Introduction

Given the increased infection risk factors in hospitalized patients, it is crucial to understand the presence of virulence genes and the prevalence of antibiotic resistance among bacteria that cause nosocomial infections [1]. Coagulase-negative staphylococci (CoNS) isolates are important opportunistic pathogens among hospitalized patients. In addition, Staphylococcus isolates are one of the major skin and mucous membrane microbiota [2, 3]. According to a previous study, CoNS isolates can spread among healthcare workers and hospitalized patients [3,4,5,6].

Various genes, including icaADBC (intracellular adhesion) operon, fbe (fibrinogen binding protein), bhp (bap homologous- protein), embp (extracellular matrix-binding protein), aap (accumulation-associated protein), and atlE (autolysin), are involved in biofilm formation, mediate the initial adhesion, and bind specifically to fibrinogen, fibronectin, and collagen. The ica operon can be regulated by insertion sequence (IS) elements. The integration of IS elements can alter genes expression, and IS256 confers phase variation phenomena concerning biofilm formation, modulation of antibiotic resistance and virulence genes in S. aureus and CoNS isolates [7,8,9,10,11].

Biofilms has considerably affect on evasion of host immune system and resistance to antibiotics in staphylococcal infections [12,13,14]. Biofilm formation and colonization on medical devices make the CoNS isolates one of the most frequent causes of nosocomial infections in hospitalized patients [12,13,14]. Nowadays, most clinical CoNS isolates, such as S. epidermidis and S. haemolyticus, are methicillin-resistant (MR-CoNS) and are commonly resistant to different antibiotic agents [3, 14].

Typically, the biofilm formation in CoNS isolates is based on the polysaccharide intercellular adhesion (PIA) matrix, which is also known as poly-N-acetyl-glucosamine (PNAG). This polysaccharide matrix is synthesized with proteins encoded by the icaADBC operon in Staphylococci species [12, 15]. On the other hand, the microbial surface components, which recognize adhesive matrix molecules (MSCRAMMs), such as Fbe and Embp, can bind to fibrinogen and fibronectin, respectively, and have an important role in attachment to the cells and biofilm production in CoNS [13]. Also, other virulence genes, such as atlE, bhp, and aap, are commonly found in S. epidermidis isolates and act as initial attachment factors in biofilm formation [6, 7]. Identifying and targeting polysaccharides and proteins involved in adhesion and biofilm formation can be suitable candidates for vaccine production that prevent staphylococcal infections [16,17,18]. The vaccine can be effective in preventing the infection. Although there is a good body of research on this issue, further practical studies should be undertaken on the vaccine to prevent infections [16,17,18].

The arginine catabolic mobile element (ACME) consists of two gene clusters: the arc-operon encoding a secondary arginine deaminase system and the opp3-operon encoding a putative oligopeptide permease system [19, 20]. These elements are homologs of virulence determinants in other bacterial species and enhance bacterial adaptability and colonization [20, 21]. The horizontal transfer of the ACME in intra-species of Staphylococci was confirmed in previous studies [19, 20]. However, the ACME was shown to contribute to the colonization and survival of CoNS isolates in hospital settings. ACME allotypes were classified as (1) ACME-I containing the arc and the opp-3 gene clusters, (2) ACME-II containing the arc, and (3) ACME-III containing the opp-3 [20, 22]. The transmission of infectious agents by healthcare workers was observed in hospital settings [2, 3]. Therefore, it is important to study the characteristics of typical flora isolates from hospital staff to control infection in hospitals. The present study determines biofilm formation ability and the prevalence of biofilm-associated and virulence genes, IS256, IS257, mecA, and ACME clusters in CoNS isolates collected from clinical staff and healthcare workers’ skin in Kerman, Iran

Results

Bacterial strains and methicillin-resistant CoNS detection

In total, 114 CoNS isolates were collected in this study. Different species of CoNS isolates, including S. epidermidis 77.2% (n = 88), S. haemolyticus 16.6% (n = 19), S. hominis 5% (n = 6), and S. saprophiticus 1% (n = 1), were detected. Among 114 clinical and healthcare workers’ CoNS isolates, 66.5% (n = 76), 10.5% (n = 12), 20% (n = 23), 6% (n = 7), 3.6% (n = 4), and 1.8% (n = 2) were collected from neonatal intensive care units (NICU), intensive care unit (ICU), infectious disease units (ID), pediatric intensive care units (PICU), bone marrow transplant (BMT), and oncology units, respectively. In addition, 96.5% (n = 56) of clinical and 70% (n = 40) of healthcare workers’ isolates were resistant to cefoxitin and considered MR-CoNS. All MR-CoNS isolates were positive for mecA.

Biofilm assay

Biofilm formation was observed at strong, moderate, and weak levels in 96.5% (n = 55) of clinical and 91% (n = 52) of healthcare workers’ isolates in this study. Among clinical samples, 46% (n = 26) of isolates were categorized as strong biofilm producers, 37% (n = 21) were moderate biofilm producers, 14% (n = 8) were weak biofilm producers, and two isolates were non-biofilm producers. In healthcare workers’ isolates, 21% (n = 12) were strong biofilm forming, 30% (n = 17) were moderate biofilm forming, 40% (n = 23) were weak biofilm forming, and 9% (n = 5) isolates were non-biofilm forming. Also, the prevalence of strong biofilm formation in clinical isolates was higher than that of healthcare workers’ isolates (P < 0.004). Among 88 S. epidermidis, 40% (n = 36), 23% (n = 20), and 36% (n = 32) were strong, moderate, and weak biofilm producers, respectively. Moreover, among 19 S. heamolyticus, 89% (n = 17) and 10% (n = 2) were moderate and weak biofilm producers, respectively. Two S. hominis were strong, and 3 were moderate biofilm producers.

Detection of ica operon, ACME type, IS, and virulence genes

The profile of icaADBC, virulence genes, and methicillin resistance among biofilm producer of CoNS isolates are presented in Tables 1 and 2. The rate of icaA and icaB was 33% (n = 38), while the prevalence of icaC and icaD was 34% (n = 39) and 37.7% (n = 43), respectively. In clinical isolates, 42% (n = 24) were positive for icaADBC, while 17% (n = 10) of healthcare workers’ isolates were positive for this operon. There was a significant relationship between icaD and mecA in both groups of isolates, among 42 isolates were positive for icaD, 93% (n = 39) were positive for mecA simultaneously (P < 0.030). The fbe and bhp were detected in 11.4% (n = 17) and 6% (n = 7) of isolates, respectively. In addition, the prevalence of embp, aap, and atlE was found to be 55% (n = 63), 53.5% (n = 61), and 51% (n = 58) in isolates, respectively as shown in Table 3 (Fig. 1).

Fig. 1
figure 1

PCR electrophoresis: A: aap (400 bp), B: atlE (682 bp), C: bhp (1917 bp), D: embp (455 bp), E: fbe (496 bp) and F: IS257 (621 bp) F: M: 100 bp DNA marker, P: Positive control, N: Negative control

Full size image
Table 1 The profile of biofilm-associated and virulence genes among 57 clinical isolates
Full size table
Table 2 The profile of biofilm-associated and virulence genes among 57 healthcare worker isolates isolates
Full size table

We did not observe a significant difference between clinical and healthcare workers’ isolates in terms of embp, aap, and atlE prevalence. The prevalence of IS257 was higher than IS256 in both clinical and healthcare workers’ isolates (83% vs. 46%). The frequency of fbe (23% vs. 7%) and IS256 (79% vs. 14%) in clinical isolates was higher than the one in healthcare workers’ isolates (P < 0.050). The distribution of icaA, icaB, icaC, icaD, and biofilm-associated virulence genes (fbe, bhp, embp, aap, and atlE) in clinical and healthcare workers’ isolates are presented in Table 3.

Table 3 Prevalence of biofilm associated, virulence, mecA genes, IS256, IS257 and ACME cluster genes among clinical and healthcare worker´ isolates
Full size table

Among 37 strong biofilm producers, 32 isolates were positive for icaADBC, and we observed a significant relationship between the presence of icaADBC operon and strong biofilm production in isolates (P < 0.001). Furthermore, in moderate and weak biofilm-forming isolates, the frequency of icaADBC was lower compared to strong biofilm-producing isolates. The prevalence of aap in the moderate biofilm producers (79%, 30 out of 38) was higher compared with other isolates. Biofilm-associated genes among the biofilm-producing clinical and healthcare workers’ isolates are presented in Table 4. Fbe, bhp, embp, and IS256 were more frequently present in icaAD-positive isolates than in icaAD-negative isolates (P < 0.030).

Table 4 Prevalence of the Biofilm-associated genes and IS elements (IS256 and IS257) among the strong (n = 38), moderate (n = 38), and weak (n = 31) biofilm producers. P1: p-value between strong biofilm and moderate biofilm isolates. P2: p-value between strong biofilm and weak biofilm isolates. P3: p-value between moderate biofilm and weak biofilm isolates
Full size table

In total, 31 isolates were positive for ACME clusters, including 40% (n = 23) of healthcare workers’ isolates and 14% (n = 8) of clinical isolates. Among the 23 healthcare workers’ ACME-positive isolates, 50% (n = 12) harbored ACME-III (opp-3AB positive), 39% (n = 9) harbored ACME-II (arcA positive), and 8.6% (n = 2) were ACME-І (arcA and opp-3AB positive). In addition, 8 clinical isolates were positive for ACME clusters, among which 37% (n = 3) were ACME-І and ACME-III, and 25% (n = 2) were ACME-II (Fig. 2). The frequency of ACME in healthcare workers’ isolates was considerably higher than that in clinical isolates (P < 0.001) (Table 3). We should note that the isolates with ACME were biofilm producers (strong, moderate, and weak), though no association was detected between ACME genes and strong biofilms or icaADBC. In ACME-harboring isolates, embp (64%, n = 20), aap (67%, n = 21), and atlE (61%, n = 19) were significantly prevalent compared to ACME-negative isolates (P < 0.010) (Fig. 3).

Fig. 2
figure 2

PCR electrophoresis: opp3AB (1183 bp) and arcA (1946 bp), M: 100 bp DNA marker. І:ACME- І, II: ACME- II and III: ACME- III. N: Negative control

Full size image
Fig. 3
figure 3

The prevalence of embp, aap, and atlE in ACME positive and ACME negative isolates. (**: P-Value < 0.010)

Full size image

Discussion

In therapeutic and infection control strategies, it is important to identify the genetic and phenotypic characteristics of CoNS that cause nosocomial infections, including the presence of virulence genes, biofilm production, and antibiotic resistance [23]. Several characteristics of CoNS are advantageous for the pathogenesis, colonization, and transmission of these bacteria [13, 20]. CoNS are the reservoir of genes in hospitals and are considered in gene dissemination to other bacteria, and healthcare workers have an important role in transmitting these bacteria to hospitalized patients [24, 25]. In addition, MR-CoNS are often resistant to other antibiotics and present challenges in hospitals, especially in intensive care units [3]. In this study, we investigated biofilm formation and identified the biofilm-associated and virulence genes as well as IS256, IS257, mecA, and ACME clusters in CoNS clinical and healthcare workers’ isolates.

Biofilm formation is the most important mechanism of pathogenicity in CoNS isolates [3]. We found a significant difference between clinical and healthcare workers’ isolates in producing strong biofilm (P < 0.001), a finding that agrees with other studies [24, 26]. In addition, the prevalence of icaADBC was higher in clinical isolates than in healthcare workers’ isolates (42% vs. 12%) (P < 0.030). our study, similar to some other studies, 90% (n = 34) of strong biofilm-producer isolates were positive for icaADBC, which showed an association between ica operon and strong biofilm formation in isolates (P < 0.001) [25, 27, 28].

The co-existence of icaA and icaD increases N-acetyl-glucosaminyl-transferase activity for synthesizing polysaccharide intercellular adhesion (PIA) oligomers and improves biofilm formation [12]. According to earlier studies, PIA is considered a putative candid for vaccine development, though more studies on staphylococcal virulence factors are warranted [16,17,18]. In this study, 91% (n = 42) of healthcare workers’ isolates were biofilm producers, and 23.8% (10/42) were positive for icaADBC. This finding should be considered in preventing and controlling infection in hospitals. Our results showed that 63% (5/8) and 43% (20/46) of wound and blood infection samples were strong biofilm producers associated with icaAD and other virulence genes, such as embp and aap. This finding is not unusual because biofilm formation of Staphylococci on a medical device is common in hospital settings [29, 30]. More than 90% of the clinical isolates of Staphylococcus epidermidis with icaD and strong biofilm phenotype also carried mecA and we observed an association between the presence of the icaD and the presence of the mecA in these isolates. Other studies have reported the presence of the ica operon genes and the simultaneous presence of the mecA gene [24, 25]. Although the presence of ica-operon have the key role in strong biofilm producer isolates in our study, other isolates formed biofilm at strong, moderate, or weak levels in non-associated ica-operon. Therefore, biofilm production in staphylococci is associated with both ica operon and proteins, such as MSCRAMMs, which play an important role in biofilm formation in these bacteria [12, 13].

The adhesive surface proteins have an important role in biofilm formation in associated or non-associated ica genes [12]. In our results, fbe was detected in 23% of clinical, which was about similar to a study in India (20%) and lower than other studies in Germany (100%), and Iran (89.9%) [5, 31, 32]. Moreover, the frequency of bhp (6%) was lower than other studies in Iran (15.3%) and Germany (18.8%), also a study in India indicated none of the strong biofilm formers was positive for bhp [5, 31, 32]. In addition, 65% (47 of 72) of strong and moderate biofilm-producer isolates were positive for aap in our study. In our findings, fbe and bhp were detected in strong and moderate biofilm-producer isolates. By contrast, other studies in India and Brazil reported these genes in moderate biofilm-producer isolates or the ubiquitous presence of these genes in strong, moderate, and non-biofilm-producer isolates [5, 13]. In our study, aap, a gene encoded accumulating associated protein, was detected in 55% (n = 63) of isolates, this rate is difference from other studies in Iran (94.8% and 64.4%), Germany (93.8%) and India (10%) [5, 31,32,33]. In the present study, 73% (56 out of 76) of strong and moderate biofilm-producer isolates were positive for aap, indicating this protein’s importance in biofilm formation. aap was detected in clinical and healthcare workers’ isolates without significant differences. We found that the prevalence of atlE (51%) and embp (55%) differed from other studies in India and Brazil (30–100%), this discrepancy may be due to different sources of specimens in studies [5, 13]. Among 37 strong biofilm isolates that were positive for icaA and icaD, 20 (54%) isolates were positive for IS256, indicating an association between icaAD, biofilm formation, and IS256. This finding is similar to the results obtained by Petrelli et al. [34]. In total, the frequency of icaA, icaD, and IS256 was significantly higher in MR and clinical isolates in comparison with methicillin-sensitive and healthcare workers’ isolates, which indicates an association among mecA, icaA, icaD, and IS256, similar to earlier studies [25, 32, 35]. In addition, our study supported the idea that IS256 acts as a marker for clinical isolates [4, 25]. In our study, 40% of healthcare workers’ isolates were ACME-positive. The high prevalence of ACME among commensal Staphylococcus species notably affects the growth, survival, colonization, and spread of CoNS among healthcare workers in hospital settings [20]. Also, eight MR-clinical biofilm-producer isolates from blood and wound infections harbored ACME elements, and more than 70% carried icaADBC, embp, atlE, and aap.

However, CoNS are part of normal skin flora. The presence of virulence factors related to the attachment alone does not confirm the pathogenicity of this bacteria, and other factors should be considered. On the other hand, the presence and expression of a set of genes and environmental factors, such as temperature and osmolality, contribute to pathogenicity and biofilm formation in Staphylococci [36,37,38]. We detected all studied genetic elements in both clinical and healthcare workers’ isolates. Our results showed that all the ACME-positive isolates were biofilm producers, which increased their survival rate in the hosts. In our study, the high ACME prevalence in healthcare workers’ isolates and its dissemination might contribute to the spread of isolates due to the direct contact of contaminated healthcare workers’ hands with patients.

Conclusion

In this study, we compared biofilm formation and the frequency of genes involved in biofilm formation, virulence genes, and IS256, IS257, and ACME in clinical and healthcare workers’ CoNS isolates. We found a significant relationship between clinical isolates and the presence of icaADBC, mecA, and IS256. We detected some transposable elements contributed to pathogenicity, colonization, and transmission in CoNS isolates. In addition, strains with similar characteristics were found in the clinical and healthcare workers’ isolates, indicating the need for strategies to control and prevent infection in hospitals.

Materials and methods

Bacterial strains and methicillin-resistant CoNS detection

In this study, 114 CoNS isolates were collected from clinical samples (n = 57) and volunteer healthcare workers’ skin (n = 57) from January 2019 to December 2021. The healthcare workers included the nurses and physicians at Afzalipour referral hospital in Kerman, Iran. We excluded staff that cleaned their hands with antiseptics at the time of analysis. Clinical samples were isolated from blood, cerebrospinal fluid (CSF), and wounds. Systemic inflammation, absolute neutrophil count (ANC), C-reactive protein (CRP) levels, mono-microbial growth in cultures, and isolate detection from the second blood culture were the laboratory criteria to distinguish true infection and contamination [39]. The isolates were identified using standard biochemical tests, including colony morphology on Baird–Parker agar (Merck, Co, Germany), Gram staining, catalase production, oxidase test, coagulase-negative tests, susceptibility to novobiocin, and the utilization of xylose, arabinose, sucrose, maltose, mannitol, lactose, ribose, fructose, and mannose. Finally, they were confirmed as CoNS by multiplex-PCR, as described by Hirotaki et al. [40]. The multiplex-PCR reaction was performed in a volume of 50 µL containing: 25 µL of Taq DNA Polymerase Master Mix RED (Ampliqon, Co, Denmark), 2 µL DNA template, 0.25 µL of each primer (10pM), and DNase and RNase free water top up to 50 µL. multiplex-PCR amplification was carried out under the following conditions an initial denaturation step 5-min at 95 °C followed by 30 cycles (30 s of denaturation at 95 °C, 30 s of annealing at 58 °C, and 70 s of extension at 72 °C) and a final elongation step at 72 °C for 5 min. Staphylococcus epidermidis RP62A was used as a positive control strain in the M-PCR technique.

According to the clinical & laboratory standards institute (CLSI), cefoxitin (FOX, 30 µg) disc was used to screen methicillin-resistant isolates (MR-CoNS) [41]. Then, mecA gene was determined using the polymerase chain reaction (PCR) method in MR-CoNS, as previously described by Ruzauskas et al. [42].

Biofilm assay

Biofilm formation assay was determined according to the microtiter plate method described by Stepanovic et al. [43]. Briefly, The culture of each isolates was prepared in BHI medium overnight and adjusted to 0.5 McFarland. After incubation, the stationary-phase culture is vortexed and diluted 1:100, in medium for bioflim cultivation (TSB supplemented with glucose). The diluted bacteria are vortexed and then inoculated into a microtiter plate (200 mL per well). The microplates were incubated overnight for 24 h at 37 ̊C. Then, the wells were washed with PBS and fixed with methanol. The wells were stained with crystal violet and then Ethanol 95% were added to each wells. Finally The optical density of each plate was measured at 570 nm. All isolates were classified into the following categories: strong, moderate, weak and non-biofilm producer. The S. epidermidis strains ATCC 35,984 (formerly RP62A) was used as positive control.

Detection of ica operon, ACME type and virulence genes

Bacterial DNA was extracted according to the boiling method [44]. The presence of icaA, icaD, icaB, icaC, atlE, aap, bhp, embp, and fbe were determined by the PCR method [13, 25, 32]. The PCR program was run as follows: the initial denaturation at 94 °C for 3 min, 2) 30 cycles of 94 °C for 1 min, 3) annealing temperature (See Table 5) 1 min at 72 °C for extension step, and 4) final extension for 5 min at 72 °C. The products were detected on 1.5% agarose gel, and the band size was compared to a DNA marker (100 bp) Fig. 1. The allotypes of ACME were determined by the PCR [13]. ACME allotypes included ACME-I consisting of the arc and the opp-3 gene clusters, ACME-II containing the arc, and ACME-III constituting the opp-3. The primers used to identify ica operon and virulence genes are presented in Table 5. Staphlococcus epidermidis RP62A (ATCC 35,984) was used as positive control for aap, bhp, embP, and ica-operon genes in the PCR method.

Table 5 Target genes, primer sequences, and annealing temperatures used in this study
Full size table

IS256 and IS257 detection

The IS256 (1103-bp) and IS257 (621-bp) were detected by the PCR technique described by Kozitskaya et al. [4] The F-5′-TGAAAAGCGAAGAGATTCAAAGC and R-5′-ATGTAGGTCCATAAGAACGGC primers were used for IS256, and the primer sequences of F-5′-GCTAATTTCGTGGCATGGCG and R-5′- GTTATCACTGTAGCCGTTGG were used for IS257. The PCR program was run based on the following steps: (1) initial denaturation at 94 ºC for 4 min, (2) 30 cycles of denaturation at 92 ºC for 30s, (3) annealing step at 60 ºC for 40s, (4) elongation at 72 ºC for 30s, and (5) final extension at 72 ºC for 5 min. The product was visualized on 1.5% agarose gel, and the band size was compared to a DNA marker (100 bp).

Statistical analysis

We analyzed the data on SPSS version 23 (IBM, Armonk, NY, USA). The chi-square and Fisher’s exact tests were used to compare categorical variables, and P values ≤ 0.05 were considered statistically significant.