Giardia lamblia is a causative parasite of Giardiasis. This disease condition is highly prevalent in human especially children and malnourished individuals [56, 57]. Currently, metronidazole is the drug of choice in the treatment of giardiasis. However, its therapeutic efficacy against giardiasis decreased due to the developing of resistance [58]. Here, we sought to investigate the effect of Artemisia annua on Giardia lamblia-infected hamster and compare its effects on the small intestinal lesion to the effect of metronidazole administration.
Parasitological analysis:
Oral application of ten thousand Giardia lamblia cysts/animal led to the establishment of giardiasis in hamsters. This was confirmed by finding Giardia lamblia cysts in feces of infected animals. Interestingly, the examination of small intestine belonging to metronidazole and Artemisia annua ethanolic extracts-treated groups for trophozoite count revealed that these animals had significant reductions in trophozoite count compared to the infected untreated group. Importantly, treating animals with Artemisia annua significantly reduced the trophozoites count compared to animals treated with metronidazole (98.3 % and 92.5%, respectively) (Table 1 and Fig. 1). These data suggest that Artemisia annua extracts were more effective in reducing Giardia lamblia trophozoite count in the small intestine than metronidazole. (Reduction of trophozoite number in Artemisia annua-treated group and metronidazole treated group was statistically significant (P<0.001)
Biochemical analysis:
Nowadays, the role of nitric oxide (NO) in the immune response to giardia infection has been well established. Therefore, we sought to investigate the serum levels of NO end-products (NOx) in animals infected with Giardia lamblia and treated with either metronidazole or Artemisia annua. In infected untreated animals, the serum levels of NOx were comparable to uninfected animals. In contrast, the treatment of infected hamsters with metronidazole had significantly elevated serum NOx levels compared to infected untreated animals (p<0.001) and uninfected control group (p<0.001). Similarly, infected animals treated with Artemisia annua extracts had their serum NOx levels significantly increased compared to infected untreated animals (p<0.001). However, we did not find a significant difference between the effect of metronidazole and Artemisia Annua extracts on serum NOx levels (Table 2 and Fig. 2a).
We next examined the serum interleukin-6 (IL-6) levels. Previously, Zhou et al. [19] had found increased IL-6 levels during the course of giardiasis in mice. Therefore, we sought to compare the effects of metronidazole and Artemisia annua on giardiasis-induced increase in IL-6 serum levels. Serum samples from animals infected with Giardia lamblia showed a significant increase in IL-6 levels compared to uninfected animals (P < 0.05). In contrast, treatment of giardia infected animals with metronidazole had significantly decreased serum IL-6 levels. Similarly, treatment of infected animals with Artemisia annua had significantly reduced serum IL-6 levels compared to untreated infected hamsters (P < 0.05). By comparing the serum IL-6 levels after treatment with metronidazole or Artemisia annua, we observed that metronidazole significantly decreased the IL-6 serum levels compared to Artemisia annua (P < 0.05) (Table 2 and Fig. 2b).
A recent study by Pacheco and colleagues [59] has demonstrated elevated serum gamma interferon (IFN-γ) in children infected with giardiasis. Therefore, we sought to compare the effect of metronidazole and Artemisia annua on giardiasis-induced elevated serum IFN-γ levels. As expected, animals infected with Giardia lamblia showed significant increase in the serum level of IFN-γ compared to the uninfected animals (P<0.05). On the other hand, the serum IFN-γ levels were significantly lower in hamsters treated with metronidazole than infected untreated animals (P< 0.05). In animals treated with Artemisia annua, the serum IFN-γ had significantly reduced compared to the infected untreated animals (P < 0.05). These levels were higher than the serum IFN-γ levels in animals infected and treated with metronidazole (Table 2, Fig. 2c).
We also investigated the serum tumor necrosis factor (TNF)-α levels in infected animals treated with metronidazole or Artemisia annua extracts. As shown in table 2 and fig. 2d, serum levels of TNF-α in infected untreated animals had significantly elevated compared to the uninfected animals (P< 0.05). In contrast, treating infected animals with metronidazole or Artemisia annua extracts had significantly reduced serum levels of TNF-α compared to the infected untreated hamsters by 4 and 3 folds, respectively.
Histopathological analysis:
To further explore the effects of metronidazole and Artemisia annua on giardiasis-induced small intestinal lesions, we stained sections taken from the proximal parts of the small intestine with hematoxylin and eosin stains. Small intestinal sections taken from uninfected animals showed normal intestinal villi and crypts (Fig. 3a, arrowheads, and arrows). In contrast, marked shortening and destruction of intestinal villi together with retraction of their connective tissue cores were observed in small intestinal sections taken from animals infected with Giardia lamblia cysts (Fig. 3b, arrowheads, and stars). On the other hand, treating infected animals with metronidazole or Artemisia annua restored the structure of the villi and their connective tissue cores (Fig. 3c and d, arrowheads, and stars). We measured the villi length in pictures taken with 10x objective lenses using ImageJ software. As demonstrated in fig. 4, significant reduction in the villi length was observed in sections taken from infected untreated animals compared to the uninfected animals (P<0.001). In contrast, treating infected animals with metronidazole or Artemisia annua had restored the normal villi length (P<0.001). We did not find a significant difference between the effects of metronidazole and Artemisia annua on the villi length.
At the higher magnification figure, small intestinal sections taken from uninfected hamsters exhibited normal histological structure of the villi. They are covered with simple columnar epithelial cells (Fig. 5a, arrowheads) and goblet cells (Fig. 5a, asterisks). Few intraepithelial lymphocytes were observed in section taken from these animals (Fig. 5a, arrows). The core of the villi is formed of connective tissue (Fig. 5a, stars). On the other hand, infecting animals with Giardia lamblia cysts lead to desquamation of epithelial cells, disruption of the epithelium covering the villi (Fig. 5b, double-headed arrows), increased intraepithelial lymphocytes (IEL) (Fig. 5b, arrows) and disintegration of the connective tissue core of the villi with few scattered connective tissue cells in this core (Fig. 5b, stars). We also observed the Giardia lamblia trophozoites in the intervillous spaces (Fig. 5b, arrowheads). However, tissue sections taken from infected animals treated with metronidazole showed normal covering epithelium of the villi except of few areas of epithelial disruptions (Fig. 5c, double headed arrow), regeneration of the connective tissue core of the villi (Fig. 5c, stars) and decreased intraepithelial lymphocytes infiltration (Fig. 5c, arrows). Additionally, few Giardia lamblia trophozoites were observed in the intervillous spaces (Fig. 5c, arrowheads). More specifically, small intestinal sections taken from infected animals treated with Artemisia annua extracts showed preserved architecture of the epithelial covering the villi, few desquamated cells (Fig. 5d, double headed arrows), few IEL (Fig. 5c, arrows) and well-preserved villous connective tissue core (Fig. 5c, stars). Few Giardia lamblia trophozoites were shown in the intervillous regions (Fig.5c, arrows).
We counted the IEL per 100 epithelial cells on images of hematoxylin and eosin-stained sections taken at 200x magnification. A total of 2500 epithelial cells were counted in each group according to previously published methods [55]. As shown in fig. 6, infecting hamsters with Giardia lamblia cysts significantly increased IEL number/100 epithelial cells compared to uninfected animals (P<0.001). However, treating infected animals with metronidazole or Artemisia annua significantly decreased the IEL number/100 epithelial cells compared to infected untreated animals (P<0.001). We did not find a significant difference between the effect of metronidazole and Artemisia annua on the number of IEL/100 epithelial cells.
To better demonstrate goblet cells, we stained our small intestinal sections with Periodic acid Schiff reagent (PAS) and hematoxylin. In uninfected animals, goblet cells are distributed among the epithelial cells in the epithelium covering the villi (Fig. 7a). Few goblet cells were observed in sections taken from infected untreated animals (Fig. 7b). However, treating infected gerbils with metronidazole or Artemisia annua reversed Gardia lamblia-induced reduction of goblet cells (Fig. 7c and d).
To confirm our observations, we evaluated the number of goblet cells/100 epithelial cells in images of small intestinal sections stained with PAS and hematoxylin taken at 200x magnification. A total of 2500 epithelial cells per group were counted according to previously published methods [60]. As demonstrated in fig. 8, infection of hamsters with Giardia lamblia cysts significantly reduced goblet cell number compared to the uninfected animals (P<0.001). On the other hand, treating infected animals with metronidazole or Artemisia annua had significantly increased the goblet cell number compared to the infected untreated animals (P<0.001). No significant difference was found between the effect of metronidazole and Artemisia annua on goblet cell number.
Earlier, Pavanelli et al. [18] have demonstrated that giardiasis reduced the muscularis externa thickness. We sought to investigate whether the treatment of infected animals with either metronidazole or Artemisia annua would reverse giardiasis-induced muscularis externa changes. In sections taken from the small intestine of uninfected hamsters, the muscularis externa is composed of smooth muscle fibers arranged in two layers, inner circular and outer longitudinal layers. These cells have acidophilic cytoplasm and elongated vesicular centrally located nuclei (Fig. 9a). As expected, small intestinal sections taken from infected untreated hamsters showed extensive vacuolation of muscle cells of the muscularis externa (Fig. 9b, arrows). Additionally, the thickness of the muscularis externa of these animals appeared reduced. In contrast, small intestinal section taken from infected animals treated with metronidazole or Artemisia annua showed decreased cytoplasmic vacuolation of smooth muscle cells (Fig. 9c and d, arrows) and the muscularis externa layers appeared thicker than those of infected untreated animals (Fig. 9c and d).
To confirm our results, we measured the thickness of the muscularis externa using ImageJ software in 15 pictures of random fields taken at 400x magnification. As shown in fig. 10, infection of hamsters with Giardia lamblia significantly reduced muscularis externa thickness compared to the uninfected animals (P<0.001). On the other hand, treating infected animals with metronidazole or Artemisia annua extracts significantly increased the muscularis externa thickness compared to the infected untreated animals (P<0.001). No significant difference was observed between the effects of the two treatments.
Immunohistochemical analysis:
A large body of evidence has associated inducible nitric oxide synthase (iNOS) expression and Giardia lamblia infection [28, 61]). To investigate whether treating infected hamsters with metronidazole or Artemisia annua would modulate the expression of iNOS, we employed the immunohistochemical technique to examine iNOS expression patterns in small intestinal tissues. In uninfected animals, we showed iNOS immunoreactivity in connective tissue cells in the core of the villi (Fig. 11a, arrows) and rarely in the enterocytes. Infection of hamsters with Giardia lamblia trophozoites caused intense iNOS signals in enterocytes covering the villi (Fig. 11b, arrowheads). In animals infected with giardia trophozoites and treated with metronidazole, moderate iNOS immunoreactivity was observed in villi-core connective tissue cells (Fig. 11c, arrows) and enterocytes (Fig. 11c, arrowheads). In contrast, tissues from the small intestine of animals infected with giardia trophozoites and treated with Artemisia annua extracts showed few cells of the villi core expressed iNOS (Fig. 11d, arrows) and barely seen signals in the enterocytes (Fig. 11d, arrowheads).
Previously, it has been demonstrated that Giardia lamblia trophozoites exhibited an apoptotic effect on cultured enterocytes [17]. Therefore, we sought to test whether treating infected hamsters with Artemisia annua extracts would attenuate giardia-induced apoptosis. We immunostained our small intestinal sections with anti-caspase-3 antibodies. In uninfected control animals, weak caspase-3 signals were observed in the enterocytes, lamina propria cells (Fig. 12a, arrows), crypt cells and muscularis externa layers (Fig. 12b, arrows). In contrast, strong caspase-3 signals were expressed by small intestinal epithelial cells and lamina propria cells (Fig. 12c, arrows) of infected untreated animals. We also observed strong caspase-3 signals in cells lining the crypts as well as smooth muscle cells of muscularis externa layers (Fig. 12d, arrows). In animals infected with Giardia lamblia cysts and treated with metronidazole, fewer epithelial cells in the villi and crypts as well as muscularis externa layers expressed caspase-3 signals compared to infected untreated hamsters (Fig. 12e and f, arrows). On the other hand, the caspase-3 expression in tissues of animals infected with giardia cysts and treated with Artemisia annua appeared lower than those of infected untreated animals or animals infected and treated with metronidazole (Fig. 12g and h, arrows).