# Application of next-generation sequencing for detecting Mycoplasma contamination in veterinary vaccines

## TL;DR

This study shows that next-generation sequencing can detect Mycoplasma contamination in vaccines more accurately than traditional PCR methods.

## Contribution

The study introduces and validates a reference-mapping NGS method that outperforms PCR and metabarcoding for Mycoplasma detection.

## Key findings

- NGS-based reference mapping detected Mycoplasma with up to 100-fold lower detection limits than PCR.
- Reference mapping outperformed metabarcoding in sensitivity and specificity for Mycoplasma detection.
- The two-step reference-mapping strategy reduced non-specific contig formation compared to single-step approaches.

## Abstract

Ensuring the safety and efficacy of veterinary vaccines requires reliable methods for detecting microbial contamination, particularly from Mycoplasma species, which pose a significant risk in cell-culture-derived vaccines. In the Republic of Korea, polymerase chain reaction (PCR) is predominantly used for Mycoplasma testing due to its faster turnaround compared to culture-based methods. However, in combination with vaccines containing Erysipelothrix rhusiopathiae and classical swine fever virus, PCR is rendered ineffective because of cross-reactivity between Mycoplasma universal primers and E. rhusiopathiae, resulting in non-specific amplification. This limitation necessitates reliance on the labor-intensive culture method, underscoring the need for more accurate and efficient alternatives. This study aimed to develop and validate next-generation sequencing (NGS)-based methods for detecting Mycoplasma contamination in veterinary vaccines and to compare their performance with that of PCR. Five species, including Acholeplasma laidlawii (genus Acholeplasma) and four Mycoplasma species—Mycoplasma fermentans, Mycoplasma orale, Mycoplasma hyorhinis, and Mycoplasma synoviae–were spiked into samples containing E. rhusiopathiae, a common vaccine component. Two NGS-based approaches were evaluated: (1) a reference-mapping method incorporating two-step alignment and de novo assembly, and (2) a 16S rRNA-based metabarcoding analysis using DADA2 and Qiime2. The reference-mapping method effectively filtered non-specific reads and accurately reconstructed Mycoplasma-derived contigs, whereas the metabarcoding approach enabled taxonomic profiling with quantitative resolution. The detection limits of NGS-based methods were substantially lower than those of PCR, demonstrating improvements of up to 100-fold depending on the species. Notably, omission of the initial mapping step resulted in excessive non-specific contig formation, highlighting the importance of the dual-step reference-mapping strategy. Although metabarcoding provided valuable abundance data, it was more prone to non-specific hits due to limited read overlap. In conclusion, the reference-mapping method demonstrated superior sensitivity, specificity, and quantification compared to both conventional PCR and metabarcoding, supporting its use as a robust tool for vaccine quality control. Implementing NGS-based detection methods could significantly enhance the safety and effectiveness of veterinary vaccines, ultimately enhancing vaccine quality control.