CA1293188C - Composition of matter and method of immunizing against viral causative agents of aids and arc - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE Composition and method for induction of a neutralizing anti-body against the viral causative agents of AIDS and ARC. The composition is a conjugate of a polyamide resin and a synthetic peptide, the amino acid sequence of the synthetic peptide being sufficiently homologous to the amino acid sequence of the gp 41 and gp 120 subunits of the gp 160 envelope glycoprotein of HTLV-III, LAV or ARC to produce an immunogenic response in an experi-mental animal and having a hydrophilic region therein. The com-position is given with or without an adjuvant in an amount effec-tive to induce an immunogenic response, thereby protecting the experimental animal from exposure to the HTLV-III, LAV and/or ARV causative agents of AIDS and/or ARC. 01/MFM2

Description

31~ 1 STATEMENT OF GOYERNMENT INTEREST
This invention was made with government support under one orj more of the following NIH grants: AI-22307-01, ~ AI- 7/~s 23619-01, and AI-23472-Ul. The government has certain rights in , th invention.

10 1, I!
. .
~, BACXGROUND O~ THE INVENTION
. ~ . ~
The present invention relates to a composition and method¦

li for vaccination against acquired immunodeficiency syndrome (AIDS)¦

,1 and AIDS-related complex (ARC). In particular, the present¦

¦; invention relates to a polyamide resin-synthetic peptide¦

conjugate which ca~ be used in a method of immunization against Il the viral causative agents of AIDS and ARC and in a diagnostic ¦ assay for AIDS and ARC.

i AIDS was first discovered as a severe immune deficiency , which resulted in reports of opportunistic infections occurring~

¦ among male homosexuals ~see Gottlieb, M.S., et al., 305 N. Engl.

J. Med. 1425-1431 ~1981) and Masur, H., et al., 305 N. Engl. J.

Med. 1431-1438 ll98l)). The incidence of this new human disease, named nacquired immunodeficiency syndromen (AIDS), is rapidly ! growing. Although sexual transmission appears to be the primary ¦¦ mode of transfer, a number of cases in which ~he disorder was ;' I
transferred by blood transfusion have been reported (see Gottlieb, et al., supra). The etiologic agent of this disease has been shown to be a human retrovirus, known variously as human I T lymphotropic virus type III (HTLV - III), lymphadenophathy-i ,, 01/MFM2 -2-'~`

31~8 .

associated virus (LAV)(Barre Sinoussi, F., et al., 220 Science 868-871 (1983)), or AIDS-associated retroviru~ (ARV).
Seroepidemiological studies have identi~ied HTLV~ spe_ cific antibodies in the serum of most patie~ts wi~h AIDS or ARC.
The predominant antigens recognized by antibodies in sera ob-tained from AIDS patients and from hemophiliacs arP associated with the envelope glycoproteins. Further, the most immunogenic proteins of the human T lymphotroph:ic viruses, HTLV-I and HTLV-II
are cell surface-expressed glycoproteins (Chen, I.S. et al., 305 Nature (London) 502 (19831).
The envelope (env) gene product of HTLV-III is synthesized as a polyprotein precursor and is subsequently glycosylated with-in infected cells. That glycosylated polyprotein, with an es-¦ timated molecular weight of 160 Kd (gp 160), is processed into¦ an amino terminus subunit gp 120 and a carboxyl transmembrane subunit, gp 41. The gp 41 subunit is one of the predominant ¦ polypeptides in purified virus preparations.
j Antibodies from AIDS and ARC pa~ients contain viral neutra-1l lizing activity; however, infection presumably occurred in those 20 ! patients prior to the development of neutralizing antibody, and whether the ind-lction of neutralizing antibody prior to infection would result in protective immunity is unknown. The general notion with retroviruses is that the antigenic determinants or epitopes associated with the induction of neutralizing antibodies are associa~ed with the glycoprotein envelope ~see Holden, H.T.
and T~ Taniyama, 150 J. Exp. Med. 1367 (1979) and Flyer, D.C. e~
al., 305 Nature (London) 815 (1983)). Until the present in-¦
vention, that association had not been established for the AIDS
associated viruses. As will be described, it has now been demon-strated that the gp 41, gp 120 and gp 160 envelope glycoproteinsare the most immunogenic epitopes in virus-exposed individuals.

.

~ z~3~8B

The present invention is premised upon the assumption that I the critical epitopes involved in the induction of protectivel ¦ virus neutralizing antibody are associated with the two viral' envelope glycoprotein subunits, gp 120 and gp 41; The present invention is also based on the ue~e of polypeptide portions of those immunogenic subunits to induce an immunogenic response to l the intact causative agents of AIDS and ARC when sonjugated to ¦I the solid phase resin on which that portion of the immunogenic l subunit was synthesized. Those polypeptide subunits arel conveniently synthesized on solid phase resins. Solid phase peptide synthesis is a valuable tool for investigating the struc-ture and mechanism of action of proteins and peptides (proteins' l and peptides are collectively referred to herein as "protides~).
¦I Most such synthetic methods involve the use of a cross-linked ¦I polystyrene based resin as the solid phase to which the pro~ide i is anchored during as~embly, usually through a linker molecule.l ¦l Assembly is accomplished by a repetitive cycle of adding a pro-¦
¦ tected amino acid to the solid phase, selectively removing~
i (deprotecting) a protective group on that amino acid, and adding;
I additional suitably protected amino acids (for a review, see¦
I Merrifield, R.B., "Solid-phase Peptide Synthesis", 32 Adv.¦
Enzymology 221 (1969~

! Although cross-linked, polystyrene based resins are most commonly used as suppor~s in solid phase protide synthesis, theirl relatively hydrophobic character in comparison to the polar¦
organic media required to solubilize reactants can be problematicl in protide chain assembly. Such media may freely solvate the , growing protide, yet incompletely swell the polystyrene matrix.
Within the polymer lattice, impaired diffusion of xeagents and steric hindrance can contribute to lowered efficiency during coupling cycles, which, on a repeated basis, lowers final yields ¦01/MFM2 -4-~ j !

~` 12~3~

appreciably. During the early stages of assembly, when the resin to protide mass ratio is high and the physical properties of the support dominate, thi~ lowered efficiency is particularly acute. I
¦ Those shortcomings led to the development of a cross-linked, ¦ polydimethylacrylamide based support which is highly polar in character and is freely permeated by the reguisite solvents for peptide synthesis. Atherton, E., D.L.J. Clive and R.C. Sheppard, "Polyamide Supports For Polypeptide Synthesis n ~ 97 J. Amer. Chem.
Soc. 6584 (1975); ALshady, R., E. Atherton, M.J. Gait, K. LPe and~
10 j R.C. Sheppard, "Easily Prepared Polar Support For Solid Phase Peptide And Oligonucleotide Synthesis". 1979 J.C.S. Chem. Comm.
; 425 (1979~. That polyamide resin, as the amino methyl ,! derivative, can accommodate synthetic schemes incorporating ¦¦ alternate protection strategies through selection of the appropriate linker molecule, which links the C-terminal residue !
I to the support. However, peptides ~ynthesized on that resin must¦
¦¦be separated from the resin after the synthesis is completed andl ¦then purified, both time-consuming steps which decrease the final ¦
llyield of the protide.
20 ¦1 A significant advantage of the composition and method of the present invention is that the protide can be used to induce an immunogenic response in an experimental animal without being separated from the resin on which it was svnthesized and then purified. The resin-protide conjugate thus synthesized can be ¦
used in a num~er of investigative applica~ions. Of particular ¦
interest to the present invention is the use of certain polyamide I
resin-peptide conjugates, specifically, conjuga~es incll~ding¦
portions of the gp 120 and gp 41 subunits of the gp 160 Pnvelope glycoprotein, as immunogens.
301 i It has previously been demonstrated that synthetic peptides analogous to sequlences contained in viral encoded proteins have ¦ 01/MFM2 -5-~Z931~
proven useful for identification of native antigen determinants associated with such proteins. Several laboratories have reported studies on the antigenic activity of various hepatitis B
antigen (HBsAg) synthetic peptides~ Dreesman, ~.R., et al., 295 Nature 158 (1982); Lerner, R.A., et al. 78 Proc. Natl. Acad. Sci.
USA 3403 11981); Prince, A.M., et al., 79 Proc. Natl. Acad. Sci.
USA ~79 (1982). The induction of an antibody response to HBsAg, using such peptides, proved to be relatively weak, but could be enhanced through coupling of pepticles to a carrier protein prior !, to immunization. Lerner, et al., ~E~; Sanchez, Y., et al., 18 ! Intervirology 209 (1982). Further, synthetic peptides corre-¦l sponding in amino acid sequence to portions of the capsid ¦i proteins of tobacco mosaic virus (Anderer, F.A. and H.D.
Schlumberger, 97 Biochim. Biophys. Acta 503-~09 (1965), Id.
Anderer, F.A. and H.D. Schlumberger, 115 Biochim. Biophys. Acta i 222-224 (1966); Fearney, F.J., C.Y. ~eung, J.D. Young and E.
Benjamini, 243 ~iochim. Biophys. Acta 509-514 (1971)), foot and mouth disease virus (Bittle, J.L., et al., 298 Nature 30-33 i (1982) and Pfaff, ~., et al., 1 EMBO J. 869-874 (1982~) and ¦poliovirus (Emini, E.A., B.A. Jameson and E. Wimmer, 304 Nature S99-703 (1983)) have been used to determine neutralizing epitopes associated with the inta~t virion.
Because the prediction of the poten~ial antigenic determinants of immunogenic AIDS and ARC pro~eins based on primary sequence analysis of those immunogens is not exact, the identification of putative epitopes ~hrough trial and error can be laborious. A composition and method which involves the delineation of antigenic sequences native to the viral causative ~ayents of AIDS and ARC with synthetic peptides which does not 30 Irequire purification of the synthetic peptide and coupling of the ¦peptide to carrier proteins offers significant advantages.
I . .

IlOl/MFM2 -6- , . I

~Z93~l8~3 ¦ SUMMARY OF THE INVENTION
¦ An object of the present invention is, therefore, to provide~
a composition of matter capable of inducing an immunogenic response to the viral causative agents of AIDS and~ARC comprising a polyamide resin and a synthetic peptide, the ~ynthetic peptide¦
comprising a chain of amino acids having a 6equence homologous to¦
a portion of the amino acid sequence of the gp 120 or gp 41 enve-¦
lope glycoprotein of ~TLV-III, ARV or LAV and having a hydro-l ¦ philic region therein. To accomplish that object, it was first¦
I necessary to identify a number of synthetic peptide candidates ¦ capable of eliciting such a response. It is, therefore, also an object of the present invention to characterize the amino acid I sequence of the most immunogenic proteins of HTLV-III (i.e., gp ¦ 120 and gp 41), and to identify the structural conformation of those proteins and the portions of the amino acid sequence of those proteins which represent the most likely binding sites for ¦ antibodies to the intact virion, and then synthesize synthetic peptides wi~h that same sequence and structure, or with al sequence and structure which is sufficiently homologous to the¦
¦ portion of the ~equence which represents the binding site as to also be immunogenic.
A further object of the present invention is to provide a polyamide resin, and a method of preparing that polyamide resin, upon which those synthetic peptides are synthesized using solid phase synthetic methods to produce a conjugate which, when injected into an expeximental animal, induces an immunogenic response to the viral causative agents of AIDS and ARC without~
separating the synthetic peptide from the resin before injection into the experimental animal.
It is another object of the present invention to proJide an assay for detecting antibodies against the viral causative agents ~ '.

93~l88 I . I
¦ of AIDS or ARC in those individuals suspected of having AXDS or !
ARC comprising conjugating a polyamide resin-peptide conjugate to the ligand of a specific binding pair wherein that binding pair is comprised of the ligand and an anti-ligand having specific affinity for that ligand and the synthetic peptide is comprised of a chain of amino acids having a sequence homologous to a portion of the gp 120 or gp 41 envelope glycoproteins of HTLV-III, ARV or LAV, contacting that conjugate with sera from an il animal, thereby causing any antibodies to the viral causative 10 ¦¦ agents of AIDS or ARC present in that sera to bind to that~
j conjugate, and then contacting the bound antibodles with the ¦l anti-ligand of the specific binding pair.
! It is another object of the present invention to provide a~
method of immunizing an experimental animal against the viral causative agents of AIDS and ARC comprising synthesizing a~
Ii peptide comprising a chain of amino acids having a sequence homo-¦¦ logous to a portion of the gp 120 or gp 41 envelope glycoprotein I' of HTLV-ITI, ARV or LAV and having a hydrophilic region thereinl on a polyamide resin and administering an immunogenically effec-¦
20 li tive amount of the polyamide resin-peptide con~ugate to an exper-~
¦¦ imental animal.
,.

! BRIEF DESCRIPTION OF THE DRAWINGS
. I
Figure 1 is an artist's rendition of the plot of the hydrophilic averages for each residue against the amino acid sequence of the gp 160 precursor glycoprotein of the gp 120 and gp 41 env glycoproteins of HTLV-III, LAV and ARV generated by a computer program utilizing the Chou-Fasman predictive scheme for ij secondary structure.

! I Figure 2 i~ an actual computer plot of a segment of the 30 ll amino acid sequence of the plot of Fig, 1.

0l/MPM2 -8-~2~3~L81 3 Figure 3 is a schematic representation of the secondaryl structure of the amino acid sequence of the plot of Fig. 1¦
showing the differences between the secondary structure of the gp 160 precursor o HTLV-III, LA~ and ARV.
Figure 4 is a graph of the optical density vs. the reciprocal dilution of the antiserum obtained from rabbits immunized with the gp 120 peptide 503-532 (Peptide ~ on Table II1 ~/~
showing the binding of the Peptide ~ by the rabbit antibodies by en~yme linked immunosorbent assay. Fig. 4A shows binding of~
~ 10 Peptide ~, Fig. 4B shows the bind:ing of a control peptide (seel '~ ~ Example 15). Data frcm anti-peptide antisera from rabbit 1 isi ? represented by a (-) and data from anti~peptide antisera from rabbit 2 is represented by a (0), Data from pre-immune sera from¦
!li each rabbit is represented by a (~) and a (a), respectively. I
Figure 5 illustrates the growth of HTLV-III virus on the¦
¦ A3.01 cell line as determined by assaying reverse transcriptase ! ~RT) activity (uptake of 3H-TTP: see Example 16). Fig. 5A shows viral replication at a lO dilution, Fig. 5B was at a 10 . ¦ dilution, Fig. 5C at 10 3, and Fig. 5D at 10 4.

2 0 3~ETAILED DESCRIPTION OF THE INVENTION
_ .
As noted above, the present invention is based in part upon the assumption that it is the gp 120 and gp 41 subunits of the envelope glycoprotein which are the most immunogenic epitopes ofl the viral causative agents of AIDS and A~C. As will be descri-¦
bed, the accuracy of that assumption has now been verifiedO It was next necessary to determine the sequence of the amino acids of the gp 120 and gp 41 subunits and ~o select the portions of~
those envelope glycoprotein subuni~s which represent the most likely antibody-binding sites. This selection was accomplished I

: II 01/MFM2 -9-iL293188 by means of computer modeling of the structure of the gp 120 and !
gp 41 subunits.
Once the most likely sites were identified, chains of amino¦
acids were synthesized on a polyamide resin to duplicate thel amino acid sequence at each of those sites. The usual method of¦
coupling a synthetic peptide to polystyrene based resin~ i8 ¦
through a benzyl ester derivative, and separation of the peptide~
from the resin is usually accomplished by ither acidic or hasic¦
¦ cleavage. Benzyl esters are s~sceptible to several such methodsl ll of cleavage, but are also stable throughout the multiple¦
I deprotection, neutralization and coupling reactions which arel ¦I characteristic of solid phase synthetic methods. ~ydrazine has¦
Il also been used to separa~e the protida from the resin ~Kessler, ¦j W. and B. Iselin, 49 Helv. Chim. Acta 1330 (1966~) as have ~ari-I! ous ~mmonolytic (Manning, M., 90 J.Am.Chem.Soc. 1348 (1968)~ and¦
¦ other methods. ~owever, ~hose methods all require that appropri- !
¦l ate steps be taken to avoid damage to th~ peptide followed by ¦I purification of the peptide from the byproducts of the syn-¦l theseis, including amino acids, short peptides, decomposition ¦
¦! products of the resin, and sometimes, peptides containing incom I
pletely removed protecting groups. Although purification can¦
sometimes be accomplished by a dixect crystallization, in synthesis in which the contaminating peptides are of approximately ~he same size and composition as the desired product, more selective techniques must be employed. RegardleSs I
of the method of separation and purification, those requirements add time-consuming steps ~o the synthesis and often lower the total yield of the cynthetic peptide. The polyamide resin and method of the present inventivn requiras n~ such separation and purification, thlereby decreasing the amount of time required to accomplish the synthesis and raising the peptide yield.

ll 1~ I
The polyamide resin of the polyamide resin-peptide conjugate of the present invention is prepared by cross-linking a commercially available dimethylacrylamide monomer in aqueous solution using a diaminoalkane, preferably a diaminoalkane having alkenoyl groups at either end of the molecule, such as N,N'-bis-alkenoyl-diaminoalkane. In a presently preferred embodiment, the cross-linker is either N,N'-bisacrylyl-1,3-diaminopropane or N,N'-bisacrylyl-1,3-diaminobutane prepared according to the method of Halpern and Sparrow (J.A. Halpern and J.T. Sparrow, "An Improved Procedure For the Synthesis of N,N'-bisacrylyldiaminoalkanes", 10 Synthetic Comm. 569 (1980)), hereby incorporated in its totality by this specific reference thereto. The ~lse of the propane analog is preferred because it yields a polymer of larger pore size and improved swelling properties during protide synthesis than the polymer obtained by use of the ethyl analog. However, it will be understood by those skilled in the art who have the benefit of this disclosure that I the other diaminoalkanes listed in that report, N~N'-bisacrylyl-¦l 1,2-diaminoethane and N,N'-bisacrylyl-1,6-diaminohexane, as well~
20~ as other diaminoalkanes, are also appropriate for use in the preparation of the resin of the present invention.
~l A functional monomer is included in the cross-linked resin.
The term "functional monomer" refers to those alkenyl amines which are used to anchor the C-terminal amino acid of a ~ynthetic~
peptide to the re~in. The functional monomer, when protectedl with the methylsulfonylethyloxycarbonyl ~MSC) group (see Tesser,¦
G.I. and I.C. Balvert-Geers, "The Methylsulfonylethyloxycarbonylj Group~ A New And Yersatile Amino Protective Function", 7 In~O J. !
IPeptide Protein Res. 295 ~1975)), is referred to as an MSC
30¦1alkenyl amine. Those functional monomers are prepared by Ireaction of the commercially available chloxide derivative with I .
!

¦1 01/MFM2 i l l l ~93~8~ 1 the alkenylamine, and the MSC protective group is subsequently removed with base. However, the MSC group is not required. The¦
polyamide resin is also prepared by simply adding an excess of the allylamine, followed by filtering or other method to remove the resulting fines. The amount of functional monomer added is selected to yield a resin substitution of between about 0.1 mmol and about 0.5 mmol per gram of resin, and preferably in the rangel ¦ of about 0.2 mmol to about 0.4 mmol per gram of resin. The ini-¦
¦ tiator can be any of the initiators known to those skilled in the !
¦~ art such as a persulfate or ribof'lavin, and is preferably ammo-¦
¦ nium persulfate. I

!. Because the above-described substances are combined in aque-¦
ous solution, they are collectively referred to as "the aqueous !
phase". The next step in the preparation of the polyamide resin¦
of the polyamide resin-synthetic peptide conjugate of the presentl vention is to combine the aqueous phase with an organic phase.¦
¦ The term "organic phase" refers to an organic solvent which, when ¦
¦I combined with the aqueous phase and stixred, results in a suspen-¦
I! sion from which the resin is obtained. In a presently preferred ¦! embodiment, the organic phase comprises a mixture of hexane and carbon tetrachloride.
¦ An emulsifier is added during the stirring to allow for the ¦
¦ formation of beads of uniform size. The emul ifier can be any ¦
detergent known to those skilled in the art, and in a presently preferred embodiment, is either sorbitan sesquiolea~e, sorbitan monolaurate or sorbitan monodecanoate. The amount o~ detergent added is ad~usted to give a spherical resin of approximately uni-¦, form si~e. A decrease in the amount of detergent results in an l¦emulsion which yields increased amoun~s of larger, amorphous 30 1l material, which could contribute to a reduction to the internal ¦~growing chains of amino acids. An increase in the amount of Il .

¦ 01~MFM2 -12-1 ~ , ~1 ` 1~93il 8~ 1 deteryent increases the amount of fine material, which is diffi-¦
cult to remove without the loss of significant amounts of the resin. Those fines clog thP reaction vessels of the peptide synthesizer as well as ~he associated lines and ~alves.
A promoter is then added to promote the polymerization of the monomers in the suspension, resulting in the formation of beads of the polyamide resin of the present invention. A number of promotors are known to those skilled i~ the art, but parti-¦
cular success in preparing the polyamide resin has been obtained with N,N,N',N'-tetramethylethylenediamine (TEMED). The resulting beads are then filtered and washed, the MSC group (if present) is removed with base, and the beads are dried. The beads may then be sifted through a mesh sieve to insure relatively uniform size.
Overall yields using the method of the present invention ranged from about 87% to about 94~ from starting monomers.
~ he resulting aminomethyl, cross-linked polydimethylacryl-amide resin, when conjugated to the synthetic peptide, provides maximum exposure of the peptide in an aqueous solution, and the¦
resin-polymer backbone does not restrict the peptide conforma-tionally. The exposure o~ the peptide is the result of the abil-ity of the polyamide resin to swell to many times its dry bed ¦ volume when highly solva~ed by water.
The synthetic peptides are synthesiz~d on the beads by¦
coupling to a linker which is attached to ~he resin with an acti- 11 vator. The term "linker" refers to a linking group which links¦
the carboxyl group of the first i~mino acid of the synthetic pep-tide to the polymeric resin. In the presently preferred embodi-ment, this linker is an oxyalkyl benzoic acid (OBA) to which an amino acid residue is coupled to serve as the ~irst amino acid in 30 the peptide chainv Because the OBA linker is used to attach the l C-terminal amino acid to the polyamide resin, anhydrous hydrogen ~1 .

fluoride can be used to remove ~ e~chain protecting groups from the peptide without significant loss of the syn-thesized peptide from the resin. In the below-described examples, -the amino acid of choice is glycine, which is protected with the t-butyloxycarbonyl (t-BOC) protecting group, but it will be understood by those skilled in the art who have the benefit of this disclosure that the amino acid could be any amino acid, particularly, the amino acid which is the first amino acid in the peptide to be syn-thesized, and -that other protecting groups are equally suitable. The glycine residue serves the additional function of a spacer between the peptide and the resin-polymer backbone.
The BOC-glycyl-4-(oxymethyl) benzoic acid which is the pres-ently preferred linker was prepared by a modification of the method described by Mitchell, et al. (Mitchell, A.R.,, S.B.H.
Kent, M. Engelhard and R.B. Merrifield, "A New Synthetic Route to tert-butyloxycarbonylaminoacyl-4-(oxymethyl) phenylacetamido-methyl-resin, An Improved Support of Solid-phase Peptide Synthe-sis", 43 J. Org. Chem. 2845 (1978)). An important modification of the Mitchell, et al., method is the elimination of the use of dimethylformamide as a solvent. That solvent is difficult to evaporate, consequently, even though evaporation can be hastened by raising the temperature, the method is still time-consuming.
The activator used to couple the linker to the polyamide resin prepared as described above is diisopropyl carbodiimide and 4-dimethylaminopyridine, but it will be understood by those skilled in the art that other activators such as dicyclohexylcarbodiimide and 4-methylpyrrolindinopyridine are equally suitable for such a purpose.

;

, " ~

~L~293~L8~ 1 After synthesis of the peptide on the polyamide resin, the polyamide resin-protide conjugates of the present invention are used for a number of purposes, including in vitro assays for thel presence of antibodies to HTLV-III, ARV or LAV, inducing an¦
immunogenic response to the viral causative agents of AIDS or ARC
in experimental animals, or mapping antigenic determinants on the viral causative agents of AIDS or ARC. For instance, an in vitro assay is conducted by crushing a beaded polyamide resin-synthetic peptide conjugate with a mortar and pestle and abqorbing the crushed conjugate onto a solid phase such as a microtiter test plate with neutral pH buffer. Serum or other body fluid suspected of containing an antibody against the viral causativel agents of AIDS or ARC is then incubated with the absorbed¦
conjugate, unbound antibodies are removed by washing, and the bound antibodies are detected by enzyme linked immunosorbent assay, biotin-avidin amplified assay or other detection methods such as are known in the art.
The polyamide resin-synthetic peptide conjugate can also be used to map antigenic determinants on the viral causative agents of AIDS or ARC by simply removing a portion of the polyamide resin at intervals during the synthesis of the peptide, depro~l tecting the peptide, and testing each removed portion in serial ¦
fashion to determine that point in the synthesis at which the¦
peptide binds to an antibody specific for the viral causative agents of AIDS and ARC. This method is made possible by the elimination of the separation and purification steps requir~d in other synthetic methods. The conjugate can al~o be tested for its ability to bind an~ibody by crushing and absorbing to a solid I
support such as a microtiter test plate and assayed as described, -above. Separation of the peptide from the resin and purification of the peptide is not required for such an assay.

~LZ931~

The polyamide resin-protide conjugates are also useful as an immunogen against the viral causative agents of AIDS or ARC. The conjugate is used directly for immunization of experimental animals with or without an adjuvant. The te~m Nexperimental ani-mal", as used herein, refers to any animal capable of an immune response. The experimental animals of primary interest are mam-mals, but an immunogenic response can be induced in other experi-I mental animals such as birds using the method of the present ¦l invention. For instance, an immune response specific for the 10 1I viral causative agents of AIDS and ARC, as measured by radio-I! immunoprecipitation, was induced by immunization of rabbits usingl ¦ a conjugate comprised of a synthetic peptide with a sequence cor-¦
responding to the protein coat of the HTLV-III virus and the¦
polyamide resin.
Those synthetic peptides which induce rabbit antibodies I which bind to AIDS virus were then tested for their ability to ¦¦ bind human anti-HTLV;III antibody, and the rabbit anti-HTLV-III
antibodies were also tested to determine whether they were Icapable of neutralizing the infective virus in tissue culture.
20 I Once the most immunogenic synthetic peptides which fulfill those criteria were identified, they are used for both a vaccine and as a diagnostic assay to identify individuals exposed to the viral causative agents of AIDS and ARC as well as AIDS and ARC
patients~ j The amino acid sequence of the gp 1~0 and gp 41 subunits was I
determined by prediction based upon the nucleo~ide sequence of HTLV-III and the verification of those sequences by analysis of ¦the sequence of the NH2~terminus by Edman degradation of ~he pro- ~
teins labeled with 3[H] leucine and 35~S] cystine, as well as i tH] valine.

ll01/MFM2 -16-11 ~

~ 318~ 1 ¦ Verification of the immunogenic nature of the gp 120 and gp 41 envelope glycoproteins (and their precursor, gp 160) was ob-¦
tained by screening serum samples rom AIDS and ARC patients to¦
identify those with antibodies against HTL~-III by indirect cell membrane immunofluorescence (MIF~ using the H9/HTLV-III cell line and by radioimmunoprecipitation and sodium dodecylsulfate-poly-acryl~mide gel electrophoresis (RIP/SDS-PAGE) with 5[S] cystine-¦ labeled H9/HTLV-III cells. Representative antibody-positive sera I were also tested on glycoprotein preparations of H9/HTLV-III
10 ¦ cells enriched through the use of a lentil lectin column. The I results indicated that the highest percentage oP antibody-posi-¦~ tive sera contained antibodies which recognized gp 120 and gp 160 ¦1 and that all of the samples which contained antibodies to otherepitopes also contained antibodies which recognized gp 120 and gp 160.
Selection of the most immunogenic sites on the gp 120, gp 41 and gp 160 envelope glycoproteins was accomplished by modifying a ¦ computer program based on the hydrophilicity index described by , ~opp, T.P. and K.R. Woods (78 Proc. Nat'l Acad. Sci. USA 3824-20 1 3828 (1981)) to predict the location of the hydrophilic regionsassociated with the HTLV-III envelope gp 160 glycoprotein from HTLV-III, LAV and ARV. The amino acid sequence of those glyco-proteins was also analyzed for secondary struc~ure using ~he¦
Chou-Fasman predictive scheme (Chou, P.Y. and E.D. Fasman, 13 Biochemistry 222 (1974)). The peak hydrophilic areas were com-pared with the predicted ~econdary structure, and those areas most likely to be exposed on the surface of the glycoprotein were ¦ identified. Those areas were also examined for the presence of a B turn because previous studies using viral envelope proteins had 30 lindicated that the hydrophilic regions exposed on the surface ¦~with predicted ,g turn secondary structure represent i= unogenic ¦1 01/MFM2 -17-1, , I.i, 1, Il 1293~
surface regions on the virus (Dreesman, G.R., et al., 295 Nature !
(~ondon) 158-160 (1982)(hepatitis B surface antigen); Hopp and Woods, supra (hepatitis B surface antigen~: Henderson, L.E., et al., ~56 J. 8iol. Chem. 8400-8406 (1981)(Raucher murine leukemia virus); Gingeras, T.R., et al., 257 J. Biol. Chem. 13475-13~91 (1983)(adenovirus spike protein); Watson, R.J., et al., 218 Science 381-384 (1982)(herpes simplex virus envelope glycoprotein D)).
Having identified the sequence of the gp 160 glycoprotein' from HTLV-III, LAV and ARV and the regions in that sequence which are likely to be immunogenic, the next step was to synthesize a polypeptide with the same amino acid sequence (or a sequence which is similar enough so as to be treated in the same manner by the antibody which binds with that epitope) as that region of the¦
glycoprotein. The synthesis was carried out by the solid-phase¦
methodology described above. A total of ~ synthetic peptides~
were synthesized, each selected on the basis of the above-described tests for predicted immunogenicity. The amino acidl sequences of each of those synthetic peptides is given in Table¦
20 j II.

The DL~nC synthetic peptidas are then used to induce an im-mune response in rabbits by injecting the rabbits with the poly-j amide resin-synthetic peptide conjugate. The rabbits were also, injected with synthetic peptide separated from the resin on which¦
it was synthe~ized and then coupled to a carrier. The antibody¦
titer of the rabbit sera was tested by the ability of the anti-¦
body to bind with the peptide con~uga~ed to bovine serum albumin (BSA). Those results were confirmed by oonducting inhibition studies in which the înhibition of the binding of the rabbit anti_peptide to the peptide-BSA was measured.

' ~Z931~
i ¦ The rabbit anti-peptide antibodies wexe then examined for their ability to recognize the native proteins associated with HTLV-III. An HTLV-III infected T-cell line labelled with 35[S]-cystine was used for immunoprecipitation to determine whether the¦
anti-peptide sera would bind any radioactively labelled HTLV-III¦
native proteins. Autoradiography with SDS-PAGE confirmed that¦
the rabbit anti-peptide antibodies specifically precipitated a !
single protein which corresponded to the gp 160 precursor enve-lope glycoprotein gp 160 of HTLV-III. The precursor gp 160 pro-duct is cleaved to yield the major gp 120 envelope glycoprotein i and gp 41, the transmembrane glycoprotein. The gp 41 envelope ¦¦ subunit does not radioactively label to the same degree with 35[S]-cystine as the amino end of the precursor gp 160 glycoprotein, and was not detected by immunoprecipitation. How-ever, when when 35[S]-methionine was used as a label, the binding ¦ was detected by immunoprecipitation~ a result which has been con-¦l firmed using Western transfer methods.
¦¦ The anti-peptide antibodies thus generated were then tested Il to determine whether they were capable of neutralizing the viral 20 ¦I causative agents of AIDS or ARC. That determination can be ma~e in a number of ways. In one method, the polyamide resin-peptide ¦ conjugate is crushed with a mortar and pestle~ and a suspension of the resin is made in buffered saline. That emulsion of peptide is absorbed to the solid phase of microtiter plates, and nonspecific ~ites are blocked with 10% normal goat serum. The binding of rabbit antibodies to the peptide is detected by using biotin-goat ~ntibody to rabbit IgG and avidin conjugated horseradish peroxidase~ Peroxidase activi~y is determined usingl 1,2'-azino-di(3-ethyl-benzthiaæoline-sulfonic acid) and H2O2 asi ~he substrate. A resin bound peptide corresponding to a hepatitis B surface antigen sequence serves as a control. The 1, , ¦1 Ol/MFM2 -l9-binding of -the rabbit anti-peptide is quantified spec-tropho-to-metrically at 410 nm with a plate reader.
In a second method, the neutralizing ability of the anti-peptide antibodies was tested by incubating purified virus and rabbit anti-pep-tide antibodies with infected T-helper cell lines, then examining the lysed cells by Western transfer and immunoprecipitation for the presence of the virus. In a third method, the neutralizing ability of the rabbit antibodies to the polyamide resin-synthetic peptide conjugate was assessed by measuring the reduc-tion of reverse transcriptase (Rl') activity. The results were verified by radioimmuno-precipitation.
The most immunogenic synthetic peptides is then used in a diagnostic assay for AIDS and ARC and as a vaccine. When used in a diagnostic assay, the preferred method involves the detection of antibody against the viral causative agent of AIDS
and/or ARC. That assay is conducted, for instance, by coating an insoluble matrix such as a column of polystyrene beads or micro-well test plate with a synthetic peptide or a synthetic peptide coupled to a carrier protein (i.e., bovine serum albumin) containing the amino acid sequence associated with the epitope(s) of one of the viral causative agents of AIDS or ARC.
Alternatively, the insoluble matrix is coated with a number of different synthetic peptides (a "cocktail") containing the amino acid sequence of several epitopes. Alternatively, the polyamide resin-synthetic peptide conjugate is crushed with mortar and pestle and absorbed onto a solid phase as described above.
A sample of biological fluid from the suspected patient is incubated with the synthetic peptide-coated matrix to immunocapture the predetermined antibody. The resultant matrix, separated from the uncaptured sample, is then incuba-ted with a quantity of biotin-labeled antibody directed -to the species of ~.;, j , I` ~f~?'33~LS~
¦ the predetermined antibody ~e.g., anti-human antibodies would be the predetermined antibody if the body fluid is taken from a human patient) sufficient to bind a measurable number of human antibodies, if present. The resultant matrix, ~eparated from uncaptured biotin-labeled antibody and the matrix, is ~hen incu-bated with a quantity of labeled avidin, preferably avidin labeled with an enzyme such as alkaline phosphatase, sufficient to bind a measurable number of antibodie3, if present. The 1I resultant matrix is separated from uncaptured avidin and a label lO 1~ detected and/or preferably quantified by adding the substrate li which is specific for that enzyme to thereby determine indirectly i the presence of antibody to AIDS virus in the sample. The anti-body could also be labeled with an enzyme directly, in which case1 1 the matrix is incubated with an enzyme-reactive substrate, and¦
¦ the change in the substrate, e.g., a color change or fluorescence~
I emission is detected. Regardless of whether the label is an ¦¦ antibody, an enzyme or an enzyme labeled with biotin-avidin, the i binding pair formed by ~he antigen and antibody or the enzyme and 1l substrate will be referred to as the "ligand" and "anti-ligand"
20 11 of the specific binding pair.
i A diagnostic assay can also be conducted for detection of the antigen rather than the predetermined antibody. To conduct an antigen test, the solid phase matrix is coated with antibodies to the viral causative agents of AIDS, i.e. r the antibodies pro-duced by immunization with a synthetic peptide or polyamide resin-~ynthetic peptide ~onjugate (or, preferably, several pep-~
tides~ such as the peptides of the present inven~ion. The sample~
of biological flllid from a patien~ suspected of having been infected with the AIDS virus is then added to the matrix, fol-¦ lowed by the addition of biotin-labeled antibody, where the anti-body is an antibody which binds to the AIDS virus produced in the ~1/MF~2 -21-11 3L~93~L8~
Il ~
, same way as discussed above. The avidin labeled enzyme is then added, followed by the substrate specific for the enzyme, and the color change or fluorescence emission is detected. Either of these assays is also conducted as an inhibition assay where, instead of adding biotin-labeled antibody to the AIDS virus to the bound antigen, a biotin-synthetic peptide or biotin-polyamide resin-synth~ic peptide conjugate is added.
To use the polyamide resin-~3ynthetic peptide conjugate fl I the present invention as a vaccine against the viral causative 10 1 agents of AIDS, approximately 100 to 1000 micrograms of synthetic Il peptide, or several synthetic peptides, prepared as a polyamide ii resin-synthetic peptide conjugate according to the teachings of~
the present invention, is administered to an individual with an¦
li adjuvant. The synthetic peptides were also administered, after , separation from the resin on which they were synthesized, coupled I to a carrier, Appropriate carriers include the toxoid ; 1~ components, any one of several large protein-containing~
¦ sub~tances which are foreign to the animal to be injected, any of ¦l se~eral small peptide preparations which have demonstratedl 1l adjuvant activity and which behave as a carrier, or liposomes. i ! I The toxoid components can be tetanus toxoid or diptheria toxoid.¦
The phrase 'llarge protein-containing substances which are foreign¦
to the animal to be in jected", refers to such substances as Keyhole limpet hemocyanin (RL~ or BSA. The small peptide prepa-rations with demonstrated adjuvant activity which also act as a carrier include muramyldipeptide, murabutidine, and the polyamino acids such as poly-~-glutamic acid or poly-L-lysine. Appxoxi-mately 10 to 100 molecules of synthetic peptide are complexed to ¦each molecule of carrier using a heterobifunctional cross-linker !such as m-maleim:idobenzyl-N-hydroxysuccinimide ester (MBS) (Liu, F.T., et al., 18 Biochemistry 690 (1979), Green, N. et al., 28 ¦lOl/MFM2 -22-~ l Cell 477 (1982)), gl~taraldehyde, a carbodiimide, succinyl anhyd-ride or N-succinimidyl-3-[2-pyridyldithio]-propionate.
Suitable adjuvants include al~ (aluminum hydroxide) and anyl of a number of additional adjuvants such as are known to those¦
skilled in the art. Both the polyamide resin-peptide conjugate and the carrier-synthetic peptide complex are administered in a pharmaceutically acceptable diluent such as distilled water, phosphate buffered saline, citrat:e buffer or any neutral pH
I buffer, i.e. a buffer with a pH of between about 6 and about 8.
¦ The polyamide resin-synthetic peptide conjugate of the pre-sent invention is also used to screen putative vaccine candidates against AIDS and/or ARC. Such screening is best conducted byl coating an insoluble matrix with crushed beads of the polyamide¦
¦ resin-synthetic peptide conjugate as described above. The i vaccine candidate is then incubated with antibodies against the peptide (with or without biotin) such as a 1:1000 dilution of ¦ Ig~-rabbit anti-peptide-biotin antibody. If biotin labeled anti-~¦ body is used, the avidin-enzyme conjugate is added ~if no biotin l! is used, add biotin-labeled anti-species (such as biotin-labeled 20 ¦ goat anti-rabbit IgG) antibody, then add avidin-enzymel, the substrate is th~n added and the reaction detected.
The polyamide resin-synthetic peptide conjugate of the pre-sent invention is also used to serotype viral isolates from AIDS¦
or ARC patients. Serotyping is conducted in the same manner as described above for screening vaccine candidates, because in both !
cases, the anti~peptide antibody must bind with the intact AIDS
viral causative ayent, ~owever, in the case of the sero~yping of ¦the viral isolate, a portion of the isolate is added, in serial Ifashion~ to a nl~nber of bound anti-peptide antibodies, each antibody being specific for a different polyamide resin-synthetic

3:~8~3 peptide conjugate and having been bound to a separate insoluble matrix.
The present invention may be better understood by reference to the following non-limiting examples.

Example 1. Maintenance and Radioactive Labeling of HTLV-III Infected Cells Two HTLV III producing cell lines, H-9 and MOLT-3, werel ~ grown in RPMI-1640 supplemented with 20% fetal bovine serum, 2 mMI
¦I glutamine, non-essential amino acids and 0.1% NaHCO3 (maintenance¦
ll medium). Cell cultures were labeled by transferring cells from¦
maintenance medium to cystine and glucose deficient medium for 1 hour before adding 35[S]-cystine (lS0 ~Ci/ml) and 3[H]-glucosamine 120 ~Ci/ml for 24 hr). Cells were separated from tissue culture supernatants by low speed centrifugation (1,000 x g for 10 minutes).

¦ Example 2. Verification of Immunogenicity of ~p 120 and ~p 41 Subunits of HTLV-III
¦I Serum samples taken ~rom subjects who came to a community ¦¦ health clinic in a high-risk area for AIDS and ARC and to hos~¦
~I pitals in that area during 1983 and 1984 were screened for anti-ij bodies to HTLV-III by indirect cell membrane immunofluorescence ¦ IMIF) using ~he H9/XTLV-III cell line as described by Essex, et ¦ al., 320 Science 859 (1933). Briefly, this method involves sep-arating the cells from the media as de~cribed in Example 1,l above, washing between 1x106 and 2X106 cells twice with phosphate¦
buffered saline ~PBS~, and exposing them ~o 40 ~l of a 1:4 dilu-i tion of previously centrifuged serum for 30 minutes at 37OC.¦
Each pr~paration was then washed twice with PBS and reacted with 40 ~l of a 1:20 dilution of fluorescein conjugated F(abl)2 frag-~

ment of goat antiserum to human immunoglobulins iIgA + IgG + IgM) ~Cappel, Cochranville, Pa.). The samples were again incubated at 33~l88 37C for 30 minutes, washed twice with PBS, and examined by flu-orescence microscopy. If at least 50 percent (or 40 percent whenl indicated) of the cells showed specific fluorescence, the serum' samples were judged positive. Samples were coded-and read in al double blind manner, and positive and negative human serum sam- !
ples were included as a reference. The results of this screening are presented in Table I.
l I
, _ _, TABLE I

j~ Number (and percent) lO j positive for Number 1 CategoryTested HTLV-III-MA gpl202 i AIDS 50 48(96) 49(98) l ARC 50 43(86) 46(92) i SEXUAL MALES 73 34(47) 36(49) jI HEALTHY LABORA 27 , TORY WORKERS

1 Assay for HTLV-III membrane antigens tHTLV-llI-MA) conducted I by MIF as described by Essex, et al., 220 Science 859 (1983).

j 2 Assay for gp 120 envelope glycoprotein of HTLV-III conducted ;~ by RIP/SDS-PAGE as described by Essex, et al., 220 Science 859 (19~3).

A`ll of the samples from the same 190 individuals were also tested by radioimmunoprecipitation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (RIP/SDS-PAGE) with 35[5]

cystine-labeled H9/HTLV-III and uninfected H9 cells. Briefly,¦
that method is as follows. After disruption of the labeled cells¦
with RIPA buffer (0.15 M NaCl~ 0.05 M tris-HCl, pH 7.2, 1~ Triton X-100, 1% sodium deoxycholate, and 0.1% SDS~, the cells were ¦centrifuged at 100,000 x g for one hour. The lysate supernatant was cleared once with 10 ~l of reference negative control serum Il 01/MFM2 -~5 * Tr~e-mark 3~l8~3 ¦ bound to Protein A-Sepharose CL-4B (Protein A beads) before portions were reacted with 10 ~1 of the human test sera.
¦ Immunoprecipitates were eluted in a fiample buffer (0.1 M
Cleland's reagent, 2% 5DS, 0.08 M tris-BCl, pH 6.8; 10~ glycerol, and 0.2% bromophenol blue) by boiling at 100C for two minutes~
Samples were analyzed in a 12.5~ acrylarnide resolving gel with¦
1 3.5% stacking gel according to the discontinuous buffer system of¦
I, Laemmli (227 Nature (London) 680 (1970)). Surface-labeling was I carried out by lactoperoxidase-catalyzed radioiodination. The I results are presented in Table IX. I
Representative antibody-positive sera were also tested on¦
,' glycoprotein preparations of H9/HTLV-III cells enriched through 1~ the use of a lentil lectin column. HTLV-III glycoproteins were~
I, incubated with lentil lectin Sepharose 4B for four hours and then il eluted with 0.2 M methyl mannoside. The resulting proteins were then immunoprecipitated with HTLV-III reference serum, and the ji precipitates bound to protein A-Sepharose were dissociated from I antibody by boiling for two minutes in the presence of 0.1~ SDS !
1 and 0.15 M sodium citrate pH 5.5. Equal portions werP then in-20 1,1 cubated for three hours at 37~C in the presence or absence of0.25 ~g of endoglycosidase H. The reaction was terminated by the j addition of five volumes of cold 95% ethanol, and the proteins were precipitated overnight at -20C. The samples were then cen I trifuged at 12000 x g or 15 minutes and the proteins were recon-¦
¦ stituted with electrophoresis sample buffer, boiled for three minutes, a~d subjected to electrophoresis. Samples from four antibody-positive ~IDS patients precipitated proteins of about 120 kD, 160 kD and 41 kD. Similar results were obtained with two ~i antibody-positive ARC patients, and with two antibody-positive j healthy homosexual males~ No proteins of related sizes were de-Il tected in sera from antibody-negative healthy homosexual males or I' .

"' i~ 01/MFM2 -26-* Trade-mark lZ~3318~

l 2 3 ~ 5 6 7 ~ g 10 Re~idue 341- 304- 728- 735-~6- 503- 733- 55- 650- ~65-Nos. 370 327 752 752860 532 756 81 671 i90 s~r thr leu ~p ~ l gly al~ hi~U gly arg ~rg pro ~rg llem ~l pro thr ~erV glyx al~ pro 11~ pro arg pro Q~p thr leu asn lySn ~n pro gluX hla thr Arg ~hr ileW ~rY
trp a~n arg gly ilen ly~ pro lcu glu ann ~nb ~ gly lle pro a1A arg~ phe3 glu asnZ
~sn thr pro glu ~rg ly8 gly ~er aer glu~
~hr ~rg Q8p glu ~rg arg lle ~la gln er c ly~ arg glul lle arg glu 3er ~-n glu lys ~er pro~ gly arg val glu a8p gln lle glu lleh glu gly gln val ql~l hlnt gln phe S~- lled ~rg gly glu gly gln gly lya glu ~rg $~ence Aspe llel lle arg leu arg gly Al~ ly8 pro oer gln1 glu nsp glu glu glu tyr asn gly ly8~rg glu arg arg ly8 srg a~p glu gly leu ~ly glul asp ~rg~p thr gln gly &rg pro gly Arg~la argr glu glu a8p glu gly gly ~er val arg v~l leu met gln ~rg gl~ glyO ~er hl~ leu ~rg phe ~la ar( i~e lle asn glu asp I gly phe; ~9E gly ~rg val leu a~n a8n v~l ~rg al~ leu trp asp trp asn thr a~p leuP ~ la arg ly8 ile ~rg phe thr ~er thr ~er leu his glu il~ gly ~la leu ilef phe cy~
phe leu lysg qly : ll gln ~la a gln in ~RV
b ala in L~V
c glu in ARV
d val ln ARVs la ln LAV
e ly~ ~ n MV
~ v~l ln ARV
g ~n ~D M ~
h tyr ln ARV
i omitted ~n ~RV
j his ln ARV
~ a~p in ~RV
: 1 o~n b~ ~p in ~TLV-III
m leu in ARV
n b~a in ARV
o insert val ~etween ll~ and gly in ARV
P ~et in M V
q c~n be tslu ln ~LV~
r ~ctu~l ~Qc~ence of thiB peptide in ~TLV-III, ARV and L~V include~
a3p bet~een arg ~nd ~rg ~ CyB in 3~LV-III
t 8rg ln ARV
u tyr in ARV
thr ln ARV
leu ln A~V
x ne~t t~r~e r~ ue~ ~b~ent ln ~AV; ~ub3tltut~ ehr ~r asn in ARY
y a~ln in ~RV
~p ln ~RV
gly ln L~V
~b val ~ ~RV

. ~
I~

01/MFM2 -27- ' !. i L2~3~
¦I with sera from apparently health~ laboratory workers. None of the human serum samples tested contained antibodies to other ¦ epitopes on the HTLV~III virus without also containing readily detectible antibodies to at least gp 120 and gp 160.

Example 3. Selection oi Immunogenic Sites on gp 120, gp 41_and ~p 160 Envelope Glycoproteins The predicted amino acid sequences of the gp 160 precursor glycoprotein from the three viral isolates HTLV-III, LAV and ARV
were run through a computer program which utilizes the parameters I and hydrophilic values arrived at by Hopp, T.P., and K.R. Woods (20 Mol. Immunol. 483-489 (1983)). The computer program was written in Apple BASIC. The program was written with the ability ~ to save the amino acid sequence to disk in a format which is com-i patible with the Chou-Fasman predictive scheme (Chou, P.Y. and E.
D. Fasman, 13 Biochemistry 222 (1974)). The hydrophilicity pro-gram calculates the hydrophilic averages over a hexapeptide ! length, thereby increasing the accuracy of the predictions.
Since there are no hydrophilic values for Asx or Glx, the amide Il form of the acidic amino acid residues, those codes must be ; 20 ¦1 edited out before running the calculations. The plots of the¦
hydrophilic averages per residue against the amino acid sequence ¦ number for the three AIDS viral glycoproteins are shown in Figure 1. Fig. 1 is actually an artistls rendi~ion of the computer gra-¦phical output ~f the hydrophilicity plots from the three viral;
causative agents of AIDS/ARC which have been characterized. The ¦
highest peak (most hydrophilic) is ~hown in a similar area for all three sequences, with the maximum hydrophilic index occuring~
at residues 739, 744, and 738 for ~TLV~III, LAV and ARV respec-tively. The second highest hydrophilic region centers around the ~
amino acid residues 653-659 just to the amino terminal side of i peak 1. ~he third highest hydrophilic region was found to be in ¦close proximity to peak 1, centered around amino acid residues 733-739 for each of the three glycoproteins.

I I

~293~
An actual computer graph output of a segment of the HTLV~
I sequence is depicted in Fig. 2 Due to the length of the entirel HTLV-III sequence, only a segment iB shown. A proline residue isl shown graphically as a ~P". Two or more aromatic amino acids inl a row within the sequence are depicted as an ~On. The presence¦
of aromatic amino acids within a g:iven se~uence is indicative of regions that possess a high degree of potential for hydrogen bonding. Thus, hydrogen bonds may act to influence the overall I confirmation of the protein. These data indicate that these, i regions are likely to be expcsed on the surface of the glycopro-¦
¦¦ tein.
The predicted secondary structure of the HTLV-III glycoprot-ein, as determined by the Chou-Fasman predictive scheme, is de-l ¦ picted in Fig. 3. The major differences in predicted secondary¦
I structure between HTLV-III, LAV and ARV are shown in boxed¦
I regions. These regions include residues 127-150, 127-155, and 126-148 for HTLV-III, LAV and ARV respectively, where the residue homology is only about 40%, causing changes in ~ turn poten-I tional. In addition, significant differences were noted at I regions 319-330 and 398-408 of ARV, 323-333 and 401~415 of LAV, i and 318-328 and 396-411 of HTLV-III. Comparison of hydrophili-~city with secondary structure indicates that peak 1 contains four potential ~ turns within the region, making ~he region c~ntered around amino acids 739-744 a prime candidate as a potential~
antigenic determinant(s). ~ydrophilic peak 2 also possessed a predicted ~ turn, suggesting that ~his region is exposed on the surface of the envelope glycoprotein.
Example 4. Synthesis of Peptide 4 A synthetic peptide having the amino acid sequence shown under the "Peptid,e 4" heading in Table II, which corresponds to the sequence of residue numbers 735 through 752 of the gp 120 ., 1~9~
glycoprotein of -the viral causative agents of AIDS and ARC, was synthesized by solid-phase methodology (Merrifield, R.B., 32 Adv.
Enzymol. 221 (1969)) on a Biosearch SamII peptide synthesizer.
Butyloxycarbonyl-S-4-me-thylbenzyl-L-cystine coupled to polysty-rene using dicyclohexylcarbodiimide with a catalytic amount of 4-N,N-dimethylaminopyridine was used as the solid-phase support for the synthesis. The four amino groups were protected with tert-butyloxycarbonyl (t-BOC) and the side chain protecting groups were as follows: benzyl ether for the hydroxyl of serine, dichlorobenzyl ether for the phenolic hydroxyl of tyrosine, and the y and ~ benzyl-esters were used for the carboxyl groups on glutamic acid and aspartic acid, respectively. Trifluoroacetic acid (40% in CH2CL2) was used to remove t-BOC and the resulting salt was neutralized wi-th N,N-diisopropylethylamine (10% in CH2CL2). Diisopropylcarbodiimide was used to couple the t-BOC
amino acids. The specific steps of the synthesis are published in Sparrow, J.T., 41 J. Org. Chem. 1350 (1976).
The protecting groups were removed and the peptide was cleaved from the resin at 0 C with anhydrous hydrogen fluoride containing 10% anisole and 1% ethanedithiol as scavengers. The hydrogen fluoride reagent was removed under vacuum at 0 C and the peptide was then precipitated and washed with anhydrous ether. After extraction of the peptide from the resin with tri-fluoroacetic acid, the solvent was evaporated to 15 C and the peptide was again precipitated with ether. The ether was decanted after centrifugation and the pellet was dissolved in 5%
acetic acid with 6M guanidine HCl.
That solution was desalted on a *BioGel P2 column equili-brated in 5~ acetic acid and the peptide containing fractions ~ were pooled and lyophilized. A cysteine residue was then added * Trade-Mark ~Zg31~8 to the carboxyl terminus of the peptide to provide a functional -SH group for the coupling of the peptide to carrier proteins. A
glycine residue was added after the cysteine to provide a spacer amino acid between the coupled cysteine residue and the amino acid sequences analogous to gp 160. A tyrosine residue was added to the amino terminus for radioactive labelling with 125Iodine to determine peptide-to-carrier protein coupling efficiency and to identify the peptide during purification by adsorbance at 278 nm.
After desalting on the BioGel P2 column in acetic acid and lyophilization, the peptide was found to have the expected amino acid analysis (see Table II) and elu-ted as a single peak (92%) on Clg-reverse-phase HPLC in a linear gradient of 0.05% trifluoro-acetic acid and 2-propanol.

Example 5.
Synthesis of Peptide 6 on Polyamide Resin A synthetic peptide having the sequence shown under the Peptide 6 heading in Table II, which corresponds to the sequence of residue numbers 503-532 of the gp 120 glycoprotein of the viral causative agents of AIDS and ARC, was synthesized on a polyamide resin as follows. The numbering system used throughout this specification is taken from the numbers assigned to the amino acid residues of HTLV-III as set out in Ratner, L., et al., "Complete nucleotide sequence of the AIDS virus, HTLV-III, 313 Nature 277-284 (1985). The sequence of the polyamide resin-Peptide 6 conjugate was as follows:

HlN-VAL-ALA-P~O-THR-Lr~-ALA-LYS-AR~-ARG-~o~
VAL--YAL~LN--ARG--GLU--L~S--ARG--AIA--VAL--GLY--~LE--GLY--ALA--LEU--PHE--LEU--GLY--PHE--LEU--t;LY~ A--GLY--O CH~--~--C--N~CH2--(RESIN) H

lf~ 93~L88 A. Preparation Of Func~ional Monomer Five grams of (26.8 mmol) 2-methylsulfonyl ethyloxycarbonyl chloride (MSC chloride) ~K+R Labs, ICN) were dissolved in 15 ml acetonitrile and added dropwise over a 20 minute period to a stirred solution of 2.1 ml (28 mmol) redi~tilled allylamine (Xodak) and 4.9 ml (28 mmol1 redistilled diisopropylethylamine (DIEA) in 20 ml acetonitrile. (DIEA (Aldrich) was refluxed over Il ninhydrin and redistilled.) The solution was ~tirred an addi-¦l tional two hours and the solvent evaporated. ~he residue was 10 11 taken up in 250 ml ethyl acetate and allowed to stand for one-two¦
¦l hours. The bulk of the DIEA hydrochloride salt precipitated as¦
¦l needles. After filtration and e~aporation, the crude material ¦l was dissolved in a minimal amount of chloroform and loaded onto al ¦~ siIica gel G-60 column ~60 g) packed in the same solvent. Elu-¦
!¦ tion with chloroform yielded pure MSC-allylamine. (RF on TLC =¦
¦! .64 ~Solvent = CHC13 CH30H, 9:1).) ;~ The remaining DIEA salts adsorbed to the column under these~
,~ conditions. Occasionally, material migrating near the solventl ¦I front on TLC contaminated the MSC-allylamine column fractions.¦
20 ¦1 ~hat material was removed by crystallizing the MSC allylamine¦
I ¦ from methylene chloride-hexane at -20C. Yield was 4.8 g (86%
¦ from MSC chloride).
B. Preparation of Cross-linker ~ he cross-linker N,N'-bisacrylyl 1,3-diaminopropane was pre-pared according to the method set out in ~elpern and Sparrow, ¦
~E~. Briefly, diaminopropane lAldrich) was dissolved in ace-¦
tonitrile and added dropwi~e to an acrylyl chloride-acetonitrile ¦ solution at 4C, allowed to warm to room temperature and stirred.
1 The diaiminopropane dihydrochloride was removed by filtration, I washed with warm acetonitrile, and the combined filtrates were l¦concentrated in vacuo. N,N'-bisacrylyl-1,3-diaminopropane was Ij Ol/MFM2 -32-I! l I! i ~Z93~

crystallized at 4C overnight and the resulting plates filtered and dried in vacuo.
C. Preparation of Polyamide Resin In a glass, 2-liter cylindrical, fluted polymerization ves-sel fitted with a nitrogen inlet and mechanically driven glass stirrer were added 490 ml hexane arld 290 ml carbon tetrachloride.
The solution was purged for 15 minutes with nitrogen to remove oxygen. To this solution was added an aqueous solution contain-¦
I ing N,N'-bisacrylyl-1,3-diaminopropane (2.9 grams, 15.9 mmol) I prepared as described in Example 5.B. mixed with 18.2 ml ~175 Il mmol) of N,N-dimethylacrylamide (Poly5ciences). Ten g (48 mmol) ¦ MSC allylamine prepared as described in Example 50A. and 120 ml ¦, water were added, and the solution was filtered and degassedl ! before addition to the organic phase. The density of ~he re-ult-¦
¦ ing mixture was adjusted to obtain a uniform suspension with~
I stirring at 400-450 RPM. Ammonium persulfate (BioRad) (0.5 g in¦
¦ ~ ml H20) and 1 ml o~ either sorbitan sesquioleate or sorbitanl monolaurate ~Si~ma) were added. A solution of 3 ml N,N,N',N'-¦
I tetramethylethylenediamine (TEMED) (BioRad) in 2 ml H20, pHI
20 ~l 6.5-705 (conc. HCl) was then added to the suspension. The sus-¦
pended emulsion was stirred for two hours under nitrogen atmo-l sphere. The resultant beaded material was then filtered and washed sequentially with water ~one liter3 methanol lone liter3,1 a mixture of dioxane:me~hanol:2 N NaOH (14:5:1, two liters, tol remove MSC group), water ~two liters), 1 N HCl (two liters),¦
water ~two liters), and then methanol (two liters3. The resin¦
was defined by suspension in methanol and decan~ing (3x~. After swelling in methylene chloride (Baker HPLC grade), the resin was ¦~hrunk in hexan~ and dried in vacuo. Large amorphous material was removed by sif~ing the resin through an 80 mesh (180 micron) I sieve.

,Ol/MFM2 -33-3~

The degree of functionalization was checked by coupling BOC-alanine to 100 mg of the resin using dii~opropylcarbodiimide as activator and 4-dimethylaminopyridine (recrystallized from ethyl ¦ acetate) as catalyst. Amino acid analysis showed a ~ubstitution of 0.15 to 0.35 mmol/g resin, dep~ending on the lok, and resins were prepared with as little as about 0.1 and as much as about O.5 mmol/g resin depending upon the ~nount of allylamine added.
l The loaded resin gave no detectable staining with picryl-sulfonic ! acid, indicating the absence of unreacted free amine. When swol-len in methylene chloride, the beads occupied about 2.5 times ¦ their dry bed volume. When swollen in dimethylformamide or an ¦ aqueous solution, the beads occupied approximately four and six times their dry bed volume, respectively.
D. Preparation of Linker ¦The linker BOC-glycyl-4 (oxymethyl) benzoic acid was pre-pared by modification of the method of Mitchell, et al., supra.
Briefly, the 4-(bromomethyl) benzoic acid phenylacyle~ter was prepared by dissolving 10.3 ml redistilled diisopropylethylamine and 12.05 y ~60.6 mmol) bromoacetophenone in 450 ml ethyl ace-~0 tate. 4-(bromomethyl) benzoic acid (13.89 ~, 60.6 mmol) was added in seven equal portions over a three hour period to the¦
stirred solution at 40-50C. Stirring was continued for two more hours at room temperature. Precipitated Et3N HBr was removed by ~iltration and the ethyl acetate solution was washed four times with 50 ml each of an aqueous solution of 10~ ci~ric acid, sat-urated sodium chloride, saturated sodium bicarbona~e, and sat-urated sodium chloride. The organic phase was dried over anhy-drous magnesium ~;ulfate and freed of solvent by rotary evapo-l ration under reduced pressure. The residue was crystallized from 30 ~ CH2C12-petroleum ether (1:3 v/v) to give ~he 4-(bromomethyl) benzoic acid phenylacylester.

¦ 01/MFM2 -34-, Il lZ93~8~3 The 4-(bromomethyl) benzoic acid phenylacylester was con-verted to BOC-glycyl-4-(oxymethyl) benzoic acid by dissolving BOC-L-glycine ~25 mmol, 4.38g) in 15 ml methanol and titrating tol neutrality with tetramethylammonium hydroxide (25% in methanol).
Solvent was removed azeotropically with chloroform in vacuo, and the salt dissolved in 150 ml acetonitrile. To the stirred solu-tion was added 5.8 g (17.5 mmol) of the 4-(bromomethyl) benzoic acid phenacyl ester prepared as de~scribed. After overnight mix-¦ ing, the precipitated tetramethylammonium bromide was filtered ~ and the solvent evaporated. The residue was dissolved in 400 ml j ethyl acetate and the solution filtered. The organic phase wasthen washed successively with 10% aqueous citric acid (3 x 75 ¦, ml), 0.5 M sodium bicarbonate: 0.S M potassium carbonate ~2~
¦I pH 9.5 (8 x 75 ml), then water (3 x 75 ml). The solution was¦
dried (MgSO4) and the solvent removed in vacuo. The residue was dissolved in 200 ml of 85% acetic acid to which 23 g acid washed zinc dust was added. The mixture was stirred until the phenacyl ester was no longer visible by TLC (4 - 5 hours). The zinc was i filtered and washed with 50 ml acetic acid, and the combinedl 1I solutions were lyophilized. The residue was suspended in 100 ml¦
~! water:300 ml ethyl acetate, and the pH adjusted to 105 (conc.l HCl). The aqueous layer was extracted with a second portion of¦
ethyl acetate (200 mll, and the combined extracts were washed ~with water (100 ml). After drying (MgSO4) and evaporating, the BOC-glycyl-4(ox~methyl) benzoic acid was purified by recrystal-lization from methylene rhloride:hexane at -10~. Yield was 4.5 g (14.5 mmol, 83% from the phenacyl ester1 E. Coupling of Linker To Polyamide Resin BOC-glycyl-4-(oxymethyl) benzoic acid prepared as described I in Example 5.D. was coupled to the aminomethyl polyamide resin ~1.2 g) prepared as described in Example 5.C. on a Biosearch Sam ', , .

~93~8~
II automated peptide synthesizer using dicycl~hexylcarbodiimide a~d dimethylaminopyridine as activator in a 1:1 methylene chlo ride:dimethylformamide solution. Both methylene chloride (Baker HPLC gradel and dimethylformamide (Baker Photrex g~ade) were used~
~ithout further purification. Following treatment with hydrogen fluoride, 50 mg of the glycyl resin was found to contain 0.15¦
~ol/g by amino acid analysis. Am.ino acid analysis was performed using either (1) a Beckman Model ll9 amino acid analyzer following either a two hour hydrolysis (12 N HCl:propionic acid,¦
~0 1:1, 135C) or 24 hour hydrolysis (6 N HCl, llO~C) of resin bound peptides or (2) a Beckman Model 7300 amino acid analyzer following a two hour hydrolysis (12 N HCl:propionic acid, 1:1, containinq 0.05% phenol at 135~C. The results of the amino acid analysis are set out in Table III. I
Peptides l, 2, 7, and 10 were synthesized in the same manner¦
on the resin to give the corresponding polyamide resin-peptide conjugates.

I Examples 6-lO. Synthesis of Additional Peptides The method described in Example 4, above, was used to syn-~
thesize the Peptides 3, 5, 8 and 9 listed in Table II, each corresponding to the amino acid sequence of the residues listed.

Example 11.
Conjugation of Synthetic Peptide to Carrier Synthetic peptide 4 (see Table II) was conjugated via thel -SH group on the cysteine residue to the amino groups on Keyhole, limpet hemacyanin (KLHl~for immuniza~ion of rabbits) and bovine !
serum albumin (BSA)(for assaying anti-peptide activity) using a I heterobifunctional cross-linker, M-maleimidobenzyl~N
30 1I hydroxysuccininmide ester (MBS~. The details of this method are ilgiven at Liu, F.T., et al., 18 Biochemistry 690 (1979) and Green, ¦`N. et al., 28 Cell 477 (1982).
i i~",.
, ,. ~ ,.", .~
~, ,~ ,~, ....
01/MFM2 -36~

333L8~3 TABLE III

Residues Before HF After HF
, Treatment Treatment Thr ~75 (l) .85 ¦1) Glu/Gln 2.30 (2) 2.15 (2) Pro N.D.b (1) 1.07 (lJ

Gly 4.85 (5) 5.35 (5) Ala 4.70 (5) 5.19 (5) 1 Val 3.60 ~4) 3.59 (4) ;l Ile 0.94 (1) 1.04 (1) Leu 2.70 (3) 3.12 t3) Phe 2.00 12) 2.00 (2) Lys 2.73 (31 4.10 ~3) Arg 4.00 (4) umoles/g 71 50 iI _ .
¦ a Values are uncorrected for destruction during hydrolysis., li The number in parenthesis represents the theoretical yield for ,' each amino acid based on the particular se~uence.
i ~ Not determined.

¦ Briefly, 1 mg of either KLH or BSA in lO mM sodium phosphate, pHI

i 7.2, was incubated with 4 mg and 800 ~g of MBS in dimethylforma-j ¦ mide, respectively, for thirty minutes at 25C. Unreacted MBSI
and solvent was removed on a Sephadex PD-10 column eqllilibrated, in 50 mM sodium phosphate buffer, pH 6Ø A 100 molar excess of Peptide 4 relative to RLH or BSA~ along with approximately ¦ 500,000 cpm of 125[I~ labeled Peptide 4r was added to the reaction mixture land incubated an additional three hours at 25C.
IPeptide which was not bound to the protein carrier was removed by lO ll repeated dialysis. The coupling efficiency was determined by the amount of 125[I3 peptide associated with KLH and BSA and was 1' .

-~ ~ 01/MFM2 -37-I * Trade-mark 3L293~18 i .
~pproximately 62% and 56~ for RLH and BSA, respectively.
The sequence of the carrier-Peptide 4 conjugate was as ~ollows:

! .
¦ H2N-TYR-ASP-ARG-PRO ~ LU ~ LY ILE ~ LU-~GLU-GLU-GLY-GLY-GLU-ARG-ASP-ARG-ASP-ARG-H H _~ O
76~ N C-cH2-s- [~ -(KLH) t~YS) Peptides 3, 5, 8 and 9 (see Table 11) were conjugated to the carrier in the same manner.

Example 12 Induction 10 l of Immunogenic R_sponse in Rabbits._ Peptide-Carrier , Two rabbits were each immunized with 100 ~g per dose of Pep-tide 4-KLH complex, prepared as described above, emulsified in Freunds incomplete adjuvant. The rabbits rec~i~ed one intramu~-, cular injection every two weeks, for a total of three injections, and serum was obtained following each immunization.
A solid phase radioimmunoassay was used to titrate the rab-I bit anti-peptide antisera. Briefly, 200 ng of Peptide 4 coupled ¦l to BSA prepared a~ described in ExampIe 11 was adsorbed to thel Iwells of poly~inyl microtlter plates, and incubated o~ernight atl - 20 4C. Following the addition of 10~ n~rmal goat serum (NGtS) to block nonspecific sites, ~he rabbit anti-peptide antisera diluted in 10% NGtS was added and incubated 2 hours a~ 37C. Antisera was obtained 14 days after each immuniza~ion. The microtiter Iwells were washed with Tween 20 phosphate bufered saline (~-PBS) ¦ and 125[I] goat-anti-rabbit gamma globulin (approximately 500,000 cmp in 50 ~1) was added. Fol~owing incubation for 1 hour at ~37C, the wells were washed of excess radioactivity with T-PBS, 01/M~M2 -38-r7~ ~ ~ i * Trade-mark ~ :~293~l~8 ¦ and counted in a gamma counter. All vol~nes were 50 ~1 and the¦
¦ anti-peptide titers chown in Table IV are expressed as the ¦ reciprocal of the endpoint titer dilution (the highest dilution of antisera that qave cpm above the preimmune rab~it sera). The~
end point titers were based on fivefold dilutions and represent the mean of tripllcate value~.

TABLE IV
Rabbit Immunization Anti-P~e~ide 4 Titer¦

Pre-immunization 10 , Primary 1250 ¦! 21 Secondary 6250 Tertiary 31,250 Pre-immunization 10 ¦ Primary 1250 22 Secondary 6250 Tertiary 156,250 The results given in Table IV show that the two rabbits ¦ produced a detectible anti-peptide response (as measured by a I peptide-BSA) after a single injection of the peptide-KLH. Serum, ¦ obtained from each rabbit prior to immunization failed to signif-i icantly bind the peptide (titers of less than ten). Anti-peptide I titers increased following each injection of the peptide and ¦ ranged from 31,250 to 156,250 following the third injection.
i The specificity of the antibody response was shown by thej ¦ inability of the anti-peptide sera to bind the control peptid~l, conjugated to BSA. In addition, the HTLV-III peptide 728-745¦
(Peptide 3) completely inhibited (100%) ~he binding of the rabbit anti-peptide to peptide-BSA. The two rabbits also produced high~
I antibody titers to KLH; howe~er, rabbit anti-KLH did not bind, I peptide-BSA.
1 1.

~ 01/MFM2 -39-Z~3~8~

Example 13. Induction of Immunogenic Response in Rabbits: Polyamide Resin-Synthetic Peptide Conjugate Two rabbits were each immunized with the polyamide resin-peptide 6 conjugate (200 ~g of peptide 6 per dose) in Freunds complete adjuvant. Rabbits received an intramuscular injection every two weeks for a total of three injections. Subsequent injections were given at monthly intervals, and serum was ! obtained 30 days after the fourth injection. Control anti-¦ peptide antisera was produced in rabbits immunized ~ith either a 10 !! simian virus 40 tumor antigen peptide coupled to KLH by themethod described for the coupling of peptide 4 to KLH in Examplej 11 or a hepatitis B surface antigen resin bound peptide produced !
by the method of Example 5.

Example 14. Reco nition of HTL~-III Proteins By Rabbit Anti-Peptide Antibodies: Peptidès - Carrier j The ability of the rabbit antibodies to Peptide 4 to recog-¦¦ nize the native proteins associated with HTLV-III was examined as ¦I follows. MOLT-3, an HTL~-III infected T-cell line, was labeled 20 , with 351S]-cystine and used for immunoprecipitation as described ~i in Example 2, above, to determine whether the anti-peptide sera would bind any radioactively labeled HTLV-III native proteins.
! The rabbit anti-peptide antibody from rabbits immunized with the~
carrier-peptide 4 conjugate prepared as described in Example 11¦
specifically precipitated a single protein of ~pproximately 160,000 daltons as shown by autoradiographs of SDS-PA&E. This I
protein is the precursor envelope glycoprotein gp 160 fl HTLV-III. No reactivity to HTLV-III proteins was demons~rated ¦
when preimmune rabbit sera was used in the immunoprecipitation experiments. The rabbit anti-peptide failed to recognize the gp~
~120 envelope subunit that is detected wi~h 35[S]-cystine labeled MOLT-3 cells when human antisera from AIDS patients is used in ~93~

lmmunoprecipitation. The gp 41 envelope subunit does not radioactively label to the same degree with 35[S]-cystine as gp 120 and is difficult to detect by immunoprecipitation.
The difficulty of producing gp 41 at a relatively high level !
of specific radioactivity was circumvented as follows. HTLV~
infected MOLT-3 cells were double 1abeled by the addition of both 35[S]-methionine and 35[S]-cystine. The glycoprotein populations present in those double cystine-me!thionine labeled ly~ates were !
then enriched by affinity chromotography on lentil-lectin ¢olumns as described in Example 2, above. Both gp 160 and gp 41 glyco-proteins was observed when the rabbit anti-peptide sera were reacted with those glycoprotein enriched fractions when analyzedl by the radioimmunoprecipitation experiment described in Example !
2, above.
Western transfer methods for HTLV proteins verified that the rabbi~ anti-peptide did recognize gp 41. That method uses stock solutions of infected H9 (BioRad Laboratories, Richmond, Cal.) or MOLT-3 cell lysates as a source of HTLV-III proteins. In that assay, 5 x 106 infected c~lls are solubilized in 1 ml of a 1%
Ij Zwittergent 3-14 (Calbiochem-Behring) solution for 5 minutes ana~
centrifugated at 1000 x g for 10 minutes. The resulting supernatan~ is mixed with an equal volume of disruption buffer (10 mM Tris-~Cl, pH 6.~ glycerol and 0.01% bromphenol blue) and boiled for 3 minutes. Eight ~l of disrupted cell lysate isl electrophoresed in adjacent lanes in 4-25% linear acrylamide¦
gradient gel (1.5 x 17 x 14 cm) for twenty hours under a constant voltage of 50 V per gel. Electrophoretically ~eparated gradienti gels ar~ then transferred to nitrocellulose sheets at 1 amp constant current for 90 minutes a~ 10C using the buffer system described by Towbin, et al., 76 Proc. Natl. Acad. Sci. 4350 ~1979) Pre-stained molecular weight markers tBRL)~I

were also electrophoresed and transferred to nitrocellulose to be used as standards for estimating the molecular weights of the transferred HTLV-III peptides.
After the transfer is completed, the nitrocellulose sheets were incubated with 100 ml of 5% w/v non-fat dry milk rehydrated¦
in PBS containing 0.001% w/v methiolate and 0.0001% v/v Antifoam A (Sigma) for 30 minutes at room tPmperature. Serial dilutions of sera obtained from the rabbits immunized with Peptide 4 were then incubated with the nitrocellulose sheets for 1 hour at 37C.
Nitrocellulose sheets were then washed with 100 ml of Tween 20 phosphate bufferred saline [T-PBS). Biotinylated goat anti-humanl IgG (5 pg/ml) was then incubated with the nitrocellulose sheets¦
for 1 hour at 37C in order to detect the binding of the rabbit¦
i anti-peptide antibodies. Nitrocellulose sheets were washed againl j~ with ~-PBS followed by the addition of 1 ~g/ml of avidin-labeledj ¦ horse radish peroxidase (Av-HRP) for 20 min at room temperature.
¦¦ After washing again with T-PBS, 100 ml of a peroxidase chrom-~agen:substrate solution (0.2 mg/ml of O-dianisdine in PBS plus 1 I! ~l/ml of 30~ ~22~ was added to the nitrocellulose membranes 20 ¦¦ until precipitates were observed on the membrane (10-15 min.).
¦i The peroxidase catalyzed reaction terminated by washing the ¦ nitrocellulose sheets in 2% SDS in water. Controls for the ¦ Western transfer assay include the use of normal human sera and al ¦ side by side comparison of the reactivity of the antisera withl infected and uninfected cell lysates. Binding with the gp 41¦
protein was observed, as well as with the gp 120 subunit.

Example 15. Recognition of Polyamide Resin Synthetic Peptides Coniugate bY Rabbit An~ibodies to Viral Causative Agents of AIDS and ARC
An enzyme-linked immunosorbent assay (ELISA) was used for ¦ 01/MFM2 -42 !, 3~293~
l i detection of human antibodies against the viral causative agents of AIDS and ARC. The polyamide resin-peptide 6 conjugate pre-pared as described in Example 5 was crushed with a mortar and pestle and a suspension of crushed conjugate was made in boratel buffered saline ~BBS), pH 8~0. One hundred microliters of that¦
emulsion containing approximately 10 ,ug of peptide 6 (weight¦
basis as calculated by amino acid composition) was absorbed to¦
the solid phase of Dynatech Immunolon II microtiter plates in ¦ (BBS), pH 8.0, for eight hours at 4C. Nonspecific sites werel I blocked with 10% normal goat serum (NGtS) in Tween 20 phosphate!
I buffered saline ~T PBS) and then washed with T-PBS.
¦ Rabbit sera diluted in 10% NGtS was then added to the Pep-~I tide 6-coated plates and incubated for one hour at 37C, followed ¦¦ by washing with T-PBS. Biotin-goat antibody to rabbit IgG ~Vec-l 1I tor ~aboratories, Burlingame, CA) was then incubated with the¦
¦ I rabbit sera for one hour at 37C. The wells were then washed and¦
avidin conjugated to horseradish peroxidase (Av-~RP) was added for 20 minutes at room temperature. The wells were then washed~
I with T-PBS to remove any unbound Av~RP and peroxidase activity¦
¦ was determined using a 1 mM solution of 1,2'-azino-di(3-ethyl-benzthiazoline-sulfonic acid) ~Sigma Chemical Co.) and 0.03% H20 as substrate. The reaction was stopped with 5% (w/v) sodium dodecyl sulfate in water prior to quantifying spectrophotome-l ~rically at 410 nm using a Dynatech plate reader. Optimal !
dilutions of each reagent were selected by titration. All reagents for determining ~pecific binding except the substrate were diluted in 10% NGtS. The resin-bound hepatitis B surface antigen described in Example 13 served as a control. The results are shown in Fig. 4, in which the results for the polyamide resin-peptide 6 conjugate are shown in graph A and the results for the polyamide resin-hepatitis B peptide conjugate are shown Ol/MFM2 _43_ !

~3~
in graph B. Rabblt anti-peptlde an-tisera were obtained Erom:
rabbit no. 1 (-); rabbit no. 2 (o). All tests were performed in triplicate and the brackets refer to the range of values.
Example 16. Neutralization of Viral Causative Agents of AIDS and ARC By Antibodies to Polyamide Resin-Peptide 6 Conjugate A. HTLV-III Replication Prior to examining the ability of the rabbit anti-polyamide resin-peptide 6 antisera to neutralize HTLV-III infectivity in vitro, it was necessary to determine the kinetics of HTLV-III
replication in a susceptible human T-cell line. Various dilu-tions of an HTLV-III viral stock were incuba-ted with normal human and preimmune rabbit serum similar to the me-thods utilized to examine viral neutralization wi-th -the anti-peptide reagen-ts.
Following a 1 hour incuba-tion with -the sera, the HTLV-III viral dilutions were added to cultures of susceptible A3.01 cells maintained according to the methods set out by Folks, T. et al.
(1985 Proc. Nat'l. Acad. Sci. USA 4539-4543 (1985)). Cultures infected with HTLV-III were maintained for 15 days and super-natant fluid from the cultures was removed on days 5, 8, 10, 12 and 15. HTLV-III replication was assessed by the presence of reverse transcriptase (RT) activity in the culture supernatant fluids.
The methods for performing reverse transcriptase ac-tivity are described in detail elsewhere (Barre-Sinoussi, F., et al., 220 Science 868-871 (1983). ~riefly, 15 ~1 of each supernatant from A3.01 infected cultures were added to 96 well microtiter plates that contain 50 ~1 of virus dilution buffer (0.05 M Tris-HCl, pH 7.8, 0.1 M NaCl, 0.15 mg/ml dithiothreitol (DTT), 0.1%
Triton X-100). Fifty ~1 of -the reaction mixture (1 M Tris-HCl, 3 M KCl, 0.15 MgC12, 10% Triton, poly A, oligo DT and .~
.~ . ~,, ~ , .

-4~-lZ~33 B8 3H-thymidine tri-phosphate ( H-TTP) was added and incubated 1 hour at 37C. Following the incubation, 50~ul of the mixture was I dotted on nitrocellulose filter paper~. The filter papers were¦
¦I washed successively in beakers containing: (1) 5%
¦ trichloroacetic acid (TCA) and 5% sodium pyrophosphate; (2) 5%1 TCA; and (3) 50% ethanol. The filter papers were counted in an¦
!! automatic scintillation counter and the cpm of H-TTP was ¦ determined.
Il To prepare the various dilutions of the HTLV III viral 10 ' stock, one normal human serum and the 4 preimmune rabbit sera was heat inactivated at 56C for 1 hour and filter sterili7ed through a 0.2 ~m filter. One hundred fifty microliters of a 1:5 dilution was incubated with an equal volume of 10 1, 10 2, 10 , 10 4 and 10 5 dilutions of an HTLV-III isolate, termed NY-5 for 1 hour atl 37C. The NY-5 isolate has an infectious titer of 10 5 units as¦
determined on the human T-cell line A3.01. Following the incu-¦
bation, the antibody treated ~irus mixture was added to 106 A3.01 cells. The mixture was incubated for 2 hours at 37~C in the l presence of 1 ~g/ml of POLYBRENE ~Cal Biochem). The infected 20 1 A3.01 cells were washed and resuspended in 1 ml of RPMI media containing 10~ heat inactivated fetal calf serum and dispersed 1, into 24 well microtiter plates. Five hundred microliters of¦
¦ spent media supernatant was removed at days 5, 8, 10, 12 and 15 ¦ after infection and frozen at -135C until reverse transcriptase ¦activity was determined. Following the removal of ~upernatant, ¦ the individual cultures were fed with 500 ,ul of RPMI plus fetal I
calf serum. Each culture is performed in duplicate. RT activity was determined by the counts per minu~e of 3H-TTP incorporated. I
! Pooled human AIDS serum that tests positive by ELISA and I
30 ¦ Western blot were obtained from Dr. Thomas Folks, Laboratory of ¦ Immunoregulation, NIAID, Bethesda, MD. The human sera and rabbit ,,'j .~

~ I f '~
!01/MFM2 -4S-' * Trade-mark lZ931~18 ¦¦ anti-peptide antisera were treated as described above. In each instance, the preimmune sera of that particular rabbit served as the negative control indicative of no neutralization of HTLV-III
infectivity as determined by RT activity cpm when compared to the individual rabbit anti-peptide preparation. The preimmune and rabbit anti-peptide antisera were incubated with 10 1, 10 2, 10 3, 10 4, and 10 5 dilutions of the NY-5 isolate as described above. Each culture was performed in duplicate. Supernatants I from infected A3.01 cells were removed at days 5, 8, 10, 12 and 10 ¦ 15, frozen at -135C, and assayed for RT activity. The percent¦
inhibition of RT activity was determined by the following~
formula:

cpm RT assay of anti-peptide antisera cultures 10 1 - cpm RT assay of prelmmune sera cultures x 0 Background counts ranging from 200 to 750 cpm were subtracted ¦~ from each determination prior to calculating percent inhibition.
I The kinetics of four dilutions of HTLV-III (A, 10 1 B, ¦ 10 2; C, 10 3; D, 10 4) based on RT activity at days 5, 8, 10, 12 I and 15 following infection are shcwn in Figure 5. The points on 20 ¦ the curve reflect the mean cpm of 3H-TTP uptake based on five ¦ different determinations performed in duplicate. The range bars signify the standard error of the mean. Significant replication I of HTLV-III diluted 10 1 and 10 2 did not occur until 10 days i after infection. At 12 days the virus was also actively repli-cating and by 15 days, the decrease in RT activity indicated the cytolytic action of these dilutions of HTLY-III for the suscep-tible target cells. Replication of the 10 3 dilution of HTLV-III
was observed at days 12 and 15, whereas little or no replication ¦ was demonstrated with 10 4 dilution of virus even by day 15.
30 11 Readings of greater than 2000 cpm of 3H-uptake were selected as ¦1 an indication of HTLV-III xeplication based on the fact that 11 I~ ~Z931~8 out of 12 determinations (days 5 and 8 with all dilutions of virus, day 10 for 10 and 10 4 dilutions and days 12 and 15 for 10 4 dilution) with less than 2000 cpm had standard deviations and standard errors of the mean greater than or equal to the mean cpm.
The finding that the standard deviations and standard errors of the mean equalled or exceeded the mean cpm at those dilutions indicated that the individual cpm values were extremely variable and that a determination of whether low RT activity resulted from 10 ¦ specific neutralization or just random variation of the RT assay ~! would be difficult. Antibodies may not efficiently neutralize virus if overwhelming quantities of infectious virions are pre-i sent. Based on the kinetic studies and the nature of the HTLV-III viral stock, it was determined that neutralization of HTLV-i! III infectivity by human and rabbit anti-peptide antisera would ¦I be examined at days 10, 12, and 15 for the 10 1 and 10 2 dilution l, and days 12 and 15 ~or the 10 3 dilution of HTLV-III.
B. Neutralization of Anti-PeJ~ide Antisera 1l As indicated by the results shown in Table V, the antiserum 20 1¦ to the polyamide resin-peptide 6 conjugate from one rabbit effi-¦l ciently reduced HTLV-III replication at day 10 when compared to ¦I pooled human sera from AIDS patients at both 10 1 and 10 2 dilu-~' tions of virus. A second rabbit anti-serum to that peptide failed to reduce HTLV-III replication and served as a control antiserum throughout ~he RT assay. No anti-HTIIV-III activity was detected in ~his particular antiserum based on radioimmunopreci-pitation even though the rabbit received a similar immunogen and produced a detectable anti-peptide response. ~he antiserum that ~ neutralized HTLV-III detected both the gp 120 and gp 160 envelope 30 I glycoproteins. Rabbit no. 1 antiserum was founa to be less effi-~ I cient in neutralizing HTLV-III when compared to human ~IDS serum 1~93~8B

on day 12 and 15. The percent reduction of RT activity decreased by day 12 from greater than 90 percent (day 10) to 23 and 45 per-l cent for a 10 1 and 10 dilution of virus, respectively. The more dilute the virus, the greater the reduction of RT activity at day 12, indicating that the ability of the antisera to neutra lize is dependent on the amount of virus.
Both rabbit antisera to the polyamide resin-peptide 6 conju gate neutralized a 10 2 virus dilution at day 10 but neither were as efficient at neutralizing higher concentrations of virus (10 13 when compared with the serum from a human AIDS patient or the rabbit anti-polyamide resin-peptide 6 conjugate. Rabbit no.
3 neutralized efficiently (greater than 95%) a 10 virus dilu-tion at both 12 and 15 days. No reduction in RT activity was obtained with antiserum from rabbit no. 4 on day 15, which may be~
reflected in the fact that rabbit no, 3 had a higher antibody titer to HTLV-III when compared to rabbit no. 4. No reduction of RT activity was observed on day 15 with any of the antisera at the high concentrations of virus indicating that infectious virus was present in the culture and the antisera were not effective in inhibiting viral replication at that point in time. Both antisera from the human AIDS patient and rabbit no. 3 inhibited RT activity at day 15 with a 10 dilution of virus. Control rabbit antisera produced ag~inst non-HTLV~III envelope glycopro teins coupled to KLH and the hepatitis B surface antigen controll resin bound peptide preparation at similar dilutions employed in¦
the in vitro neutralization test demonstrated no significant inhibition oi RT aotivity (less than 25~.

~9;~

TABLE V

Human Rabbits AIDS Serum Anti-503-532 peptide Anti-735 752 peptide ¦ Virus Dilution 1 ~ 3 4 ....~
Percent reduction of RT activity (10 days) ~l 98~ 96~ 0~ 0~ 0%

1 10-2 97% ~ 0~ 90% 97 ¦1 10-3 nd nd nd nd nd ¦I Percent reduction of RT activity (12 days) 91% 23~ 0% 0% 0%

-2 ~7% 45% 0% 67% 67%

, 10 3 100% 70% 0% 100~ nda ! -' - -~- ~ ' ~ ............... ___ r _______ Percent reduction of RT activity (15 days) 0-1 0% 0% 0% 0% 0%
0-2 0% 0% 0% 0% 0 10-3 90% 24% 0% 95% a~

Il a The number of cpm of 3~ uptake in that particular culture I' was less than 2000 cpm and the percent reduction of RT activity ! was not determined.

I . __ ......... . __ .... _... ~ .................. ..
Example 17. Assay for Diagnosis of AIDS or ARC: Detection of Antibodies An insoluble support matrix is coated with 5 ~g each of the polyamide resin-synthetic peptide conjuga~es prepared as described above in Example 5 in borate buffer saline (BBS), pH
8.0, for 8 hours at 4~C. (Alternatively, the matrix may be coated for one hour at 37C). The conjugate is blocked for 20 minutes with 10% nc,rmal goat ~erum (NGtS), and washed three times¦
with Tween 20 phosphate buffered saline (T-PBS). A serum sample suspected of containing antibody to the viral causative agents of l l AIDS and/or ARC is added and incubated for one hour at 37C. The support matrix is washed three times with T-PBS, and biotin-labeled goat anti-human Ig (1:1000 of 5 mg/ml in 10% NGtS, Vector Labs, Burlingame, California) is added. The matrix is washed three times with T-PBS, and a 1:2000 of S mg/ml avidin-horse-radish peroxidase is added and incubated for twenty minutes at room temperature. The matrix is washed three times with T-PBS
and the substrate, the diammonium salt of 2, 2'-azinodi-(3-ethyl-I benzthiazoline sulfonic acid) (ABTS) with H2O2, is added. The lO ¦ enzyme reaction is stopped with 10% SDS and optical density is I read at 410 nm as described in Example 15.

I Example 18. Assay for Diagnosis of AIDS or ARC:
' Detection of Antigen To detect the presence of the AIDS antigen, the solid phase ¦¦ matrix is coated with antibodies produced by immunization of an il experimental animal with one or more of the polyamideresin-peptide conjugates pre~ared as described in Example 5, and the antibodies blocked and washed as described above. The I biological fluid sample suspected of containing the AIDS or ARC
20 ¦ viral causative agent is then added and washed. The assay can be conducted either as a direct binding assay or as an inhibition I assay. If a direct binding assay is conducted, biotin-labeled ¦ antibodies to the AIDS and/or ARC virus produced as described above are added and washed. The avidin-labeled enzyme is then added as described above and washed, and the substrate is added as described above. The reaction is stopped and the optical density is read.
If conducted as an inhibition assay, instead of adding bio-I tin-labeled antihody to AIDS vixus, the biotin-labeled synthetic~
30 I peptide is added and the insoluble support matrix is washed. The ! avidin-labeled enzyme is then added and washed. The substrate is ~ , ,.
Il I .
I 01/MFM2 _50_ lZ9318~3 added, the reaction stopped and optical density is read. The IgG
from human or chimpanzee AIDS-containing serum is purified by ion exchange chromotography on a Whatman DE--52 anion exchange column.
IgG from rabbit anti-peptide is puxified with a protein A-Seph arose 4B column (Pharmacia). The IgG is biotinylated using bio-tin-N-hydroxysuccinamide ester (Boehinger Manheim).

Example :L9.
Vaccination A~ainst AIDS and ARC
l . .. .
l To vaccinate an ~xperimenta:L animal against the viral 10 ll causative agents of AIDS and ARC, the synthetic peptides 1~9 are synthesized on the polyamide resin as described in Example 5, above. The polyamide resin-synthe~ic peptide conjugate, or a mixture of several conjugates, is injected into the animal in a bolus of between 100 to 1000 ~g of synthetic peptide in alum as an adjuvant. Thxee separate injections may be given, either intramuscularly or subcutaneously, on a biweekly basis until a j measurable antibody response to the virus is detected. Other i time inter~als such as 0, 1 and 6 months may also be used for the injection of the synthetic peptide.

20 I Example 20.
Screening of Putative AIDS Vaccines The synthetic peptide of the present invention can also be used to screen potential AIDS vaccine candidates for their abil-¦
ity to induce an immunogenic response in an animal subject. One ¦ or more of the polyamide resin-synthe~ic peptide conjugates are ¦ coated onto the insoluble matrix as described above in Example 1 15. The vaccine candidate is then incubated with antibodies ¦ against the peptide (with or without biotin labelling). If ! biotin labeled, t:he avidin-enzyme is added, if not, a biotin I anti-species antibody such as hiotin goat anti rabbi~ IgG is i added, followed by the addition of the avidin-enzyme. The sub-¦l strate is added, t:he reaction stopped and optical density read to Il . , !
01/MF~2 -S1-lZ~318B
determine the ability of the vaccine candidate to block the bind-¦ ing of the peptidec ! * * * * *
The preceding examples are presented by way of exemplifi-cation only and not by limitation. Variations in these methods will be known to those skilled in the art, and it is expected that all such variations will be made without d~parting from the spirit and scope of the present invention as claimed in the fol-lowing clsims.

I' ll ~ OltMFM2 -52- .
1~ !

Claims (14)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A composition of matter capable of inducing an immuno-genic response to the viral causative agents of AIDS and ARC com-prising a polyamide resin and a synthetic peptide comprising a chain of amino acids having a sequence homologous to a portion of the amino acid sequence of the gp 120 or gp 41 envelope glycoprotein of HTLV-III, ARV or LAV and having a hydrophilic region therein.

2. The composition of claim 1 wherein said chain of amino acids includes a .beta. turn.

3. The composition of claim 1 wherein said polyamide resin comprises a cross-linked polydimethylacrylamide resin.

4. The composition of claim 1 wherein said chain of amino acids is conjugated to said polyamide resin through a linker.

5. The composition of claim 4 wherein said linker is an oxyalkyl benzoic acid derivative.

6. A method of immunizing an experimental animal against the viral causative agents of AIDS and ARC comprising synthesizing a peptide comprising a chain of amino acids having a sequence homologous to a portion of the gp 120 or gp 41 envelope glycoproteins of HTLV-III, ARV or LAV and having a hydrophilic region therein on a polyamide resin and administering an immunogenically effective amount of the polyamide resin-peptide conjugate to an experimental animal.

7. The method of claim 6 wherein said polyamide resin-peptide conjugate is administered to said animal in a pharmaceutically acceptable diluent.

8. The method of claim 7 wherein said diluent additionally comprises an adjuvant.

9. The method of claim 7 wherein said diluent is either distilled water or a neutral pH buffer.

10. A method of detecting antibodies against the viral causative agents of AIDS and ARC comprising:
conjugating a polyamide resin-synthetic peptide conjugate to the ligand of a specific binding pair wherein aid binding pair is comprised of said ligand and an anti-ligand having specific affinity for said ligand and said synthetic peptide is comprised of a chain of amino acids having a sequence homologous to a por-tion of the gp 120 or gp 41 envelope glycoproteins of HTLV-III, ARV or LAV;
contacting said conjugate with sera from an animal, thereby causing any antibodies to the viral causative agents of AIDS or ARC in said sera to bind to said conjugate; and thereafter contacting the bound antibodies with the anti-ligand of said specific binding pair.

11. The method of claim 10 wherein the ligand is an enzyme and the anti-ligand is a substrate for which said enzyme is specific.

12. The method of claim 11 wherein the enzyme is horseradish peroxidase and the substrate is hydrogen peroxide.

1 The method of claim 10 wherein the ligand is an antibody and the anti-ligand is an antigen for which the antibody is specific.

14. The method of claim 10 wherein the ligand is an antigen and the anti-ligand is an antibody specific for said antigen.