Evaluation of the Microbial Diversity in Amyotrophic Lateral Sclerosis Using High-Throughput Sequencing

1Amyotrophic lateral sclerosis (ALS) belongs to idiopathic, fatal neurodegenerative disease of the human motor system, characterized by the loss of neurons at all levels of the motor system—from the cortex to the anterior horn of the spinal cord. The scientific and clinical interest in ALS is growing since the 1990s, and survival in ALS is now understood to be dependent on clinical presentation (phenotype), rate of disease progression, early presence of respiratory failure, and the nutritional status of patients. Unfortunately, less than 50% of patients can survive within 3 years of onset.

2The human gastrointestinal tract is home to bacterial communities, and the microbes have profound implications on human metabolism, immunity and the gut-brain axis, and numerous studies have highlighted interactions between the central nervous system (CNS) and the gastrointestinal system. The brain may modulate peripheral gut functions to modify the gastrointestinal composition via releasing gut factors (hormones, neurotransmitters, immune factors), and the gut microbes, on the other hand, interact with the CNS by releasing of neurotransmitters, e.g., nitric oxide (NO, a major neurotransmitter in the brain) and g-aminobutyric acid (GABA, neurotransmitter produced Lactobacillus and Bifidobacterium). Short-chain fatty acids (SCFAs), the specific metabolites generated by gut bacteria, can cross the blood–brain barrier and its levels in the feces could be correlated negatively or positively with autism spectrum disorders (ASD). Moreover, researchers found that lipopolysaccharide (LPS), a constituent of Gram-negative bacteria markedly affected vagal afferent neuron function, with reduced vagal afferent leptin signaling.

3Intestinal barrier dysfunction may promote the passage of toxins in the intestinal lumen into the blood, and the innate immune response and increased circulating LPS play pivotal roles in the pathogenesis of ALS. Furthermore, reduced tight junction proteins in the lumbar spinal cord, as well as the disruption of tissue barriers (the blood–spinal cord barrier and the blood–brain barrier) were identified both in ALS patients and animal models. However, the interaction of the gut microbiota with the ALS has not been investigated.

4In the past, only a small fraction of all bacteria have been isolated and characterized severely limited by available technology and the shortage of reference genomes, and recent technological advances in next generation sequencing technology has enabled elucidation of the pleiotropic effects of microorganisms on the human host. In the present study, the high-throughput sequencing analyses were used to assess the interaction of the gut microbiota and the ALS, which proves basic data for the prevention and treatment of ALS.

5The study was approved by the Ethical Committee of The First Affiliated Hospital of Nanchang University, all participants provided written informed consent and all the methods were carried out in accordance with the approved guidelines.

6Six consecutive patients with ALS (according to the revised El Escorial criteria) were recruited at The First Affiliated Hospital of Nanchang University between 07/2015 and 05/2016, and patients who were unable to communicate, either verbally or by writing, were excluded. None had additional neurological disease or previous mental illness. Respiratory function measured by forced vital capacity (FVC) was above 70% and there was no evidence of nocturnal hypoventilation (Supplementary Table S1). Five healthy people without ALS were recruited as control. All people with random diets donated their first fecal motion of the day for only one time and the samples were stored at -70°C.

7Genomic DNA from each sample was extracted using a TIANamp Genomic DNA kit (TIANGEN) combined with bead beating as previously published. Then the Genomic DNA was sent to the one of the most famous high-throughput sequencing company for high-throughput sequencing and analysis.

8The extracted genomic DNA was used as the template to amplify the V3–V4 region of 16S rRNA genes using the primer pair 338F/806R with the barcode. PCR reactions, pyrosequencing of the PCR amplicons and quality control of raw data were performed as described previously with minor modification.

9Low-quality sequences were eliminated from analysis based on the following criteria: (a) raw reads shorter than 400 bp; (b) a sequence producing more than eight homopolymers; (c) >2 mismatches in the primers, or, (d) one or more mismatches in the barcode. Pyrosequenced amplicons were removed using the PyroNoise algorithm in Mothur. Bioinformatic analysis was implemented using the Quantitative Insights Into Microbial Ecology (QIIME) platform. Briefly, 16S rRNA operational taxonomic units (OTUs) were clustered using an open-reference OTU picking protocol based on 97% nucleotide similarity with the UCLUST algorithm. ChimeraSlayer was employed to remove chimeric sequences. The relative abundance of each OTU was determined as a proportion of the sum of sequences for each sample. Taxonomic relative abundance profiles (such as, at the phylum, class, order, family, and genus levels) were generated based on OTU annotation. The microbial community structure (i.e., species richness, evenness and between-sample diversity) of bacterial samples was estimated by biodiversity. Shannon index, phylogenetic diversity, Chao1 index, and the observed number of species were used to evaluate alpha diversity, and the weighted and unweighted UniFrac distances were used to evaluate beta diversity.

10All of these indices (alpha and beta diversity) were calculated by the QIIME pipeline.

11Statistical analysis was implemented using the R platform. Principal coordinate analysis (PCoA) was performed using the “ape” package based on the UniFrac distances between samples. The difference among groups was further assessed using a non-parametric test via Metastats software1 as described previously, and statistical significance was set at p < 0.05 for correction of multiple comparisons.

12To compare the fecal microbes of healthy people (group H) and ALS patients (group A), 16S rRNA amplicon sequencing analysis was used to sequence the V3–V4 hypervariable region, and the sequencing data was filtered to get the valid data, and all the effective tags of all samples were clustered and those sequences with over 97% similarity were considered as one OTU. In total, 802695.96 filtered clean tags (72972.36 tags/sample) and 2540 OTUs were obtained from all the samples with an average of 230.91 OTUs per group (Table 1). Chao1 index had almost got saturated and the rarefaction curve of every sample could enter the plateau phase (Supplementary Figure S1).

13The Venn figure could reflect the difference between group A and group H. As shown in Figure 1, there were 386 and 279 OTUs in group A and H, and the percent of their common OUTs were 63.0% (243/386) and 87.1% (243/279), respectively. For group H, 43.78% OTUs (169/386) were identified as common OUTs among samples H1, H2, H3, H4, and H5, while the common OTUs only occupied 17.56% (49/279) among samples A1, A2, A3, A4, A5, and A6.

14The overall picture of the microbial composition of the samples in group A and H was obtained by PCoA, based on the relative abundance profiles of bacterial taxa. As shown in Figure 2, 5/5 samples in group H clustered together on the right upper of the coordinate axis, and 5/6 samples in group A gathered together on the left upper of the coordinate axis, and samples in group H were obviously deviated from the samples in group A (Figure 2A), which was future confirmed by UPGMA method (Figure 2B).

15At the genus level, data of top 10 microorganism populations was analyzed. As shown in Figure 3, Bacteroides, Faecalibacterium, Anaerostipes, Prevotella, and Escherichia constituted five common dominant genus in group A and H (7.38 vs. 9.41%, 15.32 vs. 7.02%, 23.9 vs. 7.0%, 10.42 vs. 29.86%, 2.57 vs. 3.24%), which accounted for 59.59 and 56.53% of the total sequencing number, and the bacteria did not belong to the dominant bacteria in these two groups and classified as the “others” had occupied 30.77 and 38.29%. In addition, the average ratios of Bacteroides, Faecalibacterium, Anaerostipes, Prevotella, Escherichia, and Lachnospira between groups A and H were 0.78, 2.18, 3.41, 0.35, 0.79, and 13.07 (Figure 3).

16To determine the significant increased bacteria in group A or H, supervised comparisons by LEfSE (LDA > 4.0) were performed. In Figure 4, Lachnospiraceae (at family level), Firmicutes (at phylum level), Clostridia (at class level), Oscillibacter (at genus level), Family XIII (at family level), Anaerostipes (at genus level), Lachnospiraceae (at genus level) and Clostridiales (at order level) in group H were significant higher than that in group A, while Bacteroidetes (at phylum level), Bacteroidia (at class level), Bacteroidales (at order level), Dorea (at genus level) were significant higher than that in group H.

17Accumulating clinical- and scientific research-based evidence is driving our increased awareness of the significance of the human microbiome (HM) to the healthy and homeostatic operation of the human CNS. ALS belongs to neurodegenerative disease characterized by the loss of motor neurons, and the prevalence rate for ALS is substantially lower at 3.9/100000 in the United States. To date, the pathogenesis of ALS remains unclear and is likely multifactorial, and the pathophysiology of ALS may be related to the gastrointestinal tract. The gut microbiota, which is also referred to as the second brain, may affect brain activity through the gut-microbiota–brain axis under both physiological and pathological conditions, and accumulating evidence suggests that microbiota changes in the gastrointestinal tract of individuals possessed strong connection with neurological diseases and specifically, neurodegenerative diseases.

18In this study, high-throughput sequencing technology was used to compare the microbiota in intestinal tract of healthy people and ALS patients. To evaluate the tag quality, the raw tags, clean tags, average bp and OTUs in per sample were compared, and the mean number of 72972.36 clean tags, average length of 453 bp (Table 1), and the saturated Chao1 index and rarefaction curve ensured their reliability for the future analysis (Supplementary Figure S1).

19In Figure 1, the Venn figure reflected a high percent of 43.78% of common OTUs in H group, and a low percent of 17.56% of common OTUs in A group, which indicated that the ALS, together with individual physiological status, had severely changed the microbial composition in patients feces, which deviating from the normal microbiota and characterized by the overgrowth of total OTU number and low percent of common OTUs. Moreover, the clustered samples A1, A2, A3, A4, and A6, as well as the clustered samples H1, H2, H3, H4, and H5 using PCoA and UPGMA method further conformed the microbial changes in feces of ALS patients (Figure 2).

20Then, the top 10 microorganism populations were analyzed at genus level, and average ratios of Bacteroides, Faecalibacterium, Anaerostipes, Prevotella, Escherichia, and Lachnospira between groups A and H were 0.78, 2.18, 3.41, 0.35, 0.79, and 13.07 (Figure 3). Furthermore, supervised comparisons by LEfSE (LDA > 4.0) were performed to find the significant changed bacteria, and the relative richness of Firmicutes at phylum level, Clostridia at class level, Clostridiales at order level, Lachnospiraceae and Family XIII at family level, Oscillibacter, Anaerostipes and Lachnospiraceae at genus level in group H were significant higher than that in group A, while Bacteroidetes at phylum level, Bacteroidia at class level, Bacteroidales at order level and Dorea at genus level were significant higher in group A (Figure 4).

21In healthy adults, 80% of the identified fecal microbes can be classified into three dominant phyla: Bacteroidetes, Firmicutes, and Actinobacteria, and the Firmicutes to Bacteroidetes ratio is regarded to be of significant relevance with human health, and the significant increase of Firmicutes in H group and significant increase of Bacteroidetes in A group indicated that the ALS has seriously influenced patients’ healthy, characterized by the decreased Firmicutes/Bacteroidetes ratio. At genus level, the Dorea in group A was significant higher than that in healthy people, whose major end products of glucose metabolism are ethanol. Moreover, the significant decrease of Oscillibacter (was found in significantly more samples from healthy control test subjects than from patients diagnosed with Crohn’s disease; ), Anaerostipes (represents more than 2% of total colonic microbiota in the healthy colon, and are believed to play an important functional role in the gut ecosystem due to their ability to produce butyrate from lactate; ) and Lachnospiraceae (can protect from colon cancer in humans by producing butyric acid; ) in group A further confirmed the interaction of ALS with intestinal microbiota.

22In summary, we found that host microbiota were markedly different in health and disease, and the overgrowing of pathogens and reduction of probiotic organisms in intestines of ALS patients might up-regulated or down-regulated the production of NO, GABA, SCFAs, and LPS, which eventually increased the pathogenesis of ALS, and the ALS conversely aggravated the imbalances of intestinal microbiota, causing a vicious circle for host health. In the present study, we provide basic data to clarify the key bacteria during disease occurring, which may assist our understanding and treatment of ALS by inhibiting the growth of pathogens and enhancing the number of probiotics.

23TC designed the experiment; XF, XnW, SY, FM, and XaW performed the experiments; TC and HW analyzed the data and wrote the manuscript. All authors discussed the results and commented on the manuscript.

24The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.