Antimicrobial Resistance and Pathogenicity of Corynebacterium Striatum Clinical Isolates Collected from Three Tertiary Hospitals in China

) clinical strains were frequently reported in recent years. Even worse, daptomycin resistant isolates were found in some studies and possible resistance mechanism was explored. Also, few investigations revealed the diversity of resistance feature and potential pathogenicity of C. striatum strains with different genotypes. However, less is known about the possible differences of resistance feature and pathogenicity of C. straitum clinical strains from different hospitals at a long distance. Methods C. striatum clinical strains were isolated and identified with VITEK-2 ANC card, MALDI-TOF microTyper and 16S rRNA sequencing technique. Broth microdilution method was used to detect the antibiotic susceptibility profiles of 420 C. striatum clinical isolates, and PFGE method was used to discriminate different clones. Furthermore, in vitro adherence assay and mouse toxicity assay were performed to assess the pathogenicity of the strains with different genotypes. different clones with different genotypes were investigated.

a long time in hospital environment. The in vitro adherence abilities and mouse lethality among different clones differed greatly. The resistance and pathogenicity of C. striatum clinical strains should be paid more attention to, especially for some specific clones at different hospitals.

Introduction
In recent years, more and more reports regarding Corynebacterium striatum (C. striatum)were published, since most of the C. striatum strains were found to be multi-drug resistant and the only effective antibiotics were limited to vanvomycin, lizezolid [1][2][3]. Even worse, daptomycin resistant isolates were recently reported and its resistance mechanism was explored, which pose a more knotty issue for clinicians [4][5][6]. Furthermore, some recent reports revealed that C. striatum can lead to multi-site invasive infections, such as bacteremia, monomicrobial bone and joint infection, prosthetic valve endocarditis [7][8][9], which suggest that C. striatum may possess stronger pathogenicity, especially for the patients with specific risk factors [10,11].
It is reported that C. striatum can colonize onto multiple environmental and bodily surfaces and could transmit within patients in specific wards [12,13]. Also, the multi-drug resistant C. striatum clones seemed to be with stronger adhesion abilities [14] , which was consistent with one of our previous study [15], which may partly explain the reason why C. striatum can rapidly transmit within hospital environment. Moreover, Souza C et al [16] reported that C. striatum could lead to the death of Caenorhabditis elegans, and virulence potential varied among different C. striatum clones. The pathogenicity potential of C. striatum strains isolated from different origins need to be further investigated.
In this study, we aim to perform a further investigation with a larger number of C. striatum isolates from multi-centres to better understand the actual resistance and pathogenicity features of C. striatum strains isolated from aseptic sites were collected and identified. The sputum samples were enrolled if they were evaluated to be qualified based on the numbers of leukocytes and epithelieum using microscopy [13]. All of the cultures suspected to be C. striatum were routinely identified with VITEK-2 ANC card(Biometrieum, France), MALDI-TOF microTyper (Tianrui, China) and 16S rRNA sequencing technique. Only one C.striaum strain from the same patient was selected in this study, while the repeated ones were excluded.

Antibiotics susceptibility test
The antibiotic susceptibility test was performed using broth microdilution method, and the antibiotics tested include ceftriaxone, imipenem, erythromycin, clindmycin, gentamycin, tetracycline, vancomycin, ciprofloxicin, sulfamethoxazole, rifampin and daptomycin. The operation of susceptibility test and results analysis were performed according to Clinical and Laboratory Standards Institute guidelines (CLSI) [17].

Pulsed-Field Gel Electrophoresis (PFGE)
In general, the whole genomic DNA chromosome of the C. striatum was extracted, and macrorestriction digestion (SwaI) and PFGE (CHEF-DR II apparatus; Bio-Rad, Hercules, CA, USA) were performed as previously reported [15]. Macrorestriction patterns were analysed using Dice coefficient with Bionumerics software (Applied Maths, Kortrijk, Belgium,Version 5.0). The classification criteria for PFGE analysis was subjectively designated and the clones with similarity index 100% was classified to be a single type and named with a single capital letter.

In vitro adherence assay
In vitro adherence abilities on polystyrene surfaces of different C. striatum strains were detected quantitatively in 96-well flat-bottomed microtitre plates with a commonly used method described previously [15]. Generally, aliquots of 200 µL of bacterial suspensions [0.2 optical density (OD) at λ = 620 nm] were added to the microplate wells and incubated for 24 h at 35 °C. After the suspension content removed, the remaining attached bacteria in each well were fixed with 99% methanol and stained with 2% crystal violet. The negative controls contained LB liquid medium only. Then 33% glacial acetic acid was used to dissolve bound crystal violet and the OD of the solution was determined ( λ = 620 nm) using an enzyme immunosorbent assay reader (Tecan, Sunrise). The cut-off OD (ODc) was defined as the mean OD of the negative control. The adhesion abilities was classified as nonadherent, weakly, moderately and strongly according to Souza C et al [14].

Mouse lethality assay
Six different strains with PFGE types and ATCC6940 standard strain were selected to do mouse lethality assay. Generally, overnight cultures of bacteria were washed twice with sterile PBS by centrifugation, and approximately 10 7 CFUs of bacteria for each strain were then added into 200 ml of PBS for inoculation. Eight-week-old BALB/c mice were intraperitoneally injected with the inoculums.
The survival rates of mice were monitored for 7 days. Moribund mice were anesthetized and sacrificed prior to death. Mice were housed and monitored daily at the Animal centre of Chinese Center For Disease Control And Prevention, Beijing.

Isolates distribution and Antibiotic resistance feature
For the 420 isolates analyzed in this study, 329 isolates (78.3%) were collected from hospital A, 44 isolates (10.5%) from hospital B and 47 isolates (11.2%) from hospital C.

Mouse lethality assay
After intraperitoneal injection C. striatum (10 7 CFUs), mouses died from day 1 to day 4. As shown in Table 5, seven strains showed different mouse lethality. The non-dominant clone (Strain NMGYC339, T24) showed the strongest mouse virulence (90% lethality) and strong in vitro adherence ability, which was isolated from wound secretion. No death was observed in the groups of strain NMFYC177 and strain NMFYC477.

Discussion
As a potential pathogen for hospitalized patients in recent years, C. striatum was found to be isolated from almost all types of clinical samples [18][19][20], especially for sputum, whole blood, cerespinal fluid. In this study, the 420 C.
Striatum were mainly isolated from lower respiratory tract (89.2%, 375/420), and 52.4% samples (220/420) were purely cultured with C. striatum. Five C. striatum strains were isolated from whole blood samples, and one of the strains were repeatedly isolated from a immunosuppressed patient with infective endocarditis. Some common multi-drug resistant bacteria were simultaneously isolated with C. striatum in the same sample (53.1%, 223/420), including Acinetobacter baumannii, Methicillin-Resistant Staphylococcus Aureus (MRSA), ESBLs or carbapenemase producing Klebsiella pneumoniae. This study highlights the higher prevalence of C. striatum from the long stay patients, especially for the ones with specific risk factors [14,15,21].
Consistent with the majority of the investigations previously published [1][2][3], the C. striatum strains detected in this study were all sensitive to vancomycin and linezolid, but no daptomycin resistant isolate was observed. The resistance rates of 420 C. striatum strains to penicillin, cefepime, ciprofloxacin, erythromycin, sulfamethoxazole and trimethoprim and clindamycin were all higher than 95%. Moreover, it is reported that the usage of parenteral antimicrobial drugs could promote C. striatum acquisition [22] and the resistance rate of C. striatum strains to meropenem reached up to 86.9% (365/420) in this study, which suggested that prescription of carbapenems or other non-sensitive antibiotics should be carefully considered, especially for the patients with risk factors for C.striatum acquisition or infection.
Based on the resistance biotypes of C. striatum to 12 types of antibiotics tested in this study, three dominant resistance biotypes were discriminated, which were all multi-drug resistant. To further analyze the potential differences of resistance patterns of C. striatum from different hospitals, 420 isolates from 3 different hospitals in China were enrolled and analyzed. The isolates from hospital A and hospital C were separately distributed into at least 10 different resistance biotypes, while the isolates from hospital B were intensively distributed in 4 biotypes.
The resistance features of the isolates from different hospitals varied greatly based on the results of this study.
However, owing to relatively few isolates from hospital B and hospital C, the actual differences of resistance patterns of C. striatum isolates among these 3 hospitals still need to be explored.
To further investigate whether some dominant clones prevailed in the same hospital, PFGE typing method was employed, which revealed that 2 dominant clones (T36 and T14) were isolated from different patients in hospital A and hospital C and lasted from 2016 to 2019, which suggested that some dominant clones could exist in hospitals for a long time. Consistent with our findings, Baio PV et al [23] reported that dominant C. striatum clones can rapidly spread among inpatients and can exist for a long period in the same hospital [2]. Also, it is revealed that C. striatum could transmit from person-to-persion via the hands of healthcare personel or medical environment [24][25][26]. Unfortunately, based on the current data we obtained, it is difficult to confirm whether these two clones originated from the same patient in these two hospitals at a long distance. For the remaining two dominant clones (T32 and T28), they were only isolated from hospital C and the two times of nosocomial outbreak were observed from April, 2019 to May, 2019 and both of the outbreaks occured in intensive care unit (ICU). For specific wards such as ICU, more efficient infection control measures should be implemented to better control the transmission of C. striatum.
Souza C et al [14] reported that C. striatum clinical strains can tightly adhere to several kinds of environmental surfaces or implanted medical devices [27][28][29] and lead to several kinds of invasive infections. Consistently, most of the C. striatum strains tested in this study showed moderate to strong biofilm abilities and differed greatly. It is also noteworthy that all of the 4 dominant PFGE clones showed moderate adherence abilities on polystyrene surface, while the most adhesive strains belonged to non-dominant clones and the top three strains were all collected from hospital B. More attention should be taken for preventing catheter-related infections by C.
striatum, especially for the strains with strong adherence abilities. Furthermore, as far as we know, it is the first study to reveal the significant mouse lethality of C. striatum. Also, the consistency between in vitro adherence and in vivo mouse lethality of the C. striatum strains was not observed in this study. The actual pathogenicity of C. striatum strains and possible mechanism deserves to be further investigated.

Conclusions
Most of the C. striatum clinical strains analyzed in this study showed multi-drug resistant features and could exist in hospitals for a long period. The antibiotic resistance patterns of C. striatum strains differed greatly from different districts, and most of the C. striatum strains can present with moderate to strong in vitro adherence abilities on polystyrene surfaces and some clones could lead to significant mouse lethality. The pathogenicity mechanism of multi-drug resistant C.striatum should be further explored to better prevent and control acquisition or infection of C. striatum.