Molecular and Biochemical Parasitology, 17 (1985) 343-358 343

Elsevier

MBP 00608

PHASE SEPARATION IN TRITON X-114 OF ANTIGENS OF TRANSMISSION BLOCKING IMMUNITY IN PLASMODIUM GALLINACEUM

NIRBHAY KUMAR

Malaria Section, Laboratory of Parasitic Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bldg. 5, Room 112, Bethesda, MD 20205, U.S.A.

(Received 5 February 1985; accepted 19 August 1985)

The distribution of proteins of mosquito midgut forms of Plasmodium gallinaceum in the detergent-free (aqueous) and detergent-enriched phases was studied using a phase separation technique in Triton X-114. Of the three surface proteins on gametes and newly fertilized zygotes (240, 56, and 54 kDa) immunoprecipi- tated by transmission blocking monoclonal antibodies, 240 kDa protein was recovered in the aqueous phase, whereas 56 and 54 kDa proteins were found preferentially in the detergent phase. The hydrophobic properties of the 56 and 54 kDa proteins were also shown by their strong tendency to interact with the lipid bilayers and a hydrophobic matrix phenyl-Sepharose. Monoclonal antibody IID3B 3 immunoprecipitated all the three proteins from the whole Triton extract but in the phase-separated extracts reacted only with the 240 kDa protein in the aqueous phase and not with the 56 and 54 kDa doublet in the detergent phase. In Western blot analysis also monoclonal antibody IID3B3 reacted only with the 240 kDa protein. The 240 kDa protein in the aqueous phase was retained by monoclonal antibody IID3B 3 linked to Sepbarose 4B beads and could be eluted either with 0.1 M acetic acid or 50 mM diethylamine. The 56 and 54 kDa doublet in the detergent phase could be bound to and eluted from Sepharose 4B beads-linked monoclonal antibody IID4 or rabbit anti-male P. gallinaceum gamete serum. Two stage-specific glycoproteins of 26 and 28 kDa on the surface of ookinetes of P. gallinaceum were also separated in the detergent phase following Triton X-114 extraction. Phase separation in Triton X-114 offers a simple approach to the separation of a select group of proteins from the bulk of the cellular proteins.

Key words: Plasmodium gallinaceum; Zygotes; Ookinetes; Triton X-114; Phase separation; Transmission- blocking immunity; Hydrophobic chromatography; Liposomes; Western blotting

INTRODUCTION

Antibodies directed against antigens on the surface of extracellular gametes (mos- quito midgut stages) of malaria parasites suppress infectivity of parasites to mosqui-

Abbreviations: MAb, monoclonal antibody; NET, NaCI, Tris, EDTA; NETT, NET with Triton X-100; NETTS, NET with increased NaCI, Triton X- 100; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel

electrophoresis; TS buffer, Tris-saline buffer.

344

toes, and thus interrupt transmission. In Plasmodium falciparum as well as in Plasmodium gallinaceum monoclonal antibodies (MAbs) which blocked infectivity of parasites to mosquitoes by preventing fertilization of male and female gametes in the mosquito midgut immunoprecipitated a set of three proteins from the surface of

radioiodinated gametes or from biosynthetically labelled gametocytes. Proteins im-

munoprecipitated by the antibodies are of 260, 59 and 53 kDa in P. falciparum and of 240, 56 and 54 kDa in P. gallinaceum [1,2]. Polyacrylamide gel electrophoresis (PAGE) under reducing and non-reducing conditions showed that these three proteins are not linked to each other by disulfide bonds [1, 2]. Pulse-chase analysis during biosynthetic

labelling of gametocytes of P. falciparum did not show any precursor-product rela- tionship between the high molecular weight and the two lower molecular weight proteins [3]. Moreover, in P. falciparum, only the 59 and 53 kDa polypeptides were found to be glycosylated [3].

In addition to fertilization blocking antibodies, a MAb which blocks the post-fertil-

ization development of zygotes ofP. gallinaceum has also been described [4]. Of the two surface glycoproteins of 26 and 28 kDa which appear de novo on the surface of the transforming zygotes ofP. gallinaceum, the 26 kDa protein is the target antigen [4-6].

In this paper the properties of membrane proteins, especially those of previously

identified target antigens of transmission blocking antibodies, in P. gallinaceum have been studied by phase separation technique in the detergent Triton X- 114. Triton X- 114 is similar to Triton X-100 except that Triton X-114 has two fewer ethoxy groups and hence is more hydrophobic. This chemical difference is sufficient to lower the cloud point, the temperature at which submicellar aggregates form from 64°C for Triton X-100 to 20°C or above Triton X-114. Thus at temperatures above 20°C Triton X-114

forms two phases; an upper aqueous phase which contains less than 5% of total detergent and a lower detergent phase which contains most of it. These properties of

Triton X- 114 were first used by Bordier [7] to characterize and separate membrane proteins. In these studies hydrophilic proteins were recovered in the aqueous phase whereas amphiphilic integral membrane proteins were found in the detergent phase

after the separation. The hydrophilic and hydrophobic nature of the target antigens was also investigated

by hydrophobic chromatography and reconstitution of proteins into the lipid bilayers.

MATERIALS AND METHODS

Parasites. Male gametes, newly fertilized zygotes and ookinetes of P. gallinaceum were prepared as described previously [8].

Antibodies. Monoclonal antibodies IID3B3 and IID4 were derived by Kaushal et al. [2]. MAb IID3B3 immunoprecipitated a set of three proteins of 240, 56 and 54 kDa from the detergent solubilized extracts. MAb IID4, on the other hand, immunopreci- pitated only the 56 and 54 kDa proteins. Rabbit antisera to male gametes, zygotes and

345

ookinetes of P. gallinaceum were obtained as described earlier [8]. Rabbit anti-mouse immunoglobulins were a kind gift from Dr. H. Metzger (NIADDK, NIH).

Chemicals. Lactoperoxidase, diethylamine, KSCN, Triton X- 114, Tween-20 and egg lecithin (L-u-phosphatidyl choline, type IX-E) were purchased from Sigma Chemical Company (St. Louis, MO). Triton X-100 (scintillation grade) was purchased from Research Products International Corporation (Elk Grove Village, IL). Sodium cho- late and 3-((3-cholamidopropyl) dimethyl ammonio)-/-propanesulfonate (CHAPS) were from Calbiochem-Behring (San Diego, CA). Dimethyl pimelimidate dihydro- chloride, ' Iodogen' (1,3,4,6-tetrachloro-3ct,6et-diphenyl glycoluril) were from Pierce Chemical Company (Rockford, IL). Protein A-Sepharose, PD-10 (Sephadex G-25) columns and phenyl-Sepahrose-CL-4B were purchased from Pharmacia Chemicals (Piscataway, N J). SM-2 beads and sodium dodecyl sulfate-polyacrylamide gel electro- phoresis (SDS-PAGE) reagents were from Bio-Rad (Richmond, CA). Nitrocellulose paper was obtained from Schleicher and Schuell, Inc., Keene, NH. Carrier-free t25I-Na and [35S]methionine (-- 1200 Ci mmol -L) were obtained from Amersham (Arlington Heights, IL). 125I-Protein A (2-10 taCi ~tg -t) was purchased from New England Nu- clear, Boston, MA. Triton X-114 was purified by pre-condensation twice as described by Bordier [7]. The concentration of Triton in the purified preparations was measured at 275 nm (E = 1.46 × 103 M -~ cm-t).

lodination of parasites and solubilization in Triton X-114. Male gametes, zygotes and ookinetes of P. gallinaceum were purified and iodinated by lactoperoxidase catalyzed radioiodination of surface components [8]. Iodinated cells were solubilized in 0.5 or 1% Triton X-114 in Tris-saline (TS) buffer (10 mM Tris, 150 mM NaCI, pH 7.5) at 0-4°C for 30-60 rain, and centrifuged at 40 000 × g for 30 min at 0-4°C. Supernatants were collected in the tubes on ice using prechilled Pasteur pipettes and stored at -70°C.

Phase separation. Supernatants of Triton X-114 solubilized cells were warmed for 5 min at 37°C. Bromophenol blue (0.001%) was added to the supernatants before warming to follow the phase separation. Bromophenol blue interacts with the deter- gent Triton X-114 and therefore aids visually in the assessment of phase separation (N. Kumar, unpublished observations and ref. 9). After incubation at 37°C, samples were centrifuged for 1 min in a microfuge at room temperature. The supernatant (aqueous upper phase) was separated from the blue pellet (detergent lower phase). The detergent phase was redissolved in TS to the original volume by incubating for 5-10 min at 0-4°C (LP~). To the aqueous phase were added more Triton X-114 (5% stock) and bromo- phenol blue (0.1% stock) to yield final concentrations of 0.5% and 0.001%, respective- ly (UPI). To repeat phase separation on LPI, and UP1, samples were rewarmed, and centrifuged as above. This time the supernatant of the sample LPI was discarded and the pellet was redissolved in TS (LP~.). On the other hand, the pellet of sample UP1 was

346

discarded and the supernatant (UP2) was replenished with more Triton X-114 and

bromophenol blue as above. Phase separation steps were thus repeated the desired

number of times.

Biosynthetic labelling of zygotes with [35 S]methionine and phase separation. Zygotes of

P. gallinaceum were washed with methionine-free medium 199 and labelled with

[35S]methionine as described [5]. 5 X 107 zygotes were incubated in 1 ml methionine-

free medium 199 for 4 h at room temperature in the presence of 0.5 mCi [35S]methio-

nine. They were then distributed in four tubes (12 X 75 mm) containing 4 ml medium

199 (supplemented with 1 mM L-glutamine, 125 lag ml -~ gentamycin, 100 U ml -~

penicillin and 100 tag ml -~ streptomycin) and incubated further for 20 h at 26°C. Cells

were then washed with medium 199, extracted in 1% Triton X-114 at 0-4°C for 30 min.

Lysed cells were centrifuged at 40 000 X g for 60 min at 0-4°C and supernatant used for

phase separation.

Hydrophobic chromatography. Surface radioiodinated zygotes (1 X 108) were extrac-

ted in 1.0 ml of 0.5% Triton X- 114. For hydrophobic chromatography, total extract or

phase separated aqueous and detergent phases were diluted to 10 ml in TS buffer

containing 0.8 M ammonium sulfate and applied to a column (0.7 X 16 cm) of

phenyl-Sepharose CL-4B. The column was washed with 40 ml TS buffer containing 0.8

M ammonium sulfate and finally with 40 ml TS buffer before elution with a linear

gradient of 0.1 to 2% Triton X- 100 (70 ml each) in TS buffer as previously described

[10]. Radioactivity in each fraction was measured using a Beckman y-counter. Triton

X- 100 concentration in each fraction was assessed by measuring absorbance at 275 nm.

Interaction between proteins and lipid bilayers. Egg lecithin was dissolved in chloro-

form (25 mg ml -~) in a glass vial and a thin film of the lipid was obtained by flushing with N2 gas, followed by drying under vacuum overnight. Liposomes were then formed

by swelling the lipids in TS buffer (25 mg ml -~) for 15 min with intermittent vortexing at

room temperature, followed by sonication for 15-30 min using a microtip (Model

W-375, Heat Sytstems Ultrasonics, Inc., Plainview, NY). In some experiments [ IJ4C] -

dipalmitoyl phophatidylcholine was used as a tracer for lipids and added to the bulk

of lipids before drying.

Triton X-100-solubilized extracts or Triton X-114 phase-separated fractions of

surface radioiodinated zygotes or male gametes were mixed with the sonicated lipo- somes. An additional amount of CHAPS was added to yield a molar ratio of detergent

to lipid of 10-15:1. After 2-3 h on ice, the mixture was dialyzed using 'Spectra-PoP

membranes (2000 molecular weight cut off; Spectrum Medical Industry Inc., Los Angeles, CA) against 1 000 volumes of TS buffer (two changes every day) over 8-10 days. A dialysis bag containing SM-2 beads was added to the beaker during dialysis.

The extent of reconstitution of proteins into lipid bilayers was assessed by sucrose density gradient centrifugation [ 11]. Dialyzed samples made 35% in sucrose (40 lal)

347

were underlaid in a step gradient of sucrose 5% (25 laD, 10% (20 lal), 25% (30 lal), 30% (35 lad prepared in Beckman airfuge test tubes (Ultraclear Cat. No. 344718, 5 )< 20 mm, Beckman, Paio Alto, CA). Tubes were centrifuged at 100 000 )< g for 30 min at room temperature. Fractions (20 lad were collected from the top for radioactivity counting and SDS-PAGE.

Binding studies with Sepharose beads. Monoclonal antibodies were bound to protein A-Sepharose CL-4B beads and then the complex was cross-linked with dimethyl pimelimidate [ 12]. Extracts of surface-radioiodinated cells or Triton X- 114 phase-sep- arated fractions were incubated with antibody-protein A-Sepharose beads at room

temperature for 1-2 h with gentle rocking. The beads were then washed once with 10 mM Tris (pH 7.6), 5 mM EDTA, 500 mM NaC1, 0.5% Triton X-100 (NETTS) and twice with 10 mM Tris (pH 7.6), 5 mM EDTA, 150 mM NaC1, and 0.5% Triton X-100 (NETT). Bound anLigens were eluted by incubating the washed beads in various elution buffers (mentioned in the Results) at room temperature for 30 min. Eluted samples were neutralized for further analysis by SDS-PAGE.

Immunoprecipitation, SDS-PA GE, autoradiography and fluorography. Detergent sol- ubilized or various other fractionated samples were immunoprecipitated by incubat- ing with antibodies at 4°C for 3-4 h. Immune complexes were absorbed to protein A-Sepharose beads by rocking at room temperature for 1 h. The beads were washed twice with NETTS, twice with NETT and twice with NET (10 mM Tris, 5 mM EDTA, 150 mM NaCI, pH 7.6) and eluted in SDS-sample buffer (62.5 mM Tris, 5% SDS, 10% glycerol, 0.005% bromophenol blue, pH 6.8) by boiling for 2 min. Samples were analyzed by SDS-PAGE [13]. After staining/destaining, slab gels were dried and exposed with Kodak X-omat AR films at -80°C. Gels containing [35S]methionine-la- belled samples were first soaked in 'Autofluor ' (National Diagnostics, Somerville, N J) for 1-2 h at room temperature before drying.

Western blotting. 108 zygotes of P. gallinaceum were extracted in SDS-sample buffer in the presence or absence of 2% 2-mercaptoethanol and separated by 5-15% SDS- PAGE. After electrophoresis, the gels were equilibrated by soaking in the transfer buffer (25 mM Tris, 192 mM glycine and 20% methanol) for at least 1 h, with one change of the buffer. Proteins were then transferred onto nitrocellulose paper by electroblotting in the transfer buffer (16 h, 30 V, 4°C) and detected according to Towbin et al. [ 14] with slight modifications. The nitrocellulose sheets were immersed in TS buffer containing 0.3% Tween-20 for 30-60 min. Strips (7-10 mm in width) were then incubated with continuous shaking with 1:100 dilution of rabbit antiserum or 1:50 dilution ofascites containing MAbs at room temperature for 2-3 h. Strips were washed with TS buffer containing 0.05% Tween-20 (4-5 changes, 1 h). Strips treated with MAbs were incubated with 1:1 000 dilution of rabbit anti-mouse immunoglobulins serum before incubation with 125I-labelled Protein A (50000 cpm ml -t, 1 h, room

348

t empera ture) . Washed str ips were exposed for a u t o r a d i o g r a p h y . Pres ta ined s t andards

(Bethesda Research Labs. , Ga i the r sbu rg , MD) were used dur ing e lect rophores is .

RESULTS

Solubilization and phase separation of zygote and ookinete proteins. P. gallinaceum zygotes were surface r a d i o i o d i n a t e d and ex t rac ted in Tr i ton X- 114 and var ious deter-

gents. So lubi l iza t ion of surface label led pro te ins by the detergents was comparab le ;

75% (0.5% Tr i ton X-114), 80% (0 .5-1% Tr i ton X-100), 80% (1% C H A P S ) and

50-60% (1% sod ium chola te or Tween-20). There were no appa ren t qual i ta t ive or

quant i t a t ive differences in the pa t te rn of pro te ins (as ana lyzed by S D S - P A G E ) solubi-

lized by var ious detergents (da ta not shown). A p p r o x i m a t e l y 15-16 pro te ins have been

ident if ied on the surface of zygotes of P. gallinaceum by r a d io iod ina t i on [8].

The two phases (aqueous and detergent) ob ta ined by phase sepa ra t ion o f a 0.5%

Tr i ton X-114 extract of r a d i o i o d i n a t e d zygotes were ana lyzed by S D S - P A G E . Fig. 1

shows the d i s t r ibu t ion of surface- label led zygote pro te ins dur ing four cycles of phase

separa t ion . A 240 k D a prote in as well as many o ther p ro te ins above 60 k D a remained

Mrx 10 -3

T o p - -

2 0 0 _ w ~

A

116-- B 8 a 6 "~ 97 .4 - -

66.2- -

45.0 - - mid

3 1 . o -

2 1 . 5 -

14 .4 - -

B(~B--

TOP--

97.4 - - 66.2 - -

45.0 - - W t

31.0

2 1 . 5 - -

14.4 n ~ 41~ did

<3

1 2 3 4 1 2 3 4 B d ~ B - - I I I - - J

TO UP LP TO UP1 LP1 UP2 LP2

Fig. 1. Distribution of t2Sl-labelled surface proteins of P. gallinaceum zygotes (A) and ookinetes (B) in aqueous (UP) and detergent (LP) phases in Triton X-114 after four cycles of phase separation. Samples of total extract (To) or at various cycles of phase separation (1-4) of UP or LP were analyzed using 5-15% acrylamide gradient SDS-PAGE. Samples of ookinetes were analyzed by 12.5% acrylamide SDS-PAGE. Bio-Rad high and low molecular weight markers were used to calibrate the gels. Arrows indicate the position of 240 and 56 and 54 kDa antigens. Heavy arrow in (B) indicates the position of two ookinete stage-specific proteins.

349

exclusively in the aqueous phase (Fig. 1A, UP lanes 1-4). Two proteins of 56 and 54 kDa and others below 50 kDa always partitioned in the detergent phase (Fig. IA, LP lanes 1-4). 240, 56 and 54 kDa proteins are also present on the surface of male gametes of P. gallinaceum [8] and their distribution in the aqueous and the detergent phase of Triton X-114 was the same as that with zygotes (data not shown).

During transformation into ookinetes, zygotes synthesize and express two new proteins of 26 and 28 kDa on their surface [5,6]. When surface radioiodinated ookinetes of P. gallinaceum were phase separated, these ookinete stage-specific pro- teins and others were found in the detergent phase. None were present in any signifi- cant amount in the aqueous phase after two cycles of phase separation (Fig. IB).

Alternatively proteins were also labelled biosynthetically during transformation of zygotes into ookinetes, extracted in 1% Triton X- 114 and phase separated. Only a few [35S]methionine-labelled proteins were found in the detergent phase (Fig. 2, lane 3) while most remained in the aqueous phase (Fig. 2, lane 2). Labelled proteins in the aqueous and detergent phases were immunoprecipitated with rabbit anti-zygote or

Mr x 10 - 3

T O P - - W

2 0 0 - - ~,~

1 1 6 - -

9 7 . 4 - -

6 6 . 2 - -

4 5 . 0 - -

3 1 . 0 - -

2 1 . 5 - -

1 4 . 4 - - 4 i ~ ......

a . . ~ -

BOB

1 2 3 4 5 6 7

Fig. 2. Phase separation of [35S]methionine-labelled proteins. P. gallinaceum zygotes were labelled with

[~SS]methionine during their t ransformation into ookinetes. Labelled cells were extracted in 1% Triton

X- 114 and subjected to four repeated cycles of phase separation and analyzed by 5-20% SDS-PAGE. Lane

1, total extract, lane 2, aqueous phase and lane 3, detergent phase. Phase-separated aqueous (lanes 4, 5) and

detergent phases (lanes 6, 7) were also immunoprecipitated and analyzed on the same gel. Immunoprecipita-

tions were done with rabbit anti-zygote serum (lanes 4, 6) and rabbit anti-ookinete serum (lanes 5, 7).

350

rabbi t an t i -ook ine te serum. Prote ins seen as a heavy band between 21 ~nd 31 k D a

marke r s i m m u n o p r e c i p i t a t e d only with the an t i -ook ine te serum (lane 7) and not with

the an t i -zygote serum (lane 6) were a lmos t ent i rely in the detergent phase (lane 3).

Prote ins in this heavy band ac tua l ly separa te into two pro te in bands of appa ren t 28

and 26 k D a when ana lyzed on a 12.5% S D S - P A G E (not shown) which represent the

two new pro te ins synthes ized by the zygotes and expressed on the surface o f o o k i n e t e s

dur ing t r a n s f o r m a t i o n [5].

In ano the r series of exper iments Tr i ton X- 114 extracts o f un labe l led male gametes or

zygotes of P. gallinaceum were phase separa ted and the pro te ins in the two phases were

then iod ina ted . In both cell types 56 and 54 k D a pro te ins were label led in the detergent

phase and the 240 k D a pro te in with many o ther cel lular p ro te ins were in the aqueous

phase (da ta not shown).

Results in Figs. 1 and 2 show that only cer ta in pro te ins in teract with Tr i ton X-114,

while most o thers r emain in the aqueous (relat ively detergent- f ree) phase. Phase

sepa ra t ion me thod , therefore , offers a s imple a p p r o a c h to the enr ichment for a select

g roup o f pro te ins which interact with the detergent Tr i ton X-114.

56 and 54 kDa proteins are hydrophobic proteins. Whole cell extracts p repa red in

Tr i ton X-114 or the phase - sepa ra t ed aqueous or de tergent f ract ions o f t25I-surface-la-

bel led zygotes of P. gallinaceum were ana lyzed by h y d r o p h o b i c c h r o m a t o g r a p h y on

p h e n y i - S e p h a r o s e CL-4B. Samples made 0.8 M in a m m o n i u m sulfate were p u m p e d

th rough the column. The co lumn was subsequent ly washed and e lu ted with a gradient

of Tr i ton X-100.

When the to ta l cell extract was passed th rough the co lumn, 30-35% of the t r ichlo-

roacet ic acid prec ip i tab le counts did not b ind and were recovered in the u n b o u n d

f rac t ion and washings. Bound mate r ia l was recovered in three peaks elut ing at differ-

ent Tr i ton X-100 concent ra t ions . A minor peak eluted at 0.1% Tr i ton X-100 and

con ta ined 4% rad ioac t iv i ty and two o ther peaks elut ing at 0.45% and 0.85% Tr i ton

X-100 concen t ra t ions con ta ined 20 and 44% of the counts , respect ively (Fig. 3a).

Aqueous and detergent phases f rom Tr i ton X-114 solubi l ized extracts of tzSI-labelled

zygotes o f P. gallinaceum were each ana lyzed separa te ly on the p h e n y l - S e p h a r o s e

CL-4B column. A p p r o x i m a t e l y 60% of the counts f rom the aqueous phase were

recovered in the u n b o u n d f rac t ion and washings. Rad ioac t iv i ty b o u n d to the co lumn

i, Fig. 3. Hydrophobic chromatography. Whole cell extract (a), aqueous phase (b) and the detergent phase (c) of 125I-surface labelled zygotes were applied to a phenyl-Sepharose column. Unbound sample (flo~ through, FT) was collected. The column was washed with 40 ml TS + 0.8 M ammonium sulfate (arrow I), followed by 40 ml TS alone (arrow 2). Bound proteins were eluted (arrow 3) with a 0. I to 2e~ linear gradient of Triton X-L00 (70 ml each). Fractions were counted (o o) and Triton concentration was assessed by measuring absorbance at 275 nm ( 3 - - c ) . Major peak in (c) was analyzed by 5-15% SDS-PAGE (inset in c).

‘*5L

CP

M/m

l F

RA

CT

ION

-

‘*51

KP

M/m

l F

RA

CT

ION

-

TR

ITO

N

x-10

0 (%

)o

cw

0

b N

0

TR

ITO

N

x -

100

(%I

0

1*51

-CP

M/m

l F

RA

CT

ION

-

I I

b P

0

TR

ITO

N

x -1

00

(%)

0

352

was recovered in a single peak eluting at 0.45% of Triton X-100 (Fig. 3b). In contrast, all the radioactive counts in the detergent phase were bound to the column and subsequently eluted in one major and two minor peaks. The major peak eluting at

0.8% Triton X-100 contained 92% of radioactivity (Fig. 3c). The peak eluting at 0.8% Triton X-100 was analyzed by SDS-PAGE and contained the 56 and 54 kDa proteins as well as a 50 kDa and two minor protein bands under 20 kDa (Fig. 3c, inset). The presence of high salt in the sample promotes the effectiveness of hydrophobic interac-

tions between proteins and the hydrophobic matrix. Therefore, proteins eluting at different Triton X- 100 concentrations must vary in their net hydrophobicity. The data suggest that the 56 and 54 kDa proteins are hydrophobic in nature as both in the total

extract and in the detergent phase these proteins are retained by the hydrophobic matrix and subsequently eluted by 0.8-0.9% Triton X-100.

The hydrophobic characteristics of 56 and 54 kDa proteins were further shown in the experiments where these proteins were reconstituted into lipid bilayers. Proteins in the extracts of zygotes were reconstituted into egg lecithin bilayers by detergent dialysis and analyzed by flotation on a sucrose step gradient. Fig. 4A shows a typical density gradient profile of the total zygote extract analyzed on a sucrose gradient either before or after reconstitution into the lipid bilayer. After centrifugation, liposomes floated toward the top of the gradient and more than 70%. of total liposome lipids and reconstituted proteins were found in the first three fractions (top of the gradient). Proteins which were not reconstituted remained at the bot tom of the tubes. Fig. 4A

also shows results of an important control where an aliquot of the reconstitution

mixture (detergent extract of the cells and phospholipids) was stored at 4°C in a tube (no dialysis) for the entire duration of dialysis of the remainder and analyzed by

gradient centrifugation along with other samples. In this case, all the proteins remain- ed at the bot tom of the gradient. Therefore, presence of proteins in the top fractions of the gradient after flotation of the dialyzed sample is actually due to their reconstitu- tion into lipid bilayers and not simply due to any possible flotation of proteins in the detergent micelles.

Proteins in the aqueous and detergent phases obtained after Triton X-114 phase separation of zygote extracts were reconstituted in lipid bilayers and analyzed by sucrose gradient centrifugation. The first two fractions from the top of the gradients and two fractions from the bot tom were pooled separately for analysis of proteins by SDS-PAGE. As shown in Fig. 4B, the majority of the 56 and 54 kDa proteins in the detergent phase were readily reincorporated into the lipid vesicles (lane 3). Proteins in the detergent phase which were not reconstituted are shown in lane 4. Most proteins in the aqueous phase were not reconstituted and were recovered in the bot tom fractions of the gradient (lane 2).

lmmunoprecipitation after phase separation. To learn more about the specificity of the MAbs which immunoprecipitate three proteins of 240, 56 and 54 kDa from detergent extracted cells, immunoprecipitations were repeated using the phase-separat-

353

H

40 -

Mr xlO -3 Top --

30 rt" LU > O

' " 20 n-"

,,_1 < I.- o I - 10

2 4 6 8

FRACTION NUMBER

200--

116-- 97.4-- 66.2- -

4 5 - -

B~bB--

1 2 3 4

Fig. 4. Sucrose density gradient and SDS-PAGE analysis of reconstituted samples. (A) 12SI-surface labelled

zygote extract was mixed with [~4C]dipalmitoyl phosphatidylcholine-egg lecithin vesicles and dialyzed for

reconstitution of proteins. Fractions (20 lad were collected from the top and counted for tz5I and t4C

radioactivity using a Beckman f-counter (Model 8000) and Tracor Analytic Mark III scintillation counter

(Tracor Analytic Inc.. Elk Village, IL). • *: ~4C radioactivity of lipids; • •: distribution of ~2~l-la-

belled parasite components after reconstitution and gradient analysis; m - - m : t-'51-1abelled extract (with-

out reconstitution) analyzed by gradient centrifugation. (B) Aqueous phase or the detergent phase of ~25I-la-

belled zygote extract was dialyzed for reconstitution into egg lecithin bilayers and analyzed by density gra-

dient flotation. Top two fractions and bot tom two fractions from each gradient were pooled separately

and analyzed by 5-15% SDS-PAGE. Lanes 1 and 2 are the top and bot tom pooled fractions, respectively,

of the aqueous phase and lanes 3 and 4 are the top and bot tom fractions of the detergent phase.

ed aqueous (which contains 240 kDa protein in addition to many others) and detergent (which contains 56 and 54 kDa proteins) phases. Fig. 5 shows the results with two MAbs, IID3B~ and IID4. From the detergent-solubilized total extract, MAb IID3B3 immunoprecipitated 240, 56 and 54 kDa proteins (panel A). If the extract was warmed and cooled twice (without separation of the phases from each other by centrifugation), the immunoprecipitation results with MAb IID3B3 were indistinguish- able from those in panel A (not shown). When phase-separated fractions were used, MAb IID3B3 immunoprecipitated the 240 kDa protein from the aqueous phase (panel B) but not the 56 and 54 kDa proteins in the detergent phase (panel C). MAb IID4 reacted with the 56 and 54 kDa proteins in the whole extract as well as in the detergent phase. The separated aqueous and detergent phases were also recombined for immu- noprecipitation; MAb IID3B3 reacted only with the 240 kDa protein and IID4 immu- noprecipitated 56 and 54 kDa proteins.

354

A B C D

-I t O

I I

1 2 1 2 1 2 1 2

Fig. 5. Immunoprecipitation with MAbs after phase separation. Triton X- 114 extracts of surface radioiodi- hated zygotes were subjected to two repeated cycles of phase separation, lmmunoprecipitations were done with MAb IID~B~ (lane 1) and MAb IID 4 (lane 2). Samples used were total extract (A), aqueous phase (B), detergent phase (C) and aqueous phase recombined with the detergent phase (D). Arrows indicate the positions of three proteins of interest to this study. Samples were analyzed by SDS-PAGE (5-15%) under reducing conditions.

Western blot analysis. P. gallinaceum zygote proteins were separated by SDS-PAGE

under reducing or non-reducing conditions and electroblotted onto nitrocellulose

paper. Proteins after transfer were detected by using various antibodies (Fig. 6).

MAbs IID3B3 and IID3E8 (a MAb with properties similar to IID3B3) reacted only with

the 240 kDa non-reduced protein (A, lanes 3 and 4) and MAb IID4 showed reactivity

with a protein band in the region of 45 kDa (A, lane 5). Rabbit anti-zygote serum

reacted with a number of protein bands including 240 kDa, and a doublet of 50 and 45

kDa proteins (A, lane 2). Reactivity of anti-zygote serum as well as MAbs to 240 kDa

protein was lost if proteins were reduced for SDS-PAGE (Fig. 6B). A 25 kDa band and a few others present in all the lanes including lane 1 (normal rabbit serum) are probably

non-specific.

Binding and elution of proteins from immunoaffinity beads. In view of the results in Figs. 1-3, MAbs IID3B3 and IID4 were conjugated to protein A-Sepharose beads by

chemical cross-linking to study binding and elution of antigens, eventually for the immunoaffinity purification of individual protein antigens from the aqueous and

detergent phases. About 35% of total radioactivity in the antigens from the upper aqueous phase was bound to the MAb IID3B3-protein A-Sepharose beads and none was bound from the lower detergent phase. Several solvents were tested to elute bound

355

A

It i l

FOe i,.. -.~

~ ) 0 0 J,, - '

97 ,.- : 6 8 "

26 ,.- 18,.-

: 12 - - ,'

B B"

,i m

i

1 2 3 4 5 1 2 3 4 5

Fig. 6. Western blot analysis. P. gallinaceum zygotes proteins separated by SDS-PAGE under non-reducing (A) or reducing (B) conditions were electroblotted onto nitrocellulose paper. Strips of nitrocellulose were treated with normal rabbit serum (lane 1), rabbit anti-zygote serum (lane 2), MAb IID3B3 (lane 3), MAb IID3Es; a MAb-like IID3B 3 (lane 4), and MAb IID4 (lane 5). Arrows indicate the expected positions of 240 kDa and a doublet of 50 kDa/45 kDa proteins.

antigens. The most satisfactory (85-90% elution) results were obtained either with 0.1

M acetic acid + 0.5% Triton X- 100 or with 50 mM diethylamine + 0.5% Triton X-100.

4 M KSCN caused only 65% elution of bound antigens. Fig. 7 shows the proteins

remained unbound and eluted when the aqueous phase (panel l) or total extract (panel

2) were used for binding and elution from beads conjugated with MAb IID3B3. 240

kDa protein from the aqueous phase was bound to the MAb and subsequently eluted

in acetic acid or diethylamine. On the other hand, 240, 56 and 54 kDa proteins were

bound and subsequently eluted when total extract was used. 56 and 54 kDa proteins from the total extract (panel 3) or from the detergent phase

(panel 4) could be purified when beads containing MAb IID4 (Fig. 7)were used. 56 and

54 kDa proteins were retained by the beads and subsequently eluted either in acetic

acid or diethylamine.

DISCUSSION

Phase separation of Triton X-114 results in the separation of the detergent-binding hydrophobic proteins from relatively less hydrophobic proteins, as shown by Bordier

in 1981 [4]. The phase separation technique was used in the present studies to learn more about the proteins in the sexual and mosquito midgut stages of malaria parasite.

356

I 2 3

UB E UB E UB E UB E

Fig. 7. Binding and elution of proteins from protein A-Sepharose beads containing MAb IID3B 3 or MAb IID4. Protein A-Sepharose beads containing MAb IID3B3 were incubated with the proteins in the aqueous phase (panel 1) or whole cell extract (panel 2). Beads were washed and eluted using 0.1 M acetic acid + 0.5% Triton X-100 (see Materials and Methods). In panels 3 and 4, proteins in the whole cell extract or the detergent phase, respectively, were bound to and eluted from MAb liD:containing beads. Unbound (UB) and eluted (E) samples were analyzed using 5-15% SDS-PAGE under reducing conditions. Position of 240 kDa protein (single arrow) and of 56 and 54 kDa proteins (double arrows) are indicated.

Earlier studies f rom our labora tory have shown that monoc lona l antibodies which are

efficient in blocking transmission by blocking fertilization of gametes in the mosqui to

midgut immunoprecipi ta te a set of three proteins of apparent 240, 56 and 54 kDa in P.

gallinaceum and 260, 59 and 53 kDa in P. falciparum [1,2]. In each species these

proteins were not found to be biosynthetically related to each other as shown by

pulse-chase labelling studies ([3], D.C. Kaushal and R. Carter, unpublished results).

Using Tri ton X- 114 phase separat ion technique, the 240 kDa protein ofP. gallinaceum remained in the detergent-free aqueous phase whereas 56 and 54 kDa proteins were

essentially recovered in the detergent phase. Similar results were obtained for the

distribution o f corresponding proteins in P. falciparum (N. K., unpublished results).

Phase separation of biosynthetically labelled proteins or iodination of proteins after phase separation showed that only a select group of proteins part i t ion with the

detergent. This approach offers an easy method to separate the 240 kDa protein

antigen from the 56 and 54 kDa proteins, which are otherwise immunoprecipi ta ted

together by transmission blocking monoclonal antibodies. In agreement with the phase separat ion data, hydrophobic ch roma tog raphy and

reconsti tut ion into lipid bilayer experiments also demonst ra ted that the 56 and 54 kDa proteins are highly hydrophobic , whereas 240 kDa protein is hydrophil ic or relatively

much less hydrophobic .

357

I m m u n o p r e c i p i t a t i o n af ter phase sepa ra t ion revealed that the fer t i l iza t ion b locking

M A b s in P. gallinaceum, which i m m u n o p r e c i p i t a t e all three pro te ins f rom the whole

extract , are ac tua l ly d i rec ted agains t 240 k D a pro te in and the two lower molecu la r

weight p ro te ins appea r to be co-prec ip i t a t ed by certain M A b s . This obse rva t ion was

also conf i rmed by Wes te rn b lo t analysis of p ro te ins where these M A b s reac ted only

with the 240 k D a prote in . The react iv i ty of an t ibod ies to 240 k D a target ant igen was

lost if the pro te ins were reduced. These results thus imply tha t 240 k D a pro te in on the

gametes of P. gallinaceum is the sole target ant igen of these t r ansmis s ion -b lock ing

M A b s . The fact that when phase - sepa ra t ed fract ions are r ecombined the 240 k D a

p ro te in and 56 and 54 k D a pro te ins do not reassociate also suggests that phase

sepa ra t ion results in some sor t o f i r revers ible changes ei ther in the pro te ins or in their

immed ia t e env i ronment .

A c o m b i n a t i o n of ana ly t ica l phase sepa ra t ion technique and immunoaf f in i ty purif i-

ca t ion p r o d u c e d appa ren t l y pure p repa ra t i ons o f the 240, 56 and 54 k D a doub le t

prote ins . These steps could be scaled up for p r e p a r a t o r y i so la t ion of the target p ro te in

ant igens for future studies. S imi la r app roaches could be used for the enr ichment aiad

immunoa f f in i t y pur i f i ca t ion of ook ine te ant igens.

ACKNOWLEDGEMENTS

I t h a n k Dr. R icha rd Car t e r for encou ragemen t and va luable discussions, Drs. Louis

H. Miller , R icha rd Klausne r and Lore Ann McNico l for c o m m e n t s on the manuscr ip t ,

Mrs. Rosanne H e a r n for technical suppor t and Mrs. Wi lma Davis and Mrs. Brenda

Mar t in for ed i tor ia l assistance. These studies received a par t ia l f inancia l suppor t o f the

U N D P / W o r l d B a n k / W H O special p r o g r a m m e for Research and Tra in ing in Trop ica l

Diseases and Senior Research Fe l lowsh ip f rom the Bur roughs Wel l come F o u n d a t i o n .

REFERENCES

1 Rener, J., Graves, P.M., Carter, R., Williams, J.L. and Burkot, T. (1983) Target antigens of transmis- sion blocking immunity on gametes of Plasmodium falciparum. J. Exp. Med. 158, 976-981.

2 Kaushal, D.C., Carter, R., Rener, J., Grotendorst, C.A., Miller, L.H. and Howard, R.J. (1983) Monoclonal antibodies against surface determinants on gametes of Plasmodium gallinaceum block transmission of malaria parasites to mosquitoes. J. Immunol. 131, 2557-2562.

3 Kumar, N. and Carter, R. (1984) Biosynthesis of the target antigens of antibodies blocking transmis- sion of Plasmodiumfalciparum. Mol. Biochem. Parasitol. 13, 333-342.

4 Grotendorst, C., Kumar, N., Carter, R. and Kaushal, D.C. (1984) A surface protein expressed during the transformation of zygotes of P. gaHinaceum is a target of transmission blocking antibodies. Infect. lmmun. 45, 775-777.

5 Kumar, N. and Carter, R. (1985) Biosynthesis of two stage-specific membrane proteins during transformation of Plasmodium gallinaceum zygotes into ookinetes. Mol. Biochem. Parasitol. 14, 127-139.

6 Carter, R. and Kaushal, D.C. (1984) Characterization of antigens on mosquito midgut stages of Plasmodium gallinaceum. III. Changes in zygote surface proteins during transformation to mature ookinetes. Mol. Biochem. Parasitol. 13, 235-241.

358

7 Bordier, C. (1981) Phase separation of integral membrane proteins in Triton X-114 solution. J. Biol.

Chem. 256, 1604-1607. 8 Kaushal, D.C. and Carter, R. (1984) Surface antigens on mosquito midgut stages of Plasmodium

gallinaceum. II. Comparison of surface antigens of male and female gametes and zygotes. Mol. Biocbem. Parasitol. 11, 145-156.

9 Alcaraz, G., Kinet, Jean-Pierre, Kumar, N. and Metzger. H. (1984) Phase separation of the receptor for immunoglobulin E and its subunits in Triton X-I14. 3. Biol. Chem. 259, 14922-14927.

10 Carson, S.D. and Konigsberg, W.H. (1981) PhenyI-Sepharose chromatography of membrane pro- teins solubilized in Triton X-100. Anal. Biochem. 116, 398-401.

11 Rivnay, B. and Metzger, H. (1982) Reconstitution of the receptor for immunoglobulin E into liposomes. J. Biol. Chem. 257, 12800-12808.

12 Schneider, C., Newman, R.A., Sutherland, D.R., Asser, V. and Greaves, M.F. (1982) A one-step purification of membrane proteins using a high efficiency immunomatrix. J. Biol. Chem. 257,

10766-10769. 13 Laemmli, U.K. (1970) Cleavage of structural proteins during assembly of the head of bacteriophage

T,. Nature 227, 680-685. 14 Towbin, H., Stachlin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacryl-

amide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U.S.A. 76, 4350-4354.