Comparison of qPCR protocols for quantification of "Candidatus Saccharibacteria", belonging to the Candidate Phyla Radiation, suggests that 23S rRNA is a better target than 16S rRNA

Abstract

Background: Candidate Phyla Radiation (CPR) is a large monophyletic group encompassing about 25% of bacterial diversity. Among CPR, "Candidatus Saccharibacteria" is one of the most clinically relevant phyla. Indeed, it is enriched in the oral microbiota of subjects suffering from immune-mediated disorders and it has been found to have immunomodulatory activities. For these reasons, it is crucial to have reliable methods to detect and quantify this bacterial lineage in human samples, including saliva.

Methods and results: Four qPCR protocols for quantifying "Ca. Saccharibacteria" (one targeting the 23S rRNA gene and three the 16S) were tested and compared. The efficiency and coverage of these four protocols were evaluated in silico on large genomic datasets, and in vitro on salivary DNA samples, already characterized by amplicon sequencing on the V3-V4 regions of the 16S rRNA. In silico PCR analyses showed that all qPCR primers lose part of the "Ca. Saccharibacteria" genetic variability, even if the 23S qPCR primers matched more lineages than the 16S qPCR primers. In vitro qPCR experiments confirmed that all 16S-based protocols strongly underestimated "Ca. Saccharibacteria" in salivary DNA, while the 23S qPCR protocol gave quantifications more comparable to 16S amplicon sequencing.

Conclusion: Overall, our results show that the 23S-based qPCR protocol is more precise than the 16S-based ones in quantifying "Ca. Saccharibacteria", although all protocols probably underestimate specific lineages. These results underline the current limits in quantifying "Ca. Saccharibacteria", highlighting the needs for novel experimental strategies or methods. Indeed, the underestimation of "Ca. Saccharibacteria" in clinical samples could hide its role in human health and in the development of immune-mediated diseases.

Fig 1. In silico PCR amplifications mapped on “Candidatus Saccharibacteria” 16S rRNA phylogenetic trees.

Results of in silico PCR amplifications were mapped on 16S rRNA phylogenetic trees to visualize the portions of “Ca. Saccharibacteria” known taxonomic diversity matched by the analyzed primer pairs. (a) Maximum Likelihood (ML) phylogenetic tree obtained from the 2,978 16S rRNA sequences annotated as “Saccharimonadia” in the 16S rRNA SILVA database. The five outer circles (coloured in a gradient from blue to green) represent the results of the in silico PCRs mapped on the tree, while the inner violet circle indicates the position of the lineages sequenced by D’auria et al. (2023). (b) ML phylogenetic tree obtained from 16S rRNA gene sequences retrieved from a manually curated collection of “Ca. Saccharibacteria” genomes. Similarly to above, the six outer circles (coloured from blue to yellow) map the results of the in silico PCRs experiments on the tree, while the violet inner circle maps the position of the best hits for the lineages sequenced by D’auria et al. (2023).

Fig 2. Comparison between the “Candidatus Saccharibacteria” quantification obtained by 16S amplicon sequencing and the four tested qPCR protocols.

(a-d) Linear regression graphs of the “Ca. Saccharibacteria” quantifications obtained by: (a) 16S amplicon sequencing vs the 23S qPCR protocol (SacchariF-SacchariR); (b) 16S amplicon sequencing vs 16S p1 (TM7314F/TM7-910R); (c) 16S amplicon sequencing vs 16S p2 (Sac1031-F/Sac1218R); (d) 16S amplicon sequencing vs 16S p3 (TM7_16S_590F/TM7_16S_965R). For each plot, the Pearson correlation coefficients (Rs) and p-values are reported on the top and the confidence interval area is shown in gray. (e) Boxplot graph of the “Ca. Saccharibacteria” quantifications obtained by 16S amplicon sequencing and the four qPCR protocols. The median values are compared between 16S amplicon sequencing and the other four qPCR protocols by Wilcoxon test (p-values are reported on the plot).

Fig 3. Comparison of “Ca. Saccharibacteria” quantifications in allergic vs control patients obtained by 16S amplicon sequencing and the four tested qPCR protocols.

Boxplots reporting the quantifications obtained by: (a) 16S amplicon sequencing; (b) 23S qPCR protocol (SacchariF/SacchariR); (c) 16S_p1 qPCR protocol (TM7314F/TM7-910R); (d) 16S_p2 qPCR protocol (Sac1031-F/Sac1218R); (e) 16S_p3 qPCR protocol (TM7_16S_590F/TM7_16S_965R). Values obtained from allergic patients vs controls were compared using Wilcoxon test and the p-values are reported on the bars.