ABSTRACT Chlamydia species are important pathogens that represent a paradigm for understanding successful obligate intracellular parasitism. C. trachomatis is a prevalent human pathogens, whereas C. muridarum is a murine-specific pathogen often used to study chlamydial disease in a small animal model. Species-specific variability in coding capacity within the so-called plasticity zone of the chlamydial chromosome is thought to be one source dictating differences in infection biology. The presence/absence of genes encoding glycosyltransferases designated as putative cytotoxins represents an intriguing example of this diversity. C. muridarum express multiple homologous proteins that also show similarity to a family of larger cytotoxins. The plasticity zone of common urogenital C. trachomatis serovars express a truncated protein while ocular serovars of members of the lymphogranuloma venereum (LGV) biovar lack the toxin genes completely. These cytotoxins have been proposed to mediate immediate toxicity in highly infected epithelial cells by interfering with actin polymerization. Lack of tractable genetics in C, muridarum and urogenital C. trachomatis, in particular has previously confounded definitive studies, leading to a paucity in details regarding molecular mechanisms of infection. We have overcome this barrier by deletion of putative cytotoxin genes in C. muridarum. Our data indicated that these proteins were dispensable for immediate toxicity. Combined with analysis of C. trachomatis CT166, our data implicate an alternative function for these proteins. We propose to apply genetic and biochemical approaches to delineate molecular function(s) of these putative glycosyltransferases and address their contribution(s) to species-specific infection biology.