Promising efficacy of nitrogen-containing bisphosphonates against the infection of Cryptosporidium spp.

Abstract

Cryptosporidiosis is a major diarrheal disease that affects both humans and animals. Fully effective drug for treating cryptosporidiosis is still lacking. Nitrogen-containing bisphosphonates have been reported to inhibit Cryptosporidium growth in vitro; however, the in vivo efficacy against Cryptosporidium remain unevaluated. This study determined the anti-Cryptosporidium effect of three nitrogen-containing bisphosphonates risedronate, ibandronate, and zoledronate through both in vitro and in vivo experiments. It was determined that risedronate exhibited the highest therapeutic index of 39.10 among the three compounds, with the median effective concentration low to 17.44 μM against Cryptosporidium parvum infection in vitro. In vivo experiments showed that the high dose (10 mg/kg/d) of risedronate and ibandronate significantly reduced the shedding of Cryptosporidium tyzzeri oocyst, with no toxicity in ICR mice. Histopathological examinations of ICR mice indicated that high and medium (2 mg/kg/d) doses of the bisphosphonates could reduce intestinal damage, recover the height of intestinal villi and crypt depth, led to more intact intestinal structures, and risedronate showed the most promising effects. Furthermore, the three compounds modulated the elevated levels of IL-2, IL-4, and TNF-α cytokines, induced by C. tyzzeri infection, towards normalcy in a dose-dependent manner. In conclusion, the efficacy of three nitrogen-containing bisphosphonates against the in vitro and in vivo infection of Cryptosporidium spp. was assessed. Risedronate show promising effect for further development of new anticryptosporidial drugs.

Graphical abstract

1. Introduction

Cryptosporidium causes acute or persistent watery diarrhea in humans and animals worldwide (Black et al., 2024; Checkley et al., 2015). In immunocompetent patients, the infection is self-limiting; however, in young-children and immunocompromised individuals, Cryptosporidium infection is commonly chronic and can be fatal (Certad et al., 2005; Lenière et al., 2024). Although nitazoxanide has been approved by the United States Federal Drug Administration (FDA) for the clinical treatment of cryptosporidiosis, its efficacy is limited in children and comparable to a placebo in immunocompromised patients (Khan and Witola, 2023). Paromomycin was demonstrated to show great efficacy in immunocompromised patients, and in recent years, it has often been used as a comparator in researches investigating antiparasitic drugs for Cryptosporidium treatment (Rahman et al., 2021; Rosenblatt, 1999).

Bisphosphonates, which are often used in the clinical treatment of bone metabolic diseases such as osteoporosis, Paget's disease, and hypercalcemia, have also been found to possess antimicrobial properties against parasitic protozoa such as Plasmodium, Toxoplasma, Trypanosoma, and Leishmania (Gisselberg et al., 2018; Li et al., 2017; Martin et al., 2001; Sleda et al., 2022). As metabolically stable analogs of pyrophosphate, bisphosphonates competitively inhibit key enzymes, farnesyl pyrophosphate synthase (FPPS) and nonspecific polyisoprenyl pyrophosphate synthase (NPPPS), in the non-mevalonate pathway of isoprenoid biosynthesis, which is absent in humans, thereby affecting the production of isoprenoid compounds essential for parasite survival and growth (Artz et al., 2008; No et al., 2012). Studies have demonstrated that nitrogen-containing bisphosphonates (N-BPs) effectively inhibit the in vitro growth of Plasmodium and Toxoplasma, and the in vivo infection of Toxoplasma (Sleda et al., 2024; Szajnman et al., 2017). It has also been demonstrated that N-BPs inhibit the in vitro growth of Cryptosporidium parvum (C. parvum), with NPPPS as the target molecule (Artz et al., 2008).

In vitro evaluation of anticryptosporidial efficacies of therapeutic agents has been generally carried out in the colon cancer cell line HCT-8 (Bessoff et al., 2013). However, in vivo studies have encountered the barrier of lacking suitable animal models (Ali et al., 2024; Lenière et al., 2024). In recent years, it has been demonstrated that a native mouse parasite Cryptosporidium tyzzeri (C. tyzzeri) is closely related to the zoonotic C. parvum on genetic level and resembles the pathological infection of C. parvum in humans and large animals (Sateriale et al., 2019). Therefore, immunocompetent mouse models infected with C. tyzzeri were established and proven to be cost-effective for anticryptosporidial drug evaluation (He et al., 2022; Liu et al., 2023; Sateriale et al., 2019). By using the in vitro and in vivo systems, anticryptosporidial efficacies of various therapeutic agents have been tested, including agents targeting Cryptosporidium calcium-dependent protein kinases (CpCDPK1), phospho-inositol 4 kinase (PI4K), and cleavage and polyadenylation specific factor 3 (CPSF3) (Asmare and Yun, 2024; Lunde et al., 2019; Manjunatha et al., 2024).

Here, we report the promising efficacy of N-BPs against the in vivo infection of C. tyzzeri. C. tyzzeri-infected ICR mouse model was established. By evaluating the efficacy of three N-BPs, risedronate (RIS), ibandronate (IBAN), and zoledronate (ZOL), in combating C. tyzzeri infection in vivo, this study provides evidence for further evaluation of drug efficacy of N-BPs to treat cryptosporidiosis.

2. Material and methods

2.1. Parasite strains, cell lines and bisphosphonates compounds

The oocysts of C. parvum (gp60 subtype ⅡdA19G1) and C. tyzzeri (gp60 subtype IXa), and the human colon cancer cell line HCT-8 were passaged and preserved by the International Joint Research Laboratory for zoonotic diseases of Henan, China. The oocysts used were freshly purified and collected within two weeks of passage. Risedronate sodium (RIS, ≥98 % purity, CAS No. 115436-72-1), Ibandronate sodium monohydrate (IBAN, ≥98 % purity, CAS No. 138926-19-9) and Zoledronate monohydrate (ZOL, 99.88 % purity, CAS No. 165800-06-6) were from MCE (MedChemExpress LLC, Shanghai, China). Paromomycin (PRM) was from Solarbio® Life Sciences in Beijing, China; Dexamethasone (DEX) Sodium Phosphate injection was purchased from Shanghai Quanyu Biotechnology Animal Pharmaceutical Co., Ltd. (Shanghai, China). The ELISA kits for mouse interleukin-2 (IL-2), interleukin-4 (IL-4), interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α) were provided by Shanghai Meilian Biotechnology Co., Ltd (Shanghai, China).

2.2. Detection of compound cytotoxicity by MTT assay

The reduction of 3-(4,5-dimethylthiazol-2-yl)2,5-diphenyltetrazolium bromide (MTT) by living cells was applied to quantify cytotoxicity of the bisphosphonates. Assays were conducted using >95 % confluent HCT-8 cells grown in 96-well plates for 48 h (37 °C) with serial diluted compounds in the concentrations from 800 μM to 6.25 μM. 10 μl of MTT-solution (0.5 % w/v Thiazolyl blue, Sangon Biotech, Shanghai, China) was added to each well and incubation was continued for another 4 h, followed by addition of 150 μl of DMSO to each well to dissolve formazan crystals fully. The corner wells of each plate were trypsinized to remove cells. The mixture of equal volume of PBS and cell culture medium (RPMI1640, VivaCell Biotechnology GmbH, Denzlingen, Germany) were used as blanks for measuring absorbance at 490 nm. Data were expressed as the percentage of cell viability for vehicle controls (PBS) on each assay plate. GraphPad PRISM software, version 8, was used to calculate the 50 % cytotoxic concentration (CC₅₀).

2.3. In vitro inhibition of C. parvum infection by the compounds

Oocysts were treated with 10 mM hydrochloric acid (10 min, 37 °C), followed by exposure to excystation buffer (0.25 % w/v trypsin and 0.75 % w/v taurocholate) in PBS (6 h, 37 °C). The released sporozoites were then mixed with each of the three N-BPs (RIS, IBAN, ZOL), which have been threefold diluted from 100 to 0.41 μM in RPMI1640. The mixture of sporozoites and compounds were added to >80 % confluent HCT-8 cell monolayers in 12-well plates with three biological replicates for each concentration, using the ratio of C. parvum oocysts to HCT-8 cells as 5 to 1 per well. After culturing for 3 h, the medium and uninfected sporozoites were discarded and gently washed with PBS. Fresh drug-containing medium was added and the infected cells were cultured for another 45 h. Cells treated with PRM and RPMI1640 were setup as positive and negative controls.

2.4. Determination of compound EC₅₀ by qRT-PCR

After 48 h incubation, HCT-8 cells infected with the compound-treated sporozoites from each well were collected and total RNA were extracted using RNA-easy isolation reagent (Vazyme, Nanjing, China), followed by reverse transcription into cDNA using HiScript Ⅲ All-in-one RT Super Mix Perfect for qPCR (Vazyme, Nanjing, China). Quantitative real-time PCR (qRT-PCR) was carried out to detect the parasite infection rate using forward and reverse primers 5′- TAGAGATTGGAGGTTGTTCCT-3′ and 5′- CTCCACCAACTAAGAACGGCC-3′ specific for the SSU rRNA of C. parvum. The results were analyzed and statistically processed by using the 2^−ΔΔCt calculation method. GraphPad Prism software, version 8, was used to calculate the median effective concentration (EC₅₀). Validation of the EC₅₀ values was performed by using the calculated EC₅₀ concentrations of the three compounds to inhibit C. parvum infection of HCT-8 cells. Cells treated with PRM and RPMI1640 were setup as positive and negative controls. Viability of C. parvum was determined by qRT-PCR.

2.5. Therapeutic indices of the compounds

Therapeutic Indices (TI) for RIS, IBAN and ZOL were determined at the ratio of CC₅₀ to EC₅₀.

2.6. ICR mouse model of cryptosporidiosis

All mouse experiments were performed in compliance with animal care guidelines and were approved by Henan Agricultural University Institutional Animal Care and Use Committee. SPF-grade 3-week-old female ICR mice were all purchased from Liaoning Changsheng Biotechnology Co., Ltd. (license number: SCXK Liaoning, 2020-0001). Mice were housed by experimental groups for two days for acclimatization, immunosuppressed with dexamethasone solution (DEX) (5 mg/L in drinking water) for three days, and then infected with 10⁶ oocysts (day 0). Feces were initially examined for 28 days in the preliminary experiment to determine the oocyst shedding pattern. For the subsequent experiment, the three N-BPs and PRM (drug control) were dosed by oral gavage on the day with detectable fecal oocysts, and the treatment lasted for 7 days according to the regimen specified (6 mice per group). The doses for RIS, IBAN and ZOL were prepared on the day of dosing with high (10 mg/kg body weight), medium (2 mg/kg body weight) and low (0.4 mg/kg body weight) concentrations in distilled water. PRM was freshly prepared with the dose of 200 mg/kg/d. From the day of parasite infection (day 0), oocyst shedding in feces was monitored daily using microscopic detection, to the end of experiment on 11 days post-infection (dpi). Blood samples were collected from all 6 mice in each group for cytokine detection. Three mice from each group were euthanized (intravenous injection of sodium pentobarbital) and sampled for the observation of organ indices and ileum pathology. The rest three mice in each group were euthanized on 28 dpi for the long-term observation of organ indices.

2.7. Oocyst shedding pattern in mice

Oocyst count per gram of feces (OPG) was used to monitor oocyst shedding pattern in mice. Fecal samples were collected at the amount of 1 g for each mouse, and oocysts were detected using the saturated sucrose flotation method (Duan et al., 2024). Briefly, samples were filtered through an 80-mesh sieve and washed with single-distilled water twice. After centrifugation at 2000×g for 10 min, the supernatant was discarded and 13 mL of saturated sucrose solution was added to the precipitate. The mixture was centrifuged at 2000×g for 10 min and parasites were collected from the surface. Single-distilled water was added to the collected parasites to wash away the sucrose solution. The collected oocysts were counted under microscope on a hemocytometer with 40× magnification. OPG = (average number of parasites on the hemocytometer under 40× magnification) × 10⁴ × dilution factor × total volume.

2.8. Organ indices in mice

The organ index (%) of each mouse was determined by the ratio of the weight of each organ to the total weight of the mouse, including the organ indices of heart, liver, spleen, lungs and kidneys.

2.9. Pathological observation of mouse ileum

The pathological examination included assessing intestinal damage, the height and width of intestinal villi, the depth of crypts, and the integrity of the intestinal structure. Two sections of the ileum about one square centimeter each were selected, and then fixed in 4 % paraformaldehyde and electron microscope fixative respectively, and properly labeled. After fixation at room temperature for 1 h, they were transferred to 4 °C for long-term storage. The collected ileum samples were transferred to Servicebio® Technology Co., Ltd (Wuhan, China) to make pathological sections and perform scanning electron microscope photography. Randomly, 10 fields of view for each section were selected, the villus height and crypt depth were measured, and the villus to crypt ratio was calculated for each examined field (Sateriale et al., 2021).

2.10. Detection of cytokines in blood serum

In order to determine the in vivo immunoregulatory effect of the three N-BPs, the levels of cytokines in the blood of mice were measured. At the end of compound administration, 200 μL of blood was collected from the tip of the tail for each group of mice. After being stored overnight at 4 °C, it was centrifuged at 1500×g for 10 min. The upper layer of serum was extracted and placed in a new 1.5 mL centrifuge tube and properly labeled. Cytokine levels for IL-2, IL-4, IFN-γ and TNF-α were detected according to the instructions of respective ELISA kits. For long-term storage of the serum, gradient cooling start with storage at 4 °C for 30 min, then transferred to −20 °C for 2 h, and finally transferred to −80 °C for long-term storage (Codices et al., 2013). Each experiment was conducted in three repeats to ensure the veracity of the result.

2.11. Statistical analysis

All the above data were statistically analyzed by one-way ANOVA using SPSS 24.0 software, and the Duncan's homogeneity test for significant difference comparison (P < 0.05 indicates significant difference), or the Dunnett multiple comparisons test of GraphPad Prism8 software for significant difference analysis (P < 0.05 indicates significant difference).

3. Results

3.1. The three N-BPs inhibited C. parvum infection in vitro

The anticryptosporidial effect of RIS, IBAN and ZOL was initially determined in vitro. The cytotoxicity of the three compounds on HCT-8 cells were measured by MTT method, and then the inhibitory effects of the compounds on C. parvum were detected and analyzed by quantitative real-time PCR. The results showed that the CC₅₀ of RIS, IBAN and ZOL were 681.9 μM, 621.9 μM and 542.3 μM, respectively, and the EC₅₀ values were 17.44 μM, 61.03 μM and 20.74 μM, respectively (Fig. 1). The calculated EC₅₀ of the three compounds were validated to show that the inhibitory effect of the EC₅₀ concentrations on C. parvum was consistently close to 50 % (Supplementary Fig. 1). The therapeutic indices were calculated as 39.10, 10.19 and 26.15 for RIS, IBAN and ZOL, respectively.

Fig. 1.

Evaluation of anticryptosporidial efficacy of RIS, IBAN and ZOL in vitro. The cytotoxicity of RIS (A), IBAN (B), and ZOL (C) were measured by MTT method with serial diluted compounds in the concentrations from 800 μM to 6.25 μM. The anticryptosporidial efficacy of RIS (D), IBAN (E), and ZOL (F) were measured according to the parasite loads quantified by qRT-PCR and statistically processed using the 2^−ΔΔCt calculation method. Three technical replicates were set up for both MTT and qRT-PCR. Abbreviations: CC₅₀, 50 % cytotoxicity concentration; EC₅₀, 50 % effective concentration; CI, confidence interval; qRT-PCR, quantitative reverse transcription polymerase chain reaction.

3.2. The three N-BPs reduced C. tyzzeri oocysts shedding in infected ICR mice

ICR mice infected with C. tyzzeri began to shed oocysts on the third day post infection (dpi) and reached the highest on 8 dpi (Fig. 2A). For testing the anticryptosporidial efficacy of the three N-BPs, ICR mice (6 females in each group) treated with DEX were infected with C. tyzzeri for 3 days before treatments with PRM and the three N-BPs for 7 days (see design in Fig. 2B). Absolute oocyst counts for all groups were recorded on 3, 6, 8 and 10 dpi, and the highest number of oocyst shedding was observed on 8 dpi for all groups (Fig. 2C).

Fig. 2.

Efficacy of RIS (A), IBAN (B) and ZOL (C) at indicated daily doses on oocyst shedding of C. tyzzeri infected mice. A, Oocyst shedding pattern in ICR mice infected with C. tyzzeri from 0 to 28 days post infection. B, Experimental design. ICR mice receiving DEX in drinking water for three days were infected with C. tyzzeri, and treated with 3 compounds for 7 days. Fecal samples were collected on the indicated days for determining oocyst production. C, Oocyst productions on 3, 6, 8 and 10 days post infection (dpi) in mice treated with PRM and the 3 compounds. Negative control received vehicle (distilled water). DEX, dexamethasone.

The oocyst shedding showed significant reduction in mice treated with PRM and the three N-BPs (Table 1). Compared with the vehicle group, the oocyst reduction rate on 8 dpi was 39.15 % (P < 0.0001) for mice treated with PRM. RIS reduced oocyst production on 8 dpi by > 36 % (36.76 %–69.86 %; P < 0.0001 vs vehicle). Both IBAN and ZOL reduced oocyst productions on 8 dpi by > 13 % (13.66 %–55.99 % for IBAN, 28.52 %–43.10 % for ZOL; P < 0.0001 vs vehicle). The efficacy of RIS and IBAN against C. tyzzeri appeared to be dose-dependent, but not for ZOL, where the efficacy appeared to be reversely related to the compound concentration (Fig. 2C and Table 1). The high-dose of RIS showed the strongest effect on oocyst reduction.

Table 1.

Effects of the three N-BPs for the reductions of C. tyzzeri oocyst productions in ICR mice.

Groups	Mean (OPG)	SEM	Reduction rate (%)	P Value
Vehicle	2.37 × 106	7.26 × 104	NA	NA
PRM	1.44 × 106	4.58 × 104	39.15	∗∗∗∗
RIS, 10 mg/kg/d	7.13 × 105	2.40 × 104	69.86	∗∗∗∗
RIS, 2 mg/kg/d	1.11 × 106	3.04 × 104	53.31	∗∗∗∗
RIS, 0.4 mg/kg/d	1.50 × 106	6.39 × 104	36.76	∗∗∗∗
IBAN, 10 mg/kg/d	1.04 × 106	3.32 × 104	55.99	∗∗∗∗
IBAN, 2 mg/kg/d	1.56 × 106	5.66 × 104	34.15	∗∗∗∗
IBAN, 0.4 mg/kg/d	2.04 × 106	6.44 × 104	13.66	∗∗∗∗
ZOL, 10 mg/kg/d	1.69 × 106	2.46 × 104	28.52	∗∗∗∗
ZOL, 2 mg/kg/d	1.53 × 106	5.46 × 104	35.49	∗∗∗∗
ZOL, 0.4 mg/kg/d	1.35 × 106	1.16 × 105	43.10	∗∗∗∗

Oocyst productions and reduction rates between compounds administrated and control (vehicle) mice on 8 dpi are shown. Statistical significances in responses to the treatments by Dunnett multiple comparisons test are presented. SEM, standard error of the mean; NA, not applicable. ∗∗∗∗P < 0.0001.

3.3. RIS and IBAN did not have toxic effect on ICR mice

As shown in Fig. 3, RIS and IBAN did not induce organ abnormality for all of the examined organs; however, mice administrated with the high-dose of ZOL showed significant differences for heart, spleen, lungs and kidneys (P < 0.05), indicating the potential toxic effect of high-dose ZOL on ICR mice. To determine the long-term effect of compounds on ICR mice, organ indices were monitored on 28 dpi and showed no significant difference in all groups (Supplementary Table 1), suggesting that the toxicity of high-dose ZOL disappeared after the compound was withdrawn and the mice return to normal after a period of recovery.

Fig. 3.

Mouse organ indices at the end of N-BPs administration. At the end of compound administration (11 dpi), organ indices for heart, spleen, lungs and kidney were determined for random three mice in each group. For the examined organs, column with different lowercase letters represents significant differences (P < 0.05).

3.4. The three N-BPs protected the ileum of ICR mice from the disruption of C. tyzzeri infection

Histological analysis of the mice ilea discovered the complete morphological structure of villi in the uninfected mice (Fig. 4A). In mice infected with C. tyzzeri, the structural integrity of the ileum was disrupted. As shown in Fig. 4B, the ileum villi of infected mice were loosely and disorderly arranged, shed and vacuolation were observed, the intestinal wall became thinner and the muscular and mucosal layers were separated, and the number of goblet cells was significantly increased. The structure of the intestinal villi for mice treated with PRM was relatively complete with no obvious breakage, and the thickness of the intestinal wall was increased (Fig. 4C). The ileum of mice treated with RIS showed a dose-dependent recovery, and the high dose of RIS had the best effect, showing that the villi were regularly arranged and the intestinal wall was restored (Fig. 4D–F). In mice treated with IBAN, only the high- and medium-dose of compound had recovery effect, and the disruption of villi structure was clearly observed for the low dose treatment (Fig. 4G–I). The intestinal structure of mice treated with ZOL didn't show any recovery for high-, medium- and low-doses. The pathological features included the thinned intestinal wall, the disorderly arranged, shed and vacuolated villi, the separated muscular layer, and the infiltration of inflammatory cells (Fig. 4J–L).

Fig. 4.

Histopathological changes of mice ilea by HE staining. Pathological sections of ileum from mice (n = 3/group) under the following conditions were examined: A, uninfected; B, infected; C, PRM treated; D, RIS high dose; E, RIS medium dose; F, RIS low dose; G, IBAN high dose; H, IBAN medium dose; I, IBAN low dose; J, ZOL high dose; K, ZOL medium dose; L, ZOL low dose. The black arrow indicates goblet cells, the green bidirectional arrow indicates the muscle layer, the blue arrow shows intestinal villi, the circles represent intestinal villi vacuolation, the triangle represents the separation of muscle layer, the hollow arrows represent inflammatory cells. HE, hematoxylin-eosin.

The results of electron microscope scanning of the ileum villi showed that the villi of uninfected mice had complete structure and they were arranged neatly and densely. On the other hand, there were a large number of oocysts on the ileum surface of mice infected with C. tyzzeri. The villi around the oocysts became blunt and the arrangement was disordered, with the necrosis of epithelial cells (Supplementary Fig. 2A). The therapeutic effects of PRM, high dose RIS and IBAN were the most obvious (Supplementary Fig. 2B–C). The number and arrangement of ileum villi were improved, the number of necrotic loci and oocysts were significantly reduced, and the phenomenon of villi blunting was significantly alleviated. The protective effect of RIS and IBAN was dose dependent. Unlike RIS and IBAN, ICR mice treated with ZOL didn't show significant recovery for the ileum villi. The oocysts and necrotic loci were clearly seen in mice, and the ileum villi were disordered (Supplementary Fig. 2D).

In ICR mice infected with C. tyzzeri and treated with the three N-BPs, the villus height of ileum was significantly higher (P < 0.05) in all groups treated with high- and medium-dose of compounds (Fig. 5A), and the values were relatively close to those of the uninfected mice. The villus width did not change significantly among uninfected, infected, PRM and compound treated groups (Fig. 5B). Compared to the infected mice, the crypt depth of ileum reduced significantly (P < 0.05) for mice treated with high- and medium-dose of RIS, all doses of IBAN and high-dose of ZOL (Fig. 5C). The villus-to-crypt ratio showed significant improvement in high- and medium-dose RIS, all doses IBAN, and high-dose ZOL treatment groups compared to the infected group, whereas low-dose of RIS, medium- and low-dose ZOL regimens without statistical significance (p > 0.05) (Fig. 5D).

Fig. 5.

Statistical analysis of ileum changes in mice after the administration of the three N-BPs. Three mice (n = 3) from each group were euthanized (intravenous injection of sodium pentobarbital) and sampled for the observation of ileum pathology. Villus height (A), villus width (B), crypt depth (C) and villus/crypt ratio were presented and analyzed using t-test (vs the infected group). ∗ 0.01<P < 0.05; ∗∗ 0.001<P < 0.01; ∗∗∗P < 0.001.

3.5. The three N-BPs regulated the increased cytokine levels in C. tyzzeri-infected ICR mice to the normal range

Cytokine changes in mice infected with C. tyzzeri and treated with the three N-BPs were determined through detection of IL-2, IL-4, INF-γ and TNF-α levels in the blood serum. As shown in Fig. 6, compared to uninfected mice, cytokine levels for IL-2, IL-4, INF-γ and TNF-α were significantly elevated in mice infected with C. tyzzeri. For mice treated with RIS, IBAN and ZOL, expression levels for IL-2, IL-4 and TNF-α showed marked reduction compared to mice infected but not treated. The trend of cytokine reduction in the bisphosphonates treated mice was similar to the PRM treated mice, towards to the normal range of cytokine levels in uninfected mice, and the high dose of bisphosphonates showed the strongest effect, except for IL-2 (Fig. 6A, B and D). Levels of IFN-γ didn't show significant change in mice treated with the three N-BPs (Fig. 6C).

Fig. 6.

Cytokine levels in mice after administration of the three N-BPs. The levels of IL-2 (A), IL-4 (B), IFN-γ (C) and TNF-α (D) were measured in the blood of mice (n = 6/group) in all groups. Three technical repeats were set up to ensure the veracity of the result. Statistical analysis (vs Infected group) was performed using t-test (∗ 0.01<P < 0.05; ∗∗ 0.001<P < 0.01; ∗∗∗P < 0.001). IL-2, interlukin-2; IL-4, interlukin-4; IFN-γ, interferon γ; TNF-α, tumor necrosis factor-α.

4. Discussion

Cryptosporidiosis can cause severe diarrhea, especially in immunocompromised individuals such as AIDS patients, young children, and the elderly, which may lead to dehydration, malnutrition, and even death (Khalil et al., 2018; Ryan et al., 2021). In addition, it can cause economic losses in the livestock industry due to reduced productivity and increased mortality of infected animals (Cai et al., 2019; Zhang et al., 2020). The existing treatment options, including the only FDA approved Nitazoxanide for the treatment of cryptosporidiosis in human patients and the commonly used paromomycin in animals, show limitations of less productive in immunodeficient patients and poor intestinal absorption (Gupta et al., 2020). There is still a lack of highly effective and safe drugs for cryptosporidiosis.

Bisphosphonates have been maturely applied in the clinical treatment of bone metabolic diseases (Ebetino et al., 2022; Rogers et al., 2020), and their safety and chemical spatial structure are relatively clear, which is beneficial to the development of relevant pharmaceutical preparations of nitrogen-containing bisphosphonates. In addition, bisphosphonates containing simple aromatic or alkyl groups have also been confirmed to have inhibitory effects against Apicomplexa protozoan infections. Studies have shown that nitrogen-containing alkyl bisphosphonates showed much higher therapeutic indices against T. gondii than risedronate (Ling et al., 2005; Martin et al., 2001); moreover, the combination of bisphosphonates had a synergistic effect and can significantly reduce the toxic and side effects of compounds (Li et al., 2017). This study evaluated the inhibitory effects of three N-BPs on Cryptosporidium infection through both in vitro and in vivo experiments, providing a reference for the screening and development of potential anti-Cryptosporidium drugs.

By using C. parvum-infected HCT-8 cells as the in vitro model, it was determined that three N-BPs, RIS, IBAN, and ZOL, can significantly inhibit the infection of Cryptosporidium. The cytotoxicity of the three compounds was initially evaluated using the MTT assay, a widely used method to measure cytotoxicity of potential antiparasitic drugs. It has been used to evaluate the cytotoxicity of beta, beta-dimethylacrylshikonin and isobutyrylshikonin on Vero cells in a recent study of anti-Toxoplasma gondii drug research (Guo et al., 2025), as well as the toxic activity of Artemisia Judaica against C. parvum infection (Ahmed et al., 2023). Given the need to screen three N-BPs in this study, MTT assay provided a high-throughput, objective and quantifiable readout of viability, allowed for statistical analysis and dose-response curve generation, which were central to the study. By using MTT assay, CC₅₀ of the three compounds were calculated in the range of 542.3 μM–681.9 μM.

The inhibitory efficacy of RIS, IBAN, and ZOL on the growth of C. parvum was confirmed using HCT-8 cells. Previous studies have demonstrated that RIS, IBAN, and ZOL have strong inhibitory activity against C. parvum in MDCK cells, and their half maximal inhibitory concentrations (IC₅₀) can be as low as 3–6 μM (Artz et al., 2008). Here, EC₅₀ of the three compounds were determined in the range of 17.44–61.03 μM, which was higher than the concentration range measured in MDCK cells. The different effective concentrations could be related to the different cell lines and C. parvum strains used in the in vitro experiments. In this study, the EC₅₀ of RIS for inhibiting C. parvum was 17.44 μM, which was the lowest among the three compounds and determined to show the highest therapeutic index. However, the in vitro indices determined in this study specific to the HCT-8 infection model and should be interpreted cautiously for predicting in vivo safety. Since many factors including pharmacokinetics, tissue distribution, and organ-specific toxicity are crucial for determining the true clinical therapeutic indices, future studies evaluating the efficacy of the three N-BPs in more cells lines, especially primary cells, as well as in animal models would better determine their therapeutic efficacies.

In the current study, using ICR mice, which are the most susceptible hosts of C. tyzzeri, as an animal model, the strongest inhibitory effect against C. tyzzeri infection was observed for RIS administrated at 10 mg/kg. The doses of the compounds selected in this study mainly refer to relevant studies on T. gondii, as well as the preliminary experiment of this study. Intraperitoneal injection of 10 mg/kg and 20 mg/kg RIS into T. gondii infected mice for 10 consecutive days resulted in protection rates of 40 % and 55 %, respectively, and the average survival times were 20.3 and 22.1 days (Ling et al., 2005; Yardley et al., 2002). Although ZOL has been approved by the FDA for human bone resorption therapy, 5 mg/kg and 10 mg/kg ZOL had no protective effect on T. gondii infected mice (Yardley et al., 2002). Three doses ranging from 0.4 to 10 mg/kg were used in this study. It was found that high dose (10 mg/kg) ZOL had a toxic effect on the organs of ICR mice. However, the toxicity gradually disappeared after drug withdrawal. Different doses of RIS and IBAN all had relatively high safety.

There are fewer studies on the effect of N-BPs on the in vivo infection of Cryptosporidium. In a mouse xenograft model, 100 μM RIS can effectively inhibit the in vivo growth of C. parvum, and no parasites were detected in the mice (Moreno et al., 2001). Results presented in this study provide the direct, side-by-side comparison of the anti-Cryptosporidium efficacy of RIS, IBAN, and ZOL under identical experimental conditions both in vitro and in vivo. This allows for a more robust ranking of their relative potencies. Moreover, it is worth mention that the in vivo animal model used in this study was C. tyzzeri infected mice. As wild-type mice are naturally resistant to C. parvum infection and C. tyzzeri is a native mouse parasite, it resembles the pathological infection of Cryptosporidium in humans and large animals (He et al., 2022; Sateriale et al., 2019). Compared with the high-cost knockout mice, the mouse model used in this study has cost-effective advantage in evaluating anticryptosporidial drug efficacy (Liu et al., 2023). Furthermore, this study established detailed in vivo dose-response relationships for all three compounds within the model system, providing quantitative data for comparison.

The anti-Cryptosporidium effect of the three N-BPs mainly manifested as significantly inhibiting the production of oocysts and effectively protecting the integrity of the ileal structure of mice. Oocyst shedding is an important indicator for evaluating the efficacy of compounds or drugs against Cryptosporidium infection. In a study reporting the establishment of a novel in vivo mouse model for cost-effective anticryptosporidial drug testing, oocyst shedding was monitored to verify that two established anticryptosporidial drugs (paromomycin and nitazoxanide) and three leads (vorinostat, docetaxel, and baicalein) can effectively against the infection of C. tyzzeri in the model, by reducing the oocyst production up to 90 % (Liu et al., 2023). In this study, the three N-BPs reduced the oocyst productions up to 70 %, indicating their promising efficacy against C. tyzzeri infection. It is worth noting that the reduction rate of oocyst shedding was positively correlated with the dose administrated for RIS and IBAN, but not for ZOL. This discrepancy might be induced by the high bone affinity of ZOL (Kimmel, 2007; Reid et al., 2005), which could result in lower effective drug concentration at the intestinal sites of C. tyzzeri infection. Moreover, the toxicity of high-dose ZOL (Prommer, 2009) might damage intestinal epithelium, causing enhanced parasite invasion. When the ilea structure was examined, positive correlations between the integrity of ilea structure and the dose of compounds were observed for all three N-BPs. These results clearly demonstrated the protective effect of the three N-BPs against the infection of Cryptosporidium.

This study also measured the levels of four cytokines IL-2, IL-4, IFN-γ and TNF-α. It has been reported that IL-2 activated CD4⁺ T cell immunity, IL-4 induced Th1 cytokines, IFN-γ mediated STAT1 pathway, and TNF-α related TRAIL-dependent pathway all play important roles in host immunity against Cryptosporidium infection (Cohn et al., 2024; McDonald et al., 2004; Pardy et al., 2024; Xie et al., 2022). Cytokine levels have also been evaluated in studies of anti-Cryptosporidium therapeutic researches. The immunomodulatory effects of Aloe vera gel was demonstrated through systematic monitoring of IFN-γ and IL-4 levels in immunosuppressed mice infected with C. parvum (Farid et al., 2021). The levels of IFN-γ and TNF-α were tested to reflect the therapeutic effect of chitosan on immunosuppressed mice infected with C. parvum (Rahman et al., 2021). Furthermore, IL-2 has been established as a reliable biomarker of Th1-mediated immune responses during Cryptosporidium infections (Bedi et al., 2014; Bedi and Mead, 2012). Here, we observed the elevated cytokine levels for ICR mice infected with C. tyzzeri. After the administration of the three N-BPs, a dose-dependent recovery effect was observed for IL-2, IL-4, and TNF-α, while no significant difference was detected for IFN-γ. It has been reported that IFN-γ plays an important role in controlling Cryptosporidium infection (Pardy et al., 2024). Previous studies suggest that N-BPs (especially ZOL) can further elevate IFN-γ via γδ T-cell activation (Nussbaumer et al., 2013). Therefore, persistent IFN-γ likely reflects ongoing immune activation necessary for durable parasite control (Gullicksrud et al., 2022). These findings suggest that N-BPs treatment can modulate cytokine-mediated immune responses and mitigate Cryptosporidium infection.

Different IL-2 profiles were observed between ZOL and the other two compounds. C. tyzzeri infection induced significant increase of IL-2 in ICR mice, which is consistent with the increased Th1-enhancing cytokines induced by C. parvum antigens (Bedi and Mead, 2012). For mice administrated with RIS and IBAN, the elevated IL-2 levels decreased in a dose-dependent manner. However, the effect of ZOL on IL-2 appeared to be reversely related to the dose. The disparate was consistent with the observation of oocyst shedding, and could also stem from both the toxicity and immunomodulatory effects of ZOL at varying doses. The oocyst shedding result showed that the low-dose ZOL clears parasites most efficiently, therefore eliminates antigenic drive for IL-2 production. It has been reported that ZOL activates γδ T-cells, as well as induces dose-dependent increase of antigen-specific CD8 T cell responses (Nussbaumer et al., 2013; Park et al., 2016). The varying doses of ZOL used in this study might have different immunomodulatory effect, resulting in different parasite clearance rate and the inverted IL-2 profile. Further investigation will be required for elucidating the precise mechanism behind the unique profile of ZOL.

In conclusion, RIS, IBAN, and ZOL showed significant protective effects against Cryptosporidium infection. RIS had a significant inhibitory effect on the in vivo infection of C. tyzzeri and can be further studied as a potential anticryptosporidial drug. While this study represents an early treatment scenario of Cryptosporidium infection, it lays the foundation for further exploring bisphosphonate drugs for the treatment of Cryptosporidium infection in vivo. Future studies using chronic infection models to evaluate the compound efficacy would better model clinical cryptosporidiosis.

CRediT authorship contribution statement

Wenyan Hou: Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation. Xinyi Chen: Writing – original draft, Visualization, Investigation, Formal analysis. Yingying Zhang: Methodology, Investigation. Longfei Wu: Methodology, Investigation, Formal analysis. Songying Sun: Methodology, Investigation. Jiaye Guo: Methodology, Investigation. Wenchao Zhao: Software, Methodology. Junqiang Li: Writing – review & editing, Methodology. Sumei Zhang: Writing – review & editing, Visualization, Validation, Project administration, Methodology, Investigation, Formal analysis, Data curation. Longxian Zhang: Supervision, Resources, Methodology, Funding acquisition. Xiaoying Li: Writing – original draft, Visualization, Investigation, Formal analysis.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the author(s) used Grammarly to improve text grammar and readability. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the publication's content.

Conflict of interest

The authors declare that they have no competing interests.

Appendix A. Supplementary data

The following is the Supplementary data to this article.