Probable alterations in fecal bacterial microbiota by somatostatin receptor analogs in acromegaly

Ethics, consent and permissions

All individuals were included in the study with written informed consent. The study were approved by the Hacettepe University institutional non-interventional clinical research ethics board (GO18/608, 21.06.2018) and performed in accordance with ethical standards of the Helsinki declaration and its later amendments.

Study design and cohorts

The study groups comprised individuals with the diagnosis of acromegaly without octreotide acetate (OCT) therapy (n=5), receiving OCT therapy (n=8) and healthy controls (n=5) (Supplementary File 1). The participants, originating from same geographic region, had no history of antibiotic use or dietary change in the last six months.

Specimen processing, 16S rDNA amplification and deep sequencing

Fecal samples were collected and subjected to DNA purification, using Qiagen Stool Mini Kit (Qiagen, Hilden, Germany) with a final DNA amount of 50 ng/μL for each specimen. For the amplification of 16S rDNA sequences, primers targeting the V3-V4 region were employed [5]. Adapter attachment, amplification and library preparation were performed using the Nextera XT Index and Nextera DNA Library Prep kits (Illumina, San Diego, California, United States). The sequencing runs were performed in the Illumina MiSeq system (Illumina Inc.).

Data handling, phylogenetic and statistical analyses

Sequence data handling and taxonomic assignment were carried out using Geneious v11.1 (Biomatters Ltd, Auckland, New Zealand), MALT v0.3.8 and Megan v6.11 [6], [7]. Alpha diversity metrics Shannon–Weaver and Simpson’s reciprocal indices were calculated using Megan. The trimmed bacterial reads were mapped to the NCBI-NT RefSeq 16S database. Relative abundances of different bacterial taxonomic levels were calculated using the reads numbers of the corresponding operational taxonomic units (OTUs).

Descriptive variables and total reads, reads assigned to bacteria and bacterial phyla were assessed via statistical tests. Student’s T, Kruskal–Wallis and Mann–Whitney U tests were employed for comparisons among groups where appropriate. The linear correlation between two variables was assessed via the Spearman correlation coefficient. All analyses were performed using Analyse-it software v4.20.1 (Analyse-it Software, Ltd. Leeds, United Kingdom).

Clinical and descriptive data

A total of 18 individuals (14 males, 4 females) were enrolled in the study, with an age range of 28–68 years (mean: 47.8, median: 48). Patients under therapy were receiving 10–40 mg of OCT. Five of the acromegalic patients had colon polyps and none had colorectal cancer (Supplementary File 1).

Bacterial communities in control and acromegaly groups

The initial analysis of the trimmed sequences revealed no statistically-significant difference among study groups for total, bacterial and unassigned reads. In all groups, predominant bacterial species were Bacteroidetes and Firmicutes (Figure 1). Members of the Proteobacteria and Actinobacteria phyla were also identified in all specimens, despite varying relative abundances (Table 1). Elusimicrobia, Spirochetes and Synergistetes were identified in trace copies in patients with acromegaly.

Figure 1:

Bacterial community compositions are given as relative abundance at the phylum level. Study cohorts and enrolled individuals are indicated as control (C1−5), acromegaly without OCT (OCT−) (A1−5) and acromegaly with OCT (OCT+) (O1−8). Color codes assigned to each phylum are shown.

Bacteroidetes sequences were observed to belong in Bacteroidaceae, Odoribacteraceae, Porphyromonadaceae, Prevotellaceae and Alistipes families, where Bacteroidaceae and Porphyromonadaceae predominate in the study groups (Figure 2). Members of these families were detected in all specimens, regardless of the study group except Odoribacteraceae, was absent in one individual of the control group (C3). The families with the phyla Firmicutes were identified as Clostridia, Lactobacillales, Erysipelotrichaceae and Veillonellaceae (Figure 3). Here, Clostridia were ubiquitous in all specimens (Figure 3). Bifidobacterium genus was evaluated separately, as presence and relative abundance of this genus also demonstrated variations in control and acromegaly groups (Table 1, Supplementary File 1). Furthermore, Akkermansia mucinophila was identified as the single bacterial species in the phylum Verrucomicrobia in the acromegaly groups.

Figure 2:

Bacterial community compositions of the phyla bacteroidetes, given as relative abundance at the family level. Study cohorts and enrolled individuals are indicated as control (C1−5), acromegaly without OCT (OCT−) (A1−5) and acromegaly. with OCT (OCT+) (O1−8). Color codes assigned to each family are shown.

Figure 3:

Bacterial community compositions of the phyla Firmicutes, given as relative abundance at the family level. Study cohorts and enrolled individuals are indicated as control (C1−5), acromegaly without OCT (OCT−) (A1−5) and acromegaly with OCT (OCT+) (O1−8). Color codes assigned to each family are shown.

Bacterial species detected in trace copies comprised Elusimicrobium minutum (phylum Elusimicrobia), Treponema succinifaciens (phylum Spirochetes), Pyramidobacter piscolens and Cloacibacillus porcorum (phylum Synergistetes) (Figure 1). These species were detected only in acromegaly groups, accompanied by A. mucinophila in three individuals (Supplementary File 1).

Comparative analysis of bacterial communities

Variations on the relative abundance of Bacteroidetes, Proteobacteria and Actinobacteria phyla were not statistically-significant among study cohorts (Table 1). Similarly, Bacteroidaceae, Odoribacteraceae, Porphyromonadaceae, Prevotellaceae and Alistipes families and Bifidobacterium genus revealed no significant variations. At the family level, relative abundance variations of bacteria belonging in Clostridia, Erysipelotrichaceae and Veillonellaceae were not statistically-supported among study groups. On the contrary, Lactobacillales abundance was significantly elevated in patients receiving OCT (Table 1). Interestingly, when patients with detectable A. mucinophila sequences (n=6) were overviewed, five were observed to be receiving OCT, along with one individual without OCT and receiving metformin for diabetes mellitus (patient A5, Supplementary File 1). When this individual was omitted from the analysis, A. mucinophila detection became specific for the OCT group, with statistical significance (p=0.025). No significant variation was observed between Shannon diversity indices between study groups.