INTERNATIONAL JOURNAL OF LEPROSY^ Volume 56, Number 2

Printed in ihe U.S.A.

Identification of T-cell-activatingRecombinant Antigens Shared Among Three

Candidate Antileprosy Vaccines, Killed M. leprae,

M. bovis BCG, and Mycobacterium 14/1Abu Salim Mustafa2

Leprosy is a chronic mycobacterial dis-ease caused by Mycobacterium leprae. Re-sistance against leprosy correlates with cell-mediated immune (CMI) responses to theinfectious agent. Killed Al. leprae purifiedfrom infected armadillo tissue are highlypotent in inducing CMI responses in animal

(14, 22)models and in human volunteers (5 6).Vaccination with killed Al. leprae protectsmice from viable Al. leprae challenge (24).These characteristics of killed AI. lepraealong with its nontoxic and nonpathogenicproperties make it an ideal candidate for anantileprosy vaccine. However, since Al. lep-rae has not been grown in vitro, infectedarmadillo tissue remains the major sourceof supply of the bacilli. In addition to highcost, this source may not provide enoughbacteria required for vaccinating all of thepeople who are at risk of infection. There-fore, attempts are also being made to de-velop antileprosy vaccines based on culti-vable mycobacteria, e.g., Al. boils BCG andMycobacterium w, etc. Vaccination of ani-mals with BCG (24) and Alycobacterium(20) results in the induction of M. leprae skin-test reactivity. Immunization of mice withlive BCG protects them from viable Al.leprae challenge (23). Vaccination of Al. leP-rae skin-test-negative lepromatous patientswith Mycobacterium w converts about 60%of the patients into an Al. leprae skin-test-positivity status and also results in clinicalimprovements (2)• The above effects are saidto be due to the sharing of those antigens

' Received for publication on 28 December 1987;accepted for publication on 12 February 1988.

= A. S. Mustafa, Ph.D., Laboratory for Immunology,Institute for Cancer Research, The Norwegian RadiumHospital, Montebello NO310, Norway.

Dr. Mustafa's present address: Whitehead Institutefor Biomedical Research, Nine Cambridge Center,Cambridge, Massachusetts 02142, U.S.A.

of M. leprae by BCG and Mycobacterium wwhich are involved in CMI functions. How-ever, the identity of such antigens is largelyunknown.

In the present study, some of the antigensshared among Al. leprae, BCG, and Myco-bacterium w have been identified by testingthe recombinant antigens of Al. leprae andM. tuberculosis against human T-cell linesand clones raised from BCG and killed-M./eprae-vaccinated subjects.

MATERIALS AND METHODSAntigens. Killed M. leprae were kindly

supplied by Dr. R. J. W. Rees through theWHO/IMMLEP Bank. BCG was obtainedfrom Serum Institute, Copenhagen, Den-mark. Killed Mycobacterium w was a kindgift from Professor G. P. Talwar, NationalInstitute of Immunology, New Delhi, India.These mycobacteria were used at concen-trations optimal for T-cell proliferation, i.e.,Al. leprae at 5 x 10' bacilli/ml, BCG at 10mg/m1 (wet weight) and Mycobacterium w at5 x 106 bacilli/ml.

The recombinant antigens of Al. lepraeidentified by monoclonal antibodies whichrecognize Af. leprae antigens of molecularweight 65 kilodalton (kDa), 36 kDa, 20 kDa,18 kDa, and 12 kDa (30); the Al. leprae 13B3recombinant antigen identified from the re-combinant AI. leprae Xgt 1 1 library using hu-man T-cell clones as probes (A. S. Mustafa,et al., submitted for publication); and re-combinant AI. tuberculosis antigens identi-fied by monoclonal antibodies which rec-ognize Al. tuberculosis antigens of molecularweight 65 kDa, 19 kDa and 14 kDa (2") wereused in the present study. Escherichia colilysates containing or lacking recombinantmycobacterial antigens were prepared ac-cording to the protocols described earlier

265

266^ International Journal of Leprosy^ 1988

(I 5'2'). The lysates at a protein concentra-tion of 50 pg/ml were used for T-cell pro-liferation assays.

T-cell clones. Forty-two Al. leprae-in-duced, CD4+ T-cell clones were raised fromthe peripheral blood mononuclear cells(PI3MC) of BCG-vaccinated and killed-M./cprac-immunized healthy volunteers (5).Eleven of these T-cell clones were Al. lepraespecific and the remaining crossreacted withother mycobacteria. Three of the Al. !twat,-specific and 12 crossreactive T-cell cloneshave been used in the present study.

T-cell lines. To establish T-cell linesfrom BCG-vaccinated and killed-Ai. /cmrae-immunized healthy subjects (5), 2 x106 PBMC/ml complete medium (RPMI1640 + 15% AB serum + 1% penicillin-streptomycin) were stimulated with optimalconcentrations of Al. /cprac, I3CG, and My-cobacterium w in the wells of 24-well Costarplates (Costar, Cambridge, Massachusetts,U.S.A.). The plates were incubated for 6days at 37°C in an atmosphere of 5% CO2and 95% air. To expand the antigen-reactivecells, starting from day 6 recombinant in-terleukin-2 (IL-2) (a gift from Cetus Cor-poration, Emoryville, California, U.S.A.)was added to the cultures at 100 U/ml twicea week. After 4-5 weeks, the T-cell lineswere tested for antigen responsiveness inproliferative assays.

Proliferative assay. Autologous adher-ent cells from 1 x 105 irradiated (2500 R)PBMC added to the wells of 96-well Costarplates served as antigen-presenting cells. Tenthousand cells of Al. /eprae-induced, CD4+human T-cell clones or 1 x 105 cells ofT-cell lines raised against M. leprae, BCG,and Mycobacterium w were added to thewells. Experimental cultures were stimulat-ed with antigens in triplicates. The controlcultures did not have antigen. To assess theresponse against recombinant antigens, theexperimental cultures were stimulated withE. coli lysates containing recombinant an-tigens, and the cultures with E. coli lysateslacking recombinant antigens were taken ascontrols. The plates were incubated at 37°Cin a humidified atmosphere of 5% CO, and95% air. On day 3, 0.045 MBqdine (specific activity 185 x 103 MBq/mmole) was added to each well. The plateswere further incubated for 4 hr at 37°C. The

cultures were harvested, and 31-1-thymidineincorporation was assessed by standard pro-cedures (). Median values of counts perminute (cpm) from triplicate have been usedto express the results. The proliferative re-sponse to a given antigen was consideredpositive when a T-cell clone or a T-cell lineshowed Acpm 1000 and T/C > 2, andsuch values in the tables are underlined,where Acpm = (Median cpm in experimen-tal) — (Median cpm in control cultures) and

Median cpm in

T/C —experimental cultures

Median cpm in control cultures

Cytotoxicity assay. The neutral red up-take assay as described earlier (17'1)) wasused to assess the cytotoxic potential ofT-cell clones against antigen-pulsed adher-ent cells. In brief, autologous adherent cellsfrom 1 x 106 irradiated PBMC in the in-dividual wells of 24-well Costar plates pulsedwith Al. leprae, BCG, and Alycobacteriumiv served as targets. Control wells did nothave antigen. The effector cells were M. lep-rae-induced CD4+ human T-cell clonesadded at 1 x 105 cells/well. The plates wereincubated at 37°C in a humidified atmo-sphere of 5% CO, and 95% air. On day 6,nonadherent cells were washed off, and theremaining macrophages differentiated fromadherent monocytes were allowed to takeup the neutral red. The dye taken up bymacrophages was released in a solution of0.05 M acetic acid in 50% ethanol and wasquantitated by measuring 0D540 on a spec-trophotometer. The results are expressed interms of percentage cytotoxicity which isdefined as: % cytotocixity =

OD540 control — OD540experimental

OD540 control

where 0D540 control = OD540 of cultureswith adherent cells + antigen, OD540 ex-perimental = OD540 of cultures with ad-herent cells + T-cell clone + antigen.

RESULTSResponses of M. leprae-induced T-cell

clones. Fifteen M. /eprae-induced CD4 +

x 100

56, 2^Mustafa: T-cell-activating Recombinant Antigens^267

TABLE 1. Proliferation of M. leprae-in-duced, CD4+ T-cell clones to M. leprae,BCG, and Mycobacterium w.

Proliferation (cpm x 10 3) in response to

T-cellclones"

Noanti- lepraegen

BCG

Myco-bac-

terium

Al. kin-ay-specificATT 6/10P 0.1^13.3 0.1 1.4NI-11-1^1/E7 0.4^11.7 0.1 0.2NHH 5/1D 0.1^1.2_ 0.1 0.1

CrossreactiveJM A7" 0.3^14.7 28.1 18JM 2C6 0.2^10.9 12.9 0.1JM 3C7 0.7^11.4 11.6 6.9ATT 4/IF 0.1^14 412 L6ATT 5/213 0.1^54.2 41.8 27.4ATT 5/613 0.1^3.3 17.8 1.1ATT 4/4C 0.1^39.8 49.4 0.2ATT 2/3D" 0.1^35.2 50.7 0.1ATT 1/6B 0.1^20.8 75.7 0.1ATT 6/9B 0.1^64.1 69.4 0.2AF 6/C8 0.1^2.0 34A 13AF 5/C10 0.1^1.8 1.8 0.5

Positive/tested 15/15 12/15 7/15

" ATT, NHH, JM and AF are the subjects whosePBMC were used to establish the T-cell clones.

'The T-cell clones ATT 6/10F, JM A7, and ATT2/3D, proliferated in response to the I8-kDa recom-binant antigen of M. leprae ('s), 65-kDa recombinantantigens of M. frpracand M. tuberculosis (''), and 13133antigen of M. leprae (A. S. Mustafa, et al., paper sub-mitted for publication), respectively.

c A clone showing A cpm 1000 and T/C > 2 inresponse to a given antigen was considered positive.Such values arc underlined.

T-cell clones from four BCG-vaccinated,killed-M. /eprae-immunized subjects weretested for their proliferative responses to Al.leprae, BCG, and Alycobacterium w. Threeof these T-cell clones were Al. leprae specificand others in addition showed reactivity toother mycobacteria (data not shown and").None of the M. /eprae-specific T-cell clonesresponded to BCG, but the clone ATT6/10F which recognizes Al. leprae 18-kDaprotein (15) showed a weak response to My-cobacterium w (Table 1). All of the cross-reactive T-cell clones responded to Al. lep-rae and BCG; only five of these clonesshowed response to Mycobacterium w, in-cluding the T-cell clone JM A7 (Table 1)

TABLE 2. Cvtotoxicity of M. leprae-in-duced CD4+ T-cell clones against adherentcells pulsed with M. leprae, BCG, and My-cobacterium w.

(3/o cytotoxicity of adherentcells pulsed with

T-cell clonesM. lepiae BCG ,Ilycobac-

termut w

ATT 6/10F 96 3 87JM A7 76 100 60JM 3C7 60 98 88ATT 4/1F 87 100 92ATT 6/9B 98 100 -13

which recognizes Al. leprae and Al. tuber-culosis 65-kDa antigen (19). The 13B3 Al.leprae antigen-reactive T-cell clone ATT2/3D (A. S. Mustafa, et al., submitted forpublication) proliferated to Al. leprae andBCG but not to Mycobacterium w (Table1).

Five Al. /eprae-induced T-cell clones weretested for cytotoxicity against Al. leprae-,BCG-, and Mycobacterium ti-pulsed mono-cytes. A correlation was found between pro-liferative and cytotoxic activities of theseclones. The T-cell clones A7, 3C7, and4/1F, which proliferated to Al. leprae, BCG,and Mycobacterium iv, were also cytotoxicfor monocytes pulsed with them (Table 2).The T-cell clone 6/10F did not proliferateto BCG and did not kill BCG-pulsed mono-cytes (Table 2). The T-cell clone 6/9B didnot proliferate to Mycobacterium w (Table1) and could not kill Mycobacteriumit-pulsed monocytes (Table 2).

Response of M. leprae-, BCG-, and My-cobacterium w-induced T-cell lines. M. lep-rae-, BCG-, and Mycobacterium w-inducedT-cell lines were raised from 7, 8 and 5BCG-vaccinated and killed-M. leprae-im-munized subjects, respectively. All of theseven T-cell lines raised against Al. lepraeresponded to Al. leprae and Mycobacteriumw, and six of these T-cell lines responded )BCG (Table 3). All of the five T-cell linesraised against Mycobacterium iv respondedto Al. leprae, BCG, and Mycobacterium w(Table 4). Although all of the BCG-inducedT-cell lines recognized BCG in proliferativeassays, only three and two of these T-celllines responded to Al. leprae and Mycobac-terium w, respectively (Table 5).

268^ International Journal of Leprosy^ 1988

TABLE 3. Proliferation of M. leprac-in-duccd T-cell lines to M. leprae, BCG, andMycobacterium w.

T-celllines

Proliferation (cpm x^10 `)in response to

Noantigen

M.leprae

BCGMyco-bac-

terium iv

AF ML 1.2 21.0 5.6WR ML 0.7 2.0 6.3 1.9JM ML 0.4 8.0 48.5IIK ML 35.6 162.4 256.1 145.2PO ML 77.2 146.4 134.6 176.4ATT ML 0.6 41.9 119.5 49.7NMI ML 11.4 69.4 66.7 30.1

Positive/tested 7/7 6/7 7/7

" A T-cell line showing A cpm 12: 1000 and T/C > 2in response to a given antigen was considered positive.Such values are underlined.

To identify mycobacterial antigens acti-vating the T cells, the cell lines were testedagainst the recombinant antigens of M. lep-rae and Al. tuberculosis. The Alvcohacte-rium w-induced T-cell line from donor AF(AF Al. w) and all three cell lines from do-nors JM and NH H responded to the 65-kDa recombinant antigen (Table 6). TheT-cell lines from donor NHH were mostinteresting with respect to their response torecombinant antigens. BCG- and Mycobac-teriunz w-induced cell lines from NHH pro-liferated to 19-kDa M. tuberculosis antigen,

TABLE 4. Proliferation q/Mycobacteriumw-induced T-cell lines to M. leprae, BCG,and Mycobacterium w.

Proliferation (cpm x 10in response to

T-celllines No

antigenM.

lepme13CG

Myco-bac-

terium w

AF MW 1.0 14.1' 29.9 6.9JM MW 0.9 2.8 49.7 4.5HK MW 8.0 103.5 196.8 106.3PO MW 21.6 96.3 126.4 142.0NHH MW 10.3 36.6 55.4 39.7

Positive/tested 5/5 5/5 5/5

A T-cell line showing A cpm 1000 and TIC > 2in response to a given antigen was considered positive.Such values are underlined.

TABLE 5. Proliferation of BCG-inducedT-cell lines to M. leprae, BCG, and Myco-bacterium w.

T-celllines

Proliferation (cpm x^10in response to

Noantigen

Al.leprae BUG

3/yen-bac-

lerhi iii^ii'

AF 13CG 4.0 18.8 6.2WR 13CG 0.6 1.1 8.0 0.8JM BCC, 1.8 17.1 64.7 6.1TO 13CG 1.7 1.6 4.0 1.8IIK 13CG 34.1 56.7 122.1 51.8PO 13CG 50.5 57.0 109.1 89.8ER 13CG 5.0 5.7 15.1 6.6NtIll 13CG 1.6 34.2 145.0 13.1

Positive/tested 3/8 8/8 2/8

" A T-cell line showing A cpm ^ 1000 and TIC > 2in response to a given antigen was considered positive.Such values are underlined.

Al. leprae- and Mycobacterium w-inducedcell lines responded to 18-kDa Al. lepraeantigen, BCG-induced cell line also re-sponded to 14-kDa Al. tuberculosis antigen,and Mycobacterium w-induced cell line re-sponded to 28-kDa Al. leprae antigen (Table6). The T-cell lines from the remaining sub-jects did not proliferate to any of the recom-binant antigens tested (data not shown).

From the reactivity pattern or the T-cellclone 2/3D, which responded to recombi-nant antigen 13B3, it was shown that thisantigen is shared between M. leprae andBCG (Table 1). However, only the BCG-but not Al. /eprae-induced NH H T-cell lineresponded to 13B3 antigen (Table 6). Thereason could be that the majority of the Tcells in the M. /cprae-induced NHHline recognized antigens other than 13B3,e.g., 65-kDa and I 8-kDa antigens, and the13B3 reactive T cells, although present, weretoo few in number to yield a positive re-sponse. If this were the case, it should bepossible to enrich these T cells by culturingthe cell line in the presence of the respectiveantigen and thereby gain a response to it.To test this possibility, the Al. lepaze-in-duced NHH T-cell line (NH H-M L) was cul-tured with E. coli lysates containing 13B3antigen as well as with 65-kDa Al. tuber-culosis antigen and 18-kDa M. leprae an-

56, 2^Mustafa: 7'-cell-activating Recombinant Antigens^269

TABLE 6. Proliferation of T-cell lines to the recombinant antigens of M. leprae and M.tuberculosis.

E. coli lysatescontaining

Proliferation (cpm x^10 -3) of T-cell lines

recombinantantigens of

AFML

AFBCG

AFMW

JMML

JMBCG

JMMW

NHHML

NHHBCG

NHHMW

M. tuberculosis

65-kDa" 3.7 21.6 56.21' 14.2 9.2 2.5 29.5 11.2 7.719-k Da 7.3 6.3 3.1 0.6 0.3 0.2 5.8 4.8 7.814-k Da 7.7 13.6 4.6 0.9 NT' 0.3 6.2 4.6 0.6

Al. leprae36-k Da 8.9 10.4 5.7 0.6 NT 0.6 6.8 0.6 1.028-kDa 8.1 8.9 3.7 0.4 NT 0.7 9.7 0.6 2.718-kDa 13.1 10.2 4.6 0.7 0.3 0.4 46.3 0.2 1.812-kDa 11.4 7.5 4.9 0.8 NT 0.4 5.3 0.3 0.513133 6.7 17.4 4.3 0.7 0.4 0.4 3.7 5.6 0.8

Control E.coli lysate 8.5 14.0 0.4 0.3 0.4 6.4 0.2 0.8

" The 65-kDa antigens of Al. leprae and Al. tuberculosis have greater than 950/s amino acid sequence homology(26) and in general have common antibody and T-cell-reactive epitopes '"). Therefore, to determine thereactivity against 65-kDa antigen, the T-cell lines were tested against recombinant 65-kDa antigen of 31. tu-berculosis only.

"The cpm values showing significant proliferation (cpm >: 1000 and TIC > 2) in response to E. coli lysatescontaining recombinant antigens as compared to control E. coli lysates are underlined.

'NT = Not tested.

tigen containing lysates as positive controls,and with 19-kDa AI. tuberculosis antigencontaining lysate and E. coli lysatc lackingrecombinant antigen as negative controls.The NHH-ML T-cell line after 2 weeks ofculture with E. coli lysates containing orlacking recombinant antigens maintainedits responsiveness to Al. leprae, BCG, andMycobacterium w (Table 7). Similarly, thecell line cultured with 65-kDa M. tubercu-losis antigen or 18-kDa Al. leprae antigenmaintained the responsiveness to 65-kDaAl. tuberculosis and 18-kDa AI. leprae an-tigens, respectively. The T-cell line culturedwith 13B3 antigen gained the responsive-ness to 13B3, but the cell line cultured with19-kDa M. tuberculosis antigen failed toproliferate in response to the homologousantigen (Table 7).

DISCUSSIONThe present study was undertaken with

the aim to identify T-cell-activating anti-gens shared by Al. leprae, BCG, and My-cobacterium w. Such crossreactive antigensmay be responsible for the induction of A/./eprae-reactive CMI after BCG immuni-zation (23) or Mycobacterium w vaccination

(2.20). The positive response of some of theM. /eprae-induced T-cell clones to BCG andMycobacterium w establishes the antigeniccrossreactivity for a single specificity. Theproliferative and cytotoxicity assays used inthe present study to demonstrate antigeniccrossreactivity may have biological rele-vance. Mycobactcrial antigen-induced pro-liferation of human T-cell clones parallelswith the production of important lympho-kines like IL-2, -y-interferon and granulo-cyte-macrophage colony stimulating factors("). All of these lymphokines can directlyor indirectly have a role in the regulation ofCMI responses and in the destruction ofpathogens in the host monocytes/macro-phages. However, those monocytes/mac-rophages which somehow have become in-competent to kill ingested bacilli may serveas fertile soil for unabated bacterial growth.Their killing by cytotoxic T cells will facil-itate the release of bacteria from the pro-tected intracellular sites to be taken up byfresh and competent monocytes. The inhib-itory effect of lymphokine-producing cyto-toxic T-cell clones on the in vitro growth ofmycobacteria inside macrophages has beendemonstrated by Kaufmann, et al. ())•

83.0 1 .8 34.914.422.645.6

^

3.0^49.0

^

6.0^30.8

No recombinant antigen

M. tuberculosis65-kDa antigen19-kDa antigen

leprae18-kDa antigen1303 antigen

53.3^12.8^2.1^NA

3.0^1.8

15.1^15.769.0^20.3

20.6''

42.4

31.3

31.018.1

270^ International Journal of Leprosy^ 1988

TABLE 7. Proliferative response of^T-cell line after 2 weeks of culture withrecombinant antigens.

Proliferative (cpm x 10 ') response to

T-cell line cultured withE. coli lysates containing M. leprae BCG

Aircobac-terium w

E. colilysate

lacking re-combinant

antigen

E. colilysate

containinghomologousrecombinant

antigen'

" Homologous recombinant antigens are the antigens used to culture the T-cell line NHH-ML for the 2-weekperiod.

" The cpm values showing significant proliferation in response to the antigens are underlined.

T-cell clones are excellent tools to studyantigenic crossreactivity for a single speci-ficity. However, in an individual, the T cellsactivated in response to a complex organismlike mycobacteria will respond to antigensof varied specificity ( 18, 21

). The in vitro-raised

T-cell clones which are a selected popula-tion may not represent the complete rep-ertoire of activated T cells. Therefore, inaddition to T-cell clones, the crossreactivityorin vitro-established T-cell lines against Al.leprae, BCG, and Mycobacterium w was alsodetermined. As compared to T-cell clones,these T-cell lines are easy to raise and main-tain. Like T-cell clones, the T-cell lines havealso given evidence of antigenic crossreac-tivity among M. leprae, BCG, and Myco-bacterium w. Although all except one ofAl. leprae- and Mycobacteriuni ti-inducedT-cell lines responded to BCG and 'Myco-bacterium ii', only 3 of 8 and 2 of 8 13CG-induced T-cell lines responded to Al. lepraeand Mycobacterium iv, respectively. Thismay mean that in the subjects tested, thedominant T-cell-activating antigens of Al.leprae and Mycobacterium w are crossreac-tive with BCG, but the dominant T-cell-activating antigens of BCG may not cross-react with Al. leprae and Mycobacterium iv.Such a conclusion is supported from ourearlier studies where >70% of the Al. lep-rae-reactive T-cell clones from five killed-

Al. frprae-vaccinated subjects responded toBCG (I5) but <10% of BCG-reactive T-cellclones from nine BCG-vaccinated subjectsresponded to Al. leprae (ls)•

The T-cell lines were also used to identifyT-cell-activating recombinant antigenspresent in E. coli lysates. The antigens pre-viously identified by T-cell clones, i.e., 65-kDa antigens of Al. leprae and Al. tuber-culosis (1'. 2'), 18-kDa Al. leprac antigen (15),19-kDa Al. tuberculosis antigen (21), I 4-kDaA/. tuberculosis antigen (13), and 13B3 Al.leprae antigen (A. S. Mustafa, et al., sub-mitted for publication) were also identifiedby T-cell lines. Additional reactivity wasfound with one of the Mycobacterium w-induced T-cell lines responding to the M.leprae 28-kDa antigen, for which T-cellreactivity has not yet been reported. How-ever, the T-cell lines are a mixture of dif-ferent reactivities. The bulk of their re-sponse may be directed toward a fewdominant antigens. The T cells capable ofresponding to other antigens may be presentbut be so few in number that, unless en-riched for these reactivities, they may notbe detected by the assay system used. Thisseems to be the reason for the nonrespon-siveness of the Al. /eprac-induced NHHT-cell line to 13B3 antigen in Table 6. Theresponse was specifically restored after thecell line was cultured to enrich for 13B3

56, 2^Mustafa: T-cell-activating Recombinant Antigens^')71

antigen-reactive T cells. This enrichmenttechnique may allow the identification ofother recombinant antigens with less-dom-inant T-cell-activating epitopcs, which willbe missed due to the low frequency of re-sponding T cells in either the T-cell cloningprocedures or in the assays with nonen-riched T-cell lines.

The results of testing Al. /eprae-induccdT-cell clones, and AI. leprae-, BCG-, andMycobacterizon w-induced T-cell lines withrecombinant antigens show that the 65-kDaantigen is shared among the three myco-bacteria. Amino acid sequence analysis of65-kDa antigens from M. leprae, BCG, andAl. tuberculosis indicates >95% homologyea). Most of the antibody and T-cell cpi-topes present on this protein crossreact withother mycobacteria '2' '9). However, itmay also have species-specific antibody andT-cell epitopes Seven of the 20 in vi-tro-established T-cell lines from vaccinatedvolunteers (Table 6) and 20% of the Al. tu-berculosis-reactive T cells from immunizedmice (10) responded to the 65-kDa antigen.Immunization of experimental animals withwhole bacilli induced CMI reactivity to the65-kDa antigen and vice versa (7.)' "). Thesestudies indicate that the 65-kDa protein isa dominant T-cell antigen of mycobacteria.

AI. leprae and Al. tuberculosis possessanother immunologically important proteinof almost identical molecular weight, i.e.,the 18-kDa and 19-kDa proteins. But, un-like the 65-kDa proteins, the 18-kDa pro-tein of M. leprac and the 19-kDa protein ofAl. tuberculosis are completely different fromeach other. The DNA encoding the two pro-teins does not hybridize (8), and the anti-bodies or the T-cell clones responding tothe respective proteins do not crossreact(8. '5.21). The results presented in this papersuggest that 18-kDa M. leprae protein isshared by Mycobacterium n' and that the19-kDa Al. tuberculosis protein is present inBCG and Mycobacterium w.

The mouse monoclonal antibodies andthe polyclonal sera raised against Al. lepracand Al. tuberculosis antigens react to a lim-ited number of proteins when tested eitheragainst their lysates (3.') or their recombi-nant proteins (8.25.28-30). However, the ma-jority of the T-cell clones and T-cell linesin our hands do not respond to these re-

combinant proteins (15. 19' 21 and this report).The reason may be that Xgt11 recombinantsusually produce only a portion of the pro-tein of interest in the form of 13-galactosi-dase fusion protein. The amino terminal partwhich is missing in the recombinant pro-teins may contain most of the T-cell cpi-topes. In this connection, Klatser, et al. (")have reported that of the five T-cell cloneswhich responded to 36-kDa M. leprac na-tive protein, only one T-cell clone respond-ed to recombinant 36-kDa protein. In ad-dition, some T-cell epitopes may be presenton proteins that cannot be identified by an-tibody probes. The use of T-cell clones asprimary probes to identify such antigens re-sulted in the identification of a recombinantantigen from the Al. lepraeAgt11 library ten-tatively termed as 13B3 antigen (A. S. Mus-tafa, et al., submitted for publication). Thegene for this antigen as analyzed by restric-tion enzyme analysis is different from thegenes of the other known recombinant an-tigens of mycobacteria. The availability ofcomplete recombinant proteins identifiedby antibody or T-cell probes may help toelucidate further the dominant, species-specific and crossreactive antigens of my-cobacteria. This will be important for thedevelopment of a new generation of recom-binant vaccines, specific diagnostic tools,and for the selection of the most suitablecultivable mycobacterium as a vaccine can-didate against leprosy.

SUMMARYAntigenic crossreactivity among three

candidate antileprosy vaccines, killed My-cobacterium leprae, BCG, and Mycobacte-rium w, was studied using T-cell lines andclones raised from BCG- and killed-M. /cP-rae-vaccinated subjects. To identify thecrossreactive antigens, the T-cell lines andclones were tested against Escherichialysates containing 65-, 36-, 28-, 18-, and 14-kilodalton (kDa) and 13B3 M. leprae anti-gens and 65-, 19-, and 12-kDa M. tuber-culosis antigens. The short-term T-cell lines,which compared to T-cell clones are easy toraise and maintain, were equally effectivein identifying the T-cell-activating recom-binant antigens. The reactivity pattern ofthe T-cell lines and the clones suggested that65-kDa M. leprae and Al. tuberculosis an-

172^ International .Journal of Leprosy^ 1988

tigens are present in Al. leprae, BCG, andMycobacterium w; 18-kDa Al. leprae anti-gen is shared between M. leprae and My-cobacterium w, 1 3133 Al. leprae antigen ispossessed by Al. lepraeand BCG. These andother unidentified T-cell-activating anti-gens shared among candidate leprosy vac-cines may be the basis for induction of invivo sensitization to Al. leprae antigens aftervaccination with 13CG or Mycobacterium w.

RESUMENSc estudiO la reactividad cruzada entre 3 vactinas

ant ileprosas poten ci ales (Mycobacterium 1e10.0e m tier-

to, IIC(i, y Mycobacterium w) usando !Incas y clonas

de celulas T obtenidas de sujetos vacunados con BCG

y con .11. leprae muerto. Para identiticar los antigenos

de reacciOn cruzada, las hit eas y dorms de celulas T,

Sc probaron contra lisados de Escherichia coli come-

niendo los antigenos de 65-, 36-, 28-, 18-, y 14- kilo-

daltones (kDa) del M. leprae, contra el antigen° 13113

de .1/. leprac, y contra los antigenos 65-, 19-, y 12-k Da

del M. tuberculosis. Las lineas dc celulas T, las cuales

son más fficiles de generar y mantencr que las clonzts

de cdulas T, fueron igualmente efectivas en la iden-tificaciOn de los antigenos recombinantes activadores

de las celulas T. Los patrones de reactividad de las

lineas de celulas T y de las clonas, sugirieron que los

antigenos de 65-kDa de M. leprae y M. tuberculosisesffin presentcs en M. leprae, BCG y Mycobacterium

que el antigen() de 18-kDa de Al. leprae es com-

partido por leprae y Mycobacterium IV, y que el

antigen° 13B3 de Al. lepracestA presente en M. lop/ac'y en BCG. Estos y otros antigenos no identificados,

activadores de las celulas T y compartidos por algunasvacunas con potencialidad anti-leprosa, pueden ser la

base para la inducciOn de la sensibilizaciOn in vivocontra los antigenos del .1/. leprae cuando se vacuna

con BCG o con Mycobacterium

RESUMEOn a etudie la reactivite croisee antigenique entre

trois preparations envisagees comme vaccins éventuels

contre la lepre, a savoir Mycobacterium leprae tue, le

I3CG, et Mycobacterium w. Cette etude a ête menee

sur des lignées de cellules-T et de clones developpeesa partir d'individus vaccines par le BCG et par M.leprae tues. En vue de mettre en evidence la reactivilêcroisee des antigenes, les lignees de cellules-T et les

clones ont ête testees contre des lysats d'Escherichiacell contenant les antigenes de 65-, 36-, 28-, 18-, et 14-

kilodaltons (kDa), ainsi que des antigenes 13113 de .1/.

leprae et des antigenes 65-kDa, l9-kDa et 12-kDa

.1/. tuberculosis. Les lignees de cellules-T a courte vie,plus faciles a developper et A maintenir que les clones

de cellules-T sont tout aussi efficaces pour reconnaitreles antigenes recombinants qui activent les cellules-T.

Le profil de reaction des lignees de cellules-T et des

clones sugg,erent que les antigenes 65-kDa de Al. lepraeet de AI. tuberculosis soul presents chez Al. leprae, dans

le 13C6, et chez Mycobacterium iv; l'antigene 18-kDa

de M. leprae existe chez M. leprae et chez Mycobac-terium iv, l'antigene 13113 de A/. lepraeest present chez

M. leprae et dans le 13CG. Ces antigenes, de meme que

d'autres antigencs non identities qui activent les cel-lules-T, et que l'on retrouve dans tomes les prepara-

tions actuellement envisagees coil) me pouvant servir

dc vaccin con tre la lepre, peUvent fburnir la base pourinduire une sensibilisation in vivo aux antigenes de Al.((prat' apres vaccination par le BCG ou par he Myco-

bacterium w.

Acknowledgments. The material and intellectual help

provided by Dr. Tore Godal. Prof. G. P. Talwar, Dr.

R. J. W. Rees, and Dr. Richard Young, and the expert

secretarial assistance by Carolyn C. Carpenter aregratefully acknowledged. Recombinant human inter-

leukin-2 was kindly provided by C'etus C'orporation.This work was supported by grants from the United

Nations Development Programme/World Bank/World

Health Organization Special Programme for Research

and Training in Tropical Diseases, from the Rocke-feller Foundation, Program for Research in Great Ne-

glected Diseases, and the Laurine Maarschalk Fund.

REFERENCES1. BUCHANAN, T. M., NOMAGUCHI, 11., ANDERSON,

D. C., YOUNG, R. A., Gmus, T. P., BRITTON, W.

J., IVANVI, J., KOLK, A. II. J., CLOSS, 0., BLOOM,B. R. and MEHRA, V. Characterization of anti-

body-reactive epitopes On the 65-kilodalton pro-

tein of Mmtbacterium leprae. Infect. Immun. 55

(1987) 1000-1003.

2. CHAUDHURI, S., FOTEDAR, A. and TALWAR, G. P.

Lepromin conversion in repeatedly lepromin neg-ative BULL patients after immunization with au-

toclaved Mycobacterium iv. Int. J. Lepr. 51 (1983)

159-168.

3. ENGERS, H. D., et al. Results of a World Health

Organization-sponsored workshop on monoclonal

antibodies to Mycobacterium leprac. (Letter) In-

fect. Immun. 48 (1985) 603-605.

4. ENGERS, H. D., et al. Results of a World Health

Organization-sponsored workshop to characterize

antigens recognized by mycobacterium-specificmonoclonal antibodies. (Letter) Infect. Immun. 51

(1986) 718-720.5. Gni, H. K., MUSTAFA, A. S. and GODAL, T. In-

duction of delayed type hypersensitivity in humanvolunteers immunized with a candidate leprosy

vaccine consisting of killed Mycobacterium leprae.Bull. NVI-10 64 (1986) 121-126.

6. GILL, H. K., MUSTAFA, A. S. and GODAL, T. Invitro proliferation of lymphocytes from humanvolunteers vaccinated with armadillo-derived,

killed M. leprae. Int. J. Lepr. 55 (1987) 30-35.

7. GILDS, T. P. and JOB, C. K. Purification of the 65

kD protein from Mycobacterium goalonae and use

56, 2^Mustafa: T-cell-activating Recombinant Antigens^273

in skin test response to Mycobacterium leprae. Int.J. Lepr. 55 (1987) 54-62.

8. HUSSON, R. and YOUNG, R. A. Genes for the majorantigens or mjwbacterium tubereffiosis: the etio-logic agents of tuberculosis and leprosy share animmunodominant antigen. Proc. Natl. Acad. Sci.U.S.A. 84 (1987) 1379-1383.

9. KAM:I•MANN, S. H. E., CIIIPLUNKAR, S., ELESCH,

and Di; LIBERO, G. Possible role of helper andcytolytic T cells in mycobacterial infections. Lepr.Rev. 57 Suppl. 2 (1986) 101-111.

ID. KAUFMANN, S. II. E., VATII, U., THOLE, J. E. R.,VAN EmnorN, J. D. A. and EMMRICI I, F. Enumer-ation of T cells reactive with Mycobacterium tu-berculosis organisms and specific for the recom-binant mycobacterial 64-kDa protein. Eur. J.Immunol. 17 (1987) 351-357.

I I. KLATSER, P. R., HARTSKEERL, R. A., VAN

SCHOOTEN, W. C. A., KOLK, A. H. J., VAN RENS,

M. M. and DE WIT, M. Y. Characterization of the36 K antigen of AI ycobacterium leprae. Lepr. Rev.57 Suppl. 2 (1986) 77-81.

12. LAMB, J. R., IVANYI„J., REES, A. I). NI., ROTHBARD,

J. II., I iOWLAND, K., YouNG, R.. A. and YouNG,D. 13. Mapping of T cell epitopes using recombi-mint antigens and synthetic peptides. EM BO J. 6(1987) 1245-1249.

13, LAMB, J. R., IVANVI, J., RUES, A., YOUNG, R. A.and YOUNG, D. B. The identification of T cellepitopes in Mycobacterium tuberculosis using hu-man T lymphocyte clones. Lem. Rev. 57 Suppl.2 (1986) 131-137.

14. MEHRA, V. and BLOOM, II. R. Induction of cellmediated immunity to Mycobacterium /eprae inguinea pigs. Infect. Immun. 23 (1979) 787-794.

15. MUSTAFA, A. S., GILL, H. K., NERLAND, A.,BRITToN, W. J., MEHRA, V., BLOOM, B. R., YOUNG,

R. A. and GODAL, T. Human T-cell clones rec-ognize a major .11. leprae protein antigen expressedin E. co/i. Nature 319 (1986) 63-66.

16. MUSTAFA, A. S. and GoDAL, T. In vitro inductionof human suppressor T cells by mycobacterial an-tigens; BCC; activated OKT4^cells mediatesuppression of antigen induced T cell prolifera-tion. Clin. Exp. Immunol. 52 (1983) 29-37.

17. MUSTAFA, A. S. and GoDAL. T. 13CG inducedCD4+ cytotoxic T cells from BCG vaccinatedhealthy subjects: relation between cytotoxicity andsuppression in vitro. Clin. Exp. Immunol. 69 (1987)255-262.

18. MUSTAFA, A. S., KVALHEIM, G., DEGRE, M. andGoDAL. T. llivobacterium bovis BCG-inducedhuman T-cell clones from IICG-vaccinated healthsubjects: antigen specificity and lymphokine pro-duction. Infect. Immun. 53 (1986) 491-497.

19. MUSTAFA, A. S., OFTUNG, F., GILL, H. K. andNALvic, I. Charactertistics of human T-cell clonesfrom BCG and killed M. /eprae vaccinated sub-

jects and tuberculosis patients. Lepr. Rev. 57 Suppl.2(1986) 123-130.

20. MUSTAFA, A. S. and TALWAR, G. P. Delayed hy-persensitivity skin reactions to homologous andheterologous antigens in guinea pigs immunizedwith .11. /eprae and four other selected cultivablemycobacterial strains. Lepr. India 50 (1978) 509-519.

21. GETUNG, F., MUSTAFA, A. S., HUSSON, R., YOUNG,

R. A. and GODAL, T. Human T cell clones rec-ognize two abundant Ifycobactertum tuberculosisprotein antigens expressed in Escherichia co/i. J.lmmunol. 138 (1987) 927-931.

22. SnEpARD, C. C., MINAGAWA, F., VAN LANDING-

HAM, R. and WALKER, L. L. Foot pad enlargementas a measure of induced immunity to 1/ycobac-terium leprae. Int. J. Lepr. 48 (1980) 371-381.

23. SHEPARD, C. C., VAN LANDINGHAM, R. and WALK-

ER, L. L. Searches among mycobacterial culturesfor antileprosy vaccines. Infect. Immun. 29 (1980)1034-1039.

24. SHEPARD, C. C., WALKER, L. L. and VAN

LANDINGHAM, R.. Heat stability ofMycobacteriumhprae immunogenicity. Infect. Immun. 22 (1978)87-93.

25. SHINNICK, T. M., KRAT, C. and ScHADow, S. iso-lation and restriction site maps of the genes en-coding five Mycobacterium tubercu/osis proteins.Infect. Immun. 55 (1987) 1718-1721.

26. SHINNICK, T. M., SWEETSER, D., TrioLE, J., VAN

ENUIDEN, J. and YOUNG, R. A. The etiologic agentsof leprosy and tuberculosis share an immuno-reactive protein antigen with the vaccine strainMycobacterium boils BCG. Infect. lmmun. 55(1987) 1932-1935.

27. THOLE, J. E. R., KEULEN, W. J., KOLK, A. H. J.,GROOTHUIS, D. G., BERWALD, L. G., TIESJEMA, R.

H. and VAN EMBDEN, J. D. A. Characterization,sequence determination, and immunogenicity ofa 64-kilodalton protein of Mycobacterium boils13CG expressed in Escherichia coif K-12. Infect.Immun. 55 (1987) 1466-1475.

28. YOUNG, D. B., KENT, L. and YOUNG, R. A. Screen-ing of a recombinant mycobacterial DNA librarywith polyclonal antiserum and molecular weightanalysis of expressed antigens. Infect. I mmun. 55(1987) 1421-1425.

29. YOUNG, R. A., BLOOM, 13. R., GROSSKINSKY, C.NI., IvANyi, J., THOMAS, D. and DAvis, R. W.Dissection of:I/In/bacterium tuberculosis antigensusing recombinant DNA. Proc. Natl. Acad. Sci.U.S.A. 82 (1985) 2583-2587.

30. YOUNG, R. A., MEHRA, V., SWEETSER, D.,BUCHANAN, T., CLARK-CURTISS, J., DAVIS, R. W.and 13Loom, B. R. Genes for the major proteinantigens of the leprosy parasite Mycobacteriumleprae. Nature 316 (1985) 450-452.