PCR-based detection of Helicobacter spp. in animal facilities of a University in Rio de Janeiro, Brazil

Pathogenic microbial detection and control in laboratory animal facilities is essential to guarantee animal welfare, data validity and reproducibility. Helicobacter spp. are known to affect mice health, what may interfere with experimental outcomes. This study aimed to screen for Helicobacter spp. in mice from animal facilities in Rio de Janeiro, Brazil using a PCR-based method. Primers designed to specifi cally identify Helicobacter spp. were used to amplify feces or intestine DNA extracted of mice from four different animal facilities. The expected 375 base pairs (bp) amplicon was purifi ed, sequenced and a similarity of 95% was observed when compared to deposited sequences of H. hepaticus and H. bilis. In our screening, Helicobacter spp. was detected in ~59% of fecal and ~70% of intestine samples. Our study is the fi rst to screen for Helicobacter spp. in mouse facilities of a Rio de Janeiro University using a low cost, rapid molecular diagnostic test. Although Helicobacter spp. screening is not mandatory according to Brazilian animal welfare regulation it is recommended by institutional animal health monitoring programs guidelines worldwide, including ARRIVE, AAALAC and FELASA.


INTRODUCTION
The use of rodents as experimental models in basic and pre-clinical research has been essential for scientifi c progress. Reproducible research requires laboratory animals free of diseases and other conditions that could interfere with experimental outcome. Infections that naturally occur in mice, even when subclinical, may infl uence animal's physiology, immunity and behavior. Therefore, even in the absence of clinical signs, experimental rodents may become inadequate for research due to microorganism infections that may be a threat mainly to immunosuppressed mice (Kullberg et al. 1998, Whary & Fox 2006, Besselsen et al. 2008, Sharp et al. 2008, Chichlowski & Hale 2009, Pritchett-Corning et al. 2009, Mahler Convenor et al. 2014. Although Helicobacter infection in C57/BL6 mice are usually asymptomatic, other strains such as A/JCr, BALB/c, SJL, B6C3F1, C3H/ He, Rag-deficient and SCID may developed inflammatory bowel disease (IBD), intestinal cancer, hepatitis, and hepatocellular carcinoma (Whary & Fox 2006, Fox et al. 2011. Therefore, the establishment of animal health status monitoring routines is indispensable (Mahler Convenor et al. 2014, Bracken et al. 2017. Helicobacter bacteria, a genus that includes about 45 species, are among the microorganisms that infect laboratory rodents (Mahler Convenor et al. 2014, Pere-Vedrenne et al. 2017, Menard & Smet 2019. While in humans, Helicobacter pylori infection is associated with gastric diseases, in rodents Helicobacter spp. may infect the intestine (cecum and colon), stomach, gallbladder and liver; causing diseases in specific mouse strains (Chichlowski & Hale 2009, Pere-Vedrenne et al. 2017, de Brito et al. 2019. The diseasecausing species in susceptible mice include H. rodentium, H. bilis. H. ganmani, H. muridarum and H. typhlonius (Duangchanchot et al. 2014, Pere-Vedrenne et al. 2017, Menard & Smet 2019 Helicobacter spp. are highly prevalent bacteria in animal facilities worldwide (Taylor et al. 2007, Wasimuddin et al. 2012) and were previously detected in wild rodents in Brazil (Comunian et al. 2006). To our knowledge, the prevalence of Helicobacter infection in laboratory rodents in Brazil has never been studied. The present study aimed to detect Helicobacter spp. bacteria in rodents of animal facilities of the Universidade Federal do Rio de Janeiro (UFRJ), Brazil using a PCR-based diagnostic test.

MATERIALS AND METHODS
BALB/c or C57BL/6 mice between 15 and 22 weeks of age (n=57; mean age=19 weeks) were randomly selected from four different UFRJ animal facilities (AF-A -AF-D) ( Table I). According to Animal Ethics Committee of UFRJ protocol 092/15 mice were euthanized by CO 2 prior to colon collection. Recovered feces (1 fecal pellet/animal) and intestine sections were frozen in liquid nitrogen and maintained at -80 o C. For DNA extraction of fecal samples (about 1 cm pellet, ~50-80 mg) (Beckwith et al. 1997) or lower bowel (about 1cm of colon) samples were incubated in 100 µL of 25 mM NaOH (Isofar, 1326) for 1h at 98°C. Then, 400 µL of 10 mM pH 7.4 Tris buffer (Sigma, T1503) was added, debris were centrifuged and the samples were stored at -20 o C. Since feces may contain inhibitory substances that may lead to false negatives (Monteiro et al. 1997), we included DNA extracted from the feces of a mouse naturally infected with Helicobacter spp. (provided by Dr. Rovilson Gilioli, CEMIB, Unicamp) as positive control for the PCR reaction.

RESULTS AND DISCUSSION
First, we performed a PCR reaction using a DNA previously isolated from a mouse infected with Helicobacter ssp. as a positive control. Different amounts of DNA (0.01, 1, 10 and 100ng) were analyzed and a single amplicon of 375 bp was detected in the three higher concentrations of DNA tested (Figure 1). The successful amplification using 10 ng and a detectable weak band using 1 ng of DNA indicated a sensitivity that allows amplification of DNA extracted from feces or tissue. Analysis of the sequence showed a ~95% similarity between this amplicon and the sequence of the 16S rRNA gene of H. hepaticus, H. bilis and H. muridarum deposited in NCBI database, further validating the specificity of the PCR assay. To evaluate primers specificity to Helicobacter species, we extracted DNA from E. coli bacteria using the same NaOH protocol described above. Following PCR, no amplification was observed with the input of 1, 10 or 100 ng of E. coli DNA (Figure 2).
To analyze the presence of Helicobacter spp. in different animal facilities, we collected feces or intestine DNA samples from 57 randomly selected mice and performed PCR analysis (Figure 3). Due to sampling limitations, not all animals were tested for both feces and intestines. We found that 59.6% (28 of 47) of  (Table I). According to the PCR-based diagnostic tests of the feces, the occurrence of Helicobacter spp. infection in the different animal facilities (AF) ranged from 14% (AF-D) to 84% (AF-C). The analysis of intestine samples revealed a proportion of Helicobacter-infected mice ranging from 55% (AF-D) to 90% (AF-B) ( Table I). Although, the AF-D showed the lowest prevalence it still high index of contamination what may reflect a lack of an equivalent sanitary barrier as compared to other AF studied here (Mahler Convenor et al. 2014, Schlapp et al. 2018). The use of living animal models in research requires periodic screening, therefore, laboratory mice should be monitored for parasite, bacteria, fungi and virus, using different methods such as culture, microscopy, PCR and serology (Moerth et al. 2008, Mahler Convenor et al. 2014 Here, we used primers against 16S rRNA able to identify several Helicobacter species that may infect mice. PCR sensitivity was determined using serial dilutions of NaOHextracted DNA from feces and/or intestine. Efficient amplification demonstrated that less expensive methods of DNA extraction are useful for PCR-based screening tests of intestine and feces. In addition, screenings using a single Figure 1. Sensitivity of the PCRbased assay for Helicobacter spp. detection: DNA was extracted from feces of a mouse that was previously diagnosed with Helicobacter infection was used as the input of the PCR. Labels: MW (molecular weight) = 100 bp DNA ladder; 1 = 100 ng; 2 = 10 ng; 3 = 1 ng and 4 = 0.1 ng and (-) = negative control; 1% agarose gel (ethidium bromide).
pair of primers to detect several species of the same genus is advantageous because it minimizes time and costs (Battles et al. 1995, Shames et al. 1995. However, it should be noted that besides the four species investigated here, other Helicobacter species may infect rodents (Fox et al. 2011, Wasimuddin et al. 2012, Pere-Vedrenne et al. 2017, Menard & Smet 2019, thus it is important to adequate the choice of primers and PCR strategy. A high prevalence of Helicobacter spp.infected mice was observed in the facilities screened in this study. Overall, the percentage of positive samples was higher than 50%, therefore it is possible that a widespread contamination happened in the animal facilities studied. The successful use of feces is convenient, because it avoids euthanasia, however a study that performed side-by-side comparisons reported that in some cases feces is negative for Helicobacter while intestines are positive (Cao et al. 2020). More specifically, we observed that AF-D presented the lowest level of Helicobacter DNA positivity in the intestine (55.6%) and in the feces samples (14%). The difference between intestine and feces positivity may be because Helicobacter DNA detection may vary during infection and/or colonization. In a controlled experimental infection with H. hepaticus, bacterial DNA was detected in the feces of 100% of the animals at day 14 post-infection, while in the intestine and liver, bacterial DNA was detected only after 60 and 180 days, respectively (Cao et al. 2020). Thus, since the animals used in this study were chosen randomly it is impossible to determine the period of Helicobacter infection, and it is reasonable to assume differences in DNA availability in different periods of infection. Therefore, the use of both materials may be adequate. In addition, it is important to highlight that given the small differences in positivity within the facilities analyzed and the fact that only two mice strains were screened, our data does not allow conclusions about associations between infection rates and mouse genetic background.
Regarding the relevance of our findings to basic research using mice as a model, it is now clear that different Helicobacter spp. may cause inflammatory bowel disease (IDB) in mice (Fox et al. 2011). Infection with H. hepaticus also led to colitis, which was associated with the increase of pro-inflammatory cytokines (Kullberg et al. 1998(Kullberg et al. , 2006. Interestingly, infection with H. pylori may lead to the opposite consequences in IBD. Several studies have shown that H. pylori infection in mice may protect against IBD due to inflammatory down-regulation and immunological tolerance (Yu et al. 2018). Thus, it is crucial to determine the status of Helicobacter infection for studies modeling IBD in mice.
Notably, international councils, such as FELASA, ARRIVE, AALAS and AALAC, recommend the use of sentinel's animals for the health screening routine of rodent colonies (Lipman & Homberger 2003, Kilkenny et al. 2010, Newcomer 2012, Mahler Convenor et al. 2014. While the Brazilian National Council of Animal Experimentation Control (CONCEA 2016) highlights the importance of a health monitoring programs, it does not recommend the use of sentinel's animals. Taken into consideration that few companies provide services of microbiological screening tests for laboratory rodents in Brazil, our findings highlight the need for the development of accessible monitoring routines in order to prevent the use of infected animals.