• bovine rumen bacteria;
  • feed efficiency;
  • volatile fatty acid


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Linkage of rumen microbial structure to host phenotypical traits may enhance the understanding of host–microbial interactions in livestock species. This study used culture-independent PCR-denaturing gradient gel electrophoresis (PCR-DGGE) to investigate the microbial profiles in the rumen of cattle differing in feed efficiency. The analysis of detectable bacterial PCR-DGGE profiles showed that the profiles generated from efficient steers clustered together and were clearly separated from those obtained from inefficient steers, indicating that specific bacterial groups may only inhabit in efficient steers. In addition, the bacterial profiles were more likely clustered within a certain breed, suggesting that host genetics may play an important role in rumen microbial structure. The correlations between the concentrations of volatile fatty acids and feed efficiency traits were also observed. Significantly higher concentrations of butyrate (P<0.001) and valerate (P=0.006) were detected in the efficient steers. Our results revealed potential associations between the detectable rumen microbiota and its fermentation parameters with the feed efficiency of cattle.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Ruminal fermentation of ingested feed plays a major role in the supply of energy for metabolic functions in cattle and other ruminants. Yet little is known about the association between specific ruminal microbial population and the efficiency with which feed is converted into energy for the maintenance and growth of host. The rumen microbiota is highly responsive to changes in diet, age, antibiotic use, and the health of the host animal and varies according to the geographical location, season, and feeding regimen (Stewart et al., 1997). The rumen microbial ecosystem comprises a diverse symbiotic population of anaerobic bacteria, archaea, ciliated protozoa, and fungi (Krause & Russell, 1996; Krause et al., 2003), of which only 10–50% of ruminal bacteria have been isolated and identified (Kobayashi, 2006). Rumen bacteria are known to colonize in the rumen soon after the birth and contribute to carbon and nitrogen metabolism through their fermentation. However, the data on the composition and activity of the rumen bacteria that are currently available do not enable links to be established between the biofunction of the host and the microbial composition and metabolic activities. Several studies have used molecular based culture-independent techniques to investigate the profiles of bacteria in the rumen (Edwards et al., 2005; McEwan et al., 2005), but a few studies on bovine rumen bacteria and their correlations with the host's biology have been published.

As has been proposed as a measure of feed efficiency of cattle, residual feed intake (RFI) is defined as the difference between an animal's actual feed intake and its expected feed requirements for maintenance and growth (Archer et al., 1999; Basarab et al., 2003). RFI is moderately heritable and independent of growth and body size. The more efficient animals, which are cattle with low-RFI (L-RFI), are expected to have reduced feed intake but performance similar to cattle with high-RFI (H-RFI), inefficient animals. To date, the relationship among the composition of the rumen microbiota, the microbial fermentation parameters, and feed efficiency in cattle has not been studied and reported. In this study, we hypothesized that the diversity of ruminal microorganisms and the concentration of volatile fatty acids (VFA), one of the fermentation parameters in the rumen, are associated with the feed efficiency (RFI) of cattle. The culture-independent PCR-denaturing gradient gel electrophoresis (PCR-DGGE) analysis was applied to compare the dominant bacterial diversity in rumen of steers with different RFI. The relationships among ruminal bacterial diversity, VFA, and feed efficiency trait were then evaluated by correlation of bacterial PCR-DGGE profiles, and VFA concentrations to RFI values.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Animal handling and sampling

Sampling of rumen fluid

All experimental procedures were approved by the Animal Care and Use Committee for Livestock at the University of Alberta and steers were raised following the guidelines of the Canadian Council on Animal Care (Ottawa, ON) under feedlot conditions at the Kinsella Research Station, University of Alberta, on a finishing diet as described by Nkrumah et al. (2006). Feeding intake data were collected using the GrowSafe automated feeding system (GrowSafe Systems Ltd, Airdrie, AB, Canada); a total mixed finishing composed of c. 64.5% barley grain, 20% oats, 9% alfalfa hay, 5% feedlot supplement [32% CP beef supplement containing Rumensin (400 mg kg−1)], and 1.5% canola oil (Basarab et al., 2003). The feed efficiencies of steers were ranked as inefficient (H-RFI, RFI>0.5) or efficient (L-RFI, RFI<−0.5), based on calculated RFI values as described by Nkrumah et al. (2006). Eighteen steers were selected for this study based on extreme differences in the RFI value (Table 1). Rumen sampling was performed within 1 week after RFI evaluation. Ruminal fluid was collected before feeding by inducing flexible plastic tubing into the rumen and using the suction created with a 50-mL syringe to remove the fluid from the tubing. For each animal, 100 mL of rumen fluid was collected twice and transferred into a separate sterilized container, immediately frozen with liquid nitrogen, and stored at −80 °C until processing.

Table 1.   Information of tested cattle and their related RFI values
Animal ID (tag number)BreedAge (days)RFI value (kg day−1)RFI rankingRumen sample resource
  • *

    Animals were raised in Kinsella Ranch, University of Alberta.

  • Animals were raised in Lacombe Research Station, Alberta; the bold letters were used as sample ID in PCR-DGGE analysis.

  • ANG, Angus; HYB, Hybrid; CHA, Charolais; HEAN, Hereford-Angus.

Sampling of rumen digesta

The rumen digesta including liquid and solid parts from various locations of the rumen were collected from 13 steers, which were ranked with different RFI based on feed efficiency tests performed at the Lacombe Research Center of Alberta Agriculture, Food and Rural Development on a finishing diet composed of 73.3% barley grain, 22.0% barley silage, 1.6% molasses, and 3.1% feedlot supplement [32% CP beef supplement containing Rumensin (400 mg kg−1) fed ad libitum]. Three groups of animals were defined based on their RFI value: L-RFI (RFI<−0.5), medium-RFI (M-RFI, −0.5<RFI<0.5), and H-RFI (RFI>0.5) (Table 1). The rumen content (c. 100 g each) was collected immediately after animal was slaughtered, frozen with liquid nitrogen, and stored at −80 °C until processing.

DNA extraction

DNA extraction from rumen fluid

Duplicate rumen fluid samples were pooled for total DNA extraction. The rumen fluid was thawed and centrifuged at 200 g for 5 min to remove the particulate matter. Total DNA was extracted using a bead-beating method as described by Walter et al. (2000). Briefly, the supernatant obtained was transferred to a 2-mL microcentrifuge tube containing zirconium beads (0.3 g, diameter 0.1 mm), followed by a physical disruption in a BioSpec Mini Bead-Beater-8 at 4800 r.p.m. for 3 min. After phenol–chloroform–isoamyl alcohol (25 : 24 : 1) extraction, DNA was precipitated with cold ethanol and dissolved in 30 μL of TE buffer [10 mM Tris-HCl, 1 mM EDTA (pH 8.0)]. The amount and quality of DNA were measured at the A260 nm and A280 nm using an ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington).

DNA extraction from rumen digesta material

The rumen solid samples were thawed and 1 g of each was transferred to a 15-mL tube, 9 mL of lysozyme solution (100 mg mL−1) was added to the sample, vortexed vigorously, and incubated at room temperature for 30 min. After centrifuging at 200 g for 5 min, 1 mL of the supernatant was transferred to a 2-mL microcentrifuge tube containing zirconium beads (0.3 g, diameter 0.1 mm) followed by the steps as described above.

PCR-DGGE analysis

Individual total DNA extracted from rumen samples was diluted to a concentration of 50 ng μL−1 and 1 μL of diluted DNA was used as a template in all PCR reactions. The PCR-DGGE analysis of the total detectable bacteria was performed using universal bacterial HDA-1GC and HDA2 primers, (HDA1-GC, 5′-CGCCCGGGGCGCGCCCCGGGCGGGGCGGGGGCACGGGGGGACTCCTACGGGAGGCAGCAGT-3′; HDA2, 5′-GTATTACCGCGGCTGCTGGCAC-3′) (Walter et al., 2000). PCR products (c. 200 bp) were amplified using the program as outlined by Walter et al. (2000) and were subjected to run on a 6% acrylamide gel with 22–55% gradient using the Bio-Rad DCode Universal Mutation Detection System (Hercules, CA). The gel was run at 130 V for 4 h and, after electrophoresis, gel was stained with ethidium bromide, viewed with UV transillumination, and photographed.

Cluster analysis

The similarity or the difference of the ruminal bacterial structure was determined by comparison and clustering of the whole profiles generated from PCR-DGGE with the bionumerics software package (Applied Maths, Austin, TX). The total number of bands in each sample pattern is related to the number of dominant phenotypes. Similarity matrices were produced using the Dice coefficient, which allowed construction of dendrograms using the unweighted pairwise grouping method with the mathematical averages (UPGMA) clustering algorithm (Fromin et al., 2002; Nicol et al., 2003). The similarity of the PCR-DGGE profiles was compared and clustered as shown by dendrograms generated based on the Dsc of bacterial profiles representing by band immigration patterns using the 2% tolerance position. The similarity of comparison of each profile was obtained in percentage based on the clustering results. Average Dice's similarity coefficient (Dsc) (Knarreborg et al., 2002; Guan et al., 2003) was used to compare the profiles between the L-RFI and the H-RFI group or within each group following the calculation by adding the values of single profile comparisons (similarity in percentage) for the L-RFI or H-RFI individual range stated and dividing by the total number of Dsc values.

VFA measurement and statistical analysis

The duplicate rumen fluid samples from each animal were centrifuged and the supernatant (1 mL) was transferred to a GC vial and mixed with 200 μL of 25% phosphoric acid and 200 μL of isocaproic acid solution (Internal standard, 3 mg mL−1). A standard solution containing acetate, propionate, isobutyrate, butyrate, isovalerate, valerate, and caproic acid was prepared by dissolving 1 g of each compound in 100 mL of 25% phosphoric acid (pH 3.0). The samples were run at a split ratio of 20 : 1, and at 120–170 °C at 10 °C min−1, with the temperature of the injector at 170 °C and the detector at 190 °C using a Stabilwax-DA column (30 m × 0.25 mm). Peak integration was performed using galaxie Software (Varian). A linear mixed model was fitted to determine the differences between RFI groups in VFA concentration using the MIXED procedure of sas (version 9.1; SAS Institute Inc., Cary, NC). The statistical model included RFI group, sire breed (Angus, Charolais, and Hybrid), test group, replication, their possible interactions as fixed effects, and sire as a random effect. Differences in least squares means between RFI groups were declared at P<0.05 using the PDIFF option in SAS (Table 2).

Table 2.   Comparison of different traits and VFA concentrations in ruminal fluid between L- and H-RFI* steers (LS mean ± SE)
TraitL-RFI group (efficient)H-RFI group (inefficient)P-value
RFI, kg day−1−1.38 ± 0.141.40 ± 0.12< 0.001
Total VFA, mM96.74 ± 13.0755.35 ± 13.050.059
 Acetate, mM52.67 ± 6.7431.20 ± 6.730.074
 Propionate, mM25.02 ± 5.6018.04 ± 5.590.409
 Butyrate, mM14.54 ± 1.793.35 ± 1.79< 0.001
 Isobutyrate, mM0.86 ± 0.140.62 ± 0.140.327
 Valerate, mM1.69 ± 0.260.67 ± 0.260.006
 Isovalerate, mM1.95 ± 0.381.75 ± 0.370.774

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Correlation PCR-DGGE profiling of detectable ruminal bacteria to RFI

Recent developments in molecular-based methods such as PCR-DGGE analysis have allowed for the detection of a greater range of microorganisms in environments (Tannock, 2002; Avrahami et al., 2003; Bano et al., 2004; Edwards et al., 2005) for the PCR-DGGE-banding patterns are considered to represent the dominant bacterial groups (Muyzer et al., 1993). The bacterial PCR-DGGE profiles of 18 ruminal fluid samples showed that each steer harbored a characteristic ruminal bacterial community, demonstrated by the different band patterns (Fig. 1a). These profiles possibly represented the dominant bacterial fractions in the rumen fluid, with an individual band representing a discrete bacterial population (van Hannen et al., 1999). Clustering of these profiles showed that all bacterial profiles generated from L-RFI steers were grouped into closely related clusters and were separated from those profiles obtained from H-RFI steers (Fig. 1a). The average Dsc was 79.0% when compared between L-RFI and H-RFI groups. The average Dsc among L-RFI steers was 91.0%, exhibiting increased similarity, while the average Dsc among H-RFI steers was 71.1%, exhibiting decreased similarity when compared within each group. These data reveal that the bacterial components in the rumen of L-RFI steers are more similar to each other (91% similarity) than those in the rumen of H-RFI steers (71% similarity). This suggests that specific bacterial members may be present in L-RFI but absent in H-RFI steers.


Figure 1.  PCR-DGGE profiles generated from ruminal fluid DNA from 18 steers using primers HDA1-GC and HDA2 (22–55% DGGE). The sample ID is indicated beside each lane shown by the number; H and L represent the steers with H-RFI (H-RFI>0.5, inefficient), and L-RFI (L-RFI<−0.5, efficient), respectively. RFI a parameter to measure feed efficiency in cattle (Archer et al., 1999). The comparison of the PCR-DGGE profiles was generated with the bionumerics software package as described in the text for all tested 18 animals (a) and nine animals of the Angus breed (b). The animal breed was indicated as ANG, of Angus; HYB, of Hybrid; and CHA, of Charolais, respectively.

Download figure to PowerPoint

To confirm the above phenomenon, the rumen digesta samples were collected from another group of steers raised and tested for RFI in a different farm (Table 1). The detectable bacterial PCR-DGGE profiles of rumen digesta samples from six L-RFI, four M-RFI, and three H-RFI steers showed that bacterial profiles from L-RFI steers were grouped together and were separable from those from H-RFI steers, confirming a similar grouping trend observed from ruminal bacteria from the 18 animals (Fig. 2).


Figure 2.  PCR-DGGE profiles generated from rumen solid content DNA from 13 steers, using primers HDA1-GC and HDA2 (22–55% DGGE). The sample ID is indicated beside each lane shown by the number, H, M, and L represent the steers with H-RFI (H-RFI>0.5, inefficient), M-RFI (−0.5<RFI<0.5), and L-RFI (L-RFI<−0.5, efficient), respectively. The cluster dendrogram was generated with the bionumerics software package as described in the text. The animal breed was indicated as HEAN, of Hybrid breed from Hereford–Angus; and CHA, of Charolais, respectively.

Download figure to PowerPoint

When the PCR-DGGE profiles of rumen digesta samples were compared, it was found that two large clusters were formed and each cluster showed a specific breed correlation (Fig. 2). When we added breed information of 18 animals from the first experiment and correlated to the PCR-DGGE profiles, no direct correlation was observed, although the profiles generated from Hyb animals were different from those from ANG and CHA steers, except animal 111 (Fig. 1a). However, when the PCR-DGGE profiles from steers of Angus breed were compared, clear separation patterns were observed between H- and L- RFI animals (Fig. 1b). The reasons for not showing correlations with host breed (Fig. 1) may the following: (1) the genetic backgrounds of Hyb animals are too complicated for mixture of various breeds; (2) lack of containing of H-RFI animals with CHA breed. Combining both the above experiments, the existence of a probable correlation between the host genetic factor with the ruminal bacterial structure is indicated because all the steers were fed with the same diet and raised in the same environment under the same management in each experiment. The associations observed among PCR-DGGE profiles, RFI and the breeds reveal the potential characterization of bacteria at a DNA sequence level, which are genetically selected by the host and associated with RFI.

Owing to the small numbers of sampled animals and the limitation of rumen sampling, the data obtained in this study may only represent a partial bacterial diversity in the rumen of tested animals. Rumen samples from large numbers of steers, and/or the same steer at different growth stage, different time points of feeding and diurnal fluxes, and different sections of rumen location are in progress. The PCR-DGGE analysis could also be biased by the DNA extraction methods, PCR amplification, and limitation of the existing sequence information in the database. The characterization of bacteria by sequencing analysis of the PCR-DGGE bands generated in the clusters of L-RFI animals is in progress and to date, the sequences either belong to unidentified or unculturable rumen bacteria. For example, the clone from the band extracted from L-RFI PCR-DGGE profiles as indicated by arrows (Fig. 1a) was homological to the sequence of uncultured bacterium clone (Accession no. DQ449426). It is necessary to apply advanced techniques such as metagenomics and metatranscriptomics, and the development of the bioinfomatic database to fully understand the ruminal microbial communities and its functions. However, the identification of predominant bacteria consortium using PCR-DGGE profiling provides a rapid way to screen large numbers of individuals, which can supply a possible linkage of rumen microbiology, animal nutrition, and animal production by analysis of predominant microbial consortia that may be responsible for particular functions.

Profiling of VFA in the rumen fluid and correlation to RFI

The PCR-DGGE analysis of bacteria in both liquid and digesta ruminal samples revealed that certain species of bacteria may be associated with L-RFI steers. This suggests that the bacteria in the L-RFI steers are highly related, and probably have similar metabolic pathways, which may affect feed efficiency. The microbial activities in the rumen were evaluated by comparing one of the fermentation parameters, VFA. When the concentration of VFAs in L-RFI and H-RFI steers was compared, the following phenomena were observed (Table 1): (1) significantly higher concentrations of butyrate (P<0.001) and valerate (P=0.006) were found in the rumen of L-RFI steers; (2) a higher concentration of total VFA (P=0.059) and acetate (P=0.074) seemed to be present in the rumen fluid samples of L-RFI steers, and (3) no significant difference in the concentration of propionate and other VFAs was found between L-RFI and H-RFI steers (P>0.05).

Ruminal VFA concentrations are parameters indicative of active bacterial fermentation and host rumen epithelial absorption (Brockman, 1993). The comparison of total VFA between L-RFI and H-RFI steers showed a tendentiously higher concentration of total VFA and the presence of acetate in the rumen of the L-RFI group (Table 1), suggesting increased microbial fermentation. Butyrate is converted in the rumen wall to a ketone body, β-hydroxybutyrate, which is a very important energy resource for most tissues. The significantly higher concentrations of butyrate present in the L-RFI steers (Table 1) suggested that the higher feed efficiency in L-RFI steers may have resulted from a shift in bacteria population metabolizing feed substrate to products with greater usable energy values. The findings of higher digestible and metabolic energy detected in L-RFI steers (Nkrumah et al., 2006) indicate that more active microbial fermentation in L-RFI steers and differences in energy metabolism between L-RFI and H-RFI steers may exist. The differences in these fermentation parameters could also be the result of the interactions among ruminal bacteria, fungi, and protozoa because both fungi and protozoa can affect the bacterial activities (Kamra, 2005) as well as the microbial–microbial and microbial–host interactions. The PCR-DGGE analysis of detectable protozoa and fungi showed no significant difference (data not shown), suggesting that the VFA difference may be more associated with the bacterial structure in the rumen.

In conclusion, differences in rumen microorganisms were found to be likely associated with feed efficiency (RFI) in beef cattle. This is the first report to link rumen microbial PCR-DGGE profiles and their fermentation products with cattle's feed efficiency. Recent findings of the association between gut microbial structure with obesity traits in mice and humans (Ley et al., 2006; Turnbaugh et al., 2008) have revealed the importance of gut microbial communities in host's biology. Our studies on the identification of potential associations between rumen microbiota and cattle's feed efficiency trait will add valuable information to the field of identification of the diversity and functions of gut microbial communities relevant to physiological processes in the host.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

This work was supported by Grant #ACC-99AB343, #ASRA, AARI 2002L030R, #BCRC 2002L030R, and ABP/ACC awarded to S.S.M. through the Canada/Alberta Beef Industry Development Fund, Alberta Agricultural Research Institute, Beef Cattle Research Council, Alberta Beef Producers and Alberta Cattle Commission. The authors acknowledge the technical assistance of staff from the Metabolic Research Unit and Dr B. Irving for management of the steers at Kinsella ranch, University of Alberta. The authors also thank Dr M. Gaenzle for the support of DGGE analysis and Dr K. Lien for his kind assistance with GC analysis.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  • Archer JA, Richardson EC, Herd RM & Arthur PF (1999) Potential for selection to improve efficiency of feed use in beef cattle: a review. Aust J Agric Res 50: 147161.
  • Avrahami S, Liesack W & Conrad R (2003) Effects of temperature and fertilizer on activity and community structure of soil ammonia oxidizers. Environ Microbiol 5: 691705.
  • Bano N, Ruffin S, Ransom B & Hollibaugh JT (2004) Phylogenetic composition of Arctic Ocean archaeal assemblages and comparison with Antarctic assemblages. Appl Environ Microbiol 70: 781789.
  • Basarab JA, Price MA, Aalhus JL, Okine EK, Snelling WM & Lyle KL (2003) Residual feed intake and body composition in young growing cattle. Can J Anim Sci 83: 189204.
  • Brockman RP (1993) Glucose and short-chain fatty acid metabolism. Quantitative Aspects of Ruminant Digestion and Metabolism (ForbesJM & FranceJ, eds), pp. 249265. CAB International, London.
  • Edwards JE, Bequette BJ, McKain N, McEwan NR & Wallace RJ (2005) Influence of flavomycin on microbial numbers, microbial metabolism and gut tissue protein turnover in the digestive tract of sheep. Br J Nutr 94: 6470.
  • Fromin N, Hamelin J, Tarnawski S, Roesti D, Jourdain-Miserez K, Forestier N, Teyssier-Cuvelle S, Gillet F, Aragno M & Rossi P (2002) Statistical analysis of denaturing gel electrophoresis (DGE) fingerprinting patterns. Environ Microbiol 4: 634643.
  • Guan LL, Hagen KE, Tannock GW, Korver DR, Fasenko GM & Allison GE (2003) Detection and identification of Lactobacillus species in crops of broilers of different ages by using PCR-denaturing gradient gel electrophoresis and amplified ribosomal DNA restriction analysis. Appl Environ Microbiol 69: 67506757.
  • Kamra DN (2005) Rumen microbial ecosystem. Curr Sci 89: 124135.
  • Knarreborg A, Simon MA, Engberg RM, Jensen BB & Tannock GW (2002) Effects of dietary fat source and subtherapeutic levels of antibiotic on the bacterial community in the ileum of broiler chickens at various ages. Appl Environ Microbiol 68: 59185924.
  • Krause DO & Russell JB (1996) How many ruminal bacteria are there? J Dairy Sci 79: 14671475.
  • Krause DO, Denman SE, Mackie RI, Morrison M, Rae AL, Attwood GT & McSweeney CS (2003) Opportunities to improve fiber degradation in the rumen: microbiology, ecology, and genomics. FEMS Microbiol Rev 27: 663693.
  • Kobayashi Y (2006) Inclusion of novel bacteria in rumen microbiology: need for basic and applied science. Anim Sci J 77: 375385.
  • Ley RE, Turnbaugh PJ, Klein S & Gordon JI (2006) Microbial ecology: human gut microbes associated with obesity. Nature 444: 10221023.
  • McEwan NR, Abecia L, Regensbogenova M, Adam CL, Findlay PA & Newbold CJ (2005) Rumen microbial population dynamics in response to photoperiod. Lett Appl Microbiol 41: 97101.
  • Muyzer G, De Waal EC & Uitterlinden AG (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 59: 695700.
  • Nicol GW, Glover LA & Prosser JI (2003) Spatial analysis of archaeal community structure in grassland soil. Appl Environ Microbiol 69: 74207429.
  • Nkrumah JD, Okine EK, Mathison GW, Schmid K, Li C, Basarab JA, Price MA, Wang Z & Moore SS (2006) Relationships of feedlot feed efficiency, performance, and feeding behavior with metabolic rate, methane production, and energy partitioning in beef cattle. J Anim Sci 84: 145153.
  • Stewart CS, Flint HJ & Bryant MP (1997) The rumen bacteria. The Rumen Microbial Ecosystem, 2nd edn (HobsonPN & StewartCS, eds), pp. 1072. Springer-Verlag, New York.
  • Tannock GW (2002) Analysis of the intestinal microflora using molecular methods. Eur J Clin Nutr 56: (suppl 4): S44S49.
  • Turnbaugh PJ, Bäckhed F, Fulton L & Gordon JI (2008) Diet-induced obesity is linked to marked but reversible alterations in the mouse distal gut microbiome. Cell Host Microbe 3: 213223.
  • Van Hannen EJ, Zwart G, Van Agterveld MP, Gons HJ, Ebert J & Laanbroek HJ (1999) Changes in bacterial and eukaryotic community structure after mass lysis of filamentous cyanobacteria associated with viruses. Appl Environ Microbiol 65: 795801.
  • Walter J, Tannock GW, Tilsala-Timisjarvi A, Rodtong S, Loach DM, Munro K & Alatossava T (2000) Detection and identification of gastrointestinal Lactobacillus species by using denaturing gradient gel electrophoresis and species-specific PCR primers. Appl Environ Microbiol 66: 297303.