Journal of Chromatography A, 816 (1998) 29–37
`
`Detergent extraction of herpes simplex virus type 1 glycoprotein D
`by zwitterionic and non-ionic detergents and purification by ion-
`exchange high-performance liquid chromatography
`Sytske Welling-Wester, Matty Feijlbrief, Danny G.A.M. Koedijk, Gjalt W. Welling
`Laboratorium voor Medische Microbiologie, Rijksuniversiteit Groningen, Hanzeplein 1,9713 GZ Groningen, Netherlands
`
`*
`
`Abstract
`
`Detergents (surfactants) are the key reagents in the extraction and purification of integral membrane proteins. Zwitterionic
`and non-ionic detergents were used for the extraction of recombinant glycoprotein D (gD-1) of herpes simplex virus type 1
`(HSV-1) from insect cells infected with recombinant baculovirus. The highest yield was obtained with the two alkyl
`carboxybetaine detergents
`(N-dodecyl-N,N-dimethylammonio)undecanoate
`[DDMAU,
`critical micelle
`concentration
`(CMC)50.13 mM] and (N-dodecyl-N,N-dimethylammonio)butyrate (DDMAB, CMC54.3 mM). Therefore these zwit-
`terionic detergents were used as additives to the elution buffers in ion-exchange high-performance liquid chromatography
`(HPIEC) to purify gD-1 of HSV-1 from the extracts. The non-ionic detergent pentaethyleneglycol monodecyl ether (C E )
`10
`5
`that was used in earlier studies [R.A. Damhof, M. Feijlbrief, S. Welling-Wester, G.W. Welling, J. Chromatogr. A, 676 (1994)
`43] was used for comparison. Two columns were used, Mono Q and Resource Q, at 1 and 5 ml /min flow-rates, respectively.
`The results show that the detergents DDMAU and C E are superior to DDMAB, when the detergents were used as
`10
`5
`additives to the elution buffers at 0.2% (w/v). With 0.2% DDMAB in the eluent, purification of HSV gD-1 was not possible.
`Detergents with a high CMC may be less suitable as additives in elution buffers. HPIEC at flow-rates of 1 and at 5 ml/min
`showed satisfactory results. At 5 ml/min HSV gD-1 was mainly concentrated in two eluent fractions. The highest recovery
`of gD-1 was obtained either by chromatography of a C E extract using a Mono Q column at a flow-rate of 1 ml /min or by
`10
`5
`chromatography of a DDMAU extract using a Resource Q column at a flow-rate of 5 ml/min.
`1998 Elsevier Science
`B.V. All rights reserved.
`
`Keywords: Mobile phase composition; Surfactants; Glycoproteins; Proteins; Membrane proteins
`
`1. Introduction
`
`Detergents (surfactants) are the key reagents in the
`extraction and purification of
`integral membrane
`proteins [1]. Solubilization of membranes including
`its proteins or selective extraction by detergents is
`often the first step in the purification of an integral
`membrane protein.
`Detergents are lipid-like substances. Like the
`
`*Corresponding author.
`
`major constituent of the membrane, the phospholipid
`molecule,
`they contain a hydrophilic head and a
`hydrophobic tail. They are able to compete with the
`lipids in a bilayer and are more hydrophilic than the
`lipids. As a consequence, detergent–protein com-
`plexes are soluble in aqueous solutions, and the
`detergent molecules, in mimicking the lipid mole-
`cules, help to maintain the native configuration of the
`membrane proteins during a purification procedure.
`There are several categories of detergents [1–10]:
`(a) ionic detergents e.g., sodium dodecyl sulfate
`
`1998 Elsevier Science B.V. All rights reserved.
`0021-9673/98/$19.00
`PII: S0021-9673( 98 )00288-X
`
`KASHIV EXHIBIT 1030
`IPR2019-00791
`
`Page 1
`
`(cid:211)
`(cid:211)
`

`

`30
`
`S.Welling-Wester et al. / J. Chromatogr. A 816(1998)29–37
`
`(SDS), (b) bile salts, which are mild ionic naturally
`occurring detergents, e.g., cholate,
`taurodeoxycho-
`late, (c) mild non-ionic detergents and (d) mild
`amphoteric detergents. Detergents of categories c
`and d are particularly relevant for extraction and
`purification by ion-exchange high-performance liquid
`chromatography (HPIEC) and they are listed in
`Table 1 together with their critical micelle con-
`centration (CMC).
`The choice of a suitable detergent may depend on
`several factors, i.e., CMC, hydrophile–lipophile bal-
`ance number (HLB), micellar molecular mass, cloud
`point, UV-transparency, effect on biological activity
`and price.
`In the present study we will focus on the CMC.
`The CMC is the concentration of monomer at which
`micelles i.e., spherical bilayer aggregates of de-
`tergent molecules, begin to form. Triton X-100 has a
`low CMC, 0.24–0.30 mM, and is difficult to remove
`by dialysis. Octylglucoside has a high CMC, 25 mM,
`and can easily be removed by dialysis. Therefore,
`
`further studies to be carried out with a particular
`membrane protein may determine the choice of
`detergent. Some studies require a soluble protein–
`detergent complex in order to maintain biological
`activity. In such cases the CMC is of less impor-
`tance, although the relatively high concentration of
`detergent present
`in extracts may affect
`the bio-
`logical activity to some extent. Similarly, high
`concentrations of certain detergents may interfere
`with immunological assays e.g., an enzyme-linked
`immunosorbent assay (ELISA).
`In earlier studies, we used the integral membrane
`proteins of different viruses as a model for the
`development of methodologies for the purification of
`membrane proteins with different detergents and
`different modes of high-performance liquid chroma-
`tography (HPLC) [8,12–14]. This resulted in a two-
`step elution protocol with a non-ionic detergent at
`low and high concentration in the eluent for HPIEC
`[15–17].
`In the present study, a number of non-ionic and
`
`Table 1
`CMC of non-ionic and amphoteric detergents
`
`Detergent
`
`Non-ionic
`Triton X-100
`Nonidet-P40
`Triton X-114
`Penta-ethyleneglycol monodecyl ether
`Penta-ethyleneglycol monododecyl ether
`Emulphogen BC-720
`Lubrol PX
`Thesit
`Brij 35
`Tween 80
`Octylglucoside
`Dodecyl-b-D-maltoside
`Hecameg
`
`Mega-10
`
`Amphoteric
`3-[Cholamidopropyl)dimethylamino]-1-propanesulfonate (CHAPS)
`Zwittergent 3-12 (sulfobetain SB 3-12)
`(N-Dodecyl-N,N-dimethylammonio)undecanoate (DDMAU)
`(N-Dodecyl-N,N-dimethylammonio)butyrate (DDMAB)
`Dodecyl dimethylamineoxide
`
`Description
`
`9.6
`
`9
`
`7 – 8
`
`5
`
`5
`
`8
`
`tert.-C fE
`8
`tert.-C fE
`8
`tert.-C fE
`8
`C E
`10
`C E
`12
`C E
`12
`C E
`12
`C E
`12
`C E
`23
`12
`sorbitan E
`C
`18:1
`C glycoside
`8
`C maltoside
`12
`6-O-(N-Heptylcarbamoyl)-
`methyl-O-D-glucopyranoside
`N-(D-Gluco-2,3,4,5,6-penta-
`hydroxyhexyl)-N-methyldecanamide
`
`9 – 10
`
`9
`
`20
`
`Bile acid derivative
`Sulfopropylammonium compound
`Alkyl carboxybetaine
`Alkyl carboxybetaine
`[C N (CH ) O ] (above pH 7)
`12
`3 2
`
`1
`
`2
`
`CMC (mM)
`
`0.24–0.30
`0.29
`0.20
`0.69
`0.049
`0.08
`0.02–0.1
`,0.1
`0.091
`0.012
`25.0
`0.20
`19.5
`
`6.2
`
`4–6
`3.6
`0.13
`4.3
`2.2
`
`Data are from Refs. [1–11]; C E , x refers to the number of C atoms in the alkyl chain and y to the average number of oxyethylene units; a
`x
`y
`phenyl ring is designated by f; tert.-C refers to a tertiary octyl group and C
`indicates an 18-carbon chain with one double bond.
`8
`
`18:1
`
`Page 2
`
`

`

`S.Welling-Wester et al. / J. Chromatogr. A 816(1998)29–37
`
`31
`
`amphoteric detergents will be compared with respect
`to the extraction of a recombinant integral membrane
`protein i.e., glycoprotein D of herpes simplex virus
`type 1 (gD-1) [15,18–22] from cells infected with
`recombinant baculovirus. Subsequently,
`two de-
`tergents with a large difference in CMC will be
`compared with regard to their suitability as additive
`to the eluents for HPIEC using either a Mono Q or a
`Resource Q column. The two detergents are the
`alkylcarboxybetaine compounds, (N-dodecyl-N,N-di-
`methylammonio)undecanoate
`(DDMAU, CMC5
`0.13 mM) and (N-dodecyl-N,N-dimethylammonio)-
`butyrate (DDMAB, CMC54.3 mM) [10,23].
`
`2. Experimental
`
`2.1. HPIEC
`
`Chromatography was performed with a system
`consisting of an LKB Model 2150 pump (Pharmacia
`Biotech, Roosendaal, Netherlands), a Rheodyne
`(Inacom, Veenendaal, Netherlands) Model 7125 in-
`jector and a Waters Model 441 detector (Millipore–
`Waters, Etten-Leur, Netherlands). HPIEC was per-
`formed with either a Mono Q HR 5 /5 column (50
`mm35 mm I.D.) (Pharmacia Biotech) or a Resource
`Q column (30 mm36.4 mm I.D.)
`(Pharmacia
`Biotech). The flow-rate was 1 ml/min when the
`Mono Q column was used and 1 ml/min or 5
`ml/min when the Resource Q column was used (see
`below). The samples (500 ml, containing 10–12 mg
`protein) of the infected cell extracts [containing 1%
`(w/v) of detergent] were centrifuged at 14 000 g at
`48C for 5 min and diluted with 20 mM Tris–HCl, pH
`7.8 (buffer A), to a final detergent concentration of
`0.01%, prior to application to the column. After
`sample application,
`the column was washed in
`several steps. The first wash step was isocratic
`elution for 15 min with buffer A. A second wash
`step was elution for 10 min with buffer B (20 mM
`Tris–HCl, pH 7.8 containing 0.5 M NaCl). The third
`wash step was isocratic elution for 15 min using
`buffer A to remove the salt. The fourth step was
`equilibration of the column with 20 mM Tris–HCl,
`pH 7.8, containing 0.2% detergent (buffer C), for 12
`min. The same detergent was added to the eluent as
`was used for the extraction. The fifth step involved
`
`elution of the membrane proteins. This was per-
`formed by a 12-min linear sodium chloride gradient
`from buffer C to 0.5 M NaCl in the same buffer
`(buffer D). The detergents used in this study were
`C E (Kwant-Hoog Vacolie Recycling and Syn-
`10
`5
`thesis, Bedum, Netherlands), DDMAU and DDMAB
`(both of Calbiochem-Novabiochem, La Jolla, CA,
`USA). When the chromatography was performed
`with a Mono Q HR 5/5 column all five steps were
`performed at a flow-rate of 1 ml/min. When the
`Resource Q was used, two different protocols for
`chromatography were used. In the first protocol,
`steps 1, 2 and 3 were performed at a flow-rate of 5
`ml/min, in 3, 2 and 3 min, respectively. To allow
`comparison with HPIEC on the Mono Q, steps 4 and
`5 were performed at a flow-rate of 1 ml/min. In the
`second protocol all five steps were performed at a
`flow-rate of 5 ml /min, while steps 4 and 5 were
`reduced to 3 min each. The absorbance was moni-
`tored at 280 nm.
`Fractions of 5 ml were collected during steps 1 to
`4 and fractions of 2 ml were collected during
`gradient elution of step 5. Fractions of steps 1 to 4
`were dialyzed and lyophilized before analysis. Frac-
`tions of step 5 were analyzed directly. Fractions were
`analyzed by SDS–polyacrylamide gel electrophoresis
`(PAGE) and ELISA.
`
`2.2. SDS–PAGE
`
`Dialyzed and lyophilized samples (125 ml) of
`selected HPIEC fractions were analyzed by SDS–
`PAGE on 12.5% gels under reducing conditions [24].
`After electrophoresis, gels were fixed and silver
`stained as described [25].
`
`2.3. ELISA
`
`Microtiter plates were coated for 18 h at 48C with
`serial dilutions (in 50 mM NaHCO buffer, pH 9.6)
`3
`of samples of the collected fractions. After washing
`with phosphate-buffered saline (PBS), pH 7.4 con-
`taining 1 M NaCl and 0.3% Tween-20, plates were
`incubated with 1:6400 diluted monoclonal antibody
`(mAb) HD1 for 1 h. The mAb HD1 is directed
`against gD-1 and gD-2 and conformation-dependent
`[26]. After washing, plates were incubated for 1 h at
`378C with peroxidase-labeled sheep anti-mouse IgG
`
`Page 3
`
`

`

`32
`
`S.Welling-Wester et al. / J. Chromatogr. A 816(1998)29–37
`
`(Sanofi Diagnostics Pasteur, Marnes-la-Coquette,
`France). After color development with o-phenyl-
`enediamine dihydrochloride, the optical density was
`measured at 492 nm. Glycoprotein D-1 concentra-
`tions were calculated at OD 51.2 by using a gD-2
`492
`standard in combination with amino acid analysis.
`
`2.4. Extraction of recombinant gD-1 from Sf21
`cells using non-ionic detergent C E , DDMAU,
`10 5
`DDMAB, octylglucoside, Hecameg, dodecyl-b-D-
`maltoside
`
`Sf21 cells were grown in protein-free insect cell
`culture medium (Insect X-press, Bio-Whittaker,
`Walkersville, MD, USA) containing 10 mg /ml gen-
`8
`tamicin. Insect cells (2.5?10 ) were infected at a
`multiplicity of infection of 5 plaque-forming-units
`per cell by recombinant baculovirus containing the
`gD-1 gene (designated as gD-1-baculovirus). After
`four days of infection at 278C, cells were collected
`by centrifugation (100 g, 10 min, room temperature)
`and washed three times in ice-cold PBS. For ex-
`7
`traction of membrane proteins, the cell pellet (5?10
`cells per ml) was resuspended in ice-cold 20 mM
`Tris–HCl, pH 7.8, 2 mM phenylmethylsulfonyl
`fluoride (PMSF), 1 mM tosyllysine chloromethyl
`ketone (TLCK) and subsequently an equal volume of
`the same buffer was added, containing 2% (w/ v)
`C E , 2% DDMAU, 2% DDMAB, 2% octyl-gluco-
`10
`5
`side (Boehringer Mannheim, Almere, Netherlands),
`2% Hecameg (Vegatec, Villejuif, France) and 2%
`dodecyl-b-D-maltoside (Sigma, Brunschwig Chemie,
`Amsterdam, Netherlands),
`respectively. The cell
`suspension in the detergent solution (final detergent
`concentration 1%) was incubated on ice for 1 h. Cell
`debris was removed by low-speed centrifugation (10
`min, 2000 g). The supernatants (extracts) after ultra-
`centrifugation (70 000 g, 1 h, 48C) contain gD-1 and
`were stored in aliquots at 2808C. The amounts of
`gD-1 in the extracts were quantitated by ELISA.
`
`3. Results and discussion
`
`3.1. Extraction of recombinant gD-1 with different
`detergents
`
`Extraction of membrane proteins from infected
`
`cells or from virus particles is often the first step in
`the purification of membrane proteins. The suitability
`of different detergents to extract recombinant-gD-1
`from insect cells infected with gD-1-baculovirus was
`investigated. The amount of gD-1 extracted was
`determined by ELISA. The yields together with the
`characteristics of the detergents are given in Table 2.
`Extraction of infected cells with 1% C E , 1%
`10
`5
`dodecyl-b-D-maltoside,
`1% DDMAU and
`1%
`DDMAB, respectively, yielded approximately simi-
`lar amounts of gD-1. Low yields were obtained by
`extraction with detergents, having a high CMC. A
`subsequent, second extraction of the infected cells
`with a higher concentration (a final concentration of
`2% detergent) of these detergents with a relatively
`high CMC, e.g., Hecameg and octylglucoside, great-
`ly enhanced the yields of gD-1 (data not shown). In
`an earlier study [8], in which a number of polyoxy-
`ethylene alkylethers were compared, it was shown
`that the highest yields were obtained between HLB
`values of 11.5 to 12.5. The CMC of these detergents
`seemed to be of less importance, although in that
`particular study the yield of Sendai virus membrane
`proteins after extraction with octylglucoside (CMC
`7.1 mg/ml; HLB 12.6) was only 50% of
`that
`obtained with C E .
`10
`5
`
`3.2. HPIEC of detergent extracts containing the
`HSV membrane protein gD-1
`
`studies a multi-step purification
`In previous
`strategy was developed for the purification of inte-
`gral membrane proteins from different sources, Sen-
`dai virus [17], Plasmodium falciparum [16] and
`herpes simplex virus [15]. The basic principle was a
`sodium chloride gradient elution with eluents without
`detergent, followed by a second elution (a blank run)
`
`Table 2
`Yields of recombinant gD-1 after extraction of cells infected with
`gD-1-recombinant baculovirus with different detergents
`
`Detergent
`
`CMC (mM)
`
`Yield gD-1 (mg)
`
`C E
`10
`5
`Octylglucoside
`Dodecyl-b-D-maltoside
`Hecameg
`DDMAU
`DDMAB
`
`0.69
`25.0
`0.20
`19.5
`0.13
`4.3
`
`24.3
`1.1
`23.0
`2.6
`37.8
`37.3
`
`Page 4
`
`

`

`S.Welling-Wester et al. / J. Chromatogr. A 816(1998)29–37
`
`33
`
`with a sodium chloride gradient with buffers con-
`taining 0.1% detergent. During elution with buffers
`without detergent,
`the hydrophilic proteins were
`eluted and during the blank run with buffers con-
`taining detergent, the hydrophobic membrane pro-
`teins were eluted. In this way a selective elution of
`membrane proteins could be achieved.
`Due to several wash steps, the whole procedure is
`rather time consuming. Therefore we investigated in
`the present study whether we could apply the above-
`mentioned principle in a fast chromatographic pro-
`cedure. For this we used a Resource Q column,
`which allows a higher flow-rate than the Mono Q
`column. The detergents DDMAB and DDMAU,
`which showed promising results in the extraction of
`gD-1, were investigated as additives to the elution
`buffers. The non-ionic detergent C E which was
`10
`5
`used in an earlier study [15] was included for
`comparison. Samples of the detergent extracts, con-
`taining an equal amount of gD (approximately 1.2
`mg of gD-1), were subjected to the five consecutive
`HPIEC steps. Approximately 10 to 15% of the total
`amount of protein in the extract is gD-1. HPIEC was
`performed with 0.2% of the same detergent as used
`for the extraction, added to the elution buffer for step
`5. The CMC values of the three detergents studied,
`i.e., C E , DDMAU and DDMAB, are 0.026%,
`10
`5
`
`Table 3
`Chromatographic conditions and gD-1 recovery
`
`0.005% and 0.128%, respectively. Fractions were
`collected during the chromatographic steps and
`analyzed for the presence of gD-1. Table 3 summa-
`rizes the different procedures used with respect to
`detergents, columns, flow-rates and gD-1 recovery.
`Glycoprotein D-1 was mainly eluted during the
`sodium chloride gradient in the presence of 0.2%
`detergent in the eluent (step 5). No gD-1 was found
`in the flow-through fractions, and the fractions of the
`equilibration step (step 4) with buffer C. Glycopro-
`tein D-1 was only found in the fractions eluted with
`buffer B (step 2). When the C E and the DDMAU
`10
`5
`extracts were used as starting material
`for
`the
`chromatography approximately 15% of the gD-1
`applied to the column was eluted together with other
`proteins in step 2 (the sodium chloride wash step
`without detergent). When the DDMAB extract was
`applied to the columns either no gD-1, or only trace
`amounts could be detected among the other proteins
`that were eluted. An explanation for this could be
`that
`in the case of the C E and the DDMAU
`10
`5
`extract, detergent molecules are still attached to gD-1
`bound to the column after sample application and
`washing with buffer A, resulting in partial elution of
`gD-1. Due to the higher CMC of DDMAB,
`this
`detergent probably is more easily removed during
`and after sample application with the result that gD-1
`
`Detergent
`in eluent
`
`C E
`10
`
`5
`
`DDMAU
`
`DDMAB
`
`Column
`
`Mono Q
`Resource Q
`Resource Q
`
`Mono Q
`Resource Q
`Resource Q
`
`d
`
`f
`
`Mono Q
`Resource Q
`Resource
`
`e
`
`Flow-rates
`a
`steps 1,2,3 (ml/min)
`
`Flow-rates
`a
`steps 4,5 (ml/min)
`
`Recovery gD-1
`(%)
`
`b
`
`1
`5
`5
`
`1
`5
`5
`
`1
`5
`5
`
`1
`1
`5
`
`1
`1
`5
`
`1
`1
`5
`
`72
`34
`50
`
`15
`34
`72
`
`c
`
`g
`
`nd
`20
`nd
`
`a See Section 2.1.
`b Expressed as the percentage of the amount applied to the column.
`c nd5Not determined.
`d See Fig. 1a Fig. 1d.
`e See Fig. 1b Fig. 1e.
`f See Fig. 2.
`g Not purified.
`
`Page 5
`
`

`

`34
`
`S.Welling-Wester et al. / J. Chromatogr. A 816(1998)29–37
`
`is not eluted, or only trace amounts are eluted during
`the sodium chloride wash step (step 2).
`To compare the Mono Q and the Resource Q
`column with respect
`to the purification of gD-1,
`protocol 1 as described in Section 2.1 was used.
`Briefly, this implied that the wash steps were per-
`formed at a flow-rate of 5 ml/min for the Resource
`Q column and at 1 ml/min for the Mono Q. The
`flow-rate (1 ml/min) during the sodium chloride
`gradient in the presence of the detergent was identi-
`cal for the two columns. The elution patterns using
`the Resource Q column of the sodium chloride
`
`gradients in the presence of the detergents DDMAU
`and DDMAB (step 5, protocol 1) are shown in Fig.
`1a Fig. 1b, respectively, together with the concen-
`tration (mg/ml) of gD-1 in each fraction as de-
`termined by ELISA. The corresponding SDS gels of
`the fractions are shown in Fig. 1c and d.
`Fig. 1c shows that fractions 4 and 5 contain
`mainly three polypeptide bands with molecular mass
`(M ) of 54 000–52 000, 37 000 and 22 000, corre-
`r
`sponding to gD-1, and fragments thereof. This was
`confirmed by immunoblotting (data not shown) and
`ELISA analysis of the fractions with a gD-specific
`
`Fig. 1. HPIEC elution profile of a DDMAU extract (a) and a DDMAB extract (b) of insect cells infected with a recombinant
`gD-1-baculovirus. Chromatography was performed with a Resource Q column. After several steps (see Section 2.1, protocol 1) retained
`proteins were eluted with a linear 12-min gradient from 20 mM Tris–HCl (pH 7.8), containing either 0.2% DDMAU or 0.2% DDMAB, to
`0.5 M NaCl in the same buffer. The elution profile during the sodium chloride gradient is shown. The flow-rate was 5 ml/min for the wash
`steps and 1 ml/ min for the sodium chloride gradient elution. The absorbance was monitored at 280 nm. Fractions of 2 ml were collected as
`indicated and analyzed by SDS–PAGE on 12.5% gels and by ELISA. In (c) the analysis of the fractions collected during chromatography (a)
`is shown. The SDS gel shown in (d) corresponds to the chromatogram of (b). The polypeptides were visualized by silver-staining. The
`arrows indicate the migration position of gD-1 and fragments thereof. E is the extract of the cells infected with the recombinant
`3
`gD-1-baculovirus. The molecular masses (?10 ) of the reference proteins (R) are indicated. The concentration of gD-1 (mg/ml) in the
`fractions was determined by ELISA with the gD-specific mAb HD1 and they are indicated by black columns in the elution profile. (e)
`Immunoblot of fraction 5 of (c) obtained by using mAb A16 [21], peroxidase-conjugated rabbit antimouse IgG followed by diaminoben-
`3
`zidine staining. The molecular masses (?10 ) of reference proteins are indicated.
`
`Page 6
`
`

`

`S.Welling-Wester et al. / J. Chromatogr. A 816(1998)29–37
`
`35
`
`Fig. 1. (continued)
`
`mAb (see Fig. 1a, black columns). The proteolytic
`degradation of gD-1 in extracts of insect cells is not
`unusual and has been described [22,27]. Chromatog-
`raphy in the presence of 0.2% DDMAB did not
`result in purification of gD-1 (Fig. 1d). Although
`some gD-1 was eluted (see Fig. 1b, black columns),
`it was eluted together with numerous other poly-
`peptides. Chromatography in the presence of C E
`10
`5
`(data not shown) showed results similar to those
`obtained in the presence of DDMAU. Chromatog-
`raphy of a C E extract using a Mono Q column has
`10
`5
`been described earlier [15] and resulted in relatively
`pure conformationally intact gD-1. When the chro-
`
`matography of a DDMAU and DDMAB detergent
`extract was performed using a Mono Q column with
`the same detergents as additives to the buffers,
`results were similar to using the Resource Q column,
`i.e., with DDMAU, virtually pure gD-1 was ob-
`tained, and with DDMAB no purification of gD-1
`could be achieved.
`The two detergents DDMAU and DDMAB have
`been used for
`the selective extraction [10] and
`purification [23] of membrane proteins of Myco-
`plasma gallisepticum. In these studies, the results
`show that extraction with DDMAU was relatively
`selective and that DDMAB had a higher efficiency in
`
`Page 7
`
`

`

`36
`
`S.Welling-Wester et al. / J. Chromatogr. A 816(1998)29–37
`
`membrane protein extraction. The addition of the
`detergent DDMAU (in a concentration of 2 mM, this
`is 16-times the CMC value) to the elution buffers for
`the HPIEC in that particular study resulted in
`purification of proteins p67, p52 and p77.
`the
`The
`above-mentioned results
`show that
`strategy previously used for
`the purification of
`membrane proteins, using the detergent C E [15–
`10
`5
`17], is also applicable when the detergent DDMAU
`was added to the elution buffers, but not when
`DDMAB was used as additive. It is easy to speculate
`that the difference in CMC of the detergents may
`account for these results. The elution buffers con-
`tained 0.2% detergent, which is 40-times the CMC of
`DDMAU, seven-times the CMC of C E , and 1.6-
`10
`5
`times the CMC of DDMAB. This of course does not
`exclude other possible factors like composition of the
`
`extract, and specific properties of the protein to be
`purified.
`Since chromatographic results by following proto-
`col 1, using either a Resource Q or Mono Q column
`were similar, a next set of experiments was per-
`formed, in which the gradient elution was performed
`at a flow-rate of 5 ml/min (protocol 2, HPIEC in
`Section 2.1). In Fig. 2, the results are shown of the
`chromatography of a DDMAU extract separated on a
`Resource Q column using a flow-rate of 5 ml/ min.
`Glycoprotein D-1 is eluted relatively fast (see ELISA
`results, indicated as black columns in Fig. 2b) and
`mainly present in two fractions, fractions 2 and 3.
`The corresponding gel (Fig. 2a) shows that fractions
`2 and 3 consist mainly of gD-1 and fragments
`thereof. Results obtained with the detergent C E10
`5
`using the same protocol, are similar, gD-1, in almost
`
`Fig. 2. HPIEC of a DDMAU extract of insect cells infected with a recombinant gD-1-baculovirus. Chromatography was performed with a
`Resource Q column. After several wash steps (see Section 2.1, protocol 2) retained proteins were eluted with a linear 12-min gradient from
`20 mM Tris–HCl (pH 7.8), containing 0.2% DDMAU, to 0.5 M NaCl in the same buffer. The flow-rate was 5 ml/min for the wash steps
`and also for the sodium chloride gradient elution. The absorbance was monitored at 280 nm. Fractions of 2 ml were collected as indicated
`and analyzed by SDS–PAGE on 12.5% gels and by ELISA. In (a) the analysis of the fractions by SDS–PAGE is shown. The polypeptides
`were visualized by silver-staining. The arrows indicate the migration position of gD-1 and fragments thereof. E is the extract of the cells
`infected with the recombinant gD-1-baculovirus. G1 is the analysis of consecutive samples collected during the wash step with sodium
`chloride without detergent (step 2, protocol 2). Lanes 1–6 correspond to the fractions collected during the sodium chloride gradient with
`3
`0.2% DDMAU in the elution buffers. The molecular masses (?10 ) of the reference proteins (R) are indicated. The concentration of gD-1
`(mg/ml) in the fractions was determined by ELISA with the gD-specific mAb HD1 and they are indicated by black columns in (b).
`
`Page 8
`
`

`

`S.Welling-Wester et al. / J. Chromatogr. A 816(1998)29–37
`
`37
`
`is eluted in two fractions (data not
`pure form,
`shown). Again, no purification at all was obtained
`using the detergent DDMAB as additive to the
`buffer.
`Recoveries of gD-1 (Table 3) were determined by
`ELISA. As standard a serial dilution of a known
`concentration was included. The gD-1 concentration
`of
`the standard was determined by amino acid
`analysis [15]. The highest recovery of gD-1 was
`obtained either by chromatography of a C E
`10
`5
`extract using a Mono Q column at a flow-rate of 1
`ml/min or by chromatography of a DDMAU extract
`using a Resource Q column at a flow-rate of 5
`ml/min.
`
`4. Conclusions
`
`detergents DDMAU and
`zwitterionic
`The
`DDMAB are equally effective in extracting the
`integral membrane protein gD-1 of herpes simplex
`virus from infected insect cells. In the purification of
`gD-1 from the detergent
`extracts by HPIEC,
`DDMAU (CMC 0.13 mM) was superior to DDMAB
`(CMC 4.3 mM) and similar to C E (CMC 0.69
`10
`5
`mM). This may suggest
`that detergents with a
`relatively low CMC are more useful as additive to
`elution buffers for HPIEC. The application of a
`Resource column, which allows a higher flow-rate
`than a Mono Q column reduced the chromatographic
`procedure from 64 min to 13 min.
`
`Acknowledgements
`
`We thank Mr. B. Kwant (Bedum, Netherlands) for
`the gift of
`the non-ionic detergent C E . The
`10
`5
`truncated gD-2, which was used for quantitation and
`amino acid analysis, was a generous gift of Dr. M.
`Slaoui, Smith-Kline, Belgium.
`
`References
`
`[1] J.M. Neugebauer, Methods Enzymol. 182 (1990) 239.
`
`[2] A. Helenius, K. Simons, Biochim. Biophys. Acta 415 (1975)
`29.
`[3] C. Tanford, J.A. Reynolds, Biochim. Biophys. Acta 457
`(1976) 133.
`[4] A. Helenius, D.R. McCaslin, E. Fries, C. Tanford, Methods
`Enzymol. 56 (1979) 734.
`[5] L.M. Hjelmeland, A. Crambach, Methods Enzymol. 104
`(1984) 305.
`[6] A. Gonenne, R. Ernst, Anal. Biochem. 87 (1978) 28.
`´
`[7] D. Plusquellec, G. Chevalier, R. Talibart, H. Wroblewski,
`Anal. Biochem. 179 (1989) 145.
`¨
`[8] J. van Ede, J.R.J. Nijmeijer, S. Welling-Wester, C. Orvell,
`G.W. Welling, J. Chromatogr. 476 (1989) 319.
`[9] J.E.K. Hildreth, Biochem. J. 207 (1982) 363.
`´
`[10] C. Brenner, G. Jan, Y. Chevalier, H. Wroblewski, Anal.
`Biochem. 224 (1995) 515.
`[11] A.J. Furth, Anal. Chem. 109 (1980) 207.
`[12] R. Van der Zee, S. Welling-Wester, G.W. Welling, J. Chroma-
`togr. 266 (1983) 577.
`[13] G.W. Welling, J.R.J. Nijmeijer, R. Van der Zee, G. Groen,
`J.B. Wilterdink, S. Welling-Wester, J. Chromatogr. 297
`(1984) 101.
`[14] G.W. Welling, R. van der Zee, S. Welling-Wester, Trends
`Anal. Chem. 5 (1986) 225.
`[15] R.A. Damhof, M. Feijlbrief, S. Welling-Wester, G.W. Wel-
`ling, J. Chromatogr. A 676 (1994) 43.
`[16] D.L. Narum, G.W. Welling, A.W. Thomas, J. Chromatogr. A
`657 (1993) 357.
`[17] S. Welling-Wester, M. Feijlbrief, D.G.A.M. Koedijk, M.A.
`Braaksma, B.R.K. Douma, G.W. Welling, J. Chromatogr. 646
`(1993) 37.
`[18] N. Bourne, L.R. Stanberry, D.I. Bernstein, J. Infect. Dis. 173
`(1996) 800.
`[19] W.P. Sisk, J.D. Bradley, R.J. Leipold, A.M. Stolzfus, M.
`Ponce de Leon, M. Hilf, C. Peng, G.H. Cohen, R.J. Eisen-
`berg, J. Virol. 68 (1994) 766.
`[20] H.-Y. Chiang, G.H. Cohen, R.J. Eisenberg, J. Virol. 68
`(1994) 2529.
`[21] G.A. Schellekens, E. Lasonder, M. Feijlbrief, D.G.A.M.
`Koedijk, J.W. Drijfhout, A.J. Scheffer, S. Welling-Wester,
`G.W. Welling, Eur. J. Immunol. 24 (1994) 3188.
`[22] H. Ghiasi, R. Kaiwar, A.B. Nesburn, S.L. Wechler, Arch.
`Virol. 121 (1991) 163.
`´
`[23] G. Jan, C. Brenner, H. Wroblewski, Prot. Express. Purif. 7
`(1996) 160.
`[24] U.K. Laemmli, Nature (London) 227 (1970) 680.
`[25] W. Wray, T. Boulikas, V.P. Wray, R. Hancock, Anal. Bio-
`chem. 118 (1981) 197.
`[26] L. Pereira, D.V. Dondero, D. Gallo, V. Devlin, J.D. Woodie,
`Infect. Immun. 35 (1982) 363.
`[27] K.M. Zezulak, P.G. Spear, J. Virol. 50 (1984) 258.
`
`Page 9
`
`