Comparison of methods and animal models commonly used for investigation of fecal microbiota: Effects of time, host and gender

https://doi.org/10.1016/j.mimet.2005.10.014Get rights and content

Abstract

Denaturing gradient gel electrophoresis (DGGE), terminal restriction fragment length polymorphism (T-RFLP) and plating on selective agars were used to study variation in the fecal microbiota of rats over time as well as variation between individuals. Investigated rats were either conventional and specific pathogen free (SPF), or human flora associated (HFA). A higher variation (p < 0.05) in fecal microbiota over time was observed for HFA than for SPF animals. Analysis of DGGE and T-RFLP profiles of fecal microbiota from SPF and HFA rats revealed that variation over time was less significant than variation between individuals, and that phylogenetic profiles clustered according to gender. These observations should be taken into account when designing future research addressing changes in fecal microbiota.

Introduction

While many studies address effects of endogenous and exogenous factors on the composition of the gastrointestinal microbiota (Djouzi et al., 1997, Fujiwara et al., 2001, Mai et al., 2004, Macpherson and Harris, 2004, Goossens et al., 2005), only limited data are available on the (unprovoked) variation of intestinal microbiota in animal models. Nevertheless, knowledge about variations in the intestinal microbiota (over time and between individuals) prior to intervention is often an important prerequisite for design and interpretation of cause–effect studies. The rationale for the present study was to provide such knowledge.

The intestinal and fecal microbiota of mammals is dominated by relatively few bacterial divisions, that are highly diverse at the strain/subspecies level (Backhed et al., 2005). Feces contains up to 1012 bacteria per gram (Kleessen et al., 2000), and is estimated to harbor more than 800 different microbial species (Backhed et al., 2005). Comparison of culture-based and molecular methods reveals that only 20–50% of gut microbes can be detected by cultivation (Wilson and Blitchington, 1996, Suau et al., 1999). Therefore, various molecular methods for studies of complex microbial communities have been applied, many of which involve amplification by polymerase chain reaction (PCR) of 16S ribosomal RNA genes from extracted microbial DNA (Tannock, 2001, Zoetendal et al., 2004). Separation and analysis of amplified DNA fragments can be done using e.g., denaturing gradient gel electrophoresis (DGGE) or terminal restriction fragment length polymorphism (T-RFLP) (Zoetendal et al., 2004).

We have used DGGE separation and T-RFLP analysis to study variation over time and between individuals of the fecal microbiota of rats. Additionally, conventional cultivation on selective agars was carried out. We find that our results are relevant for design and interpretation of future studies addressing effects of a given intervention on the fecal microbiota of experimental animals.

Section snippets

Specific pathogen free (SPF) rats

Ten conventional, specific pathogen free (SPF) Sprague–Dawley rats were purchased from Møllegaarden Breeding Centre, Denmark. Animals (six males and four females, aged 9 to 10 weeks at arrival and originating from two litters), were housed and fed as previously described (Schlundt et al., 1994), but were caged individually.

Human flora associated (HFA) rats

Eight germfree Sprague–Dawley rats (four males and four females, aged six months, and originating from a single litter) were used. Rats were bred at the Danish Institute for

Comparison of the fecal microbiota of two animal models

Comparison of SPF and HFA rats by selective plating for five bacterial groups revealed that a higher number of HFA animals had a significant (p < 0.05) variation of the fecal microbiota over time than was observed for SPF animals (Table 1). Variations observed in HFA animals were mainly attributed to fluctuations in total bacterial densities as well as in populations of Enterococcus/Streptococcus, and to a general decrease in the Lactobacillus population over time (Fig. 1).

Comparison of SPF rats

Discussion

Two important prerequisites for studies addressing effects of defined actions/agents on the intestinal microbiota of experimental animal models is (i) knowledge about variation naturally occurring in such models, and (ii) knowledge about the usefulness of the applied methods to detect such variations. While traditional selective plating offers the advantage of rapid sample processing, and direct linking of results to a given bacterial genus or species, molecular methods are needed to

Acknowledgements

This work was financed by the Danish FØTEK 3 program (3401-44-03-691) and The Danish Bacon and Meat Council. The work was conducted in collaboration with Chr. Hansen A/S, Danisco A/S, and the Danish Toxicology Centre. We thank Bodil Madsen and Linda Bergstrøm for excellent technical assistance.

References (25)

  • L.V. Holdeman et al.

    Human fecal flora: variation in bacterial composition within individuals and a possible effect of emotional stress

    Appl. Environ. Microbiol.

    (1976)
  • B. Kleessen et al.

    Culture-based knowledge on biodiversity, development and stability of human gastrointestinal microflora

    Microb. Ecol. Health Dis.

    (2000)
  • Cited by (39)

    • Effect of Lactobacillus reuteri on intestinal microbiota and immune parameters: Involvement of sex differences

      2019, Journal of Functional Foods
      Citation Excerpt :

      Recently, sexual dimorphism is widespread among animals, and has triggered researchers’ interest. Studies have found sex-associated differences in relation to the immune system, gut community composition (Bernbom et al., 2006) and metabolic activity (Mcorist et al., 2011). A sexual dimorphism in immune response is seemly obvious: females have a stronger innate and adaptive immune response as compared with males, which mainly contributed to sex hormones (Bouman, Heineman, & Faas, 2005; Hewagama, Patel, Yarlagadda, Strickland, & Richardson, 2009).

    • Age, sex, and TNF associated differences in the gut microbiota of mice and their impact on acute TNBS colitis

      2017, Experimental and Molecular Pathology
      Citation Excerpt :

      Despite the benefits that mouse models provide, such as allowing for more invasive studies, variable endpoints, and reducing individual variation in the microbiota that is seen in humans; descriptions of IBD-associated differences in the gut microbiome have been inconsistent in mouse models. In laboratory mice, environmental variables such as housing conditions, facility of origin, variations in handling, time of sampling, antibiotic use, and diet have all been shown to contribute to differences in the gut microbiota (Moore and Stanley, 2016; Hufeldt et al., 2010; Bernbom et al., 2006; Bolnick et al., 2014; Thaiss et al., 2014; Voigt et al., 2014). Additionally, marked differences in the microbiota have been described in mice with different genetic backgrounds.(

    • Development and genetic influence of the rectal bacterial flora of newborn calves

      2012, Veterinary Microbiology
      Citation Excerpt :

      For pigs, within one individual, the intestinal microbiota showed temporal variations up to 28 d and a significant increase in stability beyond 30 d (Thompson et al., 2008). Interestingly, Bernborn et al. (2006) reported that samples taken from the same individual animal over time clustered together in principal component analysis. They concluded, in agreement with our findings, that the composition of intestinal microbiota is determined by host factors rather than by environmental factors.

    View all citing articles on Scopus
    View full text