Deficiency of dltD contributes to increased autolysis in Streptococcus mutans

ABSTRACT

Streptococcus mutans (S. mutans) is one of the major etiological agents of dental caries. Interfering with the expression of genes associated with cariogenicity of S. mutans is an effective ecological approach for caries prevention. Thus, understanding the role of genes related to cariogenicity in S. mutans is important. dltD is a conserved gene in Gram-positive bacteria, which is involved in D-alanylation of lipoteichoic acid to modify cell surface charge. In vivo caries experiments have shown significant decreases in the caries severity in rats infected with S. mutans dltD-deficient strain. Although the impact of dltD on bacterial surface charge and division in other Gram-positive bacteria is known, its role in S. mutans biological functions remains unclear. This study aimed to investigate the biological functions of dltD in S. mutans. We examined growth curves, morphology, autolysis, biofilm formation, and cell membrane properties of S. mutans wild strain and dltD-deficient strain. Our results revealed that dltD plays a key role in the growth and autolysis of S. mutans. dltD deletion increased biofilm autolysis, thereby reducing the number of viable bacteria within the biofilm. Furthermore, dltD may have an epistatic effect on cell division and surface charge. Its regulatory network needs to be further elucidated. This study shed light on therole of dltD in the growth of S. mutans, thereby enhancing our understanding of its function.

KEYWORDS:

• streptococcus mutans
• autolysis
• dltd
• membrane potential

Introduction

Dental caries is a common disease of the oral cavity. It is a biofilm-mediated, multifactorial, noncommunicable dynamic disease resulting in the demineralization of the hard tissues of the teeth [1]. Streptococcus mutans (S. mutans) is one of the causal agents in dental caries [2]. Currently, interfering with the expression of genes associated with cariogenicity of S. mutans is an effective ecological approach for caries prevention [3].

The gene dltD, which is conserved within Gram-positive bacteria, is a part of the dlt operon in S. mutans. This operon encompasses four genes, dltA, dltB, dltC, and dltD. DltA appends phosphorylated alanine residues to the carrier protein DltC. DltB subsequently transfers D-alanine to the exterior of the cell membrane [4]. dltD, encoding a single-pass membrane protein in Staphylococcus aureus, employs a catalytic dyad to convert D-alanine into lipoteichoic acid (LTA) (Figure 1) [5]. In other Gram-positive bacteria, dltD deficiency results in aberrant bacterial division and autolysis [6], thereby decreasing pathogenicity. However, the exact role of dltD in S. mutans remains to be elucidated.

Figure 1. Graphical schema of the S. mutans dlt operon. Diagram of the D-alanylation of LTA.

The cariogenicity of S. mutans is driven by biofilm-formation capacity, acid production, and survival capability in intricate and dynamic environments [2]. Biofilms serve as a protective shield against mechanical and external stresses. Laser scanning confocal microscopy reveals a visible decrease in the number of microcolonies in the biofilm of S. mutans UA159 dltD-deficient strain strain [7]. Furthermore, in vivo caries experiments demonstrated a significant reduction in the caries severity in rats infected with S. mutans UA159 dltD-deficient strain [8].

We isolated S. mutans clinical strain 593 (S. mutans 593) from patients with severe caries and S. mutans clinical strain 18 (S. mutans 18) from caries-free patients. In vitro, the S. mutans 593 biofilm displayed a greater extracellular matrix than the S. mutans 18 biofilm. Moreover, S. mutans 593 exhibited higher survival rates under lethal acidic conditions [9]. Importantly, dltD expression was upregulated in S. mutans 593. which led us to construct an S. mutans 593 dltD-deficient strain. The results showed that S. mutans 593 dltD defects lead to decreased acid tolerance [10]. These findings suggest that dltD is a potential gene target for caries prevention, warranting a comprehensive exploration of its role in S. mutans.

The present study aimed to elucidate dltD role in biological functions of S. mutans. We selected S. mutans 593 wild strain and dltD deletion strains to explore the role of dltD in S. mutans. Our results suggested that dltD played a significant role in S. mutans growth. dltD deletion increased autolysis and decreased the number of viable bacteria in the biofilm.

Materials and methods

Bacterial strains and culture conditions

The bacterial strains and plasmids used in this study are listed in Table 1. S. mutans 593 were routinely grown in the Brain Heart Infusion Broth (BHI) broth or BHI broth containing 1% (W/V) sucrose (Difco, Sparks, MD, USA) at 37 °C under anaerobic (90% N₂, 5% CO₂, and 5% H₂) or aerobic conditions (5% CO₂). Escherichia coli (E. coli) was grown aerobically in Luria – Bertani medium (Oxoid Ltd.) at 37 °C. Spectinomycin (Sigma Aldrich, USA) was added at a concentration of 1 mg/mL for the S. mutans dltD deletion mutant and 100 μg/mL for E. coli when required. Chloromycetin (Solarbio, China) was added at a concentration of 20 μg/mL for the S. mutans dltD complemented strain and E. coli when required.

Table 1. List of strains and plasmids used in this study (Sp^r: spectinomycin resistance; CHL^r: chloromycetin resistance).

Strain or plasmid	Description	Source
Strains
S. mutans 593	Wild type	Laboratory preservation
S. mutans 593-∆dltD	S. mutans 593 dltD deletion mutant; Sp^r	Previously constructed
S. mutans 593-∆dltD-comp	Complemented strain; CHL^r	This study
Escherichia coli DH5α	For plasmid multiplication; Sp^r	Purchased from NEB
Plasmids
pFW5	E. coli-Streptococcus shuttle expression vector; Sp^r	Purchased from http://www.miaolingbio.com
pFW5-dltD	for dltD knockout; Sp^r	Previously constructed
pIB169	E. coli-Streptococcus shuttle expression vector; CHL^r	Presented by Professor Huang
pIB169-dltD	for dltD complemented; CHL^r	This study

Construction of ∆dltD-comp strains

The complementary strains were constructed as previously described [11]. pIB169 [12-complemented strain (Δ-comp). S. mutans dltD fragment was inserted into pIB169 EcoR I/BamH I (Sangon Biotech, China). The modified pIB169-dltD was transformed into S. mutans 593-∆dltD (previously constructed[10]). Positive clones of dltD complemented strain were screened using plates containing chloromycetin. The complementation of dltD was screened by PCR and further confirmed by RT-qPCR.

Growth curve assay

The overnight bacterial culture was subcultured at 1:20 in fresh BHI broth. Then, 200 µL of the bacterial suspension was added to each well of the Bioscreen C plate, with biological triplicates set up for each strain. The growth curves were measured with a Bioscreen C lab system (Helsinki, Finland). The parameters were set as follows: shaking for 10 s before reading, 30 min intervals, wavelength of 600 nm, and temperature of 37 °C. The growth curves were plotted using GraphPad Prism 8.0.

Gram stain

Smear fixation of the bacterium was performed during the exponential growth period. Staining with crystal violet (CV) and staining with iodine solution were conducted separately for 1 min each. Decolorization was performed using 95% alcohol. Then, staining with a tomato-red beam was conducted for 2 min. Water was used for rinsing before each addition of dye. The glass slide was dried and observed under a microscope.

Transmission electron microscopy (TEM)

TEM observation was performed as previously described [13]. The mid-log phase cultures were fixed with 2.5% glutaraldehyde for 4 h and washed three times before being resuspended in 1% agarose. Then, the sample was then washed and fixed in 1% osmium tetroxide in SC-Mg buffer for 2 h. The agarose cell block was dehydrated, embedded, and sliced. The sections were stained with saturated ethanolic uranyl acetate followed by 0.25% aqueous lead citrate in 0.1 M NaOH. They were examined using a Hitachi HT-7700 microscope.

Triton x-100-induced autolysis

Triton x-100-induced autolysis was performed according to Nakao A. et al [14]. The mid-log phase cultures were centrifuged to remove the medium. After washing with PBS, the pallets were resuspended in 50 mM Tris-HCl (pH7.4 containing 0.05% Triton X-100) buffer and adjusted to OD_600 nm ≈1.2. Afterward, 200 µL of the bacterial suspension was added to each well of the Bioscreen C-specific well plate, with biological triplicates set up for each strain. The autolysis curves were measured using a Bioscreen C system (Helsinki, Finland). The parameters were set as follows: shaking for 10 s before reading, 30 min intervals, wavelength of 600 nm, and temperature of 37 °C. The growth curves were plotted using GraphPad Prism 8.0.

CV assay and Colony-forming units (CFUs)

The biofilm was grown on 24-well plates containing 1 mL BHI broth with 1% sucrose. All wells were inoculated with 1/20 dilution of mid-log phase cultures suspension (OD_600nm ≈ 0.2) and anaerobically incubated for 24 h at 37 °C. The boifilms were washed with PBS, stained with 1% CV for 15 min, and decolorized with absolute ethanol. Quantification was conducted by measuring the OD_570 nm of the solution. For CFU counting, The biofilm were rinsed with PBS, and 1 mL of PBS was added per well to resuspend biofilm. They were diluted 10¹–10¹⁰ times, and each concentration of the bacterial solution was plated on BHI agar plate with 100 μL . CFUs were counted after 24 h culture under anaerobic conditions.

eDNA assay

eDNA test method reference Liao. et al [15]. The biofilms were rinsed with PBS, and 1 mL of PBS was added per well to collect biofilm. The collected biofilm suspension was centrifugated at 4000 rpm for 20 min. The supernatant was filtered through a 0.22 μm pore size filter membrane. We incubated 50 µL of supernatant with 50 µL SYTOX Green for 30 min in the dark. After incubation, the fluorescence intensity was measured by a microplate reader at λex = 488 nm and λem = 538 nm.

Scanning electron microscope (SE)

The 24 h biofilm was formed on at the bottom of a 24-well plate. After being rinsed with PBS, the biofilm was fixed with 2.5% glutaraldehyde. Serial dehydration included preparations with ethanol solutions (30%, 50%, 70%, 80%, 90% and 100%) every time stood for 15 min. The dehydrated biofilm samples were dried and coated with gold powder. Then, the biofilm samples were evaluated using an SEM (KYKY-EM8000, China).

Surface charge assay

Surface charge assay was performed according to Buchanan J. et al [16The mid-log phase cultures were centrifuged to remove the medium. After washing with PBS, the pallets were resuspended in 1×MOPS (Solarbio, China) buffer and adjusted to OD_600 nm ≈1.0. Cytochrome C (Macklin, China) was added at a final concentration of 0.5 mg/mL. Then, the samples were incubated for 15 min at room temperature. The measured absorbance was 530 nm. A higher the amount of the detected unbound cytochrome c in the supernatant corresponded with a greater net positive charge of the bacterial surface.

Membrane potential assay

Membrane potential assay was performed according to Lacombe A. et al [17]. The mid-log phase cultures were centrifuged to remove the medium. After washingwith deionized water, the pallets were resuspended in deionized water. Then, 50 µL of cell suspension of each strain was diluted for CFU counting. Another 50 µL was added to each well with 50 µL of 5 µM DiBAC₄(3) (MedChemExpress, China) and incubated in the dark for 40 min. After incubation, the fluorescence intensity was measured with a microplate reader at λex= 493 nm and λem= 516 nm. The final result was expressed as the fluorescence intensity divided by the CFU: fluorescence intensity/CFU=I₅₁₆/CFU.

Intracellular potassium ions concentration

Intracellular potassium ion concentration was performed according to Xing Y. et al [18]. The method was modified to fit this experiment. PBFI (MKbio, China) is a potassium-sensitive fluorescent probe used to mesure changes in potassium levels in cells and intracellular compartments. The mid-log phase cultures were centrifuged to remove the medium. After being washed with Hepes buffer (containing 5 mmol/L glucose), the pellets were resuspended. Then, 50 µL cell suspension of each strain was diluted for CFU counting. Another 50 µL was added to each well with 50 µL of 5 µM PBFI and incubated in the dark for 40 min at 37 °C. After incubation, the fluorescence intensity was measured with a microplate reader at λex = 380 nm and λem = 500 nm. The final result was expressed as the fluorescence intensity divided by the CFU: fluorescence intensity/CFU=I₅₀₀/CFU.

Laurdan stain

Membrane fluidity was performed according to Scheinpflug. et al [19]. Laurdan (MedChemExpress, China) is a membrane-permeable fluorescent probe that is spectrally sensitive to the phospholipid phase of the cell membrane to which it is bound. The generalized polarization (GP) function quantification of Laurdan can be used for identifying phospholipid phases [20]. The mid-log phase cultures were centrifuged to remove the medium. After washing with deionized water, the pallets were resuspended in 200 μL of 10 μM laurdan and incubated for 1.5 h at 37 °C. After incubation, the fluorescence intensity was measured with a microplate reader at λex = 350 nm and λem = 440/490 nm. Equation 1 was used to calculate GP and compare membrane fluidity.

(1) $GP=\left(I_{440}-I_{490}\right)/\hspace{0.17em}\left(I_{440}+I_{490}\right)$(1)

RNA isolation and reverse-transcription quantitative PCR

RNA isolation was performed according to a previous study. The mid-log phase cultures or 24 h biofilm samples were collected for RNA isolation. Total RNA was extracted according to the instructions (Qiagen, Valencia). DNA contamination was removed and cDNA was obtained using a PrimeScript^TM RT kit (Perfect Real Time) from Takara Bio (Otsu, Japan) with a gDNA eraser. The extracted cDNA was used as a template. Quantitative real-time PCR was performed using SYBR® Premix Ex Taq^TM (Tli RNA seh Plus) using a LightCycler 480 quantitative PCR instrument in a 96-well real-time PCR plate from Sangon Biotech (Shanghai, China). The result was calculated by normalizing the target genes to the respective reference genes through the 2^−∆∆Ct method. The primers are listed in Table 2.

Table 2. Nucleotide sequences of primers were used in this study.

Primer	Sequence(5’-3’)	Reference
16S rRNA	F: AGCGTTGTCCGGATTTATTG	This study
	R: CTACGCATTTCACCGCTACA
atlE	F: AGCTGGTCCCAAAGGAAATC	[21]
	R: GCCTGTGCCCAATAATCATC
fabM	F: TAATGGTGTGGCGACTTTAAC	This study
	R: AACAGCTCTTTGCATCTCAACC
dltD	F: GGCAAGCCTGCGCCTTATTAAGAG	This study
	R: ACATGCGGCAATACACGACGTTC

Statistical analysis

All values are expressed as the mean ± standard error for three biological replicates (n = 3) assayed in triplicate. Data statistics were analyzed using GraphPad Prism 8.0 one-way analysis of variance (ANOVA), followed by Tukey’s multiple comparison tests. Two-sided p-values less than 0.05 were considered statistically significant.

Results

dltD altered growth and had epistatic effects on division

To detect the impact of dltD deletion on the growth curve of S. mutans, growth curves were mapped. Compared with the wild-type strain, dltD deletion strain entered the exponential growth phase approximately 1 h later (Figure 2A). Gram staining demonstrated a marked increase in chain length in dltD deletion strain (Figure 2B), implying potential effects on S. mutans growth and division. Accordingly TEM was used to compare the cell morphology of the mutant and the wild type. TEM revealed increased septation, compromised cell wall integrity, and cytoplasmic leakage in dltD deletion strain (Figure 3). The growth curves and chain lengths of the complemented strain were comparable to those of the wild-type strain (Figure 2). However, the complemented strain exhibited TEM images comparable to those of dltD deletion strain (Figure 3).

Figure 2. (a) growth curve. (b) Gram stain.

Figure 3. Transmission electron microscopy. The red arrows point to increased divisions and the yellow arrows point to damaged cells.

dltD had effects on autolysis

Given that abnormal bacterial division often results from enhanced autolysis [7], and considering the TEM results, we hypothesized that bacterial phenotypic alterations and death in dltD deletion strains were due to autolysis. Triton X-100 is a surfactant that can be used to induce autolysis. TritonX-100-induced autolysis showed increased levels of autolysis in dltD deletion strains compared with the wild-type (Figure 4). The autolysis curves of S. mutans 593 complemented strain were comparable to those of the wild-type strain (Figure 4).

Figure 4. TritonX-100-induced autolysis.

dltD deletion increased biofilm autolysis and altered biofilm formation

Bacterial autolysis can release substantial amounts of eDNA, which is a crucial component of the extracellular matrix of bacterial biofilms [22]. SYTOX Green is a DNA dye that fluoresces and binds to bacterial eDNA. The fluorescence intensity of SYTOX Green is directly proportional to the eDNA content. eDNA can influence biofilm structure by interacting with extracellular polysaccharides and the cell wall [23]. Therefore, we compared the eDNA content in biofilms of the wild type and the mutant strain. The biomasses of biofilms were determined by CV assay. The CV assay demonstrated an increase in the total biofilm biomass in dltD-deficient strain (Figure 5A). Conversely, CFU counts showed a significant decrease in viable bacteria within the biofilm (Figure 5B). dltD mutant strain had a significantly increased eDNA content (Figure 6A) and upregulated autolysin gene atlE expression levels (Figure 6B). SEM revealed the biofilm of dltD-deficient strain became sparse with more pores than in the wild-type strain (Figure 7). The complementary strain exhibited a recovery trend.

Figure 5. (a) 24 h biofilm crystal violet staining. (b) 24 h biofilm colony forming unit. The data calculated from the three samples per group are expressed as the mean ± SE * p < 0.05,** p < 0.01.

Figure 6. (a) 24 h biofilm eDNA. (b) 24 h biofilm RT-qPCR analysis showed the gene related to autolysis. The data calculated from the three samples per group are expressed as the mean ± SE * p < 0.05,** p < 0.01.

Figure 7. Scanning electron microscope observed biofilm morphology.

dltD had epistatic effects on membrane properties

Autolysis is associated with cell wall turnover [24]. In Gram-positive bacteria, D-alanylation, a process that adds positively charged groups to the cell wall, plays a crucial role in membrane homeostasis [4]. Membrane potential, which is primarily maintained by ionic gradients and cell wall surface modification, is pivotal for maintaining bacterial morphology and division [25]. Changes in bacterial cell membrane fluidity, which is a critical factor in maintaining morphology, can influence bacterial division [26]. Thus, we explored the effects of dltD on the physiological properties of the cell wall. Results indicated that dltD deletion led to an increase in the negative cell surface charge (Figure 8A).

Figure 8. (a) relative positive surface charge by cytochrome c binding. (b) cell membrane depolarization. The graph shows intracellular DiBAC₄(3) fluorescence intensity. (c) potassium ion leakage. The graph shows the fluorescence intensity of the intracellular PBFI bound to potassium ions. The data calculated from the three samples per group are expressed as the mean ± SE * p < 0.05,** p < 0.01.

We assessed the membrane potential by DiBAC₄(3), a potential-sensitive fluorescent probe that enters depolarized cells and binds to intracellular proteins or membranes, resulting in enhanced fluorescence. Results showed increased levels of membrane depolarization in dltD deletion strain (Figure 8B). Given that the gradient between sodium and potassium ions is essential for maintaining the cell membrane potential [25], we used PBFI, a potassium-sensitive fluorescent probe, to measure changes in potassium levels. dltD mutant exhibited an increase in potassium, which contributed to cell membrane depolarization (Figure 8C).

We evaluated cell membrane fluidity by using Laurdan dye, a membrane-permeable fluorescent probe that is spectrally sensitive to the phospholipid phase of the cell membrane. Quantification of the generalized Generalized polarization (GP) function facilitated the identification of phospholipid phases [20]. dltD mutants exhibited increased cell membrane fluidity (Figure 9A).

Figure 9. (a) Laurdan staining followed by the calculation of GP to represent cell membrane fluidity. (b) RT-qPCR analysis showed the gene related to cell membrane fluidity. The data calculated from the three samples per group are expressed as the mean ± SE * p < 0.05,** p < 0.01.

One factor contributing to increased cell membrane fluidity is the increase in unsaturated fatty acids, which is regulated by S. mutans fabM. fabM upregulation was observed in dltD mutants (Figure 9B), as indicated by RT‒qPCR. This finding indicated that the observed increase in membrane fluidity may be related to the increase in unsaturated fatty acids.

In the complemented strain, the positive charge on the surface showed a recovery trend (Figure 8A). Membrane potential, potassium ion leakage, and cell membrane fluidity in the S. mutans 593 complemented strain showed a recovery trend (Figures 8B and C), and cell membrane fluidity was restored (Figure 9A). The expression levels of fabM exhibited a recovery trend in the S. mutans 593 complemented strain (Figure 9B).

Discussion

In the present study, we investigated the role of dltD in the growth, division, and biofilm formation of S.mutans. Our findings can lay a foundation for further exploring of the regulatory mechanisms of dltD in this S.mutans.

Our results showed that the deletion of dltD led to delayed growth in S. mutans, but the deletion strain still had a growth ability. Recent research revealed that bacterial cells transition to exponential growth after immediate division. The time to enter the logarithmic phase at the single-cell level is related to cell size [27The division of dltD deletion strain was altered, potentially contributing to increased chain length and slowed growth of the mutant. Previous reports in other Gram-positive bacteria have associated dltD with abnormal division due to heightened autolysis [6]. Notably, autolysis increases in Lactococcus lactis dltD-deficient strain, and the binding mode of the autolysin to peptidoglycan is unaltered [28dltD mutant decreases, which may lead to increased autolysin binding and abnormal division. Our data suggested a similar trend in dltD mutants, wherein dltD deletion increased autolysis. Although autolysis was restored in the complemented strain, it did not restore the division of dltD deletion strain. exhibits an epistatic (i.e., the masking of the allelic effect at one gene by the occurrence of other genes) [29mutans are affected by complex factors. In this study, we investigated the influence of cell membrane properties on division and surface charge.

S. mutans division initiates from the equatorial ring, where MapZ guides FtsZ proteins to form a Z-ring, anchoring autolysin onto the equatorial ring and cell poles. The cell expands to both sides, forming a septum in the middle, after which MapZ moves to form a new Z ring near the equatorial ring of the daughter cell. Finally, the two daughter cells are split using autolysin. S. mutans cell wall lacking Serotype c Carbohydrate (SCC) modification reported shows serious cell morphological defects and early autolysis [30]. One reason may be that lacking the carbohydrate modification reduces the negative charge in S. mutans cell wall may change the peptidoglycan density and structure, serving as a cue for MapZ recruitment. Conversely, autolysin is recruited to promote cell separation [31]. In the current work, we also observed the changes in the surface charge of dltD mutant. Abnormal divisions in dltD mutants may be driven by alterations in the peptidoglycan structure, whereas autolysin recruitment may contribute to cell separation.

In many Streptococci, the abnormal division is also linked to cell wall modification. dltD plays a role in the positive charge modification of cell walls. The membrane potential of gram-positive bacteria is largely maintained by ionic gradients and cell wall surface modifications, which affect the spatial organization of the cytoskeleton and cell division proteins. Membrane potential influences the interactions between the C-terminal amphipathic helix and lipid bilayer, thereby influencing the localization of cell division proteins [25]. Cell membrane fluidity also affects the binding of bacterial splitting proteins to the cell membrane. In Escherichia coli and Bacillus subtilis, the polarity marker protein DivIVA, a multifunctional scaffolding protein, directly binds to the cytoplasmic membrane through its N-terminal domain, recruiting cell division proteins to affect bacterial division depending on the cellular environment and species [25]. In S. mutans, DivIB plays a role similar to that of DivIVA in cell division regulation. Excessive cell membrane fluidity may potentially impede the combination of the N-terminal domain to DivIB [32], resulting in adverse effects on cell division.

Biofilm formation is a crucial virulence factor in S. mutans cariogenesis. Our study indicated that biofilm formation increased in dltD deletion strain, but viable bacteria within the biofilm significantly decreased. In Neisseria meningitidis D-lactate dehydrogenase A mutant strains, it was found that increasing autolysis would increase the release of eDNA and cell death [33atlE mutant strains exhibited reduced autolysis and reduced eDNA release, thereby affecting biofilm formation [34Normal autolysis releases nutrients that benefit other bacteria within biofilms [21]. However, excessive autolysis resulted in extensive bacterial death and an altered biofilm structure. The dltD-deficient strain biofilm became more porous due to excessive autolysis, resulting in a decrease in biofilm density. The biofilm in the complemented strain was nearly restored to that of the wild-type strain.

The epistatic effects of dltD are unreported in other Gram-positive bacteria, particularly with constructed complementary strains for validation. A limitation of this study was our inability to explore the molecular regulatory network of dltD in S. mutans. Future research should encompass transcriptomic and proteomic analyses to expand our understanding of the physiological regulatory network of S. mutans.

Conclusion

Our results show that dltD influences growth and autolysis in S. mutans. dltD deletion leads to a reduction in viable bacteria in the biofilm owing to increased biofilm autolysis. dltD exerted epistatic effects on the division, and cell surface charge of S. mutans. These findings will guide future research into the regulatory mechanisms of dltD in S. mutans.

Author contributions

Conceptualization, J.-Y.D. and Y.-J.L.; methodology, S.H. and Y.-J.L.; software, Y.-J.L.; validation, S.H.and M.-J.W.; formal analysis, J.-Y.D.; investigation, J.-Y.D.and S.H.; resources, X.-J.H.; data curation, J.-Y.D.; writing – original draft preparation, J.-Y.D.; supervision, X.-J.H.and S.C.; project administration, X.-J.H.; funding acquisition, L.Z. and X.-J.H. All authors have read and agreed to the published version of the manuscript”.

Acknowledgments

Thanks to Professor Zhengwei Huang from Shanghai Ninth People’s Hospital for the gift of pIB169 plasmid.

Additional information

Funding

This work was funded by Project supported by the Natural Science Foundation of Fujian Province [2021J01802]; Scientific research foundation for Minjiang scholars [2018-KQMJ-02].