0022- 1767/86/137 1 -0OOiSO2.00/0 THEJOURNAL OF 1MMUNOLoGY Vol. 137. No. 1. July 1, I986 Copyright 0 1986 by The American Association of Immunol~lsts Prlnted In U.S.A.

American Associationof Immunologists

PRESIDENTIALADDRESS

LEARNINGFROM :A PERSPECTIVE ONIMMUNOLOGY AND THE THIRDWORLD

BARRY R. BLOOM From the Departmentsof Microbiology and Immunology. andCell Biology, Albert Einstein Collegeof Medicine. Bronx,NY 10461

"If we take the widest and wisest view of a Cause. there is no such thingas a Lost Cause, because there is no such thingas a Gained Cause. We fight for Lost Causes because we know that our defeat and dismay may be the preface to our successors' victory, although that victory itself will be temporary; we fi ht rather to keep somethning alive than in the expectation t fl at anything will triumph. "T.S. Eliot

"A Map of the World Without Utopia on It Is not Worth Glancing At." "Oscar Wilde

Let mebegin with a case history. notof an individual. tussis,tetanus, , polio, and measles, and but ratherof a country. any of the fortypoorest nations consequently 0.5%of them became lame from polio, 1% on earth. Let me ask you to try to imagine our qualityof died from neonatal tetanus. 2% succumbedto whooping life, if life expectancy at birth in this countrywere 42 yr. cough. and 3%died from measles. We would be living in if infant mortality at birth were 140 per thousand.if 40% a country whose average gross national product per cap- of our children suffered from malnutrition. and if only ita would be $310/yr: in which 37% of males, but only 10%of children were immunized against , per-14% of females, would be literate. Access to waterwould i ii LEARNING FROM LEPROSY be enjoyed by 50%of the urbanpopulation, but only 15% monoclonal anti-interferon (IFN)-y . Recently, of therural population.Government expenditures on Nathan et al.(6) have observed skin reactions, similar to health and education would represent only 3.9%of the those induced by MIF in guinea pigs, produced interest- central budget, and there would be 15,932 people per ingly in the skin of lepromatous leprosy patients after physician. This is the quality of life, differing only in inoculation of recombinant IFN-7. detail, endured by 200 to 300 million people in Africa, As the in vitro study of cellularimmune reactions Asia, and Latin America (1). This is the Third World, in became accepted as a legitimate immunological endeavor, which 75% of the planet’s population lives, where 86% I was invited by the World Health Organization to help of all children are born and 98% of all infant and child evaluatewhether these emerging in vitro techniques deaths occur, and where 10 kids die of vaccine-prevent- might have some conceptual or practical relevance to able illnessevery minute. understanding immunity to tropical diseases. One such This wasnot my world when I embarked ona career in meeting, held in India in 197 1,introduced me to the Third immunology. That world was the Rockefeller University World and the problem of leprosy, and was one from where I was trained in cellularimmunology by Dr. Merrill which I have never quite recovered. In the course of my Chase. Dr. Chase was one of the founders of cellular wanderings in immunology and around theglobe, I have immunology, who first established that delayed-type hy- been privileged to have been able to collaborate with a persensitivity (DTH) could be passively transferred by great many wonderful scientists andcolleagues, both in cells, not serum, andwhose example impressedupon me this country and abroad, and would like to express my the importance of discipline and commitment in science. sincere appreciation for all they have contributed. I am That too was not the world of St. Mary’s Medical School deeply grateful for the dedicated efforts and valuable in London where, as a postdoc, I learned something of advice provided me by many members of the Secretariat the structure of antibodies from Rodney Porter. I came, of the World Health Organization. in addition,to learn three important things from my Leprosy-the disease. From ancient times and inev- experience in Rod Porter’s lab: i) the active site of anti- ery culture,leprosy has evoked singular images of horror bodies was not on isolated heavy chains, ii) I was not and fascination. Many aspects of the diseaseremain destined to be a great chemist,and iii) most importantly, mysterious to this day. Although the leprosy bacillus was it was ultimately possible to do the highest quality of the first identified bacterial pathogen of man (7). it re- science with in no way compromising one’s dignity, gen- mains oneof the very few pathogens of man that cannot erosity, humor,or humanity. I cannot imagine being more yet be grown in culture. Thereis generally a long latency, privileged than to have had these two wonderful and perhaps 5 yr between presumed infection and manifes- inspiring teachers. tation of disease, and as aconsequence the mode of To convey a sense of time in that “world,” the least transmissionremains largely unknown. Although 13 understood phenomenon in immunology was “delayed- million people are estimated to have leprosy around the type hypersensitivity,” not then even dignified by the world, the disease hasa relatively low prevalence, seldom term “cell-mediated immunity.” Pioneering studies in vi- exceeding 1 to 5 per 1000 in endemic areas. There is a troon lymphocytes and macrophages by Hirshhorn, unique fear and stigmaassociated with this disease, Bach, George, and Vaughan and John David were just partially deriving from the deformities that occur in ap- breaking new ground. The most burning question, in- proximately 30% of people afflicted by it. There is no credible as it may sound, that Boyce Bennett and I first other disease whosevictims were burned at the stakeby tried to address at Albert Einstein, was whether sensi- Henry I1 of England and Philip V of Spain, or buried alive tized lymphocytes or macrophages, or both, carried the by Henry’s grandson,Edward I, or given the last ritesof immunological specificity for DTH. We were able to show the dead by the Church in medieval Europe and turned in the macrophage migration inhibition reaction i) that outside the city walls (8). only immune lymphocytes had the ability to recognize The spectrum of leprosy. There has been a renais- specific and inhibit macrophage migration: in- sance of scientific interest in leprosy, one reason being deed, as few as 0.6% of sensitized cells were sufficient: the extraordinary possibilities it offers for gaining insight and ii) that the in vitro reactionwas mediated by a into immunoregulatory mechanisms in man (9, 10). For nonantibody product of the antigen-activated lympho- leprosy is not a single clinical entity, but rathera spectral cytes (2). JohnDavid made the same observation essen- disease that presents a diversity of clinical manifesta- tially simultaneously (3). Migration inhibitoryfactor, tions (Fig. 2). At one pole of the spectrum, tuberculoid MIF, thus became the first of the lymphokines. Bennett leprosy, patients develop high levels of specific cell-me- and I really first began to believe in the lymphokines diated immunity that ultimately kills and clears the ba- when we found (Fig. 1) thatpartially purified MIF injected cilli in the tissues, although often with concomitant im- into normal guinea pig skin produced reactions as early munological damage to the nerves. At the lepromatous as 4 hr consisting of almost pure mononuclear cell infil- pole, patients exhibit a selective unresponsivenessto trates resembling DTH reactions (4).Although the chem- of leprae and the organisms ical nature ofMIF has been somewhat elusive and its ineluctably multiply in the skin, often to extraordinary biological effects on macrophages may well be mediated numbers-as many as 10’’ per gram of tissue. It remains by more than one molecular entity, it was gratifying to unclear why the vastmajority of peopleexposed to infec- learn only last year from the work of Thurman etal. (5) tion with M. leprae develop no clinical disease, and why that the in vitro activities of MIF were produced on hu- only a minority who do develop clinical disease become man monocytes by cloned human IFN-y, and conversely, lepromatous and remain immunologically unresponsive that MIF activityinduced by concanavalin A (Con A) to the antigensof the organism. activation of human lymphocytes was neutralized by Mechanism of nerve damage in leprosy. To a visitor LEARNING FROM LEPROSY FROM LEARNING iii

L' L' r' Cy- s . Flgure I. Skin reactions produced in normal guinea pigs by MIF. Sephadex C-100 fractions of serum-free lymphocyte super- C. natants (approximate size range.40.000 to 65.000 daltons) from W control and Con A-stimulated lymphocytes wereinjected intrader- mally into normal guinea pig skin and werebiopsied 4 hr later. A. MIF fraction (x47.5). B. Control fraction (x47.5). C. MIF fraction (x 180).Fractions containing MIF induced mononuclear cell infil- tration and keratinocyte hyperplasia(4).

THE SPECTRUM OF LEPROSY The resultsclearly demonstrated primaryfocal demyelin- ation of these myelinated fibers as innocent "bystanders" of a specificcellular immune reaction tothe foreign antigen (Fig. 3). Togetherwith Celia Brosnan, Wendy Cammer, Gerry Stoner, SiamonGordon, and Bill Norton, we were able to demonstrate that activated macrophages secreted neutral proteinases that had the ability to de- grade purified myelin proteins in vitro(1 2). One of these was inhibitableby EDTA and wasprobably a collagenase: the other wasplasminogen activator, dependent on plas- minogen and blocked by inhibitors of plasminogen acti- vator and plasmin. Thissuggested that activated macro- phages could secrete small amountsof neutral proteases, one of which could act upon the very large reservoir of plasminogen (-1 mg/ml) found in serum and extravas- cular fluids. releasing plasminthat would have the ability locally to degrade myelin proteins. If that were true, then Figure 2. The spectrum of leprosy. specific protease inhibitors shouldbe capable of blocking demyelination in vivo. Celia Brosnan and I set up the to the Third World for the first time, themost devastating model of experimental allergic encephalomyelitis (EAE) aspect of leprosy was the horrible deformities seen in in rats, and7 days after the animalswere sensitized and people suffering from the disease. A particular paradox the spleen cellswere capable of transferring EAE to was the nerve damage in patients with tuberculoid lep- normal recipients, the animals were treated with a vari- rosy in which cellular immunity had virtually eliminated ety of protease inhibitors (13). Two of these inhibitors, the acid-fast bacilli (AFB) from the tissues. Because M. p-nitrophenyl guanidinobenzoate and transaminometh- leprae hasselectivity for growth essentially only in mac-ylcyclohexane carboxylic acid, quite effectively blocked rophages and Schwann cells, Henry Wisniewski and I demyelination histologically and reduced paralysis in the undertook todevelop a model to analyze thistype of nerve vast majority of animals. These studiesintroduced us to damage (1 1). Guinea pigs were sensitized to Freund's two new mechanisms of immunological damage to mye- adjuvant and were challenged withpurified protein deriv- lin. bystander demyelination and an -depend- ative (PPD) near myelinated fibers of the sciatic nerve or ent. cell-mediated demyelination (14) that could have ventricles. The question asked was,"What would happen relevanceto inflammatory demyelination, perhaps in to nerves as a result of a cellular immune reaction to multiple sclerosis, and to general mechanisms of tissue foreign antigens in the vicinity of myelinated fibers?" damage. iv LEPROSY FROM LEARNING

siveness. The special aspectof this unresponsiveness in lepromatous leprosy is that it appears to be acquired, occurs over a long period of time, and in general is not life threatening. Consequently, leprosy provides unique opportunities for studyof unresponsiveness in man and for immunologic intervention. The textbook description of the "anergy" of lepromatous leprosy is that such patients aretotally unresponsive to new and recall antigens. Although nonspecific anergy may in fact be seen in small numbers of late-stage un- treatedpatients, clearly the vast majority of patients exhibit normal cell-mediated immunity to a variety of recall antigens including PPD, although they are totally unresponsive to antigens of M. leprae (Fig. 4). Because antibodiesto all the knownprotein and glycoprotein antigens of M. leprae appear to be cross-reactive with homologus antigens in a variety of other mycobacteria, the immunological paradox becomes, "How is it possible to respond to the same or cross-reactive antigens when they are associated with M. tuberculosis, bacillus Cal- mette-Guerin (BCG).or othermycobacteria, yet be totally unresponsive when these same or related antigens are associated with M. leprae?" We suggested the hypothesis that there might be one or a small number of unique epitopes associated withM. leprae capable of inducing T suppressor cells that had the abilityto suppress re- sponses of potentially cross-reactive helper clones. Dr. Vijay Mehra and I developed a simple in vitromodel for examining the ability of M. leprae antigens (Dhar- mendralepromin) to induce suppressor activity(16). Lacking HLA-matched individuals with leprosy, we used M. leprae antigens to induce suppression of Con A re- sponses measured at 72hr. In over 200 patients studied, suppression was found in 84%of lepromatous and bor- derlinepatients, but not in tuberculoid patients, le- Figure3. Bystander dernvciination. Guinea pigs were sensitized to promin-positive contacts, or normal donors. Separation Freund's complete adjuvant and were challenged with PPD inoculated of the peripheral mononuclear cell populations into ad- above the sciatic nerve. Focal demyelination was produced. Top. Disrup- herent and nonadherent subsetsindicated that leproma- tion of the myelin lamellae with normal appearing axoplasm [A)of nerve adjacent to a mononuclear phagocyte (M). Bottom. Pseudopod of mono- tous and borderline patients had nonadherent cells ca- nuclear phagocyte insinuating itself between lamellae and stripping the pable of suppressing Con A responses, and a lesser pro- myelin sheath. Note ingested myelin figures are seen in phagocytic cell portion had adherent suppressorcells, presumably mac- (1 1). rophages. In collaborative experiments with Ellis Rein- There is far less information on the mechanism of herz and Stuart Schlossman,we were able to show that nerve damage in lepromatous leprosy, in which the le- the nonadherent suppressor activity was totally associ- sions indicate fibrosisand deathof sensory nerves in the ated withthe 20to 30% subset recognized by OKT5. absence of a cellular infiltrate. Perhaps themajor clue to a horse anti-human T cell serum (anti-TH2). and the pathogenesis derives from studies of Bjorvatn et al. (15) OKT8 antibody (17). When the T8 subsetfrom leproma- that showed that in lepromatous patients, particularly in tous patients was examinedfor expressionof T cell acti- reactional episodes.there wasa large excessof C3d found vation markers, namely Fc receptorsand Ia antigens, we in the circulation,suggesting that antibody-antigen com- found that approximately 50% expressedboth mono- plexes are occurring in the tissues with fixation of com- morphic HLA-DR antigens and Fc receptors, and we be- plement, probably producingedema and compression lieve these markersmay represent usefulindices of their neuropathy. functional suppressor activityin vivo (18). In our hands, Specific unresponsiveness in lepromatous leprosy. in vitro suppression was not overcome by interleukin 2 At a time in immunology when the genes and their rear-(IL 2). but removal of the T8+cells from lymphocytes of rangements that confer specificity on B and T cells are lepromatous patients resulted in very high levels of spe- being revealed, when a variety of lymphokines and im- cific antigen responsiveness to M. leprae at 6 days in munoregulatory molecules and theirreceptors have been about one-thirdof the patients,exclusively in theborder- cloned and are availablefor study andclinical trials, and line lepromatouscategory (18). This result encourages us when the major histocompatibility complex (MHC) re- to believe that the immunologic unresponsiveness in at strictions on helper and cytotoxic T cell recognition have least a proportion of lepromatous patients cannot be due been elaborated in molecular terms, in my judgment the to the total absence of antigen-responsive T cells, and most importantand fundamentalimmunologic issue that that it may be possible to overcome the suppression at remains unresolved is that of tolerance and unrespon- least in some patients withleprosy. LEARN ING FROM LEPROSY FROM LEARNING V

TUBERCULOID LEPROMATOUS Ffgure4. Selective unresponsiveness in leprosy. Left. Lepromin test in patient. Note positive indurated reaction at 28 days. Center. Lepromin test in patient with lepromatous leprosy. Note total lack of reaction. Rfght, lower arrow.Positive skin test with killed M. tuberculosfs in same lepromatous patient. The author is indebted to Dr. Tore Codal for kindly providing this figure.

The specificityof the suppressorT cells in leproma- fortunate to have available a series of chemically modi- tous leprosy. Initial studies to ascertain whether the in fied M. leprae glycolipids and structurally related glyco- vitro suppressor activity was specific only for antigensof lipids derived from other mycobacteria to probethe spec- M. leprae were confounded by relatively frequent sup- ificity of recognition of this unusual antigen by the T pression of Con A responses. even in normal individuals, suppressor cells. The resultsindicated that i)the removal induced by a varietyof species of cultivable mycobacteria of the mycocerosic acid side chains (deacylated glycolipid- added to the lymphocyte cultures. After much effort,we 1) had no effect on the in vitro suppression. ti) removal learned that most speciesof mycobacteria tested induced of the terminal 3‘ methyl group abolished the suppres- IFN-a secretion by mononuclear cells from normal, PPD- negativedonors, which correlated with the degree of CHzOY. nonspecific in vitrosuppression. Consequently, it be- came possible to examine thespecificity of suppression in lepromatous patients by testing induction of in vitro suppression by a variety of species of mycobacteria in + MYCOCEROSIC the presence of a mixture of monoclonal antibodies to three subsets of human IFN-CY(1 1). The results clearly M /eproe phenolic glycolipid - I indicated that the nonspecific in vitro suppression in- duced by killed mycobacteria in normal lymphocyte re- sponses was completely eliminated by the antibodies to on, IFN-a. Suppression was stillinduced in the lepromatous patients only by lepromin and by killed M. leprae. dem- onstrating that theT suppressor cells studied were spe- cific for antigens of M. leprae. There has thus farbeen demonstrated only one unique species of antigen associated with M. leprae. a phenolic glycolipid-1 (PGL-1) elucidated by Hunter and colleagues (19). It consists of a complex carbon backbone structure phenolically linked to a unique trisaccharide found only in M. leprae. In collaborationwith Drs. Brennan and Convit. we were able to examine the possibility that this phenolic glycolipid might stimulate suppressorcells from lepromatous patients (20).When PGL- 1. which is highly -40-30 t Y insoluble in aqueous solution, was incorporated in lipo- Lepromotour Tubwculoid Vaccinated Normals somes andadded to lymphocytesof lepromatous patients, /ContactsLapromalour it was almost as effective as lepromin in inducing sup- (50) (30) (25) (2 0) Ffgure5. Top. Structure of M. leprae phenolic glycolipid-1. It is com- pression (Fig. 5). It did not induce suppression in tuber- posed of 3.6-dl-0-methylglucose.2.3-dl-0-methylrhamnose.3-0-methyl- culoid patients, contacts. or normal donors.Depletion of rhamnose linked to phenol dimycocerosyl phthiocerol. Bottom. Suppres- sion of Con A responses of peripheral blood lymphocytes of leprosy the T8 cell subset eliminated in vitro suppression pro- patients and normals by Dharmendra leprominand by M. leprae phenolic duced both by lepromin and phenolic PGL-1. We were glycolipid- I incorporated into liposomes (1 7). vi LEARNING FROM LEPROSY sion, and iii) removal of the terminal sugar markedly Langerhans cells leading, by default, to suppressor cell reduced the suppression.No suppression wasinduced by dominance? Do the suppressor cells act by regulating IL liposomes alone or analogousglycolipids from other my- 2 or IFN-7 production or receptors for IL 2 or antigen on cobacteria differing in the terminal saccharide moiety. T helper cells, by blocking chemotaxis of T helper cells Monoclonal antibody to the terminal disaccharide com- to lesions, by blindfolding or killing T helper cells, or by pletely abolished the suppressioninduced by the phenolic killing or blocking the function of antigen-presenting glycolipid and partially inhibited that induced by M. lep- cells? I findthese questions intriguing and curiously rae. These data indicate that the T8suppressor cells in central to understanding the process of immunoregula- lepromatous leprosy have rather remarkable specificity tion. Leprosy is one of the few models in which these for the terminal trisaccharide of the phenolic glycolipid questions canbe fruitfully explored in man. of M. leprae and provides strong structuralevidence that Vaccines against leprosy. The key finding that ini- T cells are capable of recognizing carbohydrate epitopes tially permitted the possibility of producing a vaccine or possibly idiotypes reacting with them. against leprosy to be contemplated was that M. leprae Relationship of in vitro suppression to the disease can be grown in vary large amounts in the armadillo, state in leprosy. Leprosy is basically a localized disease e.g., 10' AFB/gram of tissue (24).Initially, the armadillo in which the battlebetween the immune system and thewas chosen because of its low body temperature, com- pathogen is fought out inthe tissues, primarily skin and parable with that of human skin, butit is likely that the nerves. If the T suppressor cells studied in vitro in lep- undeveloped immune system in this host contributes as romatouspatients were involved in suppressing DTH well. The first vaccine strategy was developed by one of reactions in the tissues, one might expect to find macro- the truly great experts in leprosy, Dr. Jacinto Convit in phages laden with (AFB)and a predominance of T8 cells Caracas, Venezuela, on the basisof observations hemade over T4 cells in lesions of lepromatous leprosy. Indeed earlier that killed M. leprae inoculated into the skin of this is found to be the case (2 1, 22). Robert Modlin and lepromatous patients persisted as AFB for long periods Thomas Rea have been able to isolate small numbers of of time (25). However, when live BCG was inoculated T cells from biopsies of leprosy lesions, sort them into T4 together with killed M. leprae, a granulomatous response and T8 subsets, and establish cell lines in culture.' In was produced to the BCG, and all AFB were rapidly collaborative studies with our laboratory, we were then degraded including the M. leprae. This suggested the able to establish clones that provided a unique opportu- possibility that immunization with a mixtureof live BCG nity to test the function of the cells found at the sitesof plus killed M. leprae might bring about a state of both lesions in patients. Two of these T8clones, when added cross-reactive and specific reactivity to M. leprae and either to peripheral blood mononuclear cells or to T4 thus serve as a therapeutic vaccine. clones, were able, in the presence of lepromin, to sup- Convit and his colleagues (26) have now vaccinated press Con A responses, but only when there was match- several hundred lepromatousleprosy patients with mix-a ing at HLA-D (23).Through the useof cloned cell lines, it ture of live BCG and killed purified M. leprae. Their became possible directly to ask whether T8 suppressor results have shown immunologic conversion to positive clones were capable of blocking antigen-specificre- skin test reactivity, clearance of bacilli from the skin, sponses of T4 clones and whether that suppression was histopathological upgrading towards the tuberculoid end MHC restricted. Our two T8 suppressor cloneswere able of the spectrum, and clinical improvement in approxi- to suppress proliferation of the M. leprae-specific T4 mately 85% of borderline lepromatous leprosy patients clones only when the cells shared MHC class I1 antigens. and in65% of the polar lepromatous patients. These are These results provide the first evidence for the presence rather dramatic results in patients who have been im- of suppressor cells at lesions in any human disease and munologically unresponsiveoften for long periods of provide evidence that their activity can be restricted by time. We have had theopportunity, in collaboration with Dr. Convit, to examine the in vitro suppressor activity class I1 MHC antigens. We suspect that, as in the mouse, other levels of suppressor activity will be identified, per- and activation markers on peripheral T cells from 10 such patients in a completely blind fashion, prior and haps including idiotype and anti-idiotype recognition and subsequent to immunotherapy (18). The results indicate possibly other MHC restrictions. Overall, the results sup- port the view that the spectrum of leprosy represents a that in vitro suppressor activity in all successfully vac- cinated patients disappeared after immunotherapy, and subtle regulatory balance of T helper and T suppressor expression of HLA-D was reduced to normal levels in cells within lesions that determines the ultimate course seven of eight patients examined. These results provide of the disease. objective evidence for immunological changes brought Nevertheless, these findings raisemore questions than about by this vaccine, and establish a correlation be- they answer. The basic question becomes what factors tween the in vitro T cell suppressor activity and the determine the predominant development of suppressor degree of immunologic unresponsiveness in vivo. Thus, cells in a very small percentage of people infected with leprosy remains theonly example of specific immunolog- M. leprae. Is it a matter of genetic or environmental ical unresponsiveness in man that can, with reasonable predisposition of the individual, variations inthe genetics success, be overcome by immunologic intervention. It is of the organism, the route of infection, escape from cy- important to tryto develop an understanding of the mech- tocidal mechanisms of host macrophages, or subversion anism by which this is accomplished. Theability of BCG- of antigen-presenting macrophages, dendritic cells, or responsive T cells to produce IL 2 and IFN-7 locally is Modlin. R. L.. V. Mehra, L. Wong. Y. Fujimiya, W-C. Chang. D. A. likely to be important. Horwitz, B. R. Bloom, T. H. Rea, andP. K. Pattengale. 1986.Suppressor T lymphocytes from lepromatous leprosy skin lesions. Submittedfor Although it would have been inconceivable even 5 yr publication. ago, there are currently threemajor controlled preventive LEARNING FROM LEPROSY vii vaccine trials against leprosy in progress or about to get clones it was possible to screen crude phage lysates of underway involving 450,000 people in three partsof the the lysogens expressing major antigens and to identify world. largely as a result of the work of the Committee four human T cell clones that recognized one of the on the Immunology of Leprosy (IMMLEP) of the UNDP/ proteins (18,000 daltons) (Fig. 6). This approach should World Bank/WHO Special Programme for Research and permit testing for T clone recognition in other systems Training in Tropical Diseases. On the assumption that by using crude bacterial lysates withoutthe need to purify the combined vaccine or a cross-reactive cultivable my- the antigens. Finally,by using the computer algorithm of cobacteria will be as effective as M. leprae in providing DeLisi and Berzofsky (34), predictions of amphipathic immunityto noninfected individuals and likely to be structures likely to be recognized by T cells on the 65.000 therapeutic in those in thepopulation who have subclin- dalton M. leprae protein have been made, and one pre- ical infection, all three trials are testing the efficacy of dicted peptide has been synthesized by Peter Kim and killed M. leprae + BCG. However, even if the armadillo- found by Johanne Melancon in my lab to be recognized derived vaccine should be effective,there is real question by T cell clones from lepromin-positive donors. From whether itscost and theinevitable shortageof the bacilli these molecular approaches, we hope to be able to develop would permit it to be widely used. For this reason, it is more sensitive sero-diagnostic and skin test reagents that important to search forpotentially protective antigens in could be used on a wide scale for detecting infection by M. leprae. Because M. leprae has not been cultivated in M. leprae and possibly predicting the form of disease vitro. biochemical studies of M. leprae antigens present while infection is still subclinical. aunique challenge. To meet this challenge, we were Vaccines and the Third World. Leprosy is but one of fortunate enough to be ableto collaborate withDrs. Rich- many diseasesthat selectively exact major morbidity and ard Young and Ron Davis who had just developed the X mortality in the Third World. Five million children die gt 11 expression system. In collaboration with Josephine eachyear of the simplevaccine-preventable illnesses and Roy Curtiss, who had just worked out methods for diphtheria, whooping cough, tetanus, polio, measles, and preparing high m.w. DNA from M. leprae, and Clemens tuberculosis (1) (Table I). Seven and one-half million die Grosskinsky in my lab, Rick Young produced recombi- annually from diarrheal diseases, and 4.5 million die nant X gtl 1 genomic libraries toM. tuberculosis (27) and from respiratory diseases (35).Five hundred million peo- M. leprae (28) that arefully representative of the entire ple suffer from malnutrition. With the development of genome. Although M. leprae cannot be grown, it was very monoclonal antibodies and recombinant DNA technol- exciting to knowthat its genes could be propagated infi- ogy, it is possible to identify and produce protective an- nitely and expressed in a vicarioushost, E. coli. tigens and to modify the genetics of organisms toproduce Although I indicated earlier that all the known protein effective recombinant vaccines. and glycoprotein antigens of M. leprae appear to cross- Yet even with the simple childhood vaccines there re- react with antibodies to homologous proteins in other main the major problems of timing and delivery. In the species of mycobacteria (29).the development of mono- case of DPT, in many countries 60% of the population clonal antibodieshas permitted the identification on each receives one DPT shot butonly 20%receive the full course of the major proteins of unique epitopes possessed only of three immunizations (1). The cost of vaccines from by M. leprae or shared with only a few other species (30, development to production represents only 14% of the 3 1). Screeningof the M. leprae library with38 monoclon- cost of global immunization, the remaining 86% being als permitted the identification of X-phage expressing the thecosts associatedwith the healthinfrastructure, five major protein antigens of M. leprae of m.w. 55,000 transport, and cold chain required for effective delivery, to 65,000, 36,000, 28,000, 18,000, and 12,000.Because which must generally be borne by the developing country multiple monoclonal antibodies exist to the largestof the (36) (Fig. 7). Yet vaccines are probably the most cost- polypeptides, it seemed possible to attempt to identify effective public health measure, other than hand wash- epitopes recognized by specific and cross-reacting mon- ing, available tothe ThirdWorld. Inthe caseof smallpox, oclonals and by T cells by using recombinant DNA tech- the expenditure of $46 million over a period of 12 yr nology. To do this, a mini-library was constructed by resulted in a vaccine that saves the United States an Vijay Mehra and Rick Young of fragments of approxi- estimated $500 million each year in not having to vacci- mately 250 to 1000 base pairs from the gene encoding nate all overseas travellers. The Instituteof Medicine has the 65,000 dalton protein. These were screened with six recently developed criteria for evaluating the cost-bene- monoclonals antibodies, and from the pattern of clones fits of vaccines, and has developed a sophisticated ana- reacting with a given antibody it was possible to predict lytical framework for setting vaccine priorities relevant the sites of individual epitopes (32). Because they ob- to the United States and the Third World (37). They tained the DNA sequence of the entire gene, an amino developed a means of comparing acutediseases and acid sequence for each of the fragments of the “epitope chronic diseases by normalizing to a value termed the library” was predicted. “infant mortality equivalent” (IME), Le.,values that rep- Peptides are now being synthesized, and two have al- resent the number of acute morbidity days or chronic ready been found to bind the appropriate monoclonal cases considered to be equal in undesirability theto death antibodies. The identification of these polypeptide spe- of an infant. A simplified example of the kind of priority cies containing unique epitopes permitted, in collabora- decisions that could be made on the basis of how much tion with Salim Mustafa and Tore Codal in Norway, the one is prepared to spend perIME is given in Table I1 (38). development of a simplemethod for testing whether these One of the major conceptual breakthroughs for immu- antigens could be recognized by T helper cells, whichare nizing vast numbers of children in the Third World was most likely to be critical for providing protection against the development of multi-vaccine vehicles by Moss and leprosy (33). Throughthe use of M. leprae-specific Tcell Paoletti in vaccinia (39, 40). It has been possible to im- viii LEARNING FROM LEPROSY RESPONSE OF T CELL CLONES TO RECOMBINANT hgtll LYSATES

CI Soluble M. kpme Extrscl Phage Lysate Protein IV3178 lysmta - 64K IY3160 lysate - 36K IV31641ysale 28K Figure6. T cell clones derived from 30'000 . volunteers vaccinated with killed M. lep- E3 V31791ysale - 18K rae screened against crude phage lysates 0 V3184lynala - 12K of X gtl 1 lysogens expressing major anti- 25,000 gens of M. leprae. Four M. leprae-specific T cell clonesresponded to the lysogen Y3179 expressing the 18,000dalton pro- tein (34). 10,000

5,000

0 ATT2/4F ATTEIlOF NHH212F ATT4/1F I ATT2I / EE Speclflc T Call Cloner 5

TABLE I TABLE I1 Vaccine-preventable mortality Total disease burden values and vaccine priorities" " r Deaths Total Disease Priorityb MillionsIYear Burden (cost per IME averted) Disease I. Currently existing vaccines Value ([ME Unrestricted $10.000 Neonatal tetanus 1.14 Measles 2.60 pneumoniaStreptococcal 6.612.261 1 1 Pertussis 0.84 Hepatitis B 2,394,256 6 - 4.75 Malaria 2.111.795 3 4 Typhoid 1,308,121 4 3 11. Vaccines under development Escherichia coli (ETEC) 978,248 13 7 Diarrheal diseases 7.50 Rotavirus 925,042 2 2

Acute respiratory illness ~ 6.50 828,068 5 5 14.00 (Group A) infection 81 1,477 8 9 Leprosy 657,349 11 11 Hemophilus influenzae type B 471,336 7 7 Breakdown of Costs For a Fully Cholera 229.217 10 8 "Infant mortality equivalent (IME) values represent the number of Immunized Child acute morbidity days or chronic cases consideredto be equal in undesir- ability to the deathof an infant(39). bPriorities can be assigned to specific vaccines by considering the following factors: 1) vaccine practicability (numberof strains, antigenic Salaries of health personnel drift/shift), ii) target population selection (cost effectiveness),iii) propor- tion of illnessfalling in target populations (including opportunity to deliver at optimal age), iu) efficacy (including duration of immunity), u) capacity to deliver vaccine (health system functionality),ui) characteris- tics of vaccine (stability),uii) utilization (including provider and lay atti- tudes, and numberof doses needed).uiii) amount of funds oneis prepared to allocate to avert oneIME relative to the costof each vaccine (39). OBCG The Ideal Vaccine Mleprue HLalshrnonta aM/YbblCU/sis equipment supplies 1% Figure 7. Breakdown of costs fora fully immunized child in theWHO Expanded Program for Immunization(39). munize animals with recombinant vaccinia to produce antibodies against four independent antigens simulta- neously. Recombinant vaccines capable of providing si- 0 DNA multaneous protection against Salmonella and Shigella &RNA are being developed (41). It is our hope to develop a OrB protein recombinant mycobacterial vaccine that is capable of Figure8 Conception of arecombinant mycobacterial vaccine ex- engendering cell-mediated immunity and antibodies to pressing specific protective antigensfor other infectious agents. multiple antigens (Fig. 8). A recombinant mycobacterial vaccine would have a number of unique advantages: i) it minimal toxicity (42)(reported fatal complications occur is a superb adjuvant, perhaps the most effective known at a frequency of about 60/billion), which is about 100- for inductionof cell-mediated immunity, ii)BCG has been fold less than that of vaccinia, iii) it is the only vaccine used in more than 4 billion people around theworld with that can be given at birth, iu) it requires only a single LEARN ING FROM LEPROSY FROM LEARNING ix and effectively immunize to produce long-term but re- versible reduction in fertility. Perhaps it will be possible some day simply to immunize against babies. The second impediment is a matter of will and com- mitment. When thiscountry spends (Table 111) vast amounts of funds on national security (this year's appro- priation request is $3 17 billion and undoubtedly compa- rable amounts are spent by the Soviet Union), and our """r """r people spend almost $40 billion annually merely on al- Figure 9. Spacing between births in relation to infant and childhood coholic beverages, tobacco, and toys, one wonderswhy it deaths in Bangladesh (1). is so difficult to raisethe $300 million that would permit TABLE 111 every childin the world to be vaccinated now against the A perspectfue on prforftfes" major childhood diseases. Of the $5 billion spent on ~- ." " - - . .- - . . biomedical research, less than one-fiftieth is spent on Defense appropriation $2 10 billion Alcoholic beverages $22 billion diseases of greatest concern to the developing countries Tobacco $12 billion by all agencies of the U.S. government.* It is a telling Toys $4 billion fighter planeF- 18 fighter $32 million indictment of the priorities of the developed nations that the entire WHO Special Programme for Research and Biomedical research $4.5 billion Training in Tropical Diseases still receives less funding Tropical diseases research $100 million WHO tropical diseases programme $25 million for research in immunology, chemotherapy, and epide- miology of six major tropical diseases than the cost of a WHO/UNICEF expanded programme for $300 million single jet fighter plane (46). One wonders how much of immunization (Drowsed 19861 the preoccupation of world leaders with "national secu- a Figures obtainedfrom fiscal years 1982 and 1983. rity" actually makes a significant positive impact on the dose, u) it can engender DTH that persists for 5 to 50 yr, "personal security"and daily survival of the people in the and vi) it costs $0.55/dose. One of the major difficulties poorest nations. One wonders also to what extent the one might have anticipated inthe development of such a political tensions in the Third World arise from despair vaccine would be introducing DNA through the waxy at thedisparity in the qualityof life there relative to life coat, glycolipids. and peptidoglycan cell wall structure of in developed countries. mycobacteria. Bill Jacobs in my laboratory (43) has re- The preamble to the charterof the World Health Orga- cently developed methods for spheroplasting and trans- nization states that "Access to the highest attainable fecting mycobacteriophage DNA into a cultivable myco- standard of health is a basic human right". We as im- bacterium, M. smegmatis. These experiments give hope munologists and as Americans have important contri- to the possibility of introducing appropriate vectors with butions, and commitments,to make. selectable markers expressing foreign genes into some species of cultivable mycobacteria, preferably BCG. An REFERENCES ideal vaccine would probably require, in addition to the 1. United NationsChildren's Emergency Fund (UNICEF). 1985. The adjuvant effect of the mycobacterium, both specific neu- State of the World's Children 1985. Oxford University Press. P. 131. 2. Bloom, B. R., and B. Bennett. 1966. Mechanism of a reaction fn tralizing or protective determinantsof the foreign path- uitro associated with delayed-type hypersensitivity. Scfence153230. ogen and a T helper cell-recognized epitope of the patho- 3. David, J. R. 1966. Delayed hypersensitivity in uitro: its medtation gen to confermemory that will be stimulated by infection. by cell free substances formed by lymphoid cell interaction. Proc. Natl. Acad. Scl. USA 56:72. If highly effective and low cost vaccines were devel- 4. Bennett, B.. and B. R. Bloom. 1968. Reactions in vivo produced by oped, there are two major impediments to their use. One a soluble substance associated with delayed-type hypersensitivity. is the view in the developed countries that immunization Proc. Natl. Acad. Scf. USA 59:756. 5. Thurman, G. B.. I. A. Braude. P. W. Gray, R. K. Oldham, and H. C. of children will only lead to more babiesand simply Stevenson. 1985. MIF-like activity of natural and recombinant hu- exacerbate the shortages of food and resources in the man interferon-7 and their neutralization by monoclonal antibody. J. Immunol. 134:305. Third World, and hence should not be supported. There 6. Nathan. C. F.. G. Kaplan, W. R. Levis. A. Nusrat, M. D. Witmer. S. are a number of objections to this line of argument, of A. Sherwin. C. K. Job, C. R. Horowitz, R. M. Steinman. and 2. A. which I shall mention only two. The first is a matter of Cohn. 1986. Local and systemic effectsof low doses of recombinant interferon-gamma after intradermal injection in patients with lep- historical fact:birth rates in no countryof the world have romatous leprosy. N. Engl. J. Med. In press. decreased prior to a decrease in death rates. It is simply A report of the Office of Science and Technology Planning ("Status of not possible effectively to persuade people not to have Biomedical Research and Related Technology for Tropical Diseases": large families unless one can demonstrate that children OTA-H-258. US. Govt. Printing Office, 1985) prepared by the Congres- will survive with sufficient frequencyto care for themin sional Office of Technology Assessment indicates that in 1982 the Na- tional Institute forAllergy and Infectious Diseases spent$33 million for old age. This persuasionmay nottake generations. There tropical medicine and approximately $3 million for projects on diarrheal are datafrom Bangladesh (Fig. 9) indicating that reducing diseases and acute respiratory illnesses.a large proportion of which was infant and child mortality has a profound effect in in- for influenza research directed at domestic problems. The Centers for Disease Control spends about$5 million on tropical disease research and creasing the spacing between births (1). Finally, it may medical entomology. The agency for International Development funds well be that immunology has aneven more crucial role about $15 million and the Department of Defense about $12.5 million. The budget of the WHO Special Programme for Researchand Trainingin to play. Although not widely known by immunologists, Tropical Diseases has remained essentially constant at approximately there aretwo vaccines containing antigen determinants $25 million for 4 yr. and in 1981 approximately one-third of its research from the @-chainof human chorionic gonadotrophin (44, funds. somewhat greater than the US. contributes to the program, was awarded to American scientists. TheWHO program In diarrheal diseases 45). a hormonerequired for implantation, which are awarded 25% of its research funds to American institutions. butU.S. the about to undergo clinical trials to learnif one can safely contribution represented only1% of the program's totalbudget. X LEARNING FROMLEPROSY

7. Hansen, G. A. 1874. Undersogelser angaende spedalskhedens arsa- lepraeand BCG in different forms of leprosyand inMitsuda-negative ger. Norsk Magasin for Laegevitenskapen. 4: 1. For an English ver- contacts. lnt. J. Lepr. 50:415. sion of his major work, see: Hansen. G. A. and C. Looft. (transl. N. 27. Young, R. A., B. R. Bloom. C. M. Grosskinsky, J. Ivanyi, and R. W. Walker). 1895. Leprosy: In its Clinical and Pathological Aspects. Davis. 1985. Dissecting Mycobacterium tubercuLosis antigensusing John Wright & Co. Bristol (reprinted in 1973)P. 162. recombinant DNA. Proc. Natl. Acad. Sci. USA 82:2583. 8. Brody, S. N. 1974. The Disease of the Soul. Leprosy in Medieval 28. Young, R.A.. V. Mehra. D. Sweetser, T. M. Buchanan. J. Clark- Literature. Cornell University Press. Ithaca. P. 223. Curtis. R. W. Davis, and B. R. Bloom. 1985. Genes for the major 9. Bloom, B.R., and T. Godal. 1983. Selective primary health care: protein antigens of . Nature31 6~450. strategies for control of disease in thedeveloping world. V. Leprosy, 29. Harboe. M., 0. Closs. G. Bjune, G. Kronvall, and N. H. Axelsen. Rev. Infect. Dis. 5:765. 1978. Mycobacteriumleprae specific antibodies detected by ra- 10. Bloom, B. R., and V. Mehra. 1984. Immunological unresponsiveness dioimmunoassay. Scand. J. lmmunol. 7~111. in leprosy. lmmunol. Rev. 80:5. 30. Gillis, T. p.. and T. M. Buchanan. 1982. Production and partial 1 1. Wisniewski, H. M.. and B. R. Bloom. 1975. Primary demyelination characterization of monoclonal antibodies toMycobacterium leprae. as a non-specific consequence of cell-mediated immune reactions. J. Infect. Immun. 37:172. Exp. Med. 141:346. 31. Engers. H. D.. M. Abe, B. R. Bloom. V. Mehra, W. Britton. T. M. 12. Cammer, W.,B. R. Bloom. W. T. Norton. and S. Gordon. 1978. Buchanan, S. K. Khanolkar, D.B. Young, 0. Closs, T. Gillis, M. Degradation of basicprotein in myelin by proteasessecreted by Harboe, J. Ivanyi, A. H. J. Kolk. and C. C. Shepard. 1985. Work- stimulatedmacrophages: a possible mechanism of inflammatory shop: results of a World Health Organization-sponsored workshop demyelination. Proc. Natl. Acad. Sci. USA75: 1554. on monoclonal antibodies to Mycobacterium leprae. Infect.Immun. 13. Brosnan, C. F., W. Cammer. W. T. Norton, and B. R. Bloom. 1980. 48:603. Proteinase inhibitors suppress the development of experimental al- 32. Mehra. V., D. Sweetser,and R.A. Young. 1986. Finemapping lergic encephalomyelitis. Nature 285:235. protein epitopes of Mycobacterium lepraeusing recombinant DNA. 14. Brosnan. C. F., G. L. Stoner, B.R. Bloom, and H. M. Wisniewski. Proc. Natl. Acad. Sci, USA.In press. 1977. Studies on demyelination by activated lymphocytes in the 33. Mustafa. S., H. K. Gill, A. Nerlund, W. J. Britton, V. Mehra, B. R. rabbit eye. 11. Antibody-dependent cell-mediated demyelination. J. Bloom, R.A. Young, and T. Godal. 1986. Human T cell clones Immunol. 118:2103. recognize a major M. leprae protein antigen expressed in E. coli. 15. Bjorvatn, B., R. S. Barnetson, G. Kronvall, R. H. Zubler, and P-H. Nature 319:63. Lambert. 1976. Immune complexes and complement hypercatabo- 34. DeLisi, C., and J. A. Berzofsky. 1985. T-cell antigenic sites tend to lism in patients with leprosy. Clin. Exp. lmmunol. 26: 128. be amphipathic structures. Proc. Natl. Acad. Sci. USA 82:7048. 16. Mehra, V.,L. H. Mason, J. P. Fields. and B.R. Bloom. 1979. 35. Walsh, J. A., and K. S. Warren. 1979. Selective primary health care: Lepromin-induced suppressor cells in patients with leprosy. J. Im- an interim strategy for disease control in developing countries. N. munol. 123:1813. Engl. J. Med. 301:967. 17. Mehra, V.,L. H. Mason, W. Rothman, E. Reinherz. S. F. Schlos- 36. Henderson. R. 1984. An example of vaccineapplication: the ex- man, and B. R. Bloom. 1980. Delineation of a human T cell subset panded programme on immunization. In New Approaches to Vac- responsible for lepromin-induced suppression in leprosy patients. J. cine Development. R. Bell and G. Torrigiani. eds. Schwabe & Co.. lmmunol. 125:1183. Basel. P. 506. 18. Mehra, V., J. Convit, A. Rubinstein, and B. R. Bloom. 1982. Acti- 37. Institute of Medicine. 1985. New vaccine development: establishing vated suppressor T cells in leprosy. J. lmmunol. 129: 1946. priorities. Volume 1. Diseases of Importance in the United States. 19. Hunter, S. W., T. Fugiwara, and P. J. Brennan. 1982. Structure and Natl. Acad. (U.S.) Press. P. 210. antigenicity of the major specific glycolipid antigen of Mycobacte- 38. Institute of Medicine. 1986. New vaccine development establishing rium leprae. J. Biol. Chem. 257:15072. priorities. Volume 2. Diseases of Importance in the Developfng 20. Mehra, V.. P. J. Brennan, E. Rada, J. Convit, and B.R. Bloom. World. Natl. Acad. (U.S.) Press. In press. 1984. Lymphocyte suppressionin leprosy induced by unique M. 39. Mackett, M., G. L. Smith, and B. Moss. 1982. Vaccinia virus: a leprae glycolipid. Nature 308: 194. selectableeukaryotic cloning and expression vector. Proc.Natl. 21. Modlin, R. L., F. M. Hofman. C. R. Taylor, and T. H. Rea. 1983. T Acad. Sci. USA 797415. lymphocyte subsets in the skin lesions of patients with leprosy. J. 40. Perkins, M. E., A. Piccini, B.R. Lipinskas, and E. Paoletti. 1985. Am. Acad. Dermatol.8~181. Recombinant vaccinia virus: immunization against multiple patho- 22. Van Voorhis. W. C., G. Kaplan. E. N. Sarno, M. A. Horwitz, R. M. gens. Science 229~981. Steinman, W. R. Levis, N. Nogueira. L. S. Hair, C. R. Gattass. B. A. 41. Formol, S. B., and T. L. Hale. 1984. Immunization against shigellosis Amck. and2. A. Cohn. 1982. The cutaneous infiltratesof leprosy. and the concept of vector strains to carry antigens. In New Ap- Cellular characteristics and the predominant T cell phenotypes. N. proaches to Vaccine Development. R. Bell and G. Torrigiani, eds. Engl. J. Med. 307: 1593. Schwabe & Co.. Basel. P. 178. 23. Modlin, R. L., H. Kato. V. Mehra, E. E. Nelson, X-D. Fan, T. H. Rea, 42. Lotte, A.. 0. Wasz-Hockert, N. Poisson, N. Dumitrescu, M. Vernon, P. K. Pattengale, and B.R. Bloom. 1986. Genetically restricted and E. Couvet. 1984. BCG Complications. Adu. Tuberc. Res.21:107. suppressor T cell clones derived from lepromatous leprosy lesions. 43. Jacobs, W. R., and B. R. Bloom. 1986. Transfection of Mycobacte- Nature In press. rium smegmatis spheroplasts with mycobacteriophage DNA.Am. 24. Kirchheimer, W. F., and E. E. Storrs. 1971. Attempts to establish Soc. Microbiol. Abstr. Annual Meeting U-16, P. 122. the armadillo (Dasypusnouemcinctus Linn) as a model for the study 44. Stevens, V. C. 1986. Use of synthetic peptides as immunogens for of leprosy. Int. J. Lepr. 39:693. developing a vaccineagainst humanchorionic gonadotropin. In Syn- 25. Convit. J., M. E. Pinardi. G.Rodriguez-Ochoa. M. Ulrich, J. L. Avila, thetic Peptides as Antigens. Ciba Foundation Symposium 119. P. and M. Goihman-Yahr. 1974.Elimination of Mycobacterium leprae 200. subsequent to local in vivoactivation of macrophages in lepromatous 45. Talwar, G. P. 1984. Structural vaccines to control fertility and com- leprosy by other mycobacteria. Clin. Exp. Immunol. 17:261. municable diseases. Crit. Rev. Trop. Med.2~245. 26. Convit. J., N. Aranzazu. M. Ulrich, M. E. Pinardi, 0. Reyes. and J. 46. UNDP/World Bank/WHO. Tropical Disease Research. Seventh Pro- Alvarado. 1982. Immunotherapy with a mixture of Mycobacterium gramme Report. TDR Seventh Report. P. 265.