CA2185651A1 - Stress protein-peptide complexes as prophylactic and therapeutic vaccines against intracellular pathogens - Google Patents

Abstract

Disclosed is a family of vaccines that contain stress protein-peptide complexes which when administered to a mammal are operative to initiate in the mammal a cytotoxic T cell response against cells infected with a preselected intracellular pathogen. Also disclosed are methodologies for preparing and administering vaccines containing such stress protein-peptide complexes.

Description

w0 9~l24923 2 1 8 5651 STRESS PROTEIN-PEPTIDE COMPLEXES AS PROPHYLACTIC AND
THERAPEUTIC VACCINES AGAINST INTRACELLULAR PATHOGENS

held of the Invention The rnvention relates generally to the field of vaccine d.~ .lu,u~ . More ~II Li~ -ly, the invention relates to the d- v .lùluI~ I of 1~l u~ yLq~ and 10 Il.~,,.~,~..l;~vaccineseffectiveagainstintrqrqlll-lqrpathogens.
Background of the Invention The d. ~ ~lu~ull~ lll of vaccines drrected against intrqrY~ lqr pathogens, for 15 example, viruses, bacteria, protozoa, fungi, and intrqrPIlIlIq~r parasites, is ongoing.
The development and use of vaccmes has proved invaluable in preventing the spread of disease im man. For example, in 1g67, smallpox was endemic in 33 countries with 10 to 15 million cases being reported annually. At that time, theWorld Health O~ l~Liul~ introduced a program to eradicate smallpox.
20 Al~lu7~ 1y one decade later, smallpox was successfully eradicated from the human population.
ThPrr~tir.qlIy, an ideal vaccine has a long shelf life, is capable of inducing with a srngle dose long lasting immunity against a l,-~sel~ed pathogen and all of 25 its ~ I,uly~i~ variants, is incapable of causing the disease to which the vaccrne is directed against, is effective ~ Ally and lu~uiullylr~li~lly, is prepared easily and~.. l.. ,;.,.llyusingstandardm~qth~lrlrlr~gies~andcanbeadministeredeasilyin the field.

WO gs/249z3 2 1 8 ~ 6 5 1 r~ r~

Presently four major classes of vaine have been developed against m~mm~ n diseases. These include- Iive-attenuated vaccines; non living whole vaccines; vector vacdnes; and subunit vaccines. Several reviews discuss the ,uala~iOn and utility of these classes of vaccines. See for example, Subbarao et al.
5 (1992) in Genetically En~ineered Vacdnes, edited by Ciardi et al., Plenum Press, New York; and Melnick (1985) in Hi~h Technology Route to Virus Vaccines, edited by Dreesman et al., pub~ished by the Ameri~an Society for Microbiology, the disclosures of which are ill-uliuula~e l herein by reference. A summary of the advall~a~,~ and di~advalltà~s of each of the four classes of Yaccines is set forth 10 below.
Live attenuated vacdnes comprise live but attenuated pathogens, i.e., non-virulent pathogens, that have been "crippled" by means of genetic mutations. Themutations prevent the pathogens from causing disease in the recipient or vaccinee.
15 The primary advantage of this type of vacdne is that the attenuated organism stimulates the irnmune system of the recipient in the same manner as the wild type pathogen by mimicking the natural infection. r~ ~ vl e, the attenuated pathogens replicate in the vaccinee thereby presenting a continuous supply of antigenic~ I~.,..;,,,..-l~totherecipient'simmunesystem. Asaresult,livevaccines 20 can induce strong, long lasting immune responses against the wild type pathogen.
In addition, live vaccines can stimulate the production of antibodies which neutralize the pathogen. Also they can induce resistance to the pathogen at its natural portal of entry into the host. To date, live attenuated vaccines have been developed against: smallpox; yellow fever; measles; mumps; rubella; pulioll~y~ is, 25 ad~l,uvilus, and Iub~ulùsib.
Live attenuated vacdnes, however, have severâl inherent problems. First, there is always a risk thât the attenuated pathogen may revert back to a viruient phenotype. 1~ the event of pl~ u~ylui~ reversion, the vâccme may actually induce ~ w0951z49z3 2 1 85651 F~

the disease it was designed to provide immunity against. Second, it is expensiveand can be imrrArfirAl to develop live vaccines directed against pathogens that ly change their antigenic ~ For example, I ~a ~ ..a have been umable to develop a practical live vaccine agamst the influenza virus because 5 the virus continually changes the antigenic ~F ~. . I . . ~ IA 111~ of its coat proteins. Third, live attenuated vaccines may not be developed against infections caused by l~lluvi~ sandl,A"~t.-..,~ viruses. Thenucleicacidsfromthesevirusesmay integrate into the recipients genome with the potential risk of inducing cancer in the recipient. Fourth, during the " ~ 11 r~ of live attenuated vaccines 10 a.l Y ~ Li~iuu~ agents present in the cells in which the vaccine is -, ~ rA~ d may be copurified along with the attenuated pathogen. Alien viruses that have been detected in vaccine ~uleluala~iOIl~ to date include the avian leukosis virus, the simian u~lluùvavil u~ SV40, and the simian cytomegalovirus. Fifth, live vaccine ,UI _IUala~iUlls can be unstable therefore limiting their storage and use in the field.
15 Presently, attempts are bemg made to develop stabilizing agents which enhance the longevity of the active vaccines.
Non living whole vaccines comprise nûn viable whole organisms. The pathogens are routinely inactivated either by chemical treatment, i.e., formalin20 illa~Livaliull, orby treatment with lethal doses of radiation. Non living whole vaccines have been developed against: pertussis; typhus; typhoid fever;
paratyphoid fever; and particular strains of influenza.
In principle, non living vaccmes usually are safe to administer because it is 25 unlikely that the organisms will cause disease in the host. I' UI Ll ~ lur~, since the organism is dead the vaccines tend to be stable and have long shelf lives. Thereare, however, several disadvantages associated with non living whole vaccines.
First, ~ care is required in their , IA 11- ~ r~ to ensure that no live pathogens remain m the vaccme. Second, vaccines of this type generally are w0 95/24923 2l 85651 0 ineffective at sfimll1Rtin~ cellular responses and tend to be ineffective against infrRrPlll1lAr pathogens. Third, the immunity elicited by non viable vaccines isusually short-lived and must be boosted at a later date. This process repeated]yentails reaching the persons m need of ~d..il,a~iol~ and also raises the concernS about l~y~ the vaccinee against the wild type pathogen.
Vector vaccines, also known as live l; ., ....1,;, ~ .I vehicle vaccines, may beprepared by ill~ullJula~ a gene encoding a specific antigenic ~IPfPrminRnf Of interest into a living but harmless virus or bacterium. The harmless vector 1û organism is in turn to be injected mto the intended recipient. In theory, the. ~.. 1.: . ,,. " 1 vector organism replicates in the host producing and presenting the antigenic ~i. 1 ~", ....,- " I to the host's immune system. It is ~ ~ that this type of vaccine will be more effective than the non-replicative type of vaccine. For such a vaccine to be successful, the vector must be viab~e, and be either naturally 15 non-virulent or have an attenuated phenotype.
Currently preferred vectors include specific strains of: vaccinia (cowpox) virus, adenovirus, adeno-associated virus, salmonella and llly~ubd~ ia. Live strains of vaccinia virus and Ill~uba-~elid have been adlllillialele l safely to20 humans in the form of smallpox and ~u~el.ulObia (BCG) vaccines, lealJe~ ely.
They have been shown to express foreign proteins and exhibit little or no conversion into virulent ~ .JIyl./eS. Several types of vector vaccines using theBCG vector currently are bemg developed against the human; , " .... " ~ ti.; Pnry virus (HIV). For example, the HIV antigenic proteins: gag; env; HIV protease;
reverse 1,,.I.~.l, ;I.I~ce, gpl20 and gp41 have been introduced, one at a time, into the 8CG vector and shown to induce T cell mediated immune responses against the HIV proteins in animal models (Aldovini et al. (1991 ) Nature 351:479-482; Stover ef al. (1991) Nature 351:45~460; Colston (1991) Nature 351:442-443).

WO 95/24923 P~

s Vector vaccines are capable of carrying a plurality of foreign genes thereby permitting ~;,.",l~ ",c v~ dLiull against a variety of preselected antigenic C For example, researchers have engineered several HIV genes into the vaccinia virus genome thereby creating ..,ulLiv~ vaccines which therefore 5 are, in theory, capable of ~ y ctim~ tin~ a response against severa HIV proteins.
There are several dia~lv~ Ld~es associated with vector vaccines. First, it is necessary to identify suitable strains of viable but non-u.,~l,o~.,ic organisms that 10 may act as carriers for the genes of interest. Second, vector vaccines can be prepared only when a potentially protective antigenic ~ c has been identified and 1,~. ,,. 1~, ;,~i Accordingly, vector vaccines cannot be prepared against pathogens whose antigenic ~iPtPrmin~nt has not yet been identified or are so variable that the prospect of identifying the antigenic determinant for each 15 variant is ;1.~ Third, the genes encoding the ~ul~:Del~l~l antigenic ~iPtprmin;~nt must be stably transfected and expressed in the preferred carrier organism. ('~ cp~ ly, the mPth~ iPc required for developing this type of vaccine are both labor intensive and time corlsuming. Fourth, it has not yet been PCt~hliChP I that rPr~mhin~nt vector vaccines effectively immunize a recipient 20 against a preselected pathogen.
Subunit vaccines usually comprise a subcellular ~ 1 purified from the pathogen of interest. Subunit vaccines usually are safe to administer because it is unlikely that the subcellular ~:v" .~ c will cause disease in the recipient. The 25 purified subcellular ~ may be either a defined subcellular fraction, purified protein, nucleic acid or polyD~ e having an antigenic ~iPtPrmin~nt capableof~1;.,."1~1;,~animmuneresponseagainstthepathogen. Theantigenic ~UIIIIUUlI~:lILD can be purified from a ,u.~:,u~ iu-, of disrupted pathogen.
AlL~..,aLiv~l~,thea~tigenicproteins,nucleicacidsorpol,y;,~ lidesmaybe wo gsi24923 2 1 8 5 6 5 1 ~ ' ' using u~u ed~ well known in the art. Diseases that have been treated with subunit type vaccine6 include: cholera; dirh~hDn~- hepatitis type B;
F ~ ,~ tetanus; and specific strains of influenza.
S There are, however, several ~' l v v associated with subunit vaccines.
First, * is important to identify ar~d .1. -- -. t .; - the protective antigenic.1. t. . - - ~ This can be a labor intensive and time g process. As a result it rnay be irr~rrP~irPI to develop subunit vaccines against pathogens with highly variable antigenic ~' Second, subunit vaccines generally are ineffective at ctirn~ cytotoxic T cell responses and so they may be ineffective at~ anunmuneresponseagainst; I,_ -11~1 pathogens. Third,the immunity elicited by subunit vaccines is usually short-lived, and like the non living whole vaccines must be boosted at a later date therefore raisir~g the concern about l.ylJ. . -.- .. -;1; ;..~ the vaccinee against the wild type pathogen.
~ r~trf~r~, mary of the inactivated whole and subunit vaccine6 have not been ~ulrl~ t.ly O by i' ~ 1~._ to induce strong, protective responses. As a result,; -- .. ~I; .. l_ . ~ including, for example, aluminum hydroxide; intact ....~ul~ ~, and/or I~ ul~a-t~ l r ' have been co-20 ~ 1 with these vaccines to enhance the immune response stimulated by the vaccine. Recently, ~ r ' ' have shown that ~ ubc~t~l~l heat shock proteins may act as carriers for peptide vaccines thereby enhancing the . Ø ..: ;Iy of the peptide6 in vivo (Lussow et al. (1991) Eur. ~. Irnmunol.
~:2297-2302). Further studies have shown that ~ h ;..g a ~ I: --- to 25 mice ~ g an antigenic peptide chemically ~ 1 to a pur*ied .J ~..I~.t..~l stress protein stimulates a humoral (antibody mediated) rather than a temporal (cell mediated) responæ against the antigenic peptide ~Barrios et al.(1992) Eur. T. Inununo . 22:1365-1372).

W0 9s/24923 2 ~ 8 $ 6 ~ 7 P ~

However, because it is generally believed that celluldr responses are required for ;~ against; ~ pathogens (see for example, "Advanced T.. .-- . rI- ~;y," Male et al. (1991) Gower Medical PuLLal i~
Rdychaudhuri et ~. (1993)1--1-l,ul-010~ Today 14: 344-348) it is ~ 1 that S ~U~ iO- dl subunit and inactivated whole organism vaccines may be ineffective at .I;, ....I ~ I; ..~ immune responses, specifically cytotûxic T cell responses, against i..l.,... 11:.1,.1 pathogens.
It is an object of the instant invention to provide a safe subunit vaccine ..... ,I., ;~;, .~ a stress protein-peptide complex for R~ ., to a mammal that is capable of inducing, by means of a cytotoxic T cell response, resistance to infection by a preselected intrArPIIIllAr pathogen. The vaccines prepared in aordance with the invention may be used to eLcit an immune response against an intrAcrIIIlIRr pathogens whose antigenic ~--l r., . .; . ~ have been identified, have not yet been 15 identified, or where it is imrrArtirRI to isolate and ~ each of the antigenic .1..l ~. . 1,; . -_ - . l -~ The vaccines prepared in arrr,rrlRnrP with the invention mdy be r~ Lyandlll .,-I----I;-"llyeffectiveagainst~ le ~-lpathogens.
Another object of the invention is to provide a method for inducing in a 20 mammal resistance to infection by an intrArr~ -IRr pathogen by a~ l .; . .g to the mammal a stress protein-peptide subunit vaccine. Another object is to provide a method for rapidly and cost effectively producing commercially feasible quantities of the stress protein-peptide vaccines from a cell or cell line infected with the intrRrPlllllAr pathogen or al~..-d~iv~ly from a cell or cell line transfected with, 25 and expressing a gene encoding a specific antigenic determinant. Still another object is to provide a method for preparing an illl~llullO~;elli~ stress protein-peptide subunitvaccinebyl .~".~I;I"I;..~invitroimm11nrlrgirRllyunreactivestress proteins and peptides thereby to produce; ...., ... .... r~ I; v ~' complexes capable of ~I;.. l,,li.. ~animmuneresponseagainsta~ eleL~dintrArr-lllllArpathogen.

W0 95/24923 2 l 8 5 6 5 t F~~

These and other objects and features of the invention will be apparent from the ~ crrir~i~n, drawings, and daims which follow.

~ wo95/24923 1~~

2 1 8565~
g Summary of the Invention It has now been discovered that a subunit vaccine containing a stress S protein-peptide complex when isolated from cells infected with a l l~a~ d intr~rPlllllAr pathogen and then a.l..li-u~L~I ~ to a mammal can effectively stimulate cellular immune responses against cells infected with the same pathogen.
Specifically, the immune response is mediated through the cytotoxic T cell cascade which targets and destroys cells containing intracellular pathogens.
The vaccines prepared in ArrnrrlAnrP with the mPthr,flolr,~ir~ described herein provide an alternative approach for ~ cellular immunity thereby obviating the use of live (AttPmlAtPr~ or otherwise) intrAr^ll llAr pathogens. In addition, the vaccmes described herein are ideal for inducing immune responses 15 against intrArPll-llAr pathogens having either defined or as yet undefined illUllUI~OE;l:lliC r~ Ful ~ the vaccines may be used to induce immune responses agamst;, 1 . ~ ,1,. . pathogens whose antigenic rl l~
are either diverse or constantly changing thereby making the isolation and , . of antigenic d~L~I Illilldlll:7 impractical.
In a preferred aspect, the invention comprises a vaccine that can be administered to a mammal for inducing in the mammal a cytotoxic T cell response against a 1~ IP.1 intrArPlllllAr pathogen. Also, it is ~ P~ that the vaccines may induce rn the mammal, by means of a cytotoxic T cell response, 25 resistance to infection by the ~ le l~d ;~ pathogen. The vaccrnes d in Arrr~rdAnrp with the principles described herein contain an ,.".""nr,g.-, ;, stress protein-peptide complex that is capable of 5tim~1Atin~ in the recipient a cytotoxic T cell response directed against cells infected with the pathogen of ~terest. The complex when combined with a pl,,.",.,.. ~ ;. ally WO 95124923 r~
2~ 8565~ --acceptable carrier, adjuvant, or excipient may be ad..,il-.~t~,~ to a mammal using techniques well known in the art.
The term "Yaocine", as used herein, is lln~lPrc~nod to mean any composition containing a stress protein-peptide complex havmg at least one antigenic which when a.l.~ d to a mammal stimulates in the mammal an irllmune response agamst the antigenic determinant.
Tlle term "stress protein" as used herein, is lln~Prctr~od to mean any cellular 10 protein which satisfies the following criteria. It is a protein whose in~r~rP~ Ar rrnrPn~rA~irn increases when a cell is exposed to stressful stimuli, is capable of binding other protems or peptides, and is capable of releasing the bound proteins or peptides in the presence of adenosine t- i,ul~o~ (ATP) or low pH. Stressful stimuli include, but are not limited to, heat shock, nutrient deprivation, metabolic 15 disruption, oxygen radicals, and infection with in~rArPlllllAr pathogens.
It will be apparent to the artisan upon reading this disclosure that other ,~ "" .1.; ",., .1 stress proteins, includmg non native forms, truncated analogs, muteins, fusion proteins as well as other proteins capable of mimicking the peptide 20 binding and immllnr,g~nir properties of a stress protein may be used in the ur. l~l~iul~ of stress protein-peptide vaccines disclosed herein.
The first stress proteins to be identified were the heat shock proteins IHsp).
As their name suggests, Hsps are induced by a cell in response to heat shock.
25 Three major families of Hsp have been identified and are called Hsp60, Hsp70 and Hsp90 because of their respective molecular weights of about 60, 70, and 90 kD.
Many members of these families c~hcP~IIIPn~ly were found to be induced in response to otller stressful stimuli, such as those mentioned above.

Wo 95n4923 r~
21 85b51 Stress proteins are found in all ,UlUI~lyUI~ and eukaryotes and exhibit a remarkable level of ~ ~ uluLiu~ -y UIl,_. V~l~iUII. For example, DnaK, the Hsp7û
from E. coli has about 5û% amino acid sequence identity with Hsp7û proteins fromeukaryotes (Bardwell et al. (1984) PrQc. ~lAtl ArA~I rad. 81:848-852). The Hsp60 and 5 Hsp90 &milies also exhibit similarly high levels of intrAf~miliAl ~ulla~l VclLiul, (Hidcey et al. (1989) Mol. Cell Biol. 9:2615-2626;Jindal (1989) Mol. Cell. Biol. 9:2279-2283). In addition, it has been discovered that the Hsp-60, Hsp-70, and Hsp-9û
families are composed of proteins that are related to the stress proteins in sequence, for example, having greater than 35% amino acid identity, but whose expression 10 levels typically remain unaltered under conditions stressful to the host cell. An example of such a protein includes the .ull~iluli~ .Iy expressed cytosolic protein Hsc 70 to which is related in amino acid sequence to the stress-induced protein Hsp 70. Accordingly, it is ~ul~L~Illpla~d the definition of stress protein, as used herein, embraces other proteins, muteins, analogs, and variants thereof having at least 35'~
to 55%, preferably 55% to 75%, and most preferably 75% to g5'7. amino acid identity with members of the three families whose expression levels in a cell are stimulated in response to stressful stimuli.
The term "peptide", as used herein, is llnrlPrctrlod to mean any amino acid 20 sequence that is present in a eukaryotic cell infected with an intrArPlllllAr pathogen but which is not present in a similar cell when the cell is not infected with the same pathogen. The definition embraces peptides that not only originate from the pathogen itself but also peptides which are ~yllLll~ l by the infected cell in response to infection by the intrArPll11lAr pathogen.
The term "i..,...~ stress protein-peptide complex", as used herein, is lln~iPrctrod to mean any complex containing a stress protein and a peptide that is capable of indudng an immune response in a mammal. The peptides preferably are non cova~ntly a~sociated with the stress protein. The complexes may include, w095l24923 P~l/.l~.._. Il ~18565~ --but are not limited to, Hsp60-peptide, Hsp70-peptide and Hsp90-peptide complexes. In a preferred aspect of the invention a stress protein belonging to the Hsp90 family, namely gp96 can be used to generate an effective vaccine containing a gp96-peptide complex. Since the peptides can be rliccrriAtr~ from the complex in S the presence of ATP or low pH potentially antigenic peptides can be isolated from cells infected with a ~ ele- ~ d in trArP~ Ar pathogen~ r~ y, the antigenic ~ for potentially any intrArp~ lAr pathogen of interest can be identified readily using the mPthr.~lrlr,giPc described herein.
The term "cytotoxic T cell", as used herein, is ~n~lPrctrrci to mean any T
IyIllpllo~y~ expressing the cell surface gly~u~IuL~ marker CD8 that is capable of targeting and lysing a target cell which bears a class l l ~ ;ly complex on its cell surface and which is infected with an intrArPIl~lAr pathogen. The term "cytotoxic T cell response" is lm~lPrc~rrrl to mean any cytotoxic activity that is 15 mediated by cytotoxic T cells.
As used herein, the term "intracellular pathogen" is understood to mean any viable organism, including, but not limited to, viruses, bacteria, fungi, protozoa and intrArPll--lAr parasites, capable of existing v~ithin a mAmmAliAn cell and caus~ng a 20 disease in the mammal.
In a preferred aspect of the invention, the stress protein-peptide vaccines have particular utility in treating human diseases caused by intrArPlllllAr pathogens. It is ~ l,-I-YI that the vaccrnes developed using the principles 25 described herein will be useful in treating diseases of other mammals, for example, farm animals including: cattle; horses; goats; sheep; and pigs, and household pets including: cats; and _gs.

wo ssn4s23 ~ 'Q~

Vaccmes may be prepared that stimulate cytotoxic T cell responses against cells infected with viruses includmg, but not limited to, hepatitis type A, hepatitis type B, hepatitis type C, influenza, varicella, adenovirus, herpes simplex type I
(HSV-I), herpes srmplex type 11 (HSV-II), rinderpest, rhinovirus, echovirus, 5 rotavirus, I~lv_alvIy synctial virus, papilloma virus, papova virus, cytomegalovirus, e l~ vvil u~, arbovirus, huntavirus, coxsachie virus, mumps virus, measles virus, rubella virus, polio virus, human immI-nr,~lPfiriPnry virus type I (HIV-I), and human immIlnrrlPfiriPnry virus type Il (HIV-II). Vaccines also may be prepared that stimulate cytotoxic T cell responses a~ainst cells infected witl~
10 intrArPl~ r bacteria, including, but not limited to, MycobncteriQ, RickettsiQ, MycoplasmQ, Neissetia and Legiot~ella. Vaccines also may be prepared that stimulate cytotoxic T cell responses against cells infected with intr~rPlllllar protozoa, including, but not limited to, Letshmania, Kokidioa, and T~." . Vaccines may be prepared that stimulate cytotoxic T cell responses against cells infected with 1~ i.,l,,., ~'1: ,1,., parasites mcluding, but not limited to, Chlnmydin and l~ickettsia.
In another preferred ~, . ,h". I;, . ,~ " I of the invention, the stress protein-peptide vaccme may also contain a ~ a~ u~i~dUy effective amount of a cytokine.
As used herein, the term "cytokine" is meant to mean any secreted polypeptide that 20 influences the function of other cells mediating an immune response. Currently, preferred cytokines include: interleukin-1a (IL-la), interleukin-l~ (IL-I~), interleukin-2 (IL-2), interleukin-3 (IL-3), interleukin-4 (IL-4), interleukin-5 (IL-5), interleukin-6 (IL-6), interleukin-7 (IL-7), interleukin-8 (IL-8), interleukin-9 (IL-9), interleukm-10 (IL-I0), interleukin-11 (IL-II), interleukin-12 (IL-12), interferon 2~ (IFN ), interferon ,f~ (IFN,~), interferon r, (IFN~), tumor necrosis factor (TNF) ), tumor necrosis &ctor ~ (TNF,f~), granulocyte colony StinnIlIatin~ &ctor (G-CSF),granulocyte/Illa~lvlvlId~;e colony 5tim~ tin~ factor (GM-CSF), and I, ,. . .ir- .. " ,~
growth factor ~ (TGF-~). It is ~ that other but as yet UllVi~ U ~
cytokines m y be effective in the mvenvion~ In addition, ~ullvt:Illiull~l antibiotics

3 P`'"`~
21 8~651 may be co-a~l.il.;.,t~.~d with the stress protein-peptide complex. The choice of a 6uitable antibiotic or a ~ thereof, however, will be dependent upon the disease in question.
It has been discovered that the vaccine stimulates the cytotoxic T cell response via the major 1.;~1,,1 ...,,I.,.I;hility complex (MHC) class I cascade. Thus, it is ....,1~,..1.l,.l_.l that the cytotoxic T cell response may be enhanced further by co-~.l,.,;.,;~l~,;.,~thevaccmewitha 11,~".1.~"1;1 ,lllyeffectiveamountofoneormoreofcytokines that potentiate or modulate cytotoxic T cell responses.
Anotherpreferred ~,l,l-v,l;l"~"l, the invention provides a method for ctimlllAtin~ in a mammal a cellular immune response, specifically a cytotoxic T cell response, against cells mfected with a ~l~a~le-l~d intrAr~ -lAr pathogen. The method involves ~,1,.,;,-;~1~,;-,~ to the mammal a vaccine made in accordance with lS the principles disdosed herein in an amount sufficient to elicit in the mammal a cytotoxic T cell response against the pl~a~ d intr;~r~ Ar pathogen.
The vadne may be a.lll lil lia ~:1 ~1 lul ulullrldL ~i~dlly to a mammal in order to stimulate in the mammal a cytotoxic T cell response that prevents âllhcr~l~Pnt 20 infection of the mammal by the intrAr~ llAr pathogen. All~ dliv~ly, the vacdne may be adlllil.ia~l~ 11-~,,-l-~,-l;l ,-lly to a mammal having a disease caused by an ;ntrAr~ llAr pathogen. It is f~-nt~mrlA~r~ that the vaccine may stimulate a cytotoxic T cell response against cells presently infected with the intr~r~ lAr pathogen.
2~
The dosage and means of ,1l-";, ;~ of the family of stress protein-peptide vaccines necessarily will depend upon the nature of the complex, the intrAr/~lllllAr pathogen and the nature of the disease in question. The complex should be ad~ ;al~ in an amount sufficient to init~ate a cytotoxic T cell wo 95/24923 P~IlU.,. .'^~
~8565~

response against the intrArP~ lAr pathogen. In general, the amount of stress protein-peptide complex a~ L~I~d may range from about 0.1 to about 1000 Illi~lU~;lalllS of complex/kg body weight of the mammal/immllni7Atinn and preferably in the range of about 0.5 to 100 Illi~lU~;ldllL~ of complex/kg body weight S of the mammal/;"~ The recipient preferably should be vaccinated four times at weekly intervals. If necessary, the responses may be boosted at a later date by ~ .,l ' ' " of the vaccine. It is ~ 1, however, that the optimal dosage and V~ ldliUII schedule may be determined empirica~ly for each stress protein-peptide vaccmecomplex by an artisan using CuIlv~ iulldl techniques 10 well known in the art.
In another aspect, the invention provides a variety of mP~hmlnln~iPC for preparing ~UIllllltl~idlly available amounts of the stress-protein peptide vaccines which when ad,.,il.i~ .~ to a mammal induce in the mammal a cytotoxic T ceL
15 response against cells infected with a ~,~æle.L~:d antigen. In one approach, the stress protein-peptide complex may be harvested using conventional protein FllrifirA~inn mP~hn~inlngiPc from a sample of tissue, an isolated cell or immortalized ceL line infected with the lu~el~I~d in~rA~PlllllAr pathogen, or an isolated cell or rmmortalized cell hne transfected with, and expressing a gene encoding a 20 ~ le.Le l antigenic ~ l The purified complex 5llhcPqllPn~ly may be stored or combined with a ~ lly acceptable carrier for ~ a ir,n aS
a vaccine.
Al~ d~iv~ly~ the stress protein-peptide complex may be prepared by 25 , ~. ., .~l; l " l; . ,~ a potentially antigenic peptide and a stress protem in vitro. For example, the antigenic peptide may be eluted from either a purified stress protein-peptide complex or a MElC-peptide complex using mP~hn-ir,lngiPs well known in the art. Specifically, the peptides may be eluted from the stress protein-peptide complex by incubat~g the complex in the presence of ATP or low pH.

WO 95/24923 ~ _ IIU~

AlLelllalivel~, the peptides may be eluted from the MHC-peptide complex by mcubating the complex in the presence of trifluoroacetic acid (TFA). The resulting peptides may be purified by reverse phase HPLC and their amino acid sequences d.~ ,P l by standard protein c~ n-l~ Peptides of defined 5 sequence then may be ~y~ eai~ using ul,ve~ al peptide synthesis mPthl.finl, giPc Stress proteins may be purified directly from cells naturally expressSng the stress proteins. AlLelllali~ely, l~ stress proteins, including non native forms, truncated analogs, muteins, fusion proteins as well as other constructs capable of mimicking the peptide binding and i~
10 propertiesofstressproteinsmaybeexpressedusmgconventiona~l~-."l,;.,,.,.l DNA mPth~ giPc For example, a ~ "".l,;,.",.l stress protein may be expressed from I rl 1 ll l ,hi .~ ~ ,I DNA in either a eukaryotic or prokaryotic expression system and purified from the expression system. The two purified .., ,.I,,,,,~,,l~ then may be combmed in ~z~ to generate a synthetic and completely defined stress protein-peptidecomplex. The ;-,.. ",-y, ll;. ily and specificity of the 1- l ,,-,.l.;.. ,,l-complexes sllhcPqllPntly may be assayed in vitrû and in vivo to identify useful candidate complexes that stimulate cytotoxic T cell responses against a preselected intri~Pll~ r pathogen. Once identified, the synthetic complexes may be prepared on any scale, stored as is, or combined with ph~rm:lrPI ti~lly acceptable carriers for 20 ~,I"".,i~lIr i.", to rnarnmals.

- ~ WO 9~24923 Brief Pescription of the Drawings The foregoing and other objects and features of the invention, as well as the 5 invention itself, may be more fully - rlr~ PrC~ood from the following description, when read together with the d~u~ uculyillg drawings, in which:
Figure 1 shows antigen specific cytotoxic T cell activity of a,ul~lloLyL~a derived rTom mice i,..,.."..i,P.I with a gp96-peptide complex harvested from 10 BALB/c fibroblasts transfected with the ...~.leul,.uL~i., (NP) gene from the PR8 inrluenza virus. The cytotoxic activity was assayed by the release of 51 Cr fromBALB/c fibroblasts expressing the NP gene (filled circles), bALB/c fibroblasts expressing the NP gene but treated with the anti-MHC type I antisera K44 (empty circles) and rTom the syngeneic non-NP tTansfected cell line 5117 (asterisks).
Figure 2 shows antigen specific cytotoxic T cell activity of sp~enocytes derived from mice i .. " . .: ~ with gp96-peptide complex harvested from SV40 r", ".~1 SVB6 cells. The cytotoxic activity was assayed by the release of 5ICr rTOm SVB6 cells (filled circles) and rTOm a non-SV40 tr;lncfnrrn~l syngeneic cell line, 20 MCA (empty circles).
Figure 3A-3D shows antigen specific cytotoxic T cell activities of splenocytes derived from two mice i " .. l . l . ": P~1 with a r~nncf i~ Pfi Hsp70-peptide complex where the peptide has the sequence SLSDLRGYVYQGL (SEQ. ID. NO. l ). Prior to 25 p~.r....-.;,.~theassay,thesplenocytesderivedrTomeachmousewerestimulated either once (3A and 3C) or twice (3B and 3D) m vitro with lethally ilTadiated ceDs tTansfected with, and expressing the peptide SLSDLRGYVYQGL (SEQ. ID. NQ 1).
Cytotoxic acti~ity wa~ assayed by the release of 51Cr rTom EL4 cells expressing the wo 95124923 T~
21 ~5~51 peptide (filled triangle) and from EL4 cells not expressing the peptide (empty triangles).

~ WO 95124923 P~,l/LV~
21 8~65~

DPt~ilptl Description.
The invention is based on the discovery that a stress protein-peptide S complex when isolated from a eukaryotic ceD infected with a ~-.J~
;"1,~., '~ .1_. pathogen amd then ~. I.. .; .. ~I . ~1 to a marmnal can stimulate a cytotoxic T ceD rffponse directed against ceDs infected with the same pathogen.
This discovery provides a significant advance to the field of vacdne d~ . ~Iulu~
In accordance ~vith the invention, the afu- r ~ discovery is exploited to provide a family of vaccines which may be used to immunize mammals against diseases caused by intrA~P~ r pathogens. In principle, the vaccines can be prepared againstany ;"~ ,1", pathogen of interest, for example: viruses;
bacteria; protozoa; fungi; or i rl~rArpl~ ar parasites. Generic ~ d~ useful 15 for preparing vaccines against aD of these classes of pathogens are discussed in detail I ~ ~
As wiD be ~I,U~ by those skilled m the art, the stress protem-peptide vaccmes described herein have several adv~.L~ a over the vaccines currently 20 available. First, the stress protein-peptide vaccines provide an alternative approach for ~ ;..g ceDular immunity and obviate the use of intact intrA~PlllIlAr (AttPmlAtP~l or otherwise) pathogens. Second, since the vaccines do not-contain intact organisms this reduces the risk of causing the disease the vaccine was designed to induce immunity agamst. Third, the vaccines described herein are 25 ideal for inducing immune responses against either defined antigenic rl~l ~" " i ",.. , isolated from an " ' pathogen or as yet undefined antigenic ~ ., .,; "" "
FulLllellllul~:, vaccines may be prepared that are effective against pathogens that normally evade the immune system by evolving new antigenic coat proteins, i.e., the ir~fluenza virus. Fourth, vaccmes of this type ~ m principle b~ prepared wo 9sn4923 ~ ~ 8 ~ 6 ~

agarnst any in~rArr~ llAr pathogen of interest. Fifth, the vaccines may be prepared synthetically using the mPthndrlrgir-c described heremafter thereby providimg completely defined vaccines that are suitable for ~ to humans.
It is ",. ~ 1 that the vaccines may be ad~ iaLeled either lululullylc~ allyOrl~ ""I;~ y. Whenad~ islelei~lu~ a~ dllythe vaccine may stimulate in the mammal a cytotoxic T cell response that permits thevaccinee to resist b' ~ infection by the in~rArr-lllllAr pathogen. Alternatively, when allllillislelrd Illel~ u~ lly the vaccine may stimulate in the mammal a 10 cytotoxic T cell response against a pathogen which is presently infecting and causing disease in the mammal.
The specific ~ -l of the vaccine that induces in the recipient a specific cytotoxic T cell response against the pathogen is a stress protein-peptide complex.
15 The peptide may be any amino acid sequence that is present in a eukaryotic cell infected with an i. ,1, ,",/,ll- ,1,. . pathogen but which is not present when such a cell is not mfected with the same pathogen. This includes peptides that not only originate from the pathogen itself but also are ~yllll~e~i~el by the infected cell in response to infection by the intrArr-lllllAr pathogen.
The; " ., ..., ..r,~,.. I i. complexes may be purified from any eukaryotic cell,including: whole tissues; isolated cells; and immortalized eukaryotic cell linesinfected with the in~rArrlllllAr pathogen. The complexes may be purified by using conventional protein purification techniques well known in the art. For example, it 25 is ~ l that an illllllul-O~el,i complex capaWe of c~imlllA~in~ a cytotoxic T
cell response against the influenza virus may be harvested from a eukaryotic cell line that is infected with the influenza virus.

~ WO 95124923 P_ll~J~ _. . 11 ~ 1 8565 ~

In addivion, it has been found that the peptide can be eluted from the st}ess protein-complex either in the presence of ATP or low pH. Neither the peptide northe stress protein individualIy are effective at inducing a cytotoxic T cell response These ~ conditions, however, may be exploited to isolate peptides from 5 infected cells which may contain potentially useful antigenic ~ " ~ c Once isolated, the amino acid sequence of each antigenic peptide may be fl~ rmin~
using .v~ Liol~al amino acid c~~ n~ 'nncp~ n~ly~ the antigenic ~ 1~"..;."..,1`. for potentially any il~l, ,.. .~11..1,., pathogen of interest can be identified readily using the mPth~ giPC described herem. As discussed in detail 10 hereinafter, this property may be exploited in the ~ Livll of completely synthetic vaccines.
Similarly, it has been found that potentially i, . . .~ peptides may be eluted from MHC-peptide complexes using techniques well known in the art. See for example, Falk et al. (1990) Nature 34~3:248-251; Rotzsche et al. (1990) Nature 348:252-254; Elliott et al. (1990) Nature 348:195-197; Falk et al. (1991) Nature351:290-296, Demotz et al. (1989) Nature 334:682-684; Rotzsche et al. (1990) Science 249:283-287, the disclosures of which are incorporated herein by reference.
Although the peptides eluted from the MHC complexes may define a potentially 20 protective anvigenic rll~.ormin~n~, it is ~ L~l that ~ of the isolated peptide in a conventional subunit vaccine may be ineffective at ~
a cytotoxic T cell response in the recipient. ~ l ly~ it is ~ l that the peptides eluted from MHC-peptide complexes may be ,~ . ,.)~1;1 "1.~.1 with astress protein, using the methodologies described herein, thereby to generate a 25 sLress protein-peptide complex effective at c~inn~ in~ a cytotoxic T cell response capable of targeting and lysing cells expressmg the antigenic peptide.
Stress proteins useful in the practice of the instant invention may be defined as any cellul ~ prote~ that satisfies vhe following criteria. It is a protein whose wo 9~/24923 1 ~ I/.J~ . . I I
2l85651 intr~r~ r ~ .... increases when a cell is exposed to a stressful stimuli, it is capable of binding other proteins or peptides, and it is capable of releasing the bound proteins or peptides in the presence of adenosine l,i~ ale (ATP) or low pH.

The first stress proteins to be identified were the heat shock proteins (Hsp).
As their name implies, Hsps are byl~ a;~e~ by a cell in response to heat shock To date, three major families of Hsp have been identified based on molecular weight.
The families haYe been called Hsp60, Hsp70 and Hsp90 where the numbers reflect 10 the r~ ul~ll: molecular weight of the stress proteins in kD. Many members of these families ~ lly were found to be induced in response to other stressful stimuli including, but not limited to, nutrient deprivation, metabolic disruption, oxygen radicals, and infection with intracellular pathogens. See for example: Welch (May 1993) Scientific American 56-64; Youn~(l990) Annu. Rev. Immunol. 8:401-420; Craig (1993) Science 260:1902-1903; Gething et al. (1992) Nature 355:33-45; and Lindquist et al. (1988) Annu. Rev. Genetics 22:631-677, the disclosures of which are incorporated herein by reference. Aordingly, it is ~ ,pl~-L~l that stress proteins belonging to all three families may be useful in the practice of the inst~nt invention.
The major stress proteins can ~rrllmlll~tr- to very high levels in stressed cells, but they occur at low to moderate levels in cells that have not been stressed.
For example, the highly inducible m:lmm~li; n Hsp70 is hardly detectable at normal ul~ but becornes one of the most actively byll~ll~l proteins in the cell upon heat shock (Welch et al. (1985), 1. Cell. Biol. 101:1198-1211 ). In contra8t~
Hsp90 and Hsp60 proteins are abundant at normal ~ UI~ in most, but not all, ~ cells and are further induced by heat (Lai et al. (1984), Mol. Cell.
Biol. 4:2802-10, van Be~gen en H~ uw~-- et al. (198;, Genes Dev. 1:525 531).

Wo 95124923 r~,l,.,~,. . Il 2~ ~5651 Stress proteims are among the most highly conserved protems in existence.
For example, DnaK, the Hsp70 from E. coli has about 50% amino acid sequence identity with Hsp70 proteins from eukaryotes (Bardwell et al. (1984) Proc Natl.
Acad. Sci. 81:848-852). The Hsp60 and Hsp90 families also show similarly high levels of i "~ UI~Se'l valiull (Hickey et al. (1989) Mol. Cell Biol. 9:2615-2626;
Jindal (1989) Mol. Cell. Biol. 9:2279-2283). In addition, it has been discovered that the Hsp60, Hsp70 and Hsp90 families are composed of proteins that are related tothe stress proteins in sequence, for example, having greater than 35% amino acididentity, but whose expression levels typically remain unaltered under conditions stressful to the host cell. An example of such a protein includes the ~u.,~iLuLively expressed cystolic protein Hsc 70 which is related in amino acid sequence to thestress-induced protein Hsp 70. It is, therefore, ~ that the definition of stress protem, as used herein, embraces other proteins, muteins, analogs, and variants thereof having at least 35% to 55%, preferably 55% to 75%, and most preferably 75% to 95% ammo acid identity with members of the three families whose expression levels in a cell are enhanced m response to a stressful stimulus.
The ~uuliri~aLiull of stress proteins belonging to these three families is described below.
The;, .. " ~ stress protein-peptide complexes of the invention may include any complex containing a stress protein and a peptide that is capable ofinducing an immune response in a mammal. The peptides preferably are non covalently associated with the stress protein. Preferred complexes may include, but are not limited to, Hsp60-peptide, Hsp70-peptide and Hsp90-peptide complexes.
25 For example, a stress protein called gp96 which is present in the endoplasmicreticulum of eukaryotic cells and is related to the ~ylul~la~ Hsp90s can be usedto generate an effective vaccine containing a gp96-p~ptide complex.

W0 9~24923 E ~
`~8!~

Another family of low molecular weight heat shock proteins has now been identified and is called Hsp 25/Hsp 27. The ~ - of these proteins is discussed below. It is ~ that these low molecular weight proteins may also have utility in the instant invention.

It has been discovered also that the stress protein-peptide complexes of the invention can be prepared from cells infected with an intracellular pathogen as well as cells that have been ~ r ." "r~1 by an i " I, ,~ pathogen. For example, i."".~"~ ir stress protein peptide complexes may be isolated from eukaryotic 10 cells Ll~-,b~u~ l with a l ~ g virus such as Sv4o~ see below~
In a preferred aspect of the invention, the purified stress protein-peptide vaccines may have particular utility in the treatment of human diseases caused by r pathogens. It is ~ e-l, however, that the vaccines developed 15 using the principles described herein will be useful in treating diseases of other mammals, for example, farm animals including: cattle; horses; sheep; goats; and pigs, and household pets including: cats; and dogs, that similarly are caused byintr:lrrll~ r pathogens.
In accordance with the methods described herein, vaccines may be prepared that stimulate cytotoxic T cell responses against cells infected vvith viruses including, but not limited to, hepatitis type A, hepatitis type B, hepatitis type C, influen~a, varicella, adenovirus, HSV-I, HSV-II, rinderpest II.il,uvi.uu~, echovirus, rotavirus, respiratory synctial virus, papilloma virus, papova virus, cytomegalovirus, echinovirus, arbovirus, huntavirus, coxsachie virus, mumps virus, measles virus, rubella virus, polio virus, HIV-I, and HIV-II. Similarly, vaccines may also be prepared that stimulate cytotoxic T cell responses against cells infected with intr~r~ r bacteria, including, but not limited to, M~ ' ~, iu, Rickettsia, M; , ' Neisseria and Legionella. In addition, vaccines may also be ~ wo 95J24923 prepared that stimulate cytotoxic T cell responses against cells infected with i,.l,,.,.~ll.,l,., protozoa,including,butnotlimitedto,Leis71mania,Kokidioa,and ~13-t Fu~ lul ~, vaccines may be prepared thdt stimulate cytotoxic T
cell responses ag~unst cells infected with in~r~rr-lllllAr parasites including, but not 5 Lmited to, Chlamydia and Kickettsia.
1. rlUUa~:~liUI- of infected eukdryotic cells.
As v~ill be ~ u. t:Lidl~l by those sl<illed in the art, the protocols described lO herein rnay be used to isolate stress protern-peptide complexes from any eukaryotic cell, for example, tissues, isolated cells or immortalized eukaryotic cell linesinfected with a ~ - pathogen.
When illllllU~ L,e~:l animal cell lines are used as a source of the stress 15 protein-peptide complex it is of course important to use cell lines that can be infectecl with the pathogen of interest. In addition, it is preferable to use cells that are derived from the same species as the intended recipient of the vaccine.
For example, in order to prepare a stress protein-peptide complex for 20 ~ ;r.~ to humans that may be effective against HIV-I, the virus may be propagated in human cells which include, but are not limited to, human CD4+ T
cells, HepCi2 cells, and U937 lululllullu~yLi~ ceLs. In order to prepare a stress protein-peptide complex for ~ to humans mat may be effective against HIV-II, the virus may be lululu~ ;l m, for example, human CD4+ T cells.
25 Similarly, influenza viruses may be ~u.upa~aled in, for example, human fibroblast cell lines and MDCK cells, and I~ ub~ l id may be cultured in, for example, human Sch~aan cells.

WO 95/24923 r~
218~5~

If the i,,l, ~. ~11 ,1,.. pathogens do not Iyse the infected cells then the infected cells are cultured under the same conditions as the normal uninfected cells. Forexample, ~ uba~ lia may be IJlu~ua~d~l in nerve cultures of the sensory ganglia of newborn Swiss white mice. The nerve cells are cultured in a growth medium containing 70% Dulbecco modified Eagle minimal essential medium (DMEM) with 0.006% glucose, 20% fetal calf serum, 10% chicken embryo extract and cytosine zlrzlhinoci~lp After eight to ten days, the cultures are inoculated with 5-8x106~I~.uba~ ia isolated from fresh nodules of untreated l~lvllla~uu~ leprosy patients. The infected cells may be cultured at 37C, for up to 6 weeks, after which the infected cells are harvested and the stress protein-peptide complexes isolated.
See for example, Mukherjee et al. (1985) ~. Clin. Micro. 21:808-814, the disclosure of which is ill~ulluulal~l herein by reference.
If, on the other hand, the host cells are Iysed by the pathogen of interest (as in the case of influenza virus) the cells may still be grown under standard conditions except the cells are washed and harvested just prior to Iysis of the host cell. For example, during the ~ ;ri~ of stress protein-peptide complexes from influenza infected cells, fibroblasts (or other cell types) are infected for 1 hour at 37C with 5 1D plaque forming units (PFU) of virus per cell. The infected cells may be cultured in plain DMEM medium for 24 hours at 37C. After 24 hours the cells are washed and harvested prior to Iysis. The stress protein-peptide complexes may be isolated using the ~- U~lUlt~ set forth below.
In addition, when the gene encoding a particular antigenic ~ r ~ " ~ has been identified, the gene of interest may be transfected and expressed in an immortalized human or other m:~mm~ n cell line usimg techniques well known in the art. See for example "Current Protocols in Molecular Biolo~v" (1989), eds.
Ausubel FM, brent R, Kingston RE, Moore DD, Seidman JG, Smith ~A and Struhl K, Wiley 1~ , .. i.... P, th~ disclosure of which is ill~ull~ulaL~I by reference herein. The ~ wo95r24923 r~l,u~
2 1 8565 ~

transfected cells may be g}own under standard conditions and the complexes isolated~ ly, I~. P~ aLiullofStressProteinsandl~ oL~ ;rstressProtein-~eptide 5 complexes.
Methods for preparing Hsp70-peptide complexes, Hsp90-peptide complexes, gp96-peptide complexes, Hsp70, Hsp25/Hsp27, and Hsp60 are set forth below.
(a) Pu-iri~ aliull of Hsp70-peptide complexes.
A pellet of rnfected cells is rrcllcppnr~r~l rn 3 volumes of IX Lysis buffer consisting of 5mM sodium phosphate buffer (pH7), 150mM NaCI, 2mM CaC12, 15 2mM MgC12 and ImM phenyl methyl sulfonyl fluoride (PMSF). The pellet is sonicated, on ice, until >g9% cells are Iysed as judged by llli.l u,.ului. P~ - . " .., ;....
Allt llldliv~ , the cells may be Iysed by mr-rhanirRI shearing. In this procedure, the cells are I~uD,u~l.dP i in 30mM sodium l,i-.lll,oll~,le pH 7.5, ImM PMSF, incubated on ice for 20 min. and then homogenized in a dounce holl-o~ l until >95% cells 20 are Iysed.
The Iysate is ~ , ;r- 1~;~1 at lOOOg for 10 minutes to remove unbroken cells, nuclei and other debris. The ~U~ llL from this rPntrir~l~Atirn step is then ,~1 at lOO,OOOg for 90 minutes.
The ~U~I~L~I-l is mixed for 2-3 hours at 4cC with Con A Sepharose "~1~, ;1;1.. ~ ~ P. I with PBS containing 2mM Ca2+ and 2mM Mg2+ . When the cells are Iysed by mrrhAnirAl shearing, the ~ .l is diluted with equal volume of 2X
Lysis Buffer ~_ re ~,- u~ Th~ the slurry is packed mto a column and wo 95/24923 r washed with lX lysis buffer. The material that does not bind is dialyzed for 36 hours (three times, 100 volumes each time) against lOmM Tris-Acetate pH 7.5, 0.1mM EDTA, lOmM NaCl, ImM PMSF. The dialyzate is ~ ; t- ~,~ for 20 min. at 17,000 rpm (Sorvall SS34 rotor~ and the resulting bUUt~ Ldlll applied to a Mono Q
FPLC column (Pharmacia) r~ in 20mM Tris-Acetate pH 7.5, 20 mM NaCl, O.lmM EDTA and 15mM 2-llle~ -l Then the proteins are eluted with a 20 mM to 500 mM NaCl gradient. ~he fractions are ~llal~eli~èd by sodium dodecyl sulfate-poly~l-,ylc,ll ide gel elé~lulul~ule~;s (SDS-PAGE) and immlln~ t~in~usinganaplulu,u,i~l~eanti-Hsp70antibody(suchascloneN27F3-4 frûm StressGen).
Tlle fractiûns that are strûngly ill ~ ullul ~ i ve with the antibûdy are pûûledandtheHsp70-peptidecûmplexes~ d~e~with ~ "i.""sulfate.
The complex is ~ule iluita~ed in the 50%-70% Allllllllll;~llll sulfate cut. The protein pellet is harvested by . r. .l - i r- ~ at l7,000 rpm (5534 Sûrvall rotor) and washed with 70% A 111111.111' 1 l " sulfate. Then the pellet is solubilized and the residual A 1111111111;11111 sulfate removed by gel filtration on a Sephadex~ G25 column (Pharmacia).
The Hsp70-peptide complex can be purified to apparent hvlll~ellei~y using this method. Up to Img of Hsp70-peptide complex can be purified from Ig of cells/tissue.
(b) ruliL~ iu~ of Hsp7û.
The Hsp70 polypeptide may be purified from the Hsp70-peptide complex byATPagarose.1""".~ y. Seeforexample,Welchetal.(1985)Mol ~ell.
. 5:1229~ the disclosure of which is iu~u~luulA~ed herein by reference. Briefly, Mgcl2isaddedtothepreviouslyisolatedcomplextoafinalllllllrllllAl~ of 3mM~

~ wo 95124923 r~
21 ~5~51 Then, the complex is applied to an ATP agarose column (Sigma Chemical Co.) .I.,ilil,-,.l~lin20mMTris-Acetate(pH7.5),20mMNaCl,O.lmMEDTA,15mM2-l, 3mM MgC12. The column is washed è~lellsivæly with the ~qllilihrA~ n buffer containing 0.5M NaCI, and then washed with buffer without 5 the NaCl. Then the Hsp70 is eluted from the column with eu,uili11~,Liù.l buffer containing 3mM ATP (Sigma Chemical Co.).
(c) Puuiri~c~Liul~ of Hsp90-peptide complexes.
A pellet of infected cells is rPcllcpPnrlP~l in 3 volumes of IX Lysis buffer consisting of 5mM sodium phosphate buffer (pH7), 150mM NaCI, 2mM Ga2, 2mM MgC12 and ImM PMSF. The cell pellet is sonicated, on ice, until >95% cells are Iysed as 3udged by IlliLlUs~U~ e~ l l AlLel .laLi v ely, the cells may be Iysed by . "~1 ,," , ;. ,.l shearing, as before.
The Iysate is centrifuged at lOOOg for 10 minutes to remove unbroken cells, nuclei and other debris. The :IU,Ut-lll~ from this ~llll iru~aLion step "lly is recentrifuged at lOO,OOOg for 90 minutes.
Then, the ~ ,l is mixed for 2-3 hours at 4C with Con A Sepharose pq~ n7rA~oll with PBS containing 2mM Ga2+ and 2mM Mg2+. When the cells are Iysed by Illf~. l.,...;. ,.l shearing, the ~UIJelll.lL~lll is diluted with equal volume of 2X
Lysis Buffer before l~lu~eèdiulg. Then, the slurry is packed into a column and washed with IX Iysis buffer. The material that does not bind is dialyzed for 36 hours (three times, 100 volumes each time) against 20mM sodium phosphate pH
7.4, ImM EDTA, 250mM NaCI, ImM PMSF. The dialyzate is ~lllliru~ed at 17,000 rpm (Sorvall SS34 rotor) for 20 min. The resulting :~ul~el l~dlcllll is applied to a Mono Q FPLC column (Pharmacia) -~ with lysis buffer and the bound proteins eluted with a salt gra:ient of 200mM to 600mM NaCI.

wo 95/24923 P~
2t ~5651 The eluted fractions are analyzed by SDS PAGE and the Hsp90 complexes identified by ....- - ....~l-.1~11;..g using an anti-Hsp90 antibody (for example, 3G3 from Affinity r ~ ) Hsp90 can be purified to apparent ~ O ~ using this procedure. A~ , 150-200 llg of Hsp90 can be purified routinely from Ig of cells/tissue.
(d) I`~uir.~liu-, of Fo9~peptide complexes.
A pellet of infected cells is r~CllcpPnrl~ in 4 volumes of buffer consisting of 30mM sodium 1.;~ ' e buffer (pH7.5) and l mM PMSF and the cells allowed to swell on ice for 20 min. The cell pellet then is h - ~O- ~ -l in a Dounce h~nn~g~ni7~r (the ~ clearance of the h~ o~ , will vary according to each cells type) on ice until >95% cells are Iysed.
The Iysate is . . ~ .; r. .~, -.1 at lOOOg for 10 minutes to remove unbroken cells, nucleiandothffdebris. The~ fromthis~ t~l;r';~"stepthenis L,. .1 at lOO,OOOg for 90 minutes. The gp9~peptide complex can be purified either from the lOO,OOOg pellet or from the When purified from the ~ l, the ~u~ u-l is diluted with equal volume of 2X Lysis Buffer and the ~U~ mixed for 2-3 hours at 4C with Con A Sepharose Pqllil ' ~ with PBS containing 2mM Ca2+ and 2mM Mg2+. Then, the slurry is packed rnto a column and washed with IX Iysis buffer until the OD2go drops to baseline. Then, the column is washed with 1/2 column bed volume of 10% a-methyl mannoside (o-MM) dissolved in PBS containing 2 mM Ca2+ and 2mM Mg2+, the column sealed with a piece of parafilm, and incubated at 37C for 15 min. Then the column is cooled to room ~ UI~ and the parafilm removed from the bott ~m of the column. Five column volumes of the c~-MM buffer are WO 9~124923 P~,.,.. . ll 21 8565~

applied to the column and the eluate analyzed by SDS-PAGE. Typically the resulting material is about 60 - gsæ pure, howeYff this depends upon the cell type and the tissue-to-lysis buffer ratio used. Then the sample is applied to a Mono Q
FPLC column (Pharmacia) P~ l', il il ., ~ l ~ with a buffer contaming 5mM sodium5 phosphate, pH7. The proteins then are eluted &om the column with a 0-IM NaCI
gradient and the gp96 fraction elutes between 400mM and 550mM NaCI.
This procedure, however, may be modified by two additional steps, used either alone or in ~ , to ~ul.,;,L~.~Lly produce apparently h~m~gpnly)lls IO gp96-peptide complexes. One optional step involves an ~ "~ sulfate lul e i~alàLiull prior to the Con A purification step and the other optional step involves DEAE-Sepharose l,u,iri.atiu-- after the Con A purification step but before the Mono Q FPLC step.
In the first optional step, the ~ resulting from the I00,000g . ,.l . ir..~ . step is brought to a final ~ of 50% ~ sulfate by the addition ~ sulfate. The ~ sulfate is added slowly while gently stirring the solution in a beaker placed in a tray of ice water. The solution is stirred for about 2 to 12 h. at 4C and the resulting solution ~-on~ri r~ 1 at 6,000 rpm 20 (Sorvall SS34 rotor). The ~ l - ,l resulting from this step is removed, brought to 70% ,. " . " ,~ , ., sulfate saturation by the addition of a~ ul~iulll sulfate solution, and ~ell~ u~l at 6,000 rpm (Sorvall SS34 rotor). The resulting pellet from this step is harvested and suspended in PBS containing 70% :~mm, sulfate in order to rinse the pellet. This mixture is - ~ ir"~,- -l at 6,000 rpm (Sorvall SS34 rotor) and the pellet dissolved in PBS containing 2 mM Ca2 and Mg2'.
Undissolved material is removed by a brief .e-lLliruga~ion at 15,000 rpm (Sorvall SS34 rotor). Thffn, the solution is mixed with Con A Sepharose and the procedurefollowed as before.

WO ~5~24923 218~51 In the second optional step, the gp96 containing fractions eluted from the Con A column are pooled and the buffer exchanged for 5 mM sodium phosphate buffer, pH 7, 300 mM NaCI by dialysis, or preferably by buffer exchange on a Sephadex G25 column. After buffer exchange, the solution is mixed with DEAE-5 Sepharose previously H ~ i with 5 rnM sodium phosphate buffer, pH 7, 300mM NaCI. The protein solution and the beads are mixed gently for I hour and poured into a column. Then, the column is washed with 5 mM sodium phosphate buffer, pH 7, 300 mM NaCI, until the l ~ at 280 nM drops to baseline.
Then, the bound protein is eluted from the column with five volumes of 5 mM
10 sodium phosphate buffer, pH 7, 700 mM NaCI. Protein containing fractions are pooled and diluted with 5 mM sodium phosphate buffer, pH 7 in order to lower the salt ~,.. l,, ~ ;" to 175 mM. The resulting material then is applied to the Mono Q FPLC column (Pharmacia) Hl~ ,il;l ,. ,.~ -1 with 5 mM sodium phosphate buffer, pH 7 and the protein that binds to the Mono Q FPLC column (Pharmacia) is15 eluted as described before.
It is ,-I,l" H; 'I ' -'1, however, that one skilled in the art may assess, by routine e,~ , thebenefitofi....,l~,u.dlil,gtheoptionalstepsintothe~...;ri.,l;.~..
protocol. In addition, it is a~ e i~ ;l also that the benefit of adding each of the optional steps will depend upon the source of the starting material.
When the gp96 fraction is isolated from the lOO,OOOg pellet, the pellet is suspended in 5 volumes of PBS containing either 1% sodium d~o;.g~lloL,l~ or 1%
octyl ~lu~u~ u~osi~e (but without the Mg2+ and Ca2+) and incubated on ice for I
h. The suspension is,, ,l, ;r~H1 at 20,000g for 30 min and the resulting dialyzed against several changes of PBS (also without the Mg2+ and Ca2+) to remove the detergent. The dialysate is ~ H1 at IOO,OOOg for 90 min, the Du~ell ~k-l 'l harvested, and calcium and " .~ , are added to the Du~ell-al~lll to give final -, - - ~ of 2mM, IeD~e.~ . Then the sample is wo gsl24923 r~l,~.,. Il ~185651 ; .

purified by either the .. "r"liri.~.~ or the modified method for isolating gp96-peptide complex from the lOO,OOOg ~ , see above.
The gp96-peptide complexes can be purified to apparent homr,grnr-ify using this procedure. About 10-20 llg of gp96 can be isolated from Ig cells/tissue.
(e) Pu-irl.alivll of HSP25 and HSP27.
The pllrifirAfir,n of Hsp25 and Hsp27 polypeptides has been di3closed 10 previously and so is not di3cussed in detail herein. See for example Jakob et al.
(1993)I.Biol.Chem.268:1517-1520,thedisclosureofwhichisi"-ul~ul~1~dherern by reference.
Briefly, the cell lysates are ~ nl with 35% ~ sulfate. The 15 pellet is harve3ted by ~ ;r ~ / solubilized in buffer and rrAriirlnA~r~l by ion exchange ~ using a DEAE Sepharose CL-6B column (Pharmacia Biotechnology, Inc.). The proteins are eluted with 50-200 mM NaCI gradient. The fractions containing Hsp25 and Hsp27 are identified by immllnrhlotting using suitable antibodies. The fractions are combined and frAr~irnAIrr, by size exclusion 20 ~ ul~Lo~ lully on a Superose 6 gel filtration column (Pharmacia).
(fl rulirl~ ivllofHsp60.
The ~ ri~ ~a ;~ of Hsp6o has been discussed in detail previously and so is 2~ not discussed in detail herein. See for example, Vitanen et al. (1992) T. Biol. Chem.
267: 695-698, the disclosure of which i3 ill~Ul~Ul~l~ed herein by reference.
Briefly, a ~ l iill matrix Iysate is applied to a Mono Q FPLC column rrLIlilihrA~rrl with 50mM sodium phosphate, ImM Mga2, ImM EGTA, pH 6.9.

Wo 95/24923 P~
21 856~1 --The proteins are eluted with a 0-IM Naa gradient. The fractions containing Hsp65are pooled and ~ 1 by ATP agarose chromatography as discussed above.
TTI P-t:y~ of Rp~ Stress Proteins ~ _ It is ~ l that ~ l stress proteins and amino acid sequence variants thereof may be prepared using conventional ~ DNA
mP~h~ r,lr,gif~c For example~ l DNAs encoding either a known stress protein or a homologue can be inserted into a suitable host cell, the protein 10 expressed, harvested, renatured if necessary, and purified. Stress proteins currently known in the art are ~ - - "" ., - i, ~ l in Table 1, below.
The processes for m~nir~ in~, r~ll-~Jli~yillg~ and ~ DNA which encode amino acid sequences of interest are generally well known in the art, and15 therefore, not described in detail herein. Methods of identifying and isolating genes encodmg members of the stress protein families also are well understood, and are described in the patent and other literature.
Accordingly, the construction of DNAs encoding biosynthetic constructs as 20 disclosed herein can be performed using known techniques involving the use ofvarious restriction enzymes which make sequence specific cuts in DNA to produce blunt ends or cohesive ends, DNA ligases, techniques enabling enzymatic additionof sticky ends to blunt-ended DNA, ~ u~ of synthetic DNAs by assembly of short or medium length r,l;g.,." ~- lPr~ P`., cDNA synthesis techniques, and 25 synthetic probes for isolating genes of members of the stress protein families.
Various promoter sequences and other regulatory DNA sequences used in achieving expression, and various types of host cells are also known and available.
Conventional tr~nc~r~ir~n techniques, and equally conventional techniques for cloning and ~ bclomng DNA are useful in the practice of this invention and known W0 9sl24923 r~
2 1 ~565~

Table I
Farnilies of Stress Proteins from Gething, ~ ~., C~, /Or~anelle Hsp 60 ~ ~
E. coli GroEL DnaK HtpG (C62.5) ~ Hsp 83/Hsc83 /cytosol Karp2 (BiP) reticulum ~ Hsp 60 (Mif4p) Drosovhila Hs~ 68 Hsp 70 ~L2 Mammals /cytosol Hsp 70 (p73) Hsp 90 (Hsp83) Hsc 70 (p72) Hsp 87 . .. 1. .~.1_~, . reticulum BiP (Grp 78) Grp 94 (Erp99) gp96 /111- ~ 11 1 .. .1.. 1 . ;A Hsp 60 (Hsp 8) Hsp 70 (Grp 75) Plants / .,.1~.~,1-~,..... ;. reticulum ~i~
/chloroplasts RUSBP
Alternative na _ are s~ovJn in I ' wo 95/24923 to those skilled in the art. Various types of vectors may be used sucb as plasmids and viruses including animal viruses and bacteriophages. The vectors may exploitvarious marker genes which impart to a successfully transfected cell a detectable phenotypic property that-can be used to identify which of a family of clones hasS successfully ill~UlUUldl~d the l~ ~-,..~,' ,,-"l DNA of the vector.
DNA molecules encoding potentially useful stress proteins may be obtained by a variety of methods. Genes of interest may be purified from standard cDNA
libraries usmg colony or plaque llyblidi~dliul~ technologies or by using polymerase 10 chain reaction (PCR) mP~hrrlrlr,giPc, all of which are well known in the art. See for example, "Molecular Cloning: A Laboratory Manual, 2nd Edition" Sambrook et al.
(1989), Cold Spring Harbor Press, the disclosure of which is ill~ulluOla~nl herein by reference. AlLcll-al;~ , the preferred genes cân be generated by the assembly of syntheticr~ p~l;rl-~producedina~ul~vc~ d~ ltrm~p~ lylluLleulide 15 byll~lle~ l followed by ligation with a~ulu~uplidl~ ligases. For example, U~ CI Id,u,Uil lg, l ~ l y DNA fragments comprising l 5 bases may be synthesized semi manually using phosphoramidite chemistry, with end segments left ullphù~luhulrldL~l to prevent luolyl~lcli~d~ion during ligation. One end of the synthetic DNA is left with a "sticky end" u-- Cb,UUlldillg to the site of action of a0 particular restriction Pnrlr,~ lP~cr-, and the other end is left with an end uulldillg to the site of action of another restriction Pnrlr,nllrlP:lcP
AlLclllaLi~ely, tbis approach can be fully ~ m~Prl The DNA encoding the l,io:,y.lLll~ic constructs may be created by byll~llcai~illg longer single strand fragments (e.g., 50-I00 nllrlPnfi~lPc long) in, for example, an Applied Biosystems5 r~li~;~., ...~ I.~r~ P Yyll~llC~ .CI, and then ligating the fragments.
The .-~ DNA constructs then may be integrated into an expression vector and transfected into an d,U~.I UIU- iaLc host cell for protein expression. I lseful host cells incl~de ~ i. Sd~ d-Ul~ly~eS. the insect/b-culovirus cell system, w09sl24923 . P_l/u~

myeloma cells, amd various other ~mAmmAIiAn cells. In E. coli and other microbial hosts, the synthetic genes can be expressed as fusion proteins. Expression in eukaryotes can be ~ 1 by the ~ r~ of DNA sequences encoding the b;~y~ Li~ protein of interest into myeloma or other type of cell line.
s The vector additionally may include various sequences to promote correct expression of the ~ A~-l protein, including lld-~s~ ol~al promoter and tPrminAti~n sequences, enhancer sequences, preferred ribosome binding site sequences, preferred mRNA leader sequences, preferred protein processing 10 sequences, preferred signal sequences for protein secretion, and the like. The DNA
sequence encoding the gene of interest also may be mAnir::lAtP-I to remove potentially inhibiting sequences or to minimi7e unwanted secondary structure formation. The 1~ " . "1,; . .A " I protein also may be expressed as a fusion protein.
After being translated, the protein may be purified from the cells themselves or15 recovered from the culture medium.
For example, if the gene is to be expressed in LSQ~ it may first be cloned into an expression vector. This is A- 1(((11111;~.11~ by pOaili~ulill~ the engineered gene downstream of a promoter sequence such as Trp or Tac, and a gene coding for a 20 leader peptide such as fragment B of protein A (FB). The resulting fusion proteins Ar~llml:lAtP in refractile bodies in the cytoplasm of the cells, and may be harvested after disruption of the cells by French press or sonication. The refractile bodies are c.~ll-hili7Pd, and the expressed proteins refolded and cleaved by methods already established for many other ,~. - ., . ,1.;, .-, .1 proteins.
Expression of the engineered genes in eukaryotic cells requires the of d~ Jlidll: cells and cell lines that are easy to transfect, are capable of stably " ,_; " 1_; - l;, -~ foreign DNA with an Ulll t~ ldll~ sequence, and which have th~ neces~ary cellular "" ,~ for ~ Icient ~Idl1~1ipliUII, w0 95l24923 ~ r~

translation, post-translation " - lir;~ , and secretion of the protein. In addition, a suitable vector carrying the gene of mterest also is necessary. DNA vector design for~ rr~ into 1 cellsshouldincludea,ulul u~ul ia~esequencesto promote expression of the gene of mterest as described supra, including S alulu-uul iale hr~ncrrirhrn initiation, I rl 1 l l; ~ and enhancer sequences, as well as sequences that enhance translation efficiency, such as the Koak consensus sequence. Preferred DNA vectors also include a marker gene and means for amplifying the copy number of the gene of interest. A detailed review of the state of the art of the production of foreign proteins in mAmm~ n cells, including useful 10 cells, protein expression-promoting sequences, marker genes, and gene amplification methods, is disclosed in Çenelic Eneineerin~ Z:9~l-127 (1988).
The best-. 1-,-, - Irl ;~r~ promoters useful for expressing a foreign gene in a particular m~mm~ n cell are the SV40 early promoter, the 15 adenovirus promoter (AdMLP), the mouse mPt~llr,th;~ nPin-l promoter (mMT-I), the Rous sarcoma virus (RSV) long terminal repeat (LTR), the mouse mammary tumor virus long terminal repeat (MMTV-LTR), and the human cytomegalovirus major ;IIlrl ..,~li,-~r-early promoter (hCMV). The DNA sequences for all of these promoters are known in the art and are available rr,mmPn^i:~,lly.

TheuseofaselectableDHFRgeneinadhfr'cellhneisawell.l.,.,,.~lr,;,r.l method useful in the amplification of genes in m:~mm~liAn cell systems. Briefly, the DHFR gene is provided on the vector carrying the gene of interest, and addition of increasing .. ".. , ~ I;....c of the cytotoxic drug lllr-lllo~lr-r~a~ leads to ~mplifir~tirn 25 of the DHFR gene copy number, as well as that of the associated gene of interest.
DHFR as a selectable, ~ 1;t;~ lP marker gene in transfected Chinese hamster ovary cell lines (CHO cells) is ~alli~ulally well 1,,-,,-- Ir~ in the art. Other useful, ..l-l;t;~ Pmarkergenesincludetheadenosinedeaminase(ADA)and glutamine synthetase (GS) genes.

Wo 95~24923 2 1 ~5651 The choice of cells/cell lines is also important znd depends on the needs of the e~ -. Monkey kidney cells (COS) provide high levels of transient gene expression, providing a useful means for rapidly testing vector ~UII~ I U~liUII and S the expression of cloned genes. COS cells are transfected with a simian virus 40 (SV40) vector carrying the gene of interest. The transfected COS cells eventually die, thus ~ iUI~ the long term 1,l u-lu-liu-~ of the desired protein product.
However, transient expression does not require the time consuming process required for the d. ~ .lo~ l of a stable cell line. Among established cell lines, 10 CHOcellsmaybethebest~ r.; ~ltodate. CHOcellsarecapableof expressing proteins from a broad range of cell types. The general cll~L~ y of CHO cells and its successful l~ludu~Liol~ for a wide variety of human proteinS in unrelated cell types ".~ the underlying similarity of all ~ , cells.
The various cells, cell lines and DNA sequences that can be used for "~ cell expression of the l~.ul-lbillhllt stress protein constructs of the invention are weU .II~,.a.L~ d in the art and are readily available. Other promoters, selectable markers, gene ~rnp~ifi~tinn methods and cells also may be usedtoexpresstheprotemsofthisinvention. ParticulardetailsoftheL,,.,.~I;,....
20 expression~ and ~ of ~ l proteins are well ~ in the art and are llnrl.~rc~/-o~ by those having ordinary skill in the art. Further details on the various technical aspects of each of the steps used in 1 ~ ", 1 production of foreign genes in m:lnnm~7i~n cell expression systems can be found in a number oftexts and laboratory manuals in the art, such as, for example, Current Protocols in 25 Molecular ~iolo~v, (1989) eds. Ausubel et al., Wiley ~

wo 95/24923 r~
2l 85651 IV. Isolation of Potentially ~ ullOyelli~ Peptides.
As mentioned previously, potentially immunogenic peptides may be isolated from either stress protein-peptide complexes or MHC-peptide complexes.
5 Protocols for isolating peptides from either of these complexes are set forth below.
(a) Peptides from Stress Protein-peptide Complexes.
Two methods may be used to elute the peptide from a stress protein-10 peptide complex. One approach involves incubating the stress protein-peptide complex m the presence of ATP, the other involves incubating the complexes in a low pH buffer.
Briefly, the complex of interest is ~n~nt~l~pl through a Centricon 10 15 assembly (Millipore) to remove any low molecular weight material loosely associated with the complex. The large molecular weight fraction may be removed and analyzed by SDS-PAGE while the low molecular weight may be analyzed by HPLC as described below. In the ATP incubation protocol, the stress protein-peptide complex in the large molecular weight fraction is incubated with l0mM
20 ATP for 30 minutes at room L~ p~ldlul~. Irl the low pH protocol, acetic acid is addedtothestressprotein-peptidecomplextogiveafinal-l""~llll,,l;.~"oflO~
(vol/vol) and the mixture mcubated in a boiling water bath for 10 minutes. See for example, Van Bleek et al. (1990) Nature 348:213-21 6; and Li et al. (l 9_3) EMBOTournal 12:3143 3151, the disclosures of which are ill...l,ol~ed herein by reference.
The resulting samples are l ~,.l, ;t-.~;~1 through an Centricon 10 assembly as mentioned previously. The high and low molecular weight fractions are recovered.The remaming large molecular weight stress protein-peptide complexes can be with ATP or low pH to remove any remaining peptides.

w0 95/24923 P~~
2 1 ~5~5 1 The resulting lower molecular weight fractions are pooled, ~ 1 by ~aluul~Liullanddissolvedino.l%~ uulua~ acid(TFA)~ Then,thedissolved material is t,, ~ ,1 by reverse phase high pressure liquid ~ Y
S (HPLC), using for example a VYDAC C18 reverse phase column ~ 1 with 0.1% TFA. The bound material ~ ll y is eluted by developmg the column with a linear gradient of 0 to 80% ~ tr)nitril~ in 0.1% TFA at a flûw rate of about 0.8 ml/min. The elution of the peptides can be monitored by D210 and the fractions containing the peptides collected.
(b) Peptides from MHC-peptide Complexes.
The isol~tion of potentially illullùllO~el.i. peptides from MHC molecules is well known in the art and so is not described in detail herein. See for example, Falk et al. (1990) Nature 348:248-251; Rotzsche et al. (1990) Nature 348:252-254; Elliott et al. (1990) Nalure 348:195-197; Falk et al. (1991) Nature 351:290-296, Demotz et A,l.
(1989) Nature 343:682-684; Rotzsche et Al. (1990) Science 249 283-287.
Briefly, MHC-peptide complexes may be isolated by a ~ul~ v~ iul~
immlln~-~ffinity procedure. Then the peptides are eluted from the MHC-peptide complex by incubating the complexes in the presence of about 0.1% TFA in A~.~tt~nitrilf~ Theextracted peptides maybe r,,.. 1;..,.~l~1 and purified by reverse phase HPLC, as before.
The amino acid sequences of the eluted peptides may be determined either by manual or automated amino acid ~ techniques well known in the art.
Once the amino acid sequence of a potentially protective peptide has been tf.rmin~l the peptide may be ~y~ in any desrred amount using w~v~ ~liu~al ~eptide ~ynthesis or other protocols well known in the art.

Wo 95/24923 1 ~, I / ~ . I I
2t8565~ --V. Synthesis of Potentially Useful Immunogenic Peptides~
Peptides having the same amino acid sequen~e as those isolated above may 5 be by~ bi ~l by solid-phase peptide synthesis using ~ rdUI ~b similar to thosedescribed by Merrifield (1963) I. Am Chem. Soc., 85: 2149. During synthesis, N-o-protected amino acdds having protected side chains are added stepwise to a growing polypeptide chain bnked by its C-terminal end to an insoluble polymeric support i.e., I~uI~D~yI~-~e beads. The peptides are byII~ Di~d by linking an amino 10 group of an N-o-deprotected amino acid to an o-carboxy group of an N-ceprotected amino add that has been activated by reacting it with a reagent such as dicyclohe~yl.dIl~udiill~ide. The ~tt~fhmDnt of a free amino group to the activated carboxyl leads to peptide bond formation. The most commonly used N-o-protecting groups include Boc which is acid labile and Fmoc which is base labile.
Briefly, the C-terminal N-a-protected amino acid is first attached to the polrD~yI~Ile beads. The N-o-protecting group is then removed. The deprotected o-amino group is coupled to the activated o-~Ibu~yl~e group of the next N-o-protected amino acid. The process is repeated until the desired peptide is 20 byI~ bi~ed. The resulting peptides then are cleaved from the insoluble polymer support and the amino acid side chains deprotected. Longer peptides can be derivedbyffnflPncltinnofprotectedpeptidefragments. Detailsof~ JliD~
fhf-mictri~c, resins, protecting groups, protected amino acids and reagents are well known in the art and so are notdiscussed in detail herein. See for example, 25 Atherton et al., Solid Phase Peptide Synthesis: A Practical Approach, IRL Press, (1989), and Bodanszky, Peptide Chemistry, A Practical Textbook, 2nd Ed, Springer-Berlog (1993), the disclosures of which are ilI~ul~ L~d herein by reference.

WO 95124923 F~
21 8565~

rul ~ ul, of the resulting peptides is ~ 1 using ~UI I V -:l l LiUl ldl lul o.~ul ~, such as ~ udla~iv~ HPLC using gel permeation, partition and/or ion exchange .Iu~ y. The choice of dlu,vlv,vlid~ matrices and buffers are well known in the art and so are not described in detail herein.
s V1. Rl-- ,. ,~l;l ~ ~ I ;v~ ~ of Stress Protein-peptide Complexes.
As will be ''l'~ fl by those skilled in the art, the peptides, either isolated from the complexes using the arul ~ - "P~1 u~ u~lul ~ or chemically0 sy~ v', may be ~ with a variety of naturally purified or stress proteins in vitro to generate i~ UIIO~ stress protein-peptide rl~mrlPY~c A preferred protocol for l~ , a stress protein and a peptide ia vitro is discussed below.
Prior to r ~ the stress proteins are pretreated with ATP or low pH
to remove any peptides that may be associated with the stress-protein of interest.
When the ATP procedure is used, excess ATP is removed from the ~ ,u~ Liul, by the addition of apyranase as discussed in Levy et al. (1991 ) Cell 67:265-274, the disclosure of which is iul-ul~uuldLev herein by reference. When the low pH
20 procedure is used the buffer is readjusted to neutral pH by the addition of pH
modifying reagents.
The peptide (Img) and the pretreated stress protein (9mg) are admixed to give an dl"u.u~iu,.al~ molar ratio of 5 peptides:l stress protein. Then, the mixture is 25 mcubated for 3 hours at room ~ uel~llul~ in a binding buffer containing 20mM
sodium phosphate, pH 7.2, 350mM NaCI, 3mM MgC12, ImM PMSF. The ,UI~:,Udl~l~iUl~:t are ~ liru~ev through Centricon 10 assembly (Millipore) to remove any unboumd peptid~. The association of the peptides with the stress proteins can WO 95/24923 2 t 8 5 ~ 5 ~ r~

be assayed by SDS PAGE and ~ y when r~ hi~llr~ peptides are used to l~UI~ U~ the complexes.
Following ,~ , the candidate illllllL'IIU~ stress protein-peptide 5 complexes can be tested ~L vitro using for example the mixed Iymphocyte target cell assay (MLTC) described below. Once potential i ~ constructs have been isûlated they can be ~ rlrlr~ 1 further in animal models using the preferred ~ a ;r,n protocols and excipients discussed below.
10 Vll. DeL~", i.-,~ivn of Immuno~enicity of Stress Protein-Peptide Complexes.
The purified and .~ stress protein-peptide complexes can be assayed for illllllUllOy,..liLi~y using the mixed l~ h-~-y ~ target-culture assay (MLTC) well known in the art.

Briefly, mice are injected cllh~ y with the candidate stress protein-peptide complexes. Other mice are injected with either other stress protein-peptide complexes or whole infected cells which act as positive controls for the assay. The miceareinjectedtwice,7-lOdaysapart. Tendaysafterthelast;..,."..,.;,.,.li..",the 20 spleens are removed and Iymphocytes released from the excised spleens. The released Iymphocytes may be rPC~im~ tr~l iL vitro by the cllhcPrlllrn~ addition of dead cells which prior to death had expressed the complex of interest.
For example, 8X106 immune spleen cells may be stimulated with either 4x104 mitomycin C treated or ~irradiated (5-10,000 rads) cells (the cells having been infected with the i~ pathogen or transfected with an r~lu~lulJlid~e gene) in 3ml RPMI medium containing 10% fetal calf serum. In certain cases 33%
secondary mixed l~ ,ho-y~ culture ~ l may be included in the culture medium as a source of T cell growth factors. See for example, Glasebrook et al.

~ W095124923 I~l/~J~ c ll 2 1 ~5651 -- 4s -(1980) T. Exp. Med. 151:876. In order to test the primary cytotoxic T cell response after i, ~ , spleen cells may be cultured without ~inmll~firln In some P^l,~.;..,~..l~spleencelisofthe~ micealsomaybe~ lwith ~n~i~PnirRlly distinct ce31s, to determine the specificity of the cytotoxic T cell 5 response.
Six days later the cultures are tested for ~y ~o~u~d~i~y in a 4 hour 51 Cr-release assay. See for example, Palladino et al. (1987) Cancer Res. 47:5074-5079 and Blachereç~l. (1993) I. T~ y 14;352-356, the disclosures of which are 10 incorporated herein by reference. In this assay, the mixed Iymphocyte culture is added to a target cell suspension to give differen~ æ~ o-.~a,~r~ (E:T) ratios (usually 1:1 to 40:1). The target cells are prelabelled by incubating 1x106 target cells in culture medium containing 20û mCi 51Cr/ml for one hour at 37C. The cells arewashed three times following labeling. Each assay point (E:T ratio) is performed in 15 triplicateandthe~.v~ controlsi-I u-,uu-~ tomeasure~ P~ 5lCr release (no l~ ,ul~o-y~ added to assay) and 10û% release (cells Iysed with detergent). After incubating the cell mixtures for 4 hours, the cells are pelleted by , rl ~ l ;nn at 2oog for 5 minutes~ The amount of 51 Cr released rnto the bul,~..,a~..~ is measured by a gamma counter. The percent~ u~u;d i~y is 20 measured as cpm in the test sample mmus Dl!UIl~ UU~Iy released cpm divided by vhe total detergent released cpm minus ~IUU~ VUbly released cpm.
In order to block the MHC class l cascade a - - -"- ~ I P~-i hybridoma ~ulu~ a~d~I~ derived from K-44 hybridoma cells (an anti-MHC class I hybridoma) is 25 added to the test samples to a final ~I-nrPn~ n of 125%.

Wo 95/24923 21 ~5651 -- ~6 --VIII. Formulation and Vaccination Re~imes.
Once candidate stress protein-peptide complexes have been identified they S may be ~ 1 either to an animal model or to the intended recipient to stimulate cytotoxic T cell responses against the IJle~eléLle l in~rA~ r pathogen.
The stress protein-peptide complexes of the invention may be either stored or prepared for ~.] ".:,.:~l .,.li.~.. by mixing with physiologically acceptable carriers, excipients, or stabilizers. These materials should be non-toxic to the intended 10 recipient at dosages and - - - . .. ~ c employed.
If the complex is water soluble then it may be formulated in an ~ v- V,VI ia~
buffer, for example phosphate buffered saline (5mM sodium phosphate, 150 mM
NaCI, pH7.1) or other physiologically compatible solutions. AIL~ llively, if the15 resulting complex has poor solubility in aqueous solvents then it may be formulated with a non-ionic surfactant such as Tween, or polyethylene glycol.
Useful solutions for oral or parenteral ~-lminic~r~ n may be prepared by any of the methods well known in the ~ a-~u~ l art, described, for example, 20 in Remin~tQn's Pl .,., ., .,.~ ~, . I i- ,.l Sciences, (Gennaro,. A., ed~.), Mack Pub., 1990.
F~rm~ nc may include, for example, polyalkylene glycols such as polyethylene glycol, oils of vegetable origin, I~vl-u~ Lt d 1~ c and the like.
Ft~rm~ ir-nc for direct - ' f n, in particular, may include glycerol and other"-..,~ ;l;....c of high viscosity. ~ , preferdbly 'viu-ebv-'vdvle 25 polymers, including, for example, hyaluronic acid, collagen, tricalcium phosphate, polybutyrate, polylactide, polyglycolide and lactide/glycolide copolymers, may be useful excipients to control the release of the stress protein-peptide complexes in vivo.

Wo 9~124923 r~
2 1 856~1 -- ~7 --Fnrrmll~hnnc for mhalation ~ may contain as excipients, for example, lactose. Aqueous solutions may contain, for example, pvlyv~y~ ylr-lle-9lauryl ether, glycocholate and deoxycholate. Oily solutions may be useful ~,I",;.,i~l,..linn mtheformofnasaldrops.Gelsmaybeappliedtopically S intranasally.
The ~ ,u. . - ,~l~ provided herem can be fnrm~ t~i into l~
c by admixture with r~ y acceptable non toxic excipients and carriers. In addition the fnrm~ hnnc may optionally contain one or more 10 adjuvants. Preferred adjuvants include, but are not limited to, pluronic tri-block ~ulvoly-~Irla, muramyl dipeptide and its derivatives, detoxified endotoxin, saponin and its derivatives such as QS-21 and liposomes. The present invention further envisages sustained release fnrrmllAfinnc in which the complex is released over an extended period of trme.
The dosage and means of a-l~ dlivl, of the family of stress protein-peptide vaccines prepared in accordance with the invention will necessarily depend upon the na~re of the complex, the intr~r~ lAr pathogen and tlle nature of the disease in question. The complex should be administered in an amoun~ sufficient 20 to initiate a cytotoxic T cell response against the intr~ r pathogen. The preferred dosage of drug to be a.ll..illi~L~I~d also is likely to depend on suchvariables as the type of disease, the age, sex and weight of the intended recipient, the overall health status of the particular patient, the relative biological efficacy of the compound selected, the t~ of the compound, the presence and types of excipients in the rl .. 1~1 ;. ., and the route of ~. l ., .; .. ;~l .. l ;
In general terms, the ~ of this invention may be provided in an aqueous pllyaiOlvgi~l buffer solution containing about û.001 to 10% w/v compoundfo~paren rali~,l-,.;-l;~l-,~l;---l TypicaldosesrangefromaboutO.1 to Wo 95/24923 PCTIUSg5/03311 ~
~ 85651 about 1000 llli~lU~I.IIII~ of complex/kgbody weight of recipient/i".,.",.,i,,.l;,....
and preferably range from about 0.5 to about 100 Illi.lu~;lalllD of complex/kg body weight of recipient/i~ 7n It is ~ that between about 10 to about 250 Illi.l Ugl ~.llls of complex will be ad~ L~I ni per dose to a human subject 5 weighing about 75kg. These quantities, however, may vary according to the adjuvant-co-ad.llil~ d with the complex.
The vaccines may be adll,illis~ .l using standard protocols which include, but are not limited to, ;..I,,."",~,Ilar, c~ ""~, intradermal, illLInp~liLun~nl,10 intravenous, ill~ldvngilldl, intrarectal, oral, sublingual, transcutaneous, and intranasal ~ ".;";~ ,--, Preferably the recipient should be vaccinated four times at weekly intervals. If necessary, the responses may be boosted at a later date by cl1hsPqll~nt ~,1,.,:..: .1..-l;.... of the vaccine. It is ~ 1 that the optimal dosage and vn~ilmLiùn schedule may be fl~t~rminPcl empirically for each stress 15 protein-peptide vaccine using techniques well known in the art.
Various cytokines, antibiotics, and other bioactive agents also may be ad~ L~I~d with the stress protein complexes. For example, various known cytokines, i.e., IL-la, IL-I,B, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-11, 20 IL-12, IFNa, IFN~, IFNy, TNFa, TNF,f~, G-CSF, GM-CSF, and TGF-fj may be co-adl.lilli~ d with the complexes in order to maximize the physiological response.It is anticipated, however, that other but as yet undiscovered cytokines may be effective in the invention. In addition, conventional antibiotics may be co-a.l.llil~is~ with the stress protein-peptide complex. The choice of suitable 25 antibiotic i will also depend upon the dl~ease b que~horl --wo 95124923 P~
21 8565`1 ~, 9 ~ xample 1. r,~ c~ y of StressPrQtein-peptide Com~lexes Isolated from Cells Transfected with a Gene ~ncodin~ an Antigenic D5~
S hgure 1 shows the antigen specific cytotoxic T cell activity of ~lulellu~y lès derived from mice;,..."....: .1 with a gp96-peptide complex harvested from BALB/c fibroblasts transfected with the l~.leu~ul uLeilI (NP) gene from the PR8 mfluenza virus.
Briefly, gp96-peptide ~- elJal~lLions were isolated from BALB/c cells transfected with and expressing the NP gene of the PR8 influenzs virus. The gp96-peptide complex was isolated from 100,000g su~ellIaklllt by the unmodifed gp60-peptide complex ~.., . i r;. ,~ ., protocol. Then, the preparations were used toimmunize naive BALB/c mice. The mice were injected twice ~ ",~ y with 15 the gp96-peptide complexes at ten day intervals. The mice were sacrificed and the spleen cells obtained. The spleen cells were stimulated twice in vitrQ by the additional lethally irradiated BALB/c cells expressing the NP gene using the mixed target Iymphocyte culture (MLTC) assay described above. Six days later the cultures were tested for ~y~oIo~i~iLy using the 5ICr release assay. In order to blûck 20 the MHC type I cascade the spleen cells were incubated with the su~el .la~llI derived from K 44 hybridoma (ront linin~ anti-MHc type ~ n~ c) culture.
The cytotoxic activity was assayed by the release ûf 51Cr from BALB/c 25 fibroblasts expressing the NP gene (filled circles), BALB/c cells expressing the NP
gene but treated with the anti-MHC type I antisera (empty circles) and from the syngeneic non-NP transfected cell line 5117 (asterisks). The spleens of the micei . " " . . ", i, ~ with the gp96 complex showed strong MHC class l-restricted cytotoxic T cell activity ~gainst ~ALB/c cells expressing the NP gene, but not against the Wo gS/24923 r~
2 1 8~6~1 --syngeneic non-NP transfected cell line 5117. Fuull~ ol~, the anti MHC type I
antisera blocked the response. Therefore, it is apparent that i~.,.,,,,,;,~l;, ,,, with a stress protein-peptide complex elicits a specific cytotoxic T cell response a~ainst the peptide in the complex and that the MHC class I cascade plays an integral role in 5 .c~im~ in~ the cytotoxic T cell response against cells infected with in~r:~rrl~ r pathogens.
Example 2. Ill,l..ul.o~ y of Stress Protein-peptide Complexes Isolated from SV40 TldllDrUIIII~ Cells.
Figure 2 shows the antigen specific cytotoxic T cell activity of sp~enocytes derived from mice ;..""...,;,.~.1 with gp96-peptide complex harvested from SV40 Drullll~-l SVB6 cells.
Briefly, gp96-peptide complexes were isolated from SV40 lldl~arc Ill,ed SVB6 cells and used to immunize naive (57BL/6) mice. The gp96-peptide complex was iso~ated from lOO,OOOg supernatant by the unmodifed gp60-peptide complex purification protocol. The mice were injected twice ~..I,. I,~,..-,~,..~Iy with the complex at ten day intervals. The mice were sacrificed, the spleen cells isolated and 20 stimulated in vitro by the addition of lethally irradiated SV40 transformed SVB6 oells by the MLTC procedure. Six days later the cells were assayed for ~yluLu~d~iLy using the 51Cr release assay. The cytotoxic activity was assayed by the release of 5ICr from SVB6 cells (filled triangles) and from a non SV4û transfected syngeneic cell line, MCA (empty triangles). MHC class I mediated activity was assayed also25 by adding anti-MHC class I immI~nn~lnbulins derived from the K-44 hybridoma cell line to the spleen cells.

WO 95124923 P~
2 1 8~651 The spleen cells isolated from mice; " .. , .;, ~1 with the gp96-peptide complex showed strong MHC class l-restricted activity against the SV40 transfected SVB6 cells but not agamst the non transfected cerls.
s Example 3. Rerrnstit lti~ of Immllnog~nir Stress Protein-pçr~tide Complexes In Vitro.
Figures 3A-3D show antigen specific cytotoxic T cell activities of splenocytes 10 derived from two mice i~ 1 with ~ "~ l Hsp70-peptide complex.
Briefly, ~ ~1 Hsp70 was purified by the procedure described above amd the peptide (CiLSDLRGYVYQGL, SEQ. ID. NO.: I~ was ~y~ d by sorid phase peptide synthesis. The peptide (Img) and ATP treated Hsp70 (9mg) were 15 admrxed and incubated for 3 hours at room ~ e~ ul~ in a binding buffer containing 20mM sodium phosphate, pH 72, 350mM NaCI, 3mM MgC12, ImM
PMSF. The resulting lu~ laliul~ was ~ liru~eLl through Centricon 10 assembly (Millipore) to remove umbound peptide.
The resulting complex was used to immunize two naive mice. The spleen cells were isolated from the mice and stimulated twice in vitro by the addition of lethally irradiated EL4 cells transfected with, and expressing a minigene encoding the peptide SLSDLRGYVYQGL (SEQ. ID. NO.: 1), using the MLTC procedure. The ~:ylù~u~ of spleen cells from both mice were assayed after the first (3A and 3C) and second (3B and 3D) s~inmll:~tirnc by the 5ICr release assay. The release of 5ICr was measured from EL4 cells (hollow triangles) and from EL4 cerls transfected with, and expressmg the peptide SLSDLRGYVYQGL (SEQ. rD. NO.: I) (firled b~an~les) h n~sult-show~hatstr~ssp~otemsardp ptdssca be.~.. Cl l..1 WO 95l24923 r~l,~J.,. . Il ~

successfully rn vitro to give specific immunogenic stress protein-peptide complexes.
The invention may be embodied in other specific forms without departing 5 from the spirit or essential ~ thereo The present ~ o~ , are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

wo 95n4923 F~ r~
~1 85651 ., SEQUENCE LISTING
(I) GENERAL INFORMATION:
s (i) APPLICANT:
(A) NAME: MOUNT SINAI SCHOOL OF MEDICINE
(B) STREET: I GUSTAVE L. LEVY PLACE
(C) CITY: NEW YORK
(D) STATE: NEW YORK
(E) COUNTRY: USA
(F) POSTAL CODE (ZIP): 10029 (G) TELEPHONE: 212 241-0826 (H) TELEFAX:
(I) TELEX:
(ii) TITLE OF INVENTION: STRESS PROTEIN-PEPTIDE COMPLEXES AS
PROPHYLACTIC AND THERAPEUTIC VACCINES AGAINST
INTRACELLULAR PATHOGENS
(iii) NUM3ER OF SEQUENCES: I
(iv) ~ oNDENCE ADDRESS:
(A) ADDRESSEE: PATENT ADMINISTRATOR, TESTA HURWITZ &
THIBEAULT
(B) STREET: 53 STATE STREET, EXCHANGE PLACE
(C) CITY: BOSTON
(D) STATE: MA
(E) COUNTRY: USA
(F) ZIP: 02109 (v) COMPUTER READABLE FORM:

WO 9~/24923 PCTNS9~/03311 21 85~7 5~
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: Patentln Release #1.0, Version #1.25 (vi) CURRENT APPLICATION DATA:
~A) APPLICATION NUMBER:
(B~ FILING DATE:
(C) CLASSIFICATION:
(viii) ATTORNEY/AGENT INFORMATION:
(A) NAME: PITCHER EDMUND R
(B) REGISTRATION NUMBER: 27,829 (C) REFERENOE/DOCKET NUMBER: ARM-001 (ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: (617) 248-7000 (B) TELEFAX: (617) 248-7100 (2) INFORMATION FOR SEQ ID NO: 1:
(i) æQUENCE CHARACTERISTICS:
(A) LENGTH: 13 arnino acids (B) TYPE: arnino acid (C) STRANDEDNESS: slngle (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (ix) FEATURE:
(A) NAME/KEY: Peptide (B) LOCATION: I ..~ 3 WO 95124923 1~
21 8~65~
ss (D) OTHER INFORMATION: /label= PEPTIDEI
/note= "ANTIGENIC PEPTIDE 1"
(~u) SEQ13ENCE L1k~1' ~ 1: SEQ ID NO: 1:
Ser Leu Ser Asp Leu Arg Gly Tyr Val Tyr Gln Gly Leu

Claims (44)

What is claimed is:

1. A composition comprising:
(a) an amount of a purified immunogenic mammalian stress protein-peptide complex sufficient to induce in said mammal an immune response against a preselected intracellular pathogen, said complex comprising, a mammalian stress protein noncovalently associated with a peptide that is present in an eukaryotic cell infected with said pathogen but not present in said cell when said cell is not infected with said pathogen; and (b) a pharmaceutically acceptable carrier.

2. A composition comprising:
(a) an amount of a purified immunogenic mammalian stress protein-peptide complex sufficient to induce in said mammal an immune response against a preselected intracellular pathogen, said complex comprising, a mammalian stress protein noncovalently associated with a peptide that is present in an eukaryotic cell transfected with a gene encoding an antigenic determinant of said pathogen but not present in said cell when said cell is not transfected with said gene; and (b) a pharmaceutically acceptable carrier.

3. The composition of claim 1 or 2, wherein said stress protein is a member of the stress protein families selected from the group consisting of Hsp60, Hsp70, and Hsp90.

4. The composition of claim 1 or 2, wherein said stress protein is gp96.

5. The composition of claim 1 or 2 further comprising a cytokine.

6. The composition of claim 5, wherein said cytokine is selected from the group consisting of IL-1.alpha., IL-1.beta., IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-11, IL-12, IFN.alpha., IFN.beta., IFN.gamma., TNF.alpha., TNF.beta., G-CSF, GM-CSF and TGF-.beta..

7. The composition of claim 1 or 2, wherein said eukaryotic cell is an immortalized eukaryotic cell.

8. The composition of claim 1 or 2, wherein said stress protein is a human stress protein.

9. The composition of claim 1 or 2, wherein said pathogen is a virus.

10. The composition of claim 9, wherein said virus is selected from the group consisting of hepatitis type A, hepatitis type B, hepatitis type C, influenza, varicella, adenovirus, herpes simplex type 1, herpes simplex type II, rinderpest, rhinovirus, echovirus, rotavirus, respiratory syncytial virus, papilloma virus, papova virus, cytomegalovirus, echinovirus, arbovirus, huntavirus, coxsackie virus, mumps virus, measles virus, rubella virus, polio virus, human immunodeficiency virus type I and human immunodeficiency virus type II.

11. The composition of claim 1 or 2, wherein said pathogen is a bacteria.

12. The composition of claim 11, wherein said bacteria is selected from the group consisting of Mycobacteria, Rickettsia, Neisseria and Legionella.

13. The composition of claim 1 or 2, wherein said pathogen is a protozoan.

14. The composition of claim 13, wherein said protozoa is selected from the group consisting of Leishmania, Trypanosoma and Kokzidioa.

15. The composition of claim 1 or 2, wherein said pathogen is an intracellular parasite.

16. The composition of claim 15, wherein said parasite is selected from the group consisting of Chlamydia and Rickettsia.

17. A composition for use as a medicament, said composition comprising a pharmaceutically acceptable carrier and a purified immunogenic mammalian stress protein-peptide complex, said complex comprising a mammalian stress protein noncovalently associated with a peptide that is present in an eukaryotic cell infected with a preselected intracellular pathogen but not present in said cell when said cell is not infected with said pathogen.

18. A composition for use as a medicament for inducing in a mammal an immune response against a preselected intracellular pathogen, said composition comprising a pharmaceutically acceptable carrier and an amount of a purified immunogenic mammalian stress protein-peptide complex sufficient to induce in said mammal an immune response against said pathogen, said complex comprising a mammalian stress protein noncovalently associated with a peptide that is present in an eukaryotic cell infected with said pathogen but not present in said cell when said cell is not infected with said pathogen.

19. A composition for use as a medicament for inducing in a mammal an immune response against a preselected intracellular pathogen, said composition comprising a pharmaceutically acceptable carrier and an amount of a purified immunogenic mammalian stress protein-peptide complex sufficient to induce in said mammal an immune response against said pathogen, said complex comprising a mammalian stress protein noncovalently associated with a peptide that is present in an eukaryotic cell transfected with a gene encoding an antigenic determinant of said pathogen but not present in said cell when said cell is not transfected with said gene.

20. The composition of claim 18 or 19, wherein said immune response is a cytotoxic T cell response mediated by the class I major histocompatibility complex.

21. The composition of claim 17, 18 or 19, wherein said stress protein is a member of the stress protein families selected from the group consisting of Hsp60, Hsp70, and Hsp90.

22. The composition of claim 17, 18 or 19, wherein said stress protein is gp96.

23. The composition of claim 17, 18 or 19, wherein said composition further comprises a cytokine.

24. The composition of claim 23, wherein said cytokine is selected from the group consisting of IL-1.alpha., IL-1.beta., IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-11, IL-12, IFN.alpha., IFN.beta., IFN.gamma., TNF.alpha., TNF.beta., G-CSF, GM-CSF and TGF-.beta..

25. The composition of claim 17, 18 or 19, wherein said eukaryotic cell is an immortalized eukaryotic cell.

26. The composition of claim 17, 18 or 19, wherein said mammal is a human.

27. The composition of claim 17, 18 or 19, wherein said use is prophylactic, for stimulating in said mammal a cytotoxic T
cell response for preventing subsequent infection of said mammal by said pathogen.

28. The composition of claim 17, 18 or 19, wherein said use is therapeutic, for stimulating in said mammal a cytotoxic T
cell response against said pathogen presently infecting said mammal.

29. The composition of claim 17, 18 or 19, wherein said composition is for administration to said mammal in an amount in the range of about 0.1 to about 1000 micrograms of complex/kg body weight of mammal/immunization.

30. The composition of claim 29, wherein said amount is in the range of about 0.5 to about 100 micrograms of complex/kg body weight of mammal/immunization.

31. A method for preparing a composition for inducing in a mammal an immune response against a preselected intracellular pathogen, said method comprising:
(a) isolating from an eukaryotic cell infected with said pathogen or transfected with a gene encoding an antigenic determinant of said pathogen an immunogenic mammalian stress protein-peptide complex comprising, a mammalian stress protein noncovalently associated with a peptide that is present in said cell infected with said pathogen or transfected with a gene encoding an antigenic determinant of said pathogen but not present in said cell when said cell is not infected with said pathogen or transfected with a gene encoding an antigenic determinant of said pathogen; and (b) combining an amount of said complex which is sufficient to induce an immune response in a mammal with a pharmaceutically acceptable carrier.

32. A method for preparing a composition for inducing in a mammal an immune response against a preselected intracellular pathogen, said method comprising;
(a) reconstituting in vitro, a peptide that is present in an eukaryotic cell infected with said pathogen or transfected with a gene encoding an antigenic determinant of said pathogen but not present in said cell when said cell is not infected with said pathogen or transfected with a gene encoding an antigenic determinant of said pathogen and a stress protein, thereby generating a noncovalent stress protein-peptide complex; and (b) combining an amount of said complex which is sufficient to induce an immune response with a pharmaceutically acceptable carrier.

33. The method of claim 32 wherein said stress protein is isolated in the presence of ATP prior to reconstitution.

34. The method of claim 32, wherein said stress protein is treated with low pH prior to reconstitution.

35. The method of claim 31 or 32, wherein said stress protein is a member of the stress protein families selected from the group consisting of Hsp60, Hsp70 and Hsp90.

36. The method of claim 31 or 32, wherein said stress protein is a gp96.

37. The composition of claim 4, wherein said stress protein is a human stress protein.

38. The composition of claim 17, 18 or 19 wherein said mammal is a human and said stress protein is a human stress protein.

39. The composition of claim 17, 18 or 19 wherein said pathogen is selected from the group consisting of a virus, a bacterium, a fungus, a protozoan and a parasite.

40. The composition of claim 39 wherein said mammal is a human and said stress protein is a human stress protein.

41. A method of inducing in a mammal an immune response against a preselected intracellular pathogen that causes disease in said mammal, said method comprising:
administering to said mammal a composition comprising:
(a) an amount of a purified immunogenic mammalian stress protein-peptide complex sufficient to elicit in said mammal an immune response against said pathogen and comprising, a mammalian stress protein noncovalently associated with a peptide that is present in an eukaryotic cell infected with said pathogen or transfected with a gene encoding an antigenic determinant of said pathogen but not present in said cell when said cell is not infected with said pathogen or transfected with a gene encoding an antigenic determinant of said pathogen; and (b) a pharmaceutically acceptable carrier.

42. A method of inducing in a mammal an immune response against a preselected intracellular pathogen that causes disease in said mammal, said method comprising:
administering to said mammal a composition comprising:
(a) an amount of a purified immunogenic mammalian stress protein-peptide complex sufficient to elicit in said mammal an immune response against said pathogen and comprising, a mammalian stress protein noncovalently associated with a peptide, said complex having been isolated from an eukaryotic cell infected with said pathogen or transfected with a gene encoding an antigenic determinant of said pathogen; and (b) a pharmaceutically acceptable carrier.

43. A composition comprising:

(a) a purified mammalian immunogenic stress protein, peptide complex, said complex comprising, a mammalian stress protein noncovalently associated with a peptide that is present in an eukaryotic cell infected with a preselected intracellular pathogen but not present in said cell when said cell is not infected with said pathogen; and (b) a pharmaceutically acceptable carrier.

44. A composition for use as a medicament for inducing in a mammal an immune response against a pharmaceutically acceptable carrier and a preselected intracellular pathogen, said composition comprisinq an amount of a purified immunogenic mammalian stress protein-peptide complex sufficient to induce in said mammal an immune response against said pathogen, said complex having been isolated from an eukaryotic cell infected with said pathogen, and said complex comprising a mammalian stress protein noncovalently associated with a peptide.