1. The Microbiome as a Gateway to Detect the Microbiota Composition: 16S Sequencing

The constant increase in the number of studies on the microbiota and its interactions with the human organism is largely due to the use of new technologies involving sequencing analysis. In fact, before the adoption of next-generation sequencing (NGS) in this field, the detection of microorganisms from different human districts was conducted by culturing methods, but, beyond being time-consuming, these tools were greatly hampered and unsuccessful due to “unculturable” bacteria ^[1].

In time, it became evident for the necessity to identify markers allowing for the detection of microorganisms in a particular substrate and, together, their identification at the species level with high confidence. A very good response to this unmet need was represented by a specific class of RNAs. In bacteria, the 5S, 16S, and 23S ribosomal RNAs are organized in a gene cluster which is expressed as a single operon. The size, sequence, and secondary structures of these three rRNA genes are highly conserved between different bacterial species. In particular, the 16S rRNA ^[2] originates from a gene of about 1500 bp present in all bacteria and contains nine hypervariable and species-specific regions (V1—V9) flanked by highly conserved and well-known portions of the genome. Thus, by using specific PCR primers, it is possible to amplify parts of the gene containing both the constant and the variable sequences, which are very useful for microbial classification (Figure 1). Indeed, this gene has been used for phylogenetic studies for many years and is considered the gold standard for microbiome detection and classification ^[3].

Figure 1. Structure of the 16S gene. The nine variable regions are depicted in green. Purple bars indicate the portions of the gene mostly used for bacterial classification upon PCR-based amplification and sequencing.

The first example of 16S amplification and sequencing using the Sanger method was presented in 1990 but was soon followed by the advent of high throughput sequencing ^[4][5], which represented a turning-point and a revolution in terms of costs and speed of detection, boosting the development of microbiome-based studies. Indeed, Sanger sequencing following PCR amplification was successful in identifying bacterial strains in monomicrobial infections, for example, outperforming conventional culture methods or diagnostic body fluid markers, but failed to identify unique molecular targets in polymicrobial infections. This is due to the intrinsic property of the Sanger method, which can be performed only on DNA molecules with the same sequence, while NGS allows for massive parallel sequencing of many different DNA templates.

In 2007, the Human Microbiome Project (HMP) was started to use next-generation sequencing methods to identify the abundance, diversity, and functionality of the microorganisms that live in different sites of the human body, thus generating a reference database for subsequent comparative analyses ^[6]. With this aim, 16S-focused investigations were implemented. The 16S analysis typically starts with the amplification of specific variable regions of the 16S rRNAs (usually V3—V4), to be massively sequenced in parallel; then, the obtained sequences are clustered into Operational Taxonomic Units (OTUs) ^[7]. These are defined according to their similarity to each other based on a threshold, usually defined as a sequence similarity of at least 97%. In time, several bioinformatics pipelines have been implemented to analyze the results of 16S sequencing, such as QIIME (Quantitative Insights Into Microbial Ecology, http://qiime.org/) ^[8] and MOTHUR (https://mothur.org/) ^[9], which were specifically designed for examining microbial communities.

The two principal parameters describing the complexity of microbiota in a definite environment are the α and β diversity. Alpha diversity describes the richness and evenness, i.e., the number of different organisms and the homogeneity of their distribution within a sample. Beta diversity is a measure of absolute or relative overlap in taxa shared between samples. There is a wide range of microbial β diversity in the microbiota between individuals since particular species could be widely abundant in some individuals and may be minimally represented in others ^[10]. In particular, more specific indexes belonging to both alpha and beta metrics are often used during the assessment and classification of bacterial communities. Among the most used there are the Shannon-Weaver, the Simpson, the Jaccard, and the Bray-Curtis, which quantify the taxonomic dissimilarity. Since all diversity indices have specific biases, they must be selected appropriately ^[11][12].

2. Circulating 16S in a Disease Context

As previously said, numerous studies analyzed the role of the microbiome in relation to the onset of different pathologies, most of them focusing on the intestinal microbiome. Lately, though, an increasing number of researchers investigating connections between dysbiosis in the blood microbiome and human disease. In a work from 2011, Amar et al. showed, for the first time, that the blood microbiome composition might predict the onset of diabetes, in a 6–9-year follow-up ^[13]. About a year later, the authors performed 16S qPCR on the blood of individuals without CVD at baseline and found a decrease in blood bacterial DNA and an increase of Proteobacteria in subjects who suffered from cardiovascular complications during the follow-up ^[14]. In a subsequent work in 2014 on circulating human microbiome in CVD subjects, Dinakaran et al. found an increase in microbial diversity and bacterial DNA concentration in the blood of patients ^[15]. Here, they observed a predominance of Actinobacteria, while the most abundant phylum in healthy subjects was Proteobacteria.

In a different study, Lelouvier et al. investigated the association between blood microbiota and the onset of liver disease ^[16]. The authors performed both 16S qPCR and sequencing to unveil the relationship between blood bacterial population composition and liver fibrosis in obese patients. The results showed higher concentrations of 16S in the blood of patients with fibrosis than in healthy subjects, thus identifying a specific microbial cluster associated with liver fibrosis that was suggested as a biomarker for its early detection.

Qian et al. in 2018 investigated the possible association between blood microbiota alteration and Parkinson’s disease ^[17]. The taxonomic diversity was assessed by performing 16S sequencing and some genera resulted associated with the pathology, with Cloacibacterium, Isoptericola, Paludibacter, and Saccharofermentans genus showing a correlation with disease duration.

More recently, a manuscript from Hammad and co-authors described the profiles of circulating microbial DNA in patients with rheumatoid arthritis (RA) in comparison with patients affected by ankylosing spondylitis (AS) or psoriatic arthritis (PA) and healthy control subjects ^[18]. Bacterial community members were identified by sequencing of the 16S rRNA variable region 4 in all samples. At the phylum level, the blood microbiome was predominated by four phyla, i.e., Proteobacteria, Firmicutes, Bacteroidetes and Actinobacteria, supporting the notion of a core blood microbiome as reported in previous publications ^{[19][14][16][20]}. In 2016, Santiago and coworkers presented an investigation assessing serum microbial composition in patients with and without ascites ^[21]. They performed 16S rDNA high-throughput sequencing, evidencing complex and specific microbial communities in serum and ascitic fluid of patients with cirrhosis. Interestingly, sera obtained from healthy controls resulted in an almost complete absence of bacteria. Of note, the authors indicated the presence of an unknown phylum belonging to Cyanobacteria in the serum of patients with ascites.