INTRODUCTION
Ascaris is a large parasitic roundworm (nematode) of the small intestine of both humans and pigs. Ascariasis, the disease caused by this parasite genus, is of major human health importance in many parts of the world (WHO, 1987; Peng et al. 1998a; O'Lorcain and Holland, 2000; Crompton, 2001). Ascaris infects a quarter of the world's human population and clinically affects ~100–200 million people, particularly children (WHO, 1987; Crompton, 1989, 2001; Peng et al. 1996, 1998a,b; O'Lorcain and Holland, 2000; Stephenson, 2002). Also, ascariasis of pigs is of major economic significance due to production losses linked to reduced feed conversion efficiency and losses to the meat industry associated with the condemnation of ‘milk-spot’ livers (Stewart and Hale, 1988). Central to the implementation of control programmes against Ascaris is a clear understanding of fundamental aspects of its biology, genetics and epidemiology. While there is knowledge about the epidemiology and population biology of Ascaris for a number of countries (Crompton, 1989; O'Lorcain and Holland, 2000; Anderson, 2001; Crompton, 2001; Peng et al. 2002), there has been controversy as to whether Ascaris-infected pigs represent a significant reservoir for human infection in endemic and non-endemic regions (Anderson, 1995, 2001; Anderson et al. 1993, 1995; Peng and Zhou, 2001). Particularly for large endemic countries, such as China, questions regarding the genetic structure of Ascaris populations, host specificity and transmission patterns had scarcely been addressed (see Peng et al. 1996, 1998a,b). Clearly, investigating these areas is interesting scientifically and, importantly, contributes towards the improved prevention and control of Ascaris.
There has been debate about Ascaris lumbricoides Linnaeus, 1758 of humans representing a different species to Ascaris suum Goeze, 1782 of pigs, and various studies have attempted to address this problem (Barry and O'Rourke, 1967; Crompton, 1989; Anderson, 2001). Experimental cross-host species transmission studies have shown that Ascaris of human origin can infect the pig, and vice versa (Takata, 1951; Galvin, 1968). Other studies (Sprent, 1952; Ansel and Thibaut, 1973; Maung, 1973; Kurimoto, 1974; Kennedy et al. 1987; Nadler, 1987) have applied morphological, karyotypic, immunological and/or biochemical approaches to attempt to address this problem. More recently, molecular methods have been employed to investigate the genetic make-up of Ascaris populations (Anderson, 2001; Anderson et al. 1993, 1995; Peng et al. 1998c; Nadler et al. 1995; Nadler, 1996; Anderson and Jaenike, 1997; Nadler and Hudspeth, 1998). For example, PCR-coupled restriction fragment length polymorphism (PCR-RFLP) analyses of a nuclear ribosomal DNA (rDNA) region spanning the first (ITS-1) and second (ITS-2) internal transcribed spacers (=ITS) and the intervening 5.8S gene and of mitochondrial DNA regions had been used to study Ascaris populations (Anderson, 1995; Anderson et al. 1995; Anderson and Jaenike, 1997; Peng et al. 1998c), but there was no clear genetic delineation between human Ascaris and pig Ascaris (Anderson, 2001). While PCR-RFLP analysis of nuclear (including ribosomal) and of mitochondrial DNA markers has been used, nucleotide sequence data sets were lacking. However, Zhu et al. (1999) demonstrated a nucleotide sequence difference of 1·3% in a ~300 bp region of the ITS-1 (designated pITS-1) of nuclear ribosomal DNA between human and pig Ascaris. Due to the limited number of samples (from widely separated geographical locations) used, the interpretation from this study was guarded, but the sequence data had suggested population substructuring within the genus (based on sequence data).
Extending from previous studies, a series of recent investigations has been conducted to gain a better insight into the genetics and epidemiology of Ascaris in China. The main aims were (1) to study nucleotide variation in ribosomal and mitochondrial DNAs within and among Ascaris population from human and pig hosts from a range of different geographical regions in China employing a mutation scanning-coupled sequencing approach, (2) to establish whether there is a particular relationship between Ascaris genotype and host species, and (3) to establish an experimental infection of Ascaris of human origin in pigs, to be able to investigate the genetic and reproductive relationships between human Ascaris and pig Ascaris. The present article reviews recent findings and, based on them, makes conclusions regarding the biology and epidemiology of Ascaris and its prevention and control in China.
The genetic make-up of Ascaris populations from humans and pigs in China
Using the mutation scanning-based approach (i.e, single-strand conformation polymorphism (SSCP) and selective sequencing), Peng et al. (2003a) investigated nucleotide variation in a part of the first internal transcribed spacer (pITS-1) of nuclear ribosomal DNA within and among a large number (n=815) of Ascaris individuals from humans and pigs from 6 endemic regions in China, and recorded the frequency of the different genotypes of Ascaris in relation to host species and geographical origin. Five different genotypes (G1–G5) were detected for human Ascaris, of which 3 (genotypes G1–G3) were detected for pig Ascaris. The detection of more genetic variation than in some previous reports (Peng et al. 1998c; Zhu et al. 1999) was not unexpected, given the large numbers of Ascaris individuals examined and the use of the SSCP approach, which is highly sensitive to detecting nucleotide alterations (Gasser et al. 2002; Gasser, 2006). Of the five Ascaris genotypes detected, genotype G1 infected mainly humans (~63–74%), whereas genotype G3 infected predominantly pigs (~79–86%). Since the frequencies of the 3 other genotypes were substantially lower for the 2 host species, the results indicated that the rate of cross-infection of Ascaris between humans and pigs was low and that gene flow between the predominant genotypes (i.e., G1 and G3) was limited. These findings support other studies indicating that human and pig hosts tend to harbour a particular Ascaris genotype (Anderson et al. 1993; Peng et al. 1998c; Zhu et al. 1999), except in non-endemic regions (Anderson, 1995), and that gene flow between the Ascaris populations from humans and pigs is limited (Anderson, 1995; Peng et al. 2005). While it is not clear why pig Ascaris seems to be detected more frequently in humans in non-endemic than in endemic areas (Anderson, 1995, 2001; Nesjum et al. 2005), it may relate to a lack of exposure of humans to Ascaris eggs (of any host origin) and/or ease of sampling and detection (as infected individuals are more likely to consult a medical doctor).
The occurrence of particular genotypes of Ascaris in both the human and pig hosts has been interpreted to relate to one or more population genetic processes, such as introgression (=incorporation of genes of one species into the gene pool of another) or lineage sorting and retention of ancentral polymorphism (Anderson, 2001). However, the existence of genotypes of Ascaris (such as G2 and G5) which contain 2 distinct pITS-1 sequence types (i.e., with a G/C polymorphism at alignment position 133, consistent with a mixed profile in PCR-RFLP of ITS-1; Anderson et al. 1993; Peng et al. 1998c) had suggested that hybridization may occur between subpopulations of the human- and the pig-associated Ascaris. It had been proposed that interbreeding may occur between individuals representing different genotypes (e.g., G1 from humans and G3 from pigs) (Peng et al. 2003a), thus allowing variants of ITS-1 to be introduced and dispersed to a degree which exceeds the molecular ‘homogenisation process’. Alternatively, it is possible that different host species-associated populations have diverged genetically, whereby the processes of concerted evolution (Elder and Turner, 1995) have led to polymorphism at particular nucleotide positions and to sequence length variation. The latter explanation is plausible, given that Ascaris has been reported previously to contain distinct types of ribosomal DNA (Back et al. 1984a,b). Theoretical studies have suggested that DNA turnover mechanisms, such as gene conversion, transposition, unequal crossing-over or/and slippage during DNA replication are involved in the process of homogenization (Elder and Turner, 1995), and these mechanisms may apply to Ascaris, but would need to be evaluated.
Since no unequivocal nucleotide differences were detected in the genetic marker (pITS-1) between human- and pig-derived Ascaris (in naturally infected hosts), cross infection (between hosts) could not be reliably detected (Peng et al. 2003a). Hence, under these circumstances, the proposal for the existence of separate species of Ascaris (i.e., A. lumbricoides and A. suum) cannot be supported or refuted (Peng et al. 2003a). However, the significant frequency differences in the genotypes detected between human- and pig-derived Ascaris suggested that cross-infection occur at a low level and that they are reproductively isolated. These findings are consistent with those from previous molecular studies conducted, for example, in Guatemala (Anderson et al. 1995) and in China (Peng et al. 1998c, 2005) using nuclear or mitochondrial gene markers. Taken together, these findings appear to support the ‘two host-specialist parasite population with some cross- infection’ model, as proposed by Anderson (2001).
Experimental infections of pigs and mice with eggs from Ascaris of human or pig origin
Extending recent molecular and epidemiological studies (Peng et al. 2003a,b, 2005), experimental infections of pigs and mice with eggs representing particular genotypes of Ascaris were carried out in China (Peng et al. 2006). Ascaris of human origin (genotype G1) did not establish readily in experimental pigs whereas Ascaris of porcine origin (genotype G3) did (Peng et al. 2006). These preliminary findings are in accordance with those from a recent molecular-epidemiological investigation in China, showing that the prevalence of genotype G1 in pigs is very low (2·4%, n=329) and inferring a limited level of cross-infection of Ascaris between human and pig (Peng et al. 2003a). Although a small number of different egg batches from single Ascaris females representing the 2 genotypes were used in the infection experiments (Peng et al. 2006), the findings support the proposal that Ascaris of human origin does not readily infect pigs. This statement is reinforced by other results that Ascaris eggs of genotypes other than G1 (being consistent with genotype G2 from human Ascaris; see Peng et al. 2003a) do not establish as adults in pigs (using appropriate control groups) (Peng et al. unpublished findings). It is also supported by various molecular-epidemiological investigations (Anderson, 2001; Peng et al. 2003a, 2005). Nonetheless, further study, using more egg batches from single female worms representing the different Ascaris genotypes/haplotypes from each of the two host species is required.
CONCLUDING REMARKS
The research findings of recent molecular-epidemiological and experimental investigations in China (Peng et al. 2003a,b, 2005, 2006) support the hypothesis that there is a specific affiliation of ‘pig Ascaris’ to pigs and ‘human Ascaris’ to humans and a limited gene flow between the 2 main types of Ascaris, in spite of a long history of contact between the two host species. This information reinforces other evidence (Peng et al. 1998c; Anderson, 2001; Peng and Zhou, 2001) that, in endemic regions of China (and Guatemala), there is very limited cross-infection between human and pig or between pig and human, and provides support that pigs are not a significant reservoir of infections for human ascariasis. Interestingly, these results contrast evidence for non-endemic regions, such as Denmark and North America, where the cross-infection of Ascaris from pigs to humans can occur (e.g., Anderson, 1995; Nejsum et al. 2005). Recently, Nejsum et al. (2005) studied Ascaris from Danish patients, with a history of contact with pigs or pig manure, employing amplified fragment length polymorphism (AFLP) and PCR-RFLP analyses of a range of Ascaris samples from different countries. Together with epidemiological information, the findings supported the proposal that Ascaris infection in Danish patients was acquired from domestic pigs. While the reasons for the differential ability of the parasite to cross-infect and establish in the ‘heterologous’ host species in non-endemic but not so readily in endemic areas are unclear, they are likely to relate to host- and parasite-factors. For instance, host genes linked to susceptibility (Wakelin and Bradely, 2002; Williams-Blangero et al. 2002) and/or the immune status of the host (Cooper, 2002; Jungersen, 2002) may be associated with the ability of some Ascaris genotypes to establish preferentially in humans and others in pigs. In non-endemic regions, humans may not have been exposed previously to pig Ascaris, whereas in endemic regions (possibly due to the tradition of using pig excrement as a fertilizer) humans may have adequate immunological exposure to dead or viable porcine Ascaris eggs, thus blocking or inhibiting cross-infection. Also, susceptibility to cross-infection may be increased in hosts compromised by environmental or nutritional stresses, and/or concurrent infections with other infectious agents (Holland and Boes, 2002; Williams-Blangero and Blangero, 2002). Parasite factors could include differences in molecular processes governing the capacity of Ascaris to invade the ‘appropriate’ host. Another possibility is that the different genotypes of Ascaris induce varying immune responses via differentially expressed antigens (Abebe et al. 2002), and/or that the host defence and parasite invasion mechanism(s) or migratory routes differ in pigs compared with humans. The distinctiveness in recovery rates of worms from pigs and larvae from mice infected with eggs of genotypes G1 (‘human’) and G3 (‘pig’) (cf. Peng et al. 2006) suggests that investigations of the relationship between the genetic diversity of Ascaris and host susceptibility/resistance as well as immune responses will be worthwhile and informative. For example, the mouse model developed by Lewis et al. (2006) could be useful for the dissection of early host responses to experimental infection, given the limitation with reagents and cost associated with experiments in pigs. Another mouse model (Johnston et al. 2005) has been shown to be useful for exploring different adaptive immune responses against distinct isolates of Trichuris muris. At the molecular level, immunological responses and the parasite-host relationship of Ascaris could be explored by transcriptional profiling (temporal and spatial) utilizing a microarray platform, which could be extended to naturally infected humans of different ages groups and from non-endemic and endemic regions and, in the future (when the resources become available) to pigs experimentally infected with Ascaris of human or pig origin. With the major advances in genomic and proteomic expertise and resources, well-controlled comparative analysis of molecular differences between human Ascaris and pig Ascaris (employing a relatively large numbers of individual worms at the same and different stages of development, with biological replication) may also provide clues regarding the parasite-host interplay.
Support from the Wellcome Trust, the National Natural Science Foundation (WP), Elchrom Scientific AG and the University of Melbourne (RBG) is gratefully acknowledged. Thanks are also due to Xianmin Zhou, Guohua Peng, Ningyan Hu, Xingquan Zhu, Youssef Abs EL-Osta, David Crompton, Malcolm Kennedy and Celia Holland for their support, and to local governments and clinics for assisting in the collection of Ascaris in China. Two anonymous reviewers are thanked for their constructive comments.
