First genomic analysis of a Clostridium perfringens strain carrying both the cpe and netB genes and the proposal of an amended toxin-based typing scheme

Takashi Mada¹Kenta Ochi²Mariko Okamoto¹Daisuke Takamatsu^1*

• ¹National Institute of Animal Health, National Agriculture and Food Research Organization, Tsukuba, Japan
• ²Ehime Prefectural Toyo Livestock Disease Diagnostic Center, Imabari, Japan

Clostridium perfringens strains are classified into seven toxinotypes (A-G) based on the profiles of the six typing toxin genes. Among these toxins, NetB is known as an important virulence factor for necrotic enteritis in chickens, and its gene, netB, is present only in type G strains. CPE is the enterotoxin that causes food-borne affections in humans, and its gene, cpe, is carried by type F strains and occasionally by type C, D, and E strains. However, strains with both netB and cpe are extremely rare; thus, they are not assigned to either toxinotype under the current typing scheme. In 2022, a 69-month-old female Holstein cow in Japan died suddenly, and a C. perfringens strain (CP280) possessing both netB and cpe was isolated for the first time in Japan from the bovine intestinal contents. The CP280 genome was composed of one chromosome and six circular plasmids, and netB and cpe were carried on different plasmids, pCP280-82k and pCP280-55k, respectively. Multilocus sequence typing analysis assigned CP280 to ST21, and all other reported ST21 strains were type G strains. In the phylogenetic analysis using the genomes of 553 C. perfringens strains, CP280 was clustered into a group along with the type G strains from affected birds. The deduced amino acid sequences of NetB and CPE from CP280 were identical to those of NetB and CPE from avian necrotic enteritis cases and human food poisoning cases, respectively, implying the potential of CP280 to cause these diseases. The genetic relatedness of CP280 and type G strains strongly suggests that CP280 was originally type G with the netB-positive plasmid pCP280-82k and later acquired the cpe-positive plasmid pCP280-55k; therefore, CP280 should be treated as a type G strain. We propose to change the requirement for this toxinotype in the toxin-based typing scheme from cpe(-) to cpe(+/-).