen tussen de copolymeerblokken en tussen de
verschillende delen van het polymeer vastgesteld.
In hoofdstuk 4 wordt een geminiaturiseerd en geautomatiseerd systeem beschreven om de
kinetiek van de afbraak op een snelle en efficiënte manier te bestuderen en de verdere
afbraak van de geproduceerde verbindingen in de enzymoplossing te voorkomen. De
afbraakproducten van PEA, dat is aangebracht op de binnenkant van een roestvrijstalen
buisje, kunnen met dit systeem worden gevolgd onder statische, maar ook onder
dynamische (niet-evenwichts) omstandigheden. Het systeem is onderzocht met
verschillende reactietijden, enzymconcentraties en lengtes van het capillair. De vrijkomende
afbraakproducten werden on-line gescheiden en geïdentificeerd met LC-MS. Onderscheid
tussen primaire en secundaire afbraakproducten bleek mogelijk door analyse met continue
of gepulseerde toevoer van enzymoplossing. Het gewichtsverlies en het concentratieprofiel
van specifieke afbraakproducten vielen vaak niet samen. De kortere analysetijden,
ontbreken van de noodzaak voor monstervoorbereiding en gunstige reactieomstandigheden
maken het systeem tot een veelzijdig analytisch hulpmiddel voor het bestuderen van de
afbraak van polymeren. Andere prominente voordelen van dit systeem zijn de geringe
hoeveelheden polymeer en lage enzymactiviteit die vereist zijn.
Dit promotieonderzoek heeft geresulteerd in de kwantitatieve analyse van de samenstelling
van complexe polyacrylaatnetwerken en een handig, innovatief systeem voor de snelle
analyse van afbreekbare polymere coatings.
Acknowledgements
115
Acknowledgments
With the name of ALLAH, the Beneficent, the Merciful. Countless and humble thanks to
Almighty God, who blessed and bestowed me the courage and patience to fulfil this
assignment.
Finally the time has arrived. I can appreciate, and acknowledge the support and help of those
special people I met and worked with. First of all, I owe my deepest gratitude to my
promotor, Sjoerd van der Wal, for his kind support, effort and worthy advices during the
accompanied period. This thesis would not have been possible without his brilliant guidance
and the highly constructive critical discussions I had with him during the research work.
Despite the distance, he always responded to my queries by e-mail, even on weekends. I
sincerely appreciate the very kind, friendly and motivating attitude of my co-promotor, The
Big Peter, who provided me with an opportunity to work in his very dynamic and
international research group. Indeed, highly useful and informative discussions with him
always added value to my research work. Sjoerd and Peter, thank you very much for your
kind help and valuable contributions to my education and research. It was an honour for me
to work under your supervision.
I am very grateful to Prof. W. E. Hennink, Prof. D. W. Grijpma, Prof. J. G. M. Janssen, Prof.
C. G. de Koster, Dr. W. Th. Kok, and Dr. A. A. Dias for being members of my promotion
committee. The feedback and comments from Hans-Gerd Janssen were very much
appreciated.
Definitely, I cannot forget the wonderful time I spent at the University of Amsterdam and in
the cosmopolitan Amsterdam. My special thanks go to Wim, a true Amsterdam native, for
his expertise, cooperation, and precious suggestions throughout my PhD. Your warm
reception in the very early morning of 1st November, 2007 at Schiphol is highly regarded.
I am truly indebted and thankful to my wonderful colleagues of the Analytical-Chemistry
Group (former Polymer-Analysis Group) for making the work environment very friendly. To
Gabriel, for helping me out with data analysis and understanding MATLAB, whenever I
visited your office. To Erwin, for guidance in understanding the art of connecting and
disconnecting different HPLC systems and detectors to make suitable combinations. To
Peter Pruim, for providing me the opportunity to use the nano-pump, which was very fruitful
for me. To Rob, for providing useful tips on SEC/GPC. To Linda, for her hospitality, shukria
olivia. To Elena and Dominique, for useful discussions. To Rudy, for proofreading my thesis
Acknowledgements
116
and for being very supportive. To Peter verschuren, for pleasant technical discussions to help
tackle the difficulties during my research work. To Tom, for his generosity and tremendous
amount of cooperation that definitely facilitated my research. To Petra Aarnoutse, for her
kind help. My warmest and special thanks go to Aleksandra and Daniela, for their utmost
help and for being very caring throughout my stay. Importantly, I cordially thank Rashid for
the wonderful time we spent together in Gein and at the University of Amsterdam. Playing
tennis, cooking, and roaming around the Gasperplas with you will be a part of my memories.
Thanks for helping me in designing the cover of my thesis in photoshop. Your absence was
very much felt during the last couple of months of my stay in Amsterdam. I also enjoyed the
time I spent with to Stella, Francisco, Filippo, Sonja, Arend, Dini, Xulin Jiang, Hanneke,
Ngoc A, Katya, Maria, Jana, Chuchu, and Henrik (the viking). I also thank my students
Friso, Rolf, Roy, Merel, and Paul for working with me at different levels of my PhD and for
their contributions to this dissertation. I am grateful to Azmatullah Khan Baloch for inviting
me frequently for tasty pakoras and tea. I am privileged to have had Hannan Tahir – a very
kind, humble and sober person – as my housemate during the last year of my stay. To Sajjad
Zahir, thanks for bringing delicious food several times.
I owe sincere and earnest thankfulness to Marianne (from international office), Gerda, Petra
Hagen, Maureen, Renate (from HIMS office), and Marijke Duyvendak (from UvA Library)
for their indispensable assistance. Great discussions with the skilled Jan van Maarseveen and
benign/kind support from Hans Bieraugel (from the synthetic-organic-chemistry group), Jan
Geenevasen and Jan Meine (from the NMR department), Louis Hartog (biocatalysis and bio-
organic chemistry group), and Marjo Mittelmeijer-Hazeleger (from heterogeneous catalysis
and sustainable chemistry) were very much appreciated. I would like to mention Marco
Scholten (from Agilent Technologies) for his help in understanding different aspects of the
time-of-flight mass spectrometer.
It was a pleasure to collaborate with Janine Jansen and Dirk Grijpma; I always enjoyed my
visits to the University of Twente and to have useful discussions with you.
I am thankful for the support of DSM during my research. In particular, I would like to take
the opportunity to thank Aylvin for his critical remarks and highly useful suggestions; Ron
Peters, Bart Plum, and Tristan Handels for their kind assistance with the polyester-urethane-
acrylates project. The knowledge, expertise, and the exceptional support of George Mihov
and Guy Draaisma helped me a lot during the analysis of multi-block poly(ester amide)s. I
also would like to convey my thanks to Marc Hendriks.
Acknowledgements
117
I sincerely acknowledge the financial support for my PhD from the Higher Education
Commission (HEC) of Pakistan and Nuffic, The Netherlands. Certainly, the kind and caring
responses from Rana Shafiq Ahmad, Muhammad Ashfaq (from HEC), Charlene and Loes
(from Nuffic) are highly appreciated.
I am also grateful to my employer “University of Engineering and Technology Lahore” for
granting me the study leave to pursue my PhD at the University of Amsterdam. I would like
to express my gratitude to my colleagues at the Chemistry Department of UET for their best
wishes.
I am thankful to Prof. Robert Langer and Dr. Bozhi Tian (Massachusetts Institute of
Technology) for their kindness to provide me the high resolution picture and for the
generosity in granting me the permission to use it on the cover of my thesis.
I also acknowledge the support of my friends Shahzad Ahmed, Adnan Tahir, Mian Tariq,
Ijaz-ul-Mohsin, Khurram, Muhammad Abu Naeem (Lionel Gomes), and many more, who
helped me from the beginning and who made my stay comfortable.
If I forgot anybody, I humbly apologize. Thanks anyway.
Finally, I would like to thank my parents and family members for their prayers,
encouragement, and support…..
شايد .مجهے اپنی دعاۇں ميں ياد رکها وقت ہر نے گزار ہوں جنہوں شکر کا والدين پيارے ميں، ميں اپنے آخر..…
اهللا تعالى آپ کو صحت اور تندرستی .ميرے پاس وہ الفاظ نہيں جن سے ميں انکے احسانات کا شکريہ ادا کر سکوں
حيات اور اپنی پياری بيٹی اور بيٹے کو شريک ميں اپنی .دير ہمارے سروں پر قائم رکهے دے اور آپ کا سايہ تا
ميں .ہوں کرتا پيار بہت سے سب آپ ميں. ہوں سراہتا کو ہمت اور حوصلے کے ان اور ميں کبهی نہيں بهول سکتا
گزار ہوں جنہوں ئيوں سے شکراور اپنی بہنوں کا دل کی گہرا....) عمير حاکم، عامر، آصف،( بهائيوں اپنے پيارے
احمد باسم اور بهانجوں پيارے سے ہی بہت ميں اپنے. نے ميری غير موجودگی ميں والدين کا خيال رکها
ہوں اٹهتا مسکرا کر ميں ہميشہ سن سکتا، جن کو بهول کو کبهی نہيں باتوں لطف، پر ميٹهی، ميٹهی کی
.ہوں جاتا بهول پريشانياں تمام اپنی اور
Abdul Ghaffar
Amsterdam, September 2011
Bibliography
Printed by Ipskamp Drukkers, Enschede, The Netherlands 118
Bibliography
[1] A. Ghaffar, P.G. Verschuren, J.A.J. Geenevasen, T. Handels, J. Berard, B. Plum, A.A. Dias, P.J. Schoenmakers, S. van der Wal, Fast in vitro hydrolytic degradation of polyester urethane acrylate biomaterials: structure elucidation, separation and quantification of degradation products, J. Chromatogr. A, 1218 (2011) 449-458.
[2] A. Ghaffar, G. J. J. Draaisma, G. Mihov, A. A. Dias, P.J. Schoenmakers, Sj. van der Wal, Monitoring the in vitro enzyme-mediated degradation of degradable poly(ester amide) for controlled drug delivery by LC-ToF-MS, Biomacromolecules 12 (2011) 3243-3251.
[3] A. Chojnacka, A. Ghaffar, A. Feilden, K. Treacher, H.-G. Janssen, P. Schoenmakers, Pyrolysis–gas chromatography–mass spectrometry for studying N-vinyl-2-pyrrolidone-co-vinyl acetate copolymers and their dissolution behaviour, Analytica Chimica Acta (2011) doi:10.1016/j.aca.2011.05.052
[4] A. Ghaffar, G. J. J. Draaisma, G. Mihov, P.J. Schoenmakers, Sj. van der Wal, A versatile system for studying the enzymatic degradation of multi-block poly(ester amide)s, to be submitted.
[5] A. Ghaffar, P.J. Schoenmakers, Sj. van der Wal, Methods for the chemical analysis of degradable synthetic polymeric biomaterials, to be submitted.
[6] Janine Jansen, Abdul Ghaffar, Thomas N.S. van der Horst, George Mihov, Sjoerd van der Wal, Jan Feijen, Dirk W. Grijpma, Controlling the kinetic chain length of the crosslinkages in photo-polymerized biodegradable networks, to be submitted.
Course Attended
[1] Molecular Spectroscopy, 2009, organized by AIO Network Analytical Chemistry,
The Netherlands.
[2] Chemometrics, 2010, organized by AIO Network Analytical Chemistry, The Netherlands.
[3] Polymer Chemistry, RPK-A 2010, organized by National Dutch Graduate School of Polymer Science and Technology (ptn).
top related