Genetic Relationships Among Phlebotomine Sand Flies (Diptera: Psychodidae) in the verrucarum Species Group R. D. KREUTZER,1 M. T. PALAU,2 A. MORALES,2 C. FERRO,2 D. FELICIANGELI,3 AND D. G. YOUNG4

KEY WORDS

Insecta, Lutzomyia, electrophoresis, phylogeny

THE Lutzomyia SAND FLIES in the verrucarum species group (Theodor 1965, Martins et al. 1978) occur in many parts of the neotropics and are especially abundant in mountainous areas where females of some species are highly anthropophilic. More important, several species have been incriminated as vectors of Leishmania and Bartonella baciliformis to humans. There are several species complexes (series) within the verrucarum group, two of which have been defined previously—the series verrucarum and serrana (Theodor 1965). We recognize a third complex, the series townsendi, that includes L. townsendi (Ortiz) and its close relatives (Table 1). The males in this series, unlike those in other series, are characterized by an isolated basal spine and three distal spines on the style of the terminalia. The females are remarkably similar in structure, and it

is difficult or impossible (isomorphic) to identify specimens in the absence of conspecific males, Epidemiological and population studies of sympatric species in the townsendi series would be facilitated if these taxa could be separated. Accordingly, it is necessary to develop alternative methods to distinguish them. Electrophoretic data of sand flies from the Old and New World have been used to identify cryptic species and to study intraspecific variation. For example, the species pair Lutzomyia carrerai (Barretto) and L. yucumensis LePont, Caillard, Tibayrenc, and Desjeux in Bolivia (Caillard et al. 1986) and the pair L. wellcomei (Fraiha, Shaw, and Lainson) and L. complexa (Mangabeira), in Brazil (Ready & Da Silva 1984) represent cryptic species that can be distinguished by such methods. In addition, a study of nine phlebotomine species from France and Tunisia indicated that isozyme analysis might be useful for separating species among which one sex is • Biology Department, Youngstown State University, Youngs- Structurally identical (Ward et al. 1981). Allomorph frequency differences were reported among town, Ohio 44555. 2 Instituto Nacional de Salud, A.A. 80080, Bogota, Colombia. allopatric populations of PhlebotomuS perfiliewi ^FacultaddeCienciasdelaSalud.UniversidaddeCarabobo, perjMewi Corradetti (Ward et al. 1981) and L.

Maracay, Venezuela,

r

J

.

/r,i

\

/r>

.

,

n m

v

T

.%

' Department of Entomology and Nematology, University of panamensis (Shannon) (Peterson 1982). In other Florida, Gainesville, Fla. 32611. studies, no correlation could be found among en0022-2585/90/0001-0008$02.00/0 © 1990 Entomological Society of America

Downloaded from http://jme.oxfordjournals.org/ by guest on June 9, 2016

J. Med. Entomol. 27(1): 1-8 (1990) ABSTRACT Isozyme data were used to identify populations of certain Lutzomyia taxa in the verrucarum species group, mostly in the series townsendi. Lutzomyia youngi Feliciangelis and Murillo and L. spinicrassa Morales, Osorno, Osorno, and Hoyos each have diagnostic allomorphs for phosphogluconate dehydrogenase (6PGDH) and fumarate hydratase (FUM). The 6PGDH and FUM data and those from 6-phospho-fructokinase and phosphoglucomutase distinguish Lutzomyia sp., a new species from Colombia to be described and named later. Data from these enzymes and glucose phosphate isomerase will separate L. townsendi (Ortiz) from the others and from L. longiflocosa Osorno, Morales, Osorno, and Hoyos, L. quasitownsendi Morales, Osorno, Osorno, and Hoyos, and L. sauroida Osorno, Morales, and Osorno (three species that are inseparable using enzyme data). Three other species, L. serrana (Damasceno and Arouck) (series serrana) and L. columbiana (Ristorcelli and Van Ty) and L. andina Osorno, Osorno, and Morales (both in series verrucarum), are morphologically distinct using conventional characters and have fixed diagnostic differences at several enzyme loci. Statistical analyses of the enzyme data using genetic identities (I), differences (D), and the amount of genetic variation among these taxa indicated that such statistics can be as useful in the study of sandflyphylogeny and population genetics as they have been for other organisms. I and D values indicated that L. longiflocosa, L. quasitownsendi, and L. sauroida are very similar (I = 0.991 and D = 0.010) and possibly are populations of the same conspecific species. The levels of divergence, based on combined enzyme data for up to 21 gene loci among the taxa, are discussed, and a dendrogram based on genetic distance is presented. The genetic data confirmed established phylogenetic relationships among the sand fly taxa based on structural similarities.

Vol. 27, no. 1

JOURNAL OF MEDICAL ENTOMOLOGY Table 1.

Lutzomyia taxa and population data

Taxaa Series serana L. serrana (Damasceno and Arouck) Series verrucarum L. columbiana (Ristorcelli and Van Ty) L. andina Osorno, Osorno, and Morales Series townsendic L. spinicrassa Morales, Osorno, Osorno, and Hoyos L. sp. A

L. townsendi (Ortiz) L. longiflocosa Osorno, Morales, Osorno, and Hoyos L. quasitownsendi Morales, Osorno, Osorno, and Hoyos L. sauroida Osorno, Morales, and Osorno

Collection method

No. and sex

Arboledas (NS); cool, shaded hillsides, few large trees, dense understory of coffee plants; 1°37'N, 72°55'W

Human bait and direct aspiration from resting sites such as tree buttresses

186 99, 66

Samaniego (NA); mountainous areas with large trees; 1°14'N, 77°35'W El Chariquito (CU); type locality; rocky outcrops, semi-shaded by eucalyptus trees; 4°50'N, 74°20'W

Direct aspiration from sites such as tree buttresses Direct aspiration from sites such as rock crevasses

105 29, 66

Arboledas (NS); same locality as L. serrana

Direct aspiration from resting sites such as tree buttresses

126 66

Reventones (CU); type locality; cool, shaded, rolling hills with many large trees and understory of coffee plants; 4°45'N, 74°10'W Las Calderas (TR), nearby Los Curos (ME); type locality; cool, shaded hillsides; coffee plantation; 9°22'N, 70°22'W Rancho Grande (AR); type locality; mountainous, evergreen cloud forest; 10°2rN, 67°40'W Tello (HU); type locality; hilly with large trees; 3°5'N, 75°11'W

Human bait and direct aspiration from resting sites such as tree buttresses

186 66

Los Calderas, direct aspiration from resting sites such as tree holes; Los Curos, Shannon traps

126 66

Shannon traps

126 66

Human bait and direct aspiration from resting sites such as tree holes Direct aspiration from resting sites such as tree holes and buttresses

126 66

Barbosa (SA); type locality; rolling hills with large trees; 3°5'N, 75°11'W Santa Ana (BO); type locality; hilly with few large trees, dense understory; 5°40'N, 74°20'W

Direct aspiration from resting sites such as tree holes

105 99, 66

126 66 126 66

a

Voucher specimens of all taxa are stored in the cryobank at Youngstown State University, Youngstown, Ohio. State: Colombia: BO, Boyaca; CU, Cundinamarca; HU, Huila; NA, Narino; NS, Norte de Santander; SA, Santander. Venezuela: AR, Aragua; ME, Merida; TR, Trujillo. c Adult females of these species are structurally indistinguishable (isomorphic). b

zyme polymorphism and size heterogeneity in populations of L. longipalpis (Lutz and Neiva) from Bolivia (Bonnefoy et al. 1986), in two populations of L. flaviscutellata (Mangabeira) (which differ in their adaptability to laboratory conditions [Miles & Ward 1978]), or between canopy and ground level populations of L. umbratilis Ward and Fraiha from French Guiana (Tibayrenc et al. 1980). Studies dealing with Drosophila spp. have shown that electrophoretic data from about 20 gene loci can be combined to calculate reliable statistical values of genetic identity (I) and distance (D) (Avise 1974, Ayala 1975). This information can be used to indicate the levels of evolutionary divergence between and among populations and taxa. The D value of the cryptic species L. carterai and L. yucumensis was 0.256, which suggested they were closely related species and not phenotypic variants of the same species (Caillard et al. 1986). In this paper, we expand these studies to include populations of 10 Lutzomyia taxa in the verrucarum group to determine if electrophoretic data might be useful for distinguishing isomorphic females and the degree of genetic divergence among

populations and taxa as indicated by the statistics I and D. Materials and Methods Collection data on the sand fly populations studied are included in Table 1. Except for about 25 L. youngi Feliciangelis and Murillo that were reared to adults from egg batches, specimens were field collected, immediately placed in liquid nitrogen containers, transported to the laboratory, identified, and stored in liquid nitrogen until needed for electrophoresis. The cellulose acetate electrophoresis procedures used to obtain the enzyme data from sand flies are similar to those first reported for mosquitoes (Kreutzer & Galindo 1980) and later adapted for protozoan parasites (Kreutzer & Christensen 1980, Kreutzer & Sousa 1980, Kreutzer et al. 1983). All equipment was manufactured by Helena Laboratories, Beaumont, Tex. The enzymes studied are listed in Table 2. Enzyme data were obtained from individual, identified adult males or females (or both) of L. serrana,

Downloaded from http://jme.oxfordjournals.org/ by guest on June 9, 2016

L. youngi Feliciangelis and Murillo

Locality''

January 1990

KREUTZER ET AL.: GENETIC RELATIONSHIPS AMONG SAND FLIES

Results Taxon Identification by Genetic Data. Specimens of L. spinierassa Morales, Osorno, Osorno, and Hoyos, L. youngi, L. townsendi, and L. sp. A were distinguished by the enzymes 6PGDH, PFK, ALD, PGM, FUM, GPI (Table 3). The L. longiflocosa Osorno, Morales, Osorno, and Hoyos, L. quasitownsendi Morales, Osorno, Osorno, and Hoyos and L. sauroida Osorno, Morales, and Osorno could not be separated from each other reliably using the enzyme data included in this study; however, they were separated as a trio from the other verrucarum group species. 6PGDH and FUM separated L. youngi and L. spinicrassa, because each species has a different fixed (diagnostic) allomorph for these enzymes. Data from PFK and PGM (added to those from 6PGDH and FUM, which excluded L. youngi and L. spinicrassa) separated L. sp. A. The combined data from these four enzymes plus GPI separated L. townsendi from the species trio with a high level of confidence. The other species

Table 2. tained

Enzymes from which genetic data were ob-

Enzyme Oxidoreductases Glycerol-3-phosphate dehydrogenase Malate dehydrogenase Malic enzyme Isocitrate dehydrogenase Phosphogluconate dehydrogenase Xanthine dehydrogenase Glucose-6-phosphate dehydrogenase Glyceraldehyde-phosphate dehydrogenase Glutathione reductase Transferases Glutamate-oxaloacetate transaminase Hexokinase 6-Phosphofructokinase Adenlate kinase Phosphoglucomutase Hydrolases Acid phosphatase

Enzyme Enzyme commission number0 abbreviation 1.1.1.8

GDP

1.1.1.37 1.1.1.40 1.1.1.42 1.1.1.44 1.2.1.37 1.2.1.49

MDH ME ICD 6PGDH XDH G6PDH

1.2.1.12

GAPDH

1.6.4.2

GSR2

2.6.1.1.

5.4.2.2

GOTi and GOT 2 HK PFK AKj and AK2 PGM

2.7.1.1 2.7.1.11 2.7.4.3

3.1.3.2

ACP

Lyases Aldolase Fumarate hydratase

4.1.2.13 4.2.1.2

ALD FUM

Isomerases Mannose phosphate isomerase Glucose phosphate isomerase

5.3.1.8 5.3.1.9

MPI GPI

a

Commission of Biochemical Nomenclature (1973).

are distinct morphologically and have fixed genetic differences at several enzyme loci. Thus, enzyme data in conjunction with established morphological characters can be used to identify most of these sand fly species. I and D Values. The I and D data for the taxonomically similar species L. longiflocosa, L. quasitownsendi, and L. sauroida (Table 5, I = 0.991 and D = 0.010) indicated that they are genetically similar. The average values for all comparisons of isomorphic female pairs in the series townsendi (minus the above species trio) are I = 0.831 and D = 0.188. In Table 5, these data have been separated into two groups, 0.95 > I < 0.80 and 0.80 > I < 0.60. The average statistics for comparing structurally distinct species were I = 0.412 and D = 0.932. These also were divided into two categories, I = 0.60 > I < 0.50 and I < 0.50. The results of the statistical analysis of the genetic data in most cases confirmed established taxonomic relationships. The I and D values shown in Tables 4 and 5 for the species trio L. longiflocosa, L. quasitownsendi and L. sauroida indicate that they are similar (I = 0.991 and D = 0.010). Equivalent high levels of identity were reported among local populations of the horn fly Haematobia irritans (L.) (I = 0.989) (McDonald et al. 1987),

Downloaded from http://jme.oxfordjournals.org/ by guest on June 9, 2016

L. columbiana (Restorcelli and Van Ty), and L. andina Osorno, Osorno, and Morales, but only identified males were used for the other species. Allomorph (alleles observed by electrophoresis) frequencies for GPI and PGM enzymes were calculated separately from groups of 100, 75, 50, and 20 individually tested L. serrana and L. sp. A. The frequencies calculated from each group were not significantly different when compared by x2 tests (F > 0.05); therefore, the allomorph frequencies used to obtain I and D values were each calculated from samples of at least 20 sand flies (Tables 3 and 4). To confirm that the samples were randomly selected, genetic equilibrium tests were made for all polymorphic loci (most common allomorph 0.05) (genetic data from multiple enzymes can be obtained from individual sand flies). Each population was tested for all the enzymes listed in Table 2. Some of the enzymes (ACP, GAPDH, MPI, PFK) require large or undissected specimens to produce clear bands, but only small and dissected males that fail to produce clear bands for these enzymes could be studied. Therefore, the number of enzymes used to calculate I and D values varies for pair comparisons and is noted in parentheses after each I value in Table 4. The average number of all pair comparisons was 18.4 enzymes or gene loci (range, 16-21). I and D were calculated by the method proposed by Nei (1972). Expected heterozygosity (H) was calculated as described by Nei (1978). The mean proportion of polymorphic loci (P) per population was calculated and includes as polymorphic those enzymes when the most common allomorph had a frequency of I < 0.50 and I < 0.50) which approximate the average values reported for sibling (I = 0.563) and nonsibling species (I = 0.352) in Drosophila (Ayala 1975). The data for paired comparisons, which included L. columbiana and each of the other nine taxa (I = 0.549 and D = 0.605), indicated that L. columbiana could be at the sibling species level of divergence compared with the other taxa. The comparisons of L. serrana or L. andina with species of the series townsendi indicated the two are not sibling species and are genetically distant (I = 0.339 and D = 1.093). The amount of genetic variation was measured by variation per population of sand flies (P) and by average expected and observed heterozygosity per individual (H). When the most common allomorph was

Genetic relationships among phlebotomine sand flies (Diptera: Psychodidae) in the verrucarum species group.

Isozyme data were used to identify populations of certain Lutzomyia taxa in the verrucarum species group, mostly in the series townsendi. Lutzomyia yo...
749KB Sizes 0 Downloads 0 Views