|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
In Peru, both CL and MCL are endemic. Leishmania (Viannia) braziliensis and L. (V.) peruviana are most frequently associated with CL, although L. (V.) guyanensis, L. (V.) lainsoni and L. (Leishmania) amazonensis have also been reported.1 MCL is attributed to L. braziliensis. However, lesions from L. braziliensis and L. peruviana are not distinct in the early stages of CL, and species have often been incriminated on the basis of known geographical range, with L. peruviana found mostly in the western Andes and inter-Andean valleys and L. braziliensis occurring predominantly at lower altitudes in the Amazonian region. Mucosal leishmaniasis (ML), with involvement of the mucosae by contiguity from a primary lesion, has been described for all species causing CL in Peru.1 However, ML is distinct from MCL, which involves metastatic spread to mucosal sites some time after a primary infection. Control of MCL depends predominantly on passive or active case finding, diagnosis, and effective treatment.2 Dogs are commonly infected with L. braziliensis and/or L. peruviana, and they may act as a peridomestic reservoir.3,4 The sylvatic reservoir hosts of L. peruviana and L. braziliensis are incompletely known, although terrestrial small rodents have been implicated for some strains.4,5
Where the Amazonian forest and Andean regions meet, as in the Department of Huánuco, both L. braziliensis and L. peruviana may occur sympatrically. Leishmania (Viannia) isolates that appear to be hybrids between L. braziliensis and L. peruviana have been reported from this region of Peru.6
Herein we describe the phenotypes (obtained by multilocus enzyme electrophoresis, MLEE), and genotypes (obtained by microsatellite multilocus typing, MLMT) of 59 isolates from the Department of Huánuco. The results show that putative hybrid phenotypes and genotypes are common. A remarkable degree of diversity is revealed, suggesting that propagation is not entirely clonal and indicative of some form of genetic exchange in the Huánuco L. (V.) population.
| MATERIALS AND METHODS |
|
|
|---|
Isolates were phenotyped and genotyped against the following L. (V.) reference strains: L. (V.) braziliensis (MHOM/ BR/84/LTB300); L. (V.) peruviana (MHOM/PE/94/LC1152; MHOM/PE/84/LC26; MHOM/PE/84/LC39); L. (V.) panamensis (MHOM/PA/71/LS94); L. (V.) guyanensis (MHOM/ BR/75/M4147); L. (V.) shawi (MHOM/BR/94/M15065); L. (V.) lainsoni (MHOM/BR/81/M6426); and L. (V.) sp. n.7 (ISQU/BR/86/IM2832).
Isolation and in vitro cultivation of parasites. Reference strains were retrieved from liquid nitrogen storage onto bi-phasic 4N blood slopes.8 Peruvian stocks were dispatched from Lima on 4N blood slopes and were passaged onto fresh slopes and incubated at 23°C upon receipt. Isolates from this first-passage culture were stored under liquid nitrogen; subsequent passages were kept to a minimum to reduce culture selection of parasite strains. Promastigote cultures were expanded in alpha-modified minimal essential medium (Sigma-Aldrich Ltd., Gillingham, Dorset, UK) supplemented with 10% heat-inactivated fetal bovine serum, 50 µg/mL gentamicin, 30 mM NaHCO3, 40 mM HEPES, 20 mM D-glucose, 4 mM L-glutamine, 10 µM hemin, 30 µM adenine, 10 µM folic acid, and 10 µM D-biotin (all supplements from Sigma Chemical Co., Gillingham, Dorset, UK). Enzyme lysates were prepared from logarithmic-phase bulk cultures according to the method of Evans and others.8
Isoenzyme electrophoresis. Diversity of the Leishmania isolates was initially analyzed by MLEE using thin-layer starch gels.9 The enzymes applied were mannose phosphate isomerase (MPI, EC 5.3.1.8); glucose phosphate isomerase (GPI, EC 5.3.1.9); proline dipeptidase (PEPD, EC 3.4.13.9); phosphoglucomutase (PGM, EC 2.7.5.1); nucleoside hydrolase using 2 different substrates, inosine (NHi, EC 3.2.2.1; 2 loci: NHi1 and NHi2) and deoxyinosine (NHd, EC 3.2.2.x); 6-phosphogluconate dehydrogenase (6PGD, EC 1.1.1.44); glucose-6-phosphate dehydrogenase (G6PD, EC 1.1.1.49); esterase (ES, EC 3.1.1.1); aspartate aminotransferase (ASAT, EC 2.6.1.1); and alanine aminotransferase (ALAT, EC 2.6.1.2).
Microsatellite genotyping and DNA sequencing. Subsequent analyses involved multilocus microsatellite typing (MLMT)10 and DNA sequencing. Micosatellites AC01, AC16, and AC5211 were amplified from promastigote genomic DNA. The micosatellites were amplified with primers AC01F and AC01R-FAM (GAGAGGCCACCAGACACG-TCAGCACAC and CCCCCTTCCTTCGCCTTCAACAC-CTTTAC, respectively), AC16F and AC16R-TET (CTTCT-TCTCATGCTGCACGGTCTCCTCCTT and CCATGG-GCGGGCTTGTTTCGTTACTTTTTA, respectively), and AC52F and AC52R-HEX (CCACCGCCGGCTTCACTAC and GCGGCAATCGTCTGGCTAAA, respectively). Reverse primers were fluorophore-labeled as indicated (Perkin-Elmer, Beaconsfield, UK). Amplification reactions were carried out according to a protocol modified from that of Russell and others.11 PCR amplification for each sample was done in a 10 µL reaction mix containing 10xNH4 reaction buffer (160 mM (NH4)2SO4, 670 mM Tris-HCl, pH 8.8, 0.1% Tween-20 [Bioline, London, UK]), 1 mM (AC01, AC52) or 2 mM (AC16) MgCl2, 0.2 mM each of dATP, dCTP, dGTP, and dTTP (Pharmacia LKB, Upsala, Sweden), 5 pmol of each primer, 0.5% formamide (v/v), 1 U of Taq DNA polymerase (Bioline), and 25 ng of genomic DNA. PCR amplification was carried out in microtiter plates with sealed lids using the heated-lid option in an MJ Research PTC-200 Peltier thermocyler (Genetic Research Instrumentation Ltd., UK) using the following parameters: 35 cycles of 95°C for 30 s, 62°C (AC01 and AC52) or 60°C (AC16) for 30 s, 72°C for 1 min, followed by a final extension period of 10 min at 72°C. The multiplexed microsatellite products were sized on an ABI 377 automated sequencer by Genescan® and Genotyper® software (Applied Biosystems, Warrington, UK).
The AC01 products were sequenced by dye-terminator cycle sequencing and aligned using Sequence Navigator (Applied Biosystems).
Population genetics analysis. Resultant data from MLEE and MLMT were tested for genetic recombination and segregation using five population-genetics analyses: Hardy-Weinberg (HW) equilibrium,12 fixation index (Fis),12 D' index,13, R2 index,13 and index of association (IA)14
| RESULTS |
|
|
|---|
|
|
Five population-genetics analyses, Hardy-Weinberg (HW) equilibrium,12 fixation index (Fis),12 D' index,13, R2 index,13 and index of association (IA),14 were applied to the data for 58 isolates, including (due to apparent sharing of alleles) L. (V.) sp. n.7 but excluding L. lainsoni. The results were as follows:
0.01, Q test).13 Corresponding r2 values of linkage disequilibrium were lower, as is normal for this index.13 In summary, overall the results from these tests indicated that the Huánuco L. (Viannia) population diverges from clonality or linkage disequilibrium. Firstly, the partial lack of significant deviation from HW equilibrium and, secondly, the significant D' values imply that some form of genetic exchange has occurred among the Huánuco L. (Viannia) population.
| DISCUSSION |
|
|
|---|
Perceptions that T. brucei and Trypanosoma cruzi are entirely clonal have changed dramatically. With the aid of drug-resistance markers, genetic exchange was demonstrated to occur in the tsetse fly salivary glands, both within and between the T. brucei subspecies.16 The mechanism appears to be Mendelian, with occasional aneuploid progeny.17 Dependent on the locality under study, the population genetics of T. brucei ranges from panmixia in some undisturbed natural hosts and habitats through epidemic clonality, to true clonality.15 Among T. cruzi populations, phylogenetic analysis revealed genetic exchange18 and hybrids were produced experimentally in the laboratory.19 The T. cruzi experimental hybrids were derived from mammalian cells, not from the triatomine bug vector19 (although this does not exclude occurrence of genetic exchange within the vector). The mechanism of hybridization in T. cruzi, involving fusion, genome erosion, and recombination, was unusual but compatible with hybrid genotypes and the extensive range of DNA content found among natural populations. Heitman20 has drawn attention to striking parallels between T. cruzi and the fungus Candida albicans: cell fusion in C. albicans yields tetraploid progeny, which in appropriate growth conditions undergo random chromosome loss to revert to diploidy.21
Similarly, perceptions of Leishmania as entirely clonal have been questioned by reports of several instances of naturally occurring hybrid strains, especially from the New World. On the basis of phenotypic and genotypic markers, hybrids have been described between L. braziliensis and L. panamensis in Nicaragua9; between L. braziliensis and L. guyanensis in Venezuela22; between L. braziliensis and L. guyanensis in Ecuador23; and between L. braziliensis and L. peruviana in Peru.6 In the Old World, L. major/arabica hybrids were described.24 Putative parental and hybrid phenotypes of the L. donovani complex (L. donovani; "L. archibaldi") occur, sympatrically in East Africa, and sequencing of housekeeping genes encoding enzymes shows mosaic characters across such strains.25 Preliminary genetic analysis suggests that genetic exchange may occur among L. tropica populations in the Middle East.26 Most recently, genetic hybrids between L. infantum and L. major have been described, from immunocompromised patients in Portugal.27 Mixed Leishmania infections occur in natural hosts and in vectors, and infections in mammals may last for decades, providing ample opportunities for interactions between distinct genotypes. Overall, this is substantial circumstantial evidence that an extant mechanism of genetic exchange remains to be described for Leishmania, with epidemiologic implications, for example, in the spread of emergent strains or drug resistance.
L. peruviana was recorded in the Department of Huánuco prior to the epidemic from which the isolates in this study were derived. However, prior to this epidemic, CL was seldom encountered and no cases of MCL were recorded (Llanos-Cuentas, unpublished data). It is likely that introduction of L. braziliensis6 resulted in the increased prevalence of CL, and outbreak of MCL. L. braziliensis may have been introduced by human immigration from another region or by human or canine intrusion into an unidentified sylvatic transmission cycle.
The 59 Huánuco isolates analyzed here showed a remarkable degree of isoenzyme diversity considering that they originated from such a small geographical area. L. guyanensis and L. amazonensis, previously reported from Peru,1 were not identified. However, 4 speciesL. peruviana, L. braziliensis, L. lainsoni, and L. (V.) sp. n.7were found to occur sympatrically in Huánuco. In addition, L. braziliensis/L. peruviana phenotypic hybrids were common and almost as abundant as L. braziliensis. This local prevalence of hybrid strains is reminiscent of the predominance of the genetic hybrids of T. cruzi among human cases of Chagas disease in Paraguay and adjacent regions.28 The gene encoding the enzyme MPI has recently been sequenced from both L. braziliensis and L. peruviana, and phenotypic differences have been shown to correspond with a single nucleotide polymorphism (SNP), changing a threonine to an arginine, which has been used as the basis of a PCR identification assay.29 It would be of interest to confirm that the current L. braziliensis, L. peruviana, and L. braziliensis/L. peruviana hybrid isolates, and those from a wider geographical range, conform to the predicted SNP genotypes.
It was surprising to find 4 MLEE phenotypes and 7 microsatellite genotypes among the L. braziliensis/L. peruviana hybrids. This might be a consequence of the rapid evolution of microsatellites, or it could be consistent with the occurrence of more than one hybridization event. It is very unlikely that these genotypes are explicable by mutation, with no hybridization event. By analogy, multilocus sequence typing (MLST) of the L. donovani complex has revealed multiple heterozygous sites within a gene and at several loci, with sharing of alleles within and across genetic groups: recombination,25,30 rather than mutation,31 is considered to be the most parsimonious explanation.
The majority (12) of the L. braziliensis/L. peruviana hybrids had a single genotype (Table 1
). The expansive clonal propagation of one of the putative hybrid genotypes suggests that a hybrid agent has emerged with increased fitness relative to the parental strains, although a neutral event, such as a population bottleneck, unlikely in view of the diversity within the population, may also have been involved. One comparison of promastigote growth rates found no evidence that L. braziliensis/L. peruviana hybrids had an enhanced growth rate in vitro.32. Nevertheless, it would be of interest to compare the metastatic potential of the hybrid and nonhybrid genotypes in the hamster model of MCL33 or perhaps in the mouse ear model for dissemination.34 The occurrence of L. braziliensis/L. peruviana hybrids in dogs and humans (Tables 1
and 2
) from the same area indicates that both are exposed to the same infective sand fly population. This suggests that dogs might act as a reservoir host and enhance propagation of the emergent hybrid genotype.
Six zymodemes and at least 6 microsatellite genotypes were isolated from patients with MCL. Isolation of L. peruviana from a single case of MCL is an interesting finding but, in the absence of more cases, must be interpreted with caution. Previously in Peru, a single case of mucosal leishmaniasis (ML) was attributed to L. peruviana.1 Four of the MCL patients yielded a hybrid isolate, with no evidence of mixed infection, so patients carrying such isolates must be considered at risk of developing MCL. Thus, the finding of unequivocal genotypic markers for isolates that carry the risk of progression to MCL is still an elusive goal.35
As mentioned above, MLST applied to the L. donovani complex has already revealed the genetic basis of MLEE and provided a higher resolution approach to resolving genetic groups and to understanding relationships between them.25,30 MLMT has given even higher resolution of intraspecific population structure.10 Diversity of the subgenus L. (Viannia) is seen in some endemic foci,36 although the level observed in Huánuco is extraordinary. Genetic diversity has also been recorded from individual patients.37 Furthermore, it is apparent that there is some overlap of boundaries between the perceived species of the subgenus L. (Viannia). In combination, MLST and MLMT will be a powerful approach to understanding the complex molecular epidemiology and population genetics of L. (Viannia) and for further investigation of the extent of genetic exchange in natural or experimental populations. For the future MLMT of L. (Viannia), a much wider panel of microsatellite markers is required, and such a panel is in preparation (R. Oddone, G. Schönian, and K. Kuhls, personal communication and unpublished data). In localities where L. (Viannia) species overlap, we anticipate the discovery and possible emergence of other human infective hybrid genotypes, some with potential to generate severe disease.
Received May 10, 2006. Accepted for publication October 31, 2006.
Acknowledgments: The authors thank Rachel Gregory (LSHTM) for technical assistance, David Conway (LSHTM), Helen Roberts (University College, London), and Isabel Mauricio (LSHTM) for helpful discussions. Dr. R. Naiff (Instituto Nacional de Pesquisas Amazonas, Manaus, Amazonas, Brazil) and Prof. J. Shaw (Instituto Evandro Chagas, Belém, Pará, Brazil) generously provided some of the reference strains used in this study.
Financial Support: This work was supported by European Commission International Scientific Co-operation (grant no. C11-CT93-0036-NR), the Sir Halley Stewart Trust, and the Wellcome Trust.
* Address correspondence to Debbie Nolder, Department of Infectious & Tropical Diseases, London School of Hygiene & Tropical Medicine, Keppel Street, London WC1E 7HT, United Kingdom. E-mail: debbie.nolder{at}lshtm.ac.uk ![]()
Authors addresses: Debbie Nolder, Clive R. Davies, and Michael A. Miles: Department of Infectious & Tropical Diseases, London School of Hygiene & Tropical Medicine, Keppel Street, London WC1E 7HT, United Kingdom, Telephone: +44 (0)20 7927 2427, Fax: +44 (0)20 7637 0268, E-mails: debbie.nolder{at}lshtm.ac.uk, clive.davies{at}lshtm.ac.uk, and michael.miles{at}lshtm.ac.uk. Norma Roncal: Instituto de Medicina Tropical "Alexander von Humboldt", Universidad Peruana Cayetano Heredia, AP5045, Lima 100, Peru, E-mail: nroncal{at}upch.edu.pe. Alejandro Llanos-Cuentas: Facultad de Salud Pública y Administración, Carlos Vidal Layseca, Universidad Peruana Cayetano Heredia, AP5045, Lima 100, Peru, Telephone: +511 482 7739/381 4100, Fax: +511 382 0338, E-mail: allanos{at}upch.edu.pe.
Reprint requests: Debbie Nolder, Malaria Reference Laboratory, London School of Hygiene & Tropical Medicine, Keppel Street, London WC1E 7HT, United Kingdom.
| REFERENCES |
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |