Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Novel liquid-liquid scalable extraction/chromatography technology
December 3rd, 2008
Professor I.A. Sutherland
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Delft, Utrecht, Rotterdam, BPP2005
• Introduction to industrial scale centrifugal liquid-liquid extraction and its potential for the purification and manufacture of proteins and biologics
• Reviewed Hydrostatic & Hydrodynamic Countercurrent Chromatography
• Prospect for true moving bed chromatography
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Outline of Today’s Talk
• Some advantages of working with liquid stationary phases
• Some additional not well known versatilities
• Quick review of high-resolution, hydrodynamic counter-current extraction
• Quick introduction to aqueous two-phase systems (ATPS)
• Focus on high-resolution, hydrostatic counter-current extraction using ATPS in isocratic mode
• Introduction to intermittent true moving bed extraction
• Other high resolution continuous extractions schemes in the offing
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Key Advantages of a Liquid Stationary Phase
• Total (100%) sample recovery – liquid-liquid nature of the process means it is gentle and versatile and there is no irreversible adsorption to a solid support
• Improved sample solubility – both upper and lower phase available
• Quick due to higher throughput – High flow, low pressure system
• Inexpensive & a Green Technology compared to HPLC (significantly less solvent usage and no expensive solid supports)
• Crude extracts, including particulates can be handled reducing the number of separation steps in a process and reduced sample preparation requirements
• High loading capacity due to liquid nature of stationary phase
• Easy & predictable scale-up so that scale-down to scale-up philosophy can be applied
• Low pressure system – so columns can be lengthened to increase resolution and widened to increase capacity and throughput
• Normal and reversed-phase operation and the switching between them gives expanded operational capability – elution extrusion, dual mode, true moving bed
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Other Little Known Advantages
• Isocratic operation– Either phase can be mobile (normal or reverse phase
– The process can be switched half way through a run
– The process can be stopped and the retained components pumped
out (elution/extrusion – Berthod)
• Co-current Chromatography/Extraction– Both phases can be pumped at the same time in the same direction
• Counter-current Chromatography/Extraction– Both phases are simultaneously flowed in opposite direction
– Continuous processing of closely related binary compounds
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Hydrodynamic Counter-current Extraction
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
New Dynamic Extraction/Chromatography Technology
• Developed now as a fast, robust, continuous process with separations in minutes as opposed to hours with previous technology
• Small molecules – Natural Products
• Potential for application to proteins, peptides, plasmids and other biologics
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Coil Planet Centrifugation producing mixing
and settling zones along a continuous tube
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Twist free connection to rotor
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
DE Centrifuges - Predictable Scale-up
Milli-CCC – R=50mm, Vc=5-17ml Brunel-CCC – R=110mm, Vc=0.1-1 litre Maxi-CCC – R=300mm, Vc=4.6 litre
CCC-Device
Bore Area B Mean
Flow
92.5 90 85 80 70
Mini-DE 0.8 0.50 -37.01 0.50 0.04 0.07 0.16 0.29 0.66
Midi-DE1 1.6 2.01 -11.22 2.00 0.45 0.79 1.79 3.18 7.15
Midi-DE2 3.68 10.64 -3.13 10.00 5.74 10.21 22.97 40.83 91.87
Maxi-DE 10 78.54 -0.82 80.00 83.66 148.72 334.62 594.88 1338.49
Flow at SP Retention (Sf) given below
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Hydrodynamic Counter-current Extraction
• Continuous tube with wave mixing
• Stop the pump & liquid phases move to opposite ends of the tubing
• Ideal with low viscosity aqueous organic phases
• Hydrodynamic equilibrium between the mobile phase and the retained liquid stationary phase
• Kg scale in days have been demonstrated with aqueous/organic phase systems
• Feasibility of continuous countercurrent extraction has now been demonstrated at a pilot scale
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Aqueous Two-Phase Polymer Systems (ATPS)
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Aqueous Two-Phase Systems (ATPS)•Numerous polymer-polymer, polymer/salt combinations
•Gentle host medium for biologics (ref Albertsson, Kula, Hubbuch)
•Low interfacial tension
•ATPS more viscous than aqueous/organic phase systems
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Cascade v. Wave Mixing
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Cascade v. Wave Mixing
Hydrodynamic systems retain ATPS
but wave mixing is too gentle for
large proteins
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Cascade v. Wave Mixing
Hydrodynamic systems retain ATPS
but wave mixing is too gentle for
large proteins
Large scale toroidal hydrostatic
systems being constructed but
not yet tested
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Hydrostatic Counter-current Extraction
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Hydrostatic Counter-current Extraction
• Often discrete chambers with narrow bore interconnecting tubes
• Cascade mixing (like waterfalls)
• Stop the pump and the stationary phase stays where it is – trapped in the chambers
• The focus today - Centrifugal Partition Chromatography (CPC)
Channel
Duct
CPC: ... Without solid support ...
G
Descending modeFlow
Channel
DuctFlow
Ascending mode
PUMP
PUMP
1 channel ≈ 90 µL
Sanki HPCPC, analytical scale
Armen Elite, industrial scale
Twin cells ≈≈≈≈ 12 mL
For the 12.5 liter instrument
Earth gravity modulates the Coriolis acceleration and promotes spray formation
Earth gravity :possible sedimentation of the lower phase in deep channels
Vertical pack Horizontal pack
Visual CPC II
g
g gΓcor.
Γcor.
Stroboscopic looking at the same cell, observing
the spray formation
Stroboscopic flash and
fast response camera
1) Marchal L, Foucault AP, Patissier G,
Rosant J, Legrand J (2002) Centrifugal partition chromatography: an engineering
approach, In: Comprehensive Analytical
Chemistry (A Berthod, Ed), Elsevier, 115-
157.
2) Marchal, L., Legrand, J., Foucault, A. (2003)
Centrifugal partition chromatography: a
survey of its history and our recent
advances in the field, Chem. Rec. 3(3),
Flow rate (mL/min)
BuOH / Water
Ascending Mode
From 0 to 500 mL / min in 1 min then back to 0 mL / min in 1 min
02550751001251501752002252502753003253503754004254504755004754504254003753503253002752502252001751501251007550250
1290 rpm506 rpm
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Advanced Bioprocessing Centre (ABC)
Hazard’s Lab
Applications Lab
Analytical Lab
Control Room
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
1 litre Hydrostatic CPC Set-up at BIB
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Run Conditions
• Centrifuge: Armen Elite 1 litre CPC
• Capacity: One rotor only - volume 500ml, cell volume 429ml
• Phase System: ATPS: 12.5% PEG1000-12.5% K2HPO4
• Sample: 90mg lysozyme, 90mg myoglobin in 40ml (~10% CV) 50:50 ATPS mix
• Speed: 2000 rpm (224g)
• Flow: 10 ml/min
• Mobile Phase: lower salt phase
• Stationary phase retention (Sf) at breakthrough: 52%
• Stationary phase retention (Sf) at end: 19%
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Analytical Runs
0
10000000
20000000
30000000
40000000
50000000
60000000
70000000
80000000
90000000
100000000
0 10 20 30 40 50 60 70 80
Fraction Number
HP
LC
Peak
Are
a (
uV
s)
Lysosyme
Myoglobin & Apomyoglobin
k=0.59, Sf=40%, 80 CCD steps
k=1.91, Sf=19.5%, 80 CCD steps
Rs=1.28
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Analytical Runs
0
10000000
20000000
30000000
40000000
50000000
60000000
70000000
80000000
90000000
100000000
0 10 20 30 40 50 60 70 80
Fraction Number
HP
LC
Peak
Are
a (
uV
s)
Lysosyme
Myoglobin & Apomyoglobin
k=0.59, Sf=40%, 80 CCD steps
k=1.91, Sf=19.5%, 80 CCD steps
Note: theoretical prediction
from test tube experiments
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Transfer to Pilot Scale Unit in France
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Armen 12.5 litre CPC at Archimex
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Run Conditions
• Centrifuge: Armen Continuum 12.5 litre CPC
• Capacity: One rotor only - volume 6.25 litre
• Phase System: ATPS: 12.5% PEG1000-12.5% K2HPO4
• Sample: 1.1g lysozyme, 1.1g myoglobin in 500ml (~10% CV) 50:50 ATPS mix
• Speed: 1293 rpm (224g)
• Flow: 125 ml/min
• Mobile Phase: lower salt phase
• Stationary phase retention (Sf) at breakthrough: 63%
• Stationary phase retention (Sf) at end: 22%
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Some of the team at Archimex
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Fraction collection & on-line monitoring
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Armen 12.5 litre CPC at Archimex
0
200
400
600
800
1000
1200
1400
1600
0 10 20 30 40 50 60 70 80
Fraction Number
HP
LC
Pe
ak A
rea
Lysozyme
Myoglobin + Apomyoglobin
Rs = 1.88, sample 10%CV
ECM172 - 10%CV Sample Loading
0
10000000
20000000
30000000
40000000
50000000
60000000
70000000
80000000
90000000
100000000
0 10 20 30 40 50 60 70 80
fraction Number
HP
LC
Pe
ak
Are
a
Lysosyme
Myoglobin & Apomyoglobin
Rs=1.28
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Theoretical Predictions – Process Scale
0
200
400
600
800
1000
1200
1400
1600
0 10 20 30 40 50 60 70 80
Fraction Number
HP
LC
Pe
ak A
rea
Lysozyme
Myoglobin + Apomyoglobin
Theory - k=0.59, Sf=52.5%, 80 CCD steps
Theory - k=1.91, Sf=24.2%, 80 CCD steps
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
ATPS Stripping Characteristic – Scale-up
ATPS Stripping Characteristic During Scale-up
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 10 20 30 40 50 60 70 80 90
Time (mins)
Fra
cti
on
of
Co
lum
n V
olu
me
6.25 litre rotor
0.5 litre rotor
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Conclusions – Batch Isocratic
• Linear scale-up feasible
• Robust technology – easily transferred from the test tube to a production run
• Note that with Hubbuch’s robotic high throughput screening with ATPS systems, combined with this rapid scale-up approach – small footprint, rapid new approaches to manufacture are becoming feasible
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Intermittent – True Moving Bed (i-TMB)
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
True Moving Bed
T(P)
Heavy Phase Out
Light Phase In
Light Phase Out
Heavy Phase In
Force Field
(ng)
H(C)
Sample In
Distal – outer layer
Proximal – inner layer
Sutherland, Scientific Update, San Francisco
March 21-22, 2005
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Injection
Ascending mode period
Descending mode period
Injection of A + B
BA
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Run Conditions
• Centrifuge: Armen Elite 1 litre CPC
• Capacity: Two rotors - volume 1000ml, cell volume 858ml
• Phase System: ATPS: 12.5% PEG1000-12.5% K2HPO4
• Sample: Continuous flow of 2.2mg/ml lysozyme, 2.2mg/ml myoglobin at 3ml/min in UP (ascending) and LP (descending) –0.79g/hr – total 1.2g in 1.5 hours
• Speed: 2000 rpm (224g)
• Flow: 10 ml/min (Ascending) 5 minutes; 10ml/min (Descending) 5 minutes and so on …..
• Mobile Phase: alternating between between UP & LP
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Armen CPC i-TMB System at Archimex
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Descending Elution
0
1000
2000
3000
4000
5000
6000
7000
8000
0 20 40 60 80 100 120 140 160
Fraction Number
HP
LC
Pe
ak
Are
a (
arb
. u
nit
s)
Lysozyme
Myoglobin & Apomyoglobin
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Ascending Elution
0
1000
2000
3000
4000
5000
6000
0 20 40 60 80 100 120 140
Fraction Number
HP
LC
Pe
ak
Are
a (
arb
. u
nit
s)
Lysozyme
Myoglobin + ApoMyoglobin
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
D e s c e n d i n g E l u t i o n
0
1 0 0 0
2 0 0 0
3 0 0 0
4 0 0 0
5 0 0 0
6 0 0 0
7 0 0 0
8 0 0 0
0 2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0
F r a c t i o n N u m b e r
HP
LC
Pe
ak A
rea
(a
rb.
un
its)
L y s o z y m e
M y o g l o b i n & A p o m y o g l o b i n
A s c e n d i n g E l u t i o n
0
1 0 0 0
2 0 0 0
3 0 0 0
4 0 0 0
5 0 0 0
6 0 0 0
0 2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0
F r a c t i o n N u m b e r
HP
LC
Peak A
rea (
arb
. u
nit
s)
L y s o z y m e
M y o g l o b i n + A p o M y o g l o b i n
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Elution control monitoring during run
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Descending Elution - Myoglobin
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Summary – Batch v. i-TMB
Centrifuge* g/hour g/day
1 litre CPC Batch 0.14 3.24
i-TMB 0.79 19.01
12.5 litre CPC Batch 1.69 40.50
i-TMB 9.90 237.60
* Non-Optimised
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Conclusions
• Rapid scale up from the test tube (Hubbuch) to pilot scale (kg/day) is feasible for model systems in batch and continuous process mode
• Now needs evaluating with real systems– Paula Rosa, Lisbon, Portugal (BPP2007)
– Emma Bourton, Brunel, UK (PhD – Syngenta)
– Other volunteers welcome
Optimising Biomanufacturing
December 2-3, 2008Novel Liquid-liquid Extraction Technology
Challenges
• New liquid-liquid phase systems– Ionic liquids (QUILLS)
– ATPS with higher solubility (Hans-Olef Johansson, Sao Paulo, Brazil)
• Increased sample concentrations in ATPS
– Non-equilibrium sample loading & extraction (NESLE)
– New continuous tube hydrostatic approach
• Development of small footprint, versatile, multi-product facilities for “niche markets” highlighted by Dr Dana Andersen of Genentech
– Initially for high value added products
• UK Government – future emphasis on basic and “user-driven” research