Characterization of TM4 lysin A and generation of non-lytic mycobacteriophages

PAYASLIAN, FLORENCIA; URDANIZ, ESTEFANIA; MARTIN, MARIANO; PIURI, MARIANA

Congreso; Viruses of Microbes; 2022

Bacteriophage endolysins are crucial for progeny release at the end of the lytic cycle. In TM4, a bacteriophage that infects mycobacteria including M. tuberculosis, a cassette encoding putative LysA, LysB, and holin (gp29-30-31) was identified.Through bioinformatic analysis we were able to establish that LysA consists of 3 modules; a C-terminal domain that probably binds to the cell wall, a central domain with high similarity to an amidase-2, and an N-terminal domain proposed to encode for a peptidase. In the amidase domain, the catalytic Zn ion is coordinated by His226, His335, and Asp347 and we also identified the amino acid Glu290 as the catalytic residue.Four derivatives of the protein containing a mutation on each of these key residues were constructed.Purified fractions of LysA were incubated with MDP, a synthetic molecule that emulates the bonds on peptidoglycan. The products of the reaction were analyzed by HPLC-MS, and N-acetyl-muramic-acid and L-Ala-D-isoGlutamine dipeptide were detected, confirming that LysA has an amidase activity, as predicted in silico.We also assessed the ability of LysA to lyse E. coli or M. smegmatis from within, monitoring the optical density of cultures transformed with a plasmid expressing either LysA WT or the mutants. There was a significant decrease in the optical density of the cultures expressing the LysA WT version, but no lysis was observed in any of the mutants. These results indicate that the four predicted residues are essential for the function of the protein in vitro and when expressed in a homologous or heterologous host.In the laboratory, we work with a TM4 derivative that carries in its genome the mCherrybomb gene. This reporter bacteriophage reveals the metabolic state of mycobacteria, since a few hours after infecting them, the bacteria expresses and accumulates fluorescent protein, becoming easy to detect by fluorescence microscopy or fluorimetry. Through bacteriophage engineering, we modified the genome of this reporter phage and generated phages either carrying the E290 or the H266S mutations. Interestingly, no difference in the efficiency of plating was found between the mutated versions of the phage and the previous one. This could indicate that either these mutations were not sufficient to abrogate lysis in vivo, or that the lytic machinery is more complex than first described. To answer that question, we created two new versions of phages with stronger mutations. On one hand, we incorporated the two mutations, H226S and E290Q, simultaneously on the gp29 coding sequence, and the other hand, we constructed a bacteriophage that included four premature stop codons on the amidase domain. We were not able to isolate the double mutant, but we succeeded in obtaining the phage with the shorter version of the protein, that we called LysA***. When we assessed the phenotype of the LysA*** phage we found that, as expected it had a notoriously decreased efficiency when lysing a lawn of M. smegmatis mc2155. At the same time, it was able to generate lytic plaques in a lawn of a strain of M. smegmatis mc2155 that expresses the lysin A of another mycobacteriophage, that complemented the mutation. This last result corroborates that it is possible to abrogate the activity of LysA by interrupting the coding region. Leveraging the reporter properties of this phage, we determined the mutated phage is still capable of infecting the bacteria, showing that the lack of lytic plaques it is not due to a defective infection. To date, it is pending to further understand the biology of the mutated phage and the dynamics of the infection.