_:_ Essentialsin "

ASBMB/AAI '90Poster Presentation

A Comparison of Preparative Reverse PhasePacking Materials for the Isolation

of Proteins and Peptides

Charles Phoebe Jr. and George Vella (Sponsor: J. Li)

Waters Chromatography Division, Milford, MA 01757

Waters. The absolute essential MilliporeWatersChromatographyDivisiOncorporation Waters /for bJoresearch. 34 Maple St. Divisionof MILLIPORE

Milford, MA 01757(508) 478-2000

1_o

t '

r

.r

INTRODUCTION:

High-performance chromatographic materials, representingvarious chromatographic modes of separation, provide themost powerful preparative separation tools for the isolationand purification of proteins and peptides. One of thesemodes (reverse phase) separates proteins and peptides bytheir hydrophobic retention on the surface of a bonded-silicapacking material and desorption with eluents of increasingorganic content. Packing materials utilized in this modegenerally use pore enlarged silicas (~300 A) bonded withalkyl-silanes of different alkyi chain lengths (C4, C8 or C18).In this work, three different reverse-phase packing materials(Delta-PakTM C18, Bondapak® C18, and VydacTM C18), withparticle sizes ranging from 15 to 20 microns and packed in 8x 100 mm RadiaI-Pak TM cartridges, have been evaluated

/ with standard protein and peptide mixtures. Threeexamples of scaled-up preparative separations on 25 x 100,200 or 300 mm stackable Prep-Pak TM Cartridges, packedwith Delta-Pak TM C18 material, have been used for theisolation of human milk whey proteins, the purification of asynthetic peptide (neurotensin) and the isolation ofglycopeptides derived from a tryptic digest of fetuin.

L

1,

PEPTIDE SEPARATIONS:

Reverse-phase chromatography is recognized as thepreferred mode for the separation of peptides. Figure 1shows the separation of a peptide standard mixture which isutilized for monitoring column performance, efficiency,selectivity and resolution (1). As can be seen, all threecolumns give similar retentions for these peptides with theVydac material being slightly less retentive and the Delta-Pak being slightly more retentive of these three packingmaterials. All three materials fail to resolve the $1/$2 pair(structures shown in Figure 2) due to the decreasedefficiency of these 15-20 micron preparative materials.Higher efficiency 5 micron materials easily resolve this pairof peptides (Figure 3). A similar comparison of all three

_i packing materials utilizing a tryptic digest of reduced andalkylated fetuin is illustrated in Figure 4. Once again, littledifference is observed between the three packing materials.

A general scale-up strategy is demonstrated with the threeexamples in this poster where the first step is thedevelopment of a high resolution separation using a 5 micronC18 packing material which can be used to evaluate thestarting mixture as well as the fractions obtained from thepreparative isolation• The second step is the optimization ofthe separation on 15-20 micron preparative packing materialin scaling !00 mm ...,,,,a 8 x RadiaI-Pak cartridge _,_ thedetermination of the maximum sample load with a loadingstudy. Using scale-up equations (Figure 5) the separation isthen transferred to preparative 25 mm x 100, 200, or 300 mm

1

6

Prep-Pak cartridges packed with the same 15-20 micronpacking material.

The first example (Figures 6-9) illustrate the identificationand scale-up isolation of glycosylated peptides from thetryptic digest of fetuin using the Delta-Pak C18 material.Figure 6 shows a high resolution peptide map using a 5micron Delta-Pak C18, 300 ,&,column. In order to identifywhich peptides are glycosylated, the tryptic digest of fetuin istreated with N-Glycanase®. This enzymatic treatmentremoves the N-linked oligosaccharides present on anypeptide resulting in peptides which will be more hydrophobicand will elute later _nthe chromatographic peptide map.Figure 7 shows a high resolution peptide map of fetuinfollowing N-Glycanase treatment where two peaks (comparewith Figure 6) are now missing and two new peaks haveappeared. It is now known that peaks 1 and 2 (Figure 6) areglycosylated and will be targeted for preparative isolation.Figure 8 illustrates the optimized separation on a 8 x 100 mmRadiaI-Pak cartridge packed with 15 m_cronDelta-Pak C18material. Using the scaling equations (Figure 5) theseparation is now transferred to a 25 x 100mm Prep-Pakcartridge packed with 15 micron Delta-Pak C18 material(Figure 9).

Example two (Figures 10-12) illustrates the purification of asynthetic peptide (neurotensin) from the.deletion ....._,=,,uencesusing the Delta-Pak C18 material in the same manner asexample one.

1. C.T. Mant and R.S. Hodges, LC.GC Magazine of Liquid and GasChromatography 4:250-254 (1986)

Comparison of Preparative Reverse 1_ase Solvent A: Water/0.1%TFA '

Palfings- Peptide Retention Standar_l Solvent B: Acetonitrile/0.1°/o TFA .- -. "Gradient: 10%B to 40%B in 40 rain. ;

O. t0 - Flow Rate: 2 ml/min. ,-. ,.

- VydacTM C18 300,& Peak ID: 1. $1/$22. $3- (15-20 H) RadiaI-Pak TM 1 2 3 A 4 3, $4

____ _ Cartridge (8x100 mm_ _,_ JL 4.$5

0.00 ]- ..... 1 t .............. I.........................

' E

,_- 0._.0 4

- Delta-Pak TM C18 300A 1 3

OILl_ _ (15 g) Radiai.PakTM _ ___

z Cartridge (8x100 mm)

= \rr _

m -< 0.00 I _ _................... r- .............

0._.0

-- Bondapak® C18 300A 4_. (15-20 It) RadiaI-Pak TM 1

._._ Cartridge (8x100 mm) _ __,. ,_

0.00 .....,--r-Tr--,--r--r--, ,-l---r-,-_--,---r---,----'r-T---r---l----r--r , , , r--F1.---r--l--ir--1-.,-,.-r--1---,-,-T--I-_--, , , ,. ,-, ,--,._

0 5 _.0 '15 20

MINUTES

Figure 1

• o

Peptide Retention Standard

Figure 2

C_i_. arison of Analytical Reverse Pi_e SolventA: Water/0.1% TFAPairings- Peptlde Retention Standar_ SolventB: Acetonitrile/0.1"%TEA .: ""

Gradient: 0%B to 60%B in 60 rain.- Col.Temp." 30° C

300 Peak ID: 5 1. $12. $2

my 4 3. $34. $4

200 - VydacTM C18 300A 3 5. $5(218TP5414, 4.6 x 150 mm) 12

E Flow Rate: 0.7 ml/min.(--.

,,ooo a_a_ VL.Z J_< -O0 0 i I I i ! ]n-'0rn<t:

300 5

mV 4

200 - Delta-PakTM C18 300A 3(3.9 x 150 ram, HPI) 12Flow Rate: 0.5 rnl/min.

100

_.,o_arison of Preparative Reverse Solvent A: Water/0.1% TFA .,.

Palmngs - Tryptic Digest of Fetuin Solvent B: Acetonitrile/0.1%TEA. . :Gradient: 0%B to 60%B in 135 min_

-7 Flow Rate: 2 ml/min. ".

0.15 ,. Vydac TM C18 300A _ II I_. j_

__ (15-20 li) Radial-Pak TM Jl II ^, II Cartridge (8x100 ram) I I Jl I'l I

0.00 ........ _.................. i .....

Er"

,xt"

e_ 0.t5

Delta-Pak TM C18 300A A

LUOz - (15 g) RadiaI-Pak TM

< Cartridge (8x100 mm) J__v__) L_

rnn-"

0 - _ _ U _'N_'_" I

, .133

0.00 ............................................................................... I

0. t5 Bondapak® C18 300A j... (15-20 I1) RadiaI-Pak TM

I

_ Cartridge(Sxl00mm) __ _' --.,jL _0.00 - ---r--lr--_--r--l---r-t---l--1 -_- -i-Tr--I-T-l---i--r-ll---i--l-, ,--"r--l-r 1 l -i S-l- i-i , --i -1---i l--i i i -i , , , _ _ -

0 20 40 60 80

MINUTES

Figure 4

SCAUNG.OPERATING PARAMETERS

Sample Load

Scale the sample load accordingto the internal volumesof the columns, as follows:

(D1)2L1Loadprep - LoadanalyticaI x

(D2)2L2

Where: D1 - Internaldiameter of the preparative AP column (cm).L -- Length of the preparativeAP column (cm).

D2 = Internaldiameter of analyticalAP column(cm).L2 -- Lengthof the analyticalAP column (cm).

Row Rate

Scale up the flow rate to maintain the same linear velocityin the preparative as in thepreviouslydeveloped analyticalseparation,as follows:

(D1)2Qprep -- Qanalytical x

(D2)2

Where: Q = Flow rate (ml/min).

D1 = Diameter of preparativecolumn (cm).

D2 - Diameter of analytical column(cm).

Gradient DuraJon

Scale up the durationof the gradient so the preparative gradient occursover the samenumber of column volumesas the analyticalgradient, as follows:

(Vp)(GDa)(Qa)GDprep =

(Va)(Op)

Where: Vp = Void volume of preparative column(ml).GDa = Gradient durationof analytical method(min).

Qa = Flow rate of analyticalmethod (ml/min).

Va = Void volumeof analytical column(ml).

Qp = Flow rate of preparative method (ml/min).

Figure 5

...... _.,,, , I-.I I I1-/I.-- IVlt'_l U_JII_I_ _JU _UU/'_

DELTA 8 3.9mmx150mm() (,I)

I'

...... ".,

',.JO................................. ".................. 'J#

]" 1' 1 2

0 I- i

lI t-_:-" ............ :......... r

II

o

I

>

- ' ' f ' 1 1 .... I .... I ' I .... I .... 1 ' I '-' '-' -'--(mi,,) l) Ill 2l) 3(} 4l) r_() (,(I

Figure 6

l) I () 2() 40 _(I -(}1)

After Treatment With N-Glycanase®

'J_ X

• lI,

o

o L • . ............................. 2

- 1 i-.,.4

• I' ' _ I ' I l , r 1 .... I ' '' "' ' I I ' ' ' I ' I ' ' -' I ' ' I '--"- _-i ll) } ] 2i) 30 4t) c;I} {,()

Figure 7

I]CM 8 x 10

)PTIMIZ/tlON OF PEPTIDE MAP USING 15u co,urn.-Delta-PakC18, 15micron0300ASample: Reducedandalkylatedfetuintrypticdigest

DELTAPAK C18 8x10 CARTRIDGE COLUMN Injection: 10ulMobile Phale: SolventA: Waterwith .1% TFA

SolventB: Acelonildlewith.1% TFA ' ....__ Gradient Conditions: 0-29.2 mln.,0-25% B

29.2-122.5rain.,25-60% B122.5-134.1rain.,holdat 60"/,B134.1-135.5min.,60-0%B

FlowRate: 2.0 mPmin.Detection: UV at214 nm

2

6.00- 1t

ta 4 oo-4J "r-t

g

x 2.00- _

0.00-

, i , I ' ' ' I ' ' _ I ' ' ' I ' ' b I ' ' ' I ' ' ' I ' ' 't.00 2.00 3.00 4.00 5.00 6.00 7.00

x J0t mtnutes

F' 8

OS_ILE-UP OF PEPTIDE MAP USING 1_ 300A co,.m...De,,a-Pa,C,e.,Sm,c.en.3oo^w,i,g ,umn,.Sample: ReducedandalkylaledfetuintrypticdigestDELTA PAK C18 25x10 CARTRIDGE COLUMN ,.Jec,o..SOOu,

Moblle Phase: SolventA: Waterwith .I % TFASolventB: Acetonltrilewilh.1%TFA

Gmdlent Conditions: 0-29.2 mln.,0-25% B29.2-122.5rain.,25-60% B122.5-134.I min.,holdat 60"/, B134.1-135.5rain.,60-0%B

FlowRate: 19.5 ml/mln.2 Detectlon: UVat214 nm

' 3.00 1

,...,2.00-o

.x t.00-

O.

I , I ' I , I ' ' ' I ' ' '2.00 4.00 6.00

x t0 ! minutes

Figure 9

ANALYTICAL SCALE A Crude syntheticpeptide

COLUMN: Delta-PakC18 5u, 300A, 3.9mmx150mm Neurotensin5mg/ml '.SAMPLE: Synthetic Neurotensin 10ul injected

_ SOLVENT A: Water/0.1%TFA250 SOLVENT B: Acetonitrile/0.1%TFA

mV GRADIENT:5%B hold2min, 5%B to 90%B in 30rain.- FLOW RATE: I ml/min

200 DETECTION • UV at 214mn

150

, iO0 - I I

50

0 '- -_'- i l I ! I

B Purifiedfraction from prep run100ul inject

250

mV2O0

50 ___.____,_,_, J _____ _._..___.__,___0 _._ , .|ll. | ' ' '-- "r'--r" , ,l l , , ! I ' ' ' ' ' I i I | I .... ' I ' | | I "'' I | I r--TT-I---T--[---I ' I-T--I-- I l ,. -;-':T--

0 5 10 15 20 25

Minutes

Figure 10

LOADING STUDY (OPTIMIZED FOR 8X1001T_I_ARTRIDGE)COLUMN" Delta-PakC18' 15u, 300A, 8x100mm cartw_eSAMPLE: Synthetic neurotensin

SOLVENT A: Water/0.1%TFA A - LOAD: 40ul (200ug) .IllV SOLVENT B: Acetonitrile/0.1%TFA .

GRADIENT:5%B hold2min,to 27%B in 8min, holdfor 5min27%B to 90%B in 12min,holdfor 3min.

1dO0 FLOW RATE:2ml/minDETECTION: UV at 230nm

t

,ooi.0 "l-......--_...... i.......... I "...."---'-"----_- I_ I ........--I..............

I

mVLOAD: 100ul (500ug)

,400

200

0 T ..... r- -T ---I......... I............

mY C - LOAD: 200ul (1000ug)

200 ,

0 -+-1 _,', 1.-:,_ , , , , 1-,-,'--I _, -_ , i , i', ,-r- I ,-,--r-,-,--rr-"-,--,--V-, ,1-,, ,-,-TT- V, ,,-,--,-, , , ,--

0 5 10 '15 20 25

Hinutes

Figure 11

bL;ALI::-UP TO 25X100mm CARTRIDGE

COLUMN::Delta-PakC18, 15u, 300A, 25x100mm ca_e .-SAMPLE: SyntheticNeurotensin .-SOLVENT A: Water/0.1%TFASOLVENT B: Acetonitrile/0.1%TFAGRADIENT:5%B hold2min,to 27%B in8min, holdfor 5min

._ 27%B to 90%B in 12 min, holdfor 3min. LOAD: 1500ul (7.5mg)500 FLOW RATE: 19.5 ml/min.

DETECTION: UV at 230nmmV

450

350

300

-4

250 12O0l

5O

I--I_ r-l-t'1 i r1-VF'iT-lFr_-r_|-_-rT-r--lr-l_i i r]--i-rr_-]'r'i , , ] i-l--r_--r _--r 177-= r-ir__ = ]7--

0 5 iO 15 20 25

Minutes

Figure 12

Cgiliparison of Preparative Reverse/ase Solvent A: Water/0.1%TFAPdWings - Standard Proteins Solvent B: Acetonitrile/0.1%TEA" -,

Gradient: 25%B to 46%B in 10 min.- to 70%B in 5 rain

0.6 3 FlowRate: 2 ml/min._ VydacTM C18 300A /_ Peak ID: 1. Ribonuclease A/B

(15-20 I_)Radial-PakTM 2 ,_l 4 2. Bovine InsulinCartridge (8x100 mm)1 A _,_:. Lysozyme- ./':_._._ _k_ 5 4. Horse Myoglobin

• Ovalbumin

0.0 I i r i I I T I I

' E _ 3'- 0.6

Delta-PakTM C18 300A 4uJ - (15 p) RadiaI-PakTM 2

<zO Cartridge (8x100 mm) _ j_

- 1rrO

< 0.0 I ! l ] ] r [ i..........

__ 30.6

Bondapak® C18 300A 4- (15-20 _) RadiaI-PakTM 2

Cartridge (8x100 mm) A

0.0

0 2 4 6 8 I0 12 14 16 18

MINUTES

Figure 13

RP-HPLC Separation _I Human Milk Whey

_80

mV _ Column: Delta-PakTM C18 5p.,300A(3.9x150 mm) PEEK

Sample: Human MilkWheySolventA: Water/0.1%TFA

- SolventB: Acetonitrile/0.1%TFAGradient: 0% to 60%B in 90 min.FlowRate: 0.5 ml/min.Load 20 plDetection UV at 214 nm

0 20 40 60 R0

MINUTES

Figure 14

------ ellO lized 8x10 Scaling Cartridge S ration Column: Delta-PakTMC18 151x','300A ,,

of iman Milk Whey (sxloomm)cartridgeSample: Human Milk Whey , ..Solvent A" Water/0.1% TFA

0 16 I Solvent B: Acetonitrile/0.1% TFA

• 1 Gradient: 30%B to 48%B in 12 min.,

to 80%B in 8 min.Flow Rate: 2 ml/min.

Load: 35 pl

0.00-1F I rl-I T-FI'-I |-I-I-FIT! rt-t- l I-FIIT-FT lrl-i-ll-I t I'1 i] i | i i i !'1 i i ! i I i t'rl"il i i i I I-1F II-FIF N I-r'TTI"I ITI--ITITI i'Ti I I'it-FI i i I I Frl-r-

0 2 4 6 8 10 12 14 _6 18

MINUTES

Figure 15

mlmmmmmi_.'--

Lo ing Study - 8x10 Scaling Cartrid_t column: Delta-PakTM C18 , )A"(8x100 mm) cartridge ' ,,-

Sample: Human Milk Whey .,.

0.6

- // Load: 175 I_1

;_ (1.6 mg protein)2L0.0

I 1 r I T I I I............ T....

"-7

E 1 5o Load: 700 plcooj - (6.3 mg protein)LUoz<O0

" ......... I................ I ....... 1 _ i I ............. t t -I0 0CO<

"_" ,.,uj,uvdn,son o! fix10 Scaling Cartridg_o a Column:Delta-PakTM C18 1_, 300A.25q_0 Preparative Cartridge (8xloomm)cartridge

Sample: Human MilkWhey oO

- Load: 700 I_1(6.3 rag)`1.5 FlowRate: 2 ml/min.

0

,(N

,,, /z 0 0<_ " I I i I i I I I I

n--0

,-n Column: Delta-PakTM C18 151_,300A<2.0 (25x100 mm) cartridge

Sample: Human Milk WheyLoad: 7 ml (63 mg protein)

- _ Flow Rate: 19.5 ml/min.

0 2 4 6 8 10 '12 14 `16 18

MINUTES

Figure 17

S_-Up To 25x10/20/30 Stackable _rtridges Column: Delta-Pak_MC18 1! I,_ "._

(CIWl'stantLinear Velocity) sample: Human MilkWiley _ "SolventA: Water/0.1% TFA ,

- Solvent B: Acetonitrile/0.1% TFA •0.20 Flow Rate: 19.5 ml/min.

A-ONE CARTRIDGE (25x100mm)- Gradient: 30%B to 80%B in 20 min.

Load: 7 ml (63 mg protein)

0.00 i i i i

,EcOco 0 20

" B-TWO CARTRIDGES (25x200mm)w - Gradient:30%B to 80%B in 40 rain.Oz _ Load: 14 ml (126 mg protein)<oD

0 -

< 0.00 I I I i

O. 3 C-THREE CARTRIDGES (25x300mm)Gradient: 30%B to 80%B in 60 min.

- Load: 21 ml (189 mg protein)

0.0 l _' ' '-' "

0 20 40 60 80

MINUTES

Figure 18

o OScikt-Up To 25x10/20/30 otackable tridges Column:Delta-PakTMC181 )_, .

(Inllbasing Linear Velocity) Sample: HumanMilkWheySolvent A: Water/0.1%TFA ' ""

0.6 - Solvent B: Acetonitrile/0.1% TFA *Gradient: 30% B to 48% B in 24 min.,

_ to 80% B in 16 min.

A-o.ECA.T.,DGEI2 x00m l- Load: 7 ml (63 mg protein)

Flow Rate: 9.75 ml/min._-- --.......

J

0.0 I i J = i I i I

Et-

_o 0 1504

LU B-TWO CARTRIDGES (25x200mm)Oz Load: 14 ml (126 mg protein)< _ _ Flow Rate: 19.5 ml/min.rn

Yn"

o \O3

<

0.0 I i i i I I l t

0.6

- C-THREE CARTRIDGES (25x300mm)

Load: 21 ml (189 mg protein)

Flow Rate: 29.25 ml/min.

0.0 , , , , , , , i J i , , , , , , , , , i ,', , , , , , , , I , , , ,., , , , , i _ , , , , , , , , i , , ,', , , , , , I _ _ _ _ , , , , , i , , , , , _'_ _ _ i _ _ _

0 5 10 t5 20 25 30 35 40

MINUTES

' Figure 19

•°

PROTEIN SEPARATIONS:

Although not utilized to the same extent as in the purificationof peptides, reverse-phase chromatography is a powerfultool for the purification of proteins. Figure 13 shows theseparation of a standard mixture of proteins on the threepreparative 15-20 micron C18 packing materials. Asobserved with the peptides, little difference is noted betweenthe three materials for this separation.

The third example in this presentation (Figures 14-19) showsa preparative isolation of human milk whey proteins usingthe Delta-Pak C18 packing material. Figure 14 shows a highresolution profile of the constituents of human milk whey with

_t a 5 micron Delta-Pak C18 column. In order to preparativelyisolate the major milk whey proteins and to isolate sufficientquantities of the minor proteins and peptides present, theseparation was optimized on a 8 x 100 mm scaling Radial-Pak cartridge packed with 15 micron Delta-Pak C18 material(Figure 15). A loading study on the 8 x 100 mm cartridgewas performed (Figure 16) with the maximum loaddetermined to be 700 _1of milk whey. The separation wasthen transferred to a 25 x 100 mm Prep-Pak cartridge(Figure 17). Using the scale-up equations (Figure 5) , thescale-up factor based on the differences in column internaldiameter (since both columns are 100 mm in length) is 9.7.Therefore the flow rate will be increased from 2 ml/min, to19.5 ml/min and the sample load increased from 700 !_1to 7ml of milk whey. In order to increase the sample load,

- • d

increasing 100 mm lengths of Prep-Pak cartridges wereadded so as to make either a 200 or 300 mm column. Thisallowed the increased injection of 14 ml (126 mg protein) ona 200 mm column or 21 ml (189 mg protein) on a 300 mmcolumn (Figures 18 and 19). Two different modes ofpreparative scale-up are illustrated in Figures 18 and 19.Figure 18 utilizes a constant flow rate of 19.5 ml/min.(constant linear velocity) with increasing gradient times asthe column length increases (20 min. for 100 mm column, 40min. for 200 mm column, and 60 min. for 300 mm column).Figure 19 shows a constant gradient time of 40 min. withincreasing flow rates (increasing linear velocity) as columnlength increases (9.25ml/min for 100 mm column, 19.5ml/mnnfor 200 mm column, and 29.25 ml/min for 300 mm

O column). Both modes of scale-up result in increased sampleload and resolution with increasing column length. However,using the increasing linear velocity mode the overall analysistime _s the shortest with the highest sample load andresolution (21 ml of milk whey with 300 mm column length,last chromatogram, Figure 19).