Curr Microbiol DOI 10.1007/s00284-014-0592-x
Comparative Diversity Analysis of Gut Microbiota in Two Different Human Flora-Associated Mouse Strains Xiaojing Zhang • Benhua Zeng • Zhiwei Liu Zhenlin Liao • Wenxai Li • Hong Wei • Xiang Fang
•
Received: 16 January 2014 / Accepted: 1 March 2014 Ó Springer Science+Business Media New York 2014
Abstract The Kunming (KM) mouse is a closed colony mouse strain widely used in Chinese pharmacology, toxicology, and microbiology research laboratories. However, few studies have examined human flora-associated (HFA) microbial communities in KM mice. In this study, HFA models were built from germ-free KM and C57BL/6J mouse strains, and gut microbial diversity was analyzed by denaturing gradient gel electrophoresis (DGGE) and DNA sequencing. We found that the two strains of HFA mice were significantly different based on the UPGMA dendrogram and the Richness index, but dice similarity coefficients of mouse replicates were not significantly different between HFA-KM and HFA-C57BL/6J. Most of the dominant phyla of human gut microflora could be transferred into the guts of the two mouse strains. However, the predominant genus that formed in HFA-KM was Clostridium sp. and that in HFA-C57BL/6J was Blautia sp. These results imply that genotypes difference between the two mice strains is a critical factor in shaping the intestinal microflora. However, genetic differences of individuals within KM mouse populations failed to lead to individual difference in microflora. Successful generation of HFAKM mice will facilitate studies examining how diet affects
Benhua Zeng is the co-first author. X. Zhang Z. Liu Z. Liao X. Fang (&) College of Food Science, South China Agricultural University, Guangzhou 510642, China e-mail: [email protected] B. Zeng W. Li H. Wei (&) Department of Laboratory Animal Science, College of Basic Medicine Science, Third Military Medical University, Chongqing 400038, China e-mail: [email protected]
gut microbial structure, and will enable comparative studies for uncovering genetic factors that shape gut microbial communities. Keywords Human flora-associated mice C57BL/6J KM DGGE Gut microbiota
Introduction The numerous microfloral communities in the human gut play a positive role in pathogen defense, nutrient absorbance, energy metabolism, and immune response [1–3]. In addition, genotype, birth mode, diet, and drugs affect gut microbiota; and even cause changes in health status [4, 5]. The relationship between human gut microbial diversity and physical condition has become a hot research topic. Addressing this issue directly in humans is challenging because of numerous uncontrolled variables, such as host genotype, microbial community composition, diet, and housing conditions. Human flora-associated (HFA) animals provide a pipeline toward a mechanistic of understanding host–microbial interactions, and can be used to determine how human intestinal microbiota affect the host’s health conditions, such as obesity, allergy, and other diseases [6], and how diet or other environmental conditions affect the gut flora. HFA animals are established by intragastric gavage into germ-free animals using a suspension of human feces, which allows the human gut floral community to colonize the animal gut. HFA rats, mice, pigs, dogs, chickens, and cows have all been established, and comparative studies of bacterial composition in the gut of HFA animals and conventionally raised animals have also been undertaken [7, 8]. Mice have numerous advantages as model
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
organisms, including the availability of genetically modified inbred lines, the ease of animal husbandry, and their many similarities to human physiology and genetics [7]. Studies comparing HFA mice reared on a low-fat, plant polysaccharide-rich diet versus those reared on a high-fat, high-sugar Western diet have demonstrated that dietary shifts can have major and rapid effects on the gut microbiome [9]. HFA mice or rats have been widely used because they have been confirmed to be highly humanized in a consistent manner and, most importantly, the gut floral communities are vertically transmissible from HFA parent animals [9]. C57BL/6J is an inbred strain with high gene homogeneity, so different individuals in the group have almost the same genetic background. The C57BL/6J HFA model has been established and widely used for studying the relationship of the gut microbiota to energy accumulation, and for researching how diet or other environmental factors affect the gut flora and the host’s physiological status. KM mice, a closed colony strain with high gene heterogeneity, which have different genetic backgrounds between individuals, is widely used in pharmacology, toxicology, and microbiology research in Chinese laboratories. The two mice strains have differences in genetic backgrounds, environmental sensitivities, and biological characteristics. Additionally, the KM mice used for studies are easier to operate on and more convenient because the body of KM mice is larger than that of C57BL/6J mice and the survival rate of germ-free KM mice is higher than that of C57BL/6J mice. Considering the advantages of HFA-KM and some disadvantages of HFA-C57BL/6J mice, this research focused on analyzing the gut floral diversity of HFA-KM and HFA-C57BL/6J to compare gut microbiota colonization in the two strains to determine whether HFA-KM mice could be another alternative HFA mice model.
Table 1 The diet formula of the two strains of mice
Ingredient Corn flour Soybean flour
In addition, 0.03 % multivitamin was added based on the diet formula
% 35.00 5.00
Soybean expeller
15.00
Wheat bran
15.00
Wheat flour
15.00
Yeast powder
2.00
Bone meal
2.50
Sesame expeller
7.00
Milk powder
2.00
Salt
0.50
Vegetable oil Total
1.00 100
irradiation while the water and bottles were high-pressuresteam sterilized at 121 °C for 60 min. Microbiological testing was conducted with the feces and hair of mice to ensure that the animal was germ free prior to gavage. Human Fecal Floral Inoculation Fresh human feces was collected from a 25-year-old, healthy, female adult, who was not a vegetarian and had not taken any antibiotics for at least 3 months prior to the fecal sample. The donor feces, taken in the morning and stored in a -80 °C freezer, were diluted 100-fold in sterile pre-reduced 0.1 M PBS (pH 7.2) under anaerobic conditions. Each mouse was inoculated by gavage with a 0.5 mL suspension. Each of the HFA-KM and HFA-C57BL/6J feces was collected by provoked defecation at 1, 2, 3, and 4 weeks after inoculation and frozen immediately at 80 °C until DNA extraction. Extraction of Total Genomic DNA
Materials and Methods Animals Germ-free C57BL/6J (n = 10, 5 male, 5 female) and KM (n = 8, 4 male, 4 female) mice, aged 6–8 weeks, were provided by the Department of Laboratory Animal Science, College of Basic Medical Sciences Third Military Medical University, Chongqing, China. All mice were raised and fed with the same diet in sterile Trexler-type plastic film isolators (Fengshi Laboratory Animal Equipment, Suzhou, China), in which temperatures were maintained at 20–26 °C and humidity between 40–70 %, and a rhythmic light cycle of 12 h (light):12 h (dark). The diet formula of the the two strains of mice was described in Table 1. The bedding and food were sterilized by 40 kGy Co-60 gamma
123
Metagenomic DNA was extracted from the fecal samples according to the methods of Turnbaugh et al. [10]. Fecal samples (50 mg) were resuspended in 700 lL lysis buffer (500 mM NaCl, 50 mM Tris–HCl, 50 mM EDTA, and 4 % SDS, pH 8.0) by adding 0.2 g 0.5-mm zirconia/silica beads (BioSpec Products) and 250 lL of phenol/chloroform/isoamyl alcohol (25/24/1). Microbial cells were lysed by mechanical disruption with a mini beadbeater (BioSpec Products) for 2 min. Samples were centrifuged at 20,000 9 g for 5 min, and 250 lL 10 M ammonium acetate was added after the aqueous phase was collected, and was then incubated on ice for 5 min, centrifuged at 20,000 9 g for 10 min, and the supernatant was collected. Extraction was performed twice with 250 lL phenol/ chloroform/isoamyl alcohol (25/24/1) and centrifugation was performed for 2 min at 20,000 9 g. DNA in the
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
aqueous phase was precipitated by the addition of an equal volume of pre-cooling isopropanol and stored at -20 °C for 30 min. Next, following centrifugation at 4 °C at 20,000 9 g for 10 min, the pelleted DNA was washed twice with 1 mL of 70 % (v/v) pre-chilled alcohol, and the supernatant was removed. DNA pellets were dried and then resuspended in 50 lL of Tris–EDTA containing 2 lL RNAase (10 mg/lL) (Qiagen). PCR Amplification of 16S rDNA The bacterial 16S rRNA gene was amplified with primers targeting the DNA region between positions 968 and 1401 of Escherichia coli, corresponding to the V6–V8 region. The primers were GC-F968: 50 - CGCCCGGGGCGCGCC CCGGGCGGGGCGGGGGCACGGGGGGAACGCGAA GA ACCTTAC -30 , and R1401: 50 -CGGTGTGTACAAGACCC-30 (the GC-rich clamp is underlined in GCF968) [11]. The amplification of the V6–V8 region was performed in a 25 lL mixture containing 100 ng templates of DNA, 1 9 PCR Buffer, 2.5 mM MgCl2, 200 lM deoxynucleoside triphosphates, 0.2 lM of each primer, and 0.625 U of Ex-Taq DNA polymerase (Bioteke, Beijing, China). The cycling parameters for touchdown PCR, which reduced the formation of non-specific amplification during the reaction [12, 13], were followed by an initial denaturation step at 94 °C for 5 min. Then, the following steps were performed: 30 cycles of denaturation at 94 °C for 30 s, annealing for 30 s with a 0.5 °C/cycle decrement from 65.5 to 56 °C for the first 20 cycles, then annealing was done at 56 °C for the remaining 10 cycles, and extension was performed at 72 °C for 1 min. Cycling was ended by an 8-min incubation at 72 °C. To eliminate heteroduplexes from mixed-template PCR products, Reconditioning PCR [14] was performed. This process involves diluting the amplification products 10-fold and then using 1 lL of that dilution to serve as the new template to add into a fresh mixture to be used for an additional five cycles of amplification using the same mixture and the same parameters used in the last 10 cycles specified above. DGGE Analysis DGGE was performed in a D-Code universal mutation detection system (Bio-Rad, Hercules, California, USA) using 8 % (wt/v) polyacrylamide (37.5:1 acrylamide:bisacrylamide) in 1 9 TAE (40 mM Tris base, 20 mM acetic acid, and 1 mM EDTA. The denaturing gradient ranged from 32 to 56 % for the V6–V8 region, with 100 % denaturants corresponding to 7 M urea and 40 % [v/v] deionized formamide). The gel was run at 220 V for 10 min, and then at a constant voltage of 85 V and a temperature of 60 °C for 16 h. After that, the gel was
stained with silver nitrate and photographed with a digital camera (Canon, Japan). Target bands were excised from the polyacrylamide gel, rinsed, and mashed in 50 lL of sterile deionized water at 4 °C overnight. A total of 5 lL of the DNA extract was used as a template for additional amplification [15], and the reaction mixture and parameters were the same as for touchdown PCR described in paragraph 1.4. DGGE was conducted again to confirm the position of the band. Then, a PCR reaction using the non-GC-clamp primers was performed again in the same condition, and the products were excised from a 1.0 % agarose gel, purified with a DNA Gel Extraction Kit (Omega Bio-Tek, Norcross, GA, USA), and then subjected to sequencing (Sangon, Shanghai, China) [16]. The 16S rRNA sequences were compared with the GenBank database using the BLAST algorithm of the National Center for Biotechnology Information. Statistical Analysis The DGGE pictures were digitalized with quantity one (Bio-Rad, Hercules, California, USA) software, which removed the lane background and determined the migration position and peak density of each band, and then acquired a two-axis matrix of peak density for the band position. Similarities were displayed graphically as a dendrogram using a dice coefficient based on the UPGMA method. Microbial diversity index of each lane, containing band number, Evenness Index (Pielous, E), Shannon–Wiener Index (H’), Richness Index (Margalef, R), and Dominance Index (Simpson, D), was calculated based on the digital statistical information from the DGGE profile. Then, similarities between the human donors and the two mouse strains were analyzed using the one-way ANOVA. The similarity coefficients for the sample replicates of each mouse strain were further analyzed and compared between the two strains of HFA mice. Principal components analysis (PCA) was conducted by SPSS version 13.0 using the two-axis matrix of peak density and band position, the three principal components that contributed most to the difference between samples were extracted, and a three-dimensional scatter plot was prepared using Origin 8.0 software (Originlab, Northampton, MA, USA).
Results DGGE Profile Changes in microbial diversity of the guts of HFA mice were analyzed over time by PCR-DGGE analysis using
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
Fig. 1 a 16S rRNA V6–V8 region PCR-DGGE profiles of fecal samples from human donor (13–14) and those from HFA-KM mice (1–12) and HFA-C57BL/6J mice (15–26) after incubated for 1, 2, 3,
and 4 weeks, at each time point, three samples of the same mice were analyzed due to the limited gel holes in one piece of gel; b UPGMA dendrogram of V6–V8 region of DGGE profiles 92 9 111 mm
primers targeting the V6–V8 region (position 968–1401) of the 16S rRNA gene (Fig. 1a). At each time point, three samples coming from the same mice were analyzed due to
the limited gel holes in one piece of gel. Population fingerprint profiles were numerically analyzed by Quantity One software. The PCR-DGGE profiles of fecal samples
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota Fig. 2 a 16S rRNA V6–V8 region PCR-DGGE profiles of fecal samples from human donor (9–10) and those from eight HFA-KM mice (1–8) and ten HFA-C57BL/6J mice (11–20) at the fourth week after inoculation, the identified bands are indicated by arrows and numbered by 1–17; b UPGMA dendrogram of DGGE profiles 91 9 119 mm
showed that the dominant gut microbiota changed over time, with the gut microbiota profiles of the last 2 weeks of the analysis more stable than those of the first 2 weeks. Comparing the gel samples between weeks 1 and 2, band 9 in HFA-KM and bands 4, 7, and 16 in HFA-C57BL/6J disappeared; however, bands 7, 8, and 10 in HFA-KM and bands 3, 5, 9, 10, 13, 15, and 18 in HFA-C57BL/6J were enhanced at weeks 3 and 4. Cluster analysis using DGGE population profiling and the UPGMA dendrogram confirmed that the predominant gut microbiota from the human donor and those from two strains of mice were different (Fig. 1b).
There were distinct differences in gut microflora composition between the human and mouse samples, which was confirmed by the DGGE profiles (Fig. 2a) and the UPGMA dendrogram (Fig. 2b). The total dice similarity coefficient of the human donor and the mouse model was 38 %, indicating that the HFA models became humanized to a certain degree. Bands 4, 7, and 12 in the lanes of human fecal samples were present in the lanes of the two strains of mice. However, bands 1, 2, 5, and 8 in human feces were not detected in the feces of the mice. In addition, the gray levels of some of the bands from the HFA mice, such as bands 7, 11, 12, 13, and 14, were obviously
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
Fig. 4 The dice coefficient (similarity index) for the DGGE profiles of HFA-KM eight replicates and HFAC57BL/6J ten replicates of fecal microbiota at the fourth week (mean ± SEM plotted), P [ 0.05. Error bars indicate the standard error of the mean (SEM)
Fig. 3 a A comparison of microbial diversity between the two strain HFA mice and human at the fourth week after inoculation (mean ± SEM plotted), *, P \ 0.05; **, P \ 0.01; b The dice coefficient (similarity index) for the DGGE fingerprint of gut microbiota of HFA-KM (KM) or HFA-C57BL/6J (C57) compared to human donor (Human) at the fourth week after inoculation, **, P \ 0.01. Error bars indicate the standard error of the mean (SEM)
which indicated that only a subset of gut microbes from the human donor colonized the HFA mice. However, the analysis suggests there is no difference between the microbial diversity indices, except in the case of richness, between HFA-KM and HFA-C57BL/6J. The similarity index of the gut microbiota in the HFA-C57BL/6J mice compared with that of the human donor (47.41 % ± 1.05) was significantly higher than that of HFA-KM mice (28.45 % ± 0.74) (P \ 0.01) (Fig. 3b). However, there was no significant difference in the dice similarity coefficient between HFA-KM (69.92 %) and HFA-C57BL/6J (68.66 %) (P [ 0.05) (Fig. 4). Principal Component Analysis
different from those bands with the same migration position as the donor, indicating that these microbes had immigrated into the mouse gut from the donor but the ratio in the host gut had changed. Furthermore, some bands were present only in one of the HFA mouse strains, such as bands 3, 6, 7, 10, 14, and 17, which may imply that the diversity of the gut microbiota was affected by the genetic background of the host. Diversity Analysis Microbial diversity indices of each lane in Fig. 2a were calculated based on the digital statistical information from the DGGE profile. Each microbial diversity index for the two strains of HFA mice and the human donor was compared using one-way ANOVA, and the result is shown in Fig. 3a. Highly significant differences were evident when comparing the gut microbial diversity in the two HFA mice using the Shannon–Wiener (H’) and richness indices,
123
Three principal components were extracted by principal component analysis, and scatter plots were generated in Origin 8.0 (Fig. 5). Each point indicates the spatial position of each sample and the entries are distributed according to their relatedness [17]. The HFA-KM, HFA-C57BL/6J, and human samples localized to the different areas in the plot, respectively (Fig. 5). The PCA separated the different groups completely, indicating that there were distinct differences in gut microflora composition between the human donor and the two mouse strains. Identification of Bands in DGGE Profiles To investigate the similarities and differences in gut microbial community composition between the human donor and the two HFA strains, 17 bands in the gel (Fig. 2a) were extracted corresponding to the 16S rRNA
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
Fig. 5 Principal component analysis of mice fecal bacteria community based on 16S rRNA gene PCR-DGGE profile at the fourth week. Values on the axis indicate variation percentages contributed to the first three principal components, respectively
Table 2 Closest relatives associated with bands in DGGE profile Band no.
Length (bp)
Closest match
Gene bank accession no.
Similarity (%)
1
436
Lactonifactor longoviformis
KC140198
93
2
437
Acinetobacter calcoaceticus
KC140199
99
3
365
Ruminococcus gauvreauii
KC161373
95
4
369
A. gyllenbergii
KC140197
91
5
350
R. torques
KC140200
91
6
364
Clostridium clostridioforme
KC140196
96
7
353
C. clostridioforme
KC140204
99
8
348
Blautia luti
KC172380
93
9
373
B. wexlerae
KC172381
97
10
359
B. wexlerae
KC172382
98
11
348
B. wexlerae
KC140195
100
12
369
C. hathewayi
KC140194
98
13 14
344 347
B. wexlerae C. lavalense
KC140203 KC140202
94 99
15
437
C. lavalense
KC140193
99
16
437
B. schinkii
KC140192
96
17
436
Stenotrophomonas maltophilia
KC140191
98
base sequences with those in NCBI were between 91 and 100 % (Table 2). It is interesting that there were six bands (Bands 8, 9, 10, 11, 13, and 16) belonging to the genus Blautia, and another five sequences (Bands 6, 7, 12, 14, and 15) belonging to the genus Clostridium, indicating that these two genera were readily implanted into the guts of mice. Moreover, four of the six sequences (Bands 9, 10, 11, and 13) of Blautia spp. were from HFA-C57BL/6J samples with weak or no signal in HFA-KM and the five sequences of Clostridium spp. were from the gel of the HFA-KM samples with weak or no signal in HFA-C57BL/6J, suggesting that the Blautia spp. was better adapted to the C57BL/6J mice, whereas the Clostridium spp. was more adapted to the intestinal environment in KM mice. Bands 1, 2, 5, and 8 from the human samples, whose nearest relatives were Lactonifactor longoviformis, Acinetobacter calcoaceticus, Ruminococcus torques and B.luti, respectively, were not found in the mouse samples. Bands 4 and 13, which were most closely related to A. gyllenbergii and B.luti, respectively, were only found in the humanized mice, implying that the colonic microfloral composition was affected by genetic factors.
Discussion
Bands numbers are indicated by arrows in Fig. 2a
base sequences (NCBI accession number: KC140191 to KC140200, KC140202 to KC140204, KC161373, and KC172380 to KC172382). These sequences were used to perform a BLAST search in NCBI. The identities of these
This study focused on the intestinal microbial diversity in two strains of HFA mice to determine if HFA-KM mice, a closed colony mouse strain, could be an alternative HFA model used to study environmental interventions for altering the gut microbiota to promote health. After inoculation with a suspension of human feces, a substantial difference in the band types present in individuals from the same HFA mouse strain was observed in the first 2 weeks. However, 2 weeks later, each of the two mouse strains gradually formed a unique band pattern, demonstrating that there is a difference in the banding types of the two strains of HFA mice. This was corroborated by the observation that the samples of the three groups clearly separated into a hierarchical clustering dendrogram at the 4th week (Fig. 2b). PCA scatter of the DGGE profile also made it clear that the spatial distribution of the mouse samples was consistent with the results of the cluster analysis of the DGGE profile presented in Fig. 2b. Compared with gut microbiota of the human donor, the similarities ratio of the inbred HFA-C57BL/6J mice (47.41 %) was much higher than the closed colony HFAKM mice (28.45 %). However, KM mice have advantages over the C57BL/6J strain such as their larger size, which allows researchers to obtain more blood samples, and they survive longer and reproduce at a higher rate than C57BL/ 6J mice. What is more, our study proved that there was no
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
significant difference (P [ 0.05) in the dice similarity coefficient between HFA-KM mouse replicates (69.92 %) and HFA-C57BL/6J mouse replicates (68.66 %) (Fig. 4). In theory, the variation of the microbial communities of individuals within HFA-KM mice may be more than that of HFA-C57BL/6J mice because of the lower gene homogeneity of KM mice compared to C57BL/6J mice. However, the above result indicated that the diversity of colonic flora was not affected in a significant way by the gene heterogeneity of the KM strain. There are three dominant phyla (Firmicutes, Bacteroidetes, and Actinobacteria) in the adult human gastrointestinal tract. Firmicutes and Bacteroidetes are predominate and Actinobacteria is a lesser but significant contributor to the overall community. Each of Firmicutes and Bacteroidetes comprise *30 % of bacteria in feces and the mucus overlying the intestinal epithelium [18, 19]. Proteobacteria are common but usually not dominant [19, 20]. In our analysis, we identified 17 sequences. Bands 2, 4, and 17 in Fig. 2b were related to the phylum Proteobacteria and the remaining 14 belong to the phylum Firmicutes. We found that the two phyla of the human gut microflora were stably implanted into the colon of the two strains of HFA mice, and Proteobacteria emerged as a dominant member of the flora in the HFA mouse colon. However, the predominant genera in the colon of HFA-KM mice were different from those of HFA-C57BL/6J mice; the predominant genus in HFA-KM was Clostridium spp., in contrast to HFAC57BL/6J whose predominant genus was Blautia spp; it is also notable that Blautia spp. was recently discovered to be a novel genus and B. wexlerae is proposed to be a new species isolated from human feces [21]. This further confirmed that the composition of the dominant microbial community is affected by host genetic backgrounds, which have been verified by PCR-DGGE/TTGE, ERIC-PCR [15], T-RFLP [22], metagenomic analysis, and other cultureindependent methodologies [23].
Conclusion In conclusion, three important points can be inferred. Firstly, a reliable HFA model should be exposed to donor microbiota for at least 3–4 weeks before used in experiments. Secondly, hereditary differences between strains affect not only the overall gut microbial structure, but also lead to the distinct predominated genus. Furthermore, HFA-KM mice may be an attractive alternative HFA model for researches related to microbial communities for the little variation of the microbial communities of individuals within the group, especially when certain microbial genera cannot be implanted into the colons of HFAC57BL/6J mice.
123
Acknowledgments This work was supported by Grant 31071528 and 81370906 from the National Natural Science Foundation of China, Grant 2013CB531406 from the National Basic Research Program of China, and Grant 2012B091100429 and 2011B090400260 from Production-Learning-Research Program of Guangdong Science and Technology Department.
References 1. Round JL, Mazmanian SK (2009) The gut microbiota shapes intestinal immune responses during health and disease. Nat Rev Immunol 9(5):313–323. doi:10.1038/nri2515 2. Sanz Y, Santacruz A, Gauffin P (2010) Probiotics in the defence and metabolic balance of the organism: gut microbiota in obesity and metabolic disorders. P Nutr Soc 69:434–441. doi:10.1017/ S0029665110001813 3. Xu J, Gordon JI (2003) Honor thy symbionts. P Natl Acad Sci Usa 100(18):10452–10459. doi:10.1073/pnas.1734063100 4. Martens EC, Koropatkin NM, Smith TJ, Gordon JI (2009) Complex glycan catabolism by the human gut microbiota: the Bacteroidetes sus-like paradigm. J Biol Chem 284(37):24673– 24677. doi:10.1074/jbc.R109.022848 5. Perrin-Guyomard A, Poul JM, Laurentie M, Sanders P, Ferna´ndez AH, Bartholomew M (2006) Impact of ciprofloxacin in the human-flora-associated (HFA) rat model: comparison with the HFA mouse model. Regul Toxicol Pharm 45(1):66–78. doi:10. 1016/j.yrtph.2006.02.002 6. Perrin-Guyomard A, Poul JM, Corpet DE, Sanders P, Ferna´ndez AH, Bartholomew M (2005) Impact of residual and therapeutic doses of ciprofloxacin in the human-flora-associated mice model. Regul Toxicol Pharm 42(2):151–160. doi:10.1016/j.yrtph.2005. 03.001 7. Gootenberg DB, Turnbaugh PJ (2011) Companion animals symposium: humanized animal models of the microbiome. J Anim Sci 89(5):1531–1537. doi:10.2527/jas.2010-3371 8. Pang XY, Hua XG, Yang Q, Ding DH, Che CY, Cui L, Jia W, Bucheli P, Zhao LP (2007) Inter-species transplantation of gut microbiota from human to pigs. ISME J 1(2):156–162. doi:10. 1038/ismej.2007.23 9. Turnbaugh PJ, Ridaura VK, Faith JJ, Rey FE, Knight R, Gordon JI (2009) The effect of diet on the human gut microbiome: a metagenomic analysis in humanized gnotobiotic mice. Sci Transl Med 1(6): 6ra14. doi:10.1126/scitranslmed.3000322 10. Turnbaugh PJ, Hamady M, Yatsunenko T et al (2009) A core gut microbiome in obese and lean twins. Nature 457(7228):480–487. doi:10.1038/nature07540 11. Huws SA, Edwards JE, Kim EJ, Scollan ND (2007) Specificity and sensitivity of eubacterial primers utilized for molecular profiling of bacteria within complex microbial ecosystems. J Microbiol Meth 70(3):565–569. doi:10.1016/j.mimet.2007.06.013 12. Gafan GP, Lucas VS, Roberts GJ, Petrie A, Wilson M, Spratt DA (2005) Statistical analyses of complex denaturing gradient gel electrophoresis profiles. J Clin Microbiol 43(8):3971–3978. doi:10.1128/JCM.43.8.3971-3978.2005 13. 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 Microb 59(3):695–700 14. Thompson JR, Marcelino LA, Polz MF (2002) Heteroduplexes in mixed-template amplifications: formation, consequence and elimination by ‘reconditioning PCR’. Nucleic Acids Res 30(9):2083–2088. doi:10.1093/nar/30.9.2083 15. Leung K, Topp E (2001) Bacterial community dynamics in liquid swine manure during storage: molecular analysis using DGGE/
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
16.
17.
18.
19.
20.
PCR of 16S rDNA. FEMS Microbiol Ecol 38:169–177. doi:10. 1111/j.1574-6941.2001.tb00895.x Yuan J, Zeng B, Niu R, Tang H, Li W, Zhang Z, Wei H (2011) The development and stability of the genus Bacteriodes from human gut microbiota in HFA mice model. Curr Microbiol 62(4):1107–1112. doi:10.1007/s00284-010-9833-9 Boon N, De Windt W, Verstraete W et al (2002) Evaluation of nested PCR-DGGE (denaturing gradient gel electrophoresis) with group-specific 16S rRNA primers for the analysis of bacterial communities from different wastewater treatment plants. FEMS Microbiol Ecol 39(2):101–112. doi:10.1111/j.1574-6941.2002. tb00911.x Backhed F, Ley RE, Sonnenburg JL, Peterson DA, Gordon JI (2005) Host-bacterial mutualism in the human intestine. Science 307(5717):1915–1920. doi:10.1126/science.1104816 Ley RE, Backhed F, Turnbaugh P, Lozupone CA, Knight RD, Gordon JI (2005) Obesity alters gut microbial ecology. P Natl Acad Sci Usa 102(31):11070–11075. doi:10.1073/pnas. 0504978102 Seksik P, Rigottier-Gois L, Gramet G, Sutren M, Pochart P, Marteau P, Jian R, Dore´ J (2003) Alterations of the dominant
faecal bacterial groups in patients with Crohn’s disease of the colon. Gut 52(2):237–242. doi:10.1136/gut.52.2.237 21. Liu C, Finegold SM, Song Y, Lawson PA (2008) Reclassification of Clostridium coccoides, Ruminococcus hansenii, Ruminococcus hydrogenotrophicus, Ruminococcus luti, Ruminococcus productus and Ruminococcus schinkii as Blautia coccoides gen. nov., comb. nov., Blautia hansenii comb. nov., Blautia hydrogenotrophica comb. nov., Blautia luti comb. nov., Blautia producta comb. nov., Blautia schinkii comb. nov. and description of Blautia wexlerae sp. nov., isolated from human faeces. Int J Syst Evol Micr 58(Pt8): 1896–1902. doi:10.1099/ijs.0.65208-0 22. Kibe R, Sakamoto M, Yokota H, Ishikawa H, Aiba Y, Koga Y, Benno Y (2005) Movement and fixation of intestinal microbiota after administration of human feces to germfree mice. Appl Environ Microbiol 71(6):3171–3178. doi:10.1128/AEM.71.6. 3171-3178.2005 23. Gill SR, Pop M, DeBoy RT, Eckburg PB, Turnbaugh PJ, Samuel BS, Gordon JI, Relman DA, Fraser-Liggett CM, Nelson KE (2006) Metagenomic analysis of the human distal gut microbiome. Science 312(5778):1355–1359. doi:10.1126/science. 1124234
123
Comparative Diversity Analysis of Gut Microbiota in Two Different Human Flora-Associated Mouse Strains Xiaojing Zhang • Benhua Zeng • Zhiwei Liu Zhenlin Liao • Wenxai Li • Hong Wei • Xiang Fang
•
Received: 16 January 2014 / Accepted: 1 March 2014 Ó Springer Science+Business Media New York 2014
Abstract The Kunming (KM) mouse is a closed colony mouse strain widely used in Chinese pharmacology, toxicology, and microbiology research laboratories. However, few studies have examined human flora-associated (HFA) microbial communities in KM mice. In this study, HFA models were built from germ-free KM and C57BL/6J mouse strains, and gut microbial diversity was analyzed by denaturing gradient gel electrophoresis (DGGE) and DNA sequencing. We found that the two strains of HFA mice were significantly different based on the UPGMA dendrogram and the Richness index, but dice similarity coefficients of mouse replicates were not significantly different between HFA-KM and HFA-C57BL/6J. Most of the dominant phyla of human gut microflora could be transferred into the guts of the two mouse strains. However, the predominant genus that formed in HFA-KM was Clostridium sp. and that in HFA-C57BL/6J was Blautia sp. These results imply that genotypes difference between the two mice strains is a critical factor in shaping the intestinal microflora. However, genetic differences of individuals within KM mouse populations failed to lead to individual difference in microflora. Successful generation of HFAKM mice will facilitate studies examining how diet affects
Benhua Zeng is the co-first author. X. Zhang Z. Liu Z. Liao X. Fang (&) College of Food Science, South China Agricultural University, Guangzhou 510642, China e-mail: [email protected] B. Zeng W. Li H. Wei (&) Department of Laboratory Animal Science, College of Basic Medicine Science, Third Military Medical University, Chongqing 400038, China e-mail: [email protected]
gut microbial structure, and will enable comparative studies for uncovering genetic factors that shape gut microbial communities. Keywords Human flora-associated mice C57BL/6J KM DGGE Gut microbiota
Introduction The numerous microfloral communities in the human gut play a positive role in pathogen defense, nutrient absorbance, energy metabolism, and immune response [1–3]. In addition, genotype, birth mode, diet, and drugs affect gut microbiota; and even cause changes in health status [4, 5]. The relationship between human gut microbial diversity and physical condition has become a hot research topic. Addressing this issue directly in humans is challenging because of numerous uncontrolled variables, such as host genotype, microbial community composition, diet, and housing conditions. Human flora-associated (HFA) animals provide a pipeline toward a mechanistic of understanding host–microbial interactions, and can be used to determine how human intestinal microbiota affect the host’s health conditions, such as obesity, allergy, and other diseases [6], and how diet or other environmental conditions affect the gut flora. HFA animals are established by intragastric gavage into germ-free animals using a suspension of human feces, which allows the human gut floral community to colonize the animal gut. HFA rats, mice, pigs, dogs, chickens, and cows have all been established, and comparative studies of bacterial composition in the gut of HFA animals and conventionally raised animals have also been undertaken [7, 8]. Mice have numerous advantages as model
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
organisms, including the availability of genetically modified inbred lines, the ease of animal husbandry, and their many similarities to human physiology and genetics [7]. Studies comparing HFA mice reared on a low-fat, plant polysaccharide-rich diet versus those reared on a high-fat, high-sugar Western diet have demonstrated that dietary shifts can have major and rapid effects on the gut microbiome [9]. HFA mice or rats have been widely used because they have been confirmed to be highly humanized in a consistent manner and, most importantly, the gut floral communities are vertically transmissible from HFA parent animals [9]. C57BL/6J is an inbred strain with high gene homogeneity, so different individuals in the group have almost the same genetic background. The C57BL/6J HFA model has been established and widely used for studying the relationship of the gut microbiota to energy accumulation, and for researching how diet or other environmental factors affect the gut flora and the host’s physiological status. KM mice, a closed colony strain with high gene heterogeneity, which have different genetic backgrounds between individuals, is widely used in pharmacology, toxicology, and microbiology research in Chinese laboratories. The two mice strains have differences in genetic backgrounds, environmental sensitivities, and biological characteristics. Additionally, the KM mice used for studies are easier to operate on and more convenient because the body of KM mice is larger than that of C57BL/6J mice and the survival rate of germ-free KM mice is higher than that of C57BL/6J mice. Considering the advantages of HFA-KM and some disadvantages of HFA-C57BL/6J mice, this research focused on analyzing the gut floral diversity of HFA-KM and HFA-C57BL/6J to compare gut microbiota colonization in the two strains to determine whether HFA-KM mice could be another alternative HFA mice model.
Table 1 The diet formula of the two strains of mice
Ingredient Corn flour Soybean flour
In addition, 0.03 % multivitamin was added based on the diet formula
% 35.00 5.00
Soybean expeller
15.00
Wheat bran
15.00
Wheat flour
15.00
Yeast powder
2.00
Bone meal
2.50
Sesame expeller
7.00
Milk powder
2.00
Salt
0.50
Vegetable oil Total
1.00 100
irradiation while the water and bottles were high-pressuresteam sterilized at 121 °C for 60 min. Microbiological testing was conducted with the feces and hair of mice to ensure that the animal was germ free prior to gavage. Human Fecal Floral Inoculation Fresh human feces was collected from a 25-year-old, healthy, female adult, who was not a vegetarian and had not taken any antibiotics for at least 3 months prior to the fecal sample. The donor feces, taken in the morning and stored in a -80 °C freezer, were diluted 100-fold in sterile pre-reduced 0.1 M PBS (pH 7.2) under anaerobic conditions. Each mouse was inoculated by gavage with a 0.5 mL suspension. Each of the HFA-KM and HFA-C57BL/6J feces was collected by provoked defecation at 1, 2, 3, and 4 weeks after inoculation and frozen immediately at 80 °C until DNA extraction. Extraction of Total Genomic DNA
Materials and Methods Animals Germ-free C57BL/6J (n = 10, 5 male, 5 female) and KM (n = 8, 4 male, 4 female) mice, aged 6–8 weeks, were provided by the Department of Laboratory Animal Science, College of Basic Medical Sciences Third Military Medical University, Chongqing, China. All mice were raised and fed with the same diet in sterile Trexler-type plastic film isolators (Fengshi Laboratory Animal Equipment, Suzhou, China), in which temperatures were maintained at 20–26 °C and humidity between 40–70 %, and a rhythmic light cycle of 12 h (light):12 h (dark). The diet formula of the the two strains of mice was described in Table 1. The bedding and food were sterilized by 40 kGy Co-60 gamma
123
Metagenomic DNA was extracted from the fecal samples according to the methods of Turnbaugh et al. [10]. Fecal samples (50 mg) were resuspended in 700 lL lysis buffer (500 mM NaCl, 50 mM Tris–HCl, 50 mM EDTA, and 4 % SDS, pH 8.0) by adding 0.2 g 0.5-mm zirconia/silica beads (BioSpec Products) and 250 lL of phenol/chloroform/isoamyl alcohol (25/24/1). Microbial cells were lysed by mechanical disruption with a mini beadbeater (BioSpec Products) for 2 min. Samples were centrifuged at 20,000 9 g for 5 min, and 250 lL 10 M ammonium acetate was added after the aqueous phase was collected, and was then incubated on ice for 5 min, centrifuged at 20,000 9 g for 10 min, and the supernatant was collected. Extraction was performed twice with 250 lL phenol/ chloroform/isoamyl alcohol (25/24/1) and centrifugation was performed for 2 min at 20,000 9 g. DNA in the
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
aqueous phase was precipitated by the addition of an equal volume of pre-cooling isopropanol and stored at -20 °C for 30 min. Next, following centrifugation at 4 °C at 20,000 9 g for 10 min, the pelleted DNA was washed twice with 1 mL of 70 % (v/v) pre-chilled alcohol, and the supernatant was removed. DNA pellets were dried and then resuspended in 50 lL of Tris–EDTA containing 2 lL RNAase (10 mg/lL) (Qiagen). PCR Amplification of 16S rDNA The bacterial 16S rRNA gene was amplified with primers targeting the DNA region between positions 968 and 1401 of Escherichia coli, corresponding to the V6–V8 region. The primers were GC-F968: 50 - CGCCCGGGGCGCGCC CCGGGCGGGGCGGGGGCACGGGGGGAACGCGAA GA ACCTTAC -30 , and R1401: 50 -CGGTGTGTACAAGACCC-30 (the GC-rich clamp is underlined in GCF968) [11]. The amplification of the V6–V8 region was performed in a 25 lL mixture containing 100 ng templates of DNA, 1 9 PCR Buffer, 2.5 mM MgCl2, 200 lM deoxynucleoside triphosphates, 0.2 lM of each primer, and 0.625 U of Ex-Taq DNA polymerase (Bioteke, Beijing, China). The cycling parameters for touchdown PCR, which reduced the formation of non-specific amplification during the reaction [12, 13], were followed by an initial denaturation step at 94 °C for 5 min. Then, the following steps were performed: 30 cycles of denaturation at 94 °C for 30 s, annealing for 30 s with a 0.5 °C/cycle decrement from 65.5 to 56 °C for the first 20 cycles, then annealing was done at 56 °C for the remaining 10 cycles, and extension was performed at 72 °C for 1 min. Cycling was ended by an 8-min incubation at 72 °C. To eliminate heteroduplexes from mixed-template PCR products, Reconditioning PCR [14] was performed. This process involves diluting the amplification products 10-fold and then using 1 lL of that dilution to serve as the new template to add into a fresh mixture to be used for an additional five cycles of amplification using the same mixture and the same parameters used in the last 10 cycles specified above. DGGE Analysis DGGE was performed in a D-Code universal mutation detection system (Bio-Rad, Hercules, California, USA) using 8 % (wt/v) polyacrylamide (37.5:1 acrylamide:bisacrylamide) in 1 9 TAE (40 mM Tris base, 20 mM acetic acid, and 1 mM EDTA. The denaturing gradient ranged from 32 to 56 % for the V6–V8 region, with 100 % denaturants corresponding to 7 M urea and 40 % [v/v] deionized formamide). The gel was run at 220 V for 10 min, and then at a constant voltage of 85 V and a temperature of 60 °C for 16 h. After that, the gel was
stained with silver nitrate and photographed with a digital camera (Canon, Japan). Target bands were excised from the polyacrylamide gel, rinsed, and mashed in 50 lL of sterile deionized water at 4 °C overnight. A total of 5 lL of the DNA extract was used as a template for additional amplification [15], and the reaction mixture and parameters were the same as for touchdown PCR described in paragraph 1.4. DGGE was conducted again to confirm the position of the band. Then, a PCR reaction using the non-GC-clamp primers was performed again in the same condition, and the products were excised from a 1.0 % agarose gel, purified with a DNA Gel Extraction Kit (Omega Bio-Tek, Norcross, GA, USA), and then subjected to sequencing (Sangon, Shanghai, China) [16]. The 16S rRNA sequences were compared with the GenBank database using the BLAST algorithm of the National Center for Biotechnology Information. Statistical Analysis The DGGE pictures were digitalized with quantity one (Bio-Rad, Hercules, California, USA) software, which removed the lane background and determined the migration position and peak density of each band, and then acquired a two-axis matrix of peak density for the band position. Similarities were displayed graphically as a dendrogram using a dice coefficient based on the UPGMA method. Microbial diversity index of each lane, containing band number, Evenness Index (Pielous, E), Shannon–Wiener Index (H’), Richness Index (Margalef, R), and Dominance Index (Simpson, D), was calculated based on the digital statistical information from the DGGE profile. Then, similarities between the human donors and the two mouse strains were analyzed using the one-way ANOVA. The similarity coefficients for the sample replicates of each mouse strain were further analyzed and compared between the two strains of HFA mice. Principal components analysis (PCA) was conducted by SPSS version 13.0 using the two-axis matrix of peak density and band position, the three principal components that contributed most to the difference between samples were extracted, and a three-dimensional scatter plot was prepared using Origin 8.0 software (Originlab, Northampton, MA, USA).
Results DGGE Profile Changes in microbial diversity of the guts of HFA mice were analyzed over time by PCR-DGGE analysis using
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
Fig. 1 a 16S rRNA V6–V8 region PCR-DGGE profiles of fecal samples from human donor (13–14) and those from HFA-KM mice (1–12) and HFA-C57BL/6J mice (15–26) after incubated for 1, 2, 3,
and 4 weeks, at each time point, three samples of the same mice were analyzed due to the limited gel holes in one piece of gel; b UPGMA dendrogram of V6–V8 region of DGGE profiles 92 9 111 mm
primers targeting the V6–V8 region (position 968–1401) of the 16S rRNA gene (Fig. 1a). At each time point, three samples coming from the same mice were analyzed due to
the limited gel holes in one piece of gel. Population fingerprint profiles were numerically analyzed by Quantity One software. The PCR-DGGE profiles of fecal samples
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota Fig. 2 a 16S rRNA V6–V8 region PCR-DGGE profiles of fecal samples from human donor (9–10) and those from eight HFA-KM mice (1–8) and ten HFA-C57BL/6J mice (11–20) at the fourth week after inoculation, the identified bands are indicated by arrows and numbered by 1–17; b UPGMA dendrogram of DGGE profiles 91 9 119 mm
showed that the dominant gut microbiota changed over time, with the gut microbiota profiles of the last 2 weeks of the analysis more stable than those of the first 2 weeks. Comparing the gel samples between weeks 1 and 2, band 9 in HFA-KM and bands 4, 7, and 16 in HFA-C57BL/6J disappeared; however, bands 7, 8, and 10 in HFA-KM and bands 3, 5, 9, 10, 13, 15, and 18 in HFA-C57BL/6J were enhanced at weeks 3 and 4. Cluster analysis using DGGE population profiling and the UPGMA dendrogram confirmed that the predominant gut microbiota from the human donor and those from two strains of mice were different (Fig. 1b).
There were distinct differences in gut microflora composition between the human and mouse samples, which was confirmed by the DGGE profiles (Fig. 2a) and the UPGMA dendrogram (Fig. 2b). The total dice similarity coefficient of the human donor and the mouse model was 38 %, indicating that the HFA models became humanized to a certain degree. Bands 4, 7, and 12 in the lanes of human fecal samples were present in the lanes of the two strains of mice. However, bands 1, 2, 5, and 8 in human feces were not detected in the feces of the mice. In addition, the gray levels of some of the bands from the HFA mice, such as bands 7, 11, 12, 13, and 14, were obviously
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
Fig. 4 The dice coefficient (similarity index) for the DGGE profiles of HFA-KM eight replicates and HFAC57BL/6J ten replicates of fecal microbiota at the fourth week (mean ± SEM plotted), P [ 0.05. Error bars indicate the standard error of the mean (SEM)
Fig. 3 a A comparison of microbial diversity between the two strain HFA mice and human at the fourth week after inoculation (mean ± SEM plotted), *, P \ 0.05; **, P \ 0.01; b The dice coefficient (similarity index) for the DGGE fingerprint of gut microbiota of HFA-KM (KM) or HFA-C57BL/6J (C57) compared to human donor (Human) at the fourth week after inoculation, **, P \ 0.01. Error bars indicate the standard error of the mean (SEM)
which indicated that only a subset of gut microbes from the human donor colonized the HFA mice. However, the analysis suggests there is no difference between the microbial diversity indices, except in the case of richness, between HFA-KM and HFA-C57BL/6J. The similarity index of the gut microbiota in the HFA-C57BL/6J mice compared with that of the human donor (47.41 % ± 1.05) was significantly higher than that of HFA-KM mice (28.45 % ± 0.74) (P \ 0.01) (Fig. 3b). However, there was no significant difference in the dice similarity coefficient between HFA-KM (69.92 %) and HFA-C57BL/6J (68.66 %) (P [ 0.05) (Fig. 4). Principal Component Analysis
different from those bands with the same migration position as the donor, indicating that these microbes had immigrated into the mouse gut from the donor but the ratio in the host gut had changed. Furthermore, some bands were present only in one of the HFA mouse strains, such as bands 3, 6, 7, 10, 14, and 17, which may imply that the diversity of the gut microbiota was affected by the genetic background of the host. Diversity Analysis Microbial diversity indices of each lane in Fig. 2a were calculated based on the digital statistical information from the DGGE profile. Each microbial diversity index for the two strains of HFA mice and the human donor was compared using one-way ANOVA, and the result is shown in Fig. 3a. Highly significant differences were evident when comparing the gut microbial diversity in the two HFA mice using the Shannon–Wiener (H’) and richness indices,
123
Three principal components were extracted by principal component analysis, and scatter plots were generated in Origin 8.0 (Fig. 5). Each point indicates the spatial position of each sample and the entries are distributed according to their relatedness [17]. The HFA-KM, HFA-C57BL/6J, and human samples localized to the different areas in the plot, respectively (Fig. 5). The PCA separated the different groups completely, indicating that there were distinct differences in gut microflora composition between the human donor and the two mouse strains. Identification of Bands in DGGE Profiles To investigate the similarities and differences in gut microbial community composition between the human donor and the two HFA strains, 17 bands in the gel (Fig. 2a) were extracted corresponding to the 16S rRNA
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
Fig. 5 Principal component analysis of mice fecal bacteria community based on 16S rRNA gene PCR-DGGE profile at the fourth week. Values on the axis indicate variation percentages contributed to the first three principal components, respectively
Table 2 Closest relatives associated with bands in DGGE profile Band no.
Length (bp)
Closest match
Gene bank accession no.
Similarity (%)
1
436
Lactonifactor longoviformis
KC140198
93
2
437
Acinetobacter calcoaceticus
KC140199
99
3
365
Ruminococcus gauvreauii
KC161373
95
4
369
A. gyllenbergii
KC140197
91
5
350
R. torques
KC140200
91
6
364
Clostridium clostridioforme
KC140196
96
7
353
C. clostridioforme
KC140204
99
8
348
Blautia luti
KC172380
93
9
373
B. wexlerae
KC172381
97
10
359
B. wexlerae
KC172382
98
11
348
B. wexlerae
KC140195
100
12
369
C. hathewayi
KC140194
98
13 14
344 347
B. wexlerae C. lavalense
KC140203 KC140202
94 99
15
437
C. lavalense
KC140193
99
16
437
B. schinkii
KC140192
96
17
436
Stenotrophomonas maltophilia
KC140191
98
base sequences with those in NCBI were between 91 and 100 % (Table 2). It is interesting that there were six bands (Bands 8, 9, 10, 11, 13, and 16) belonging to the genus Blautia, and another five sequences (Bands 6, 7, 12, 14, and 15) belonging to the genus Clostridium, indicating that these two genera were readily implanted into the guts of mice. Moreover, four of the six sequences (Bands 9, 10, 11, and 13) of Blautia spp. were from HFA-C57BL/6J samples with weak or no signal in HFA-KM and the five sequences of Clostridium spp. were from the gel of the HFA-KM samples with weak or no signal in HFA-C57BL/6J, suggesting that the Blautia spp. was better adapted to the C57BL/6J mice, whereas the Clostridium spp. was more adapted to the intestinal environment in KM mice. Bands 1, 2, 5, and 8 from the human samples, whose nearest relatives were Lactonifactor longoviformis, Acinetobacter calcoaceticus, Ruminococcus torques and B.luti, respectively, were not found in the mouse samples. Bands 4 and 13, which were most closely related to A. gyllenbergii and B.luti, respectively, were only found in the humanized mice, implying that the colonic microfloral composition was affected by genetic factors.
Discussion
Bands numbers are indicated by arrows in Fig. 2a
base sequences (NCBI accession number: KC140191 to KC140200, KC140202 to KC140204, KC161373, and KC172380 to KC172382). These sequences were used to perform a BLAST search in NCBI. The identities of these
This study focused on the intestinal microbial diversity in two strains of HFA mice to determine if HFA-KM mice, a closed colony mouse strain, could be an alternative HFA model used to study environmental interventions for altering the gut microbiota to promote health. After inoculation with a suspension of human feces, a substantial difference in the band types present in individuals from the same HFA mouse strain was observed in the first 2 weeks. However, 2 weeks later, each of the two mouse strains gradually formed a unique band pattern, demonstrating that there is a difference in the banding types of the two strains of HFA mice. This was corroborated by the observation that the samples of the three groups clearly separated into a hierarchical clustering dendrogram at the 4th week (Fig. 2b). PCA scatter of the DGGE profile also made it clear that the spatial distribution of the mouse samples was consistent with the results of the cluster analysis of the DGGE profile presented in Fig. 2b. Compared with gut microbiota of the human donor, the similarities ratio of the inbred HFA-C57BL/6J mice (47.41 %) was much higher than the closed colony HFAKM mice (28.45 %). However, KM mice have advantages over the C57BL/6J strain such as their larger size, which allows researchers to obtain more blood samples, and they survive longer and reproduce at a higher rate than C57BL/ 6J mice. What is more, our study proved that there was no
123
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
significant difference (P [ 0.05) in the dice similarity coefficient between HFA-KM mouse replicates (69.92 %) and HFA-C57BL/6J mouse replicates (68.66 %) (Fig. 4). In theory, the variation of the microbial communities of individuals within HFA-KM mice may be more than that of HFA-C57BL/6J mice because of the lower gene homogeneity of KM mice compared to C57BL/6J mice. However, the above result indicated that the diversity of colonic flora was not affected in a significant way by the gene heterogeneity of the KM strain. There are three dominant phyla (Firmicutes, Bacteroidetes, and Actinobacteria) in the adult human gastrointestinal tract. Firmicutes and Bacteroidetes are predominate and Actinobacteria is a lesser but significant contributor to the overall community. Each of Firmicutes and Bacteroidetes comprise *30 % of bacteria in feces and the mucus overlying the intestinal epithelium [18, 19]. Proteobacteria are common but usually not dominant [19, 20]. In our analysis, we identified 17 sequences. Bands 2, 4, and 17 in Fig. 2b were related to the phylum Proteobacteria and the remaining 14 belong to the phylum Firmicutes. We found that the two phyla of the human gut microflora were stably implanted into the colon of the two strains of HFA mice, and Proteobacteria emerged as a dominant member of the flora in the HFA mouse colon. However, the predominant genera in the colon of HFA-KM mice were different from those of HFA-C57BL/6J mice; the predominant genus in HFA-KM was Clostridium spp., in contrast to HFAC57BL/6J whose predominant genus was Blautia spp; it is also notable that Blautia spp. was recently discovered to be a novel genus and B. wexlerae is proposed to be a new species isolated from human feces [21]. This further confirmed that the composition of the dominant microbial community is affected by host genetic backgrounds, which have been verified by PCR-DGGE/TTGE, ERIC-PCR [15], T-RFLP [22], metagenomic analysis, and other cultureindependent methodologies [23].
Conclusion In conclusion, three important points can be inferred. Firstly, a reliable HFA model should be exposed to donor microbiota for at least 3–4 weeks before used in experiments. Secondly, hereditary differences between strains affect not only the overall gut microbial structure, but also lead to the distinct predominated genus. Furthermore, HFA-KM mice may be an attractive alternative HFA model for researches related to microbial communities for the little variation of the microbial communities of individuals within the group, especially when certain microbial genera cannot be implanted into the colons of HFAC57BL/6J mice.
123
Acknowledgments This work was supported by Grant 31071528 and 81370906 from the National Natural Science Foundation of China, Grant 2013CB531406 from the National Basic Research Program of China, and Grant 2012B091100429 and 2011B090400260 from Production-Learning-Research Program of Guangdong Science and Technology Department.
References 1. Round JL, Mazmanian SK (2009) The gut microbiota shapes intestinal immune responses during health and disease. Nat Rev Immunol 9(5):313–323. doi:10.1038/nri2515 2. Sanz Y, Santacruz A, Gauffin P (2010) Probiotics in the defence and metabolic balance of the organism: gut microbiota in obesity and metabolic disorders. P Nutr Soc 69:434–441. doi:10.1017/ S0029665110001813 3. Xu J, Gordon JI (2003) Honor thy symbionts. P Natl Acad Sci Usa 100(18):10452–10459. doi:10.1073/pnas.1734063100 4. Martens EC, Koropatkin NM, Smith TJ, Gordon JI (2009) Complex glycan catabolism by the human gut microbiota: the Bacteroidetes sus-like paradigm. J Biol Chem 284(37):24673– 24677. doi:10.1074/jbc.R109.022848 5. Perrin-Guyomard A, Poul JM, Laurentie M, Sanders P, Ferna´ndez AH, Bartholomew M (2006) Impact of ciprofloxacin in the human-flora-associated (HFA) rat model: comparison with the HFA mouse model. Regul Toxicol Pharm 45(1):66–78. doi:10. 1016/j.yrtph.2006.02.002 6. Perrin-Guyomard A, Poul JM, Corpet DE, Sanders P, Ferna´ndez AH, Bartholomew M (2005) Impact of residual and therapeutic doses of ciprofloxacin in the human-flora-associated mice model. Regul Toxicol Pharm 42(2):151–160. doi:10.1016/j.yrtph.2005. 03.001 7. Gootenberg DB, Turnbaugh PJ (2011) Companion animals symposium: humanized animal models of the microbiome. J Anim Sci 89(5):1531–1537. doi:10.2527/jas.2010-3371 8. Pang XY, Hua XG, Yang Q, Ding DH, Che CY, Cui L, Jia W, Bucheli P, Zhao LP (2007) Inter-species transplantation of gut microbiota from human to pigs. ISME J 1(2):156–162. doi:10. 1038/ismej.2007.23 9. Turnbaugh PJ, Ridaura VK, Faith JJ, Rey FE, Knight R, Gordon JI (2009) The effect of diet on the human gut microbiome: a metagenomic analysis in humanized gnotobiotic mice. Sci Transl Med 1(6): 6ra14. doi:10.1126/scitranslmed.3000322 10. Turnbaugh PJ, Hamady M, Yatsunenko T et al (2009) A core gut microbiome in obese and lean twins. Nature 457(7228):480–487. doi:10.1038/nature07540 11. Huws SA, Edwards JE, Kim EJ, Scollan ND (2007) Specificity and sensitivity of eubacterial primers utilized for molecular profiling of bacteria within complex microbial ecosystems. J Microbiol Meth 70(3):565–569. doi:10.1016/j.mimet.2007.06.013 12. Gafan GP, Lucas VS, Roberts GJ, Petrie A, Wilson M, Spratt DA (2005) Statistical analyses of complex denaturing gradient gel electrophoresis profiles. J Clin Microbiol 43(8):3971–3978. doi:10.1128/JCM.43.8.3971-3978.2005 13. 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 Microb 59(3):695–700 14. Thompson JR, Marcelino LA, Polz MF (2002) Heteroduplexes in mixed-template amplifications: formation, consequence and elimination by ‘reconditioning PCR’. Nucleic Acids Res 30(9):2083–2088. doi:10.1093/nar/30.9.2083 15. Leung K, Topp E (2001) Bacterial community dynamics in liquid swine manure during storage: molecular analysis using DGGE/
X. Zhang et al.: Comparative Diversity Analysis of Gut Microbiota
16.
17.
18.
19.
20.
PCR of 16S rDNA. FEMS Microbiol Ecol 38:169–177. doi:10. 1111/j.1574-6941.2001.tb00895.x Yuan J, Zeng B, Niu R, Tang H, Li W, Zhang Z, Wei H (2011) The development and stability of the genus Bacteriodes from human gut microbiota in HFA mice model. Curr Microbiol 62(4):1107–1112. doi:10.1007/s00284-010-9833-9 Boon N, De Windt W, Verstraete W et al (2002) Evaluation of nested PCR-DGGE (denaturing gradient gel electrophoresis) with group-specific 16S rRNA primers for the analysis of bacterial communities from different wastewater treatment plants. FEMS Microbiol Ecol 39(2):101–112. doi:10.1111/j.1574-6941.2002. tb00911.x Backhed F, Ley RE, Sonnenburg JL, Peterson DA, Gordon JI (2005) Host-bacterial mutualism in the human intestine. Science 307(5717):1915–1920. doi:10.1126/science.1104816 Ley RE, Backhed F, Turnbaugh P, Lozupone CA, Knight RD, Gordon JI (2005) Obesity alters gut microbial ecology. P Natl Acad Sci Usa 102(31):11070–11075. doi:10.1073/pnas. 0504978102 Seksik P, Rigottier-Gois L, Gramet G, Sutren M, Pochart P, Marteau P, Jian R, Dore´ J (2003) Alterations of the dominant
faecal bacterial groups in patients with Crohn’s disease of the colon. Gut 52(2):237–242. doi:10.1136/gut.52.2.237 21. Liu C, Finegold SM, Song Y, Lawson PA (2008) Reclassification of Clostridium coccoides, Ruminococcus hansenii, Ruminococcus hydrogenotrophicus, Ruminococcus luti, Ruminococcus productus and Ruminococcus schinkii as Blautia coccoides gen. nov., comb. nov., Blautia hansenii comb. nov., Blautia hydrogenotrophica comb. nov., Blautia luti comb. nov., Blautia producta comb. nov., Blautia schinkii comb. nov. and description of Blautia wexlerae sp. nov., isolated from human faeces. Int J Syst Evol Micr 58(Pt8): 1896–1902. doi:10.1099/ijs.0.65208-0 22. Kibe R, Sakamoto M, Yokota H, Ishikawa H, Aiba Y, Koga Y, Benno Y (2005) Movement and fixation of intestinal microbiota after administration of human feces to germfree mice. Appl Environ Microbiol 71(6):3171–3178. doi:10.1128/AEM.71.6. 3171-3178.2005 23. Gill SR, Pop M, DeBoy RT, Eckburg PB, Turnbaugh PJ, Samuel BS, Gordon JI, Relman DA, Fraser-Liggett CM, Nelson KE (2006) Metagenomic analysis of the human distal gut microbiome. Science 312(5778):1355–1359. doi:10.1126/science. 1124234
123
