Rapid whole blood assay for HIV-1 seropositivity using an Fab-peptide conjugate

111

Journal of Immunological Methods, 138 (1991) 111-119 © 1991 Elsevier Science Publishers B.V. 0022-1759/91/$03.50 ADONIS 002217599100132F

JIM 05908

Rapid whole blood assay for HIV-1 seropositivity using an Fab-peptide conjugate * K i m M. Wilson 1, Michael G e r o m e t t a 2, D e n n i s B. R y l a t t 2, Peter G. B u n d e s e n 2, Dale A. M c P h e e 3, C a r m e l J. Hillyard 2 a n d Bruce E. K e m p 1 I St. Vincent's Institute of Medical Research, 41 Victoria Parade, Fitzroy, Victoria 3065, Australia, 2 Agen Limited, l l Durbell St., Acacia Ridge, Queensland 4110, Australia, and 3 AIDS Cell Biology Laboratory, Macfarlane Burnet Centre for Medical Research, Fairfield Hospital, Fairfield, Victoria 3078, Australia (Received 15 November 1990, revised received 17 January 1991, accepted 21 January 1991)

A rapid whole blood test has been developed for circulating antibodies to human immunodeficiency virus type 1 (HIV-1), based on agglutination of autologous red blood cells. Evaluation of the test revealed that 100% of seropositive HIV-1 patients (both asymptomatic and AIDS cases) were detected (n = 94) with a specificity of 99.5% in healthy blood donors (n = 596). The assay uses an Fab fragment of a monoclonal antibody specifically directed against glycophorin (a transmembrane glycoprotein present on the surface of human red blood cells). This anti-red blood cell Fab is conjugated via the inter-heavy chain cysteines to a synthetic peptide corresponding to the immunodominant epitope of the HIV-1 viral coat protein gp41 (579-613). Addition of this reagent to 10 /~1 of whole blood results in the Fab-peptide conjugate coating the red blood cells with peptide. In the presence of circulating antibodies to the HIV-1 peptide, red cell agglutination occurs within 2 rain. The sensitivity and specificity of this reagent indicate that it is appropriate for use as a rapid diagnostic test for HIV-1 seropositivity. Key words: HIV seropositivity; Antibody; Agglutination test

Introduction

In order to prevent the spread of AIDS via contaminated blood products and protect health care workers, a simple, reliable means of detecting HIV-1 infected individuals is essential. The most

Correspondence to: B.E. Kemp, St. Vincent's Institute of Medical Research, 41 Victoria Parade, Fitzroy, Victoria 3065, Australia. * This work was supported by a Commonwealth AIDS Research Grant, Agen (Biomedical Ltd.), the NH&MRC and an Industrial Research and Development grant.

widely used serological test for antibodies to HIV-1 is the enzyme-linked immunosorbent assay (EIA). If an individual is repeatedly reactive a supplemental test is required, usually an immunoblot (Schochetman et al., 1989). These tests utilize partially purified viral antigens derived from whole disrupted virus, recombinant proteins or synthetic peptides. Currently there are rapid tests based on particle agglutination (Francis et al., 1988), hemagglutination (Scheffel et al., 1990) and immunoblotting (Santos et al., 1987) which compare favourably in sensitivity to the EIA, but require the use of serum rather than whole blood. The urgent need for simpler test methods has been highlighted by reports of rapid increases in the

112 numbers of infected people in Africa and the Asian continent (De Cock et al., 1990). Previously, we described the principle of an autologous red cell agglutination test for antibodies to HIV-1. The test is somewhat similar to the classical Coombs test (Coombs et al., 1950) except that the patient's own red blood cells are used as the indicator rather than exogenous red blood cells. It is simple to perform requiring only the addition of a single reagent to 10 /~1 of whole blood. The blood requires no separation or treatment and an answer is obtained within 2 min. In the assay a synthetic peptide corresponding to the immunodominant region of the HIV-1 envelope glycoprotein gp41 (residues 579-601) is chemically coupled via the heterobifunctional reagent N-succinimidyl-3(2-pyridyldithio) p r o p i o n a t e (SPDP) to an intact antibody which recognises human red blood cells (Kemp et al., 1988; Rylatt et al., 1990). The heterobifunctional reagent used in this procedure results in random crosslinking of the peptide antigen to lysines on the antibody. This paper describes the preparation of a chemically defined, specific, stable reagent through the use of antibody Fab fragments rather than the intact antibody. We have coupled an extended gp41 immunodominant peptide, corresponding to residues 579-613 of the HIV-1 envelope glycoprotein, via the cysteine residues normally used to link the antibody heavy chains which are distal from the Fab fragment's glycophorin binding site. This has resulted in a greatly improved reagent with excellent sensitivity, specificity and stability.

Methods

equilibrated in phosphate-buffered saline (PBS) to remove the cysteine. Activated papain (0.76 U) was added per mg of antibody and digestion at 37°C in the absence of added thiols was monitored by gel permeation HPLC. If antibody digestion with papain was carried out in the presence of cysteine, Fab rather than F(ab')2 fragments were generated. When digestion was 90% complete (1-4 h) 30 mM iodoacetamide was added and incubated for 30 min at 25°C to inactivate the papain. The resultant F(ab')2 was purified by gel filtration chromatography on a Sephacryl S-200 column equilibrated in PBS.

Preparation of Fab-TNB derivative The papain F(ab')2 fragments were reduced with 5 mM mercaptoethylamine, 10 mM EDTA in PBS at 37°C and the reaction monitored by gel filtration HPLC to confirm 50% reduction at 30 min. At 60 min 10 m g / m l of 5,5'-dithiobis-(2nitrobenzoic acid) (DTNB) was added (Brennan et al., 1985). After 5 min any residual sulphydryl residues were blocked with iodoacetamide (30 mM) for 30 min at 25°C. The Fab-TNB derivative was purified by gel filtration on Ultrogel ACA44 in PBS. The stoichiometry of TNB protection varied between 1-3 m o l / m o l of antibody determined by measuring the absorbance at 412 nm following reduction of an aliquot of the Fab-TNB derivative with 30 mM dithiothreitol (DTT).

Peptide synthesis The gp41 (579-613) synthetic peptide was synthesized by the Merrifield solid phase synthesis procedure as described previously (Kemp et al., 1988) or supplied by IAF BioChem International (Montreal, Canada).

Production of F(ab')2 fragments An anti-glycophorin monoclonal antibody was raised by immunizing B A L B / c mice with washed human erythrocytes and screened by spin agglutination and glycophorin specific EIA (Kemp et al., 1988). F(ab')2 fragments were prepared from intact antibody by papain digestion (Parham et al., 1982) because very poor yields were obtained with pepsin digestion of this particular antibody. Papain (95 U) was activated with 6 mg of cysteine for 30 min at 37°C then passed through a gel filtration column (PD-10 from Pharmacia)

Fab-peptide conjugate construct Peptide gp41 (579-613) was dissolved in 100 mM Tris- HCI, pH 8.0 containing 1 mM EDTA, 4 M guanidine hydrochloride and reduced with 50 mM D T T for 40 min at 25°C. Reduced peptide was isolated by batch chromatography on a C18 Sep Pak from Waters Associates and eluted with 60% acetonitrile containing 10 mM HC1 then rotary evaporated to dryness at 40°C. Reduced peptide was redissolved in 100 mM Tris. HC1 pH 7.4, 1 mM E D T A and added in ten-fold excess to

113 the Fab-TNB for 1 0 m i n at 25°C. For larger scale preparations, where large amounts of synthetic peptide are consumed, equimolar amounts of reduced peptide gave satisfactory yields. Peptide substitution resulted in the release of T N B and this was quantitated by monitoring at 412 nm. The reaction was terminated by blocking free sulphydryls with the addition of 30 m M iodoacetamide for 60 min at 25°C and the conjugate purified by cation exchange chromatography on a Mono S F P L C column from Pharmacia (0-300 m M NaC1 gradient in 20 m M sodium phosphate p H 7.4 run over 60 min at 1 m l / m i n ) or by affinity chromatography using a monoclonal antigp41 (579-601) peptide antibody coupled to CNBr activated Sepharose. Purity of the product was assessed by electrophoresis on a 10% polyacrylamide gel run in sodium dodecyl sulphate (10% SDS PAGE) at 100 V for 16 h at 10°C. The Fab-peptide conjugate was identified using Coomassie blue staining and immunoblotting. Apparent molecular weights (Mr) were determined using prestained, low range,

molecular weight standards from Bio-Rad. Bands from the gel were transferred to nitrocellulose for 2 h at 60 V, blocked with 5% skim milk powder and reacted with heat inactivated human plasma positive for HIV-1 antibodies. The immunoblot was developed with horseradish peroxidase coupled to rabbit anti-human I g G antibody and 3,3'diaminobenzidine was used as a substrate according to the manufacturer's instructions.

Storage and assay of Fab-peptide conjugate The Fab-peptide conjugate was stored at 10 or 25 /~g/ml in PBS containing 0.05% w / v sodium azide, 0.5 m g / m l mouse blocking antibody, 0.5% v / v fish gelatin and 0.1 m g / m l Fab-TNB which had been blocked with iodoacetamide and derived from the unrelated blocking antibody. The unrelated mouse antibody (to Brucella abortus) and its F a b - T N B derivative were included to block any heterophile antibody present in patients' blood and suppress aggregation of the reagent that may give rise to a false positive result. The assay format involved the addition of 25 ~1 of the Fab-

papain

mercaptoethylamine "--

"-

F(al~)2

INTACT ANTIBODY

S - S peptide S - S peptide ~ S - S peptide

peptide SH [ l iodoacetamide

I-SH

Fab

S - TNB - S - TNB I- S - TNB

I PEPTIDE-Fab

TNB-Fab

CONJUGATE DERIVATIVE Fig. 1. Flow diagram for the preparation of Fab-peptide conjugate. Intact anti-human erythrocyte antibody was digested with papain to produce F(ab')2 fragments, which in turn were reduced with mercaptoethylam/ne to Fab fragments. The Fab fragments were stabilized with DTNB and the TNB substituted with reduced synthetic peptide gp41 (579-613). Any remaining free sulphydryls were alkylated with iodoacetamide resulting in a stable Fab-peptide conjugate.

114 peptide conjugate reagent to 10/~l of whole blood on a plastic tray containing 4 X 6 shallow wells. The reagent and blood sample were mixed with a plastic stirrer, rocked manually for 2 min and then read visually. Agglutination was scored in arbitrary units (0-4) with 0 being no agglutination and 4 for the highest level of agglutination. The agglutination specificity was assessed by competition with the free peptide gp41 (579-613). Various amounts of peptide were dissolved in PBS and 10 ffl added to the HIV-1 positive whole blood prior to the addition of Fab-peptide conjugate. This resulted in inhibition of agglutination. All blood was collected into heparinized tubes and tested within 24 h of collection.

Results

Reagent preparation The strategy employed in the preparation of the Fab-peptide conjugate is illustrated in Fig. 1. It is based on the procedure developed by Brennan et al. (1985) for the preparation of bispecific antibodies. In the case of the anti-red blood cell antibody, conventional cleavage with pepsin to generate F(ab')2 fragments resulted in very poor yields. It was found for this particular antibody that digestion with activated papain in the absence of sulphydryl reagents gave better yields of F(ab')2 than pepsin. The resultant F(ab')2 fragments were reduced to produce Fab fragments and these were protected with DTNB. Addition of reduced gp41 (579-613) displaced the TNB to give an Fabpeptide conjugate with a yield of 67%. Any residual sulphydryl groups were blocked with iodoacetamide. This proved to be a vital step as it prevented disulphide exchange and the rapid loss of peptide from the Fab fragment. The Fab-peptide conjugate was separated from Fab-TNB, free peptide and small molecular weight contaminants on a Mono S cation exchange column (Fig. 2). The three initial peaks (Fig. 2) corresponded to batch loading (three injections prior to running the gradient) of the reaction mixture. Excess iodoacetamide and TNB were not retained on the column. Analysis of the eluted peaks with 10% SDS P A G E stained with Coomassie blue (Fig. 3A) revealed that peak I contained

10"

.10

09-

"'9

'0~"

"'8

-07-

-.7

06

-.6

g .05-

-.5

o

~ 04-'3 ~

03-02-

-'2

01-

-'1

O"

20

~0 Tirne(rnin)

6O

0

Fig. 2. Cation exchange purification of the Fab-peptide conjugate. The Fab-peptide conjugate reaction mixture was applied to a Mono S column with three injections and developed with a gradient 0 300 mM NaC1 in 20 mM sodium phosphate run over 60 min at l ml/min. Peak I contained unconjugated Fab-TNB. Peak II contained conjugated Fab-peptide. The three flow through peaks corresponding to the three sample injections did not contain protein (see Fig. 3A).

the unconjugated F a b - T N B at M r 41 kDa (Fig. 3A, lane D) and peak II contained the Fab-peptide conjugate M r 45 k D a (Fig. 3A, lane E). Conjugation of.the peptide to the Fab caused an apparent increase in M r of 3.9 kDa. The elution of the peptide conjugate later than the Fab-TNB derivative is consistent with it having additional basic residues contributed by the immunodominant peptide. Comparison of the SDS P A G E fractions from the Mono S chromatography with the corresponding immunoblot (Fig. 3B) shows that the Fab-TNB (Fig. 3B, lane A) was not recognized but the Fab-peptide conjugate was specifically recognized by heat inactivated HIV-1 positive plasma (Fig. 3B, lane E). The unconjugated peptide (Fig. 3B, lane F) running near the dye front gave a strong reaction. There was some immunoblot positive material present in lane D (Fig. 3B) due to carry-over of Fab-peptide conjugate into peak I. This material comigrated with

115 TABLE I CLINICAL EVALUATION OF AUTOLOGOUS RED CELL AGGLUTINATION TEST Blood samples were obtained from HIV-1 positive patients (both asymptomatic and AIDS cases). These patients had been confirmed positive by Abbott EIA and immunoblot. Healthy blood donors were tested by Genetic Systems EIA. HIV-1 negative hospitalized patients exhibiting a variety of infectious diseases were tested by Abbott EIA. False positives were retested by Abbott EIA and immunoblot. 100% of HIV-1 positive patients were detected. A specificity of 99.5% was observed in healthy blood donors and 97.4% in HIV-1 seronegative hospitalized patients. Subject

Total number

Agglutination assay

EIA test

Positive

Positive

Negative

Negative

HIV-1 seropositive patients

94

94

0

94

0

Hospitalized patients

76

2

74

0

76

596

3

593

0

596

Healthy blood donors

A) A B C

D

E

F G

H

kDa A B C

D E F G H

kDa

--110

--110

--84

--84

--47

--47

--33

--33

--24 --24

-

--16 "--'

.

.

.

.

.

.

.

16

Fig. 3.A: analysis of Mono S Peaks on 10% SDS PAGE stained with Coomassie blue. (A) Fab-TNB starting material, (B) Fab-peptide conjugate prior to purification, (C) flow through peaks, (D) peak I Fab-TNB, (E) peak II Fab-peptide conjugate, (F) peptide gp41 (579-613), (G) ion-exchange purified Fab-peptide conjugate, (H) affinity purified Fab-peptide conjugate; molecular weight markers are indicated to the right. Fig. 3.B: analysis of Mono S Peaks with an immunoblot following 10% SDS PAGE. The immunoblot was prepared as described under Methods. (A) Fab-TNB starting material, (B) Fab-peptide conjugate prior to purification, (C) flow through peaks, (D) peak I Fab-TNB, (E) peak II Fab-peptide conjugate, (F) peptide gp41 (579-613), (G) ion-exchange purified Fab-peptide conjugate, (H) affinity purified Fab-peptide conjugate; molecular weight markers are indicated at right.

116

Neg

1+

2+

3+

4+ t/

Fig. 4. Visual agglutination scale for agglutination of HIV-1 seropositivepatient's red blood cells. The Fab-peptide conjugate was added to either control HIV-1 seronegativewhole blood or HIV-1 seropositivewhole blood with various levels of antibody and scored after 2 min.

the conjugate at M r = 45 kDa and not Fab-TNB with M r = 41 kDa (Fig 3A, lane D). A densitometric scan of the Coomassie blue stained gel showing the conjugate prior to purification (Fig. 3A, lane B) indicated that the yield of conjugate was approximately 67%. The immunoblot revealed small amounts of higher molecular weight species corresponding to the multiple addition (from 2 to 4) of 3.9 kDa peptides to the Fab fragments. Since there are three inter-heavy chain sulphydryl residues in mouse IgG1 the maximum degree of substitution at these sites is 3 mol of peptide/mol of Fab. The minor top band in lane B at 54.4 kDa representing four peptides per Fab, indicates that some peptide was binding through additional sulphydryls. It is apparent from the Coomassie blue stained gels and the immunoblot of the conjugate prior to purification that the major product was mono-

substituted with smaller amounts of di- and trisubstituted Fab. The di- and tri-substituted Fab fragments were not evident after cation exchange chromatography. Although the higher molecular weight conjugates were quantitatively minor species on a Coomassie blue stained gel, they appeared as stronger bands on the immunoblot due to the number of peptides present rather than the amount of protein. Peak I (Fig. 3A, lane D) and peak II (Fig. 3A, lane E) suggest that the purification step results in a homogeneous conjugate. Ion exchange purified conjugate (Fig. 3A, lane G) and affinity purified conjugate (Fig. 3A, lane H) were both free of unconjugated peptide and Fab-TNB. The immunoblot also revealed the presence of trace amounts of pepfide coupled to free light chain in the purified Fab-peptide conjugate. The specificity of agglu*ination was assessed by competition with excess free synthetic HIV-1

117 peptide which was added to the HIV-1 positive whole blood prior to the addition of Fab-peptide conjugate. The concentration of conjugate used in the assay was 0.42 /~M. A two fold excess of free peptide over conjugate produced detectable inhibition of agglutination while a thirty fold excess was sufficient to inhibit agglutination completely. A variety of unrelated synthetic peptides did not inhibit the agglutination reaction. The amount of peptide required to block agglutination varied with the level of antibody present in the whole blood sample. Eoaluation of the test The agglutination reaction was scored visually and the scale used is illustrated in Fig 4. An agglutination level of I + is readily apparent when visualized with the plastic agglutination plate on a shaker but is less evident in a still photograph. The dependence of agglutination on conjugate concentration was evaluated. The test was performed using the conjugate at both 25 ~tg/ml and 10/~g/ml. At 25/~g/ml conjugate 80% of samples had a score of 3 or better whereas at 10 /~g/ml conjugate 58% of samples had a score of 3 or better. There was only one sample out of 94 detected at 25 btg/ml that failed to be detected at 10 t~g/ml. The choice of peptide to be used in the test was based on results obtained from EIA and performance in the agglutination assay. Previously we had used a shorter synthetic peptide gp41(579601) but the peptide gp41(579-613) displayed a stronger agglutination score in 25% of seropositive patients tested (n = 71). The Fab-peptide conjugate was tested against 94 HIV-1 seropositive patients (both asymptomatic and AIDS cases), 76 hospitalized patients exhibiting a variety of infectious disease and 596 healthy blood donors (Table I). This resulted in 100% detection of the HIV-1 seropositive patients while a specificity of 99.5% was found in the healthy blood donors and 97.4% in the seronegative hospitalized patients. The two hospitalized patients who gave false positive results were tested by Abbott EIA and found to be seronegative for antibodies to HIV-1. The three blood donors out of 596 who were detected, tested negative by Abbott EIA and did not exhibit antibodies against

the peptide, the conjugate or the unconjugated antibody as assessed by immunoblot analysis. However, addition of free peptide gp41(579-613) completely inhibited red cell agglutination while two unrelated peptides had no effect. The reason for the false positive results of the three blood donors is not known. A blood sample, not included in Table I, was obtained from an AIDS patient who was HIV-1 antibody negative. Virus had been isolated from the patient's peripheral blood mononuclear cells in culture and assessed by reverse transcriptase assay and detection of antigen by EIA (Innotest, Wellcome, England). This patient's blood sample was also negative with a peptide EIA, and commercial tests including Wellcome EIA, Immunoblot and Fujirebio Serodia particle agglutination test.

Discussion

We have adapted the strategy developed by Brennan et al. (1985) for preparing bispecific antibodies to construct an Fab-peptide conjugate. The peptide is linked by a disulphide bond to one or more of the cysteines 221, 224 and 226 of the amino acid sequence (Sakano et al., 1979) in the murine Ig4-1 heavy chain hinge region (Fig. 1). This results in the peptide being linked to the Fab distal from its binding site for glycophorin on the red cells. Thus when the Fab-peptide conjugate binds to red blood cells they are coated so that the peptide antigen is directed away from the red blood cell, thereby ensuring ready access to circulating antibodies. Previously we had used intact anti-erythrocyte antibody to prepare conjugates with a heterobifunctional crosslinking reagent but this approach had the disadvantage of linking the peptide randomly to the antibody via lysines. This random coupling may interfere with binding of the conjugate to the red blood cell. An important feature of the autologous red cell agglutination test reagent is that it does not cause false positive agglutination in the absence of circulating antibodies directed against the peptide antigen. The Fc region of antibodies is involved in numerous effector functions including complement binding, macrophage recognition, interactions with rheumatoid factor and heterophile anti-

118

bodies. By removing the Fc region and using Fab fragments the likelihood of obtaining a false positive agglutination through non-specific interaction is reduced. This has also meant that the amount of blocking antibody required to eliminate heterophile reactions can be reduced. We have used both ion-exchange chromatography and affinity chromatography to prepare purified conjugate. The removal of unconjugated FabTNB has allowed the optimal level of Fab-peptide conjugate required for maximum agglutination to be assessed. A further improvement was obtained by using the extended synthetic peptide gp41 (579-613). Previously we had used a shorter version gp41 (579-601) but the peptide gp41 (579613) encompasses the disulphide loop of the immunodominant region of gp41. Virtually all seropositive patients have been shown by EIA (Wang et al., 1986; Gnann et al., 1987) to recognize this region containing the loop. A direct comparison in which both of these peptides were coupled to the Fab fragments revealed that gp41(579-613) had the greater sensitivity. While it seems reasonable that only one of the peptide cysteines forms a disulphide with the Fab fragment, thereby preventing loop formation, it is possible that the construct may mimic the presence of the native disulphide loop within the immunodominant peptide, since all HIV-1 seropositive samples were detected by the agglutination test. In the absence of blocking with iodoacetamide there was a rapid loss of the peptide from the Fab conjugate possibly due to disulphide exchange. This was prevented by alkylating free sulphydryls following conjugation. We were unable to improve conversion of the Fab-TNB to Fab-peptide conjugate over a yield of 67%. The reason for this is not clear, but may result from steric effects, oxidation of the heavy chain cysteine residues during the conjugation reaction or the presence of irreversibly blocked cysteines produced when the Fab-TNB derivative is alkylated with iodoacetamide. Clinical trials have detected 100% of HIV-1 seropositive patients. The HIV-1 positive individual who was not detected by the autologous red cell agglutination test was also found to be negative by four other currently accepted antibody

tests, indicating the absence of any antibodies to HIV-1. However, isolation of virus by co-culture of patient and donor peripheral blood mononuclear cells on numerous occasions confirmed HIV-1 infection in this patient. The reason for the three false positive agglutination tests among the 596 blood donors is not clear. The reaction was blocked by free synthetic peptide gp41 (579-613) implying that agglutination was due to crosslinking of the Fab conjugate via the peptide. This does not appear to be due to circulating antibody because the patient's sera failed to detect the Fabpeptide conjugate using a sensitive immunoblot. Further work is required to establish the nature of the material responsible for the infrequent false positive tests among blood donors. The major advantages of the red cell agglutination test are its speed and simplicity while still providing adequate sensitivity. The Fab-peptide reagent described here would be useful as a first line test to screen blood donors, casualty patients, high risk individuals and populations in third world countries. The test also offers the potential for undertaking simple serology screens useful in epidemiology studies. Because of the 'on the spot' nature of the test it permits the immediate testing of blood samples that may be used for blood transfusions or where a knowledge of the HIV-1 antibody status is important. Outside the area of HIV-1 diagnostic testing the autologous red cell agglutination test has potential applications in identifying antibody responses to a wide range of infectious diseases as well as monitoring vaccination programs.

References Brennan, M., Davison, P.F. and Paulus, H. (1985) Preparation of bispecific antibodies by chemical recombination of monoclonal immunoglobulin G 1 fragments. Science 229, 81. Coombs, R.R.A., Mynors, L.S. and Weber, G. (1950) The use of the indirect sensitization technique as a method of titrating antibodies against a chemical compound. Br. J. Exp. Pathol. 31,640. De Cock, K.M.., Barrere, B. and Diaby, L. (1990) AIDS - The leading cause of adult death in the West African city of Abidjan, Ivory Coast. Science 249, 793. Francis, H.L., Kabeya, M., Daufuana, N., et al. (1988) Comparison of sensitivities and specificities of latex agglutina-

119 tion and an enzyme-linked immunosorbent assay for detection of antibodies to the human immunodeficiency virus in African sera. J. Clin. Microbiol. 2, 2462. Gnann, J.W., Nelson, J.A. and Oldstone, B.A.. (1987) Fine mapping of an immunodominant domain in the transmembrane glycoprotein of human immunodeficiency virus. J. Virol. 61, 2639. Kemp, B.E., Rylatt, D.B., Bundesen, P.G., et al. (1988) Autologous red cell agglutination assay for HIV-1 antibodies: Simplified test with whole blood. Science 241, 1352. Parham, P., Androlewicz, M.J., Brodsky, F.M., Holmes, N.J. and Ways, J.P. (1982) Monoclonal antibodies: Purification, fragmentation and application to structural and functional studies of class I MHC antigens. J. Immunol. Methods 53, 133. Rylatt, D.B., Kemp, B.E., Bundesen, P.G., et al. (1990) A rapid whole-blood immunoassay system. Med. J. Aust. 152, 75. Sakano, H., Rogers, J.H., Huppi, K., et al. (1979) Domains and the hinge region of an immunoglobulin heavy chain are encoded in separate DNA segments. Nature 277, 627.

Santos, J.I., Galvao-Castro, B., Mello, D.C., Pereira, H.G. and Pereira, M.S. (1987) Dot enzyme immunoassay: A simple cheap and stable test for antibody to human immunodeficiency virus (HIV). J. Immunol. Methods 99, 191. Scheffel, J.W., Wiesner, D., Kapsalis, A., Taylor, D. and Suarez, A. (1990) Retrocell HIV-1 passive hemagglutination assay for HIV-1 antibody screening. J. Acquir. Immune Defic. Syndr. 3, 540. Schochetman, G., Epstein, J.S. and Suck, T.F. (1989) Serodiagnosis of infection with the AIDS virus and other human retroviruses. Annu. Rev. Microbiol. 43, 629. Wang, J.J.G., Steel, S., Wisniewolski, R. and Wang, C.Y. (1986) Detection of antibodies to human T-lymphotropic virus type III by using a synthetic peptide of 21 amino acid residues corresponding to a highly antigenic segment of gp41 envelope protein. Proc. Natl. Acad. Sci. U.S.A. 83, 6159.