Acta Tropica, 53(1993)319-330 319 © 1993 Elsevier Science Publishers B.V. All rights reserved 0001-706X/93/$06.00

ACTROP 00282

Molecular variation in Trichinella

Christopher Bryant Division of Biochemistry and Molecular Biology, School of Life Sciences, Australian National University,

Canberra, ACT, Australia

The taxonomic status of variants within the genus Trichinella is problematical. Some authors recognise no fewer than four species (Trichinella spiralis, T. pseudospiralis, T. nativa and T. nelsoni), others regard T. nativa and T. nelsoni as strains of T. spiralis (T. spiralis var nativa or sylvatica), while others consider the genus to be monospecific, with a variety of more or less well defined isolates. Much of the current evidence adduced to support these various positions is similar to that used pre-1983. It derives from studies of the incidence of Trichinella infections in wild and in domestic animals, comparisons of infectivity of different isolates in laboratory animals and studies of immunity. However, it has become clear that infectivity and epidemiological studies are unreliable tools for discriminating between isolates of Trichinella and it has been shown that differences in the elicitation of immune responses are as much a function of the host as of the parasite. The introduction of monoclonal antibody technology has, however, permitted the identifica- tion of specific antigens in different isolates. The information is as yet scant, and one antigen does not a species make. Isozyme analysis provides some support for separating the various isolates of Trichinella into distinct groups, but cannot of itself shed light on the species problem until certain conditions are met. These conditions are difficult to achieve even in organisms abundantly available and without the baggage of the parasitic habit. Isozyme analysis is probably best used to support the newer studies of genomic DNA. Recent analyses of DNA by restriction endonucleases and dot-blot hybridisation techniques show ample promise of insights into speciation, and a new technique for amplifying the DNA from a single larva by the polymerase chain reaction offers exciting prospects. However, the position yet remains as stated in the first section of this abstract.

Key words: Trichinella spiralis; Intraspecific variation; Enzyme electrophoresis; Monoclonal antibody; genomic DNA; Restriction endonuclease; Polymerase chain reaction

Introduction

D i c k (1983) was s t a n d i n g on the t h r e s h o l d o f the m o d e r n era o f m o l e c u l a r b i o l o g y

w h e n he w r o t e his m u c h - q u o t e d rev iew on v a r i a t i o n in Trichinella. I t is the p u r p o s e

o f this pape r , wh ich does n o t p r e t e n d to be a c o m p r e h e n s i v e review, to c o n s i d e r

recen t ev idence fo r the poss ib i l i ty o f spec ia t ion wi th in Trichinella a n d to see i f the

p o t e n t i a l o f the new b i o l o g y has, for Trichinella at least , been real ised. In the pas t

decade , the re has still been a g rea t e m p h a s i s on the aspec ts o f e p i d e m i o l o g y r ev iewed

by D i c k (1983), b u t i sozyme pa t t e rns , an t igen ic di f ferences and nucle ic ac id s t ruc tu re

o f the v a r i o u s i so la tes h a v e b e g u n to a s s u m e i m p o r t a n c e .

Correspondence to: C. Bryant, Division of Biochemistry and Molecular Biology, School of Life Sciences, Australian National University, Canberra, ACT, Australia. Tel: + 61 (6) 249 4815; Fax: + 61 (6) 249 0313.

320

The parasite

The family Trichinellidae contains only one genus, Trichinella. It remains a matter for debate whether this genus can be further divided. Characteristically, adult worms are found in the small intestines of many different hosts, including rodents, carnivo- rous mammals and omnivorous mammals such as humans and pigs. Trichinella is sometimes known as the garbage worm, and the infection in pigs, in particular, is a cause of great concern because of the important role pig products have in the diets of many peoples. Birds may also be infected (Cheng, 1973).

In the small intestine, adults lie embedded in the mucosa and in the crypts of Lieberkfihn. After copulation, the males die and the females penetrate the crypts. The females are ovoviviparous, and may produce as many as 1500 larvae, after which they die. The larvae enter blood vessels and are carried to the heart and thence to the peripheral circulation. When one encounters striated muscle, it becomes partially intracellular, forming a syncytial cyst (Behnke at aL 1992). The cycle is continued when a suitable host eats the raw muscle; more rarely, perhaps because of lowered immunity, a host may support several cycles within itself.

In humans, trichinosis is a cause of much morbidity and some mortality. Its incidence in western countries is diminishing, because of improved surveillance techniques, but it still figures in the WHO's 'top ten' parasites. Heavy infections of Trichinella are often an immediate cause of diarrhea, with mucus and blood in the stool and severe abdominal pain. Lighter infections may have a delayed onset of the abdominal syndrome. As well, muscle soreness and stiffness (myositis) may ensue (Ozeretskovskaya and Tumolskaya, 1974). However, the severity of the disease appears to vary with geographical location, which may be due to the presence of more or less distinct strains of either parasites or people.

The debate as to whether different species or strains of Trichinella exist is one that is still awaiting a proper resolution. The subject was well reviewed by Dick (1983), who bewailed the lack of reliable morphological characteristics but, from the literature and from his own work, assembled substantial evidence from infectivity studies, cross-breeding experiments, and geographical location which suggests that variants do exist.

The speciationist position

The genus Trich&ella contained one species, T. spiral&, until Forrester et al. (1961) provided evidence for the existence of an African variant. It was later shown that a number of other geographical variants could be identified. Subsequently, Britov and his colleagues (see, for example, Britov, 1980) put forward the view that there existed three species - T. spiral&, confined to the north-temperate regions; a tropical sylvatic form, T. nelsoni; and an arctic sylvatic form, T. nativa. A fourth species, T. pseudospi- ral&, similar in most respects to T. spiralis , except that the larva fails to form a cyst on penetrating host muscle, was proposed by Garkavi (1972). A new category of T. spiral&, the subspecies T. spiral& spiral& has been erected for the pig parasite (Lichtenfels et al., 1983).

There is by no means general agreement about the scheme given above. Variously, one finds in the literature reference to several species (see, for example, Flockhart

321

et al., 1982) - or to one species with several subspecies (see, for example, Marinculic et al., 1991). In the present paper, the names given to the various isolates are the names the authors themselves use. Although this may give rise to some confusion, it is no worse than the confusion endemic in the literature.

Studies in epidemiology, infection and immunity

Although much of the literature differentiates between sylvatic cycles and domestic cycles of T. spiralis, there is some evidence that the distinction may be a false one. Murrell et al. (1987) found that the parasite established in a domestic swine herd had marked similarities (data from infectivity trials, isoenzyme analyses and repetitive DNA sequence analyses) to isolates recovered from rats, skunks, opossums and raccoons. On the other hand, Marinculic et al. (1991) found that a sylvatic isolate of T. spiralis (T. s. nativa) inoculated into pigs resulted in a lower 'take' than that observed when T. s. spiralis, the pig strain, was used. Nonetheless, there was a high cross-immunity. More detailed analysis showed that pigs inoculated with the sylvatic strain recognised more antigens than those inoculated with the pig strain.

Trichinella has recently been found in mammals native to Tasmania (Henry, 1989). The parasites have been recovered from three carnivorous species (the Tasmanian devil, Sarcophilus harrisii, and two species of quoll, Dasyurus maculata and D. viverrinus). Extensive testing of meat samples from mainland Australia has failed to detect the organism in live-stock. Following the discovery of Trichinella in Tasmania, surveillance was stepped up, but although an 25-30% incidence in native mammals was recorded, all pig meat samples from Tasmania and the Australian mainland were negative. It appears that the cycle is purely sylvatic in Tasmania.

A more detailed study of the parasites suggested that they belong to the pseudospio ralis group (Obendorf et al., 1990). Characteristics of the worms, including a weak PAS reaction, and strong reactions for sorbitol and lactic dehydrogenase, considered by Garbryal et al. (1981) to be at least partly diagnostic, are consistent with this identification, and an independent assessment, using a DNA hybridisation technique, supports it. A retrospective museum study indicates that the Tasmanian cycle has existed for at least 16 years. It would be interesting to know the origin of this parasite; whether it was imported in livestock by the first colonists, whether it was imported earlier from the north or, less likely, whether it is a Gondwanaland survival.

The use of inbred mice has proved a valuable tool for the analysis of the immune response against a whole variety of parasites. Dick et al. (1988) compared the responses of five strains of mice, two of which were susceptible and three of which were resistant to infection with T. spiralis. Three isolates of T. spiralis were used. They were designated pig, arctic fox and pseudospiralis strains. The resistant mice expelled over 80% of worms by the tenth day of infection, although the response was strongest against the arctic fox isolate. The susceptible strains of mice were much slower to develop a response but, once again, the arctic fox isolate was most effective. This work lends additional support to the views of Wassom et al. (1988) that the kinetics of immune responsiveness to different T. spiralis isolates is dependent on the expression of different antigens and to those of Bolas-Fernandez and Wakelin (1989, 1990) that infectivity of Trichinella isolates depends as much or more on the immunoresponsiveness of the host than on the intrinsic properties of the parasite.

322

Tuohy et al. (1990) also explored the aspects of the immune response to T. spiral& in high and low responder mice. In high responder mice, worm loss from a primary infection is complete by about day 15 post infection; low responders are still harbour- ing about 25% of the initial worm burden after day 30. These workers found that the ability to generate a rapid mast cell response was characteristic of high responder mice. This response was mirrored in the blood levels of intestinal mast cell proteinase and there was an inverse relationship between this measurement and worm burden, which suggested that worm expulsion was related to mast cell activity.

Further confirmation that different T. spiralis isolates induce different effects in different strains of mice derives from the work of Stewart et al. (1988). They measured plasma corticosterone levels in mice receiving either T. spiralis or T. pseudospiralis. They found that male ICR swiss albino mice receiving T. pseudospiralis had higher levels during the course of the infection, at least up to day 20. This was related to a direct effect on the zona fasciculata of the adrenals, which exhibited increased mitochondrial biogenesis and lipid depletion. Interestingly, mice receiving T. pseudo- spiralis did not show signs of the myositis which normally accompanies infection with T. spiralis and, in a concurrent infection of both isolates, the symptoms of myositis were suppressed. It appears that at least part of the immunosuppressive effect of T. pseudospiralis is mediated through increased corticosteroid levels. A similar immunosuppressive phenomenon is observed in the Chinese hamster. Further, the suppression of the systemic and inflammatory response in this animal by T. pseudospiralis results in a much higher reproductive success than for T. spiralis; the ratio of the reproductive capacity indices is greater than 10:1 (Stewart and Larsen, 1989; Larsen et al., 1991). Of considerable interest in this regard is the observation by Minchella et al. (1989) is that, in twenty T. spiralis isolates from wild mammals, parasite fecundity was ten to thirty times greater in wild mice (Peromyscus) than in white mice.

Bolas-Fernadez and Wakelin (1992) used partially purified antigens from six Trichinella isolates in a comparative study of immunogenicity in mice. The surface antigens were all produced in the same way, by incubating the larvae in cetyl- trimethyl ammonium bromide in an attempt to standardise the preparations. Immunogenicity was determined by antibody and lymphocyte responses, and the establishment of protective immunity against challenge infections. They preparations showed considerable cross-reactivity, and two types of immunogenicity, poor or effective, could be discerned. This clearly has important implications for the develop- ment of vaccines against widely distributed forms of T. spiralis.

Monoclonal antibodies have not long been added to the experimental armoury against T. spiralis. Wassom et al. (1988) showed that it was possible to prepare monoclonal antibodies that recognised an antigen produced by T. pseudospiralis (or, in their terms, T. spiralis var. pseudospiralis) that was not shared by T. spiralis. On the other hand, Rodriguez-Perez et al. (1989) derived saline extracts from T. spiralis and T. pseudospiralis and subjected them to Western blot analysis with a monoclonal mouse antibody ES/TA2, previously produced against T. spiralis larvae. The monoclonal antibody recognised seven antigenic components from T. spiralis but only five from T. pseudospiralis.

More recently, monoclonal antibodies (mAb 3G6 and mAb 3El0) have been incorporated into an ELISA test (Kehayov et al., 1991) which has been used to investigate differences between Trichinella species. The first of the monoclonals gave

323

positive results with T. spiral&, T. nelsoni, T. nativa and T. pseudospiralis and proved to be associated with both the stichosomes and the surface cuticle of T. spiralis and T. pseudospiralis. The second did not recognise T. pseudospiralis, and was strongly associated with the surface cuticle and capsule of T. spiralis. As the authors remark, 3El0 could become a useful diagnostic tool for distinguishing between T. spiralis and T. pseudospiralis in taxonomic studies.

Other evidence for the separation of T. spiralis from T. pseudospiralis rests on the different effects the parasites have on the expression of myofibrillar proteins in muscle cells of the host mouse (Jasmer, 1990: Jasmer et al., 1991). Myofibrillar protein expression is greatly reduced or absent from muscle cells infected with T. spiralis. This is not true for T. pseudospiralis, as antibody reactions to myosin heavy chain and tropomyosins persist throughout the infection. If, as the authors point out, this can be verified, it is an important observation for it implies that T. spiralis secretes factors that are unique to itself. It is tempting to speculate that these are the two additional antigenic components detected by Rodriguez-Perez et al. (1989) but there is no evidence for this.

Leiby and Bacha (1987) obtained isolates of T. spiralis from black bear (Ursus americanus), grey fox (Urocyon cinereoargenteus) and the domestic pig and passaged them for ten generations in outbred Swiss Webster mice. They found the following differences: the bear isolate was always more posterior in the intestine than the pig isolate; the sex ratio of the fox isolate was lower than those of the other two; females of the pig isolate produced more larvae than the other two and had a higher reproductive capacity index; and finally, male worms from the fox isolate were significantly smaller.

The fact that different isolates occupy different portions of the intestine may be of considerable significance. The characteristics of the small intestine vary along its length, and the environments provided by the small intestines of different host animals also vary. These simple observations beg the question of the extent to which differences in morphology and growth of different T. spiralis isolates are intrinsic to the parasite and which can be ascribed to environment they inhabit Sukhdeo (1991) has shown that adult female T. spiralis recovered from the jejunum of rats were significantly more fecund that worms recovered from the distal ileum. This was confirmed by surgical implantation of worms and it was concluded that the physico- chemical conditions of the small intestine were optimal for parasite growth.

It is also probable, however, that the immune expression varies along the intestine. Recent studies on the tapeworm, Hymenolepis diminuta, have shown that the mor- phology and metabolism of the parasite varies, depending on the immune status of the host rat (Bennet et al., 1990) and similar changes occur in Echinococcus granulo- sus, depending on whether the parasites are in the anterior or posterior part of the small intestine of the dog. This is probably due to the presence of different popula- tions of cells of the immune system (E.M. Bennet, unpublished). Effects of this sort operating on Trichinella may account for some of the variations in infectivity that are often observed.

The possible impact of environmental effects on apparent strain variation is well emphasised by some interesting and challenging observations on T. pseudospiralis by Stewart et al. (1990). They found that the establishment of larvae in mice was significantly reduced by exposure to low pH or 1% pepsin at low pH, when compared with larvae maintained in buffered saline at pH 7. This effect was independent of

324

the immune status of the host and was reflected in longevity and fecundity of the parasite. It would be interesting to document other simple changes in the environment of the infective larvae and their effect on these parameters, and to what extent the geographical distribution of experimenters using slightly different techniques has an effect on variation in Trichinella!

That altered environmental conditions may indeed be important in nature is also indicated by a study of the infectivity of T. pseudospiralis isolated from carrion (Stewart et al., 1990). The reproductive capacity index of worms recovered from mice carcases at 24°C remained at a high level for five days after the mice were killed but declined to one fifth after 10 days. After 10 days, viable worms were not recovered but there is evidence that at lower temperatures (4°C) worms remain infective for 30 days (Ooi et al., 1986).

Protein polymorphism - isozyme analysis

Harris (1980) defined polymorphism as "a term used to specify a situation in which members of a naturally occurring population can be categorised into sharply distinct phenotypes which are determined by two or more allelles at a given gene locus and in which the different phenotypes are each relatively common in the population". Unfortunately, this definition leaves a great deal to the judgement of the individual worker and it is expert judgement exercised in different ways that bedevils the whole topic of speciation in the genus Trichinella. As remarked above, the lack of gross morphological differences in the parasites has driven workers to look for other characteristics that might substitute. Protein polymorphism is one such character.

Unfortunately, there is little agreement whether protein polymorphism is an appropriate character upon which to erect a new species. The arguments for and against have been reviewed for helminth parasites by Bryant and Flockhart (1986). In general, they concluded that, to form the basis of a case for speciation, protein polymorphism should be shown to be adaptive, and not the result of genetic drift. Clarke (1975) and Koehne (1978) separately set up the criteria which must be satisfied before a particular enzyme polymorphism can be shown to be due to selective processes and, therefore, adaptive. To summarise these criteria briefly, they are that:

i. there should be a strong correlation between phenotype and the presence of given isozyme.

ii. the different molecular function should be relevant to the metabolic economy of the organism.

iii. a specific locus on the genome must interact with a specific ecological compo- nent by means of the enzyme produced at that locus.

iv. the phenotype differences must confer some measure of fitness. This is is a tall order indeed; very rarely, if ever, are all the criteria seen to be

satisfied because the causality is so difficult to establish. However, if a large number of polymorphisms can be demonstrated, it does at least provide prima facie evidence for speciation to have occurred. But what is a sufficient number to allow the erection of a species?

If these criteria are difficult to satisfy with easily accessible organisms, it is well nigh impossible with an organism like T. spiralis. For example, some 21 isolates

325

from a broad sample of hosts and from widely separated geographical locations were used to compare T. spiralis, T. nelsoni and T. nativa (Flockhart et al., 1982; Flockhart, reported in Bryant and Flockhart, 1986). Three enzymes, phosphoglucose isomerase, esterase and lactate dehydrogenase showed promise as markers, and showed that pig, human and rat isolates were closely related. The status of T. nativa and T. nelsoni remained unresolved. A similar suite of enzymes has been used more recently to establish that an isolate from Thailand is similar to T. spiralis (Pozio, 1987; Pozio and Khamboonruang, 1988). In none of these cases was any functional relationship between isozyme patterns and the ecology of Trichinella established.

Fukumoto et al. (1987) investigated eight enzymes by electrofocusing on polyacryl- amide gel. Lactate dehydrogenase, malate dehydrogenase, 6-phosphogluconate dehy- drogenase and adenylate kinase were not variable enough to discriminate between isolates of T. pseudospiralis. Malic enzyme, phosphoglucoisomerase, superoxide dis- mutase and phosphoglucomutase, however, permitted the establishment of four types of T. pseudospiralis. They were as follows: type 1 - a raccoon strain; type 2 - a polar bear strain; type 3 - the Iwasaki strain from a Japanese black bear and the Yamagata strain from a raccoon dog; type 4 - a Polish strain from a wild pig, a USA strain from a domestic pig, and a human Thai strain. Type 3 was considered to be genetically intermediate between types 2 and 4.

The relationship between the four types of T. pseudospiralis, described above, and T. spiralis was investigated further using an additional suite of six enzymes (Fukumoto et al., 1988). These enzymes consisted of two esterases, acid phosphatase, N-acetylglucosaminidase, peptidase and mannosephosphate isomerase. Combining the results from twelve enzymes enabled the authors to compile a molecular phyloge- netic tree. Coefficients of genetic similarity showed that types 2 and 3 were most closely similar, and they resembled type 4 more than they resembled type 1. These two papers (Fukumoto et al., 1987, 1988) seem to embody the confusion experienced by every worker when confronting speciation in Trichinella. The earlier paper refers to all seven strains as variants of T. pseudospiralis. The later paper describes six of them as variants of T. spiralis.

Other authors have had different successes, and perhaps the best use of isozyme patterns is in conjunction with other techniques for establishing differences, such as DNA fingerprinting and infectivity studies. The infectivity studies of Murrell et al. (1987) have already been referred to, but they were backed up by isozyme and DNA studies. Murrell et al. (1987) surveyed 28 enzyme reactions in domestic and sylvatic isolates of T. spiralis. Six enzymes, aldolase, fumarase, glucose-6-phosphate dehydro- genase, phosphoglucoisomerase and phosphoglucomutase gave 'good' - that is to say, clear, sharp and reproducible - results. Of these, fumarase, phosphoglucose isomerase and phosphoglucomutase gave results that reflected differences among the isolates. Pig, skunk, raccoon and opossum isolates from New Jersey had identical isozymes, whereas an isolate from a Pennsylvanian fox, collected a good distance from New Jersey, showed marked differences. It is, however, impossible even to speculate on the functional significance of these differences.

DNA studies

One of the earliest attempts to employ an aspect of DNA structure in the elucidation of the species problem in the genus Trichinella was that of Feldman et al. (1975).

326

These authors measured the guanine + cytosine content of T. spiralis, and found it to be significantly lower (35.7 moles%) than that of the mouse (41.8 moles%). In this it was close to a range of other helminths, including nematodes, trematodes and cestodes, and the somewhat naive hope was expressed that such measurements might be useful in showing phylogenetic relationships between closely related species. In the intervening period, the study of DNA has advanced to the stage where it can be extracted, purified, cut by restriction enzymes, analysed in detail and rebuilt if necessary. In 1988, DeVos et al. published the sequence analysis of a 1.6 kilobase repetitive DNA element of an isolate of T. spiralis from pigs. The DNA was ligated to the EcoRI site of the plasmid pPra and, when expressed, hybridised with DNA from a range of T. spiraIis isolates. It shows no significant homologies with othe eukaryotic DNAs or structural RNA sequences.

This example is given to illustrate the potential power of the DNA technology in resolving questions of speciation within taxonomic groups. Unfortunately, no doubt due to the intractability of Trichinella material, the resolution of the problem of the speciation of Trichinella is not yet in sight. There are, however, many hopeful signs. Dame et al. (1987) investigated the structure of genomic DNA from seventeen American and one Asian isolate of Trichinella, Thirteen were of American sylvatic origin, four came from domestic pigs and the last was derived from a pig from Thailand. After extraction, the DNA was cut with HindIII and the restriction fragments run out on agarose gel. Thirteen isolates gave nearly identical patterns: these included eight sylvatic isolates (from a polar bear, a bobca t , two black bears, two raccoons, a skunk and an opossum) and five domestic isolates (four domestic pigs from different states in USA and the Thai pig). The remaining five isolates were distinct; three originated in black bears from different states, one in a polar bear and one in a fox from Pennsylvania. All isolates from the first group were highly infectious for pigs and were classified as T. spiralis spiralis (pig); the second group was unassigned. From this the authors concluded that the pig worm occurs in wild mammals and formed a severe impediment to attempts to eradicate the disease. This work, in effect, confirmed and extended the earlier study of Murrell et al. (1987) and lends powerful support to the view that at least two strains (species?) of TrichineIla exist. Additional evidence derived from DNA technology indicates that, of two isolates of Trichinella from humans in France, one was the pig type and one was a sylvatic type. Pattern comparisons also showed that T. nelsoni is similar to North American sylvatic variants (Dick et al. 1990).

Minchella et al. (1989) used dot-blot hybridisation with DNA probe pBP2 to separate twenty one isolates (twenty from wild mammals - six coyotes, six raccoons, seven red foxes and a mink - and one from a domestic pig) into two groups. The probe, which is specific for the swine parasite (T. spiralis spiralis), recognised only the pig and one coyote of the twenty one isolates. Subsequent analysis with twelve restriction endonucleases gave identical banding patterns for these two. Restriction enzyme analysis of the remaining sylvatic isolates, using ClaI, gave a series of unique banding patterns suggesting considerable genetic heterogeneity (Dick et al. 1990).

Conclusion

The starting point for this survey of speciation and strain variation in Trichinella was the review by Dick (1983). A decade has since passed, a time during which the

327

landmarks of molecular biology have seemed to appear on the horizon and disappear into history with bewildering and increasing frequency. It is ironic now that while so much is known about the genetic material at least one nematode - Caenorhabditis elegans - that nematode happens to be a free-living species. This is, I suppose, a statement about the difficulty of working with parasitic material. So many of the adaptations of a parasitic helminth are concerned with adaptation to a host. Many of those adaptations are not accessible of study in the absence of the host and are concerned with what Behnke et al. (1992) call the host-parasite arms race.

In the case of Trichinella, the promise of 1983 has yet to be realised. A cold look at the evidence indicates that most of our information is derived from studies of infection, immunity and epidemiology. The information gained from these is neither sufficient nor rigourous enough to permit the description of isolates of Trichinella as anything other than isolates, except possibly for the division between T. spiralis and T. pseudospiralis. However, molecular biology has had little to do with that decision, as differences in larval behaviour are probably sufficient to discriminate between them. The status of the isolates T. nelsoni and T. nativa remains problemati- cal. On the basis of the evidence presented here it seems most likely that they are variants of T. spiralis. The problem is that many of the characteristics that have been used to distinguish them from T. spiralis are of host origin rather than parasite origin - for example, pathogenicity and the elicitation of immune responses. It does seem possible, however, that T. spiralis spiralis, specialising in the domestic pig, is on its way to separation from the main stock.

It is hard to see how an indiscriminate and opportunistic parasite like Trichinella could speciate unless it began to specialise. To describe a separate sylvatic species implies that there is something about wild mammals - from polar bears to opossums - that makes them collectively similar and somehow sets them apart from domestic species. This is absurd. Each species parasitised will present special problems; there may be local selection pressures set up. If isolation or geographical separation ensues each isolate will develop its own minor differences. Australian blackbirds sing with an Australian 'accent', so to speak, but they still belong to the same species as European ones. And Alsatian dogs belong to the same species as foxhounds. Because Trichinella is such a successful generalist and as long as there is the possibility of recruitment from the larger Trichinella gene pool, speciation is unlikely to occur

These general statements are supported by such molecular studies as have been completed. At best, there is evidence for three variants, T. spiralis spiralis, T. spiralis var. pseudospiralis and T. spiralis var. nativa. None of these, in the opinion of the writer, seems sufficiently distinct to warrant the creation of separate species.

Perhaps the next decade will provide the definitive answer. It therefore seems appropriate to launch it with another paper by Dick and coworkers (Dick et al., 1992). They have developed a polymerase chain reaction-based method which has the potential to amplify DNA fro,u-a a single larva. They used oligonucleotide primers based on sequences from the 1.6 kilobase repetitive DNA clone pPRA from a pig isolate (deVos et al., 1988). This proved specific for pig isolates, as amplification of DNA did not occur with sylvatic or pseudospiralis isolates. If it does prove to be possible to amplify DNA from a single larva, and other primers can be developed, it will be interesting to learn whether infections within a single host are homogeneous.

One final point needs to be made. Trichinosis may be attacked in three ways. The first depends on a thorough knowledge of the so-called sylvatic cycles, so that they

328

may be broken at their most vulnerable point . The second is the development of new antiparasi te drugs to treat humans and livestock in the immediate future. Third, and for the longer term, is the development of new vaccines. All three are predicated on an accurate knowledge of the parasite. Of the first, no more need be said. New drugs depend on an accurate knowledge of potent ial targets within the parasite. Experience shows that different species of parasite respond in different ways to anthelminthics. Fur ther , the gene pool of Trichinella is so widespread and diverse that it is likely that resistance will rapidly emerge unless any new drug is effective across the whole range of isolates. And finally, any new vaccine will have to be mul t ivalent to cope with the different antigenicity of different isolates. In the end, molecular biology will help us describe the parasite and all its variants, but control is likely to depend more on an accurate knowledge of epidemiology.

References

Behnke, J.M., Barnard, C.J. and Wakelin, D. (1992) Understanding chronic nematode infections: evolu- tionary considerations, current hypotheses and the way forward. Int. J. Parasitol. 22, 861-907.

Bennet, E.M., Behm, C.A. and Bryant, C. (1990) The role of the host in the regulation of end product formation in two strains of the rat tapeworm, Hymenolepis diminuta. Int. J. Parasitol. 20, 841-848.

Bolas-Fernadez, F. and Wakelin, D. (1989) Infectivity of Trichinella isolates is determined by host immune responsiveness. Parasitology 99, 83-88.

Bolas-Fernadez, F. and Wakelin, D. (1990) Infectivity, antigenicity and host responses to isolates of the genus Trichinella. Parasitology 100, 491-497.

Bolas-Fernadez, F. and Wakelin, D. (1992) Immunization against geographical isolates of Trichinella spiralis in mice. Int. J. Parasitol. 22, 773-781.

Britov, V.A. (1980) the species of Trichinella, their specificity and their role in initiating disease in humans and animals. Helminthologia 17, 63-66.

Bryant, C. and Flockhart, H. A. (1986) Biochemical strain variation in helminths. Adv. Parasitol. 25, 276 319.

Clarke, B. (1975) The contribution of ecological genetics to evolutionary theory: detecting the direct effects of natural selection on particular polymorphic loci. Genetics 79, I01-113.

Cheng, T.C. (1973) General Parasitology. Academic Press, New York. Dame, J.B., Murrell, J.D., Worley, D.E. and Schad, G.A. (1987) Triehinella spiralis : genetic evidence

for synanthropic subspecies in sylvatic hosts. Exp. Parasitol. 64, 195-203. DeVos, T., Klassen, G.R. and Dick, T.A. (1988) Sequence analysis of a 1.6kb repetitive element from a

porcine isolate of T. spiralis. Nucleic Acids. Res. 16, 3114-3115. Dick, T.A. (1983) Species and intraspecific variation. In Trichinella and Trichinosis (W.C.Campbell, ed.),

pp. 31 73, Plenum Press, New York. Dick, T.A., DeVos, T. and Dupouy-Camet, J. (1990) Identification of two isolates of Trichinella from

humans in France. J. Parasitol. 76, 41-44. Dick, T.A., Dougherty, D.A. and Wassom, D.L. (1988). Trichinella spiralis infections of inbred mice:

genetics of the host response following infection with different Trichinella isolates. J. Parasitol. 74, 665-669.

Dick, T.A., Lu, M., DeVos, T. and Ma, K. (1992) The use of the polymerase chain reaction to identify porcine isolates of Trichinella. J. Parasitol. 78, 145-148.

Feldman, A., Rosenkrantz, H.S. and Despommier, D. (1975) Guanine-cytosine content of DNA from the mature muscle larva of Trichinella spiralis as determined from buoyant density and thermal-helix coil transition measurements. J. Parasitol. 63, 570-571.

Forrester, A.T., Nelson, G.S. and Sander, G. (1961). The first resord of an outbreak of trichinosis in Africa, south of the Sahara. Trans. R. Soc. Trop. Med. Hyg. 55, 503-513.

Flockhart, H.A., Harrison, S.E., Dobinson, A.R. and James, E.R. (1982) Enzyme polymorphism in Trichinella. Trans. R. Soc. Trop. Med. Hyg. 76, 54l 545.

329

Fukumoto, S., Takechi, M., Kamo, H. and Yamaguchi, T. (1987) Comparative studies on soluble protein profiles and isozyme patterns of seven Trichinella isolates. Parasitol. Res. 73, 352-357.

Fukumoto, S., Nagai, D., Yazaki, S., Kamo, H. and Yamaguchi, T. (1988) The molecular phylogenic tree of the genus Trichinella constructed from isozyme patterns. Parasitol. Res. 74, 574 580.

Garbryel, R., Gustowska, L. Rauhut, W. and Blotna- Filipiak, M. (1981) In "Trichinellosis" (Kim, C.W., Ruitenberg, E.J. and Teppema, J.S. eds) pp231-234. Reedbooks, Surrey, England.

Garkavi, B.L. (1972) Species of Trichinella isolated from wild animals. Veterianariya 10, 90-94. Harris, H. (1980) The Principles of Human Biochemical Genetics, 3rd Edn., Elsevier/North-Holland,

Amsterdam. Henry, W. (1989) Trichinella pseudospiralis in Tasmanian native fauna. Aust. Vet. J. 66, 336. Jasmer, D.P. (1990) Trichinella spiralis: expression of muscle proteins in trichinosis. Exp. Parasitol.

70, 45-55. Jasmer, D.P., Bohnet, S. and Prieur, D.J. (1991) Trichinella spp.: differential expression of acid phospha-

tase and myofibrillar proteins in infected muscle cells. Exp. Parasitol. 72, 321-331. Kehayov, I., Tankov, C., Komandarev, S. and Kyurkchiev, S. (1991) Antigenic differences between T.

spiralis and T. pseudospiralis detected by monoclonal antibodies. Parasitol. Res. 77, 72-76. Koehne, R.K. (1978) Physiology and biochemistry of enzyme variation; the interface of ecology and

population genetics. In "Ecological Genetics: the Interface" (Bussard, P.F., ed.), pp51-72, Springer, Berlin.

Larsen, E., Stewart, G.L. and Niederkorn, J.Y. (1991) Trichinella pseudospiralis overcomes the innate resistance of the Chinese hamster to Trichinella spiralis. Parasitology 103, 465-470.

Leiby, D.A. and Bacha, W.J. (1987) A comparison of three geographical isolates of T. spiralis from the mid-atlantic United States. J. Parasitol. 73, 207-213.

Lichtenfels, J.R., Murrell, K.D. and Pilitt, P.A. (1983). Comparison of three subspecies of Trichinella spiralis by scanning electron microscopy. J. Parasitol. 69, 1131-1140.

Marinculic, A., Gamble, H.R., Urban, J.F., Rapic, D., Zivicnjak, T., Smith, H.J. and Murrell, K.D. (1991) Immunity in swine inoculated with larvae or extracts of a pig isolate and a sylvatic isolate of Trichinella spiralis. Am. J. Vet. Res. 52, 754-758.

Minchella, D.J., Branstetter, B.A. and Kazacos, K.R. (1989) Molecular characterisation of sylvatic isolates of Trichinella spiralis. J. Parasitol. 75, 388-392.

Murrell, K.D., Stringfellow, F., Dame, J.B., Leiby, D.A., Duffy, C. and Schad, G.A. (1987) Trichinella spiralis in an agricultural ecosystem II. Evidence for a natural transmission of Trichinella spiralis spiralis from domestic swine to wildlife. J. Parasitol, 73, 103-109.

Obendorf, D., Handlinger. J.H., Mason, R.W., Clarke, K.P., Forman, A.J., Hooper, P.T., Smith, S.J. and Holdsworth, M. (1990) Trichinella pseudospiralis infection in Tasmanian wildlife. Aust. Vet. J. 67, 108 110.

Ooi, H.K., Kamiya, M., Ohbayashi, M, and Nakazarva, M. (1986) Infectivity in rodents and cold resistance of Trichinella spiralis isolated from pig and polar bear and Trichinella pseudospiralis. Jap. J. Vet. Res. 34, 105-110.

Ozeretskovskaya, N.N. and Tumolskaya, N.I. (1974) Clinical pattern and pathogenesis of the abdominal syndrome in trichinellosis. In "Trichinellosis" (Kim, C.W., ed.), pp389-398. Intext Educational Publishers, New York.

Pozio, E. (1987) Isoenzymatic typing of 23 Trichinella isolates. Trop. Med Parasitol. 38, 111 116. Pozio, E. and Khamboonruang, C. (1988) Trichinellosis in Thailand: epidemiology and biochemical

identification of-4.he aetiological agent. Trop. Med. Parasitol. 40, 73-74. Rodriguez- Perez, J., Gomez-Garcia, V., Rodriguez-Osorio, M., Rojas-Gonzalez, J. and Gomez-Morales

M.A. (1989) Differentiation between T. richinella spiralis and Trichinella pseudospiralis infective larvae by a monoclonal antibody. J. Helminth. 63, 275-279.

Stewart, G.L., Kennedy, R.R. and Larsen, E. (1990) Alterations in the longevity and fecundity of adult Trichinella pseudospiralis related to method of isolation of infective larvae. J. Parasitol. 76, 297-301.

Stewart, G.L. and Larsen, E. (1989) Infection of Chinese Hamster with Trichinella pseudospiralis. J. Parasitol. 75, 1006-1007.

Stewart, G.L., Kennedy, R.R. and Larsen, E. (1990) Infectivity of Trichinella pseudospiralis isolated from carrion. J. Parasitol. 76, 750-751.

Stewart, G.L., Mann, M. A., Ubelaker, J. E., McCarthy, J.L. and Wood, B.G. (1988) A role for elevated plasma cortisone in modulation of host response during infection with Trichinella pseudospiralis. Par. Immunol. 10, 139-150.

330

Sukhdeo, M.V.K. (1991) the relationship between intestinal location and fecundity in adult Trichinella spiralis. Int. J. Parasitol. 21,855-858.

Tuohy, M., Laminas, D.A., Wakelin, D., Huntly, J. F., Newlands, G. J. F. and Miller, H. R. P. (1990) Functional correlations between mucosal mast cell activity and immunity to Trichinella spiralis in high and low responder mice. Par. Immunol. 12, 675-685.

Wassom, D.L., Dougherty, D.A. and Dick, T.A. (1988) Trichinella spiralis infections in inbred mice: immunologically specific responses induced by different Trichinella isolates. J. Parasitol. 74, 283-287.