Dormancy Induces the Formation of Persisters in the Candida Albicans-staphylococcus Epidermidis Mixed Species Bio lm

jichen Yang 960th Hospital of PLA Bingquan Xiang Yunnan Cancer Center: Yunnan Cancer Hospital Wei Wang Yunnan Cancer Hospital Sunyin Rao Yunnan Cancer Hospital Run Cao Yunnan Cancer Hospital Shouyong Xiao Yunnan Cancer Hospital Xin Cui Yunnan Cancer Hospital Yunchao Huang Yunnan Cancer Center: Yunnan Cancer Hospital lianhua ye (  yelianhuaa@163.com ) Yunnan Cancer Center: Yunnan Cancer Hospital https://orcid.org/0000-0002-7748-5830


Background
The development of medicine makes more and more medical devices used to treat human disease, the incidence of bio lm infection in these patients with indwelling medical devices is also increasing [1], which can cause the dysfunction of the device that is implanted [2] and serious complications [3]. For example, the incidence of infective endocarditis after transcatheter aortic valve implantation was 5.5%, after the occurrence of infective endocarditis, the in-hospital mortality was 36.4%, and the one-year mortality rate was as high as 54.5% [4]. S.epidermidis has now become one of the most prevalent causes for implant-associated and nosocomial infections [5], such as endotracheal intubation devices[6] and arti cial heart valves [7]. Candida is a kind of opportunistic pathogen, which causes systemic infection through the colonization of oral, gastrointestinal and urinary tract [8]. C.albicans can form bio lms on almost any medical device, including urinary catheters, prosthetic joints, cardiac valves, pacemakers [9].
Of all types of nosocomial infections, approximately 27% of nosocomial candida infections are polymicrobial [10]. During long-term infection, multiple pathogenic microorganisms will produce mixed infections, and form mixed bio lm on the surface of biomaterials, the pathogenic bacteria will have a synergistic effect to resist the killing of antibiotics, making the infections di cult to eradicate [11].
Persister cells are a slow-growing semi-dormant fraction of cells within a bio lm with low levels of transcriptional activity [12], the reductions of anabolism and energy metabolism are their important characteristics [13]. The induced dormancy that characterizes the persistent cell population makes them resistant to classical antibiotics, and is one of the main reasons for bio lm recalcitrance [14]. But persisters comprise less than 1% of cells in a bio lm, one of the key challenges in the study is the di culties with isolation maintenance of the persistent cell phenotype during culture [15]. However, there have been no reports of large number of induced persisters of S.epidermidis and C.albicans. Study found that inhibiting cellular transcription, translation, or the formation of adenosine triphosphate increases persister cell formation inEscherichia coli [16], aminoglycoside antibiotics combined with bacterial metabolites could reduce the formation of persisters and increase the killing capacity of antibiotics [17]. Based on these studies, we speculate, the use of speci c drugs to block the energy metabolism of bacteria and fungi can induce cells to enter a dormant state, it is possible to produce a large number of persistent cells.
This study builds upon on our previously established polymicrobial bio lm model, in which a mixture of S.epidermidis and C.albicans were co-cultured on a biomaterial (e.g. PVC, titanium)[18], to determine the level of persisters in the mixed bio lms. The structural relationship between persisters and the bio lm was observed using SEM and CLSM. Our research solved the technical dilemma in studying the rare phenotypes of microorganisms, and contribute to understand the formation of the persister phenotype in polymicrobial bio lms. China) overnight at 37°C, respectively. Isolated colonies were picked and inoculated into TBS and grown at 37°C in a 150-rpm shaker for 24h until the logarithmic growth phase. The optical density (OD) of C.albicans suspension and S.epidermidis suspension were used to adjust cultures to a nal concentration of 2×10 6 CFU/mL by adding sterilized TSB medium. The bacterial and the fungal suspensions were then mixed in 1:1 volume ratio to obtain a polymicrobial mixture containing each microbe at concentration of 1×10 6 CFU/mL. Ten milliliters of the polymicrobial mixture was used for bacterial identi cation and drug sensitivity detection in a fully automated system. The polymicrobial mixture was divided into 9 test tubes, each containing a sterilized 1cm×1cm×1mm titanium coupon (sterilized at 121°C for 30min), and incubated at 37°C in a 150-rpm shaker for 24h to allow C.albicans and S.epidermidis to adhere to the surface of the coupon.

Dormancy induction
Titanium coupons with adhered cells were transferred to the fresh TSB containing tetracycline, gentamicin, rifampicin, 5-FC, caspofungin, uconazole, and CCCP(Beijing Solaibao Technology Co., Ltd. China) solutions(all at 10,000μg/mL, 100μL)for dormancy induction, and TSB without antibiotics was used as control. By inhibiting the anabolism of microorganisms, C.albicans and S.epidermidis in the mixed bio lm were induced to enter dormant state. After incubating in a 30-rpm shaker at 37°C for 1h, the media was removed and replaced with TSB containing 10μg/mL cipro oxacin and amphotericin B in the experimental groups and the negative control group, or sterilized TSB in the blank control group, all the above antibiotics were 10 times minimal inhibitory concentration(MIC). Cells were subsequently incubated in a 30-rpm shaker at 37°C for 6h. By inducing dormancy treatment, non persistent bacteria cells will be killed by high concentration of antibiotics, and only persistent bacteria cells can survive. The 9 titanium coupons were then transferred to a new centrifuge tube containing 10mL sterilized TSB medium, and were sonicated to dislodge the polymicrobial bio lms into the aqueous phase. Ten microliters of this polymicrobial solution was spread on an MH agar plate and a Sabouraud agar plate, and incubated at 37°C for 24h for enumeration. C.albicans cannot grow on MH agar plate, and S.epidermidis can not grow on Sabouraud agar plate. By this method, the number of S.epidermidis and C.albicans living on the surface of each biomaterial could be counted respectively. The count of the blank control group shows the number of S.epidermidis and C.albicans on the surface of the biological material after 24h of normal culture. The count of the negative control group showed the level of naturally occurring persister cells in the C.albicans-S.epidermidis mixed bio lm. The counts of each experimental group represent the proportion of persistent cells produced in the mixed bio lm after induction treatment with each antimetabolites. The above experiment was repeated 15 times. Based on initial results, tetracycline, gentamicin, 5-FC, caspofungin were selected for subsequent experiments.

Drug sensitivity tests
The microbial colonies in the agar plate before sterilization and after dormant sterilization were used to prepare microbial suspensions, which were subjected to microbial identi cation/drug sensitivity in an automated system to exclude the possibility of antimicrobial drug-resistance (Table 2).

Sample preparation and observation with SEM
After the induction of dormancy, titanium coupons from the 4 experimental groups and the blank control group were placed in a 24-well plate, and washed 3 times in 1mL cold PBS. Bio lms were xed in 2.5% glutaraldehyde solution for 24h. After washing 3 times in PBS, bio lms were xed in 1% osmium acid solution at 4°C for 2h, followed by dehydration in gradient ethanol solution for 20min, isoamyl acetate replacement for 20min, tert-butanol penetration at 40°C for 2h, CO 2 critical drying, ion-sputtered surface xed coating, and SEM(Laica, Germany) observation of the ultrastructure of the bio lms.

Sample preparation and observation with CLSM
Bio lms on titanium coupons prepared as described above were divided into three groups: blank control group (non-dormancy-induced and non-sterilized), negative control group (non-dormancy-induced, but sterilized), and experimental group (dormancy-induced and sterilized). Fluorescent dyes for bacterial viability assay were prepared using Live/Dead Baclight TM Bacterial Viability Kit 7017(Invitrogen, USA) following manufacturer's instructions. Coupons were washed 4 times in 4mL 4°C cold PBS and stained using 400μL Live/Dead microbial staining reagents in the dark at room temperature for 20min according to the manufacturer's instructions. The bio lm ultrastructure of samples were observed by CLSM(CLSMFV1000, Olympus Corporation, Japan) equipped with an argon laser (514/488 nm). The number of bacterial colonies in a unit of eld of view was counted by randomly examining one eld of view for samples of each group. The thickness of bio lm was measured by collecting confocal images, with resolution of 1024 pixels, in serial sections at 1μm intervals from interior toward exterior of one bacterial colony randomly selected from each eld of view.

Statistical analysis
SPSS23.0 statistical software (IBM SPSS Inc., Chicago, IL) were used to compare the counts of each experimental group with the counts of the negative control group and the blank control group. The comparison between groups was analyzed by independent sample t-test; P< 0.05 was considered statistically signi cant.

dormancy induction increased the formation of persisters in the mixed bio lm
Compared with the negative control group, the dormancy induced by tetracycline, gentamicin, rifampicin, 5-FC, uconazole and CCCP signi cantly increased the level of S.epidermidis persisters (P < 0.01), and the level of persisters of the tetracycline group was the highest (P < 0.01). The dormancy induced by 5-Fc, and caspofungin signi cantly increased the level of C.albicans persisters (P < 0.01), and the level of persisters of the 5-Fc group was the highest (P < 0.01)( Table 1).The bacterial solution of mixed bio lm in blank control group (not induced dormancy, not sterilized), negative control group (after induced dormancy, but not sterilized), and the mixed bio lm bacterial solution obtained by reculture in the experimental group(after dormancy induction, and after sterilization), the results of strain identi cation were S. epidermidis and C.albicans. There was no contamination of bacteria in the operation process. According to the drug sensitivity test, the S.epidermidis strain before inducing dormancy was resistant to erythromycin, clindamycin, oxacillin, and penicillin; and was sensitive to cipro oxacin, tetracycline, gentamicin, rifampin, linezolid, moxi oxacin, nitrofurantoin, levo oxacin, tigecycline, and vancomycin. *Compared with the negative control group, the dormancy induced by tetracycline, gentamicin, rifampicin, 5-FC, uconazole and CCCP signi cantly increased the level of S. epidermidis persisters(P < 0.01), and the level of persisters of the tetracycline group was the highest(P < 0.001).
**Compared with the negative control group, the dormancy induced by 5-Fc, caspofungin, andcaspofunginsigni cantly increased the level of C.albicans persisters(P < 0.01), and the level of persisters of the 5-Fc group was the highest(P < 0.001).
The C.Albicans strain was sensitive to amphotericin B, uconazole, 5-FC, caspofungin, and itraconazole. Induction of dormancy did not result in resistance of S.epidermidis and C.albicans to the antimicrobials tested. After treatment with cipro oxacin and amphotericin B, the survival of some S.epidermidis and C.albicans was not caused by antibiotic resistance due to drug-resistant mutation ( Table 2). Table 2 The agents susceptibility to S.epidermidis and C.albicans before and after dormancy induction The sensitivity and speci city of PCR-SSCp assay were 71.4% and 100%, respectively.

Morphological and structural changes of persisters in the mixed bio lm under SEM
Under SEM, the blank control group, the negative control group, and the experimental group all showed that mixed growth of C.albicans and S.epidermidis in the mixed bio lms. The hyphae and spores of C.albicans formed the reticular structure skeleton of the mixed bio lm. In the skeleton, S.epidermidis adheres to the surface of C.albicans, gathers together to form a mass, dense and complex hybrid bio lm. The morphological characteristics are consistent with the previous research results of the research group[18]. In the experimental group, the number of mixed bio lm reduced, the density of structure decreased, the clumpy colonies decreased, and the number of bacteria and fungi is reduced (Fig. 1A, B).
The mixed bio lm treated with 5-FC and caspofungin to induce dormancy, although the ratio of yeast phase to mycelium phase of C.albicans did not change, but the surface of C.albicans has a convex morphological change, with a diameter of about 70%-80% of the host bacteria (Fig. 1C, D), the change was noted to occur in both the yeast and the mycelial phases. In the negative control group, the C.albicans in the mixed bio lm had almost no such morphological changes (Fig. 1E). Afterinduced by tetracycline and gentamicinin the experimental group, the S.epidermidis showed reduced cell division and smaller volume than normal cells (Fig. 1F, G). In the mixed bio lm in the blank control group, there were a large number of mitotic cells with uniform size (Fig. 1H).

Changes of persisters and mixed bio lms under CLSM
In the blank control group, the CSLM images showed that mixed bio lm adhered to the surface of biomaterials, the bacteria on the surface of the bio lm are mainly living, which emit green uorescence, and are densely distributed in clumps. The average thickness of bio lm is about 37.1µm( Fig. 2A). In the negative control group, the mixed bio lm was dissolved and destroyed, and a large number of dead bacteria were seen, which emit red uorescence. The live bacteria were distributed on the surface of the mixed bio lm in a dot shape, and hardly gather in clumps, the average thickness of bio lm is about 6.3µm (Fig. 2B). In the experimental group, the structure of the mixed bio lm was partially destroyed, the average thickness is about15.7µm. The mixed bio lm with high content of persistent bacteria did not dissolve under the killing of antibiotics. The structure of the mixed bio lm was complete and dense. On the surface of the biomaterials, there were still living bacteria gathered into clumps. The mixed bio lm with low content of persistent bacteria dissolved in large area (Fig. 2C).

Discussion
Persisters in the bio lm have become a di cult point in the treatment of infection, how to remove persisters is particularly important for anti infection treatment. The low content of persisters of S.epidermid and C.albicans in the mixed bio lm brings many di culties to the research. The key to solve the problem is how to induce enough persistent cells for research. The dormancy of bacteria and fungi is mainly carried out by the inhibition of protein, ATP, nucleic acid synthesis, and the ribosome assembly. If the drugs with the above mechanism are used to induce bacteria and fungi, they can be put into a dormant state to extract persister bacteria. In this study, C.albicans ATCC10231 and S.epidermidis ATCC35984 were used to culture mixed bio lm on the surface of titanium sample. The drugs that inhibit the synthesis of bacterial protein and fungal nucleic acid, such as tetracycline, gentamicin, 5-Fc, and caspofungin, were used to induce the dormancy of the mixed bio lm, and made a lot of S.epidermidis and C.albicans to enter a dormant state, which laid the foundation for studying the drug resistance mechanism and phenotypic characteristics of persistent cells.
The study showed that, tetracycline and gentamicin can make S.epidermidis enter a dormant state in large numbers, 5-Fc and caspofunin can induce a large number of C.albicans to enter dormancy state. However, tetracycline and gentamicin could not induce the formation of dormant cells of C.albicans. Rifampicin, which interferes with mRNA synthesis, and CCCP, which inhibits ion transport, can only slightly increase the level of persisters of S.epidermidis in the mixed bio lm, although there is a signi cant difference between them and the negative control, and both of them could not induce the formation of dormant cells of C.albicans. 5-Fc inhibits DNA synthesis of fungi, while caspofungin damages the cell membrane and cell wall of fungi. Therefore, fungi can be induced to form persister cells by damaging the cell membrane and cell wall. These were different from those previously reported by Kwan et al[16], they pretreated E.coli with rifampicin, tetracycline and CCCP, and successfully induced the formation of persisters. Study [19]showed that tetracycline and rifampicin induced a viable but nonculturable state in S.epidermidis bio lm. Iron starvation and hypoxia induced persister cells of S.epidermidis in planktonic culture [20]. The number of persister bacteria obtained is more than 0.1%-1% reported by the study [21]. Therefore, inducing S.epidermidis and C.albicans into dormancy is not simply to inhibit its anabolism. Growth stagnation and anabolism decrease are only the results of the formation of persister cells rather than their mechanisms. Different microorganisms have different mechanisms for producing persister cells. More research is needed to clarify the mechanism by which cells enter the dormant state [13].
The research found that, after the induced persisters were re-cultured, there was no signi cant change in antibiotic sensitivity compared with before induction. The survival of some S.epidermidis and C.albicans after cipro oxacin and amphotericin B treatment was not caused by antibiotic resistance due to drugresistant mutation. This indicates that the essence of persistent cells is a group of cell subgroups in a dormant state [13], which can escape the effect of antibiotics without changing their genetic characteristics.The tolerance of the bacteria to antibiotics is obtained by changing the environmental conditions and making them enter the non growth state (dormant state). Antimicrobial resistance is generated in the classical sense by genetic mutation of the corresponding antimicrobial drug target, which manifests as an increased MIC. In contrast, antimicrobial drug tolerance is not genetic, but rather an inducible phenotypic switch. The characteristics of non growth state and slow metabolism of persister bacteria make it possible to relapse again after avoiding the killing of the antibiotics, which makes the clinical cure more di cult. With the extension of the course of the disease, the level of persister bacteria will gradually increase, even more than 100 times [22]. Antibiotic tolerance is associated with failure of antibiotic treatment, recurrence of bacterial infection and prolonged disease [23].
Under the SEM, in the mixed bio lm inducedby 5-FC and caspofungin, some C.albicans cells showed bulge-like morphological changes. Benmansour W et al [24] used amphotericin B to screen C.albicans ATTCC10231 and showed similar bulge-like morphological changes on the surface of C.albicans under electron microscope. These indicate this morphological change may occur, regardless of the C.albicans being treated with dormancy-inducing drugs or directly killed by antimicrobial drugs. However, in the blank control group, only a few strains of C.albicans had this morphological change, and the proportion of persisters was far less than the experimental group. Yang et al [25] used vancomycin with the lowest bactericidal concentration to screen S.epidermidis, after 48h, the cell division phase of S.epidermidis living in the bio lm was reduced and the volume was smaller than that of ordinary bacteria, which was similar to the volume change of dormant S.epidermidis in this study, this change in volume reduction may be related to the cessation of metabolic activities and the dormancy, but the exact mechanism needs further study. CLSM observations showed that the mixed bio lms of the experimental group had clumpy structures, which mostly showed mixed growth of live and dead bacteria, these persister cells scattered in the mixed bio lm can maintain the bio lm structure, and resist the killing of antibacterial drugs, this may be an important reason why bio lm infection is di cult to cure. A study [26] showed that tRNA acetylation is one of the formation mechanisms of persisters. By using the hydrolase peptidyl-tRNA hydrolase (Pth) to hydrolyze the hetero-acetylated tRNA, the normal tRNA is released to reactivate the persisters,thus signi cantly reducing the level of persisters and achieving the purpose of controlling the persisters. Interfering with the dormancy process of persisters from the molecular mechanism, inhibiting the dormant state, and improving the effectiveness of antibiotics, may be a new direction for the treatment of bio lm infection. Whether there is such a mechanism in the persistesin S.epidermidis-C.albicans mixed bio lm, it needs to be further studied on the basis of successful induction of a large number of persisters.

Conclusions
Induction of dormancy is a reliable method for the formation of persistent bacteria in C.albicans-S.epidermidis mixed bio lm.In this study, we successfully induced S.epidermidis-C.albicans mixed bio lm to produce a lots of persistent cells, which provided a basis for further research on the phenotypic characteristics and the drug resistanceof persisters.

Declarations
Ethics approval and consent to participate Our manuscript does not report on or involve the use of any animal or human data or tissue, the"Ethics approval and consent to participate" was not applicable in our manuscrip. SEM image of the mixed bio lm (Fig.A,B,C,D,E,H×2000 times ) In the blank control group(A), the bio lms showed mixed growth of C.albicans and S.epidermidis, the hyphae and spores of C.albicans formed the reticular structure skeleton of the bio lm. In the experimental group(B), the number of bio lm, bacteria and fungi is reduced, the density of structure decreased. Induction with 5-FC(C)and Caspofungin(D) (Arrow shows morphological change of C.albicans). In the negative control group(E), the C.albicans had almost no convex morphological change. After inducing with tetracycline(F) and gentamicin (G)(Circle shows the S.epidermidis with reduced cell division and smaller volume than normal cells)(×4000 times).The blank control group (H) had a large number of mitotic cells with uniform size.

Figure 2
CLSM image of the mixed bio lm In the blank control group(A), the bio lm adhered to the surface of biomaterials, the living bacteria that emit green uorescence are densely distributed on the surface of bio lm. In the negative control group(B), the bio lm was dissolved and destroyed, and a large number of dead bacteriae mitting red uorescence were seen, the live bacteria were distributed on the surface of the bio lm in a dot shape. In the experimental group(C), the structure of the bio lm was partially destroyed, the bio lm with high content of persisters did not dissolve under the killing of antibacterial drugs, the bio lm with low content of persisters dissolved in large area.