Construction of mazE and mazF mutants by pnCasSA-BEC system

The pnCasSA-BEC plasmid system was used in the conversion of nucleotide base ‘C’ to ‘T’ in ST 239 (P-1780) clinical MRSA strain to generate a premature stop codon. The spacers of mazE and mazF genes that contain editable ‘C’ at position ‘7’ were assembled (Fig. 1a). The editable ‘C’ was converted to nucleotide ‘T’ for generating a premature stop codon. Since the transformation efficiency in ST239 MRSA strains is very low, assembled plasmids were transduced to wildtype P-1780 with the help of phage phi85. The CFU/ml count was significantly reduced (t-test, p-value < 0.001) for transduced mazE mutant cells compared to mazF mutant as shown in Fig. 1b. Thus, the active MazF toxin confers programmed cell death (PCD) on the P-1780 cells. The Q 16 of mazE and Q 14 of mazF, were mutated to stop codon successfully, as shown in Fig. 1a. The pnCasSA-BEC possesses the temperature-sensitive origin of replication (repF). Therefore, the same method was used to cure the plasmid as mentioned previously [14]. After growing the cells at a non-permissive temperature (42 °C), all the arbitrarily chosen colonies were seen to grow in the absence of chloramphenicol but were unable to grow with its presence, suggesting the effective removal of the pnCasSA-BEC from the cells.

Fig. 1
figure 1

Mutation of mazEF TA gene and its effect on growth rate in MRSA P-1780 strain. a Sequencing chromatogram revealed successful point mutation of C to T (marked in red) converting Q14 of mazE and Q16 of mazF, into a premature stop codon respectively. b Reduction in CFU count/ml post 16 h growth, after the transduction of the pnCasSA-BEC-mazEsp plasmid generating mazE mutants compared to mazF mutants. c Growth curve analysis of the wildtype, mazE and mazF mutants grown at 37 °C for 8 h in the TSB broth, for analyzing any growth defects

Growth characteristics of mazE and mazF mutants

The wildtype, mazE and mazF mutant strains were further tested for their growth characteristics by generating bacterial growth curves at various time intervals. The growth curve analysis (Fig. 1c) shows no change in the growth pattern in the case of both the mutants when compared to the wildtype strain, suggesting that mutants do not have any growth defects.

Differentially expressed genes from transcriptomics analysis

From the RNA-sequencing data, we found that a total of 178 genes were differentially regulated in the case of wildtype vs. mazE mutant strain, which was statistically significant. Out of the 178 genes, 65 genes were found to be upregulated, while 113 genes were downregulated. The differentially expressed genes are indicated in the volcano plot (Fig. 2a). This observation indicates that mutation of mazE generates an alteration in the pattern of gene expression profile in the clinical MRSA strain (Additional file 1: Table S2). Some of the genes affected by the mazE mutation are shown as heat map representation (Fig. 2b). Mutation in mazE gene lead to significant upregulation of the holin encoding cidA gene responsible for cell lysis induced biofilm formation with a concomitant reduction in the expression of lrgB gene encoding the anti-holin protein. The CidA protein forms a pore in the membrane through which CidB autolysin can easily pass, thus rupturing it and causing PCD [34]. On the other hand, the LrgB protein works in a contrary manner whereby it blocks the murein hydrolase activity of CidA [34]. Interestingly, in the mazE mutant, cell-lysis induced by atl gene encoding the autolysin protein was found to be downregulated. These autolysins are responsible for ica and PIA independent biofilm formation in clinical MRSA strain by its catalytic activity of the amidase region [35, 36]. On the other hand, sarA, another biofilm regulatory gene, known to contribute to persistent endovascular infection by promoting biofilm formation in MRSA cells showed upregulation in the mazE mutant strain [37, 38]. Also, the increased expression of the sarA and cshA gene, along with the cidA gene, in the mazE mutant strain remarkably points towards a tendency to form biofilm [37, 39] The alsS and the budA gene, responsible for mitigating nitrosative stress in S. aureus, were also seen to be upregulated in the generated mutants signifying that mazF helps in managing nitrosative stress [40]. Similarly, the betB gene was upregulated in the mazE mutants, which acts as a compatible solute in response to osmotic shock [41]. The urease BCDEFG operon, responsible for maintaining intracellular pH homeostasis by converting urea to ammonia and carbon dioxide, was found to be highly upregulated in the mazE mutants, showing that mazF promotes tolerance to acid stress, which arises from formate metabolism [42]. The rot (repressor of toxins) acts as a global regulator of virulence genes was downregulated [43]. Genes like mecA, vraS, vraR, tcaA and multidrug resistance proteins responsible for promoting antibiotic resistance were also significantly upregulated in the mazE mutant strains. On the other side, genes responsible for host–pathogen interaction and virulence like hlgA, empbp, hla, essD, esxA, rot, sbi, saeP, saeR, sak, spa, and fibrinogen binding protein showed significant downregulation in the mutants studied. The azoR gene responsible for providing resistance to thiol-specific stress was upregulated in mazE mutants, suggesting the role of mazF in the thiol stress response system [44]. Thus, a mutation in the mazE gene caused a change in the expression patterns of a plethora of genes that take part in various molecular and cellular pathways in the clinically isolated ST239 MRSA strain.

Fig. 2
figure 2

Differentially expressed genes in the P-1780 MRSA mazE mutant strain. a Volcano plot showing the differentially expressed genes in case of wildtype vs mazE mutant, after transcriptome analysis. b Heat map representation of some of the important genes, showing significant (p-value < = 0.05) differential expression. c qPCR analysis to check the differentially expressed genes in P-1780. Relative gene expression of different genes after mazE and mazF gene mutation in the MRSA strain. Error bars represent the standard deviation. Stars (*) above the error bars indicate the genes which are significantly differentially expressed (0.5 ≥ Fold change ≥2)

Differentially expressed genes from qPCR analysis

It has already been reported that mazF selectively cleaves mRNAs in vivo, avoiding essential transcripts like recA, gyrB and sarA in mazF induced cells [45]. So, we performed the qPCR analysis of the sarA, saeP, atl, hlgA, hla, cidA, lrgB mecA, femA and vraS gene using gyrB as the reference gene. From the qPCR analysis (Fig. 2c), the sarA gene was found to be 4.88-fold upregulated in the mazE mutant, compared to wildtype strain. No significant change in sarA expression was observed in the mazF mutant strain. The saeP gene was found to be downregulated in both the mazE and mazF mutant strains. The change in the atl (autolysin) gene in both the mutant strain was not significant. The mecA gene responsible for cefoxitin/oxacillin resistance was upregulated around eightfold and tenfold in the case of mazE and mazF mutants respectively. For the fem gene, no significant change was observed for both the mutant strains. The hlgA and the hla gene were found to be downregulated in the case of both the mazE and mazF mutants. The cidA and vraS genes were seen to be upregulated 10.9-fold and 12.3-fold respectively in the mazE mutant strain whereas no significant change in expression was observed for the mazF mutant strain. Also, mazE mutant revealed significant downregulation of the lrgB gene with no expression change in case of mazF mutant. The validity of the qPCR data was confirmed by melting curve analysis of the above mentioned genes (Additional file 1: Fig. S3). Thus, the qPCR analysis clearly highlights the change in gene expression caused by mutation of the mazE and mazF gene.

The mazF induced stressful condition deregulates cidA, leading to cell lysis promoting biofilm formation in MRSA P-1780

Biofilm formation assay revealed that both the wildtype and the generated mutants were successfully able to form biofilm. However, as shown in Fig. 3a and b, the biofilm formation was increased in the mazE mutant strain (Absorbance570/600 3.62, p-value < 0.01, t-test). On the other hand, the mazF mutant showed decreased biofilm formation (Absorbance570/600 2.71, p-value > 0.05, t-test). The biofilm formation in the mazE complement (mazE mutant/pCL55 carrying mazE gene) and the mazF complement strain (mazF mutant/pCL55 carrying mazF gene) were found to be similar to the wildtype (Absorbance570/600 2.88). These results shows that mazF toxin enhances biofilm formation in the P-1780 strain. To evaluate the importance of cidA mediated cell lysis for the release of eDNA, the biofilms of wildtype, mazE and mazF mutants along with the complement strains were treated with DNaseI at the time of inoculation and was further grown for 24 h. DNaseI treatment tremendously decreased mazE mutant biofilm compared to untreated biofilm (p-value < 0.001, Students t-test) (Fig. 3b). This observation pointed to the fact that a higher percentage of the cells involved in the biofilm formation in mazE mutant was dead and lysed, which might be due to the activated MazF, which leads to upregulated cidA and downregulated lrgB (as obtained from the transcriptomics data). To ascertain the fact that the increased biofilm formation in mazE mutant is due to upregulated cidA, we tried to overexpress lrgB gene (inhibitor of cidA) in the mazE mutant background and checked the biofilm formation [34]. The biofilm formation was significantly reduced (Absorbance570/600 2.5) in the lrgB overexpressing mazE mutant strain compared to only mazE mutant (Fig. 3b). However, DNaseI treatment did not affect the biofilm formation in the lrgB overexpressing mazE mutant strain. To further examine whether the biofilm formation by the mazE and mazF mutants correlates with cell death, we measured the dead cells in the biofilm by propidium iodide (PI) staining (Fig. 3c). In the wildtype MRSA strain, the percentage of dead cells was found to be 50.7%. However, in the case of the mutant strains, the percentage of dead cells was 69.4% (mazE mutant) and 11.6% (mazF mutant), respectively (Fig. 3d). FACS analysis also showed a similar pattern where the percentage of dead cells is higher in mazE mutant strain (26.2%) when compared to mazF mutant (7.9%) (Fig. 3e–h). Thus it can be hypothesized that mazF toxin works synergistically with cidA in regulating cell death and lysis mediated biofilm formation.

Fig. 3
figure 3

Cell death-induced biofilm formation by the different test strains. a Microtitre-well showing biofilm formation in the wildtype along with both the mutants, complement strains and the lrgB overexpressing mazE mutant after crystal violet staining: with untreated and DNaseI treatment. b The mazE antitoxin mutant showed increased biofilm formation in case of untreated while, DNaseI treatment drastically reduced the biofilm formation in case of wildtype and mazE mutant strain. The graph depicts the correlation of the measured crystal violet absorbance of the attached cell (Absorbance 570 nm) to the planktonic cell growth (Absorbance 600 nm). Each point and the standard deviation are the measures of three independent samples per condition. (t-test: ‘*’ denotes p-value between 0.01–0.05: significant; ‘**’ denotes p-value between 0.001–0.01: highly significant; ‘n.s’ denotes not significant). c Confocal laser scanning microscopy of the three test strains obtained from the biofilm, after staining with 4 µM propidium iodide (PI). Dead cells appear as discrete red puncta when excited with a 522-nm argon-krypton laser and emission collected with a 580-nm to 630-nm bandpass filter. d Percentage of dead/total biomass were calculated for the MRSA cells from the biofilm sample. Each point and the standard deviation is the measure of three independent samples per condition. Flow cytometry analysis for e Control (unstained), f wildtype, g mazE mutant, h mazF mutant, were done with the following settings: threshold set to side scatter (SSC) and flow rate set to the lowest possible, 6 µl/min. All measurements were run for 10,000 events. Data were averaged over three identical experiments. The percentage of dead cells was analyzed using BD FACS Diva™ (Becton Dickinson)

MazF toxin changes MRSA P-1780 cell susceptibility towards β-lactams, glycopeptide vancomycin, and the lipopeptide daptomycin

From the VITEK2 antibiogram profiling, the susceptibility pattern of the five different test strains was checked in presence of different antibiotics. For the wildtype and the complement strains (the mazE mutant/pCL55 carrying mazE and the mazF mutant/pCL55 carrying mazF), the MICs for oxacillin, daptomycin, and vancomycin observed were 1 µg/ml, < = 0.5 µg/ml and < = 0.5 µg/ml respectively, but for mazE mutant, the MIC for oxacillin, daptomycin and vancomycin was increased to > = 4 µg/ml, > = 8 µg/ml, > = 32 µg/ml respectively, confirming the reduction in oxacillin, daptomycin and vancomycin susceptibility. But rest of the antibiotics including gentamicin, ciprofloxacin, levofloxacin, erythromycin, clindamycin, linezolid, teicoplanin, tetracycline, tigecycline, rifampicin, trimethoprim/sulphamethoxazole showed no significant changes in MIC values (Table 1). Similar results supporting antibiotic resistance were obtained from transcriptome analysis (RNA-seq) which showed significant downregulation in transcription of cell surface protein genes (spa), second immunoglobulin-binding protein (sbi), regulatory genes (saePRS), PP2C phosphatase gene with upregulation in multi-drug resistance protein, betaine aldehyde dehydrogenase (responsible for the accumulation of the compatible solute glycine, betaine) and ure genes of the urease operon. RNA-sequencing results showed characteristic expression profiles for vraS and vraR, which were up-regulated in the mazE mutants, confirming the changed phenotype of glycopeptide antibiotic resistance [46]. The penicillin-binding protein 2A (mecA), responsible for methicillin resistance was upregulated in the mazE mutant, suggesting a possible role of the mazF toxin in regulating resistance to beta-lactams or cephalosporin group of antibiotics. Thus, it can also be demonstrated that the mazF toxin is responsible for making the P-1780 strain resistant to beta-lactam, glycopeptides and lipopeptide group of antibiotics.

Table 1 Table showing different antibiotic response of the WT P1780 strain and the generated mazE mutant, mazF mutant and the complement strains: mazE mutant/pCL55 carrying mazE gene and mazF mutant/pCL55 carrying mazF gene

mazEF system affect the hemolysis activity of P1780

Spot plate hemolysis assay on blood agar plates revealed that the wildtype, mazE and mazF mutants, as well as both the complement strains, were capable of lysing red blood cells (discoloration of red color surrounding the colonies) (Fig. 4a). However, the beta-hemolysis shown by the mazE mutant strain and the mazF mutant strain was found to be less than that of the wildtype. Moreover, both the complement strain showed a similar hemolysis pattern as the wildtype. Since the spot plate assay was qualitative, we performed a quantitative blood hemolysis assay using a 3% human RBC solution. We observed different percentages of hemolytic activity for all the tested MRSA strains. The percentage hemolysis calculated for the wildtype, mazE mutant and mazF mutant strains was 58.32%, 32.76% and 35.51%, respectively (Fig. 4b). While the percentage hemolysis for the mazE mutant/pCL55 carrying mazE and the mazF mutant/pCL55 carrying mazF were observed as 54.02% and 56.23% respectively. From the graph, it is evident that there was a significant change in the percent hemolysis of the mazE and mazF mutants when compared to the wildtype. This observation from both experiments suggests that the mazEF toxin–antitoxin system affects the hemolysis activity of the MRSA P-1780 strain. This was also confirmed by the decreased expression of hla and hlgA gene responsible for hemolysis (Fig. 6, Additional file 1: Table S2) [38].

Fig. 4
figure 4

Beta hemolytic activity shown by the P-1780 MRSA strains. a Liquid cultures of wildtype, mazE mutant, and mazF mutant along with both the complement strains were spotted onto blood agar plates containing 5% sheep blood and incubated overnight at 37 °C. b Percentage hemolysis of red blood cells were calculated for the five test strains by measuring the absorbance at 540 nm. Error bars represent standard deviation. Each point and standard deviation are the measures of three independent samples per condition

Urease operon is positively upregulated by mazF toxin

From RNA-sequence analysis, it was seen that, unlike rot, the mazF toxin positively regulates various urease genes like ureB, and ureC, that encode different subunits of the urease enzyme, along with the accessory genes like ureD and ureE, essential for its enzymatic activity [43]. Therefore, we also tested the urease activity on urea agar slants for the wildtype, mazE mutant, and mazE mutant/pCL55 carrying mazE strains. As shown in Fig. 5a, the mazE mutant strain possessing the activated MazF toxin was urease positive within 24 h but the wildtype and the complement strain (mazE mutant/pCL55 carrying mazE) were urease negative when kept for 24 h. The mazF mutant was also found to be urease negative (data not shown).

Fig. 5
figure 5

Virulence activity shown by the P-1780 MRSA strains. a Overnight grown cultures of wildtype, mazE mutant and the mazE complement strains were streaked onto urea agar slant. The urease activity was monitored by the color change of the urea agar slant after 24 h of growth. bThe wildtype, mazE, and mazF mutant were administered into ten C57BL/6 mice via retro-orbital injection. Three days post-infection, the mice were euthanized, the kidneys and livers were harvested, and CFU in the organs was counted. The statistical significance was determined by the Mann–Whitney test. n.s.: not significant

The mazEF TA system is not involved in host–pathogen interaction mediated virulence

Earlier studies have identified EssD, a novel secreted effector involved in the ESS pathway, which functions as a DNA nuclease, whereby it stimulates interleukin (IL-12) signaling to confer pathogenicity in the S. aureus, within deep-seated abscess lesions [47]. On the other hand, in the ESAT-6-like system, S. aureus mutants that failed to secrete EsxA and EsxB exhibited pathogenesis defects in murine abscesses, signifying that this specialized secretion system might play an important role in human bacterial pathogenesis [48]. Both the essD and the esxA genes were found to be downregulated significantly in the mazE mutant strain. Since the mazE and mazF mutations affect the transcription of multiple genes (hlgA, empbp, hla, essD, esxA, rot, sbi, saeP, saeR, sak, spa, etc.) involved in staphylococcal pathogenesis, we examined whether those gene expression alterations, affect the virulence of S. aureus. The wildtype, mazE, and mazF mutant cells (~ 2 × 107 cfu) were injected into ten mice via retro-orbital route. Three days later, the mice were sacrificed, and bacterial loads in the kidneys and liver were measured. As shown in Fig. 5b no significant change was observed. The mazE and mazF mutants showed altered biofilm formation (Fig. 3a and b). To examine whether the alteration of biofilm formation affects the bacterial virulence in persistence phase, we administered the reduced number of the bacterial cells (~ 1 × 107 cfu) via a retro-orbital route and observed the infected mice for 14 days. As shown in Additional file 1: Fig. S4, all infected mice showed similar survival, suggesting that the gene expression alterations caused by mazEF mutations do not significantly affect the overall virulence of S. aureus.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.


This article is autogenerated using RSS feeds and has not been created or edited by OA JF.

Click here for Source link (