MOLECULAR DISSECTION AND ENGINEERING OF GROUP B STREPTOCOCCAL ENDOLYSINS

Abstract

Group B Streptococcus (GBS) is an opportunistic pathogen that asymptomatically colonizes the genital tract of pregnant women. It may ascend through the cervix to the placenta and cause invasive infections, or it may be passed to the neonate through the vaginal tract at the time of birth. GBS is responsible for an array of severe outcomes such as stillbirth, preterm birth, neonatal meningitis, and even neonatal and maternal death. Despite advances in vaccination efforts, none of the vaccines have yet reached phase III clinical trials, let alone been approved. Furthermore, antibiotic resistance is increasing among bacteria. One alternative to antibiotics is bacteriophage-encoded endolysins, which are cell wall hydrolases produced at the end of a bacteriophage infection cycle, causing bacterial lysis from within. Endolysins are also capable of lysing bacteria from the outside when applied externally. Endolysins targeting Gram-positive bacteria, such as Streptococcus, typically comprise of an enzymatically active domain (EAD) and a cell-wall binding domain (CBD). Frequently, endolysins may have multiple enzymatic domains. PlyGBS, an endolysin targeting GBS, has two EADs and a CBD. We use rational engineering approaches to understand the mechanisms and importance of each domain and answer a long-standing question regarding the necessity of CBD for bacteriolytic activity. We identify a calcium-binding loop in one of the EADs (the CHAP domain), which might explain the bacteriolytic activity of the CHAP domain even in the absence of a CBD. We then evaluate the binding efficacy of the CBD to GBS and compare it with a CBD from another endolysin, PlySs2, and conclude that the native CBD binds poorly to GBS as compared to the PlySs2 CBD. The fusion of the PlySs2 CBD to the CHAP domain significantly increases the bacteriolytic activity as compared to the fusion of native CBD to the CHAP domain. Taken together, these results show that the presence of a calcium-binding loop in some CHAP domains may explain why they retain bacteriolytic activity in the absence of CBD. Further, an efficiently binding CBD will most likely increase the endolysin's bacteriolytic activity, and it may prove to be indispensable. While engineering endolysins for improved bacteriolytic activity and an extended host range, our group previously created a chimeric endolysin, ClyX-2, through the fusion of two endolysins, PlyC and PlySs2. We extensively characterize ClyX-2 against GBS in vitro and evaluate its efficacy in vaginal epithelial and mucosal models that mimic the vaginal microenvironment, concluding that ClyX-2 retains bacteriolytic activity in cervicovaginal mucosal environment, while reducing inflammatory response caused by GBS infection. Furthermore, ClyX-2 specifically targets GBS and not the commensal Lactobacillus. The results suggest that ClyX-2 can be an effective therapeutic in vivo. Our results answer key questions about the interplay between the EAD and CBD domains, which may guide further engineering efforts. We also demonstrate the efficacy of endolysins to eradicate GBS biofilms under physiological conditions and propose future avenues for its development in vivo.