Vol. 58, No. 9INFECTION AND IMMUNITY, Sept. 1990, p. 2821-28270019-9567/90/092821-07$02.00/0Copyright X) 1990, American Society for Microbiology

Selection and Characterization of Recombinant Clones That ProduceMycobacterium leprae Antigens Recognized by Antibodies in Sera

from Household Contacts of Leprosy PatientsRUDY A. HARTSKEERL,* RIA M. VAN RENS, LINDA F. E. M. STABEL, MADELEINE Y. L. DE WIT, AND

PAUL R. KLATSERN. H. Swellengrebel Laboratory of Tropical Hygiene, Royal Tropical Institute, Meibergdreef 39, 1105 AZ Amsterdam,

The Netherlands

Received 21 February 1990/Accepted 29 May 1990

A Mycobacterium leprae expression library was constructed in the vectors EX1, pEX2, and pEX3 andscreened with a pool of 19 well-absorbed sera from household contacts of leprosy patients. Twelve selectedrecombinants that were further characterized differed clearly from recombinants selected with murinemonoclonal antibodies. Whereas the monoclonal antibodies recognized mainly six recombinant antigens, thehuman sera from contacts reacted with a range of different recombinant antigens. None of the contactrecombinant antigens was identical or related to well-characterized antigens from M. leprae or othermycobacteria selected with monoclonal antibodies, including proteins of the heat shock families. Two groupsof recombinant antigens could be distinguished: one that was recognized by all sera used in the pool and onethat was recognized by only a limited number of sera. These antigens, selected with sera from householdcontacts of previously untreated lepromatous leprosy patients, may be relevant to the immune responses duringthe early phase of infection with M. leprae.

Leprosy is still a major health problem in many parts ofthe world; despite the great efforts that have been made tobetter control the disease, the prevalence of the diseaseremains at a constant level (41). Studies on the immunologyof leprosy have mainly focused on clinical aspects of thedisease. Relatively little is known about the immunologicalevents during the preclinical phase of the disease.

Serological studies among household contacts of leprosypatients, in which higher seropositivity rates were reportedthan the incidence rates expected in these populations,suggest that seropositivity reflects subclinical infection (2,12). Considering the low sensitivity of these tests whenrelatively small numbers of bacilli are present (2), theseropositive contacts are probably only a small proportion ofthe total pool of infected individuals. The dynamics ofinfection would likely be, as for many other infectiousdiseases, as follows. A person becomes colonized withMycobacterium leprae and subsequently either gets rid ofthe bacterium or becomes infected (invasion of tissue bybacteria). Subsequently, the individual could either over-come the infection (by an effective immune response or bytreatment) or develop the disease. Little is known about thefactors that determine the dynamics of infection. Availablemethodology to determine the immunological status of in-fected individuals lacks either sensitivity, as in the case ofserological tests (2, 13), or specificity, as in the case of thelepromin skin test (41). However, it is important to broadenour knowledge of the immunological reactions during theearly stages of infection. Such information could, for exam-ple, lead to the identification of antigens involved in protec-tion or in pathogenesis and, equally important, may result inthe development of new tools for the detection of preclinicalleprosy.Most of the recombinant antigens from M. leprae were

* Corresponding author.

initially identified either with murine monoclonal antibodies(45) or with sera from diseased individuals (9, 29). However,the immune responses in inbred mice and in persons withclinical leprosy may well be distinct from those during earlyinfection. As a first step to identify antigens that may play arole during the early phase of infection with M. leprae, wescreened an M. leprae gene library with sera obtained fromhousehold contacts of untreated multibacillary leprosy pa-tients. These contacts, who had lived in close proximity withthe index cases for several years, are likely to be infectedwith M. leprae. In this report we describe the selection andcharacterization of recombinant clones obtained with suchsera.

MATERIALS AND METHODSStrains and plasmids. M. leprae was isolated from spleen

tissue of experimentally infected armadillos (Dasypusnovemcintus Linn.) as recommended by the World HealthOrganization (40). Escherichia coli K-12 strain POP2136 wasused as a host for construction of recombinant plasmids andfor gene expression. POP2136 is a nonexcisable lambdalysogenic strain carrying the cI ts856 gene (Genofit, Geneva,Switzerland). Strain S1036 is POP2136 carrying plasmidpEX2. Plasmids pEX1, pEX2, and pEX3 have been de-scribed by Stanley and Luzio (34).Media and reagents. Strain POP2136 and derivatives were

grown in LB medium (22). When appropriate, 100 ,ug ofampicillin per ml was added to the medium. Difco agar(1.5%) was added to solidify the medium. Restriction endo-nucleases and T4 DNA ligase were from Boehringer GmbH(Mannheim, Federal Republic of Germany) and New En-gland BioLabs, Inc. (Beverly, Mass.). Alkaline phosphatase-labeled anti-human polyvalent immunoglobulins were fromSigma Chemical Co. (St. Louis, Mo.).

Sera. Sera from 19 household contacts of previouslyuntreated lepromatous leprosy patients (coded in this paperas a to s) were kindly provided by R. V. Cellona (Leonard

2821

on March 30, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

2822 HARTSKEERL ET AL.

Wood Memorial Center for Leprosy Research, Cebu City,The Philippines). The contacts included 8 women and 11men whose mean age at the time of blood withdrawal was34.7 years (standard deviation, 14.8 years). The mean dura-tion of contact with the index case was 17.6 years (standarddeviation, 8.4 years). During follow-up of these contacts upto 3 years after blood withdrawal, none of them had becomea clinically and bacteriologically established leprosy patient.None of the contacts had been vaccinated with Mycobacte-rium bovis BCG.

Sera were pooled and extensively absorbed as describedby Sathish et al. (29). Briefly, the pool of sera was incubatedovernight at 4°C with whole POP2136 and heat-inducedS1036 cells. Subsequently, the pool was absorbed overcolumns of lysates of strain POP2136, heat-induced strainS1036, and purified beta-galactosidase from E. coli (Sigma)coupled to cyanogen bromide (CNBr)-activated Sepharose4B (Sigma). The pool was subjected to four cycles ofabsorption. Pooled sera were used at a final dilution of 1:200in blocking buffer (see below) to screen the libraries. Ab-sorption of individual sera was essentially done by the sameprocedure, except that the elution over the columns wasomitted.DNA technology. Standard procedures were used for re-

striction enzyme digestion, ligation, and transformation (22).Preparation of genomic DNA from nonirradiated M. lepraewas done as described previously (15). DNA sequencing ofthe pEX1, pEX2, and pEX3 derivatives was done by thedideoxy-chain termination method (28) with a T7 sequencingkit (Pharmacia LKB, Uppsala, Sweden) and a ThermalbaseTaq sequencing kit (Stratagene, La Jolla, Calif.). For DNAsequencing, plasmid DNAs were isolated by alkaline extrac-tion (5) and further purified by CsCl gradient centrifugationor by use of the Geneclean kit (BIO 101, La Jolla, Calif.).

Synthetic oligonucleotides S10 (CAACATCAGCCGCTACAGTC), 86.25-2 (GGCGACGACTCCTGGAGCCCG),and 17.5.88-2 (CAGCAAGCTTGGCTGCAGGTCG) wereused as primers for DNA sequencing. Primers S10 and86.25-2 are homologous to the region of lacZ from 145 to 125base pairs (bp) and 33 to 12 bp, respectively, upstream of theEcoRI site (18). Primer 17.5.88-2 is homologous to the region13 to 35 bp downstream of this EcoRI site in pEX2 (34). Allprimers were synthesized on a 381A DNA synthesizer(Applied Biosystems) and were used without further purifi-cation.For dot blot hybridization, 1-,ug samples of plasmid DNA

were spotted onto Duralose-UV membranes (Stratagene).The subsequent procedure was essentially the same as forcolony hybridization (22). Insert DNA used as a probe wasexcised from hybrid plasmids with Sail and SmaI. Smallinsertion fragments (100 to 400 bp) were amplified as previ-ously described (15) with oligonucleotides 86.25-2 and17.5.88-2 as the set of primers in the amplification reaction.Amplified fragments were cleaved with Sall and SmaI.Unique cleavage sites for these restriction enzymes arepositioned to the left and right of the BamHI cloning site ofthe pEX plasmid vectors. Fragments were separated byelectrophoresis on agarose gels and isolated from the aga-rose with the Geneclean kit. Some of the DNA insertsapparently contained one or more Sall-SmaI cleavage sites,since more than one insert fragment was visible on theagarose gels. Preparation of 32P-labeled probe DNA wasdone with a random priming DNA labeling kit (Boehringer)according to the instructions of the manufacturer. Addition-ally, probe DNA was labeled by using a DNA digoxigenin-dUTP labeling and detection kit (Boehringer). Prehybridiza-

tion, hybridization, and washing of all membranes were doneas described previously (15). DNASIS and PROSIS software(Pharmacia LKB) was used for the homology search ofnucleotide and amino acid sequences.

Construction of the M. leprae expression libraries in pEX1,pEX2, and pEX3. M. leprae genomic DNA was partiallydigested with Sau3A as follows: 5 U of Sau3A was added to10 ,ug of M. Ieprae DNA in 100 RI1 of Sau3A buffer. After 30,60, and 120 s of incubation at 37°C, samples of 30 ,u1 weretaken and immediately transferred to phenol saturated with10 mM Tris hydrochloride (pH 8.0)-10 mM EDTA to stopthe reaction. After phenol-ether extraction, DNA was recov-ered by ethanol precipitation. Two fractions (30 and 60 s)contained fragments of 0.5 to 4 kbp, judged from electropho-retic patterns of the fractions on agarose gels. These frac-tions were pooled and used for cloning into the pEX plas-mids.Samples (1.5 p.g) of partially digested M. leprae DNA

were ligated to 0.15 p.g of pEX1, pEX2, and pEX3 digestedwith BamHI. Transformation of strain POP2136 with ligatedDNA resulted in 2 x 104 to 5 x 104 transformants per ligationmixture.

Screening of the pEX1, pEX2, and pEX3 libraries. Coloniesgrown on nitrocellulose filters (type BA 85; Schleicher &Schuell, The Netherlands) at 30°C were replicated to asecond filter. Induction of expression and preparation of thereplica filters for colony enzyme-linked immunosorbent as-say was essentially done as described by Stanley and Luzio(34). For the colony enzyme-linked immunosorbent assay,filters were washed in wash buffer (100 mM Tris hydrochlo-ride [pH 8.0], 0.1% [vol/vol] Tween 80, 0.02% [wt/vol]NaN3) and in blocking buffer (wash buffer containing 0.25%[wt/vol] bovine serum albumin and 0.25% [wt/vol] gelatin).Filters were then incubated overnight with the pooled seraand washed again in wash buffer before incubation withalkaline phosphatase-labeled anti-human immunoglobulinsdiluted 1:500 in blocking buffer. The filters were then washedand developed in substrate buffer (100 mM Tris hydrochlo-ride [pH 9.6]-100 mM NaCl-50 mM MgCl2 containing 0.1 mgof nitro blue tetrazolium grade III per ml and 0.05 mg of5-bromo-4-chloro-3-indolylphosphate per ml).

Immunological techniques. Sodium dodecyl sulfate-poly-acrylamide gel electrophoresis was done on 8% acrylamidegels as described by Laemmli (21), and Western immuno-blotting was done by the method of Burnette (7) as modifiedby Van Embden et al. (38). For dot blot analysis, 5 p.1 oflysate (optical density at 600 nm, 5) was spotted ontonitrocellulose or Immobilon-P (Millipore B.V., Etten-Leur,The Netherlands) membranes. Substrate buffer for alkalinephosphatase-labeled anti-human immunoglobulins was asdescribed above for the colony enzyme-linked immunosor-bent assay.

RESULTS

M. leprae expression library. Restriction enzyme analysisof plasmid DNA isolated from randomly picked colonies ofthe gene libraries in pEX1, pEX2, and pEX3 revealed thatapproximately 40% of the plasmids contained an M. lepraeDNA insert larger than 400 bp, with a mean of about 1.0 kbp.Colony hybridization with 32P-labeled M. Ieprae genomicDNA as a probe showed that up to 60% of the plasmidscontained an M. leprae DNA insert. Apparently, a substan-tial number of plasmids contained DNA inserts smaller than400 bp, which are difficult to detect on agarose gels. Basedon these data, the number of transformants, and a genomic

INFECT. IMMUN.

on March 30, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

M. LEPRAE ANTIGENS RECOGNIZED BY CONTACTS' SERA 2823

Aa b c d e f g h j k ni n

. 4b,.116--."_ _ "

Ba b c d e f g h k ni ri

116-

FIG. 1. Coomassie brilliant blue-stained sodium dodecyl sulfate-polyacrylamide gel (A) and Western blot (B) of crude lysates (5 ,ul,optical density at 600 nm of 5 for sodium dodecyl sulfate-polyacryl-amide gel electrophoresis and Western blotting) of the followingheat-induced strains: a, S1036; b, S1143; c, S1142; d, S1141; e,

S1140; f, S1130; g, S1129; h, S1123; i, S1120; j, S1119; k, S1116; 1,S1115; m, S1114; n, S1113. The Western blot was developed with a

pool of 19 sera from household contacts; the sera were absorbed as

described in Materials and Methods.

size of 3.3 x 103 kbp for M. leprae (10), each pEX1-, pEX2-,or pEX3-derived recombinant contained three to six myco-

bacterial genome equivalents.Selection of recombinants. About 1.7 x 105 colonies of the

M. leprae gene libraries in pEX1, pEX2, and pEX3 were

screened with the absorbed contact serum pool. We selected36 potentially positive recombinants. These recombinantswere characterized with respect to DNA inserts and expres-

sion products. Recombinant plasmids contained insertsranging in size from 0.2 to 2.5 kbp. All recombinantsexpressed M. leprae antigens as fusion proteins with ,-galactosidase, with apparent molecular weights ranging from116,000 (116K proteins) to 145,000 (145K proteins). Westernblot analysis revealed that 12 of the recombinants producedantigens that clearly reacted more strongly with antibodies inthe sera from contacts than did the control antigen, cro-,B-galactosidase, produced by strain S1036 (Fig. 1). Sevenrecombinants produced proteins that were reactive with theconjugate in the absence of serum and the lysates of theother 17 recombinants produced bands on Western blotswith intensities that were comparable to that of the controlantigen (data not shown). It is possible that these latterrecombinants expressed conformational epitopes. To inves-tigate this, we performed a dot blot analysis in which, incontrast to Western blot analysis, conformational epitopesremain intact. However, no positive reactions could bedetected on the dot blots with these clones (data not shown).Therefore, we believe that these 17 recombinants are arti-facts. We continued our study with the 12 clearly positiverecombinants and one conjugate-reactive recombinant forfurther analysis (Table 1). All fusion proteins appeared as

clear bands migrating at a position of 116K or higher, both on

TABLE 1. Characterization of recombinant clonesselected with contacts' sera

Strain Plasmid Vector Mol wt' DNA insertG%

S1113 pTHL1031 pEXl 116,000, 120,000 0.6 75.6S1114 pTHL1032 pEX2 140,000 1.0 62.2S1115 pTHL1033 pEX3 135,000 1.1 69.9S1116 pTHL1034 pEX3 119,000 0.213* 63.8S1119 pTHL1037 pEX1 116,000 1.5 72.7S1120 pTHL1038 pEX2 116,000 0.146* 60.6S1123 pTHL1041 pEX2 118,000, 135,000 >0.3 70.6S1129 pTHL1051 pEX3 145,000 0.6 61.7S1130 pTHL1052 pEX3 140,000 0.7 65.0S1140d pTHL1055 pEX1 116,000, 119,000 1.9 77.5S1141 pTHL1056 pEX2 120,000 1.0 67.8S1142 pTHL1057 pEX3 123,000 1.2 62.2S1143 pTHL1058 pEX3 119,000, 120,000 0.3 66.1

a Apparent molecular weights deduced from electrophoretic patterns onsodium dodecyl sulfate-polyacrylamide gels. Two values indicate that twobands appeared on sodium dodecyl sulfate-polyacrylamide gels and onWestern blots (Fig. 1).

b Sizes of DNA inserts were estimated from electropheretic patterns ofrestriction enzyme digests on agarose gels. Sizes derived from establishedDNA sequences are indicated by asterisks.

c Content of nucleotides G+C in the coding parts of the fragments depictedin Fig. 2.

d Antigen was reactive to conjugate in the absence of serum (see Results).

sodium dodecyl sulfate-polyacrylamide gels and Westernblots (Fig. 1).

Nucleotide and deduced amino acid sequences. To charac-terize the recombinant antigens, we determined part of thenucleotide sequence of one strand of the various DNAinserts, adjacent to lacZ (Fig. 2). Both strands of the smallinserts of plasmids pTHL1034 and pTHL1038 were se-quenced. The correct reading frame of the M. leprae part ofthe fused genes could be derived from the reading frame ofthe lacZ counterpart (18). The coding regions of the variousestablished DNA sequences had an overall G+C content of60% or more (Table 1). The established sequences shown inFig. 2 do not have regions of homology and thus representdifferent antigenic determinants. Moreover, as can be de-duced from the apparent molecular weights (Table 1) and theestablished DNA sequences (Fig. 2), eight of the recombi-nant plasmids (pTHL1031, pTHL1034, pTHL1037,pTHL1038, pTHL1055, pTHL1056, pTHL1057, andpTHL1058) encode C-terminal parts of different antigens,suggesting that a high percentage of the recombinants ex-pressed different M. leprae antigens. Consistent with this,dot blot hybridizations revealed that none of the DNAinserts hybridized to DNA from a pEX clone other than itsown (data not shown). Apparently, the pEX-derived plas-mids contained distinct DNA inserts. Therefore, we con-clude that the selected recombinants expressed parts ofdifferent antigens.Do the immunologically reactive clones express known

mycobacterial antigens? Comparison of nucleotide and de-duced amino acid sequences with known sequences wasused to establish homologies and relationships with a num-ber of well-characterized mycobacterial antigens. The DNAsequences of the clones selected with contacts' sera werecompared with the sequences of the genes encoding thefollowing proteins: 70K (14), 65K (26), 36K (PRA; 37), 28K(9), 18K (6), 12K (16), and manganese superoxide dismutase(35) of M. leprae; 65K (30), 19K (3), 12K (4, 31), and Pab (1)of Mycobacterium tuberculosis; 65K (32), MPB70 (27),MPB64 (42), MPB57 (43), and the a antigen (24) of M. bovis

VOL. 58, 1990

on March 30, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

2824 HARTSKEERL ET AL. INFECT. IMMUN.

pTHL 1031 1 GGATCGGAACCGCGAAAACGCGACGGCCCGCCGCGCGGTGCGCGGCGGGCCGACAAGCTT 601 G S E P R K R D G P P R G A R R A D K L 20

61 GCCGGTGGCGGAAGCGTCCGCCGG21 A G G G S V R R

pTHL1032 1 GGGATCGAACAGCTTGACGCCCGTGTCGAGGCGGATCTACCGCCAGTTCTCGCTGACCAT 601 G E Q L D A R V E A D L P P V L A D H 20

61 CGTCTCCTCGATGGCGCTGTCGGTGCT21 R L L D G A V G A

pTHL1033 1 GAAACCGTGCGCCTGGCCGTGCGCCAGCTCGACCGCGTGATCGACCTGAACTTCTATCCG 601 E T V R L A V R Q L D R V D L N F Y P 20

61 ATCGAGACCGCACGCCGCGCCAACCTGCGCTGGCGCCCGGTCGGCCTCGGCGCGATGGGC 12021 E T A R R A N L R W R P V G L G A M G 40

121 CTGCAGGACGTGTTCTTCAAGCTGCGCCTGCCGTTCGACTCCGAGCCGGCG41 L Q D V F F K L R L P F D S E P A

pTHL1034 1 GATCGGCGACGTCGACCAAGACCGCGTGGTCGCGGTGACCAACGTCGGCCACCTGCTCGC 601 D R R R R P R P R G R G D Q R R P P A R 20

61 GTTCCCGGTCGCGGAGCTGCCGGAACTCGACAAGGGCAAGGGCAACAAGATCCTCCACAT 12021 V P G R G A A G T R Q G Q G Q Q D P P H 40

121 ACCATTCCGAACCGCGTTGAGACAGCCGAGCGTGCCGATTTGACGCATAGTCGTCGGTAA 18041 T P N R V E T A E R A D L T H S R R * 60

181 GTACCAAGGTGGAGTCGTCATCCGCCGCCGATCC

pTHL1037 1 GGATCACAGCTTCCGCGCGTCGGGGCACCACGGTGATCGATCCGGGCTTCCTCGCCGTGT 601 G S Q L P R V G A PR *

61 ACGAGGAAGGCAAGGACGCCAAGGCCGCCGAGGACGAGGACGAAGGCCGCAAGCTGCCGC 120

pTHL 1038 1 GGGATCGAGACGCGCAACAGCGATGGCCTTCTCGACGGGAAAGCACGGGGCACTGTCGAT 601 G E T R N S D G L L D G K A R G T V D 20

61 TTCTACTAACGCGTGTCGGCCTTCCAGTTGCAGGTAACGAAGGACGATGTACCCGTCACG 12021 F Y *

1 21 CAATTTCAGTGGCTTCTGACGCTCGATCC

pTHL1041 1 GGGATCTCGCCGTTCGGGCTGCCGGTGAACTTGTACGCTGCGTCCAGCGCCTTGAGCTGC 601 G S P F G L P V N L Y A A S S A L S C 20

61 TCGACGCTGAGCGTTTCGGGCATGCCCTCGATGAAGTGCACCCACTCCTGCGTGCTCCAC 12021 S T L S V S G M P S M K C T H S C V L H 40

1 21 TTCGCGGTCGCCGCGGCCGCGGCAGCGTGCCGCCCTCGACCCAGGCCTTGCGCGCGGCGT 18041 F A V A A A A A A C R P R P R P C A R R 60

pTHL1051 1 GATCACCACGCCGCCCTTGTCGCCCTTGTACATGTGCGGCAACCGGGCCTTGGTGGTCAG 601 D H H A A L V A L V H V R 0 P G L G G Q 20

61 GAAGGCGCCGTCGAGATGGATGGCGAGCATCTTCTTCCAGTCGGCGAAGGCGTAGTTCTC 12021 E G A V E M D G E H L L P V G E G V V L 40

121 GATCGGATTGACGATCTGGATGCCGGCGTTGGAGACCAGGATGTCGATGCCGCCGAGTTG 18041 D R D D L D A G V G D 0 D V D A A E L 60

pTHL1052 1 GATCGCGGCGAATGCGGCCGAGGTCATGATCGCGTAGCTGATCGCGTAGAACATCGCCGC 601 D R G E C G R G H D R V A D R V E H R R 20

61 GGCGAAGCCTGCGCGCCGCCGCCGGCCATGCCGATGAACAGGAAGCCGACATGCGAGACC 12021 G E A C A P P P A M P M N R K P T C E T 40

pTHL 1055 1 GGATCCCGCCCAGCGCAGAGCCTGCGCGCAGGCCTATCTCAACACCCTGCGCCTGGCCGT 601 G S R P A Q S L R A G L S Q H P A P G R 20

61 CGGCCGCGCCGCGCCGGCCGGCGCAGCGCGTTCACCCGTGCCGCGATCCGGGCCAACACC 12021 R P R R A G R R S A F T R A A I R A N T 40

pTHL1056 1 GGGATCGGCGATTACGACAGCGCTGGCGCAGCCAAACTGCAGGCGGTGTGGCTCGGCAAC 601 G G D Y D S A G A A K L 0 A V W L G N 20

61 GCGCCTGTCGGTTTGGTGGTCGGGAGCGGCAGCACGCAGACGCTGCAATATGTGCAGCCG 12021 A P V G L V V G S G S T Q T L Q Y V Q P 40

121 GATCGCCAGCGCCGGCGCGCCGACCACCACGCCCAGCGCATAGGCGCTGATGACATGGCC 18041 D R Q R R R A D H H A 0 R G A D D M A 60

pTHL 1057 1 GATCAGCCAGTCCAGGCTCATGCCGTCCTTCGCGAAAATGCGCAGAAGCTGTTCGAGGTG 601 D Q P V 0 A H A V L R E N A 0 K L F E V 20

61 CCGCAAACTCGGCTCGCTTCTGCCGTTTTCCCACTTGGATACAGTGGCAGAGGTGATCAA 12021 P Q T R L A S A V F P L G Y S G R G D 0 40

121 CGCGTCGATGCGCTTGCCGGCGGCGGCCTTGGGGATGTCCTGCACGGCGGCGAGGAACTC 18041 R V D A L A G G G L G D V L H G G E E L 60

pTHL1058 1 GATCGCGGCCTGGATGCGGCTTTGCAGCGCGCTCACGTCCAGGCCTTCGAGCAAGGTGCC 601 D R G L D A A L 0 R A H V Q A F E Q G A 20

61 GGCCAGGTCGGGAATGCCGTAGTTCAGTACGCTGGCGGCGACGTAGGGATGCGCCTCGAG 12021 G 0 V G N A V V Q Y A G G D V G M R L E 40

121 CGCTTCGTCAGACCAGTGCCGGGTGCAGTTGAGCAGCCACGACAGGTCGCGACGATGCAT 18041 R F V R P V P G A V E Q P R Q V A T M H 60

FIG. 2. Nucleotide sequence and deduced amino acid sequence of a part of the M. leprae DNA inserts of the various hybrid plasmidsencoding M. leprae antigenic determinants recognized by contacts' sera. The DNA sequence of the part of the inserts adjacent to lacZ wasdetermined for one strand by using primers S10 and 86.25-2. Primer 17.5.88-2 was used to sequence the complementary strands of the insertsof plasmids pTHL1034 and pTHL1038. Boldface type indicates nucleotides belonging to the vector or the BamHI-Sau3A cloning sites.pTHL1033 contains a mutation at the cloning site; the first nucleotide in the sequence belongs to the vector. The translation of all nucleotidesequences is in accordance with the open reading frame of lacZ in the pEX plasmids.

BCG; and the a antigen of Mycobacterium kansasii (23). In antigens are related at the protein level. To investigate suchall cases, the complementary strands were also involved in a relationship, deduced translation products of all genes asthe comparisons. No significant homology was found with well as the amino acid sequence of the tuberculin-activeany of these genes. Consequently, none of the cloned protein of M. tuberculosis (20) were compared. Errors in thecontact antigens is identical with any of the known M. leprae nucleotide sequence due to the sequencing of only oneantigens, nor is any antigen encoded by possible open strand could have led to frame shifts and hence to a partlyreading frames on the complementary strands of the corre- incorrect deduction of the translation products. To preventsponding genes. Furthermore, at the DNA level no relation- misinterpretations caused by the use of such wrongly de-ship could be found between the contact antigens and the duced amino acid sequences, the translation products fromother mycobacterial antigens. However, it could be that the all three reading frames were used in the homology search.

on March 30, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

M. LEPRAE ANTIGENS RECOGNIZED BY CONTACTS' SERA

B a bode f gh±i j k1 m no p q r a

ili

FIG. 3. Western blot analysis of lysates of recombinant cloneswith the individual contact sera from the pool. (A) Lysate of strainS1143 as a typical example of group 1 lysates, reacting with allindividual sera and (B) lysate of strain S1120 as a typical example ofgroup 2 lysates, reacting with a limited number of individual sera.Lanes a through s were developed with individual sera coded a

through s, respectively. Individual sera were absorbed with E. colias described in Materials and Methods.

No significant homologies between any one of these trans-lation products and any of the other mycobacterial antigenswere found.Taking these results together, we conclude that the recom-

binant M. leprae antigens selected with sera from contactsare not homologous or related to any other well-character-ized mycobacterial antigen.

Reactivity of recombinant antigens with individual sera. Toinvestigate which of the sera in the pool contained antibodiesto the fusion proteins expressed by the selected recombi-nants, lysates of these recombinant clones were subjected toWestern blotting with the individual sera as probes. Basedon the results from the Western blotting experiments, theantigens could be divided into two groups. Group 1 antigensshowed a weak reaction with all tested sera, i.e., S1113,S1114, S1115, S1116, S1119, S1123, S1141, S1142, andS1143. Group 2 antigens reacted strongly with a limitednumber of sera, i.e., S1120 (reacting with serum c), S1129(reacting with sera e, h, m, and n), and S1130 (reacting with

serum k). As typical examples for each group of antigens,Western blots of lysates of S1143 (group 1) and S1120 (group2) are shown in Fig. 3.

DISCUSSION

In this paper we report the selection and characterizationof several antigenic determinants of M. leprae that arerecognized by sera from well-documented contacts of lep-rosy patients. For this selection, we constructed a genomiclibrary of DNA from nonirradiated M. leprae.The previously described lambda gtll::M. leprae expres-

sion library (45) was shown to contain a relatively highpercentage of a number of identical recombinants (37; un-published observations). This is probably caused by effec-tive multiplication of these recombinants during amplifica-tion of the library. Recombinants with a low replication rateare present in low numbers in the library and hence caneasily be missed in the screening. Instead of using anamplified library, we preferred to construct a new expressionlibrary and to use it without significant amplification in thescreening procedure.From this library we selected 12 recombinants that pro-

duced antigenic determinants clearly recognized by serafrom contacts on Western blots. All antigens were producedas fusion proteins ranging from 116K to 145K. This meansthat parts of 2K to 31K M. leprae antigens were expressedby the various recombinants. The corresponding DNA in-serts had sufficient coding capacities.To further characterize the recombinant antigens, we

established the nucleotide sequence of a 84- to 180-bpsegment of the various inserts adjacent to lacZ. Based onthese nucleotide sequences and on the apparent molecularweights of the recombinant antigens, as well as on the resultsof DNA hybridizations, we conclude that all 12 recombi-nants probably encode different antigens. One of sevenrecombinants showing conjugate binding in the absence ofserum was characterized in a similar way (S1140). Whethersuch expressed recombinants might bind to the Fc part ofimmunoglobulins needs further investigation.

Screening of M. leprae libraries with sera of leprosypatients (9, 29) or their contacts (this report) has resulted inthe selection of recombinants producing a variety of anti-gens. A relatively large number of antigens had been identi-fied earlier in M. leprae by using blotting techniques (8, 11,19). In contrast, murine monoclonal antibodies have pre-dominantly recognized six antigens, 12K, 18K, 28K, 36K,65K, and 70K (39), which have also been frequently selectedfrom recombinant libraries. This discrepancy in the numberof antigens recognized by murine monoclonal antibodies andhuman sera might be a reflection of both quantitative andqualitative differences in their immune responses to M.leprae.Comparison of nucleotide and amino acid sequences of the

contact clones did not reveal any significant homology withother well-characterized antigens from M. leprae or fromother mycobacteria, including the 12K, 18K, 65K, and 70Kheat shock proteins (33, 39, 44). Apparently, the sera con-tained only limited amounts of antibodies to these heat shockproteins. At first glance this seems surprising, since thesestress-related proteins are supposed to be major immunolog-ical targets (44) and antibody responses to some of theseantigens have been reported in leprosy patients (25). Apossible explanation for the absence of heat shock protein-producing recombinants could be that heat shock proteinsform a well-conserved group of molecules also present in E.

VOL. 58, 1990 2825

on March 30, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

2826 HARTSKEERL ET AL.

coli (33, 36). Hence, cross-reactive antibodies against heatshock proteins have very likely been depleted during theabsorption step. In agreement with this explanation, no heatshock protein-producing recombinants could be selectedwith E. coli-absorbed patient sera (29). On the other hand, itmight be possible that no such antibodies are present in mostsera from contacts.Two groups of recombinant antigens could be distin-

guished: one group was recognized by the sera of all contactsused in this study; the other group was recognized by thesera of only a limited number of contacts. Apparently, some

mycobacterial antigenic determinants are capable of induc-ing antibody responses independent of host-related factors,whereas others are not.

In this study we describe the identification of M. leprae

antigenic determinants that may play a role at the early stageof infection. Household contacts used in this study havelived in the close vicinity of untreated lepromatous leprosypatients for long periods and are very likely to be infectedwith M. leprae. The contacts have not been vaccinated withM. bovis BCG. Therefore, we assume that the antibodiespresent in the absorbed sera are the products of a humoralresponse to M. leprae. However, the possibility cannot beexcluded that a portion of the antibodies is directed tocross-reactive determinants and results from a response toother mycobacteria. The established DNA sequences had a

G+C content that was much higher than that reported for theM. leprae genome and more comparable to those of a

number of other mycobacteria (10, 17). Although the se-

quenced parts of the genes may not be representative for thecomplete genes, this might indicate that the expressed M.leprae determinants are not specific for this bacillus butcould be common in a number of other pathogenic andnonpathogenic mycobacteria.The contacts used in this study have not yet developed

leprosy, which suggests that they have some form of immu-nity to the disease. As a continuation of this study, completegenes of contact antigens will be selected from a cosmidlibrary. Reactivity patterns of sera of leprosy patients andtheir contacts with the complete recombinant antigens, as

well as T-cell responses to these antigens, will be deter-mined.The available methodology for the study of infection with

M. leprae and immunity to such infection is inadequate. Therecent development of a polymerase chain reaction for thesensitive and specific detection of M. leprae (15) has openedwider perspectives for the future. In this report we describepreviously unidentified antigens, which may be relevant toimmune responses during the early phase of infection. Thecombination of studies of infection and immunity shouldprovide valuable information and contribute to the develop-ment of tools necessary for the eradication of leprosy.

ACKNOWLEDGMENTS

We thank Caroline Hermans for her assistance in part of the work,Jelle Thole for his advice, R. V. Cellona (Leonard Wood MemorialCenter for Leprosy Research, Cebu City, The Philippines) for hisgenerous gift of the sera, and Pamela Wright for critically readingthe manuscript.

This investigation received financial support from the NetherlandsLeprosy Relief Association and from the Commission of EuropeanCommunities Directorate General for Science, Research and Devel-opment (grant TS2-0111-NL).

LITERATURE CITED1. Andersen, A. B., and E. B. Hansen. 1989. Structure and mapping

of antigenic domains of protein antigen b, a 38,000-molecular-

weight protein of Mycobacterium tuberculosis. Infect. Immun.57:2481-2488.

2. Anonymous. 1986. Serological tests for leprosy (editorial). Lan-cet i:533-535.

3. Ashbridge, K. R., R. J. Booth, J. D. Watson, and R. B. Lathigra.1989. Nucleotide sequence of the 19 kDa antigen from Myco-bacterium tuberculosis. Nucleic Acids Res. 17:1249.

4. Baird, P. N., M. C. Hall, and A. R. M. Coates. 1988. A majorantigen from Mycobacterium tuberculosis which is homologousto the heat shock proteins groES from E. coli and the hptA geneproduct of Coxiella burneti. Nucleic Acids Res. 16:9047.

5. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extractionprocedure for screening recombinant plasmid DNA. NucleicAcids Res. 7:1513-1524.

6. Booth, R. J., D. P. Harris, J. M. Love, and J. D. Watson. 1988.Antigenic proteins of Mycobacterium leprae: complete se-quence of the gene for the 18-kDa protein. J. Immunol. 140:597-601.

7. Burnette, W. N. 1981. "Western-blotting": electrophoretictransfer of proteins from sodium dodecyl sulfate-polyacryl-amide gels to unmodified nitrocellulose and radiographic detec-tion with antibody and radioiodinated protein A. Anal. Bio-chem. 112:195-203.

8. Chakrabarty, A. K., M. Maire, and P.-H. Lambert. 1982.SDS-PAGE analysis of M. leprae protein antigens reacting withantibodies from the sera from lepromatous patients and infectedarmadillos. Clin. Exp. Immunol. 49:523.

9. Cherayil, B. J., and R. A. Young. 1988. A 28-kDA protein fromMycobacterium leprae is a target of the human antibody re-sponse in lepromatous leprosy. J. Immunol. 141:4370-4375.

10. Clark-Curtiss, J. E., W. R. Jacobs, M. A. Docherty, L. R.Ritchie, and R. Curtiss III. 1985. Molecular analysis ofDNA andconstruction of genomic libraries of Mycobacterium leprae. J.Bacteriol. 161:1093-1102.

11. Converse, P. J., T. H. M. Ottenhoff, N. Gebre, J. P. Ehrenberg,and R. Kiessling. 1988. Cellular, humoral, and gamma interferonresponses to Mycobacterium leprae and BCG antigens inhealthy individuals exposed to leprosy. Scand. J. Immunol.27:515-525.

12. Douglas, J. T., R. V. Cellona, R. M. Abalos, M. G. Madarang,and T. T. Fajardo. 1987. The serological reactivity and earlydetection of leprosy among contacts of lepromatous patients inCebu, Philippines. Int. J. Lepr. 55:718-721.

13. Fine, P. E. M., J. M. Ponnighaus, P. Burgess, J. A. Clarkson,and C. C. Draper. 1988. Seroepidemiological studies of leprosyin northern Malawi based on an enzyme-linked immunosorbentassay using synthetic glycoconjugate antigen. Int. J. Lepr.56:243-254.

14. Garsia, R. J., L. Hellqvist, R. J. Booth, A. J. Radford, W. J.Britton, L. Astbury, R. J. Trent, and A. Basten. 1989. Homologyof the 70-kilodalton antigens from Mycobacterium leprae andMycobacterium bovis with the Mycobacterium tuberculosis71-kilodalton antigen and with the conserved heat shock protein70 of eucaryotes. Infect. Immun. 57:204-212.

15. Hartskeerl, R. A., M. Y. L. de Wit, and P. R. Klatser. 1989.Polymerase chain reaction for the detection of Mycobacteriumleprae. J. Gen. Microbiol. 135:2357-2364.

16. Hartskeerl, R. A., L. F. E. M. Stabel, C. J. Hermans, P. R.Klatser, and J. E. R. Thole. 1990. Nucleotide and deducedamino acid sequence of a Mycobacterium leprae 12K protein.Nucleic Acids Res. 18:1294.

17. Imaeda, T., W. F. Kirchheimer, and L. Barksdale. 1982. DNAisolated from Mycobacterium leprae: genome size, base ratio,and homology with other related bacteria as determined byoptical DNA-DNA reassociation. J. Bacteriol. 150:414-417.

18. Kalnins, A., K. Otto, U. Ruther, and B. Muller-Hill. 1983.Sequence of the lacZ gene of Escherichia coli. EMBO J.2:593-597.

19. Klatser, P. R., M. M. van Rens, and T. A. Eggelte. 1984.Immunochemical characterization of Mycobacterium leprae an-tigens by the SDS-polyacrylamide gel electrophoresis immuno-peroxidase technique (SGIP) using patients' sera. Clin. Exp.Immunol. 56:537-544.

INFECT. IMMUN.

on March 30, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

M. LEPRAE ANTIGENS RECOGNIZED BY CONTACTS' SERA 2827

20. Kuwabara, S. 1975. Amino acid sequence of tuberculin-activeprotein from Mycobacterium tuberculosis. J. Biol. Chem. 250:2563-2568.

21. Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature (London)226:680-685.

22. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

23. Matsuo, K., R. Yamaguchi, A. Yamazaki, H. Tasaka, K.Terasaka, and T. Yamada. 1990. Cloning and expression of thegene for the cross-reactive a antigen of Mycobacterium kan-sasii. Infect. Immun. 58:550-556.

24. Matsuo, K., R. Yamaguchi, A. Yamazaki, H. Tasaka, and T.Yamada. 1988. Cloning and expression of the Mycobacteriumbovis BCG gene for extracellular a antigen. J. Bacteriol. 170:3847-3854.

25. Meeker, H. C., D. L. Williams, D. C. Anderson, T. P. Gills, G.Schuller-Levis, and W. R. Levis. 1989. Analysis of humanantibody epitopes on the 65-kilodalton protein of Mycobacte-rium leprae by using synthetic peptides. Infect. Immun. 57:3689-3694.

26. Mehra, V., D. Sweetser, and R. A. Young. 1986. Efficientmapping of protein antigenic determinants. Proc. Natl. Acad.Sci. USA 83:7013-7017.

27. Radford, A. J., B. J. Duffield, and P. Plackett. 1988. Cloning ofa species-specific antigen of Mycobacterium bovis. Infect. Im-mun. 56:921-925.

28. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequenc-ing with chain terminating inhibitors. Proc. Natl. Acad. Sci.USA 74:5463-5467.

29. Sathish, M., R. E. Esser, J. E. R. Thole, and J. E. Clark-Curtiss.1990. Identification and characterization of antigenic determi-nants of Mycobacterium leprae that react with antibodies inleprosy patients' sera. Infect. Immun. 58:1327-1336.

30. Shinnick, T. M. 1987. The 65-kilodalton antigen of Mycobacte-rium tuberculosis. J. Bacteriol. 169:1080-1088.

31. Shinnick, T. M., B. B. Plikaytis, A. D. Hyche, R. M. vanLandingham, and L. L. Walker. 1989. The Mycobacteriumtuberculosis BCG-a protein has homology with the Escherichiacoli GroES protein. Nucleic Acids Res. 17:1254.

32. Shinnick, T. M., D. Sweetser, J. Thole, J. van Embden, andR. A. Young. 1987. The etiological agents of leprosy andtuberculosis share an immuno-reactive protein antigen with thevaccine strain Mycobacterium bovis BCG. Infect. Immun. 55:1932-1935.

33. Shinnick, T. M., M. H. Vodkin, and J. C. Williams. 1988. TheMycobacterium tuberculosis 65-kilodalton antigen is a heatshock protein which corresponds to common antigen and to the

Escherichia coli GroEL protein. Infect. Immun. 56:446-451.34. Stanley, K. K., and J. P. Luzio. 1984. Construction of a new

family of high efficiency bacterial expression vectors: identifi-cation of cDNA clones coding for human liver proteins. EMBOJ. 6:1429-1434.

35. Thangaraj, H. S., F. I. Lamb, E. 0. Davis, and M. J. Colston.1989. Nucleotide and deduced amino acid sequence of Myco-bacterium leprae manganese superoxide dismutase. NucleicAcids Res. 17:8378.

36. Thole, J. E. R., P. Hindersson, J. de Bruyn, F. Cremers, J. vander Zee, H. de Cock, J. Tommassen, W. van Eden, and J. D. A.van Embden. 1988. Antigenic relatedness of a strongly immuno-genic 65 kDa mycobacterial protein antigen with a similar sizedubiquitous bacterial common antigen. Microb. Pathogen. 4:71-83.

37. Thole, J. E. R., L. F. E. M. Stabel, M. E. G. Suykerbuyk,M. Y. L. de Wit, P. R. Klatser, and R. A. Hartskeerl. 1990. Amajor immunogenic 36,000-molecular-weight antigen from My-cobacterium leprae contains an immunoreactive region of pro-line-rich repeats. Infect. Immun. 58:80-87.

38. Van Embden, J. D. A., H. J. van Donk, R. V. van Eijk, H. G. vander Heide, J. A. de Jong, M. F. van Olderen, A. D. Osterhaus,and L. M. Schouls. 1983. Molecular cloning and expression ofTreponema pallidum DNA in Escherichia coli K-12. Infect.Immun. 42:187-196.

39. Watson, J. D. 1989. Leprosy: understanding protective immu-nity. Immunol. Today 10:218-221.

40. World Health Organization. 1980. Purification of M. leprae.Report of the 5th meeting of the scientific working group on theimmunology of leprosy (IMMLEP). TDR/IMMLEP-SGW(5)/80.3. World Health Organization, Geneva.

41. World Health Organization. 1988. WHO Expert Committee onLeprosy: sixth report. Technical report series, no. 768. WorldHealth Organization, Geneva.

42. Yamaguchi, R., K. Matsuo, A. Yamazaki, C. Abe, S. Nagai, K.Terasaka, and T. Yamada. 1989. Cloning and characterization ofthe gene for immunogenic protein MPB64 of Mycobacteriumbovis BCG. Infect. Immun. 57:283-288.

43. Yamaguchi, R., K. Matsuo, A. Yamazaki, S. Nagai, T. Terasaka,and T. Yamada. 1988. Immunogenic protein MPB57 from My-cobacterium bovis BCG: molecular cloning, nucleotide se-quence and expression. FEBS Lett. 240:115-117.

44. Young, D., R. Lathigra, R. Hendrix, D. Sweetser, and R. A.Young. 1988. Stress proteins are immune targets in leprosy andtuberculosis. Proc. Natl. Acad. Sci. USA 85:4267-4270.

45. Young, R. A., V. Mehra, D. Sweetser, T. Buchanan, J. Clark-Curtiss, R. W. Davis, and B. R. Bloom. 1985. Genes for themajor protein antigens of the leprosy parasite Mycobacteriumleprae. Nature (London) 316:450452.

VOL. 58, 1990

on March 30, 2020 by guest

http://iai.asm.org/

Dow

nloaded from