Regional genetic variation in the major sperm protein genes ofOnchocerca volvulus and Mansonella ozzardi (Nematoda: Filarioidea)q

Ramiro Morales Hojasa,b,*, Rory J. Posta

aDepartment of Entomology, The Natural History Museum, Cromwell Road, London SW7 5BD, UKbPest Management Department, Natural Resources Institute, University of Greenwich, Chatham Maritime, Kent ME4 4TB, UK

Received 2 August 2000; received in revised form 1 September 2000; accepted 1 September 2000

Abstract

Onchocerca volvulus and Mansonella ozzardi are two human ®larial parasites present in South and Central America. In the Brazilian

Amazonia they are found in sympatry, and the lack of clear morphological diagnostic characters in the micro®lariae hinders their identi®ca-

tion. The major sperm protein (MSP) gene of both species has been sequenced and characterised to determine its potential as a molecular

diagnostic character. The length of the MSP gene is different in each species, and this could be used to detect and differentiate them by

running the polymerase chain reaction (PCR) product in an agarose gel. Two major gene groups were identi®ed in O. volvulus with a genetic

distance of 6% between them. In M. ozzardi only one major group of genes was observed. The high similarity between the protein amino acid

sequence of both ®larial species con®rms that the MSP has been highly conserved through nematode evolution. q 2000 Australian Society

for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved.

Keywords: Filarial parasites; Onchocerciasis; Major sperm protein genes; Species identi®cation; Major sperm protein evolution

1. Introduction

Onchocerciasis or river blindness disease is a major

problem in West Africa affecting a large proportion of the

human population. It is caused by the ®larial nematode

Onchocerca volvulus which is transmitted by ¯ies of the

Simulium damnsoum complex [1]. This nematode is also

present in South and Central America where it is also trans-

mitted by simuliid ¯ies [2,3]. Here the symptoms of infec-

tion are generally milder, blindness being almost absent.

Two theories have been postulated to explain the presence

of O. volvulus in America. One suggests that it was imported

with the African slaves taken to America by the Europeans

within the last 500 years [4]. The other proposes the auto-

chthonous origin of the parasite in America [5]. Recent

work based on the phylogenetic analysis of a repeated

DNA sequence (O±150 satellite) gives support to the ®rst

hypothesis [6]. Also, analysis of the nuclear rDNA internal

transcribed spacer (ITS) region of samples from Brazil and

Cameroon support a recent separation (unpublished data).

Mansonella ozzardi is a ®larial nematode parasite of

humans, present in South and Central America and which

is generally considered to be symptomless [7]. It is trans-

mitted by Culicoides in the Caribbean region and parts of

South America [8,9], and simuliids in the Amazonia [10]. It

has been hypothesised that these populations could be two

different species [11], but morphological analysis of micro-

®lariae from Colombia (simuliid transmitted) and Haiti

(culicoid transmitted) showed no taxonomic differences

[12]. Furthermore, analysis of the rDNA ITS2 of samples

transmitted by different vector types, from North Argentina

(where culicoids are the most ef®cient vectors, see Ref. [9])

and from Brazilian Amazonia (simuliid transmitted),

supports the single species hypothesis (unpublished data).

The main importance of M. ozzardi is the morphological

similarity between its micro®lariae and those of O. volvulus,

which is of greatest relevance in those areas of the Brazilian

Amazonia onchocerciasis focus where both nematodes have

been found in sympatry, and where they are also transmitted

by the same vector species, Simulium oyapockense [2,10].

Speci®c identi®cation of ®larial nematodes in hosts and

vectors is crucial for epidemiological studies (including the

monitoring of control, such as that introduced by the WHO

Onchocerciasis Elimination Programme for the Americas,

based on Ivermectin) as well as for systematic and evolution-

ary studies. Since micro®lariae of O. volvulus are skin-dwell-

International Journal for Parasitology 30 (2000) 1459±1465

0020-7519/00/$20.00 q 2000 Australian Society for Parasitology Inc. Published by Elsevier Science Ltd. All rights reserved.

PII: S0020-7519(00)00117-X

www.parasitology-online.com

q Note: Nucleotide sequence data reported in this paper are available in

the EMBL, GenBank TM and DDJB data bases under the accession

numbers AJ404204±AJ404225.

* Corresponding author. Tel.: 144-20-7942-5595; fax: 144-20-7942-

5229.

E-mail address: r.morales-hojas@nhm.ac.uk (R. Morales Hojas).

ing and micro®lariae of M. ozzardi are blood dwelling, infec-

tions have in the past been detected by microscopic analysis

of samples of skin or blood, respectively. However, M.

ozzardi micro®lariae have been found in the skin [13,14]

and O. volvulus micro®lariae in the blood [15,16] making

diagnosis more dif®cult because of the lack of morphological

characters. Immunological and DNA methods have been

developed to detect O. volvulus [17], however, they are

time consuming and give no information about the presence

or absence of M. ozzardi. Immunological methods are based

on the detection of O. volvulus triggered antibodies using a

cocktail of recombinant antigens [18], however, it is not

known for how long the these antibodies remain in circula-

tion after the infection ends, and the cross-reactivity to M.

ozzardi antibodies has not been thoroughly tested. The DNA

method for the detection of O. volvulus is based on the ampli-

®cation of the repeated satellite family O±150 by polymerase

chain reaction (PCR) and its subsequent detection by South-

ern blot with a speci®c probe [19,20].

The major sperm protein (MSP) is a ®ne ®bre protein

speci®c to nematode sperm. The sperm of nematodes is

uniquely amoeboid and MSP plays an important role in

cell motility [21±23]. It is usually encoded by a multigene

family which varies in copy number from one to four in

some parasitic nematodes and more than 50 in Caenorhab-

ditis [24,25]. The MSP has been highly conserved during

evolution, and it is possible to detect MSP genes in several

parasitic and free-living nematodes using MSP cDNA from

Ascaris and Caenorhabditis as probes [24]. The nucleotide

coding sequence homology between Ascaris and Caenor-

habditis is 72%, and between O. volvulus and these two

nematodes is approximately 83 and 79%, respectively, and

the protein is 90% homologous in sequence between Ascaris

and O. volvulus, 82% between Caenorhabditis and O. volvu-

lus, and over 82% between Caenorhabditis and Ascaris

[26,27]. The MSP genes are potentially useful sequences

for the identi®cation of nematode species because they are

short and there are several copies in each genome so ampli-

®cation can be performed easily, and because they are orga-

nised into two conserved exons with a single variable intron,

in all species studied, except Caenorhabditis [28]. These

features make the MSP genes, like the rDNA, a potentially

useful sequence for identi®cation of nematodes at different

taxonomic levels [24,29].

In this study we compare the O. volvulus MSP gene

sequences from Liberia published by Scott et al. [26] with

new MSP gene sequence data from Brazil. We also present

the MSP gene sequence for the ®larial nematode M. ozzardi

and compare it with that of O. volvulus.

2. Materials and methods

2.1. Parasite material

A single O. volvulus nodule from a Yanomami Indian

from the Brazilian Amazonia onchocerciasis focus, was

supplied by Dr Maia-Herzog (FIOCruz, Brazil), and a

pool of M. ozzardi micro®lariae in isopropanol from a heav-

ily infected person from the Jujuy province in North Argen-

tina was obtained from Dr CoscaroÂn (Museo de La Plata,

Argentina). A piece of O. volvulus nodule was shredded

with a scalpel, and 500 ml of the M. ozzardi micro®lariae

pool were sedimented by centrifugation and the supernatant

removed, prior to DNA extraction by a standard sodium-

dodecylsulfate (SDS)/proteinase K treatment (brie¯y, 500

ml of ethylene-diamine tetra acetic acid (EDTA), 5 ml of 14

mg/ml proteinase K and 5 ml of 10% SDS were added to the

sample and this was incubated at 568C for 1 h and then at

1008C for 30 min), followed by ethanol precipitation. The

DNA pellet was resuspended in 25 ml of sterile water.

2.2. MSP genes ampli®cation and sequencing

MSP genes of O. volvulus and M. ozzardi were ampli®ed

using two primers designed from the conserved exons of the

O. volvulus gene sequences from Liberia (OVGS-1 and

OVGS-2, GenBank accession numbers J04662 and

J04663, respectively), [26]. The forward primer, MSPEx1

5 0-ATGGCGCAATCGGTTCCACC-3 0, is located in the

®rst 20 bases (position of alignment 1±20). The reverse

primer, MSPEx2 5 0-CTTAAGATTTTTGCGACGAAC-

CAT-3 0, is located in the position 501±524 of the aligned

OVGS-1 sequence, 15 bases upstream of the stop codon.

The reaction consisted of a total volume of 25 ml containing

1 £ buffer (Promega), 2 mM MgCl2, 60 mM of each dNTP,

0.2 mM of each primer and 0.5 U of Taq polymerase

(Promega). As template, 1 ml of the extracted genomic

DNA was added. For the negative control, sterile water

was added instead of DNA. The cycle conditions included

an initial denaturarion at 948C for 3 min, and this was

followed by 35 cycles of 948C for 30 s (denaturation), 55

or 538C for 45 s (annealing), 728C for 30 s (extension), and a

®nal extension time of 10 min at 728C. The optimum anneal-

ing temperature was 558C for O. volvulus and 538C for M.

ozzardi. PCR products were run in a 1% agarose gel in 1 £TBE buffer. Gels were stained with ethidium bromide and

visualised on a UV light transilluminator. The size of the

PCR products was estimated by using a DNA marker (1 kb

plus, Gibco Life Technologies). DNA bands were cut out of

the gel and the PCR products were puri®ed using a gene-

clean kit (Anachem).

PCR products were cloned into a pCRw2.1-TOPO vector

and transformed into TOP10 competent cells using the

TOPO TA cloning kit (Invitrogen) following the manufac-

turer's instructions. Recombinant plasmids were recovered

using Hybaid's plasmid midi prep recovery kit. Five clones

of the O. volvulus nodule from Brazil and 17 clones of the

pool of M. ozzardi micro®lariae were sequenced to comple-

tion in both directions using Big Dye (ABI) chemistry in a

Techne thermocycler. The sequencing cycle consisted of an

R. Morales Hojas, R.J. Post / International Journal for Parasitology 30 (2000) 1459±14651460

initial denaturation at 948C for 2 min followed by 35 cycles

of 958C for 15 s, 508C for 15 s, and 608C for 4 min.

2.3. Sequence analyses

MSP gene sequences were aligned using CLUSTAL W

1.7 (Higgins D.G., Thomson J.D., Gibson T.J., CLUSTAL

W Multiple Sequence Alignment Program, version 1.7.

1997. www address: http://www.sgi.com/chembio/

resources/clustalw/) and corrected by eye. The limits of

the intron were determined by comparison with the Liberian

O. volvulus MSP gene sequences OVGS-1 and OVGS-2

(GenBank accession numbers J04662 and J04663) [26].

The primer sequences were removed from the analyses

because they would be identical in all ampli®cations. P-

distances (% differences) were calculated with the program

molecular evolutionary genetics analysis (MEGA) (Kumar

S., Tamura K., Nei M. MEGA, version 1.02. The Pennsyl-

vania State University, University Park, PA 16802, 1993)

for each pair of sequences.

To study the genetic similarity between the different

sequences within species a phenetic tree was constructed

for each species using the unweighted pair group method

with arithmetic averages (UPGMA) in PAUP 4.0b4a (Swof-

ford D.L, 2000. PAUP*. Phylogenetic analysis using parsi-

mony (and other methods). Version 4. Sinauer Associates,

Sunderland, MA). Pairwise comparison of the sequences

was done using the Kimura 2-parameter model with transi-

tion/transversion ratio set to 2:1.

In order to establish phylogenetic relationships within

species, trees were constructed using neighbour joining

(NJ) and maximum parsimony (MP). Both types of trees

were constructed using PAUP 4.0b4a. Alignment gaps

were treated as missing data. For the NJ method pairwise

distances were determined using the Kimura 2-parameter

model with the transition/transversion ratio set to 2:1. Boot-

strap analysis was performed using 100 replicates to test the

support for each branch in the tree. For the MP method

heuristic (with tree bisection-reconnection as the branch-

swapping algorithm) and branch-and-bound searches were

conducted. All characters had the same weight and were

treated as unordered. Bootstrap analysis with 100 replicates

was used to test the support for the branches.

Translation of the partial nucleotide sequences obtained

for exon 1 and 2 into amino acid sequences and comparison

of the partial protein sequences within and between species

was done with the program MEGA.

3. Results

Ampli®cation with primers MSPEx1 and MSPEx2 gave a

single product of different size for each species which could

be distinguished in 1% agarose gels (Fig. 1). The sequenced

fragment of the MSP gene is 555 bp in M. ozzardi and 477±

478 bp in O. volvulus (not including the primers) (Fig. 2).

All sequences were deposited in the nucleotide sequence

data base (GenBank TM, EMBL) and have the following

accession numbers: AJ404204±AJ404208 for the O. volvu-

lus sequences, and AJ404209±AJ404225 for the M. ozzardi

sequences. All the different MSP gene sequences have been

given names following the Filarial Uni®ed Nomenclature

Kommittee (FUNK) recommendations [30] (Ov-msp-3 to -

7 for the Brazilian O. volvulus MSP gene sequences, and

Mo-msp-1 to -17 for the Argentinean M. ozzardi MSP gene

sequences). Limits of the intron in both species were located

by comparison with OVGS-1 and -2 [26] (called hereafter

Ov-msp-1 and -2, following FUNK) (Fig. 2). The difference

in length found between the MSP sequences of both nema-

todes is restricted to the intron, which is 233 bp in M.

ozzardi and 153±156 bp in O. volvulus (153 bp in Liberian

sequences, 156 bp in Ov±msp-7 and 155 bp in the other

Brazilian sequences). The length of exon 1 in both species

is 99 bp when the primer MSPE £ 1 is included, the length

of the sequenced part of the exon 2 is 267 bp (including

MSPE £ 2) for both species.

The p-distance between the MSP sequences of O. volvu-

lus ranges from 0.42±7.2%. When the distances are calcu-

lated independently for the coding regions and the intron, p-

distances range from 0.93±5.30% in the coding region and

from 0±12.33% in the intron. Mostly single base substitu-

tions account for this variation, however, several indels of

one nucleotide and one indel of three nucleotides were

found in the intron when the Brazilian samples were aligned

with the Liberian ones (Fig. 2). Most of the substitutions

found in both exons fall in the third codon position, ®ve out

of a total of 22 fall in the ®rst or second codon position (four

in the ®rst and one in the second). For M. ozzardi the p-

distances between the sequences range from 0±1.44% (Mo±

msp-7,-11, -14 and -17 are identical to each other). When

the distances are calculated for the coding regions and the

R. Morales Hojas, R.J. Post / International Journal for Parasitology 30 (2000) 1459±1465 1461

Fig. 1. Photo of a gel showing the ampli®cation products of M. ozzardi and

O. volvulus MSP genes. In lanes 1 and 2 are shown the PCR products of

mixed DNA from both species; in lane 3 the PCR product of M. ozzardi; in

lane 4 the PCR product of O. volvulus; and in lane 5 the negative control. M

is the DNA ladder, with the numbers on the left indicating the size of the

marker bands.

intron, p-distances range from 0±1.87% in the coding region

and from 0±1.73% in the intron. All variation amongst the

sequences of M. ozzardi are single base substitutions, there

is no variation in length. Contrary to what was found for O.

volvulus, in M. ozzardi most of the substitutions fall on the

®rst or second codon position (eight are in the second posi-

tion and three in the ®rst), only ®ve out of 16 fall in the third.

The inter-speci®c genetic distance in the MSP gene is

approximately 25%, the p-distance drops to 14±18% if

calculated for the coding regions only.

The UPGMA, NJ and MP trees obtained for both species

were identical (MP trees in Fig. 3). For O. volvulus two

R. Morales Hojas, R.J. Post / International Journal for Parasitology 30 (2000) 1459±14651462

Fig. 2. Alignment of the MSP gene sequences (incomplete at the 5 0 and 3 0 ends) of O. volvulus (a) and M. ozzardi (b) (primer sequences are not included).

Exons are in bold. (.) Represents an identical nucleotide, (:) represents an alignment gap.

evolutionary lineages were identi®ed, one with the Liberian

Ov-msp-2 and the Brazilian Ov-msp-7, and the other lineage

including Ov-msp-1 and the other four Brazilian sequences.

The genetic distance between these two groups is approxi-

mately 6% on average. In the second group the Brazilian

samples are genetically more similar to one another (0.6%

on average) than to the Liberian Ov-msp-1 (approximately

2%), and in the ®rst one Ov-msp-2 differs from Ov-msp-7 by

1.5%. Both lineages were highly supported in the NJ tree

(82 and 100%) by bootstrap analysis. Two MP trees of

length 140 were retained (consistency index (CI) 0.97,

retention index (RI) 0.87, rescaled CI (RC) 0.846). The

number of parsimony-informative characters was 24. In

the MP strict consensus tree both lineages were also highly

supported (96 and 100%) by the bootstrap analysis. For the

M. ozzardi sequences no distinct evolutionary lineages are

identi®ed. The genetic distances between the sequences

range from 0±1.4%. In the MP analysis 26 trees were

retained of length 132 (CI 0.955, RI 0.33, RC 0.32). The

number of parsimony-informative characters was seven.

Bootstrap analysis in the NJ or MP method failed to give

more than 50% support to any branch.

The partial sequences of the two exons sequenced in this

study were translated into the corresponding protein, giving

a sequence of 107 amino acids (the entire protein in O.

volvulus is 127 amino acids in length) (Fig. 4). The intra-

speci®c similarity of the partial amino acid sequences

ranges from 97±100% in O. volvulus and from 95±100%

in M. ozzardi. Among the O. volvulus protein sequences

six amino acid changes were observed, of which ®ve were

non-conservative and one conservative. Among the M.

ozzardi protein sequences 11 amino acid changes were

seen, of which six were non-conservative and ®ve were

conservative. All the amino acid substitutions in the protein

in both species corresponded to changes in the ®rst or

second codon positions of the genes (see above). The third

codon position substitutions were all synonymous except

that of nucleotide 79 of the O. volvulus sequence Ov-msp-

1 (G±C) which corresponds to an amino acid change of

asparagine instead of a lysine in position 26. The inter-

speci®c similarity of the protein partial sequences ranges

from 93.5 to 98%. The amino acid differences between

R. Morales Hojas, R.J. Post / International Journal for Parasitology 30 (2000) 1459±1465 1463

Fig. 3. MP trees showing the genetic relationships of the O. volvulus (a) and

M. ozzardi (b) sequences. Scale-bars in both trees represent the number of

nucleotide substitutions, and numerals above branches represent number of

changes too. Numbers in italics below the branches show bootstrap values.

Fig. 4. Alignment of the major sperm protein sequences (incomplete at the

N and C terminals) of M. ozzardi (Mo-MSP-1 to -17), O. volvulus from

Brazil(Ov-MSP-3 to -7) and from Liberia (Ov-MSP-1 and -2). The abbre-

viations used for the amino acids are the IUPAC single-letter codes. (.)

Indicates same amino acid.

the consensus protein sequence of both species are two,

positions 45 and 75, and they are both conservative (see

Fig. 4).

4. Discussion

The UPGMA phenogram and the phylogenetic trees all

showed that there are two distinct groups of MSP genes in

the Brazilian samples of O. volvulus, the genetic distance

within each of the two groups being much smaller than that

between them. The fact that the two African sequences Ov-

msp-1 and -2 are each included in a different cluster indi-

cates that the two distinct MSP genes originated and

diverged before the O. volvulus populations from the two

continents became isolated from each other. Within each

group, the O. volvulus MSP nucleotide sequences from

Brazil are more similar to each other than to the Liberian

ones. This suggests a reduction in gene ¯ow between the

two populations which would result in a decreased homo-

genisation of the genetic composition of the two populations

[31], and this could have been the result of the introduction

to the Americas.

The pattern of variation that we observe in O. volvulus

MSP genes suggests that both groups of sequences have

evolved, to a certain extent, independently from each

other, which could be the result (1) of neutral drift between

two loci, (2) of neutral drift between two clusters of loci and

concerted evolution within each cluster [32], or (3) of a

functional divergence of the proteins coded by each MSP

gene after duplication of the gene occurred [33]. Different

selective pressures could act on the different MSP genes if

the proteins coded by each had evolved a slightly different

function in the sperm cell. However, empirical evidence

suggests that although there are different isoforms of the

MSP in other nematode species (three in Caenorhabditis

and two in Ascaris), these do not have different functions

[34,35]. Also the proteins coded by each MSP gene in O.

volvulus are highly similar in amino acid sequence (97±

100%), thus, we would favour one of the two earlier expla-

nations over the latter. Some of the MSP genes in O. volvu-

lus are grouped in clusters because the intergenic sequences

were successfully ampli®ed (Post R.J., unpublished results),

and given the known clustered-repetitive structure in some

other species [24,25], suggests that the second explanation

is more likely.

The MSP genes of M. ozzardi do not fall into differen-

tiated groups, as shown by both the phenogram and the

phylogenetic trees, but rather a continuous variation in

distance among the sequences was found. This suggests

that instead of having two distinct MSP gene sequence-

groups in M. ozzardi there is only one which shows poly-

morphism in the population. The maximum level of varia-

tion in M. ozzardi was similar to the distance between

Liberian and Brazilian sequences in the O. volvulus groups,

being higher than the within-group level of variation of the

Brazilian sequences. This degree of variation in M. ozzardi

sequences could be the result, under mutation-drift equili-

brium, of a larger effective population size or of a higher

mutation rate [36], or even of a slower rate of molecular

drive. The rate of evolution of the MSP genes in both

species is not likely to be much different, and populations

of M. ozzardi are almost certainly larger than the American

populations of O. volvulus, hence, the higher level of varia-

tion in M. ozzardi could be explained by the size of the

population. Another likely explanation for the higher level

of variation found in the M. ozzardi sequences compared to

that in the O. volvulus group could be the greater number of

sequences obtained for the former species.

The level of intra-speci®c variation of the MSP coding

sequences in M. ozzardi and O. volvulus is similar to what

was found in Caenorhabditis (87±98% among 14 clones)

[25] and in the plant parasitic nematode Globodera rosto-

chiensis [37]. In O. volvulus the majority of the substitutions

are located in the third codon position and are silent, while

in M. ozzardi most nucleotide substitutions fall in the ®rst or

second codon positions. Similarly, in Caenorhabditis most

of the differences between 14 MSP genes fell in the third

codon position [25] and in G. rostochiensis most nucleotide

substitutions were located in the ®rst or second codon posi-

tions [37]. In the two ®larial species all nucleotide substitu-

tions in the ®rst or second codon positions resulted in amino

acid changes, and there was a high proportion of non-

conservative changes (®ve out of six in O. volvulus and

six out of 11 in M. ozzardi). This is similar to what was

found in O. volvulus from Liberia [26], where three out of

®ve amino acid changes between the protein coded by Ov-

msp-1 and -2 were non-conservative. The intra-speci®c

protein variation was also similar to that found in Caenor-

habditis, Ascaris and G. rostochiensis [25,35,37]. Three

isoelectric forms of the protein are known in Caenorhabditis

and two in Ascaris with no apparent signi®cant difference in

function [34,35]. In the same way, the differences seen in

the amino acid sequences in both ®larial species could

correspond to distinct forms of the protein with no differ-

ence in function in O. volvulus and M. ozzardi.

The MSP protein has an important and unique role in

sperm locomotion [21±23], and thus it has been highly

conserved in nematode evolution. Protein similarity

between the two ®larial species is 93±98%, which is higher

than the similarity of O. volvulus to Ascaris or Caenorhab-

ditis elegans protein (approximately 90 and 80%, respec-

tively), [26]. The functional unit of the MSP is a dimer

formed by the union of two proteins, and these dimers inter-

act to form a sub-®lament. The residues that appear to be

important in the interaction between the dimers are 112±119

[38], and these are conserved between all the species for

which the protein sequence is known (Caenorhabditis,

Ascaris, O. volvulus and G. rostochiensis). The partial

sequence of the protein obtained for O. volvulus and M.

ozzardi in the present study ®nishes in residue 114,

however, the three last residues in both show the same

R. Morales Hojas, R.J. Post / International Journal for Parasitology 30 (2000) 1459±14651464

conservation. While these residues would have to be

conserved, the rest could be more free to vary without

affecting the protein function and making different isoforms.

Acknowledgements

We would like to thank Dr A.J. Shelley, Dr M. Maia-

Herzog, and Dr S CoscaroÂn for supplying the parasite mate-

rials. RMH was supported by a studentship from The

Natural History Museum.

References

[1] Blacklock DB. The development of Onchocerca volvulus in Simulium

damnosum. Ann Trop Med Parasitol 1926;20:1±48.

[2] Shelley AJ, Luna Dias APA, Moraes MAP, Procunier WS. The status

of Simulium oyapockense and S. limbatum as vectors of human onch-

ocerciasis in Brazilian Amazonia. Med Vet Entomol 1987;1:219±34.

[3] Shelley AJ, Lowry CA, Maia Herzog M, Luna Dias APA, Moraes

MAP. Biosystematic studies on the Simuliidae (Diptera) of the

Amazonia onchocerciasis focus. Bull Nat His Mus Lond (Entomol)

1997;66:1±121.

[4] Damalt HT. The black ¯ies (Diptera, Simuliidae) of Guatemala and

their role as vectors of onchocerciasis. Smith Misc Collect

1955;125:1425.

[5] MarroquõÂn HF. Historia de la enfermedad de Robles y de su descu-

brimiento en Guatemala. Guatemala: Luz, 1963.

[6] Zimmerman PA, Katholi CR, Wooten MC, Lang-Unnasch N,

Unnasch TR. Recent evolutionary history of American Onchocerca

volvulus, based on analysis of a tandemly repeated DNA sequence

family. Mol Biol Evol 1994;11:384±92.

[7] McNeeley DF, Raccurt CP, Boncy J, Lowrie Jr RC. Clinical evalua-

tion of Mansonella ozzardi in Haiti. Trop Med Parasitol 1988;40:107±

10.

[8] Natham MB. Transmission of the human ®larial parasite Mansonella

ozzardi by Culicoides phlebotomus (Willinston) (Diptera: Ceratopo-

gonidae) in coastal North Trinidad. Bull Entomol Res 1981;71:97±

105.

[9] Selley AJ, Coscaron S. Simuliid black¯ies (Diptera: Simuliidae) and

ceratopogonid midges (Diptera: Ceratopogonidae) as vectors of

Mansonella ozzardi (Nematoda: Onchocercidae) in Northern Argen-

tina. Mem Inst Oswaldo Cruz 2000 (in press).

[10] YarzaÂbal L, BasaÂnÄez MG, RamõÂrez Perez J, RamõÂrez A, Botto C,

Yarzabal A. Experimental and natural infection of Simulium sanchezi

by Mansonella ozzardi in the middle Orinoco region of Venezuela.

Trans R Soc Trop Med Hyg 1985;79:29±33.

[11] Nelson GS, Pester FRN. The identi®cation of infective larvae in

Simuliidae. Bull WHO 1962;27:473±81.

[12] Kozek WJ, Raccurt C. Ultrastructure of Mansonella ozzardi micro-

®laria, with comparison of the South American (Simuliid-transmitted)

and the Caribbean (Culicoid-transmitted) forms. Tropenmed Parasitol

1983;34:38±53.

[13] Ewert A, Smith JH, Corredor A. Micro®lariae of Mansonella ozzardi

in human skin biopsies. Am J Trop Med Hyg 1981;30:988±91.

[14] Moraes MAP. Mansonella ozzardi micro®lariae in skin snips. Trans R

Soc Trop Med Hyg 1976;70:16.

[15] Fuglsang H, Anderson J. Micro®lariae of Onchocerca volvulus in

blood and urine before, during, and after treatment with diethylcarba-

mizine. J Helminthol 1974;48:93±97.

[16] Anderson RI, Fazen LE, Buck AA. Onchocerciasis in Guatemala. II.

Micro®lariae in urine, blood, and sputum after diethylcarbamizine.

Am J Trop Med Hyg 1975;24:58±61.

[17] Bradley JE, Unnasch TR. Molecular approaches to the diagnosis of

onchocerciasis. Adv Parasitol 1996;37:58±106.

[18] Bradley JE, Trenholme KR, Gillespie AJ, et al. A sensitive serodiag-

nostic test for onchocerciasis using a cocktail of recombinant anti-

genes. Am J Trop Med Hyg 1993;48:198±204.

[19] Meredith SEO, Lando G, Gbakima AA, Zimmerman PA, Unnasch

TR. Onchocerca volvulus: application of the polymerase chain reac-

tion to identi®cation and strain differentiation of the parasite. Exp

Parasitol 1991;73:335±44.

[20] Zimmerman PA, Toe L, Unnasch TR. Design of Onchocerca DNA

probes based upon analysis of a repeated sequence family. Mol

Biochem Parasitol 1993;58:259±68.

[21] Roberts TM, Pavalko FM, Ward S. Membrane and cytoplasmic

proteins are transported in the same organelle complex during nema-

tode spermatogenesis. J Cell Biol 1986;102:1787±96.

[22] Roberts TM. Fine (2±5 nM) ®laments: new types of cytoskeletal

structures. Cell Motil Cytoskeleton 1987;8:130±42.

[23] Roberts TM, Stewart M. Nematode sperm: amoeboid movement

without actin. Trends Cell Biol 1997;7:368±73.

[24] Scott AL, Dinman J, Sussman DJ, Ward S. Major sperm protein and

actin genes in free-living and parasitic nematodes. Parasitology

1989;98:471±8.

[25] Ward S, Burke DJ, Sulston JE, et al. Genomic organization of major

sperm protein genes and pseudogenes in the nematode Caenorhabdi-

tis elegans. J Mol Biol 1988;199:1±13.

[26] Scott AL, Dinman J, Sussman DJ, Yenbutr P, Ward S. Major sperm

protein genes from Onchocerca volvulus. Mol Biochem Parasitol

1989;36:119±26.

[27] Bennett KL, Ward S. Neither a germ line-speci®c nor several soma-

tically expressed genes are lost or rearranged during embryonic chro-

matin diminution in the nematode Ascaris lumbricoides var. suum.

Dev Biol 1986;118:141±7.

[28] Klass MR, Kinsley S, Lopez LC. Isolation and characterization of a

sperm-speci®c gene family in the nematode Caenorhabditis elegans.

Mol Cell Biol 1984;4:529±37.

[29] Setterquist RA, Smith GK, Jones R, Fox GE. Diagnostic probes

targeting the major sperm protein gene that may be useful in the

molecular identi®cation of nematodes. J Nematol 1996;28:414±21.

[30] Blaxter ML, Guiliano DB, Scott AL, Williams SA. A uni®ed nomen-

clature for ®larial genes. Parasitol Today 1997;13:416±7.

[31] Futuyma DJ. Evolutionary biology. Sunderland, MA: Sinauer, 1986.

[32] Dover G. Molecular drive: a cohesive mode of species evolution.

Nature 1982;299:111±7.

[33] Li WH. Evolution of duplicate genes and pseudogenes. In: Nei M,

Koehn RK, editors. Evolution of genes and proteins. Sunderland, MA:

Sinauer, 1983. pp. 14±37.

[34] Burke DJ, Ward S. Identi®cation of a large multigene family encoding

the major sperm protein of Caenorhabditis elegans. J Mol Biol

1983;171:1±29.

[35] King KL, Stewart M, Roberts TM, Seavy M. Structure and macro-

molecular assembly of two isoforms of the major sperm protein

(MSP) from the amoeboid sperm of the nematode, Ascaris suum. J

Cell Sci 1992;101:847±57.

[36] Kimura M. The neutral theory of molecular evolution. Cambridge,

MA: Cambridge University Press, 1983.

[37] Novitski CE, Brown S, Chen R, Corner AS, Atkinson HJ, McPherson

MJ. Major sperm protein genes of Globodera rostochiensis. J Nema-

tol 1993;24:548±54.

[38] Bullock TL, McCoy AJ, Kent HM, Roberts TM, Stewart M. Struc-

tural basis for amoeboid motility in nematode sperm. Nat Struct Biol

1998;5:184±9.

R. Morales Hojas, R.J. Post / International Journal for Parasitology 30 (2000) 1459±1465 1465