Journal of Dermatological Science (2007) 46, 101—110

www.intl.elsevierhealth.com/journals/jods

Recombinant Mycobacterium leprae proteinassociated with entry into mammalian cells ofrespiratory and skin components

Naoya Sato a, Takao Fujimura a,*, Mikio Masuzawa a, Yasuko Yogi b,Masanori Matsuoka b, Maho Kanoh a, Lee W. Riley c, Kensei Katsuoka a

aDepartment of Dermatology, Graduate School of Medical Science, Kitasato University, 1-15-1 Kitasato,Sagamihara, Kanagawa 228-8555, Japanb Leprosy Research Center, National Institute of Infectious Diseases, 4-2-1 Aobacho, Higashimurayama,Tokyo 189-0002, Japanc Program in Infectious Diseases and Immunity, School of Public Health, University of California, Berkeley,140 Warren Hall, Berkeley, CA 94720, USA

Received 1 December 2006; received in revised form 18 January 2007; accepted 18 January 2007

KEYWORDSMycobacterium leprae;mce1A gene;Recombinant protein;Mammalian cell entry;Nasal epithelial cells;Endothelial cells;Leprosy

Summary

Backgrounds: The transmission of Mycobacterium leprae, the causative pathogen ofleprosy, has been postulated to occur mainly through upper respiratory route ratherthan skin-to-skin contact via minor injuries. The M. leprae genome contains mce1Agene, which encodes a putative mammalian cell entry protein. However, to date,there have been no functional analyses of the M. leprae mce1A gene product.Objective: The aim of this study was to elucidate a possible relationship between thistransmission mechanism and the mce1A gene product.Methods: We analyzed the cell uptake activity in vitro of polystyrene latex beadscoated with a purified recombinant (r-) protein expressed by a 849-bp locus within themce1A gene.Results: The r-protein promoted uptake of the beads into human nasal epithelialcells derived from nasal polyps, human bronchial epithelial cell line, normal humandermal fibroblasts, normal human microvascular endothelial cells and normal humankeratinocytes cultured at 0.01 mM extracellular calcium concentration [Ca]; nouptake occurred with keratinocytes cultured at 1.2 mM [Ca].Conclusion: These results suggest that the mce1A gene product can mediate M.leprae entry into respiratory epithelial cells as their natural target cells, which maybe the main mode of transmission. Endothelial cells, on the other hand, may serve asthe reservoir of the bacilli for long-term infection. The M. leprae Mce1A protein has

* Corresponding author. Tel.: +81 42 778 9216; fax: +81 42 778 8628.E-mail address: fujimura@med.kitasato-u.ac.jp (T. Fujimura).

0923-1811/$30.00 # 2007 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.jdermsci.2007.01.006

102 N. Sato et al.

potential important implications for mode of transmission and pathogenesis ofleprosy.# 2007 Japanese Society for Investigative Dermatology. Published by Elsevier IrelandLtd. All rights reserved.

1. Introduction

Leprosy is a chronic granulomatous infection of theskin and peripheral nerves with the intracellularbacterium Mycobacterium leprae, and a great dealof research has been conducted on its transmissionmode. Two routes — through the nasal mucosa andthrough minor skin injuries [1] — have been postu-lated as the principal modes of transmission of M.leprae, but the nasal mucosa has been recentlysuggested to be the principal route [2]. Themechan-ism of entry into the nasal mucosa or skin by M.leprae, however, has not been clearly elucidated.

Themce1A gene inmce1 operon (mammalian cellentry) of Mycobacterium tuberculosis genome is avirulence gene involved in epithelial cell entry andsurvival inside macrophages [3]. The recombinant(r-) M. tuberculosis Mce1A protein has been shownto promote epithelial cell uptake of latex micro-spheres coated with the r-protein [4], and the abil-ity of epithelial cell entry can be conferred uponnon-pathogenic Escherichia coli by expressing it ontheir surface [5]. The M. leprae genome sequence[6] has revealed a region containing eight genes(yrbE1A, yrbE1B, mce1A, mce1B, mce1C, lprK,mce1E and mce1F) with close identity of theencoded proteins to the M. tuberculosis mce1operon. The protein encoded by M. leprae mce1Agene is a putative epithelial cell entry protein,suggesting that it participates in the entry of M.leprae into the nasal mucosa and skin. However, todate, there have been no functional analyses of theM. leprae mce1A gene product.

To elucidate the role of M. leprae mce1A gene inthe transmission mode of leprosy, we generated a r-protein and investigated its entry activity in respira-tory epithelial cells as well as the cells of skincomponents, which comprise the possible initialtransmission routes of M. leprae.

Fig. 1 DNA sequence and purification of the recombi-nant M. leprae truncated Mce1A protein (r-Mce1a). (A)The full-length Mce1A protein is encoded by 1326-bp ORFindicated as mce1A locus in the M. leprae strain TNgenome sequence. The r-Mce1a represents a truncatedprotein with 283-amino-acid residues deleted at the N-and C-terminus of the full-length Mce1A protein. (B)Coomasie brilliant blue stain of a 12% SDS-PAGE gel revealsa purified 37-kDa protein (lane 2) expressed by 849-bp ORFDNA fragment cloned fromM. leprae strain Thai 53. Lane 2contains about 2 mg of a protein. Lane 1 contains amolecular weight marker.

2. Materials and methods

2.1. Bacterial strain and plasmid

The genomic DNA used in the study was isolatedfromM. leprae strain Thai 53, which was maintainedat Leprosy Research Centre, National Institute ofInfectious Diseases, Japan, as previously described[7]. The pQE30 plasmid and E. coliM15 [pREP4] were

purchased from Qiagen (Valencia, CA, USA). ThepQE30 plasmid was used as expression vector. E.coliM1 5 [pREP4] was used as a host for the vector, asrecommended by the manufacturer. The use ofpQE30 vector allowed the expression of Mce1A pro-tein ofM. lepraewith a polyhistidine (6 � His) tag atthe N-terminus (r-Mce1a).

2.2. Construction of vector

In Sanger Centre M. leprae strain TN complete gen-ome sequence, mce1A gene is a 1326-bp putativeopen reading frame (ORF) located between positions3092446 and 3093771 (NCBI-GeneID: 910890). Themce1A DNA sequence of strain Thai 53 was identicalto that of strain TN. It was subcloned into pQE30vector in a truncated reading frame. The 849-bp ORFdeleted at 50 and 30 ends of mce1A gene is locatedbetweenpositions 73and921 (Fig. 1A).This sequencewas amplified by polymerase chain reaction (PCR)directly from the genomic DNA of M. leprae strainThai 53 with oligonucleotide primers designed

Recombinant M. leprae Mce1A protein 103

to introduce SacI and HindIII endonucleaserestriction sites at the ends. The sequences of usedprimers and PCR condition were as follows: sense,50-CTGAAGCTTGGGTTCGATGTCATGATAATT-30, anti-sense, 50-CGGGAGCTCGTGGCTGTAGTGATTTTC-30,35 amplification cycles at 94 8C for 30 s, 65 8C for30 s and72 8C for90 s, followedbyafinal extensionat72 8C for 7 min. The amplified products were ligatedinto the pQE30 vector linearized with SacI—HindIII.The resultant plasmid (pQE30/mce1a) was clonedinto E. coli M15 [pREP4] by electroporation (GenePulser1 II, Bio-Rad, Hercules, CA, USA), according tothe manufacturer’s instructions. The cloned frag-ment in the plasmid was sequenced by ABI PRISM1

310 Genetic Analyzer (Applied Biosystems, FosterCity, CA, USA) to confirm the correct sequence,orientation and frame of insert DNA fragment.

2.3. Protein expression and purification

Recombinant protein was expressed and purifiedaccording to manufacture’s instruction. Briefly, E.coli M15 [pREP4] containing pQE30/mce1a plasmidwas grown overnight in 10-ml superbroth contain-ing 100 mg/ml ampicillin and 50 mg/ml kanamycin.A 500 ml aliquot of bacterial suspension was pel-leted, resuspended in 30 ml of superbroth andincubated at 37 8C for 1—2 h until OD600 = 0.6.Then isopropyl b-D-thiogalactoside was added tofinal concentration 1 mM and incubated for 3 h at37 8C. The induced and uninduced r-E. coli strainswere analyzed by SDS-polyacryamide gel electro-phoresis (SDS-PAGE). The newly expressed proteinformed an inclusion body in the r-E. coli host. Theinclusion body was therefore purified under dena-turing conditions according to the instructions ofthe expression vector’s respective manufactures.The polyhistidine tagged Mce1a solubilized withlysis buffer (8 M urea, 10 mM Tris—HCl, 100 mMNaH2PO4, pH 8.0) was bound to a Ni—NTA resincolumn equilibrated with lysis buffer, and waseluted by elusion buffer (8 M urea, 10 mM Tris—HCl, 100 mM NaH2PO4, 20—250 mM imidazole, pH6.3). The eluted protein were subsequentlyrefolded with 1 mM dithiothreitol (Sigma, St.Louis, MO, USA) and 0.1 mM phenylmethylsulfonylfluoride (Sigma) by dialysis, gradually removingurea. The r-Mce1a was finally purified and refoldedas a soluble protein. The protein was separated bySDS-PAGE and was analyzed for purity by Coomasiebrilliant blue R-250 staining. The recombinantM. tuberculosis truncated Mce1A protein (r-Mtb-Mce1a) encoded by 606-bp locus in its mce1Agene (NCBI-GeneID: 886823) of the strain H37Rvgenome was expressed and purified as previouslydescribed [4].

2.4. Cell culture

HeLa cells and BEAS-2B were purchased from Amer-ican Type Culture Collection (ATCC, Manassas, VA,USA). Normal human dermal fibroblasts (NHDF),normal human microvascular endothelial cells(HMVEC) and normal human epidermal keratino-cytes (NHEK), all of them were derived from neo-natal foreskin, were purchased from Kurabo (Osaka,Japan). HeLa cells (ATCC CCL-2) were maintainedwith Dulbecco’s modified Eagle’s media (Invitrogen,Carlsbad, CA, USA) supplemented with 50 mg/mlgentamicin (GM) (Invitrogen) and 10% fetal bovineserum (FBS) (JRH Biosciences, Lenexa, KS, USA).BEAS-2B (ATCC CRL-9609) is a normal human bron-chial epithelial cell line infected with adenovirus12-simian virus-40 hybrid virus. BEAS-2B was main-tained in bronchial epithelial cell growth media(BEGM1, Cambrex Bio Science, Walkersville, MD,USA) supplemented with 10% FBS, 13 mg/ml bovinepituitary extract (BPE), 0.5 ng/ml recombinant epi-dermal growth factor (hEGF), 0.5 ng/ml hydrocor-tisone, 0.5 ng/ml epinephrine, 10 mg/ml retinoicacid, 6.5 ng/ml triiodothyronine, 50 mg/ml GMand 50 ng/ml amphotericin-B (Cambrex BioScience). NHDF was grown in basal media (Medium106S1, Kurabo) supplemented with 2% FBS, 1 mg/mlhydrocortisone, 10 ng/ml hEGF, 3 ng/ml humanbasic fibroblast growth factor (hFGF), 10 mg/mlheparin and 50 mg/ml GM. The third-passage NHDFwas used in this study. HMVEC was grown in basalmedia (HuMedia-EB21, Kurabo) supplemented with5% FBS, 1 mg/ml hydrocortisone, 10 ng/ml hEGF,5 ng/ml hFGF, 10 mg/ml heparin, 39.3 mg/ml dibu-tyryl cyclic AMP, 50 mg/ml GM and 50 ng/ml ampho-tericin-B. The fifth-passage HMVEC was used in thisstudy. NHEK was grown in serum-free keratinocytegrowth media (HuMedia-KB21, Kurabo) supplemen-ted with 0.4% BPE, 10 mg/ml bovine insulin,0.18 mg/ml hydrocortisone, 0.2 ng/ml hEGF,50 mg/ml GM and 50 ng/ml amphotericin-B (Kurabo)containing 0.01 mM calcium at the second-passage.The third-passage NHEK was used at various con-centrations of calcium as indicated in this study. Allcells were used 1 day after replating and werecultured at 37 8C in a 5% CO2 incubator.

2.5. Preparation of primary human nasalepithelial cells

Primary human nasal epithelial cells derived fromnasal polyps were prepared according to the methoddescribed by Yankaskas et al. [8]. The tissueswere obtained from a patient with nasal polypsoperated at the Department of Otorhinolaryngology,Kitasato University Hospital (Kanagawa, Japan)

104 N. Sato et al.

after informed consent was obtained for this study.The tissues of nasal polyps were washed by McCoy’s5A media (Invitrogen), and incubated with McCoy’s5A media supplemented with 50 mg/ml GM and 0.1%protease type XIV (Sigma) at 4 8C for 16 h. Then 10%FBS was added to neutralize the protease. Theepithelial cells were detached by gentle agitation,filtered through a 70-mm nylon mesh cell strainer(Becton Dickinson, Franklin Lakes, NJ, USA), cen-trifuged at 800 � g for 5 min and the pellet wasresuspended in McCoy’s 5A supplemented with50 mg/ml GM and 10% FBS. The cells (3.5 � 105)were harvested on a 12.5-cm2 flask (Becton Dick-inson) coated with type I collagen (Vitrogen1001,Collagen, Palo Alto, CA, USA) for cell uptake assayby electron microscopy, and 105 cells were har-vested on a type I collagen-coated well of 24-wellplate (Becton Dickinson) for cell uptake assay byfluorometer. The cells were incubated at 37 8C with5% CO2 for 24—48 h to near-confluence.

2.6. Cell uptake assay of protein-coatedlatex beads by electron microscopy

A 30 ml of stock suspensions of 1.1-mm-diameterpolystyrene latex beads, containing 5 � 108 beads/ml (Sigma), weremixed in 150 ml of phosphate buffersaline (PBS) containing 50 mg/ml of each set of pro-tein and incubated for 16 hat37 8C.After incubation,the samples were centrifuged at 7000 � g and resus-pended in 750 ml of PBS. A 500-ml sample of thissuspension was added to a near-confluent culturedcell monolayer grown in a 25-cm2 flask containing7 mlof appropriatemedia for culturedcells. Thecellswere incubated for 5 h at 37 8C in a CO2 incubator,washed four times with PBS and one time with 0.1 Mcacodylate phosphate buffer (pH 7.6), and then col-lected with cell-scraper (Becton Dickinson). The col-lected cells were fixed with 2% glutaraldehyde in0.1 M cacodylate phosphate buffer (pH 7.6) at 4 8Covernight, post-fixed with 1% osmium tetroxide inPBS, dehydrated through graded ethanol solutionsand embedded in Spurr’s low-viscosity embeddingmedia. The ultrathin sections were stained withuranyl acetate and lead citrate and examined withtransmission electron microscope (JEM-1200EX,JEOL, Tokyo, Japan). Uncoated beads and coatedbeads with bovine serum albumin fraction V (BSA)were used as negative controls.

2.7. Cell uptake assay of protein-coatedfluorescent latex beads by fluorometer

Fluorescence emission by protein-coated 1.0-mm-diameter fluorescent polystyrene latex beads (Poly-sciences, Warrington, PA, USA) was measured to

quantify the level of association with each set ofcultured cells. A 7 ml of stock suspensions of fluor-escent beads containing 4.55 � 1010 beads/ml weremixed in 100 ml of PBS containing each set of 50 mg/ml of protein and incubated for 16 h at 37 8C. Afterincubation, the samples were centrifuged and thepellet was resuspended with 250 ml of PBS. A 100-mlsample of this suspension was added to a near-confluent cultured cell monolayer grown in eachwell of a 24-well polystyrene tissue culture platecontaining 1 ml of appropriate media for cells (105

cells per well). The cells were incubated for 4 h at37 8C in a CO2 incubator, washed five times with PBSand fluorescence emission (fluorescence units per105 cells) per well was detected by fluorometer(Spectrafluor Plus, Tecan, Grodig, Austria).Uncoated beads and coated beads with BSA wereused as controls. Each set of protein-coated beadswas tested in triplicate, and a mean (�S.D.) fluor-escence emission (fluorescence units per 105 cells)determination was calculated for each set. Thestatistical significance of the mean values was com-pared by the Student’s t-test. P-values of <0.05were considered as statistically significant.

3. Results

3.1. Expression and purification of therecombinant M. leprae truncated Mce1Aprotein (r-Mce1a)

The full-length Mce1A protein of M. leprae containsa possible N-terminal signal sequence. Because ofthe high hydrophobicity and poor solubility of thefull-length protein expressed in E. coli (data notshown), we cloned an 849-bp segment of the M.leprae mce1A gene, which yielded a truncated deri-vative (r-Mce1a) that was soluble and enabled us tocarry out its large-scale expression and purification.The r-Mce1a is an approximately 37-kDa proteinwith 283-amino acid residues deleted at N- and C-terminus of the full-length Mce1A protein (Fig. 1A).The r-Mce1a was purified and refolded as a solubleprotein of approximately 37 kDa (Fig. 1B).

3.2. HeLa cell uptake of protein-coatedbeads is mediated by r-Mce1a

To investigate the mammalian cell associationinvolved in attachment and entry activity of purifiedr-Mce1a, we initially performed HeLa cell uptakeassay of protein-coated beads by transmission elec-tron microscopy and fluorometer. Since the recom-binant M. tuberculosis truncated Mce1A protein(referred to as r-Mtb-Mce1a) encoded by a 606-bp

Recombinant M. leprae Mce1A protein 105

locus in its mce1A gene of the strain H37Rv wasshown previously to promote cell uptake of theseprotein-coated beads into HeLa cells [4], it was usedas a positive control in this assay. Internalization of

Fig. 2 HeLa cell uptake assay of protein-coated latex beaTransmission electronmicroscopy of HeLa cells incubated for 5uncoated (D) polystyrene latex beads. The oval-shaped electrinside cells (A and B). No beads were visible inside cells (Cassociation of fluorescein isothiocyanate labeled polystyreneMce1a, r-Mce1a and BSA by fluorescence emission determinatioand 50 mg/ml of each protein was used to coat the beads. Bafrom three wells for each protein. Asterisks indicate significavalues < 0.01). Double-asterisk represents P-value = 0.14.

r-Mce1a-coated beads into HeLa cells was confirmedby transmission electron microscopy, as well as r-Mtb-Mce1a-coated beads (Fig. 2A and B). To com-pare levels of cell association of the beads into HeLa

ds by transmission electron microscopy and fluorometer.h with r-Mtb-Mce1a- (A), r-Mce1a- (B), BSA- (C) coated andon-dense particles representing the beads were observedand D). Bars indicate 1 mm. (E) Comparison of HeLa celllatex beads coated with various preparations of r-Mtb-n. The incubation period for each set of the beads was 4 h,rs represent mean � S.D. fluorescence units per 105 cellsnt difference compared with each of negative control (P-

106 N. Sato et al.

Fig. 3 Transmission electron microscopy of various cultured cells incubated with protein-coated latex beads. Variouscultured cells were incubated for 5 h with r-Mce1a- (A—C, G—I) and BSA- (D—F, J—L) coated polystyrene latex beads. (Aand D) HNEpC as primary cultured nasal epithelial cells derived from nasal polyps. (B and E) BEAS-2B as a bronchialepithelial cell line. (C and F) NHDF as dermal fibroblasts. (G and J) HMVEC as dermal microvascular endothelial cells. (Hand K) NHEK as epidermal keratinocytes cultured at 0.01 mM extracellular calcium concentration [Ca]. (I and L) NHEKcultured at 1.2 mM [Ca]. The tonofibrils (arrows) indicating cell differentiation were observed inside NHEK (I and L). Theoval-shaped electron-dense particles (arrowheads) representing the beads were observed inside cells (A—C, G and H). Nobeads were visible inside cells (D—F, I—L). Fiftymicrograms permilliliter of each protein were used to coat the beads. Barsrepresent 1 mm.

Recombinant M. leprae Mce1A protein 107

cells, wemeasured fluorescence, expressed as fluor-escence units per 105 HeLa cells, with a fluorometer.The level of fluorescence emission of the beadscoated with r-Mce1a was equivalent to that of r-Mtb-Mce1a (Fig. 2E).

Fig. 4 Various cultured cell uptake assay of protein-coateemission determinations of protein-coated fluorescent polystyNHDF (C), HMVEC (D), NHEK (E) cultured at 0.01 mM extracel1.2 mM [Ca]. The incubation period for each set of the beads wbeads. The r-Mce1a-coated beads showed significantly higherBSA-coated and uncoated beads, used as negative controls, incultured at 0.01 mM [Ca] (E) except for NHEK cultured at 1.2fluorescence units per 105 cells from three wells for each proteeach of negative control (P-values < 0.05).

Uncoated and BSA-coated beads did not exhibitHeLa cell association. The attachment and entryactivity into HeLa cells of r-Mce1a was nearly iden-tical to that of r-Mtb-Mce1a. Thus, we used r-Mce1ain the following experiments.

d fluorescent latex beads by fluorometer. Fluorescencerene latex beads associating with HNEpC (A), BEAS-2B (B),lular calcium concentration [Ca] and NHEK (F) cultured atas 4 h, and 50 mg/ml of each protein was used to coat thelevels of fluorescent emission determinations than did theHNEpC (A), BEAS-2B (B), NHDF (C), HMVEC (D) and NHEKmM [Ca] (F). ND, not done. Bars represent mean � S.D.in. Asterisks indicate significant difference compared with

108 N. Sato et al.

3.3. R-Mce1a promotes the uptake ofprotein-coated beads into human nasalepithelial cells, bronchial epithelial cells,fibroblasts and endothelial cells

To elucidate whether the Mce1A protein is asso-ciated with entry of M. leprae into respiratoryand skin tissues, we analyzed the attachment andentry activity of r-Mce1a into primary culturedhuman nasal epithelial cells derived from nasalpolyps (HNEpC), human bronchial epithelial cell line(BEAS-2B), normal human dermal fibroblasts (NHDF)and human dermal microvascular endothelial cells(HMVEC). Internalization of r-Mce1a-coated beadswas confirmed by transmission electron microscopyin HNEpC, BEAS-2B, NHDF and HMVEC (Fig. 3A—C andG). BSA-coated (Fig. 3D—F and J) and uncoated(data not shown) beads as negative controls werenot observed inside any of the cultured cell types.The levels of cell association of r-Mce1a-coatedfluorescent beads were significantly higher thanthat of BSA-coated and uncoated beads in eachset of cultured cell (Fig. 4A—D) (P-values < 0.05).The r-Mce1a facilitated cell association of the beadsinto HNEpC, BEAS-2B, NHDF and HMVEC. Theseresults suggest that the Mce1A protein mediatescell entry of M. leprae into not only respiratoryepithelial cells as a respiratory tract component,but also fibroblasts and endothelial cells as dermalcomponents.

3.4. The cellular uptake of r-Mce1a-coated beads into keratinocytes dependson extracellular calcium concentration

Since M. leprae cannot enter intact skin, twotypes of normal human epidermal keratinocytes(NHEK) as epidermal components were preparedby changing extracellular calcium concentration[Ca] at 0.01 mM or 1.2 mM. In vitro, NHEK culturedat [Ca] below 0.07 mM proliferate and express abasal cell phenotype, and elevation of [Ca] to0.1 mM or above (calcium switch) leads to differ-entiation and morphological changes such as des-mosome formation, stratification and cornificationof NHEK [9—11]. In this assay, r-Mce1a promotedthe uptake of the beads into NHEK cultured at0.01 mM [Ca] (Figs. 3H and 4E). However, thisuptake was not observed in NHEK cultured at1.2 mM [Ca] (Figs. 3I and 4F). The cell uptake ofthe beads into NHEK mediated by r-Mce1a wassuppressed by the rise in [Ca] above 0.1 mM. Theseresults suggest that the entry of M. leprae intokeratinocytes mediated by the Mce1A proteindepends on the differentiated status of the kera-tinocytes.

4. Discussion

M. leprae is most commonly recognized to infect thenasal mucosa, skin, and peripheral nerves. Themechanism of Schwann cell entry of M. lepraemediated by laminin-2 and a-dystroglycan expressedin the peripheral nerve tissue, in particular, has beenanalyzed in detail by Rambukkana et al. [12—15].Concerning the cell entry mechanism of M. lepraeinto the nasal epithelial cells or skin components inthe process of reaching the Schwann cells, an alter-native mechanism is indicated by non-expression oflaminin-2 and a-dystroglycan in these components.However, this mechanism remains unclear.

The transmission route of M. leprae has beenmainly considered to be the thick skin contact withlepromatous leprosy (LL) patients in which thebacilli are thought to enter skin through minorinjuries. However, the experimental transmissionof leprosy in nude mice has been demonstrated byinoculation ofM. leprae through intact nasal mucosa[16,17], and untreated LL patients shed large num-bers of the bacilli from their nasal discharges [18].Based on the experimental and clinical findings, thetransnasal entry of M. leprae is generally taken asthe principal transmission route.

This study demonstrates that M. leprae Mce1Aprotein is involved in entry mechanism into the cellsof nasal mucosal and skin components. The r-Mce1afacilitated delivery of beads into human nasal andbronchial epithelial cells, fibroblasts, endothelialcells and keratinocytes cultured at 0.01 mM [Ca]as basal cell phenotype, but did not facilitateuptake into keratinocytes cultured at 1.2 mM [Ca]as keratinized phenotype. These results suggestedthat the Mce1A protein can allow M. leprae todirectly enter nasal epithelial cells and skin compo-nents through possible minor injuries as suggestedby the basal cell phenotype of keratinocytes, fibro-blasts and endothelial cells in the lower epidermisand dermis.

M. leprae is often observed in histopathologyinside fibroblasts and endothelial cells at sites oflesions in the nose and skin of LL patients [19,20].Endothelial cells, in particular, have been postu-lated as important reservoirs of M. leprae for long-term infection and endothelial cell entry of thebacilli has been postulated to be closely relatedto subsequent neural tissue entry [21,22]. Our find-ing that the recombinant M. leprae Mce1A proteinpromotes the endothelial cell uptake of protein-coated beads provides a potential rationale forrecent observations and speculations based onthem. The M. leprae Mce1A protein-mediated entryinto endothelial cells is crucial for the pathogenesisof leprosy as well as its transmission.

Recombinant M. leprae Mce1A protein 109

Taking into consideration the results of recentstudies on leprosy together with our own findings,the transmission mode of leprosy appears to be asfollows.M. leprae in the environmental aerosol mayattach and directly enter nasal epithelial cells or topoorly differentiated keratinocytes, fibroblasts andendothelial cells of the skin through minor injuries.This entry may be mediated by the Mce1A protein.Subsequently, M. leprae inside or escaping from theepithelial tissues may be phagocytosed by macro-phages, multiply escaping intracellular killing inmacrophages and disseminate hematogenously.After reaching endothelial cells of the vasa ner-vorum, the bacilli may invade peripheral nervetissue surrounded by Schwann cells expressing a-dystroglycan on the surface. The a-dystroglycan is atarget molecule for Schwann cell entry of the bacillithat use phenolic glycolipid-1 and 21 kDa surfaceprotein encoded by ML1683 gene in the M. lepraegenome through laminin-2-dependent pathway[12—15].

In this study we have demonstrated that theprotein encoded by M. leprae mce1A gene playsan important role in mammalian cell entry of thebacilli. However, the exact cell entry mechanism ofmycobacteria into non-phagocytic mammalian cellsmediated by the Mce protein has not been comple-tely elucidated. In M. tuberculosis, the cellularuptake of r-Mtb-Mce1a-coated beads into HeLa cellswas blocked by pre-treatment of the beads withantibody raised against r-Mtb-Mce1a or by pre-treat-ment of HeLa cells with the soluble r-protein [4]. Inaddition, internalization of E. coli-expressing M.tuberculosis Mce1A peptide by HeLa cells was pre-viously shown to require the actin rearrangement ofhost cells [5]. These observations suggest that themechanism of mammalian cell entry of M. tubercu-losismediated by its Mce1A protein involves a recep-tor-mediated zipper mechanism. However, nothingis known about receptors for the Mce1A protein.Although not known at this time, a similar mechan-ism may contribute to mammalian cell entry of M.leprae mediated by its Mce1A protein.

In conclusion, further understanding of the func-tion of Mce1A protein in M. leprae entry into nasalepithelial and endothelial cells may better eluci-date the mechanisms of transmission and pathogen-esis of leprosy, and has potential implications fornew strategies to prevent leprosy.

Acknowledgements

This study was supported in part by Grants-in-Aid forScientific Research from the Ministry of Education,Culture, Sports and Technology in Japan (No.

15591190) and by the Senior Scholar Award in GlobalInfectious Disease of the Ellison Medical Foundation(LWR).

References

[1] van Beers SM, de Wit MY, Klatser PR. The epidemiology ofMycobacterium leprae: recent insight. FEMS Microbiol Lett1996;136:221—30.

[2] Britton WJ, Lockwood DN. Leprosy. Lancet 2004;363:1209—19.

[3] Arruda S, Bomfim G, Knights R, Huima-Byron T, Riley LW.Cloning of an M. tuberculosis DNA fragment associated withentry and survival inside cells. Science 1993;261:1454—7.

[4] Chitale S, Ehrt S, Kawamura I, Fujimura T, Shimono N, AnandN, et al. Recombinant Mycobacterium tuberculosis proteinassociated with mammalian cell entry. Cell Microbiol2001;3:247—54.

[5] Casali N, Konieczny M, Schmidt MA, Riley LW. Invasionactivity of a Mycobacterium tuberculosis peptide presentedby the Escherichia coli AIDA autotransporter. Infect Immun2002;70:6846—52.

[6] Cole ST, Eiglmeier K, Parkhill J, James KD, Thomson NR,Wheeler PR, et al. Massive gene decay in the leprosy bacil-lus. Nature 2001;409:1007—11.

[7] Maeda S, Matsuoka M, Nakata N, Kai M, Maeda Y, HashimotoK, et al. Multidrug resistant Mycobacterium leprae frompatients with leprosy. Antimicrob Agents Chemother2001;45:3635—9.

[8] Yankaskas JR, Cotton CU, Knowles MR, Gatzy JT, Boucher RC.Culture of human nasal epithelial cells on collagen matrixsupports. A comparison of bioelectric properties of normaland cystic fibrosis epithelia. Am Rev Respir Dis1985;132:1281—7.

[9] Hennings H, Michael D, Cheng C, Steinert P, Holbrook K,Yuspa SH. Calcium regulation of growth and differentiationof mouse epidermal cells in culture. Cell 1980;19:245—54.

[10] Hennings H, Holbrook KA. Calcium regulation of cell—cellcontact and differentiation of epidermal cells in culture. Anultrastructural study. Exp Cell Res 1983;143:127—42.

[11] Tu CL, Oda Y, Komuves L, Bikle DD. The role of the calcium-sensing receptor in epidermal differentiation. Cell Calcium2004;35:265—73.

[12] Rambukkana A, Salzer JL, Yurchenco PD, Tuomanen EI. Neuraltargeting ofMycobacterium lepraemediated by theG domainof the laminin-alpha 2 chain. Cell 1997;88:811—21.

[13] Rambukkana A, Yamada H, Zanazzi G, Mathus T, Salzer JL,Yurchenco PD, et al. Role of alpha-dystroglycan as a Schwanncell receptor for Mycobacterium leprae. Science 1998;282:2076—9.

[14] Shimoji Y, Ng V, Matsumura K, Fischetti VA, Rambukkana A. A21-kDa surface protein of Mycobacterium leprae binds per-ipheral nerve laminin-2 and mediates Schwann cell invasion.Proc Natl Acad Sci USA 1999;96:9857—62.

[15] Ng V, Zanazzi G, Timpl R, Talts JF, Salzer JL, Brennan PJ,et al. Role of the cell wall phenolic glycolipid-1 in theperipheral nerve predilection ofMycobacterium leprae. Cell2000;103:511—24.

[16] Chehl S, Job CK, Hastings RC. Transmission of leprosy in nudemice. Am J Trop Med Hyg 1985;34:1161—6.

[17] McDermott-Lancaster RD, McDougall AC. Mode of transmis-sion and histology of M. leprae infection in nude mice. Int JExp Pathol 1990;71:689—700.

[18] Davey TF, Rees RJ. The nasal discharge in leprosy: clinicaland bacteriological aspects. Lepr Rev 1974;45:121—34.

110 N. Sato et al.

[19] McDougall AC, Rees RJ, Weddell AG, Kanan MW. The histo-pathology of lepromatous leprosy in the nose. J Pathol1975;115:215—26.

[20] Coruh G, McDougall AC. Untreated lepromatous leprosy:histopathological findings in cutaneous blood vessels. Int JLepr Other Mycobact Dis 1979;47:500—11.

[21] Scollard DM, McCormick G, Allen JL. Localization of Myco-bacterium leprae to endothelial cells of epineurial andperineurial blood vessels and lymphatics. Am J Pathol1999;154:1611—20.

[22] Scollard DM. Association of Mycobacterium leprae withhuman endothelial cells in vitro. Lab Invest 2000;80:663—9.