poster presen¢ tion faseb - waters corporation · poster presen¢_tion faseb/asbmb symposium -...

t Essentialsin "

95065_

Poster Presen¢_tion

FASEB/ASBMB Symposium - April, 1991Poster # 4788

Membrane Convective LiquidChromatography (MCLC) for the Ion

Exchange Separation of Bio-Molecules

R.A. Hamilton & R. BarryMillipore Corporation, 80 Ashby Road,

Bedford, MA. 01730 U.S.A. _.

Waters Chromatography Division Wat ,/Waters. The absolute essential Millipore Corporation ersfor bioresearch. 34 Maple St. Divisiono! MILLIPORE

Milford, MA 01757(508) 478-2000

Membrane Convective LiquidChromatography

Membrane-Based Chromatography Columnsfor the High Performance Separation ofBiomolecules.

by Richard Hamilton and Robert Barry

ABSTRACT

Ion exchange chromatography is performed everyday in thebiotech laboratory. Columns packed with soft gel and porousparticle media used in most ion exchange separations is oftencharacterized by slow flow rates, low resolution, and lowthroughput. These limits are often due to the mass transport ofmolecules inherent in diffusion dominant systems.

Membrane Convective Liquid Chromatography (MCLC) isa new approach to the separation of biomolecules. Unlike beadsor traditional soft gels, mass transport of the molecules beingseparated is achieved through convective, rather than diffusive,flow.

This allows the flow to be increased without band broadeningand loss of resolution caused by small pore diffusion that istypical of conventional porous particles. High resolutionseparations of biomolecules can be achieved in minutes, nothours.

This study focuses on the application of membrane-basedconvective chromatography devices for the rapid, highresolution separation of proteins, enzymes and other bio-molecules by ion exchange.

INTRODUCTION

MEMBRANE CONVECTIVE LIQUID CHROMATOGRAPHY(MCLC)

Membrane-Based Chromatography Columns for the HighPerformance Separation of Biomolecules.

The extensive application of liquid chromatography for bioseparations began withthe introduction of soft gels in the late 1950s. These natural polymers, based oncellulose, agarose and dextrans, made it possible to form supports that arehydrophilic, easy to functionalize, economical and sufficiently porous for theseparation of large biomolecules. However, these packing materials also sufferedfrom various chromatographic limitations.

In the middle 1970s the introduction of high performance polymer particles helpedeliminate several limitations. The HPLC materials were strong and could be madein a variety of particle and pore sizes for most applications. These new columnscould be used at faster flow rates with smaller particles to improve resolution an,throughput. Surface chemistries could be created for a full range of separationtechniques. Although these new HPLC separation materials solved a number of theproblems associated with soft gels, they created several new limitations. Thesenew high pressure columns required special equipment to pack and operate.Resolution was improved, but capacity was low and throughput reduced.

Many of the factors that limit resolution in chromatography are associated withthe efficiency of molecular transport to and from the active binding sites on thesolid support. The application of wide pore particles produced a significantimprovement in resolving power. Most importantly however, it is the diffusionof the molecules in and out of those pores that physically limits the flow rateand therefor the throughput.

The key to increasing resolution AND increasing throughput is toeliminate the mass transport limitations characteristic of conventionachromatography.

The key to high performance liquid chromatography is to reduce the contributionof different band-broadening processes to column plate height, H [2].

H = Au 0.33 + B/u + Cu + Du Equ. 1.0

where: Au = Eddy diffusion and mobile phase transferB/u = Longitudinal diffusionCu = Stagnant mobile phase transferDu = Stationary phase mass transfer

Mass transport in a chromatography column occurs primarily by two mechanisms:(1) diffusive mass transport through the stagnant liquid in the pore and(2) convective mass transport in the flowing mobile phase between the particlesin the column. There are other factors that effect resolution such as eddy diffusklongitudinal diffusion and stationary phase mass transport, but the ratio of theseother mass transport effects are insignificant. The "stagnant mobile phase masstransfer" has been recognized as one of the major causes of band spreading inliquid chromatography. There have been many approaches to solving this problemshort of eliminating the pore entirely with nonporous beads.

We have taken a new approach-we have eliminated the beads and kept the pores.

Membrane Convective Liquid Chromatography (MCLC) is a new approach to highresolution separations. By flowing through the pores and not past them, weeliminate the diffusive mass transport limitation characteristic of conventionalbead columns. By placing the active surface as close to the flowing buffer aspossible, transport of the molecule to and from the active surface is byconvective mass transport. Resolution is increased as peaks narrow,even at very high flow rates. Equilibration, separation and re-equilibration are faster, yet capacity and throughput are high.

[1] E.A. Peterson and H.A. Sober. J. Am. Chem. Soc. 78, (1956), 751.

[2] L.R. Snyder and J.J. Kirkland, 'Introduction to Modern LiquidChromatography'. 1979.

EXPERIMENTAL PROTOCOL I RESULTS

All separation were performed on a CM MemSep-1000 (1.4 ml bed volume) using aWaters 650 Advanced Protein Purification System and the Auto-Blend* methodfor rapid optimization of separations. All chromatographic runs were conductedat pH 5.25. The loading buffer was 20 mM sodium acetate. The protein wereeluted using a 20 bed volume linear gradient from 0 to 0.25M NaCI in 20 mMsodium acetate.

Flow rates were varied from 1.4 ml/min (1 bed volume/min) to 14.0 ml/min(I0 bed volumes/rain). The three protein test mixture consisted ofc_-chymotrypsinogen A, cytochrome c (bovine heart), and lysozyme. A total of

1 mg of the protein mix was applied to the MemSep for each run.

Membrane-based chromatography columns introduce a fundamentally new approacito eliminating a number of the limitations in the separation of biomolecules. Byutilizing unique membrane systems and optimized device design, an innovativechromatographic system has been developed that dramaticallydecreases separation times and increases resolution.

Three plots were constructed from the resulting data (Fig. 1,2 & 3).

Figure 1. Peak volume vs Flow Rate: In a typical porous particle column, peakvolume increases as flow increases due to the mismatch in the

convectiveand the diffusive mass transport rates.

Figure 2. Separation Efficiency (H) vs Flow Rate: Separation efficiency is relatecto several mass transport factors as described in equation 1.0.

Figure 3. Resolution (Rs) vs Flow Rate: In a porous bead system, as flow rateincreases, resolution decreases. Complete separation between two

peaks is a number greater than 1. Resolution is calculated by usingthe equation:

Rs = [(t2 tl) / !/2 (wl + w2)]

Figure 4. Separation of the three protein mix at 1.4 ml/min ( 1 bed volume/minTotal run time = 32 minutes.

Figure 5. Separation of the three proteins mix at 2.8 ml/min (2 bed volume/minTotal run time = 16 minutes.

Figure 6. Separation of the three protein mix at 5.6 ml/min (4 bed volume/min)Total run time = 8 minutes.

Figure 7. Separation of the three protein mix at 9.8 ml/min (8 bed volumes/minTotal run time = 4 minutes.

Figure 8. Separation of the three protein mix at 14.0 ml/min (10 bed volume/n"Total run time = 3 minutes.

Peak Volume vs Flow

Peak Volume (m])40Z

//

//

J

30 /o/

o/

_. --"- MemSepi000 FIG. 1

10 --°---Soft Gellx30 Col.

F " Soft Gel.... "----" = lx10 Col.0 , , , , I , , , , I , , ,

0 5 lO 5

Flow (ml/minute)

Separation Efficient_ vs Flow

5008,E f f icienty

400023000

._ _ FIG. 22000 "" .........................................

........Graph D

1000 LgsozLjme

..... Cwto

0 ..............................._.............. ' .... ' • Chwmo0 5 i0 15

Flow (ml/minute)

Resolution (Rs) vs FlowResolution (Rs)

5L.,.,

4

3 FIG. 3........Peak 2vs3

Solt Gel

2 ...............Peak ivs2So f _ Gel

........Peak 2vs3l -"' '-. MemSep

.... :"::":- ...... Peak ]vs2.......... 1:::1:::: :::::: :1:1:;:: : : :- :::::-::::-:: -:::-:: l"lemSep

FI _ ,m ,-

0.070- 11.1 21.57

CM MemSep-lO00 0.050: A15.s7,AU 0.030] /LLL_o.oio

0 _ l0 I0 20 30 Minutes

.,,,,

FIG. 4

I I I I I I

Flow Rate: 1.4 ml/min

CM MemSep-1 000 s.70-070t I 10_

,u 0.050_ n7.70

oo.oloq0.920 4_'- i ' , I _ J l , J

0 2 4 6 8 I0 12 1416 Minutes

FIG. 5


t 2.9

CM MemSep-1000 0.070 _8

AU 0.050_ " '_"0.030-_0.010 q9.51

0 ,--_-"----I , , ,0 2 4 6 8 Minutes

FIG. 6

.Jk. _ L, _,__,

i i


1.8

oo7oCM MemSep-1000 o.o5o:1 3.18

AU 0.030J0.010 -_ 0.32

0_- , I I I

0 1 2 3 4 Minutes r_T_", t

I I I I I I


0.070q

CM MemSep-1000 o.o5o1AU 0.030

O.OLO-I0.24o,-7"-, , , , FIG. 8

0 0.5 1.0 1.5 2.0 2_5 310 Minutes

L

Illlll

'i

:,: High Performance Alternative To Conventional Soft Gel

Chromatography

INTRODUCTION

Ion exchange chromatography provides a powerful separation tool that findsapplication in virtually all branches of the life sciences. The value of itsutility on the practical level is defined in terms of the speed and resolutionof the separation, and the purity and biological activity of the recoveredproduct.

MemSep® (Membrane Separation) Chromatography Cartridges form MilliporeCorporation combine the resolution capabilities of conventionalchromatography media with the speed and throughput capabilities ofmembrane systems. Here, the utility of the MemSep anion and cation exchangerin protein separation is examined by comparing them to the separation of thesame protein test mixture using conventional soft gel chromatography.

EXPERIMENTAL PROTOCOL ! RESULTS

Separation of the indicated protein mixtures was examined first on highperformance agarose columns at flow rates of 2.8, 4.2 and 5.6 ml/min.Within a reasonable run time the resolution exhibited at 2.8 and 4.2 ml/min

were comparable, whereas a significant decrease in the resolving power ofthe agarose column was observed at the higher flow rate of 5.6 ml/min. Theprotein elution profiles obtained at a flow rate of 4.2 ml/min is shown inFigure l a & 2a.

The elution sequence of the two protein mixture employing cation (CM)exchange chromatography (Fig. 1) is: 1. c(-chymotrypsinogen A

2. lysozymeThe elution sequence of the three protein mixture employing anion (DEAE)exchange chromatography (Fig. 2) is: 1. human transferrin

2. ovalbumin

3. g-lactglobulin A

The agarose column was then removed and replaced with the correspondingMemSp-1000 cartridge to obtain the chromatograms shown in Fig. l b & 2b.The flow rate, gradient, run time and sample quantity (I mg of total protein)all remained the same as indicated for the agarose column.

In Figure l c & d, and Figure 2c & d, the chromatographic separation of thecorresponding protein mixtures was examined at considerably shorter runtimes, under modified gradient conditions and increased flow rates of5.6 ml/min (Fig. l c: 2c) and 11.2 ml/min (Fig. l d: 2d) using the MemSep-I000 ion exchangers.

oo-I lOO-9o-! 90-

80- '"1 _ 80-70 -- o 'i 70 -;"';'";";, ",T'_;'";, °"+

+°- "/ I +o_ _so_: 20_ I1 ,,91A28o50 -_ c_,,: CMM,_S,_;C_OA28o 50 -- Sa_: 200_J 11 mgI Flow ra_e:l1.2 rrd/min

CoJ_'nn:CM/v_n'_p-1000 BufferA: 20 _ Sod;urnAce_e, 10H4.75Flowrote:4.2 ml/m_n Bd'_ B: 1.0 M Solon AcCrue,pH4.75

40- 8u_A: 20m_ _o(_umAcet_le,p_ 4.7._ ,4 0 __ Grodk_r_.0_w0.5 rain,0_)30_Bin 5 rain[_J_l" B: | .0 M ,_x_um A.r._at_, pH 4.75 _: 280 m'_, 0.2 AUFS

Grodi_m_. 0 IO 10% B in 4 rain, lO b 80_|n22 rain

30 - ' m.,,,:,+,,:_o ,,,,.o._̂ u_s 30 -

b 0 "rime/rain 2A 0 6..- Time/rain

FIGURE 1. Separation of a two protein mixture (c<-chymotrypsinogen Aand lysozyme) on: a. High Performance CM Agarose column(IXIO cm); b, c, & d. DEAE MemSp-lO00 Chromatography Cartridge.

1 235 6 7

_' "o _ 1 2 35._o

'"" -- 0 5 10 15 20 25 30 35 40 "o '_ -- 0O_ 0 i'll-ill III I I III I I I I Illi III| I I I I I I'1 ] I Ill J I

100- / " 100 0 2 4 s e t01214

• _ Sample: 50p.L (1 rag)

80 " - _ Column: DEAE High Sample: ,50gL (1 rag)performance Agarose _0 Column: DEAE MemSep-1000

-_. Flowr_e:. 4.2 mUmin Flowrate: 5.6 mUmlnBufferA: 20 mM Tfis-HCI, pH 8.5 Buffer A: 20 mM Tris-HCt, pH 8.5

(_0 BufferB: 1M NaCltnA Buffer B: 1M NaCAInAGradient: 0% Btor 6 rain, 60 Gradient 0 - 15% Bin 10 mtn

0 to 25% B In 38 rain _ Detec_on: 280 nm, 0.1 AUFSA280 _etec_on: 280 nra, 0.1 AUFS

2 2¢a c

0 Ttme/min 41 ( 0 Time {min) 14

_._ o_,............... ,,,, ..................... g_- o" 0 5 10 15 20 25 30 35 40 100- 0 2 4 6 8

Cot'umn: DEAE Mem_:_ep_uu,J 80 •Flow rate: 4.2 mUmln Column: DEAE MemSep-1000

80 BuffecA: 20 mMTtis-HCt, pH 8.5 Rowra.te: 112. mL/m_Buffet El: 1M NaCI inA BufferA: 20 mM Tds-HCt, pH 8.5

Gradlec¢ 0% Bfor 6 rain, 60 "_ Bullet B: 1M NaCIInA0 to 25% Bin 38 rain Gradient: 0 o 15% Bin 6 rain

[_ectlon: 280 nm, 0.1 AUFS _ Detecffom 280 nrn, 0.1 AUFS

- _ ' 40 -_

20-

b- _,_ 0 _/o..... • o 9 d0 "lqme/min 41 Time (rain)

FIGURE2. Separation of a three protein mixture (human transferrin,ovalbumin, fi-lactoglobulin A) on: a. High Performance DEAE

Agarose column (1X10 cm); b, c, & d. DEAE MemSp-1000Chromatography Cartridge.

Rapid Separation of Glucose-6 Phosphate Dehydrogenase _

From Crude Yeast Extract

INTRODUCTION

The goal in all purification protocols is the isolation of the desired product inthe purest form possible, in the shortest period of time, while maintaining thehighest degree of biological activity. Here, the utility of MemSep anion exchangechromatography is examined in the rapid separation of G6P-DH from yeast.


All chromatographic separations were performed using Waters 650 AdvancedProtein Purification System. The flow rate was 5 or 10 ml/min with a 10 minlinear gradient from 0 to 0.2M NaCI in 20 mM Tris buffer. The DEAE MemSep-1000 (1.4 ml bed vol.) and the larger DEAE MemSep-1010 (4.9 ml bed vol.) haveprotein binding capacities of 10-20 mg and 60-80 mg respectively.

Separation of glucose-6-phosphate dehydrogenase (G6P-DH) from yeast extractwas first evaluated at low concentrations (--5 rag) on the DEAE MemSep-1010

chromatography cartridge employing the Waters Auto-Blend* method. Because ofthe speed advantage offered by MemSep cartridges, injections could be made ever15 minutes, examining separations at nine different pH values, from pH 7.25 topH 9.1, in just over 2.5 hours (Figure 1).

Protein detection is at 280 nm. Enzymatic activity was determined using a

spectrophotometric assay monitoring the reduction of NADP + at 340 nm.

Glucose-6-P �NADP+ G6P-DH Gluconate-6-P + NADPH + H+

Using the information provided by the Auto-Blend pH study, the yeast extract wasthen fractionated at considerably higher sample concentrations (20 - 50 mg) on

the MemSep-1010 under optimal conditions (Fig. 2a). The active peak was collectcdesalted, concentrated and applied to a DEAE MemSep-1000 chromatographycartridge. The resulting chromatogram demonstrates a single peak whichco-chromatographs with authentic enzyme (Fig. 2b).

The purity of the final chromatographic fraction was examined using SDS-PAGEwhich showed that the isolated G6P-DH is free of major protein contaminants.

Recovery of the target enzyme was greater than 80%, demonstrating activityon the order of -340 U/mg.

*Warren et al., 'A New Strategy for Rapid Optimization of Protein Separations',American Biotechnolo_v Laboratory, June 1989.

pH- 725 pH- 7.6 pH- 7.75

10 peaks 9 peaks 10 peaks

I I I I I I I I I I I I I I I I I I I I J0 4 8 12 0 4 8 12 0 4 8 12

pH- 8.0 pH- 8.2 pH- 8.3


I I I I I I I I I I I I I I I I I I 10 2 4 6 8 10 12 0 2 4 6 8 10 12 0 2 4 6 8 10 12

pH- 8.5 pH- 8.75 pH- 9.1


I I I I I I I i i i i i I i I0 2 4 6 8 10 12 0 2 4 6 8 10 12 0 2 4 6 8 10 12

FIGURE 1. Fractionation of crude yeast extract (-5 mg) on a DEAE MemSep-1010 cartridge at nine different pHs from 7.25 to 9.1 using theWaters Auto-Blend* method. G6P-DH containing fractions are

indicated by the grey areas.

a

bG-6-P-DH

J

G-6-P-DH

I I I I I I I

0 2 4 6 8 10 12

FIGURE2. a. Separation of-35 mg of yeast enzyme concentrate at pH 7.6on a DEAE MemSep-1010 Chromatography Cartridge. b. Rerun ofthe isolated G6P-DH fraction on a DEAE MemSep-1000.

_'/ Single-Step Purification of Mab from Ascites

INTRODUCTION

No two monoclonal antibodies (Mab) are alike. When considering purificationstrategies, the variation in the physio-chemical nature of monoclonalsnecessitates a variety of separation techniques to draw upon. The considerationsfor an isolation method includes: purity requirements, which depend upon theintended use and application of the purified Mab; the scale-up potential of themethod; and the stability and bioactivity of the recovered material. Here, theutility of the DEAE MemSep anion exchanger is examined in a single-steppurification of a Mab (anti-scorpion toxin II-3C5) from murine ascites.


Employing the membrane-based MemSep anion exchange (DEAE) cartridge,variations in chromatographic conditions such as the pH of the loading buffer,and the gradient elution conditions were examined in the isolation of anti-scorpion toxin antibody from murine ascites.

The loading buffer (A) was comprised of 20 mM Tris-HCl, at pH 7.2 or pH 8.0 asindicated. The elution buffer (B) was made up of 20 mM Tris-HCI and 1.0 M NaCI,pH 8.0. The mouse ascites used in this study was developed according to thepublished procedure (Bahraui, E., et al., J. Immunology, 141, 214-220, 1988), andwas the kind gift of the author.

Before applying the ascites to the DEAE MemSep, 200 ul was filtered through aMillex-GV 0.22 um filter unit, followed by a four-fold dilution with buffer A. Forall chromatographic runs, the MemSep was equilibrated with buffer A at 3 ml/minfor at least 5 minutes prior to sample loading. A linear gradient and/or stepgradient was employed as indicated (see Fig. 1 & 2). A flow rate of 3 ml/min wasused with detection at 280 nm.

Each of the major peaks eluting from the MemSep anion exchanger was collectedand subjected to further fractionation by gel filtration chromatography using aWaters Protein-Pak 300SW column. This column was first equilibrated with therunning buffer which was comprised of 50 mM sodium phosphate buffer, 150 mMNaCI, pH 7.4. The column was then calibrated using High Molecular Weight Markers(Sigma).

The peak designated as number 2 in Figure 1 & 2 consistently resulted in asingle peak on the gel filtration column which co-chromatographed with themolecular weight of the scorpion toxin II Mab. The purity of the recovered Mabwas further examined and verified employing SDS-PAGE.

Tile purification of Mabs from ascites using DEAE MemSep is an attractivealternative to affinity techniques for several reasons. Near physiological pHconditions may potentially enhance the stability and activity of the recovered Mab.The ion exchange procedure is directly scalable. In addition, unlike protein-A, ion

exchange may be used to purify any monoclonal, and may purify monoclonal fromhost IgG.

Figure 1. Separation of anti-scorpion toxin I1 Mab (3C5) from mouse ascite:_with loading buffer at pH 7.2 under various gradient elution condition_. Thetarget Mab eluted in peak number 2.

_: Mouse_ 200 pl.l1:241_.. _: D_ Me,nS_ 1000

Burr A: Trls-HQ 20 n_'l pH - 7.2

1A. c,_- B_er I_: B.ut_ A �IM Na(_ pH ° 8.0

! : tllB. _ i

j "0% B [o,' I m;n. 0"4.B for !.75 mi_. O"4.B for 2 m_.O,50% B tor S mi_ O.lO%B for S min. 2.5% B kx"2 min.

5_ B for 2 mi_lO%B lot I m_n.

Figure 2. Separation of anti-scorpion toxin 11Mab (2,C5) from mouse asciteswith the loading buffer at pH 8.0 under the given elution conditions. Theanti-scorpion toxin ri Mab is contained in peak number 2.

s,,,_: Mo_,,_:_ 200 _. (l:,qc.,_k,nm: Of.AEM,,,,.S_ 1000e,,n%.rA: T_s.H_20 mHi:_ - e "_,ff_,'B:Bd_rA., 1.0MI_FH- 8.0

2

"1" * 213. '1.. 2C.. ,....ii I

- "N'"' "zle- • "

0'% B k_" 1 rain. 0% n (o, 1.75 mia. 0% B for 1.5 mln.0,.$O% B k:_"5 mln. O-lO_ B lot 5 rain. 2..5% 8 k_ 2_5 rain.

5% B k_" 2 m_n.

SEPARATION OF CYTOCHROME C FROM FOUR DIFFERENT SPECIES

INTRODUCTION

The separation of complex protein mixtures always presents a challenge toliquid chromatography. If the proteins are very similar in amino acid composition,shape and size the task becomes even greater. These separations require thehighest column efficiency and resolution.

In this study MemSep Chromatography was examined in the separation ofcytochrome c from four different species. A mixture of these cytochromesare difficult to separate on any chromatography system. There are three aminoacids out of 104, and a single charge difference between cytochrome c fromchicken, cow, pigeon and horse. Their molecular weights differ by I11 units. Theseparation of these four proteins require the column to perceive the very smalldifference in charge density and composition.


All separations were performed using the Waters 650 Advanced ProteinPurification System and Auto-Blend* method for rapid optimization of proteinseparations. The three chromatograms were run in 33 minutes (11 minutes each)under the following conditions.

The three separate chromatographic runs were conducted at pH 5.0, 5.25 and 5.6as indicated (Fig. 1, 2 and 3). The loading buffer in each run was 20 mM sodiumacetate. The proteins were eluted using a linear gradient from 0 to 0.15 M NaCIin 20 mM sodium acetate buffer. The flow rate was 5.6 ml/min with detection at

390 nm. A total og 5 mg of the protein mix was applied to the column for each rut,

Figure 1. Fractionation of the cytochrome mix at pH 5.0 on the CM MemSep-1000



Pigeon j

Cow Horse F_Initial Conditions: 20 mM NaAc at pH 5Final Conditions 20 mM NaAc and 0.15 M NaCI at pH 5Gradient Linear in 8.5 minutesFlow Rate: 5.6 ml/minute

Detection: 390 nm Chicken till i

FIG. 1

..s

I I I I I I I I I I -

Initial Conditions: 20 mM NaAc at pH 5.25M a 0111

Final Conditions 20 mM NaAc and 0.15Gradient Linear in 8.5 minutesFlow Rate: 5.6 ml/minute

Detection: 390 nm

FIG. 2

Initial Conditions: 20 mM NaAc at pH 5 NFinal Conditions 20 mM NaAc and 0.1Gradient: Linear in 8.5 minutesFlow Rate: 5.6 ml/minute

FIG. 3 Detection: 390 nm

O 1 _ _ ,1 £ (_ 7 }3 9 10 11

poster presen¢ tion faseb - waters corporation · poster presen¢_tion faseb/asbmb symposium -...

Documents