Introduction

Assessment of environmental health with biota has become increasingly more attractive, and the number of identified indicator species has similarly increased. One of the most well-known examples of an organism’s usefulness as an indicator of the quality of an environment is that of canaries in coal mines. Historically, canaries accompanied miners into coal mines where the sensitivity of the birds to trace amounts of noxious gases, methane and carbon monoxide (Smith and Baker 2003; Holt and Miller 2010) was used as an early warning of toxic conditions to miners. As a result of their acute sensitivity towards the gases, exposure to low concentrations resulted in death of the birds which served as an unequivocal signal for miners to evacuate mining shafts (Smith and Baker 2003). Within aquatic ecosystems, a similar principle exists for determining the suitability of aquatic ecosystems in meeting a number of different services from environmental functionality to human uses, and with that, a number of organisms have been suggested as suitable indicators in this regard. Some of these include fish (e.g. Zhou et al. 2008; Squadrone et al. 2013), macroinvertebrates (e.g. Goodyear and McNeill 1999; Wah Chu and Chow 2002; Bird et al. 2011; Martins et al. 2011; Leonard and Wood 2013), bivalves (e.g. Kádár et al. 2006; Schintu et al. 2008; Minguez et al. 2011; Le et al. 2014) and more recently parasites (Sures 2001; Retief et al. 2006, 2009; Bayoumy et al. 2008; Sures et al. 2017).

Over the past 25 years, there has been growing interest into the interactions between parasites and the macroenvironment and with that the potential application of parasites as sentinel organisms for monitoring habitat quality and ecosystem health in relation to anthropogenic impacts (e.g. Khan and Thulin 1991; Šebelová et al. 2002; Sures 2004; Pečínková et al. 2005; Retief et al. 2006, 2009; Vidal-Martínez et al. 2010). Williams and MacKenzie (2003) suggested that the suitability of parasites as indicators stems from the variety of life stages in the life cycle of parasites and the variety of hosts involved in life cycles, some of which have been shown to be suitable indicators. Interest in the inter-relationship between parasites and the environment has appeared in a number of publications since the early 1980s (Sures 2001; Vidal-Martínez et al. 2010) where Khan and Thulin (1991) reviewed the interactions and effects of different pollutants on a number of different parasites. Later, Lafferty (1997) provided a detailed review and summary of literature pertaining to faunal differences of parasites towards pollution and the relationship between parasites and pollution. There are a variety of ways anthropogenic impacts result in environmental alterations which directly or indirectly affect parasites (Lafferty 1997; Galli et al. 2001). Responses of parasites towards environmental alterations have been analysed at species, population, behavioural and biochemical levels, and from this, such findings have deepened the suggestion of their suitability as bioindicators for monitoring the state of the macroenvironment (Sures 2001). Based on the responses displayed by parasites, they can be divided into two main indicator groups, namely effect indicators and accumulation indicators (Sures 2001, 2004). These aspects have been summarised in a number of reviews which deal directly with the usefulness of parasites as effect and bioaccumulation indicators (Lafferty 1997; MacKenzie 1999; Sures 2001, 2003, 2004 2005, 2008; Williams and MacKenzie 2003; Marcogliese 2004, 2005; Blanar et al. 2009; Vidal-Martínez et al. 2010; MacLeod and Poulin 2012; Le et al. 2014; Sures et al. 2017). Collectively, the assessment of environmental pollution using parasites has been termed “environmental parasitology” (Sures et al. 2017). In the most recent review on the subject by Sures et al. (2017), a summary of work from the past 20 years dealing with parasites as pollution indicators was presented and three main facets in the field were identified: “parasites as bioindicators”, “parasites as effect indicators” and “parasites interacting with established bioindicators”, and provides a holistic view of the advances made in the field at multiple levels of biological organisation relating to parasites (Sures et al. 2017).

Infrapopulation differences in parasite communities have been linked to changes in host population size and composition (Poulin 1995; Crafford and Avenant-Oldewage 2009; Watson et al. 2012; Saliu et al. 2014; Otachi et al. 2015), seasonal variances (Hakalahti and Valtonen 2003; Mbokane et al. 2015) and water quality (Khan and Thulin 1991; Broeg et al. 1999; Brito et al. 2014). Most studies conducted are related to effect indication by parasites, where changes in the infection biology and component community of a number of parasites in relation to differences in the chemical and physical constituents of aquatic environments have been assessed in relation to different types of pollution, whereas most bioaccumulation studies relating to parasites have focussed on the uptake of metals by these organisms with few investigations into organic pollutant accumulation.

The interest into metals could be related to their persistence in aquatic environments, their ability to accumulate within organisms and the array of deleterious effects associated with their presence in ecosystems (e.g. Avenant-Oldewage and Marx 2000; Le et al. 2014). Metals are naturally occurring and ubiquitous components of aquatic ecosystems, usually present in low concentrations (Förstner and Prosi 1979; Biney et al. 1994). The ecotoxicology of metals in aquatic ecosystems is complex (Wang 1987) and toxicity can be affected by changes in physical water quality variables, such as pH, alkalinity and hardness, and chemical variables such as carbon dioxide levels, total dissolved salts and chelating agents (Förstner and Prosi 1979; Biney et al. 1994). Furthermore, the variability in toxicity of metals in natural systems is linked to changes in the speciation of metals in the environment as a result of interactions with physico-chemical variables, some of which have been identified above (Hodson 1988). Furthermore, synergistic and antagonistic interactions between the metals present in aquatic systems can further affect their toxicity towards biota (Siegel et al. 1991; Tao et al. 1999; Zargar et al. 2012a, b).

In Africa as a whole and specifically South Africa, a large volume of studies have been produced which detail the accumulation of metals by aquatic organisms and the effects on the community structures; however, comparatively fewer studies have been conducted to investigate the usefulness of parasites as indicators and the effects of pollution exposure on population, organismal and physiological processes. This has, therefore, left a large gap in knowledge for the continent and country compared to the world. The aims of this review are to highlight the aspects of pollution identification using parasites in general and provide detail on the effects of pollution exposure at the component community level and the organismal level from studies across the globe and then comparing to the subject in context to Africa and South African examples.

Physico-chemical water quality variables

In order to properly ascertain the effectiveness of parasites as indicators, adequate information of natural fluctuations within component communities is required. The influence of seasonal fluctuations in abiotic factors (e.g. temperature) influences parasite populations and plays an important role in governing helminth infections in fishes (Khan and Rasheed 1993). Because of the complexity of biological systems and the host–parasite relationship, it is almost impossible to accurately determine or predict what consequence alterations in abiotic factors may have on parasite infections (Marcogliese 2016). The relationship between abiotic environmental factors and the occurrence of parasites in aquatic ecosystems has been addressed in a number of reviews (see Harvell et al. 1999, 2002; Hayes et al. 2001; Marcogliese 2001, 2008, 2016; Marcos-López et al. 2010; Lõhmus and Björklund 2015) under the umbrella term “climate change”. Overall, the interactions between parasites and the conditions of the macroenvironment are highly complex and relate to the life cycle characteristics of the parasite (see Anderson and Sukhdeo 2010). The interplay between biotic and abiotic factors influencing parasites has been demonstrated by Anderson and Sukhdeo (2010) who found that influences by abiotic factors operated secondarily to biotic influences on parasite populations. This finding was related to the effect that abiotic factors have on intermediate host occurrence, which had a greater effect on parasitic infections and therefore indirectly affected the dynamics of parasite populations (Zander and Reimer 2002; Marcogliese 2005; Anderson and Sukhdeo 2010). For the purposes of this review, the focus will be directed towards the effects of abiotic factors on parasites.

Of the abiotic factors analysed in studies and the relationship with helminth infections, temperature has been suggested to be the principal factor mediating infrapopulation differences of parasites either directly by affecting the parasite or indirectly affecting host abundance and distribution (Marcogliese 2001, 2016). Examples of direct and indirect effects of temperature on parasite life cycles have been presented in numerous reviews and provide details on effects towards development of life stages and host susceptibility towards infection (Harvell et al. 1999; Hayes et al. 2001; Marcogliese 2001, 2008, 2016; Harvell et al. 2002; Marcos-López et al. 2010; Lõhmus and Björklund 2015). Such temperature dependence and influence has been related to parasite incubation period which is temperature dependent (Paperna 1996). Seasonal fluctuations in temperature of aquatic ecosystems have been suggested to weaken the immune response of the host, making the host more susceptible towards infection (Altizer et al. 2006; Lamková et al. 2007). This trend has specifically been related to increases in infections by endoparasites such as cestodes and acanthocephalans. For example, infections by the Asian tapeworm (Granath and Esch 1983; Retief et al. 2007), Schyzocotyle acheilognathi [(formally Bothriocephalus acheilognathi) (Brabec et al. 2015)], have been shown to increase as water temperature decreases. Similarly, gyrodactylid infections have been found to peak when water temperatures are cooler (Gelnar 1991; Blažek et al. 2008), whereas the opposite has been observed in seasonal fluctuations of dactylogyrids, with infections peaking when temperatures increase (Šimková et al. 2001). In the acanthocephalan, Pomphorhynchus laevis, Nachev and Sures (2016) indicated that the mean weight of parasite infrapopulations differed seasonally, with significantly higher in summer compared to autumn and spring. Furthermore, they found that seasonal infrapopulation differences were related to different stages of maturation of adult worms, with young preadults being most prevalent in autumn and gravid adults in summer (Nachev and Sures 2016). Such seasonal change would not relate to alterations in the immune system of the host but rather the abundance of intermediate hosts in this case during the summer period (Rohde 1984). Thermophilicity has similarly been observed for diplozoid monogeneans (Höglund and Thulin 1989; Koskivaara et al. 1991; Koskivaara and Valtonen 1991) and has been suggested to result from the influence on reproductive success (Höglund and Thulin 1989). However, in a study on Paradiplozoon homoion, MacDonald and Jones (1978) found that egg laying in this parasite was not affected by temperature but instead correlated with photoperiod. Gravidity in Paradiplozoon ichthyoxanthon was further observed to increase in colder temperatures compared to warmer ones in the Vaal Dam (Gilbert and Avenant-Oldewage 2016a).

With regard to dactylogyrids, Koskivaara et al. (1992) suggested that temperature influenced this group in two ways: firstly, by directly affecting reproduction and development; and secondly, indirectly through changes in the immunological state of the host. Temperature effects on host immunology in relation to monogenean infections were analysed by Lamková et al. (2007) in the chub (Leuciscus cephalus), and they indicate that given the diversity of monogeneans, responses differ between the different groups. However, an adequate sample size is necessary to properly elucidate seasonal changes in parasite populations as le Roux et al. (2011) demonstrated for Cichlidogyrus philander; sample size greatly influenced infection statistics. Similarly, Crafford et al. (2014) and Mbokane et al. (2015) found that the prevalence of a number of Dactylogyrus sp. from fishes inhabiting the Vaal River and Limpopo River systems, respectively, fluctuated with season, but in all cases, the prevalence increased as the temperature increased with the season. In their studies, the authors found that the highest prevalence was present in either summer or spring seasons. Tsotetsi et al. (2004) and Austin and Avenant-Oldewage (2009) indicated that the variances in seasonality of Lamproglena clariae and Lamproglena hoi from the Vaal River and the Inkomati River and Pongola River systems, respectively, were similar to trends outlined for monogeneans, with prevalence of these parasites increasing with water temperature. For the monogenean, Afrodiplozoon polycotyleus, Chapman et al. (2000) and Raymond et al. (2006) indicated that temperature fluctuations, which related to rainfall and host habitat size, were factors influencing prevalence of A. polycotyleus with higher prevalence when temperatures were high.

However, changes in parasitic infections and population structures are not solely due to changes in temperature (Marcogliese 2008), and a number of factors have been suggested to operate in synergy with temperature to result in alterations. Such additional factors include changes in acidification, water level, eutrophication and stratification (Marcogliese 2001), together with anthropogenic intrusion (Marcogliese 2008) which will ultimately influence parasites (Marcogliese 2016). Salinity has been identified as another principal abiotic factor which affects parasite population structures. Several studies in both aquatic and marine ecosystems have indicated that changes in salinity affect infection biology of parasites (Harvell et al. 1999, 2002; Hayes et al. 2001; Marcogliese 2008, 2016). Generally, increases in salinity have been shown to negatively correlate with the population density and abundance of parasites (Soleng and Bakke 1997; Schmidt et al. 2003). Zander and Reimer (2002) found that parasite infections in the Baltic Sea were variable and correlated with the salinity gradient in this region. Salinity in the Baltic Sea decreases from west to east, and parasite distribution has been found to be limited along the salinity gradient (Zander and Reimer 2002). Although freshwater, marine and brackish water parasite taxa are located in this region, the distribution is spatially variable with few freshwater species being found in the western parts where salinity levels are high (Zander and Reimer 2002). Later, Jakob et al. (2008) observed that salinity dependence of 27 parasite taxa, infecting the European eel (Anguilla anguilla), reflected the habitat preference of the host for freshwater, marine and brackish water habitats with parasites clustering according to the habitat type and differences not being related to a salinity gradient between habitats. In their study, brackish sites investigated were similarly located in the Baltic Sea where both freshwater and marine parasites were found infecting eels and differences were not related to a salinity gradient at the site (Jakob et al. 2008). Sensitivity of freshwater parasites towards variances in salinity has been experimentally demonstrated for Gyrodactylus salaris infecting Atlantic salmon (Salmo salar) (Soleng and Bakke 1997). Soleng and Bakke (1997) observed that elevated salinity negatively correlated with survival of G. salaris, and therefore, the occurrence of the parasite was affected as salinity increased. In the German Bight, Schmidt et al. (2003) found that occurrence of copepod parasite Lepeophtheirus pectoralis corresponded to seasonal changes in salinity rather than temperature. Similarly, Blanar et al. (2011) indicated that the effect of salinity on occurrence of parasites on mummichog (Fundulus heteroclitus) in the Miramichi River was determined by a salinity gradient, with greater parasite infections being recorded for freshwater sites and lower intensities being recorded in areas with higher salinity. Poulin et al. (2011) also indicated in a comparison of the community ecology of metazoan parasites infecting stickleback (Gasterosteus aculeatus) from Eurasia and North America that structural differences in parasite communities were related to salinity and geographical distance.

Effects of temperature and salinity therefore correspond to the type of environment, where temperature fluctuations correspond to changes in the population ecology of freshwater parasites, whereas salinity differences are an important abiotic variable influencing the occurrence of parasites in coastal waters (Schmidt et al. 2003).

Effect indicators

Effect indicators are those taxa which provide information about their environment based on their presence or absence (Sures 2001; Williams and Mackenzie 2003). With regard to water quality, variations in parasite communities have been found to vary with pollution levels in aquatic systems (Khan and Thulin 1991; Lafferty 1997; Sures 2005; Blanar et al. 2009). Under field conditions, the exact variable resulting in changes is often difficult to identify given the large number of factors that need to be considered, as well as any synergistic and antagonistic relationships between variables. The effects observed are therefore likely the result of a number of responses of parasites towards pollution and natural environmental influences. Studies detailing the effect of pollution at the component community level of parasites from freshwater, marine and terrestrial environments from across the globe have been summarised in Table 1. From the table, it is clear that such analyses have been conducted on a diverse group of parasites, but most studies have focussed on differences in component communities of particular parasite groups, such as Monogenea (number of studies 33), Trematoda (number of studies 21), Nematoda (number of studies 16) and Protozoa (number of studies 15). Overall, under particular conditions, parasites are more sensitive towards exposure to pollution than their hosts. Comparatively greater research interest has been devoted to freshwater and marine environments, while very few examples of terrestrial host–parasite models have been analysed.

Table 1 Table summarising literature detailing the effects of different types of pollution on parasite infrapopulations for freshwater, marine, euryhaline and terrestrial ecosystems. Effects of pollution on component communities are classed as either increasing, decreasing or unaffected and were assessed according to variations in infection variables. Lowercase superscript letters indicate which host–parasite model was analysed based on the habitat type of the host (i.e. freshwater, marine, etc.)
Full size table

Changes in the macroenvironment can either indirectly affect parasites through the hosts or directly where the parasites are affected by changes in the environment (Rohde 1984). Direct effects of pollution on the host have been linked to increases in parasitism, specifically towards the effect on immune mechanisms of the host (Bagge and Valtonen 1996), as well as increases in population density of definitive and intermediate hosts (Rohde 1984). Differences in digenean infections, for instance, as a result of pollution in the Kymijoki water system have been linked to decreased population density of clam intermediate hosts due to intolerably high levels of pollution in reaches of this water system (Valtonen et al. 1997; Karvonen et al. 2005). Nachev and Sures (2009) also suggested that greater endohelminth diversity in barbel (Barbus barbus) at locations impacted by eutrophication could occur due to higher occurrence of tolerant intermediate hosts. Shah et al. (2013) also attributed differences in helminth infections to densities of intermediate hosts between study sites. However, exposure of parasites and their hosts to particular pollutants has been conducted in an attempt to elucidate the effects that chemicals have on the population structure and infection variables of parasites. Siddall et al. (1997) exposed Rutilus rutilus (roach) infected with dactylogyrids to effluent collected from Lake Vatia and observed that in all cases the abundance and mean intensity of the infections decreased as the effluent concentration increased. Similarly, brown trout exposed to heavy metals were found to harbour more cysts of Eustrongyloides than unexposed fish (Brotheridge et al. 1998). From a population perspective, ectoparasitic monogeneans are ideal organisms to study changes in infection dynamics in relation to water quality as is evident from the large number of studies involving this group of parasites (Koskivaara et al. 1992; Bagge and Valtonen 1996; Dušek et al. 1998; Zargar et al. 2012a). Such organisms often require a single host for the completion of their life cycle (in the instance of crustaceans, only the adult is parasitic), and all life stages are in constant contact with the macroenvironment to which they have had to adapt. As a result of their ectoparasitic nature, they are exposed to both the host and surrounding environment (Sures 2004). Alterations in the community structure of ectoparasites have been identified in relation to a variety of different types of pollution and include eutrophication (Koskivaara and Valtonen 1991; Valtonen et al. 1997), pulp and paper mill effluent (Khan et al. 1994; Siddall et al. 1997) as well as changes in physical water quality variables such as temperature (Höglund and Thulin 1989), pH (Baker and Cone 2000; Halmetoja et al. 2000; MacLeod and Poulin 2012) and salinity (Soleng and Bakke 1997; Kua et al. 2013). In a meta-analysis of the effect size of pollution on parasite infrapopulations, Blanar et al. (2009) found that the largest effect sizes were associated with exposure to pulp and paper mill effluent, metals, pesticides and acidification. Of the groups of ectoparasite taxa studied, they further identified that Monogenea were the most affected by exposure to the abovementioned pollutants, whereas members of Arthropoda did not show any significant negative effect (Blanar et al. 2009). Meanwhile, for endoparasite taxa, such as the acanthocephalans, cestodes and nematodes, Blanar et al. (2009) found that there was no significant relationship between exposure to pollution and the presence or absence of these taxa. Digenea were the only endoparasitic taxa to be found to exhibit negative effects in relation to PCBs, pulp mill effluent, metals and eutrophication. In monogenean communities, alterations at the population level occur in a concentration-dependent manner and are governed by the type of pollutant exposed to. Sensitivity of some monogeneans towards metals is supported by laboratory studies where gyrodactylids have been exposed to a number of different metals (see Poléo et al. 2004; Soleng et al. 2005; Gheorghiu et al. 2007). However, pollution sensitivity of monogeneans also appears to be related to the concentration of toxicants within aquatic ecosystems. Skinner (1982) indicated that infection of fish by monogeneans increased as the concentrations of nutrients increased. Khan and Kiceniuk (1988) found that gyrodactylid infrapopulations did not significantly differ between treated and control groups, but results of their study indicated that mean intensity of infections increased as crude oil concentrations increased. Exposure to low concentrations of paper mill effluent resulted in a proliferation in the infrapopulation of Dactylogyrus (Bagge and Valtonen 1996) due to exposure compromising the immune system of the host, as well as stimulating the production of large amounts of mucous creating a favourable environment for the parasites. Valtonen et al. (1997) found that the lower prevalence of parasites in Lake Vatia compared to other sites was due to extensive pollution by pulp and paper mill effluent. Hypoxia, eutrophication and humus content have also been shown to affect gyrodactylid infrapopulations (Koskivaara et al. 1991).

Tuuha et al. (1992) found that of the five ergaslid species infecting fish from an interconnected lake system in Central Finland, Ergasilus sieboldi infections were related to the condition of the lakes, which were highest in an oligotrophic section of the system compared to other sections. Infections by other ergaslids from the same system were not related to water quality or pollution but instead were dependent on seasonal fluctuations. In the same lake system, Valtonen et al. (1997) demonstrated a similar trend for ergaslids from the same fish species, with little deviation in parasite prevalence between the different sites. Recovery in parasite infrapopulations has also been found to correlate to improvements in water quality as pollution levels decrease (Valtonen et al. 2003). Siddall et al. (1997) found that experimental exposure of several species of Dactylogyrus to a concentration of 10% pulp and paper mill effluent was sufficient to reduce the abundance and mean intensity of the parasites. In the Kymijoki River system in the Gulf of Finland, Koskivaara and Valtonen (1991) found differential effects of pulp and paper mill effluent and eutrophication on the mean abundance of monogeneans, protozoan and copepod parasites infecting R. rutilus (roach). For the monogenean, P. homoion, they identified that exposure to paper mill effluent reduced the mean abundance, whereas exposure to eutrophication led to an increase in the mean abundance when compared to the infection dynamics in an oligotrophic lake within the same river system. However, for dactylogyrid and gyrodactylid monogeneans, and the protozoan, Ichthyophthirius multifiliis, exposure to both pollution and eutrophication resulted in higher mean abundances compared to the mean abundance of the parasite at the oligotrophic site, and for the copepod, Ergasilus briani, the mean abundance at the oligotrophic site was higher than at the polluted and eutrophic sites (Koskivaara and Valtonen 1991).

Kuperman (1992) found that exposure to metals, phenol, naphthalene and oil compounds resulted in a reduction in mean abundance and intensity of Diplozoon paradoxum. Dušek et al. (1998) further indicated that the responses of specialist and generalist monogeneans differ with exposure to pollution. Specialist monogenean assemblages were found to be significantly reduced with exposure to eutrophication, whereas the opposite was identified for generalist parasites, which were found to parasitise a range of different hosts (Dušek et al. 1998).

Metal sensitivity has been demonstrated as a possible treatment for a wide range of parasite infections in aquaculture and under laboratory conditions. The sensitivity of gyrodactylids has been shown for a number of different metals under laboratory conditions (Poléo et al. 2004; Soleng et al. 2005; Pettersen et al. 2006; Gheorghiu et al. 2007). Poléo et al. (2004) found that of the metals analysed, only Al and Zn significantly reduced the infrapopulation size of G. salaris in a concentration-dependent manner, whereas exposure to Fe, Mn and Cu did not affect the parasites. The effect of Al on the survival of G. salaris was confirmed by Soleng et al. (2005). Pettersen et al. (2006) tested the sensitivity of Gyrodactylus derjavini and Gyrodactylus macronychus and similarly indicated complete elimination of the parasite from host fishes exposed to elevated metal concentrations. Gheorghiu et al. (2007) suggested that a reduction in the community size and sensitivity of Gyrodactylus turnbulli towards elevated Zn concentrations could be related to the direct effect of the metal on the in utero daughter parasites as well as stimulating the production of mucous by the host which later prevented reestablishment of the parasite. Hagen et al. (2014) found that reduced survival of G. salaris exposed to sodium hypochlorite resulted from toxicity of Cl− ions towards the parasite.

Zharikova (1993) suggested that a reduction in the abundance of D. paradoxum exposed to Cu under environmental conditions would result from the effects of the metal on the survival of oncomiracidium larvae. Reduced survival in other larval parasites exposed to metals has been studied extensively in digeneans (Cross et al. 2001; Morley et al. 2001, 2003, 2004). A reduction in the survival and swimming performance of larvae of Cryptocotyle lingua exposed to Zn and Mn was indicated by Cross et al. (2001), which ultimately led to a reduction in the infection of definitive hosts. Exposure of adult Fasciola hepatica to Pb reduced survival up to 70% when exposed to 200 mg Pb/L. In addition, exposure of snail intermediate hosts to Cd has been found to result in transference of toxicity towards digenean larval stages through reduced encystment rate within snails (Morley et al. 2004; Pietrock et al. 2008). The effect of metal exposure on the emergence rate of cercariae of a number of digeneans from snail intermediate hosts has been studied under laboratory conditions (e.g. Morley et al. 2001, 2003, 2004, 2005a, b). Exposure to Cd was further found to affect the emergence rate of cercariae of Diplostomum spathaceum from snail intermediate hosts compared to Echinoparyphium recurvatum which did not display any difference in emergence rates (Morley et al. 2003). Morley et al. (2004) then indicated that cercariae of E. recurvatum prematurely encysted within snail intermediate hosts due to unfavourable conditions of Cd-exposed snails.

Following on from this, Khalil et al. (2009) observed that S. acheilognathi coracidia would only succumb to the toxicity of Cd once hatched from the egg. Even when exposed to concentrations as high as 10 mg/L, normal coracidial development within the egg was observed. This is contrary to the findings by Morley et al. (2001) who observed reduced hatchability of the eggs of the trematode Schistosoma mansoni when exposed to 10 mg Cd/L. The authors suggested that hatching inhibition as a result of metal exposure resulted from the effect of Cd on aminopeptidase activity which facilitates the hatching process (Morley et al. 2001). In light of the similarities between monogenean and digenean eggs, it is likely that a similar trend would be observed for monogeneans exposed to metals under laboratory conditions. However, in a study by Thoney (1990), hatching of the eggs of the monogenean Benedeniella posterocolpa exposed to Cu was unaffected and was similarly shown by Khalil et al. (2009) for eggs of S. acheilognathi which were unaffected until the eggs hatched. Similarly, Gilbert and Avenant-Oldewage (2016c) showed that exposure to elevated levels of aluminium negatively affected the survival of oncomiracidia of P. ichthyoxanthon only once the larvae had hatched from the egg and became exposed to the metal. Both studies therefore indicate that the egg shell is a very effective barrier which provides a high degree of protection towards the developing larvae.

Accumulation indicators

Comparatively fewer studies have analysed pollutant levels within tissues of parasites compared to other accumulation indicators. Furthermore, as indicated above, there has been extensive work conducted on effect indication by parasites for monitoring the quality of aquatic environments with fewer studies addressing toxicant accumulation in parasites. Most of the studies addressing this aspect in parasites have analysed the levels of a number of metals within the tissues of endoparasites, particularly acanthocephalans and cestodes (summarised in Table 2). The majority of studies have focussed on endoparasites, with few ectoparasite examples being analysed in this regard. From the volume of work conducted on this aspect, it is clear based on the ability of accumulate metals that acnahtocephalans and cestodes, followed by nemtodes and lastly digeneans are of particular interest. Acanthocephalans are particularly effective accumulation indicators given their ability to bioconcentrate metals at levels orders of magnitude higher than those in their hosts (e.g. Siddall and Sures 1998; Nachev et al. 2010), and to levels which are comparatively greater than those demonstrated in other intestinal helminths, therefore showing the highest accumulation potential of the parasites studied thus far.

Table 2 Table summarising results from studies investigating the accumulation of pollutants by parasites in freshwater, marine terrestrial and euryhaline ecosystems. Observations made in the studies detailed are based on the partitioning of elements within host–parasite models. Lowercase superscript letters indicate ecosystem type of the host (freshwater (a); marine (b); terrestrial (c); brackish/euryhaline (d))
Full size table

Further to this, parasites have been shown to affect the concentrations of metals in their hosts, where comparison between infected and uninfected hosts has identified lower somatic metal levels in the infected hosts compared to uninfected ones (Sures and Siddall 1999). This is believed to occur as both cestodes and acanthocephalans interrupt the hepatic–intestinal reabsorption of metal bound bile salts by the intestinal tract of the definitive hosts (Sures and Siddall 1999). Studies which have analysed the levels of metals in intermediate hosts compared to larval stages of the parasite found the opposite trend, where larval stages did not accumulate metals to higher levels than the invertebrate intermediate host and, in fact, the host had higher metal concentrations than the parasite (Sures 2006; Tekin-Özan and Barlas 2008; Frank et al. 2013). More recently, Le et al. (2016) further provided a model for metal uptake in host–parasite systems. The authors based this on the chub–acanthocephalan model (Squalius cephalus–P. laevis) exposed to Pb over a 36-day period. Nachev and Sures (2016) further indicated that the accumulation of trace elements by P. laevis differed between seasons. When temperatures were highest, accumulation of metals by this parasite was highest compared to lower temperatures. They related this finding to accumulation differing between different life stages of the parasite. Younger life stages, which were more abundant when water temperatures were cooler (autumn–winter), accumulated lower metal concentrations than older adults which were more abundant when water temperatures were higher (spring–summer).

Metal accumulation in cestodes and acanthocephalans has been found to be variable between segments of the strobila and male and female worms, respectively (Riggs et al. 1987; Sures et al. 1997a, b; Scheef et al. 2000). In both cases, it has been suggested that metals binding to the eggs of these parasites represent a regulatory function through which the parasites are able to detoxify themselves (Scheef et al. 2000; Degger et al. 2009; Khalil et al. 2009). Studies by Riggs et al. (1987) and Sures et al. (1997a) have indicated differential metal accumulation between body segments of the cestodes S. acheilognathi and Bothriocephalus scorpii, respectively. In both studies, metals were more concentrated in the posterior body segments compared to anterior segments. In the acanthocephalan, Moniliformis moniliformis, Scheef et al. (2000) identified higher concentrations of Cd in female worms compared to males and attributed this to the high number of eggs within females. This was similarly observed in two separate studies by Degger et al. (2009) and Khalil et al. (2009) for the cestode, S. acheilognathi. In the former study, positive fluorescence for metals was observed for the egg shells, whereas the latter detected Cd in the egg shells with the use of X-ray microanalysis (EDXMA). The sequestration of metals in the egg shells of these parasites is believed to occur through the enzymatic hardening or tanning of the egg shells whereby metals become incorporated into the shell matrix. This finding corroborated with studies where incorporation of metals into the egg cases of skates and dogfish (Koob 1991; Knight et al. 1996; Jeffree et al. 2008) was observed and suggested to occur during hardening of the case via enzymatic incorporation of catechol and oxidative cross linking with phenolic compounds (Degger et al. 2009). The involvement of the egg shell in removing metals from the bodies of adult organisms has also been indicated for a number of other invertebrates (Hook and Fisher 2001; Craig and Overnell 2003; Rosa et al. 2015) and vertebrates (Richards 1997; Blanco et al. 2003; Riggio et al. 2003; Brasfield et al. 2004).

In the case of ectoparasites, few studies have investigated metal accumulation in this group. One study (Gao and Nie 2000) attempted to analyse metal accumulation in the monogenean, Ancryocephalus mogurndae, which accumulated Pb at higher levels than the host. This indicates that like their endoparasite counterparts, monogeneans are able to accumulate metals at levels which surpass those in their hosts; however, further studies incorporating other members of this group are needed to confirm this suggestion.

In ectoparasites, studies on the elemental composition of the hard structures of the attachment organ have shown that some metals are incorporated into the structural components of these organs (Shinn et al. 1995). Shinn et al. (1995) indicated that elements such as V, Si, Na, Cl and P were present in different proportions within the sclerotised marginal hooklets, anchors and ventral bar of the haptor of Gyrodactylus. However, similar to the documentation of metal accumulation in ectoparasites, little is known about the sequestration mechanisms in this group. In invertebrates, a number of studies have indicated that metals are incorporated into sclerotised or hardened structures as a means of providing functional support to these organs.

Fewer studies documenting the accumulation of organic pollutants by parasites have been conducted and, therefore, with much research being devoted to the accumulation of metals in these organisms (e.g. Le et al. 2014). Le et al. (2014) reviewed the accumulation of persistent organic pollutants in parasites and indicated that lower accumulation potential for these compounds could be attributed to lower lipid content in the parasites compared to their hosts. Differences in organic pollutant content between infected and uninfected hosts have been related to the effect of the parasite on the physiology of the host (Sures 2006; Marcogliese and Pietrock 2011), which then affects the tolerance of the host towards exposure to organic pollutants (Heinonen et al. 2000; Le et al. 2014). Le et al. (2014) further suggested that differences in organic pollutant content between hosts and parasites could relate to the feeding mechanisms of the parasites and thus the trophic level of the parasite. Ectoparasites and endoparasites feeding on host tissue are therefore expected to have higher organic pollutant levels than endoparasites which feed by absorbing toxicants bound to compounds such as bile.

Biomarker responses in parasites

The toxic effects of pollutants have been studied for a number of aquatic organisms. Most studies involving the analysis of physiological responses of parasites towards pollution have dealt with metals. The degree of damage as a result of exposure differs with the nature and level of metals (Madoni and Romeo 2006; Wei and Yang 2015). Studies have associated metal exposure in organisms with the production of reactive oxygen species or intermediates (ROS/ROI), which result in oxidative stress (Hu 2000). A limited number of studies conducted on protozoan parasites and free living ciliates have documented the production of ROS in response to exposure to toxicants (Madoni and Romeo 2006; Mehta and Shaha 2006; Shukla et al. 2012). Mehta and Shaha (2006) observed that exposure of Leishmania donovani to metalloids, AsIII and SbIII, affected cells differently, and therefore, although the two metalloids are similar, resultant toxic effects manifest via two different mechanisms.

Other biomarker responses in helminths have been indicated for larval acanthocephalans (Sures and Radszuweit 2007) and nematodes (Chen et al. 2014; Keppel et al. 2014). In relation to metals, only Sures and Radszuweit (2007) have indicated a response in the production of heat shock protein 70 (Hsp-70) due to metal exposure. In this study, the authors observed that cysticanths exposed to metals produced higher levels of Hsp-70 than unexposed larvae, and further to this, Hsp-70 production in hosts infected with cysticanths was lower than uninfected exposed hosts. The production of Hsp-70 in nematodes, Anisakis spp. and Anguillicola crassus and Anguillicola novaezelandiae, respectively, has been observed in relation to increased temperature (Chen et al. 2014) and to the immune response of the host (Keppel et al. 2016), respectively.

Water quality, ecoparasitology and ecotoxicology in southern Africa

Water quality monitoring in South Africa has occurred since the 1950s (Roux et al. 1993; Claassens et al. 2016). Initial surveys were focussed on water supply for population growth and infrastructural expansion, but more recently, the focus has shifted to management and development following the inception of the National Water Quality Guidelines to assist in the determination of the suitability of water for all users (Department of Water Affairs and Forestry 1996). This was followed by the publication of the National Water Act in 1998 which relies on the integrated water resources management (IWRM) approach (Nomquphu et al. 2007).

Water resources in Africa are under threat due to continuing impact from anthropogenic pressures such as mining, agriculture, industrial activities and urbanisation (Wepener et al. 2011). Water quality monitoring methods have therefore evolved from traditional chemical analysis of water and sediment samples to biomonitoring (Roux et al. 1993; Crafford and Avenant-Oldewage 2010, 2011; Watson et al. 2012; Claassens et al. 2016), providing a time integrated analysis of the quality of aquatic ecosystems (Crafford and Avenant-Oldewage 2009; Watson et al. 2012). Such biomonitoring approaches have similarly incorporated a number of aquatic and marine organisms, such as fish (e.g. Mhlanga 2000; Wepener et al. 2011; Otachi et al. 2014), molluscs (e.g. Wepener et al. 2005; Cannicci et al. 2009; Mahmoud and Abu Taleb 2013), mussels (e.g. Degger et al. 2011a, b; Greenfield et al. 2011, 2014; Dahms et al. 2017), macroinvertebrates (e.g. Dickens and Graham 2002; Tate and Husted 2015), birds (e.g. Baker et al. 2016; van der Schyff et al. 2016), diatoms (e.g. Taylor et al. 2007; Harding and Taylor 2014) and in a few instances artificial mussels (Degger et al. 2011b; Greenfield et al. 2014; Claassens et al. 2016). More recently, attention has been devoted to exploring the usefulness of parasites from aquatic ecosystems (e.g. Retief et al. 2006, 2007; Oyoo-Okoth et al. 2010; Otachi et al. 2014; Torres et al. 2014; Gilbert and Avenant-Oldewage 2016b, c).

Incorporation of parasites as indicators in South African environments has been conducted at both population and individual levels. Populations of parasites have been incorporated into both pollution biomonitoring in freshwater and marine ecosystems (e.g. Crafford and Avenant-Oldewage 2009; Retief et al. 2006, 2007; Madanire-Moyo et al. 2012; Gilbert and Avenant-Oldewage 2016b, c; Morris et al. 2016), as well as indicators for monitoring sardine stocks, off the southern and western coasts of the country (e.g. Reed et al. 2012; Weston et al. 2015).

Overall, parasitological surveys in Africa have mostly focussed on parasite diversity while fewer studies have addressed the effects of pollution on these organisms in freshwater ecosystems (Retief et al. 2006; Olivier et al. 2009; Madanire-Moyo and Barson 2010; Madanire-Moyo et al. 2012; Gilbert and Avenant-Oldewage 2016b) compared to studies conducted in European rivers. Such studies are, however, sparse and predominantly localised to specific aquatic systems, namely the Vaal River and Limpopo River systems in southern Africa (e.g. Retief et al. 2006, 2009; Crafford and Avenant-Oldewage 2009; Madanire-Moyo et al. 2012; Gilbert and Avenant-Oldewage 2016b, c); terrestrial, aquatic and marine environments in northern Africa (e.g. Sures et al. 2003a, b; Torres et al. 2014; Abdel-Ghaffar et al. 2015; Abdel-Gaber et al. 2016); and aquatic ecosystems in Kenya (e.g. Otachi et al. 2014).

Following the development and implementation of the fish health assessment index (HAI), the parasite index (PI) was developed as an associated index to account for parasites associated with popular fish indicator species (Crafford and Avenant-Oldewage 2009). This index has been applied in riverine systems such as the Olifants River (Marx 1996; Watson et al. 2012) and the Vaal River (Crafford and Avenant-Oldewage 2009). Crafford and Avenant-Oldewage (2009) found that the prevalence of endoparasites and ectoparasites infecting the sharptooth catfish (Clarias gariepinus) was variable between two sites in the Vaal River. Endoparasites, overall, demonstrated higher prevalence at polluted sites compared to ectoparasites, which were higher in an unpolluted site. Crafford and Avenant-Oldewage (2009) therefore concluded that from such an analysis, it was possible to discriminate between study sites of differing water quality based on the presence or absence of parasites. The sensitivity of ectoparasites towards pollution levels within the Vaal River was later confirmed for a monogenean–host model, P. ichthyoxanthon infecting smallmouth yellowfish (Labeobarbus aeneus) where a local extinction of the parasites appeared to occur at a site below the Vaal River Barrage (Gilbert and Avenant-Oldewage 2016b).

Regarding accumulation of metals in parasites infecting South African fishes, a series of papers by Retief et al. (2006, 2009) detailed the usefulness of the cestode S. acheilognathi as an accumulation indicator for an array of trace elements in the Vaal River. Following this, Degger et al. (2009) determined with the use of fluorescence microscopy that the egg shells of the S. acheilognathi were sites where metals became sequestered and possibly indicated a mechanism for regulating the levels of metals in the tapeworms. Following from these studies, Morris et al. (2016) found that two endoparasites collected from sharks off the South African coast similarly accumulated trace elements 2–6 times higher than in the tissues of their hosts. For ectoparasites, the accumulation capacity of representatives of this group has yet to be examined. However, a study by Gilbert and Avenant-Oldewage (2017) indicated that metals accumulated by P. ichthyoxanthon collected from the Vaal Dam become sequestered in the vitellaria and sclerites of this parasite.

In other African countries, trace element and metal accumulation in numerous aquatic, marine and terrestrial host–parasite models have similarly been investigated and are mostly localised within North Africa and Central Africa. For terrestrial environments, a single study by Sures et al. (2003a, b) showed that the cestode, Hymenolepis diminuta, accumulated Pb to levels which were 97–36-fold greater than in the organs of the host, Rattus norvegicus. For the marine environment, Torres et al. (2014) studied the accumulation of Se and Hg in two fish hosts (Sarpa salpa and Balistes capriscus) and their respective digenean parasites (Robphildollfusium fractum and Neoapocreadium chabaudi) and indicated that levels of Se and Hg were higher in parasites than their hosts. Abdel-Ghaffar et al. (2015) similarly indicated that Fe, Cu, Cd and Ni were accumulated at higher concentrations by the nematode, Hysterothylacium aduncum, than the host fish sole (Solea solea) collected along the coast of Alexandria City, Egypt. In aquatic ecosystems, Otachi et al. (2014) found that only Zn was accumulated to higher levels in Contracaecum multipapillatum (L3) compared to the host and other trace elements were higher in host tissues than the parasites. Abdel-Gaber et al. (2016) similarly indicated that along with Pb, Zn, Cu, Mn, Cd and Ni were accumulated in P. clarias to greater concentrations than in the host fish, C. gariepinus, collected from Lake Manzala in the Nile Delta.

Marx (1996) indicated that the incorporation of parasites into the HAI accurately represented the condition of the environment and that parasite infections were lower when water quality deteriorated and this overall correlated with a higher HAI value for fish from the polluted sites along the Olifants River. In the Manyame River catchment in Zimbabwe, Madanire-Moyo and Barson (2010) found that parasite infections in C. gariepinus were related to water quality and exposure to sewage effluent. The levels of infection for all parasites were found to generally decrease as organic contaminant levels increased from the unpolluted site towards the polluted site. Variability was further identified between different parasites with pollution-tolerant species being dominant in polluted sites. The cestode Proteocephalus sp. was the only parasite present in the polluted site, and this could possibly relate to the eutrophic conditions favouring increased population density of intermediate hosts. Tylodephys sp. and Macrogyrodactylus sp. were pollution sensitive, and exposure to altered environmental conditions resulted in decreased infections along the length of the river in relation to levels of organic pollution content. Other parasites, such as Diplostomum sp., Caryophyllaeus sp. and Polyonchobothrium clarias, had highest prevalence in areas moderately affected by pollution, but as the levels increased, so the prevalence decreased similarly to those which are sensitive towards pollution (Madanire-Moyo and Barson 2010).

In the Limpopo River system, Madanire-Moyo et al. (2012) found that parasite infections in Oreochromis mossambicus were variably affected by water quality and exposure to sewage effluent at three dams in the Limpopo River and Olifants River systems. All parasites were found more prevalent on fish in the reference site compared to the polluted site, with the exception of gryporynchid larvae, which were more numerous at the polluted site (Madanire-Moyo et al. 2012). In the Olifants River, Watson et al. (2012) similarly indicated through the use of the HAI-PI that variations between infections by endoparasites and ectoparasites were related to water quality, with more endoparasites in impacted regions of the river and fewer ectoparasites. More recently, Dos Santos and Avenant-Oldewage (2016) indicated that the infection biology of a newly described diplozoid monogenean, from the two Labeo species inhabiting the Olifants and Selati rivers, Paradiplozoon krugerense, was negatively affected by decreased water quality and trace elements. No studies as of yet have investigated metal accumulation in parasites from fishes in this river system.

From the general trends outlined above for studies detailing changes in the infection biology of parasites in relation to water quality, it is evident that recent work has focussed on monogenean parasites in particular. In an experimental study, Gilbert and Avenant-Oldewage (2016c) indicated that exposure to aluminium reduced the survival of oncomiracidia of P. ichthyoxanthon, but only after the eggs had hatched and the larvae became completely exposed to the metal. This further suggested that the egg shell affords a high level of protection to the developing larvae and effectively prevents passage of foreign particles across the egg shell and into the egg. These results corroborate with the suggestion by Zharikova (1993) that changes in the abundance of monogeneans can be related to the sensitivity of the larval stages towards environmental contamination.

Conclusion

Recent studies have shown that parasites are affected by the condition of the macroenvironment, with effects between endoparasites and ectoparasites being variable. For this reason, it has been suggested in a number of reviews that parasites are suitable biological indicators, often responding to lower levels of perturbation which the host is not susceptible toward. Many studies have been conducted on a number of parasites from aquatic and marine ecosystems which have detailed the effects of pollution exposure at the component community level, while fewer studies have detailed the suitability of parasites as viable bioaccumulation indicators for monitoring levels of metals in the environment. In their meta-analysis of the effects of different types of pollution, Blanar et al. (2009) not only highlighted the sensitivity of different parasite taxa through variations in component communities towards pollution but also provided an account of the extent of the severity of exposure to numerous pollutants. Of the taxa assessed, they indicated that monogeneans and digeneans were the most sensitive. For both groups, sensitivity towards heavy metals have been tested and confirmed under laboratory conditions. Most studies have reported on the negative effects of pollution on parasite populations and decreases in infection variables in relation to water quality, and reports have also indicated that improvements in water quality can have a positive effect on parasite component communities. In such cases, decreases in pollution levels and improvement in water quality have been linked with increases in the infections by some parasites, whereas a decrease in other parasites associated with an improvement in the host immune response with decreased pollution levels was also indicative of improved water quality (Valtonen et al. 2003). From an accumulation perspective, parasites have been shown to be able to accumulate metals at very low exposure levels, which further supports their viability as accumulation indicators. The majority of these studies have been conducted on endoparasites, such as acanthocephalans, cestodes, nematodes and digeneans. Accumulation of metals by endoparasites have further been shown to be of benefit to the host, in that the parasites receive metals from their hosts, and in this way, infections diminish the somatic metal levels in the host tissues. In cestodes and acanthocephalans, it was further indicated that these organisms are able to regulate the levels of metals in their bodies through sequestration to the egg shells. It has been suggested that this process occurs during hardening of the egg shells during their formation. In cestodes, this has been shown to result in variable metal levels between the immature and gravid segments of the strobila and, therefore, further indicates that segments of the worms may be collected when they are released from the host and analysed for metals. In this way, the need to sacrifice the host would be avoided. Further study into the viability of this method of monitoring the levels of metals in the environment is however in need of validation.

Compared to Africa and South Africa, comparatively fewer investigations have addressed the effects of water quality on parasites. However, those that have, have indicated similar trends to those outlines in other studies. Furthermore, similarly to studies in host–parasite models in European river systems, parasites infecting African fishes are suitable indicators for monitoring the condition of the environment. The distribution of the studies conducted is sporadic and sparse; in South Africa for instance, these have only been conducted in two of the country’s major river systems. Furthermore, the focus of these investigations has mostly been at the population level, with three studies documenting the accumulation of metals in endoparasites from marine and freshwater ecosystems. However, despite the poverty in studies from the African continent detailing the effects of pollution on parasites, the general trends identified are similar to other studies from across the world. With that said, more intense investigation into the effects of pollution on parasites is required to adequately understand the processes occurring in these organisms in relation to the condition of the macroenvironment.