Characterization of the development of Haemonchus contortus ZJ strain from gerbils

Background Haemonchus contortus is a serious parasitic nematode in domestic ruminants around the world, including China. Haemonchus contortus has developed extensive resistance to commercial anthelmintics, which has produced a demand for new control methods, such as more effective drugs. Gerbils infected with H. contortus have previously been used as a model for anthelmintics selections, and the growth of H. contortus had been briefly examined. To enhance the model, this study provides an additional description of the development of H. contortus ZJ strain in gerbils. Results Gerbils were infected with H. contortus ZJ strain at a dose of 2000 exsheathed infective larvae (xL3s) and sacrificed at 4, 7 and 18 days post-infection (dpi). Only fourth-stage larvae were found in the stomachs. About 2% of the inoculums were obtained at each of the three sampling time points. Larvae grew more slowly in gerbils than in sheep, but presented almost the same morphology. Rod-like crystalline inclusions were present in the intestinal cells of larvae, indicating that the metabolic rate of larvae was probably greatly reduced. Histological examination of stomach sections showed that larvae are located in the lumens or at the mucosal surfaces, with few inflammatory changes evident. Conclusions The development and features of H. contortus ZJ strain in gerbils were described. Our results provide supplementary information of H. contortus growth in gerbils, especially the presence of rod-like crystalline inclusions, and may contribute to improve the anthelmintic selection system.


Background
The blood-feeding gastric nematode Haemonchus contortus is a serious pathogen in ruminants all around the world, including China. Drug resistance and lack of effective vaccines bring difficulties to the control of the disease. Previous studies have screened potential anthelmintics utilizing a H. contortus-infected gerbil model.
Conder et al. [1] first described the model for studying anthelmintics against H. contortus, and examined the growth and development of H. contortus in gerbils [2]; this study found that artificially exsheathed larvae were able to grow and establish successfully in gerbil's stomach. However as compared to those in sheep, the larvae in gerbils exhibited slower and incomplete growth. De Jesus-Gabino et al. [3] and Suires et al. [4,5] also tested the anthelmintic effects of different plant extracts on H. contortus in gerbils. In addition, other laboratory animals, such as mice, have been utilized to set up laboratory screening systems for anthelmintic activity against H. contortus [6,7].
The slower and incomplete growth of H. contortus in gerbils was also observed as described in other laboratory animals. Hutchinson & Slocombe [8] found that H. contortus completed its development at a slower rate than in sheep when they infected laboratory rabbits with exsheathed larvae. Even though a few female worms were pregnant, a high proportion of early fourth-stage larvae were recovered and some of them contained rod-like crystalline inclusions. Wagland et al. [9] demonstrated that H. contortus stayed and developed into fourth-stage in guinea pigs. Mature females, earlier stage larvae and fertile eggs were found in Chinchilla [10].
Usually, sheep were chosen as a suitable host for H. contortus research [11], but this procedure is expensive and time-consuming. A convenient and cost effective in vivo gerbil model is urgently needed. To improve the model, this experiment details the developmental morphology of H. contortus ZJ strain in gerbils and provides additional description of the growth.

Necropsy of gerbils
All gerbils were euthanized and necropsy was carried out to obtain the stomachs. Six of the nine stomachs from groups 3-5 were used for larvae collection. Stomachs were digested in peptic-HCl and washed by phosphate-buffered saline (PBS) [11]. The number of larvae was counted with an anatomical lens (Motic, Fujian, China). Larvae were fixed in 7% formol saline and lengths (n = 10) were measured with a confocal laser scanning microscope (Olympus, Tokyo, Japan). Stomachs from groups 1-2, and three of the nine stomachs from groups 3-5, were directly fixed in 10% neutral buffered formalin for histological study.

Histological examination of stomach
Each of the fixed stomachs was embedded in paraffin, mounted on glass sides and stained with hematoxylin-eosin (HE). Stained tissue sections were examined with an optical microscope (Olympus, Japan).

Statistical analyses
The number and length of larvae were shown as means ± SD. A one-way ANOVA was conducted for length comparison between different groups.
Larvae recovered at 4 dpi displayed obvious sexual differentiation and could be identified as males and females. A swelling containing the vestigial copulatory bursa with short spicules was found at the posterior extremity of male larvae (Fig. 1a), but just a tail in the females (Fig. 1b). The well-developed excretory pore (Fig. 1c) and buccal capsule (Fig. 1d) were very prominent at 4 dpi, as was the genital primordium ( Fig. 1e). Some differentiation into functional entities, such as the vulva ovejector primordium, had occurred in the genital primordium at 7 dpi (Fig. 1f ). One confusing finding was the appearance of a mass of strange tissues below the cuticle. The tissues could be seen in Fig.1b-f. Most of them were helically pipeshaped, and some were vesicle-shaped. The exact role of the specific subcutaneous substance was unknown and warrants follow-up studies.
The rod-like crystalline inclusions were present in H. contortus after 7 dpi (Fig. 2) and could be found in almost all of nematodes. The crystals were located in the intestinal cells of larvae and were fairly uniform in their rod-like shape, but varied considerably in their sizes. In some cases, two seemingly identical crystals were found parallel to each other with a narrow space. Moreover, a small number of crystals were able to split in two, giving the appearance of a Y shape. The number of crystal was  found to increase with infection duration. In addition, the intestinal cells were also found to be solidly packed with crystals at 18 dpi (Fig. 2d). Upon histological examination, larvae were found, in the form of cross-sections with the intestinal lumen inside, in the gerbil lumens and mucosal surfaces of stomach (Fig. 3). The presence of eosinophil infiltrates and lymphocytic aggregates, or obvious inflammatory changes, was barely detected.

Discussion
Previous studies have made great efforts to search for a laboratory model to take the place of a ruminant model when conducting in vivo H. contortus infection experiments. Gerbils infected with H. contortus were first used as an anthelmintic screening model [1]. This model provided useful and credible information about hostparasite interactions. Rodent models, not just gerbil, including mouse, hamster, chinchilla and rabbit, could improve discovery efficiency, which means more work could be done prior to being completed in ruminants. Small laboratory animals offer convenience in feeding and handling, as well as lowering the research cost.
The amount of H. contortus recovered from gerbils can reach up to more than 30% of the inoculums [2]. However the larvae that survived in our experiment accounted for only a small percentage (2%) of the inoculums. The factors causing this difference might be various; the inoculation dose of larvae is probably the most important one. Conder et al. found that inocula of 500-2000 xL3s resulted in relatively constant worm recoveries, which kept 5-30% of the inoculums, and a dose of 1000 larvae was chosen for further work [2]. Squires et al. [4,5] gave 600 xL3s to gerbils and about 10% larvae recovered. But unlike the above two experiments, De Jesus-Gabino et al. [3] administered 4000 sheathed larvae to each gerbil, and the recovery rate was just 1.6-2.0%. The relationship between infection dose and recovery rate seems to be unpredictable and occasional. Other factors, such as larval exsheathment method, sex and age of gerbils, the individual difference of host and environmental factors, were all other possible factors. Further study is required to enhance the recovery rate of H. contortus.
According to the life-cycle described by Veglia [13] (cf. Georgi & Georgi [14]), once ingested by the appropriate host, a ruminant for H. contortus, the L3 exsheath and grow to early L4 at~4 dpi, and then L4 at~7 dpi. Specialized development of the buccal capsule for blood-feeding and sexual differentiation is available at the fourth stage, following by gametogenesis in the adult stage. Female adults lay eggs at~18 dpi, following copulation with males. The body length of larvae from L3 to adult could be~1 mm to~15-18 mm (male) or to~25-30 mm (female). In this experiment, the basic tissue differentiation and body length of larvae in gerbils was similar to that of normal larvae in sheep at 4 dpi. However larvae recovered from gerbils at 7 and 18 dpi were significantly shorter than those in sheep at the same time point [2]. This was probably induced by host differences, since the gastric conditions, including pH, gas composition and content, of gerbil and sheep are very different [15][16][17][18]. Even the growth extent was much lower than that found by Conder et al. [2], which also described a gerbil model. Larvae grew to the length of 1100 ± 169 μm, 1273 ± 122 μm and 1553 ± 133 μm at 4, 7 and 18 dpi, respectively, in our study, but grew to 1320 ± 333 μm, 2055 ± 290 μm (male) and 5415 ± 3589 μm (male) at 4, 7 and 21 dpi, respectively, in the study of Conder et al. [2]. This could be explained by a different parasite strain and housing environment.
Haemonchus contortus can undergo arrested development at the fourth stage in sheep abomasum when threatened by host or environment [19,20]. Arrest also refers to diapause in H. contortus. Diapause is defined by two unique characteristics [11,15]. First, it is induced by environmental cues, often seasonal changes, and the arrest is temporarily irreversible. Secondly, the body length of diapausing larvae is short and uniform. Resumption of development will not proceed unless diapause development has been completed and environmental conditions are favorable. In H. contortus, diapause ensues when infective larvae (iL3s) undergo development for 3 or 4 days to reach the early fourth-stage after host infection [21]. Larvae recovered from gerbils in our experiment were not uniform, although they were short. The most important point was that larvae became longer with increased infection length. Thus, they just grew at a much slower speed in an inappropriate host, rather than enter into diapause.
Another symbolic feature of diapausing H. contortus is the presence of rod-like crystalline inclusions in the intestinal cells [21]. While these inclusions can still be found in the adult stage when the diapausing larvae have resumed development, they were not detected in normally developing larvae. In this study, the rod-like crystalline inclusions appeared in the intestinal cells at 7 dpi and were present from this time on. This is a new discovery which was absent in the study of Conder et al. [1,2]. Earlier work has reported the presence of protein and sulphur in the crystals [15], while our recent DAPI staining results revealed positive nucleic acid staining [11]. Although the significance of these compounds remain unknown, it is possible that they were built up metabolic waste products generated during the process of development [21,22]. Colglazier et al. [23] supported this theory, in which diapausing H. contortus conferred great resistance to most anthelmintics including thiabendazole. As such, it is reasonable that the number of crystals increased with infection days, as the metabolic rate and energy requirements were probably greatly reduced in slowly developing H. contortus. Rodlike crystalline inclusions were also present in H. contortus recovered from rabbits, but without any explanation to this phenomenon [8].
Since larvae were mostly located in the lumens or on the stomach mucosal surfaces, eosinophils or other inflammatory responses were not obvious in our results. We performed routine peripheral blood examination after gerbil infection, and found few changes in the numbers of red blood cells, eosinophils or lymphocytes (data not shown). This indicated that, unlike sheep, gerbils infected with H. contortus failed to establish a significant systemic immune response, or even a local reaction. It is possible that H. contortus were being regarded as food items and subsequently digested in non-routine hosts. Only the most adaptive larvae survived, but they failed to develop to adulthood. In addition, it is possible that the small number (about 2% of the inoculums in our study) of larvae was insufficient to elicit effective immune responses. Gerbils were immunosuppressed by dexamethasone, and therefore this may also affect their immune responses. A higher larval dose would be required to stimulate a local reaction [2].

Conclusions
The present study has successfully obtained fourth-stage H. contortus larvae in a gerbil infection model. Morphological characterization of the larvae revealed shorter larval length but almost the same appearance, which suggested that slower development, rather than arrested development, has taken place in gerbils, compared to in sheep. The presence of rod-like crystalline inclusions in nematode intestinal cells was found, indicating that slowly developing H. contortus had a reduced metabolic rate. Our results provide an additional description of the development of H. contortus ZJ strain from gerbils and may contribute to enhance the anthelmintic screening model.