# Evaluation of propidium monoazide for 16S ribosomal RNA metabarcoding assessment of microbial communities in 60-day ripened raw goat milk cheese

## TL;DR

This study shows that using PMA improves the accuracy of identifying living microbes in ripened goat cheese by removing DNA from dead cells.

## Contribution

The study demonstrates that PMA treatment enhances microbiome accuracy in cheese by filtering out nonviable cell DNA.

## Key findings

- PMA treatment reduced farm-related microbes like Dickeya and Pectobacteriaceae.
- PMA did not change overall microbial diversity in cheese.
- PMA improves viable microbiota detection in ripened cheese.

## Abstract

Summary: This study evaluated whether propidium monoazide (PMA) treatment affects the characterization of microbial communities in 60-day-ripened raw goat cheese using 16S ribosomal RNA (rRNA) metabarcoding. The PMA selectively depletes DNA from nonviable cells, potentially providing a more accurate representation of living microorganisms. Paired samples (PMA-treated and nontreated controls) from the same cheese units were analyzed using Illumina MiSeq sequencing, with downstream analyses performed in QIIME 2 and R. Results showed no significant differences in alpha or beta diversity metrics between treatments, indicating that PMA did not alter the overall microbial community structure. However, PMA treatment significantly reduced the abundance of farm environment-associated taxa, specifically Dickeya and Pectobacteriaceae, suggesting these organisms were predominantly nonviable with residual DNA persisting in the cheese matrix. This demonstrates that PMA pretreatment improves the accuracy of cheese microbiome characterization by filtering dead cell DNA from environmental contaminants, providing clearer insights into viable microbial communities during cheese ripening.

Summary: This study evaluated whether propidium monoazide (PMA) treatment affects the characterization of microbial communities in 60-day-ripened raw goat cheese using 16S ribosomal RNA (rRNA) metabarcoding. The PMA selectively depletes DNA from nonviable cells, potentially providing a more accurate representation of living microorganisms. Paired samples (PMA-treated and nontreated controls) from the same cheese units were analyzed using Illumina MiSeq sequencing, with downstream analyses performed in QIIME 2 and R. Results showed no significant differences in alpha or beta diversity metrics between treatments, indicating that PMA did not alter the overall microbial community structure. However, PMA treatment significantly reduced the abundance of farm environment-associated taxa, specifically Dickeya and Pectobacteriaceae, suggesting these organisms were predominantly nonviable with residual DNA persisting in the cheese matrix. This demonstrates that PMA pretreatment improves the accuracy of cheese microbiome characterization by filtering dead cell DNA from environmental contaminants, providing clearer insights into viable microbial communities during cheese ripening.

•Environmental contaminants in ripened cheese decreased with PMA treatment.•PMA removes dead cell DNA but does not change overall cheese microbial diversity.•PMA improves accuracy of viable microbiota detection in ripened goat cheese.

Environmental contaminants in ripened cheese decreased with PMA treatment.

PMA removes dead cell DNA but does not change overall cheese microbial diversity.

PMA improves accuracy of viable microbiota detection in ripened goat cheese.

Cheese ripening is a complex microbial process marked by significant shifts in microbial composition. Considering that propidium monoazide (PMA) depletes DNA from nonviable cells, we hypothesized that PMA treatment of cheese samples could affect the microbiota characterization of 60-d-ripened raw goat curd cheese by 16S rRNA metabarcoding sequencing. After ripening, PMA-treated and nontreated (control) samples from the same cheese units were processed for DNA extraction, library preparation, and 16S rRNA metabarcoding sequencing on an Illumina MiSeq platform. Downstream bioinformatic analyses for microbial diversity assessment were performed using QIIME 2 and the phyloseq package in R. Statistical analyses included permutational multivariate analysis of variance (PERMANOVA), Wilcoxon tests, and linear discriminant analysis effect size (LEfSe). No significant differences were observed in either α or β diversity metrics between PMA-treated and nontreated samples. However, PMA treatment significantly reduced the abundance of farm environment–associated Dickeya and Pectobacteriaceae taxa in cheese samples, thus improving the accuracy of determining the cheese microbial structure using next-generation sequencing technologies. Further longitudinal studies focusing on different sampling periods during ripening, as well as other cheese types, may shed light on the potential benefits of using PMA for improving the accuracy of cheese microbial community characterization by next-generation sequencing.