Members of the phylum Myxozoa Grasse , 1970 mainly infect fish. About 2300 species of Myxosporeans have been described so far [1]. It is known that the two largest genera of Myxosporeans (Myxobolus and Henneguya , including approximately 904 and 189 species respectively) are histologic [2,3]. They are commonly described on the gills of teleost fishes [2,3]. Molnar [4] reported that Myxosporeans have host, organs or tissues specificities. The tissue specificity is the most important. Knowledge related to the preferred gill area for the establishment of the Myxosporeans may facilitate the identification of the parasite species. Such knowledge is therefore relevant for the species description [5]. Only few authors have reported data on the spatial distribution of Myxosporeans species on the gills [6-8].

The original observation that some parasites have higher affinities for specific organs within the host was first reported by Cerfontaine et al. [9,10]. This observation has been greatly extended and refined. According to Dogiel [11], the host and its environment are the overall environment of the parasite. This finding is particularly approved as concerning the gill parasites which are in direct contact with the external environment of the host. The gills, commonly known as the most infected body part of the host by parasites are deemed a rather complex organ. Numerous authors have so far investigated the microhabitat of gill-living parasites [6-8,12-21]. Therefore, in the overall situation, the parasite species coexistence is studied in the context of site segregation [19,22]. Several authors have studied the spatial distribution of various Monogenea . They have reported some specificity for particular areas of attachment of these parasites by arbitrarily dividing each gill arch into several regions.

This study aim at investigating the spatial distribution of Myxobolus pethericii and Henneguya pethericii on the gills of Ctenopoma petherici.

Fish host Ctenopoma petherici Günther, 1864 (Anabantidae) were sampled in Wouri bassin on monthly basis during 15 months (January 2008-march 2009), from the Sange River at Ntonde (between latitude 4°12ˈ N to 4°17ˈ N and Longitude 10°0ˈ E to 10°8ˈ E), located in the Littoral Region of Cameroon. Fishes were captured using a 1 cm² mesh gill net. The collected sample were then preserved in a 10% formalin solution and transported to the laboratory (The University of Yaounde I) in a plastic container. In the laboratory, each specimen was dissected for the extraction of the gills. Both sides of the gills were examined with a stereoscopic microscope (Olympus Bo 61) to search for cysts. Myxosporidia cysts found on gill filaments or bony arch were counted and crushed between a slide and cover-slide and their content identified using an objective 100X of a Wild M-20 microscope. Parasitic species were identified according to Lom et al. [23]. Prior to the observation of the parasites location on the gills, arbitrary division of the gill arches was made according to Turgut et al. [24] modified. Therefore, the gill arches from each side of the fish were numbered I, II, III, IV from the anterior gill arch below the operculum to its posterior part. The surface of each hemibranch was designated as outer (i.e. that surface being the nearest to the operculum) and inner, and each hemibranch was divided into three approximately equal segments: anterior, medial and posterior. The bony part of the gill arch was also divided into three equal sections (Figure 1). The infection rate, the percentage (or occurrence) of infection and the cyst load were also evaluated as the parasites prevalence and intensity. Prevalence and intensity were defined according to Bush et al. [25].

Figure 1: The bony part of the gill arch.

Using the statistical package SPSS version 16.0, the data analysis was based on: (1) the χ² test for the comparison of the infection rates of M. pethericii, H. pethericii and M. pethericii + H. pethericii according to the side of the gill, the gill’s arches, the inner and outer hemibranches and the gill segments of C. petherici ; (2) the Kruskal-Wallis H test for the comparison of the mean cyst load of M. pethericii, H. pethericii and M. pethericii + H. pethericii between the gill’s arches and the gill segments; (3) The Mann-Whitney U test for the comparison of the paired mean cyst load of M. pethericii, H. pethericii and M. pethericii + H. pethericii depending on the side of the gill, the gill arches, the inner and outer hemibranches and the gill segments of C. petherici.

According to Bush et al. [25], the infection rate was estimated as the number of individuals of C. petherici infected with one or more individuals of a particular Myxosporean species divided by the total number of C. petherici examined. Referring to the mean intensity defined by Bush et al. [25], the mean cyst load was calculated as the average number of cysts of a particular species of Myxosporean among the infected members of C. petherici found in the sample divided by the number of C. petherici infected with that Myxosporean species. All statistical tests were considered significant at P<0.05.

A total of 364 specimens of Ctenopoma petherici were examined, among which 245 (67.3%) were infected by Myxobolus pethericii and 201 (55.2%) by Henneguya pethericii. From this host sample, 166 (45.6%) individuals harbored both M. pethericii and H. pethericii. As for the cases of mono infection, it appears that 79 (21.7%) specimens of C. pethericii harbored M. pethericii cysts alone while 35 (9.62%) individuals were only infected by H. pethericii. Among the examined fish specimens, 84 (23.08%) were free of parasites. A total of 31952 Myxosporeans cysts were found on the gills of examined fishes; among which 3446 (10.8%) cysts were of M. pethericii and 28506 (89.2%) were of H. pethericii. The mean cyst load was 144.7 ± 214.9 (3-1470) for the xenocommunity (M. pethericii + H. pethericii), 14.1 ± 23.6 (1-196) cysts for M. pethericii and 141.8 ± 216.4 (1-1464) cysts for H. pethericii . Parasites species studied were found among the above considered gill regions, their infection rate and the mean cyst load are respectively presented in Figure 2 and Table 1. No gill area was free of parasites.

Figure 2: The infection rate and the mean cyst load of Parasites species.

Table 1: Distribution of mean cyst load of Myxobolus pethericii and Henneguya pethericii on different parts of the gill apparatus in Ctenopoma petherici .

Data analysis showed no statistically significant difference in the variation of the occurrence and the mean cyst load of M. pethericii (χ²=0.204; P>0.05) and H. pethericii (χ²=0.022; P>0.05) between the right and left gill arches of C. petherici (Figure 2A). The branchial system of C. petherici is formed of four pairs of gill arches; but due to the absence of difference in the variation of the occurrence and the mean cyst load of parasites species between the right and left gill arches of C. petherici, only one set of gill arches has been considered in the following analysis.

With no significant difference, M. pethericii (χ²=7.163; P>0.05) presented preference for arches I and II meanwhile H. pethericii (χ2=2.032; P>0.05) preferred arches II and III (Figure 2B). The mean cyst load of M. pethercii decreased in the anterio-posterior direction without any significant difference (H=5.845; P>0.05). This pattern was not the same with H. pethericii. The median arches II and III harbored more cyst than arches I and IV but there was no significant difference (H=7.799; P>0.05) (Table 1). Our observations also showed that, for all parasite species with no significant difference, arch IV always harboured few cysts for all parasite species with no significant difference. At the xenocommunity level (M. pethericii + H. pethericii), the mean cyst load obtained on arches II (U=19171; P<0.05) and III (U=17778.5; P<0.05) are significantly higher compared to arch IV (Table 2).

Table 2: Comparison of the xenocommunity mean cyst load between different gills arches (U value).

Data analysis showed no statistical significant differences in the infection rate of M. pethericii (χ²=0.022; P>0.05) and H. pethericii (χ²=0.22; P>0.05) on outer and inner hemibranches irrespective of the gill arch (Figure 2C). On Arch IV, M. pethericii significantly (U=2064; P<0.01) encysted more on outer hemibranch (Table 3).

Table 3: Comparison of parasites mean cyst load between hemibranches for the same gill arch (U value).

Our observations showed that the medial segment of the gill was significantly more colonized by M. pethericii (χ²=18.598; P<0.01) and the combination made of M. pethericii + H. pethericii (χ²=6.689; P<0.05) (Figure 2D). The same result was obtained when comparing segments at the gill arch level (Table 4). There was a significant difference between the mean cyst loads of M. pethericii on different gills arches segments (H=6.8; P<0.05). A greater mean cyst load of M. pethericii was observed on medial segment. At the gill arch level, only arch IV show significant different (H=7.36; P<0.05) mean cyst load between segments, the medial segment harbouring more cyst than the others. In single infection with H. pethericii (H=1.67; P>0.05) and mixed infection (M. pethericii + H. pethericii) (H=2.556; P>0.05), the medial segment also harbour more cysts without significant difference.

Table 4: Comparison of rate of infection between different gills segments (χ² value).

Parasite specific richness of gill of C. petherici consisting of two different species of myxosporean (Myxobolus pethericii and Henneguya pethericii ) may be responsible of significant losses in river Sange. According to Combes [26], the pathogenic effect is rarely due to a single parasite species. The sum effect of all species of C. petherici gill myxosporean could be the cause of host morbidity and even mortality. The large number of cyst observed can be the consequence of accumulation of vegetative forms of myxosporean studied on the gills of C. petherici. These cysts, which are firmly attached to the fish gill epithelium, can release their infective spore only after the death of the host.

Myxobolus pethericii and Henneguya pethericii are distributed on the entire bronchial apparatus. The present study indicated that only weak competitive relations exist between the studied parasite species. Most probably there is a reciprocal tolerance between M. pethericii and H. pethericii. On the basis of the performed study, one cannot talk about separate ecological niches occupied by different parasite species [27]. Moreover Lom et al. [28] reported in myxosporean the paucity of inter-and intra-specific competition. According to these authors, the lack of competition would promote polyparasitism in hosts.

Myxobolus pethericii and H. pethericii affecting the gill of C. petherici could be responsible for major pathological changes (haemorrhagic foci, inflammations in the gill epithelium). Fomena [29,30] noted that in an advanced stage of infection, the plasmodia of such parasites can fully occupy the gill lamellae and cause epithelial dilation and hyperplasia. A pronounced dilation of infected gill lamellae can create pressure on the neighbouring lamellae causing their deformation and ultimately a merger. In massive infection by myxosporean cysts , reduction of the epithelial surface and compression of blood capillaries by these parasites can partially impair gill functions [31].

Studies on microhabitat distribution of myxosporean on the gills of their host. Data available are less abundant and are those of [6-8]. The majority of works on the distribution of gill parasites are related to Monogenea [6,7,12-21].

The degree of colonization of different gill zones by the myxosporean studied varied from one parasite species to another. Thus, preference of some regions was observed. Combes [26] noted that biotope heterogeneity creates a series of distinct microenvironments that are all habitats options for parasites species.

Differences between rate of infection and mean cystic load on left and right site of C. petherici were not statistically significant at xenocommunity and parasite species level. Similar results were obtained by Tombi et al. [7] on Myxobolus barbi and M. njinei gill parasites of Barbus martorelli in Cameroon, and Saha et al. [8] on Thelohanelus rohita, a gill parasite of Labeo rohita in India. We believe that, the equal distribution of myxosporean on both sides of the body of C. petherici would be the consequence of bilateral symmetry of the host. Furthermore, we agree with Saha et al. [8] that this could be due to the fact that similar volumes of water flowing through the left and right sides of the gill might have brought equal amount of actinospore stages to the gill. However, preference for fish side has been recorded with some monogenes. Preference for the right side was observed on Dactylogyrus amphibothrium [32] and Microcotyle mugilis while preference for the left side was observed concerning Metamicrocotyle cephalus [14] and Dactylogyrus valeti [21].

Our observations showed that arch IV was always less colonized by almost all parasite species. Few studies had been done to determine whether all of the gill arches play an equal part in gaseous exchange or whether more of the respiratory current passes over some gill arches than others. Considering the size alone, one might suspect that at least in most freshwater fishes the most posterior gill arch, number IV, receives less water flow than the anterior ones. Paling [33] described a single method of estimating the relative volume of water flowing over the different gill arches. He found that in brown trout, most of the respiratory current flows over the second and third pair of gills, less flows over the first pair and least of all across the most posterior pair of gills. In the absence of more sophisticated methods producing more accurate results, Paling's [33] findings serve useful functions in providing estimates of the different volumes of water flowing over the four pairs of gill arches. His findings, therefore, was adopted, particularly in view of Hughes [34] work indicating that the degree of infection of the gills is directly related to the ventilation volume and the pattern of current flow over the gills. As far as differences in the water current over the different parts of gill surface can be considered important in determining the distribution of parasites on the gills [20,32], the strongest water current passes through the middle part of the gill arches, thus creating convenient conditions for parasite settlements. The volume of the passing water may influence the aerobic conditions in certain gill parts, thus facilitating parasite settlement but also reflected the greater surface area available for parasite attachment on these gills [32]. This result might explain the present findings that the greatest mean cyst load of the xenocommunity occurred on the second and third gill arches. Myxobolus pethericii average cyst load reduced gradually on gill arches without any significant difference in the anterior-posterior direction. The same observation was made by Tombi et al. [21] on the distribution of Dactylogyrus amieti, a gill parasite of Barbus camptacanthus, which follows the variation of host filaments number. This filament number decreased significantly from arch I towards arch IV, the posterior arch (arch IV) which harboured the smallest number of filaments was least infected. Although slightly more H. pethericii cysts occurred on the second and third gill arches of C. petherici, the difference was not statistically significant. The results coincide with the findings of Tombi et al. [7] who found no statistical difference in mean cyst number of M. barbi and M. njinei between gill arches of B. martorelli.

In the present study, M. pethericii mean cyst load was statistically higher on the outer rather than the inner face of the arch IV hemibranch. Different observation was made by Saha et al. [8] on the distribution of Thelohanellus rohita on the gill of Labeo rohita. For this host species, posterior hemibranch of second gill arch was the most preferred site for parasite attachment. El Hafidi et al. [14] pointed that some monogenean species tend to attach to the inner hemibranch of the gill. On arch I, II and III, M. pethericii and were randomly distributed between the outer and inner hemibranches. This can be explained by the geometry of the gills that changes constantly during a single breathing cycle [35]; therefore, parts of the gill sieve are alternately exposed to and protected from the water flow.

A high occurrence of M. pethericii and the xenocommunity (M. pethericii + H. pethericii) on median segment of the gills arches was found in this work. Bychowsky [36] reported that the Monogenean Diplozoon paradoxum was predominant in the median sector of the gills. Similar preference was noted by Suydam et al. [37,38]. When studying spatial distribution of parasites species of the genus Dactylogyrus (monogenean) on the gills of the host fish, Turgut et al. [24] found a preference for specific regions of the gill arches. The author concluded that these specific preferences might be affected by the interaction of several factors such as differences in the hydrostatic pressure of the bronchial pump [39], coughing action [40], water current over the gill surface [32,33] during the respiratory cycle [41,42].

Animal use followed a protocol approved and authorized by Institutional Animal Care and Use Committee at Animals Biology and Physiology Department, Faculty of Science, University of Yaounde 1, Cameroon.

The authors are thankful to the Faculty of Science, University of Yaounde 1, Yaounde, Cameroon, for providing all the facilities to complete this work.

The authors declared: There is no conflict of interest.