HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Stewart, J. A Articles by Murray, A. Search for Related Content PubMed PubMed Citation Articles by Stewart, J. A Articles by Murray, A. Agricola Articles by Stewart, J. A Articles by Murray, A.

Investigations into the influence of host genetics on the predominant eubacteria in the faecal microflora of children

¹Wakefield Gastroenterology Research Institute, Wakefield Hospital, Private Bag 7909, Wellington South, New Zealand ²Institute of Veterinary, Animal & Biomedical Sciences, Massey University, Palmerston North, New Zealand

Correspondence Jessica A. Stewart jess-stewart{at}paradise.net.nz

Received 5 June 2005
Accepted 16 August 2005

The eubacterial population was studied in faecal samples of related and unrelated children. Temporal temperature gradient gel electrophoresis (TTGE) provided a snapshot of the bacterial population and allowed calculation of the degree of similarity in the predominant faecal microflora of identical twin pairs, fraternal twin pairs and unrelated paired controls. The highest levels of similarity were found in genetically identical twins. Significant differences were observed between the identical and fraternal twins (P = 0.037), strongly suggesting a genetic influence over the composition of the faecal microflora. The unrelated control group had the lowest similarity and was significantly different from the twins (P = 0.001). The results of this study indicate that host genetics influence the composition of the dominant eubacterial population in children.

	INTRODUCTION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Composed of 10¹¹ bacteria per gram of faeces and as many as 800 different bacterial species (Backhed et al., 2005), the faecal microflora inhabiting the human colon is a large and diverse population of micro-organisms in close contact with the human host. The large intestine is sterile at birth, but becomes rapidly colonized with bacteria derived from the maternal microflora and the environment. The population evolves over time; however, by the age of 2 years it has reached a stable climatic community that resembles the adult microflora (Conway, 1995). The faecal microflora is highly stable within one individual over time, and a unique population is found in each individual (Zoetendal et al., 1998).

Little is known of the bacteria–host and bacteria–bacteria interactions that are required to allow colonization and succession in the community. It has been hypothesized that an initial nutrient foundation provided by the host may determine the pattern of colonization at weaning (Hooper et al., 1999). A study of the dominant eubacteria amongst adults of varying degrees of relatedness found that the host genotype has a significant effect on determining the species composition of the population (Zoetendal et al., 2001). There is further evidence to suggest that the host may genetically determine the carriage of specific bacterial species. Each individual harbours their own set of specific strains of Lactobacillus and Bifidobacterium spp. (McCartney et al., 1996), and studies in mice have shown that the major histocompatibility complex can influence the composition of the murine faecal microflora (Toivanen et al., 2001).

The large number of uncharacterized species present in the gut limits the analysis of this population using traditional culture techniques. Temporal temperature gradient gel electrophoresis (TTGE) is a rapid molecular technique that avoids the need to grow organisms in the laboratory and is ideally suited for the analysis of microbial communities in the gut. In this study, TTGE was used to investigate the influence of host genetic control over the faecal microflora in children. The degree of similarity in the predominant eubacterial populations of identical twin pairs, fraternal twin pairs and unrelated control pairs was determined.

	METHODS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Human volunteers.

Thirteen identical twin pairs, seven fraternal twin pairs and 12 unrelated control pairs provided a single faecal sample for analysis. The age of the volunteers ranged from 4 months to 10 years, with a median of 23 months. The 24 unrelated controls were divided into groups of breast- and/or formula-fed infants and weaned children. Within these groups individuals were approximately age-matched to form unrelated control pairs, generating a median age difference of 1.5 months amongst infants and 1.1 years amongst weaned children. These unrelated infants and children were all living in separate home environments. All the genetically related volunteers were living in the same home environment at the time of sample collection. This study was carried out under ethical approval granted by the Wellington Ethics Committee (application number 99/98).

Twin zygosity.

Parents were asked to indicate if their children were identical or fraternal twins. Monozygosity of all identical twins was confirmed by genetic analysis of buccal swabs, carried out commercially by DNA Diagnostics. Twins were considered monozygous if DNA matches were obtained at all 15 loci examined (D8S1179, D21S11, D7S820, CSF1PO, D3S1358, TH01, D13S317, D16S539, D2S1338, D19S433, vWA, TPOX, D18S51, D5S818 and FGA). Twins that failed to complete zygosity testing were excluded from the study.

Faecal sample collection and DNA extraction.

Samples were collected in standard faecal tubes (LabServ) and stored at –20 °C. Samples were processed within 48 h. Potassium phosphate buffer (0.05 M, pH 7) was added to the faecal sample in a ratio of 1.3 ml to 1 g of faeces. Samples were vortexed with glass beads for 10 min to thoroughly homogenize the sample. The QIAmp DNA Stool Mini Kit (Qiagen) was used to extract bacterial genomic DNA from 500 µl of homogenate following the manufacturer's instructions. DNA samples were visualized on ethidium-bromide-stained 1.5 % agarose gels under UV light.

PCR-TTGE.

Primers U968-GC (CGCCCGGGGCGCGCCCCGGGCGGGGCGGGGGCACGGGGGGAACGCGAAGAACCTTAC) and L1401 (GCGTGTGTACAAGACCC) (Nubel et al., 1996) were used to amplify the V6-V8 region of the 16S rRNA gene of eubacteria. Faecal DNA samples were amplified in 50 µl reactions using the Qiagen HotStarTaq PCR kit. The reactions were carried out with 2.5 mM MgCl₂, 200 µM dNTPs and 0.2 µM of each of the forward and reverse primers. Cycling was carried out in an MJ Research thermal cycler under the following conditions: initial denaturation and Taq polymerase activation at 95 °C for 15 min, followed by 36 cycles of denaturation at 95 °C for 30 s, annealing at 63 °C for 30 s and extension at 72 °C for 35 s. A final extension was carried out at 72 °C for 7 min. PCR products were visualized on ethidium-bromide-stained, 1.5 % agarose gels under UV light. TTGE was carried out using the Bio-Rad DCODE Mutation Detection System. PCR products were separated on 6 % polyacrylamide gels containing 8 M urea. Electrophoresis was carried out at 80 V for 14 h and 20 min, with the temperature increasing by 0.7 °C h^–1 from 56 °C to 66 °C.

Staining and analysis of TTGE gels.

Standard methods were used to silver stain the TTGE gels, which were subsequently analysed using Kodak 1D image analysis software. Comparisons of TTGE profiles were made using Sorenson's similarity coefficient, and community richness of TTGE profiles was measured using Shannon's Index as described elsewhere (Simpson et al., 1999). Pearson product moment correlation coefficients and t tests were calculated using MiniTab (version 14).

	RESULTS AND DISCUSSION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Eubacterial TTGE gel profiles of related and unrelated children

TTGE profile comparisons were made between identical twin pairs, fraternal twin pairs and unrelated individuals. While every individual's profile was unique, increased levels of similarity were evident in TTGE profiles of genetically related individuals (Fig. 1). The number of bands per TTGE profile ranged from eight to 28, with a median of 16 bands. There was a positive correlation between the number of bands and the age of the volunteer (correlation coefficient = 0.49, P < 0.001). Shannon Weaver analysis also revealed a positive correlation between community richness and age (correlation coefficient = 0.361, P = 0.003). This shows that the predominant eubacterial population evident in TTGE banding profiles is more complex in older children. This is likely to reflect the increasing complexity present in the microflora with age (Hopkins et al., 2001).

View larger version (105K): [in this window] [in a new window]   Fig. 1. Examples of TTGE profiles of the predominant eubacterial population present in faecal samples of twins and unrelated control pairs of children. All pairs harboured some common TTGE bands. However, the degree of similarity of TTGE profiles was most marked in identical twin pairs. Identical twins: pair 1, 82.14 % similarity, 2 years, males; pair 2, 75.68 % similarity, 0.58 years, females; pair 3, 91.30 % similarity, 0.58 years, females. Fraternal twins: pair 1, 68.18 % similarity, 1.25 years, males; pair 2, 66.67 % similarity, 5 years, mixed genders; pair 3, 63.64 % similarity, 10 years, mixed genders. Unrelated control pairs: pair 1, 66.67 % similarity, 6 years, mixed genders; pair 2, 34.78 % similarity, 1.42 years, males; pair 3, 46.81 % similarity, 2 years, mixed genders.

Determination and comparison of TTGE profile similarity coefficients

Sorenson's similarity coefficient was calculated to assess the similarity of banding patterns, where 100 % describes complete identity and 0 % indicates no common bands between the two TTGE profiles (Fig. 2). For identical twin pairs, similarity values ranged from 69 to 91 %, with a median value of 82 %. Fraternal twin pairs’ similarity ranged from 55 to 87 %, with a median value of 68 %. The lowest similarity was seen amongst unrelated individuals, with a median value of 45 % and a range from 22 to 71 %. Analysis using t tests demonstrated significant differences between identical and fraternal twins (P = 0.037) and between fraternal twins and unrelated individuals (P = 0.001).

View larger version (21K): [in this window] [in a new window]   Fig. 2. Box and whisker plots of TTGE profiles of the predominant eubacterial populations for each group. Similarity of 100 % corresponds to identical TTGE profiles. All groups overlapped; however, the similarity coefficients increased with increased host genetic similarity. The differences between the groups were statistically significant (P 0.05).

It has already been established that there is a positive correlation between the similarity of the predominant eubacterial population and the degree of relatedness in adults (Zoetendal et al., 2001). Undoubtedly this contributes to the intra-individual stability and inter-individual variability (Zoetendal et al., 1998) that is characteristic of this population. The work described here further supports this premise and demonstrates a significant difference between TTGE similarity values in identical twin pairs and fraternal twin pairs, and suggests that genetic influence from the host functions from an early age. These differences cannot be explained by intra-assay or intra-individual variation, which were determined to be < 5 % and < 12 %, respectively.

Twin pairs undergo simultaneous development of the microflora. Although some environmental variability may occur, the environment is likely to be highly similar for twins, which may lead to an overestimation of the effect of host genetic control. In an attempt to address this issue we also compared the predominant eubacteria amongst seven sibling pairs (three adult sibling pairs, living apart, median age 25 years; four child sibling pairs, living together, median age 3.5 years) and three mother–child pairs (living apart, median age of children 25 years). Like fraternal twins, these 10 pairs have approximately 50 % of their genes in common; however, the microflora of these individuals evolved at different times and therefore with greater environmental variability than amongst twins, for example food types and amounts, infections and environmental sources of microbes. Comparisons between fraternal twins and the group of sibling pairs and mother–child pairs found no significant differences in the TTGE profile similarity values (P = 0.10). This suggests that the highly homogeneous environment found amongst twin pairs may not be a major factor determining the increased similarity amongst these groups.

Analysis of similarity coefficients with respect to age, gender and diet

Similarity coefficients were examined with respect to the age, gender and diet of the paired individuals (Table 1). There was no correlation between the gender of the twin pairs or unrelated controls pairs and their TTGE similarity coefficients.

From the dietary information available, no influence of breast feeding or formula feeding on TTGE similarity coefficients was evident. Comparisons between breast-fed and formula-fed twins and weaned twins found no significant difference. The diets of weaned twins were classified as identical, similar or different by their parents. There was no significant correlation or significant difference in the TTGE similarity values of twins with identical diets and twins with similar diets. Amongst unrelated children, the two pairs that were formula-fed did not demonstrate increased TTGE profile similarity compared to the pairs consuming different diets.

There was no correlation between TTGE similarity coefficients and the age of the volunteers amongst identical twin pairs and unrelated control pairs. Analysis of the age difference between unrelated control pairs with respect to TTGE similarity coefficients revealed no correlation. A large negative correlation between similarity and age was evident in fraternal twins and this was statistically significant (correlation coefficient = –0.825, P = 0.022). Although the numbers of identical and fraternal twins in different age groups is small, one could speculate that identical twin pairs may maintain a constant level of high similarity during development of the microflora as they are under the same genetic constraints, while a genetic effect in infant fraternal twins may be masked due to low environmental variability. Increased environmental exposure to micro-organisms as fraternal twins grow older, may permit increased divergence due to their different genetic backgrounds. A longitudinal study following the development of the microflora in infant twins, with adequate controls for factors such as diet and gender, is essential to investigate this hypothesis.

This work provides evidence that host genetics impact the composition of the predominant faecal eubacteria in children. Given the complexity of the microflora, the difficulty in controlling environmental variables, and the assumption that many host genes are likely to be involved in this process, attempts to unravel the genetic mechanisms responsible for this process in humans will be challenging. Evidence collected in mouse studies suggests a role for MHC genes in host control of the faecal microflora composition (Toivanen et al., 2001).

Currently much research is underway investigating manipulations of the microflora with prebiotics and probiotics. Increasing evidence for genetic influence over members of the microflora suggests that host factors may influence the ability of an organism to colonize different individuals, and therefore may have implications for attempts to therapeutically manipulate the microflora.

	ACKNOWLEDGEMENTS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
This work was supported by Wellington Medical Research Foundation Grant 2003/75, and the Wakefield Gastroenterology Research Trust.

	REFERENCES

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES