β-Hemolysin, not agrA mutation, inhibits the hemolysis of α-hemolysin in Staphylococcus aureus laboratory and clinical strains

ABSTRACT Staphylococcus aureus produces various hemolysins regulated by the Agr-QS system, except β-hemolysin encoded by the gene hlb. A classical laboratory S. aureus strain RN4220 displays only the β-hemolysin phenotype. It was suspected that the 8A mutation at the end of its agrA gene delayed the expressions of hla and RNAIII, then failed to express α- and δ-hemolysins. However, hla gene expression was detected at the later culture time without α-hemolysin phenotype, the reason for such a phenotype has not been clearly understood. We created hlb knockout and complementary mutants via homologous recombination in RN4220 and NRS049, two strains that normally produce β-hemolysin and carry agrA mutation. We found interestingly that the presence or absence of α-hemolysin phenotype in such strains depended on the expression of β-hemolysin instead of agrA mutations, which only inhibited δ-hemolysin expression. The hemolysis phenotype was verified by the Christie-Atkinson-Munch-Peterson (CAMP) test. Quantitative reverse transcription PCR was carried out to evaluate the relative gene expressions of hlb, hla, and RNAIII. The construction of mutants did not affect the agrA mutation status. We demonstrate that the absence of α-hemolysin in S. aureus RN4220 and NRS049 strains is attributed to their production of β-hemolysin instead of agrA mutation. Our findings broaden the understanding of the molecular mechanisms that control hemolysin expression in S. aureus that is crucial for the development of new therapeutic strategies to combat S. aureus infections. IMPORTANCE α-Hemolysin is a critical virulence factor in Staphylococcus aureus and its expression is largely controlled by the Agr-QS system. Nonetheless, the hemolysis phenotype and the regulation of the Agr-QS system in S. aureus still hold many mysteries. Our study finds that it is the expression of β- hemolysin rather than the agrA mutation that inhibits the function of the α-hemolysin in an important S. aureus strain RN4220 and a clinical strain presents a similar phenotype, which clarifies the misunderstood hemolytic phenotype and mechanism of S. aureus. Our findings highlight the interactions among different toxins and their biological roles, combined with QS system regulation, which is ultimately the true underlying cause of its virulence. This emphasizes the importance of considering the collaborative action of various factors in the infection process caused by this significant human pathogen.

by the staphylococcal Agr-QS system (4,5).The effector of this system is the RNAIII (6), which stimulates transcription of the downstream hla gene, encoding α-hemolysin, while the hld gene encoding δ-hemolysin is located within the RNAIII locus (7).The agr locus consists of two transcriptional units driven by P2 and P3, which are activated by the regulator AgrA (8).Differently, the expression of β-hemolysin in some S. aureus strains is dependent on the presence or absence of prophage insertion in its encoding gene hlb (9) and its regulation awaits further study (10).β-Hemolysin, also known as hot-cold hemolysin, has enhanced hemolytic activity below 10°C and appears to function as a sphingomyelinase (SMase) that binds to the sphingomyelin of endothelial cells and erythrocytes (11)(12)(13).We know relatively little about β-hemolysin compared to other hemolysins, which has promoted us to study the biological function of this staphylococ cal toxin.
We choose two S. aureus strains, RN4220 (a classic laboratory strain) and NRS049 (a clinical strain), both of which only produce β-hemolysin, and a α-and δ-hemolysin producer, the USA300 strain LAC (Fig. 1A and B).We analyzed the major hemolysin regulator encoding gene agrA in these strains and found an 8A mutation in RN4220 and a 9A mutation in NRS049 at the site of 713 (Fig. 1C).Previous reports have shown that mutations in agrA result in a late transcription of RNAIII and the absence of αand δ-hemolysins in S. aureus (14,15).However, it is not clear why the α-hemolysin phenotype was not observed even though the hla message was always detected about 1 h after the expression of RNAIII.The β-hemolysin phenotype was not seen in strain LAC due to a prophage insertion in its hlb gene that was visualized via Easyfig 2.2.5 (16) (Fig. 1D).When we knocked out the hlb gene in strains RN4220 and NRS049, we expected no hemolysis on the sheep blood agar plate (Fig. 2A).Surprisingly, as shown in Fig. 2B of the CAMP (Christie-Atkinson-Munch-Peterson) test ( 17), a strong α-hemo lytic halo was observed around all hlb knockout strains, including RN4220Δhlb and NRS049Δhlb.Furthermore, the α-hemolytic halo disappeared in the hlb complementary strain RN4220Δhlb/hlb::pTSSCm.As previously reported, the 8A mutation at the end of the agrA gene can cause a delay in RNAIII expression and thus may ultimately result in a failure of αand δ-hemolysin translations (14).To ascertain if the agrA gene was affected during our hlb knockout process, we amplified the agrA gene of all wild type, hlb mutants, and hlb complementary strains by PCR and verified it by Sanger sequencing.The results showed that 8A and 9A mutations of agrA continue to exist in all tested strains (Fig. 2C).We found that the presence of α-hemolytic halo depended on the lack of expression of β-hemolysin instead of the mutations that occurred at the end of agrA gene.To further confirm our findings, we also conducted quantitative reverse transcription PCR for hlb, RNAIII, and hla genes in all RN4220 derivatives and LAC at 2 h of incubation.According to the hlb expression pattern in RN4220 wild type, Fig. 2D demonstrates that RN4220Δhlb and LAC did not exhibit any detectable hlb activity, whereas LAC and RN4220Δhlb/hlb::pTSSCm displayed high levels of hlb expression.As expected, the expression level of RNAIII was found to be significantly reduced in strain RN4220, regardless of the presence of the hlb gene, while the expression of hla remained unaffected.This was observed by comparing the expression levels in strain RN4220 to those in LAC.Demonstrating the mutations at the end of gene agrA resulted in a loss of δ-hemolysin gene expression, while α-hemolysin expression remained unaffected in all tested strains.
In summary, S. aureus strains with the 8A or 9A mutation at the end of the agrA gene fail to express the δ-hemolysin (RNAIII) but not α-hemolysin.The inhibition of α-hemolysin phenotype in such strains is due to the expression of β-hemolysin, which is an Mg 2+ -dependent neutral sphingomyelinase that hydrolyzes sphingolipids into phosphatidylcholine and ceramide (12).Since S. aureus α-hemolysin is a cell membrane pore-forming toxin, causing an outflow of intracellular material and leads to cell death (18), the binding of α-hemolysin to the human cell membrane is the first step of this process, which significantly relies on the presence of sphingolipids (19).Removal of sphingomyelin from the human airway epithelial cells plasma membrane will completely abrogate the damaging effect of α-hemolysin.Therefore, when β-hemolysin is present and consumes sphingolipids, α-hemolysin lacks the binding targets and hence fails to form pores on the erythrocyte membrane.This explains that the presence of α-hemolytic halo is rarely observed in strains that produce β-hemolysin, regardless of their agrA mutation status, unless the expression of α-hemolysin is sufficiently high (20).
The mystery of why some β-hemolysin-producing S. aureus strains (RN4220) do not show α-hemolysin phenotype despite having a functional hla gene has puzzled researchers for some time.This study has solved this mystery by demonstrating that the presence of β-hemolysin inhibits α-hemolysin phenotype activity regardless of the strains' agrA mutation status and results in a phenotype with only β-hemolysin phenotype when the expression of δ-hemolysin is inhibited in S. aureus.Our find ings broaden the understanding of the molecular mechanisms that control hemoly sin expression in S. aureus, which is crucial for the development of new therapeutic strategies to combat S. aureus infections.

Mutant construction
Mutants were constructed as described previously with slight modification (21).Briefly, allelic substitutions were made by amplifying approximately 1 kb regions upstream and downstream of the target gene hlb and cloning them into plasmid pKOR1.The plasmid was first transformed into S. aureus RN4220 and then into NRS049.To integrate the plasmids, RN4220 (pKOR1-hlb) and NRS049 (pKOR1-hlb) cultures were grown in TSB containing 10 µg/mL chloramphenicol (TSB Cm10 ) at 30°C and were transferred to fresh TSB Cm10 at 43°C overnight.The deletion mutants were verified using PCR and Sanger sequencing.For the complementary strain construction, the wild-type and mutant derivatives of hlb were cloned into pTSSCm and transferred into strain RN4220Δhlb.Briefly, the hlb gene was amplified using PCR, cloned into the SpeIand NdeI-digested plasmid pTSSCm, and transformed into S. aureus RN4220, which was then transformed into strain RN4220Δhlb by electroporation.These plasmids conferred tetracycline resistance, and tetracycline was added to growth cultures at a concentration of 12.5 µg/mL.The parental Escherichia coli DH5α and RN4220Δhlb strains carrying the empty plasmid pTSSCm were used as controls.The hlb complementary strain was only constructed for strain RN4220, due to NRS049 being a multi-drug resistant strain that lacks selection marker for the hlb complementary screening.The strains and plasmids used in this study are listed in Table 1 and the primers used are listed in Table S1.

CAMP test
According to the interaction between the hemolysins of S. aureus that β-hemolysin inhibits the lysis of α-hemolysin but enhances the lysis of δ-hemolysin (8,26), we performed a CAMP test (17) to confirm the hemolysin phenotype of each S. aureus strain.
For instance, we determined all strain hemolysin types by cross-hatching them vertically on sheep blood agar plates with RN4220 (14).The degree of hemolysis at the cross indicates phenotypes of different hemolysins: the same as for RN4220 is determined as β-hemolysin, the attenuated hemolysis is determined as α-hemolysin, and the enhanced hemolysis is determined as δ-hemolysin.

RT-qPCR
Cultures of S. aureus strains LAC, RN4220, RN4220Δhlb, and RN4220Δhlb/hlb::pTSSCm were grown to a density of 50 Klett units (1.5 × 10 8 cells/mL), diluted 10-fold times, and grown again to 50 Klett units.After 2 h of culture, samples were harvested for total RNA extraction using the E.Z.N.A.Total RNA Kit (OMEGA, Switzerland) according to the manufacturer's protocol.The quality of extracted RNA was assessed using a NanoDrop spectrophotometer (Thermo Fisher Scientific, USA).Thereafter, the total RNA was reverse-transcribed to cDNA using the PrimeScript RT reagent Kit (TaKaRa, Japan), and RT-qPCR was then carried out using the TB Green Fast qPCR Mix (TaKaRa, Japan) via Roche LightCycler 480 Real-Time PCR System.Primers used for RT-qPCR assays are listed in Table S1.The relative quantitation of mRNA expression was normalized to the constitutive expression of the housekeeping gyrB and rho genes and calculated by the comparative CT (2 −ΔΔCT ) method.A fold change over 4 was considered a significant increase or decrease.

Statistical analysis
For the RT-qPCR test results, the relative expression of each tested gene that was decreased or increased over fourfold would be considered as biologically significant down-and upregulation.Meanwhile, we also conducted one-way ANOVA tests to compare the differences between groups for each gene expression experiment, where P < 0.0001 was considered statistically significant.All analyses were performed using GraphPad Prism V9.5.0.All experiments were repeated at least three times.

FIG 1
FIG 1 Hemolysis phenotype of S. aureus strains.(A) Hemolysis phenotype of tested laboratory strain RN4220 and clinical strain NRS049 on sheep blood agar plate under 37°C and 4°C.(B) The CAMP tests for RN4220, NRS049, and LAC strain.Strains were tested against RN4220 to evaluate their hemolytic activities: β-hemolysin phenotype occurred in RN4220 and NRS049, α-hemolysin (yellow arrows marked decreased hemolysis when crossing the β-hemolysin from underlined RN4220) and δ-hemolysin (yellow arrows marked enlarged hemolysis when crossing the β-hemolysin) phenotypes occurred in LAC strain.(C) The sequences of gene agrA in LAC, RN4220, and NRS049 were compared via https://www.ebi.ac.uk/Tools/msa/clustalo/. Different from the normal agrA gene in LAC, an 8A and a 9A mutation at the site of 713 in RN4220 and NRS049 was observed, respectively.(D) β-Hemolysin encoding gene hlb was compared between LAC and RN4220/RNS049 using Easyfig 2.2.5 (https://mjsull.github.io/Easyfig/)and depicted a prophage insertion in the hlb gene of LAC that causes its failure to produce β-hemolysin.

FIG 2 β
FIG 2 β-Hemolysin inhibits α-hemolysin phenotype, rather than agrA 8A/9A mutations, in S. aureus that express both.(A) Schematic representation of the hlb gene deletion and gel electrophoresis confirmation of hlb knockout and complemen tary construction.(B) The CAMP test results presentation.Strains were tested against RN4220 to evaluate their hemolytic activities: α-hemolysin phenotype occurred in RN4220Δhlb and NRS049Δhlb and disappeared in RN4220Δhlb/hlb::pTSSCm. (C) Sanger sequencing results of the agrA gene, where the 8A/9A mutation was confirmed to continue to exist in all mutants.(D) The relative expression levels of hlb, RNAIII, and hla in LAC and RN4220 derivatives.The red dashed line indicates the cutoff of a fourfold change in each panel, which was considered a significant increase or decrease."****" represents P < 0.0001 for one-way analysis of variance (ANOVA) tests.The comparison was conducted against the gene expression level in RN4220 for hlb and in LAC for RNAIII and hla.