Genipin inhibits rotavirus-induced diarrhea by suppressing viral replication and regulating inflammatory responses

Rotavirus is the leading cause of acute gastroenteritis among young children worldwide. However, agents specifically designed to treat rotavirus infection have not been developed yet. In this study, the anti-rotavirus and anti-inflammatory effects of genipin, a chemical compound found in the fruit of Gardenia jasminoides, were evaluated. Genipin had an antiviral effect against the human rotavirus Wa and SA-11 strains in vitro, and it inhibited two distinct stages of the viral replication cycle: attachment and penetration (early stage) in pre-treatment and assembly and release (late stage) in post-treatment. Additionally, genipin downregulated nitric oxide synthase and pro-inflammatory cytokines in lipopolysaccharide-stimulated RAW264.7 cells and rotavirus-infected Caco-2 cells. Oral administration of genipin before and after viral infection with the murine rotavirus epidemic diarrhea of infant mice strain led to a reduced duration of diarrhea and faecal viral shedding and to decreased destruction of the enteric epithelium. Genipin could have potential as a natural compound with preventive and therapeutic effects against infection and colitis caused by rotavirus.

Effects of NO, PGE 2 , and pro-inflammatory cytokine generation. The inhibitory effects of genipin on accumulated production of LPS-induced NO and PGE 2 were determined in the cell supernatant of macrophage RAW264.7 cells treated with genipin at concentrations of 10, 50, 100, and 150 µM/mL in the presence of 0.1 μg/μL of LPS. Compared with the control of LPS-induced NO, accumulated NO and PGE 2 production in the cell supernatant treated with genipin was significantly inhibited at 50-150 µM/mL (Fig. S2A,B). In particular, treatment with genipin 50 µM/mL resulted in NO and PGE 2 production rates of 12% and 30%, respectively, relative to the untreated control. The levels of IL-6, IL-10, and IL-1β were significantly inhibited at genipin concentrations of 50-150 µM/mL. With 150 µM/mL of genipin, the IL-6 level was 4.0 pg/mL (compared with untreated control, 16.3 pg/mL) (Fig. S2C), the IL-10 level was 14.8 pg/mL (compared with untreated control, 84.8 pg/mL) (Fig. S2D), and the IL-1β level was 10.8 pg/mL (compared with untreated control, 43.9 pg/mL) (Fig. S2E). In contrast, the TNF-α level was slightly inhibited. At a genipin concentration of 150 µM/mL, the TNF-α level was 18.6 pg/mL (compared with untreated control, 21.0 pg/mL) (Fig. S2F).

Inhibitory effects of genipin against inflammation by rotavirus infection.
To evaluate the inhibitory effects of genipin against inflammation by rotavirus infection, NO and PGE 2 were measured in virusinfected Caco-2 cell supernatant fluid (Fig. 1). The Caco-2 cells infected with rotavirus were used as a positive control. NO and PGE 2 production were inhibited at genipin concentrations of 50-150 µM/mL in both pre-and post-treatment, compared with the positive control (Fig. 1A,B). The level of pro-inflammatory cytokines was significantly inhibited at genipin concentrations of 50-150 µM/mL in both pre-and post-treatment contexts. In detail, at a genipin concentration of 100 µM/mL, the IL-6 level was 22.7 pg/mL in pre-treatment and 16.6 pg/mL in post-treatment (compared with positive control, 67.3 pg/mL) (Fig. 1C), the IL-10 level was 50.0 pg/mL in pretreatment and 51.0 pg/mL in post-treatment (compared with positive control, 118.0 pg/mL) (Fig. 1D), the IL-1β level was 3.5 pg/mL in pre-treatment and 2.6 pg/mL in post-treatment (compared with positive control, 36.9 pg/ mL) (Fig. 1E), and the TNF-α level was 178.6 pg/mL in pre-treatment and 177.9 pg/mL in post-treatment (compared with positive control, 261.5 pg/mL) (Fig. 1F). At a genipin concentration of 150 µM/mL, the IL-6 level was 22.4 pg/mL in pre-treatment and 22.0 pg/mL in post-treatment (Fig. 1C), the IL-10 level was 45.8 pg/mL in pre-treatment and 54.5 pg/mL in post-treatment (Fig. 1D), the IL-1β level was 3.8 pg/mL in pre-treatment and 5.3 pg/mL in post-treatment (Fig. 1E), and the TNF-α level was 172.1 pg/mL in pre-treatment and 189.5 pg/mL in post-treatment (Fig. 1F).
Genipin inhibits rotavirus in a dose-dependent manner. The inhibitory effects of genipin against the human rotavirus Wa strain and simian rotavirus SA-11 strain in MA104 cells were determined by a real-time PCR viral titre assay using different strategies. As expected, viral titre was significantly reduced by genipin treatment in a dose-dependent manner. Genipin pre-treatment significantly reduced the Wa strain viral titre (5.02 log of 10 µM, 4.16 log of 50 µM, 3.17 log of 100 µM, and 0.86 log of 150 µM) and SA-11 strain viral titre (5.50 log of 10 µM, 5.22 log of 50 µM, 3.55 log of 100 µM, and 3.18 log of 150 µM), compared with untreated control (7.53 log and 6.23 log, respectively) in MA104 cells ( Fig. 2A). Post-treatment also significantly reduced Wa strain viral titre (6.14 log of 10 µM, 3.88 log of 50 µM, 2.86 log of 100 µM, and 2.94 log of 150 µM) and SA-11 strain viral titre (5.8 log of 10 µM, 4.53 log of 50 µM, 3.97 log of 100 µM, and 2.88 log of 150 µM), compared with untreated control (8.18 log and 6.23 log, respectively) in MA104 cells (Fig. 2B).
Inhibitory impact of genipin on rotavirus plaque formation in MA104 cells. Genipin was evaluated for anti-rotavirus activity in a plaque assay (Fig. S3). The wells pre-treated with genipin showed significantly lower plaque numbers at concentrations of 10, 50, 100, and 150 µM compared with untreated control wells. Genipin inhibits rotavirus in an MOI-dependent manner. MA104 cells were infected with multiplicities of infection (MOIs) of 0.1, 0.5, and 1 (Fig. 3). In genipin pre-and post-treatment, infectious virus particle production was significantly inhibited at MOIs of 0.1, 0.5, and 1. Genipin pre-treatment significantly reduced viral titre (5.92 log of MOI 0.1, 6.09 log of MOI 0.5, and 6.19 log of MOI 1). Genipin post-treatment also significantly reduced viral titre (5.89 log of MOI 0.1, 5.99 log of MOI 0.5, and 6.12 log of MOI 1), compared with untreated control (6.10 log of MOI 0.1, 6.17 log of MOI 0.5, and 6.32 log of MOI 1).
Effects on the viricidal activity of genipin. The viricidal activity of genipin against the human rotavirus Wa strain was determined via a real-time PCR viral titre assay using MA104 cells (Fig. 4). The genipin inhibited the Wa strain viral activity (6.17 log of 10 µM, 6.09 log of 50 µM, 6.07 log of 100 µM, and 5.98 log of 150 µM) relative to untreated control (6.32 log).   www.nature.com/scientificreports/ Impact of genipin on EDIM-induced diarrhea. The impact of genipin on EDIM-induced diarrhea in neonatal mice was tested at a genipin concentration of 100 µM/mouse. At 20,000 PFU/mouse, the mice pretreated with genipin (0%; p < 0.05) and those post-treated with genipin (0%; p < 0.05) had a significantly lower diarrhea score than the mice in the EDIM group (22.0%) from day 4 after infection ( Fig. 6A). On days 7-8, there were significant differences between the mice pre-treated and post-treated with genipin. The incidence of diarrhea among pre-treated mice (0%) was significantly lower than that among post-treated mice (50% on day 7 and 67% on day 8). Specifically, none of the mice pre-treated with genipin had diarrhea, compared with 100% of the mice with acute diarrhea and severe dehydration in the EDIM group after infection. The diarrhea score among mice post-treated with genipin (0; p < 0.05) was significantly lower than that among mice in the EDIM group (0.67) on day 4 after infection ( Fig. 6B). On days 5-6, the diarrhea scores showed significant differences (p < 0.0001) in both the pre-treated (0.50 and 0.23, respectively) and post-treated groups (0 and 0.72, respectively), compared with the EDIM group (2.86 and 2.00, respectively). All groups showed decreased diarrhea scores on day 8. Interestingly, mice post-treated with genipin had a diarrhea score of 0 until day 6, and a score ≤ 1 on days 7-8.

Viral shedding.
We wanted to confirm whether these approaches also influenced viral load. Total cellular RNA levels isolated from the faeces of neonatal mice pre-treated or post-treated with genipin were analysed by real-time RT-PCR to obtain a number of RNA copies of the EDIM VP6 gene (Fig. 7). From days 1-4, no significant differences in viral RNA shedding titre were observed in the four groups. On day 7, the viral RNA shedding titre in the EDIM group reached its peak by 6.95 log. In contrast, the viral RNA shedding titres in mice pre-treated (2.83 log; p < 0.0001) or post-treated with genipin (4.38 log; p < 0.0001) were significantly lower than that in the EDIM group. There were statistically significant differences (p < 0.005) between the pre-treated and post-treated groups on days 6-8.  www.nature.com/scientificreports/

Discussion
Rotavirus is still the leading cause of acute gastroenteritis among young children worldwide. Approximately 95% of children experience at least one rotavirus infection by the age of 5 years 30 . Since antibiotic treatment is nonspecific for rotavirus infection, vaccination is the most important strategy recommended by the World Health Organization to protect children against rotavirus-related hospitalization and death 31 . According to a recent study, only 53 countries offer rotavirus vaccines through their national immunization programs 32 . The efficacy of rotavirus vaccines is < 90%, so new preventive and therapeutic strategies need to be developed. Our data showed that rotavirus-infected MA104 cells treated with different concentrations of genipin secreted decreased levels of different inflammatory markers (IL-6, IL-10, IL-1β, and TNF-α); we also observed a genipin effect against the inflammatory responses of LPS-induced macrophage RAW264.7 cells. IL-6, IL-1β, and TNF-α are pro-inflammatory cytokines known as triggers of pathologic pain 33,34 .
Previous studies have indicated that some natural products and their secondary metabolites, such as pectic polysaccharides (Panax ginseng), epigallocatechin gallate, and theaflavin (green tea) have inhibitory effects against rotavirus 35,36 . However, these studies showed no inhibitory effects against rotavirus infection in either pre-or post-treatment. In the present study, pre-and post-treatment with genipin was associated with inhibitory effects against the human rotavirus Wa and SA-11 strains in MA104 and Caco-2 cells. Plaque numbers and quantification of viral titre via real-time PCR indicated significant decreases in viral infection associated with  www.nature.com/scientificreports/ both pre-and post-treatment with genipin. These results indicate that genipin can be used in the prevention and treatment of rotavirus infection. Moreover, anti-rotavirus activity was dose-dependently and MOI-dependently affected by genipin in both pre-and post-treatment contexts. Genipin also suppressed rotavirus replication at two distinct stages of the viral replication cycle: the rotavirus RNA replication cycle in MA104 cells was significantly inhibited at the early stage in pre-treatment and the late stage in post-treatment. These results are supported by previous studies, which indicated that kuraridin isolated from Sophora flavescens and iridoid glycosides extracted from Fructus Gardeniae have inhibited reovirus and influenza virus by replication cycles in pre-and post-treatment 37,38 . Additionally, genipin showed viricidal effects against rotavirus. Anti-rotaviral activity (e.g., inhibiting viral replication and viricidal effects) has previously been reported in association with natural products, including against porcine rotavirus (G5P [7]) and bovine rotavirus (G8P [7]) 39,40 .
We also evaluated the inhibitory effects and antiviral activity of genipin against the rotavirus EDIM strain in a murine model. The EDIM group exhibited clinical symptoms of rotavirus infection and high viral shedding, whereas both the pre-treated and post-treated groups exhibited significantly reduced diarrhea incidence and faecal viral shedding. These results are supported by previous in vivo studies, which showed that Glycyrrhiza uralensis and Oryza sativa, and a combination of Sophora flavescens and stevioside, have marked anti-rotavirus effects after induction of rotavirus diarrhea [41][42][43] . However, these studies investigated only preventive aspects. Taken together, our results demonstrate that genipin may be applied as a potent medication for preventing and curing rotavirus diarrhea in humans and animals.
In conclusion, genipin showed anti-inflammatory and antiviral activities against rotavirus in vitro and in vivo, which led to significantly reduced diarrhea incidence and suppressed viral shedding, by inhibiting the replication cycle (Fig. 8). Therefore, our results suggest that further studies are required for in vitro evaluation of various rotavirus types and for preclinical evaluation of genipin to apply new therapeutic strategies against rotavirus infection.

Materials and methods
Reagents. Genipin powder (≥ 98% high-performance liquid chromatography purity, G4796) was purchased Cytotoxicity assay. The cytotoxicity of genipin against macrophage RAW264.7, MA104, and Caco-2 cells was determined by MTT assay (Mosmann, 1983). Briefly, 100 µL of cell suspension containing 5 × 10 4 , 3 × 10 4 , and 3 × 10 4 of RAW264.7, MA104, and Caco-2 cells, respectively, was seeded into each well of a 96-well plate and incubated for 24 h. The cells were then treated with genipin at serial concentrations of 10 to 200 µM/mL. After 24 h of incubation, 5 µL of 5 mg/mL MTT reagent (Sigma-Aldrich) was added to each well and incubated at 37 °C for 4 h. The media were then carefully removed, and 150 µL of DMSO was added. The plates were covered with tinfoil and agitated on an orbital shaker for 15 min. The optical density at 590 nm was measured using a NanoQuant spectrophotometer (Infinite 200; Tecan, Männedorf, Switzerland). After the experiments, cells were harvested by freezing and thawing three times and centrifuged at 3000×g for 5 min, and the supernatant was stored at − 80 °C for quantitative polymerase chain reaction (qPCR) analysis. To validate the possibility of genipin anti-rotavirus activity, the simian strain SA-11 together with the Wa strain were evaluated in MA104 and Caco-2 cells. These experiments were performed using the same methods as above-the genipin antiviral activity methods (1) and (2). MA104 and Caco-2 cells untreated with genipin were used as a control.

Anti-rotavirus activity in vitro in an MOI-dependent manner.
To evaluate the efficacy of genipin against rotavirus infection, rotavirus stock was serially diluted in MOIs of 0.1, 0.5, and 1. MA104 cells were infected with different rotavirus MOIs. The minimal genipin concentration for anti-rotavirus activity was determined to be 100 µM/mL. These experiments were performed using the same methods as above-the genipin antiviral activity methods (1) and (2).
Genipin kinetics of rotavirus cell release. The rotavirus replication cycle was determined as previously described, with a few modifications [44][45][46][47] . Confluent monolayers of MA104 cells grown in a 96-well plate were washed with alpha-MEM and infected with the human rotavirus Wa strain at an MOI of 0.1. The virus was adsorbed to cells for 1 h at 37 °C. After the adsorption period, the inoculum was removed, and alpha-MEM was added. These experiments were performed using the same genipin antiviral activity methods as above, (1) and (2). MA104 cells untreated with genipin were used as a control. The infected cell supernatant was collected at different times (0, 1, 3, 6, 9, 12, 24, and 48 h).
The different categories of diarrhea reflected the amount of water lost during rotavirus infection. Diarrhea severity was reported as a diarrhea score, where level 3 was less severe, and level 4 was most severe. Faeces were collected daily, and EDIM viral shedding was assessed by qPCR analysis. Statistical analysis. Significant differences among the three groups were calculated using two-way analysis of variance with Dunnett's post-test for comparison using GraphPad Prism (v.6.0.1) (GraphPad Software Inc., San Diego, CA, USA). Data are expressed as mean ± standard error of the mean (SEM), and p values < 0.05 were considered significant. www.nature.com/scientificreports/