Genotype diversity in the honey bee parasite Nosema ceranae: multi-strain isolates, cryptic sex or both?

There is great controversy as to whether Microsporidia undergo a sexual cycle. In the paradigmatic case of Nosema ceranae, although there is no morphological evidence of sex, some meiosis-specific genes are present in its reduced genome and there is also high intraspecific variability, with incongruent phylogenies having been systematically obtained. The possibility of sexual recombination is important from an epidemiological standpoint, particularly as N. ceranae is considered to be a major factor in the current disquieting epidemic of widespread bee colony losses. This parasite apparently originated in oriental honey bees, spreading out of Asia and Australia to infect honey bees worldwide. This study had three main objectives: i) to obtain genetic markers that are not part of known multi-copy arrays for strain determination; ii) to shed light on the intraspecific variability and recombination of N. ceranae; and iii) to assess the variability in N. ceranae populations. The answers to these questions are critical to understand the capacity of adaptation of microsporidia.

The N. ceranae isolates from honey bees correspond to genotypically distinct populations, revealing that individual honey bees may not be infected by a particular clone but rather, a pool of different strains. Homologous recombination implies the existence of a cryptic sex cycle yet to be described in N. ceranae. There are no diagnostic alleles associated with Australian or European origins, nor are there differences between the two hosts, A. cerana and A. mellifera, supporting the absence of biological barriers for N. ceranae transmission. Diversity is high among microsporidia of both these origins, and the maintenance of a high heterozygosis in the recently invaded European populations, could hypothetically underlie the stronger virulence of N. ceranae observed in A. mellifera.

Biallelic polymorphisms were detected at a number of specific points in the five coding loci analyzed from European and Australian isolates of N. ceranae. Heterozygous genotypes were abundant and cloning experiments demonstrate that they reflect the existence of multiple alternative sequences in each isolate. The comparisons of different clones and genotypes clearly indicate that new haplotypes are generated by homologous recombination. Conclusions: The N. ceranae isolates from honey bees correspond to genotypically distinct populations, revealing that individual honey bees may not be infected by a particular clone but rather, a pool of different strains. Homologous recombination implies the existence of a cryptic sex cycle yet to be described in N. ceranae. There are no diagnostic alleles associated with Australian or European origins, nor are there differences between the two hosts, A. cerana and A. mellifera, supporting the absence of biological barriers for N. ceranae transmission. Diversity is high among microsporidia of both these origins, and the maintenance of a high heterozygosis in the recently invaded European populations, could hypothetically underlie the stronger virulence of N. ceranae observed in A. mellifera.