| There are two types of genome in the human body, one is the human genome obtained from the genetic parents, and the other one is the microbial genome, mainly the gut flora genome, which is gained after born. The two genomes coordinate to maintain the host health. The intestinal flora is important for the health of the host and plays a role in the development, nutrition, metabolism, pathogen resistance, and regulation of immune system. Antibiotics may disrupt these coevolved interactions, leading to acute or chronic diseases. The poor sensitivity, inadequate resolution, and significant cost of current research methods have limited our understanding of the gut flora with the antibiotics treatment. The high-throughput sequencing technology and new clustering method of bacterial community have circumvented these limitations and revealed previously unexplored "rare biosphere" of gut microbiota.BIPES and TSC are two new methods for analyzing bacterial community being developed by our research team. The methods have high efficiency and low cost for processing the high-throughput sequencing tags. In the present study, we compared the chronic effect of the berberine, ciprofloxacin and metronidazole treatments on mice gut microbiota and analyzing the roxithromycin and azithromycin-associated disturbance of the mice intestinal flora.Objective:1. The present study aims to explore the effects of berberine, ciprofloxacin and metronidazole treatments on mice gut microbiota with15days gavage treatment and characterize the berberine, ciprofloxacin and metronidazole impact on intestinal microbiota, in addition, we would compare the effects of roxithromycin and azithromycin treatments on mice gut flora as well.Materials and Methods:Sample collectionIn the experiment one,32BALB/c mice were divided into4groups, namely, the berberine group, the ciprofloxacin group, the metronidazole group and the control group. Each group of the mice were treated respectively with0.100g berberine per kilogram of mice weight,0.072g ciprofloxacin per kilogram of mice weight,0.216g metronidazole per kilogram of mice weight and0.2ml each mouse. In the experiment two,24BALB/c mice were divided into3groups, namely, the roxithromycin group, azithromycin group and the control group. Each group of the mice were treated respectively with0.043g roxithromycin per kilogram of mice weight,0.072g azithromycin per kilogram of mice weight and0.2ml each mouse. All the fecal samples were collected respectively from the mice of each group before the processing of treatment. And each fecal sample was immediately stored at-80℃until DNA extraction without treatment.DNA extraction, PCR amplification and Illumina sequencingAll fecal samples were weighed10mg and added into100μl ddH2O, then vigorously homogenized by vortex for30s. Fecal suspensions were incubated at100℃in a boiling water-bath for15min, and immediately frozen at0℃on the ice. Then the suspensions were centrifuged at13,000xg at4℃for15minutes. The supernatants were transferred to a clean500μl tube and stored at-20℃before PCR. All direct boiling extractions were carried out in replicates. We amplified the16S rRNA V6tags as described in our previous reports using the967F (CNACGCGAAGAACCTTANC) and1046R (CGACAGCCATGCANCACCT) primers. The PCR cycle condition was an initial denaturation at94℃for2min,25cycles of94℃for30s,57℃for30s and72℃for30s, and a final extension at72℃for5min. Each25μl reaction consisted of2.5μl of Takara10×Ex Taq Buffer (Mg2+free),2μl of dNTP Mix (2.5mM each),1.5μl of Mg2+(25mM each),0.25μl of Takara Ex Taq DNA polymerase (2.5units),1μl of template DNA,0.5μl of10μM barcode primer967F,0.5μl of10μM primer1406R, and16.75μl of ddH2O. All the barcode-tagged16S rRNA V6tags PCR products were detected by agarose gel electrophoresis. Then we used the BandScan software to determine the mixed volume for each sample. The Invitrogen fluorescence quantitative instrument was used to test the concentration of PCR products. Finally, all amplified PCR products were sequenced using Illumina HiSeq2000for100bp from both ends at the BGI Genomics Institute (Shenzhen, China).Data analysisAfter obtained the raw data after sequencing, we used the BIPES pipeline to process the raw sequences. Briefly, we separated the paired-end (PE) reads with a Perl script from each sample according to their barcode sequence in sequencing file,1or2. The PE reads were overlapped to assemble the final V6tag sequences with the global alignment using the Needleman-Wunsch (NW) algorithm. We removed all of the sequences that contained one or more ambiguous reads (N), those that contained any errors in the forward or reverse primers, and those with more than1mismatch within the40-70bp regions during the overlap step. The variable tags (overlapped length minus primers and barcodes) that were shorter than50bp or longer than90bp were also removed. UCHIME was used to remove chimeras from the clean with de novo mode (parameters were set as:-minchunk20--xn7-noskipgaps2).We used a Two-Stage-Clustering (TSC) algorithm to cluster tags into operational taxonomy units (OTUs). Briefly, the tags with frequencies of3or greater were clustered using a stringent hierarchical clustering algorithm with the NW distance using the complete linkage method. The rare tags that occurred1or2times were clustered using a greedy heuristic algorithm with the NW distance using the single linkage. The method clustered highly abundant tags with high accuracy, while maintaining the diversity of rare tags and mitigating the effect of sequencing and PCR errors.The taxonomy assignment of tags and OTUs was performed using the Global Alignment for Sequence Taxonomy (GAST) method. The rarefaction curve and Shannon index curve were analyzed using mothur. PCoA analysis was implemented using QIIME based on UniFrac distance, including following processes: Representative sequences of each OTU were aligned using Pynast with Greengene core set as template file. Phylogenetic tree relating the OTUs was generated using FastTree, based on which UniFrac distance was calculated. Statistical analysis was implemented using SPSS17.0.Result:1.1Alpha diversity. We obtained2921901reads after Illumina sequencing from560fecal samples using the BIPES pipeline. A total of2602251clean sequences were got after quality control. The clean reads were screened to2578873sequences for the presence of chimeras using the UCHIME. Finally, we used TSC to cluster tags into operational taxonomy units (OTUs), and we got122436OTUs. To compare the diversity indices, we normalized the sequences of each sample to820reads. Tags with97%similarity (Needleman-Wunsch alignment) then were grouped into OTUs to calculate the diversity indices. The Shannon index obviously demonstrated that the alpha diversity of the mice intestinal flora decreased significantly under the processing of berberine, ciprofloxacin and metronidazole treatment. The bacterial community of the metronidazole group had the lowest bacterial richness and evenness, and the berberine group had the medium one, while the ciprofloxacin group was the highest. An approximate return to pre-treatment conditions occurred immediately after cessation of berberine and ciprofloxacin treatment, but the bacterial alpha diversity didn’t return within1month to the pre-treatment conditions in the metronidazole post-treatment state. On the second day of the berberine treatment, the Shannon index returned gradually, while the Shannon index returned on the fourth day of metronidazole treatment.1.2Beta diversity. The bacterial structure of berberine, ciprofloxacin and metronidazole on mice gut was different. The unweighted PCoA results showed that the samples of metronidazole group in the treatment and post-treatment state was separated from that of the control group, and the samples of berberine and ciprofloxacin group just in the treatment state were separated from that of the control group. The structure of intestinal flora changed obviously under the processing of berberine, ciprofloxacin and metronidazole treatment. The bacterial community of the metronidazole group changed the most seriously, and the berberine group had the medium one, while the ciprofloxacin group was the least. The structure of gut microbiota in the berberine group returned to the pre-treatment conditionsl month after the treatment. While stopped treatment of ciprofloxacin for2weeks, the structure of mice gut microbiota returned to the pre-treatment conditions. But an approximate return to pre-treatment conditions didn’t occur within1month after cessation of the metronidazole treatment. The results of weighted PCoA were similar with that of the unweighted PCoA. In another word, the treatment of berberine, ciprofloxacin and metronidazole significantly influenced the high abundant OUTs rather than the low abundant OTUs.1.3Phylum-level taxonomic distribution. The Firmicutes, Bacteroidetes, Proteobacteria and Actinobacteria contained a total of98.85%sequences. Under the processing of berberine, ciprofloxacin and metronidazole treatment, Firmicutes and Actinobacteria reduced and Proteobacteria enriched in the treatment state, but the Actinobacteria enriched in the berberine, ciprofloxacin and metronidazole post-treatment state. In addition, the Bacteroidetes significantly reduced in the metronidazole treatment.1.4Distribution of class in Proteobacteria phylum. The Gammaproteobacteria and Betaproteobacteria significantly enriched on the metronidazole and berberine treatment, but the increase rate of the berberine was much lower than that of metronidazole. Most of pathogens relative to the hospital-acquired infections belonged to Gammaproteobacteria. The Gammaproteobacteria of metronidazole significantly enriched to the rate of39.47%, and the rate of enrichment reduced with the cessation of the metronidazole treatment.1.5Genus-level taxonomic distribution. The top10most abundant genera differed significantly among the four groups in the treatment and post-treatment state. Under the processing of berberine treatment, Escherichia, Bacteroides, Lactobacillus and Parabacteroides enriched, while Bilophila and Eggerthella reduced. Meanwhile, Lactobacillus, Bacteroides and Eggerthella enriched and Papillibacter and Helicobacter reduced in the ciprofloxacin treatment. In addition, with the metronidazole treatment, Escherichia, Enterococcus and Paenibacillus enriched. And Bacteroide, Eggerthella, Alistipes, Papillibacter, Bilophila and Sporobacter reduced. Compared to the ciprofloxacin and metronidazole, the Bacteroides and Lactobacillus enriched during the berberine treatment and post-treatment state, and there were no new bacteria to colonize the intestinal tract, In contract, the gut flora began to return after a week of the cessation of metronidazole, the presence of new bacteria was observed on the fourth day in the treatment state, such as Paenibacillus and Clostridium.1.6Bacterial groups with statistical differences. Bacterial groups were distinct to the2groups using the default logarithmic (LDA) value of2.0. As the LEfSe showed, there were8groups of bacteria enriched in the berberine treatment state, namely, Escherichia (from phylum to genus), Bacteroides (from family to genus), Lactobacillus (from class to genus), Enterococcus (from class to genus), Sporobacter, Parabacteroides and Bryantella.The bacterial lineages reduced in the berberine treatment state were Bilophila, Eggerthella (from phylum to genus), Alistipes (from family to genus), Mucispirillum (from phylum to genus), Prevotella (from family to genus), and Allobaculum.When the default logarithmic (LDA) value was2.0, bacterial groups were distinct. There were7groups of bacteria enriched in the ciprofloxacin treatment state, namely, Bacteroides (from family to genus), Parabacteroides, Sporobacter, Lachnospira, Acinetobacter (from phylum to genus), Staphylococcus (from class to genus) and Mirabella (from class to genus).The bacterial lineages reduced in the ciprofloxacin treatment state were Astipes (from family to genus) and Anaerotruncus.When the default logarithmic (LDA) value was4.0, bacterial groups were distinct. There were7groups of bacteria enriched in the metronidazole treatment state, namely, Escherichia (from phylum to genus), Enterococcus (from class to genus), Paenibacillus and Clostridium (from family to genus). The bacterial lineages reduced in the metronidazole treatment state were Bilophila (from class to genus), Bacteroides, Alistipes (from phylum to genus), Mucispirillum (from phylum to genus) and Papillibacter (from phylum to genus).1.7The increase in the frequency of bacteria belonging to the genera Clostridium and Enterococcus and the family Enterobacteriaceae was noteworthy because many of the major bacterial species causing hospital-acquired infections belong to these groups. The Enterobacteriaceae enriched by the mean rate of29%in the metronidazole treatment state, the maximal increase rate was39%, and reduced gradually after cessation of metronidazole treatment. Besides, the Enterobacteriaceae enriched from the first day and the fifth day of the berberine treatment state, and the increase rate of berberine was much lower than that of metronidazole. Homoplastically, The Enterococcus enriched in the metronidazole treatment state, the maximal increase rate was20%, while the Enterococcus enriched in the first two days of the berberine treatment. In addition, the Clostridium were observed and enriched from the eighth day of metronidazole treatment state to the fourth day of the metronidazole post-treatment state. These and the recovery of gut microbial community in the metronidazole treatment state provide potential clues for hospital-acquired infections.The results of the Shannon index, PCoA and taxa distribution demonstrated that the gut flora returned after a week of metronidazole treatment, the new bacteria of Paenibacillus and Clostridium were observed and enriched in the metronidazole treatment. While there were no new bacteria to appear and enrich in the berberine treatment state.The Clostridium, Enterococcus and Enterobacteriaceae didn’t change in the ciprofloxacin treatment state. In another word, ciprofloxacin treatment prevented intestinal colonization of Clostridium, Enterococcus, and Enterobacteriaceae. 2.1Alpha diversity. We obtained2969614reads after Illumina sequencing from396fecal samples using the BIPES pipeline. A total of2636857clean sequences were got after quality control. The clean reads were screened to2601081sequenses for the presence of chimeras using the UCHIME. Finally, we used TSC to cluster tags into operational taxonomy units (OTUs), and we got181282OTUs.To compare the diversity indices, we normalized the sequences of each sample to1000reads. Tags with97%similarity (Needleman-Wunsch alignment) then were grouped into OTUs to calculate the diversity indices. Both Shannon index and Observed OTUs obviously demonstrated that the alpha diversity of the mice intestinal flora decreased significantly under the processing of roxithromycin and azithromycin treatment. An approximate return to pre-treatment conditions didn’t occur within1month after cessation of roxithromycin and azithromycin treatment.2.2Beta diversity. The effect of roxithromycin and azithromycin on mice structure of gut microbiota is similar. The results of unweighted PCoA showed that the samples of roxithromycin and azithromycin treatment and first two weeks of post-treatment were separated from that of the control group. When stop the two kinds of macrolide antibiotics treatment for1month, the structure of gut microbiota in mice didn’t recovered completely. The results of weighted PCoA showed that there was not a clear separation between the samples of each macrolide group and control group, In another word, the roxithromycin and azithromycin treatment didn’t significantly influence the high abundant OUTs.2.3Phylum-level taxonomic distribution. We assigned the taxonomy of all tags using the GAST method, the pipeline of which showed a high performance for analyzing the V6tag. Under the processing of roxithromycin and azithromycin treatment, three bacterial phyla, the Firmicutes enriched, while the Proteobacteria and Bacteroidetes reduced in the treatment state. And the Firmicutes continued to enrich and the Proteobacteria remained to reduce in the azithromycin post-treatment state.2.3Genus-level distribution. The top10most abundant genera differed significantly among the three groups in the treatment and post-treatment state. The dominate bacteria of Escherichia, Alistipes, Parabacteroides and Anaerotruncus enriched in the roxithromycin treatment state, while the Bacteroides and Bilophila reduced. Furthermore, there were3types of bacteria enriched in the treatment state, namely, Parabacteroides, Papillibacter and Alistipes, and the Bacteroides, Bilophila and Lactobacillus reduced.2.5Bacterial groups with statistical differences. In addition to a-and β-diversities, another major aim in comparing microbial communities is to find specialized bacterial groups within each type of sample. This tool of LEfSe can analyze bacterial community data at any taxonomy level. However, because analyzing the large number of OTUs determined in the present study was too computationally intensive, we only performed statistical analysis from domain to genus levels.Bacterial groups were distinct using the default logarithmic (LDA) value of2.0. As the LEfSe showed, there were2groups of bacteria enriched in the roxithromycin treatment state, namely, Escherichia (from class to genus), Anaerotruncus and Anaerostipes (from phylum to genus).The bacterial lineages reduced in the roxithromycin treatment state were Bacteroides (from phylum to genus), Bilophila (from phylum to genus), Anaerovorax (from family to genus), Helicobacter (from phylum to genus), Faecalibacterium, Prevotella (from phylum to genus), Mucispirillum (from phylum to genus), Bulleidia (from phylum to genus), Eggerthella (from phylum to genus), Odoribacter (from phylum to genus).When the default logarithmic (LDA) value was3.5, bacterial groups were distinct. There were4groups of bacteria enriched in the azithromycin treatment state, namely, Parabacteroides, Papillibacter (from phylum to genus), Alistipes (from family to genus) and Bryantella (from phylum to genus). While the bacterial lineages reduced in the roxithromycin treatment state, namely, Bacteroides (from phylum to genus), Bilophila (from phylum to genus), Lactobacillus (from class to genus), Mucispirillum (from phylum to genus), Helicobacter (from phylum to genus) and Prevotella (from phylum to genus).Conclusion:1.1The a-diversity and β-diversity of the mice intestinal flora decrease significantly under the processing of berberine, ciprofloxacin and metronidazole treatment. The bacterial community of the metronidazole group has the lowest bacterial richness and evenness, and the berberine group has the medium one, while the ciprofloxacin group is the highest. After cessation of the berberine, ciprofloxacin and metronidazole treatment, the structure of mice gut microbiota returns to the pre-treatment conditions in the berberine and ciprofloxacin group, but that of metronidazole group dosen’t.1.2The effect of metronidazole treatment on the mice gut flora is the strongest, while the time of the mice gut microbiota returning to the pre-treatment conditions is the longest. The Escherichia, Enterococcus, Paenibacillus and Clostridium enrich obviously in the metronidazole treatment. And the metronidazole treatment promotes intestinal colonization by the Paenibacillus and Enterococcus during the return of gut flora. The increase in the frequency of bacteria belonging to the genera Clostridium and Enterococcus and the family Enterobacteriaceae is noteworthy in the metronidazole group because many of the major bacterial species causing hospital-acquired infections belong to these groups.1.3As a traditional Chinese medicine, berberine has its unique characteristics. The effect of berberine treatment on the mice gut flora is lower than that of metronidazole, while the time of the mice gut microbiota returning to the pre-treatment conditions is shorter than metronidazole group. It is amazing that the Bacteroides and Lactobacillus enrich significantly in the berberine treatment. The diversity of inherent intestinal bacteria returns after cessation of berberine treatment.1.4Compared to the berberine and metronidazole group, the gut microbiota has the highest inertia and highest restoring forces in the ciprofloxacin treatment state. And the Bacteroides enriches in the ciprofloxacin treatment state. For protecting the normal intestinal flora, the ciprofloxacin is better than berberine, and the berberine is better than metronidazole.2.1The changes of a-diversity and gut microbiota structure are similar in the roxithromycin and azithromycin treatment and post-treatment2.2Both richness and evenness of the mice intestinal flora reduced in the roxithromycin and azithromycin treatment state. In addition, the structure of gut microbiota changes significantly as well and the Bacteroides reduces observably with the treatment.The roxithromycin and azithromycin have a long-term effect, because an approximate return to pre-treatment conditions doesn’t occur within1month after cessation of roxithromycin and azithromycin treatment. |
PDF Full Text Request