J. Euk. M i c r o b i d . 44(5). 1997 pp 408-411 8 1997 by the Society of Protozoologists
Evidence for Hybridization by Multilocus Enzyme Electrophoresis and Random Amplified Polymorphic DNA Between Leishmania braziliensis and Leishmania panamensislguyanensis in Ecuador ANNE-LAURE BAfiULS,*.l FRANCOISE GUERRINI,* FRANCOIS LE PONT,** CESAR BARRERA,*** IVAN ESPINEL,*** RONALD GUDERIAN,***RAMIRO ECHEVERRIA**** and MICHEL TIBAYRENC* *Centre d'Etudes sur le Polymorphisme des Microorganismes (CEPM), Unite' Mixte de Recherche n o 9926 Centre National de la Recherche Scientifque (CNRS)/Institut Frangais de Recherche Scientifque pour le DPveloppement en Cooperation (ORSTOM), ORSTOM, BP 5045, 34032 Montpellier, Cedex I , France, and **Services Scientij'iques Centrarcx Institut Francais de Recherche Scientifque pour le Developpement en Coope'ration (ORSTOM), 70-74 Route d'Aulnay, 93140 Bondy, France, and ***Hospital Vozandes, Department of Clinical Investigations, CP 691 Quito, Ecuador, and ****Fundacion Eugenio Espejo, Quito, Ecuador ABSTRACT. The taxonomic attribution of four Leishmania stocks isolated from humans in Ecuador has been explored by both multilocus enzyme electrophoresis and random amplified polymorphic DNA. For three loci, MLEE results showed patterns suggesting a heterozygous state for a diploid organism, while the corresponding homozygous states are characteristic of the Leishmania panamensisl guyanensis complex and Leishmania braziliensis, respectively. Other enzyme loci showed characters attributable to either the L. panamensislguyanensis complex or L. braziliensis. RAPD profiles exhibited for several primers a combination of the Leishmania panamensislguyanensis complex and L. braziliensis characters. These data hence suggest that the four stocks are the result of hybridization between L. panamensislguyanensis and L. braziliensis. MLEE data show that the results cannot be attributed to either mixture of stocks, or an F1 in the framework of a simple Mendelian inheritance. Supplementary key words. Clonality, epidemiology, genetic exchange, taxonomy.
T
HE mating system of Leishmania in nature is presently under debate. By analyzing data from the literature in population genetic terms, Tibayrenc et al. [18, 191 have postulated that Leishmania sp. exhibit a basically clonal population structure, as do several other major parasitic protozoa. This model has been recently challenged in the case of Leishmania by Bastien, Blaineau and Pagbs [l], who observed apparent HardyWeinberg equilibrium in Leishmania infantum populations SUTveyed by pulse field gel electrophoresis (PFGE). The implications of these authors' assumptions have been further discussed by Tibayrenc [15, 161, and Pagks, Bastien and Blaineau [13]. In regard to this debate, the existence of hybrids between different species of Leishmania [2, 5-8, 101 is puzzling, since it does provide evidence that some kind of genetic exchange can occur in Leishmania natural populations. We present here additional evidence for hybridization in Leishmania, through a multilocus enzyme electrophoresis (MLEE) and random amplified polymorphic DNA (RAPD) analysis of four Ecuadorian Leishmania stocks, suspected to represent hybrids between the L. panamensislguyanensis complex [ 121 and L. braziliensis on the basis of limited MLEE evidence [Guemni, E 1993. GtnCtique des populations et phylogCnie des Leishmania du Nouveau Monde. Dissertation. University of Montpellier, France]. MATERIALS AND METHODS Origin of the isolates. The study involves four Leishmania stocks, namely: MHOM/EC/92/E102, MHOM/EC/92/Z5, MHOM/EC/92/Z9, MHOM/EC/92/P5, isolated from humans in the Zumba region in the south of Ecuador. The patients showed only cutaneous lesions. The three reference stocks were MHOMD30/84/LPZ595 (L. braziliensis from Bolivia), M H O W PN7 1 L S 9 4 (15.panamensis from Panama) and MHOIWBIU781 M5378 (L. guyanensis from Brazil). An eighth stock, MHOM/ PE/92/LH835, isolated from a patient in Peru, shows a natural mixture between a genotype attributed to the L. panamensisl guyanensis complex, and L. braziliensis. Isolation and culture. Isolation was performed by biopsy and aspiration. Bulk cultures were grown in RPMI medium with 10% fetal calf serum v/v. Parasites were harvested by centrif-
ugation (8,000 g for 15 min at 4"C) and washed twice in phosphate-buffered saline (PBS, pH 7.3: Na,HPO, (O.OlM), NaH,P04 (O.OlM), NaCl (0.15M)). Sample preparation and isoenzyme electrophoresis. Technical conditions for sample preparation, electrophoresis and staining procedures have been described in Ben Abderrazak et al. [3]. Cellulose acetate electrophoresis was used. A total of 15 enzyme systems were performed, namely: aconitase (ACON, EC 4.2.1.3), alanine aminotransferase (ALAT, EC 2.6.1.2), glucose-6-phosphate dehydrogenase (G6PD, EC 1.1.1.49), glucose phosphate isomerase (GPI, EC 5.3.1.9), glutamate oxaloacetate transaminase (GOT, EC 2.6.1. l), isocitrate dehydrogenase (IDH, EC 1.1.1.42), malate dehydrogenase NAD+ (MDH, EC 1.1.1.37), malate dehydrogenase NADP+ or malic enzyme (ME, EC 1.1.1.40), mannose phosphate isomerase (MPI, EC 5.3.1.8), nucleoside hydrolase, substrate deoxyinosine (NHd, EC 3.2.2.*), nucleoside hydrolase, substrate inosine (NHi: EC 3.2.2.*), peptidase 1, substrate 1-leucyl-leucine-leucine(PEP1, EC 3.4.11 or 13), peptidase 2, substrate I-leucyl-l-alanine (PEP2, EC 3.4.1 1 or 13), 6 phosphogluconate dehydrogenase (6PGD, EC 1.1.1.44), and phosphoglucomutase (PGM, EC 2.7.5.1). Random amplified polymorphic DNA. Random amplified polymorphic DNA procedures were performed according to Williams et al. [21]. Six ten-mer primers, which are fairly discriminative for Leishmania [20], were selected. Their sequences (CAGGCCCTTC; TGCCGAGCTG; AGGGGTCTTG; GAAACGGGTG; GTGATCGCAG; TTCCGAACCC) are identical to the 1, 4, 5 , 7, 10 and 15 primers of A Kit of Operon Technologies Inc., Alarneda, CA. RAPD polymorphism has been analyzed on ethidium bromide-stained agarose gels. Phylogenetic clustering of the isolates. Phylogenetic relationships among the isolates and the reference stocks were inferred from both MLEE and RAPD data, by computing Jaccard's distances [9], based on the following formula:
D,, = 1
-
[a/(a + b
+ c)]
Where a = number of bands that are common to the i a n d j stocks, b = number of bands present in the first genotype and absent in the second, and c = number of bands absent in the To whom correspondence should be addressed. Telephone: 33-4-67- first genotype and present in the second. Jaccard's distance 41-61-80; Fax: 33-4-67-41-62-99; Email:
[email protected] ranges from zero (the two genotypes exhibit exactly the same
408
BANULS ET AL.-LEISHMANIA
409
HYBRIDS IN ECUADOR
L pBnanUnsLf
L puyanensis
Fig. 1. Double dendrogram built from Jaccard’s genetic distances based on MLEE (right) and RAPD (left) for four putative hybrids between Leishmania braziliensis and the L. panamensislguyanensis complex, and for three reference stocks of L. panamensis, L. guyanensis and L. braziliensis.
profile for all systems surveyed) to one (no band is common between these two genotypes). The unweighted pair-group method with arithmetic averages (UPGMA) method [14] was used to compute dendrograms showing the genetic relationships among the putative hybrid stocks under study and the reference stocks (Fig. 1). RESULTS
Multilocus enzyme electrophoresis. The 15 enzyme systems made it possible to explore the variability of 16 putative enzyme loci, since two different loci could be inferred for the NHi system. These two loci were referred to as Nhi 1 and Nhi 2 (1 = the locus that has the fastest migration on gels). The variability recorded among the four isolates under study, E102, Z9, Z5, P5 made it possible to identify three distinct enzyme profiles, or zymodemes. These four stocks showed specific patterns at either two or three loci, namely: Nhi 2 and Pgrn for the four isolates, and Mpi for the isolate P5. For Nhi 2, the four stocks showed an identical, symmetrical five-banded pattern, typical of a heterozygous state for a tetrameric enzyme in a diploid organism (Fig. 2). A tetrameric structure has been previously postulated for this enzyme in the case of Leishmania infantum by another author (Moreno, G. 1989. Les complexes Leishmania donovani et Leishmania infantum. Implications taxinomiques, biogtographiques et CpidCmiologiques. A propos de l’analyse isoenzymatique de 548 souches de 1’Ancien et du Nouveau Monde. Dissertation. University of Montpellier, France). In the case of the Pgrn locus for the four putative hybrids, and in the case of the Mpi locus for the P5 stock only, two-banded patterns that are typical of a heterozygous state for a monomeric enzyme in a diploid organism were recorded. Interestingly, the two alleles involved correspond to alleles that are characteristic of the L. panamensid guyanensis complex on one hand, and L. braziliensis on the other hand (see Fig. 2). For the loci Alat, Idh, Acon, Pep, the putative hybrids showed homozygous one-banded patterns corresponding to either the L. panamensislguyanensis complex or to L. braziliensis. As an example, the four stocks showed an Alat 8 genotype that is specific of L. braziliensis, while they showed an Idh 8 genotype that is specific of the L. panamensid guyanensis complex. The four putative hybrids exhibited a homozygous L. braziziensis profile for the locus 6pgd. The MHOM/PE/92/LH835 stock (natural mixture between a genotype attributed to the L. panamensislguyanensis complex and L. braziliensis) revealed for each enzyme system a mere juxtaposition of the respective profiles of each species (see Fig. 2). The UPGMA dendrogram computed from MLEE Jaccard’s distances clustered the putative hibrid stocks slightly closer to
t ttttttttttt 1 2 3
4 5
6 7 8 9101112
Fig. 2. Cellulose acetate isoenzyme electrophoresis for the locus nucleotide hydrolase, substrate inosine (Nhi). From left to right: 1 , Leishmania infantum; 2 & 3 , putative hybrids between the L. panamensislguyanensis complex and L. braziliensis (stocks Z9 and Z5); 4, 5, 7, 9, L. braziliensis; 6 & 12, L. guyanensis; 8, natural mixture of L. braziliensis and L. guyanensis; 10, L. amazonensis; 11, L. forattinii.
the L. panamensisfguyanensis group than to the L. braziliensis reference stock (Fig. 1). Random amplification of polymorphic DNA. For the six primers under study, the four stocks all exhibited combinations of characters from the L. panumensislguyanensis complex and L. braziliensis (see Fig. 3). The MHOM/PE/92LH835 stock (natural mixture between a genotype attributed to the L. pana-
M
1
2
3
4
5
6
1
2
3
4
5
6
L ’ A5
A7
Fig. 3. RAPD profiles obtained with the A5 (left) and A7 (right) primers. 1: Leishmania panamensis; 2: L. braziliensis; 3-6: four putative hybrids between L. braziliensis and the L. panamensislguyanensis complex (from left to right: 29, Z5, El02 and P5). M. molecular mass ladder
410
J. EUK. MICROBIOL., VOL. 44, NO. 5, SEPTEMBER-OCTOBER 1997
mensislguyanensis complex and L. braziliensis) exhibited a similar result. Differently from the MLEE dendrogram, the RAPD dendrogram clustered the four putative hybrid stocks slightly closer to the L. braziliensis reference stock than to the L. panamensis and L. guyanensis reference stocks (see Fig. 1).
DISCUSSION Biological nature of the four Ecuadorian stocks. The most parsimonious hypothesis to account for the results obtained on the four Ecuadorian stocks is to equate them to hybrids between the L. panamensislguyanensis complex and L. braziliensis. Although stocks have not been cloned in the laboratory, these results cannot be explained by a genotype mixture. This last hypothesis could be considered for the RAPD data. Although the profiles of the putative hybrids do not correspond to a simple juxtaposition of the parental profiles (see Fig. 3), this could be explained by a differential amplification of different fragments in a mixed stock. Now, when MLEE data are considered, the Nhi 2 locus shows symmetrical five-banded patterns, while simple two-banded patterns would be expected in the case of mixtures such as the M H O M E 1 9 2 L H 8 3 5 stock (see Fig. 2). Indeed when two different species are mixed in the same isoenzyme extract, all the enzyme systems exhibit a juxtaposition of the patterns of each species. Although the Pgrn and Mpi loci show two-banded patterns that could be considered a result of stock mixtures, for the loci Alat, Zdh, Acon, Pep, the putative hybrids showed homozygous one-banded patterns that correspond to either the L. panamensislguyanensis complex or to L. braziliensis. It is more difficult to decide whether the stocks are the result of a hybridization between L. braziliensis and either L. guyanensis or L. panamensis. The 6pgd locus is considered the only locus able to distinguish between the two species L. guyanensis and L. panamensis [7]. Because the four putative hybrids exhibited a L. braziliensis profile for this locus, it is not possible to know whether the other putative parent is L. guyanensis or L. panamensis. The clearer indication on this question comes from the Nhi 2 locus. Indeed the L. panamensis reference stock shows a weak band for this locus, which suggests a low enzyme activity for this allele, while both the L. guyanensis and L. braziliensis reference stocks show a well-stained band. This last result has been verified on a broader range of reference stocks (S. S., unpublished data). If L. panamensis was one of the parents, one would expect therefore an asymmetric heterozygous pattern, with a weak top band for the four putative hybrids, while the contrary is observed here (see Fig. 2). This suggests that they are the result of a hybridization between L. braziliensis and L. guyanensis rather than L. panamensis. The results obtained by Belli et al. [2] on hybrids between L. braziliensis and L. panamensis in Nicaragua are consistent with this hypothesis. Indeed the heterozygous profiles recorded by these authors, instead of the five-banded patterns recorded in the present study, exhibited three-banded patterns for the Nhi 2 locus, which is expectable in the case of a heterozygous genotype for a tetrameric enzyme, with a functional allele (the one from L. braziliensis) and an allele with low enzymatic activity (the one from L. panamensis). An alternative explanation for the origin of these stocks deserves to be considered, and is probably worth considering for previously published examples of putative Leishmania hybrids [2, 5, 6-8, 101. The “hybrids” could represent ancestral states, while the “parents” would correspond to the descendants. This hypothesis is consistent with both MLEE and FUPD patterns, although the number of inferred evolutionary steps required makes this hypothesis less parsimonious than hybridization. Biological mechanisms at the origin of the hybrids. It is
impossible from these data to infer which precise biological processes produced these hybrids, If one infers a simple Mendelian inheritance in a diploid organism, these hybrids cannot be considered as a F1 product. Indeed only three enzyme loci, namely Nhi 2, Pgrn, and Mpi (for stock P5), show putative heterozygous patterns, while the other loci exhibit various combinations of putative parental characters. It is worth noting that diploidy in Leishmania 111, 171 is only a working hypothesis. The genome structure in Leishmania is still under debate, and other hypotheses apart diploidy can be proposed [4]. The results observed on the putative hybrids in the present study could be consistent with an aneuploid structure: the heterozygous patterns would be related to loci in a diploid state, while the patterns recorded at the other loci would reveal a haploid state. Due to the impossibility at present to perform mating experiments in Leishmania, these hypotheses cannot be conveniently tested. Hybrids between different Leishmania species have been already reported, both in the Old World [8] and in the New World [2, 5-7, 101. The present study is the first one, to our knowledge, to rely on the RAPD technique to confirm the occurrence of hybridization. The use of the RAPD technique will make it possible to explore more in depth the inheritance, in the hybrids, of the parental characters, by analyzing the sequences of the RAPD fragments apparently involved in the hybridization process. Inferences on population genetics of Leishmania. Hybridization phenomena seem to be relatively frequent in the genus Leishmania [2, 5, 6 - 8 , 10 and present results]. Nevertheless, Leishmania natural populations show typical indications of clonal structure, mainly strong linkage disequilibrium (nonrandom association among loci) [18, 191. The two facts are not inconsistent to each other. The high level of linkage disequilibrium recorded in Leishmania populations shows that recombination does not play the same role in them than in populations of sexual higher organisms, and is rare enough to permit the generation and propagation of discrete genetic lines. The hybrids show that sometimes, these discrete lines still are capable of some kind of genetic exchange. ACKNOmEDGMENTS This work was supported by a European Community International Scientific Cooperation grant no CI1*-CT9 1-0901. LITERATURE CITED 1. Bastien, €?, Blaineau, C. & Pages, M. 1992. Leishmania: sex, lies and karyotypes. Parasitol. Today, 8:174-177. 2. Belli, A. A., Miles, M. A. & Kelly, J. M. 1994. A putative Leishmania panamensislLeishmania braziliensis hybrid is a causative agent of human cutaneous leishrnaniasis in Nicaragua. Parasitology, 109:435442. 3. Ben Abderrazak, S., Guerrini, E, Mathieu-DaudC, E, Truc, P. Neubauer, K., Lewicka;K., BarnabC, C. & Tibayrenc, M. 1993. Isozyme electrophoresis for parasite characterization. In: Hyde, J. E. (ed.), Protocols in Molecular Parasitology, Humana Press, Totowa, NJ. Pp. 361382.
4. Cruz, A. K., Titus, R. & Beverley, S. M. 1993. Plasticity in chromosome number and testing of essential genes in Leishmania by targeting. Proc. Natl. Acad. Sci. USA, 90:1599-1603. 5 . Darce, M., Moran, J., Palacios, X., Belli, A. A., Gomez-Urcuyo, E, Zamora, D., Valle, S., Gantier, J. C . , Momen, H. & Grimaldi, G., Jr. 199 1. Etiology of human cutaneous leishmaniasis in Nicaragua. Trans. Royal SOC. Trop. Med. Hyg., 85158-59. 6. Dujardin, J. C., Baiiuls, A. L., Llanos-Cuentas, A., Alvarez, E., Dedonker, S., Jacquet, D., Le Ray, D., Arevalo, J. & Tibayrenc, M. 1995. Putative Leishmania hybrids in the Eastern Andean valley of Huanuco, Peru. Acfa Tropica, 59:293-307. 7. Evans, D. A,, Lanham, S. M., Baldwin, C. I. & Peters, W. 1984.
B m U L S E T AL.-LEISHMANIA The isolation and isoenzyme characterization of Leishmania braziliensis sp. from patients with cutaneous leishmaniasis acquired in Belize. Trans. Royal SOC.Trop. Med. Hyg., 78:35-42. 8 . Evans, D. A., Kennedy, W. P. K., Elbihari, S., Chapman, C. J., Smith, V. & Peters, W. 1987. Hybrid formation within the genus Leishmania? Parassitologia, 29: 165-173. 9. Jaccard, P. 1908. Nouvelles recherches sur la distribution florale. Bull. SOC.Vaud. Sci. Nut., 44:223-270. 10. Kreutzer, R. D., Yemma, J. J., Grogl, M., Tesh, R. B. & Martin, T. I. 1994. Evidence of sexual reproduction in the protozoan parasite Leishmania (Kinetoplastida: Trypanosomatidae). Am. J. Trop. Med. Hyg., 51~301-307. 11. Maazoun, R., Lanotte, G., Rioux, J. A., Pasteur, N., Killick-Kendrick, R. & Pratlong, E 1981. Signification du polymorphisme enzymatique chez les leishmanies. A propos de trois souches httkrozygotes de Leishmania infantum, Leishmania cf tarentolae et Leishmania aethiopica. Ann. Parasitol. Hum. Comp., 56:467-475. 12. Moreno, G., Pratlong, E, Velez, I. D., Restrepo, M. & Rioux, J. A. 1986. Individualisation du complexe Leishmania guyanensis. A propos de l'analyse num6rique de sept zymodkmes. Leishmania. Taxonornie et PhylogCnbse. Applications Eco-CpidCmiologiques (Colloques Int. Centre National de la Recherche Scientifiquehstitut National de la Sant6 et de la Recherche Mtdicale, 1984):165-172. 13. Pages, M., Bastien, F? & Blaineau, C. 1992. Leishmania: sex, karyotypes and population genetics: reply. Parasitol. Today, 8:306. 14. Sneath, P. H. A. and Sokal, R. R. 1973. Numerical Taxonomy.
HYBRIDS IN ECUADOR
411
The Principle and Practice of Numerical Classification. Freeman, San Francisco. 15. Tibayrenc, M. 1992. Leishmania: sex, karyotypes and population genetics. Parasitol. Today, 8:305-306. 16. Tibayrenc, M. 1993. Clonality in Leishmania. Parasitol. Today, 9:58 17. Tibayrenc, M., Cariou, M. L. & Solignac, M. 1981. InterprCtation gCnCtique des zyrnogrammes de flagell6s des genres Trypanosoma et Leishmania. C. R. Acad. Sci. Paris, 292:623-625. 18. Tibayrenc, M., Kjellberg, E & Ayala, E J. 1990. A clonal theory of parasitic protozoa: the population structure of Entamoeba, Giardia, Leishmania, Naegleria, Plasmodium, Trichomonas and Trypanosoma, and its medical and taxonomical consequences. Proc. Natl. Acad. Sci. USA, 87:24 14-24 18. 19. Tibayrenc, M., Kjellberg, E, Amaud, I., Oury, B., Brenibre, S. E, Dard6, M. L. & Ayala, E .l. 1991. Are eucaryotic microorganisms clonal or sexual? A population genetics vantage. Proc. Natl. Acad. Sci. USA, 885129-5133. 20. Tibayrenc, M., Neubauer, K., Barnabt, C., Guenini, F., Sarkeski, D. & Ayala, E J. 1993. Genetic characterization of six parasitic protozoa: parity of random-primer DNA typing and multilocus isoenzyme electrophoresis. Proc. Natl. Acad. Sci. USA, 90: 1335-1339. 21. Williams, J. G. K., Kubelik, A. R., Livak, K. J., Rafalski, J. A. & Tingey, S. V. 1990. DNA polymorphism amplified by arbitrary primers are useful as genetic markers. Nucl. Acids Res., 18:653 1-6535. Received 8-21-95, 4-16-96, 12-3-96, 4-30-97; accepted 4-30-97
