Evaluation of the Ability to Form Biofilms in KPC-Producing and ESBL-Producing Klebsiella pneumoniae Isolated from Clinical Samples

The appearance of Klebsiella pneumoniae strains producing extended-spectrum β-lactamase (ESBL), and carbapenemase (KPC) has turned into a significant public health issue. ESBL- and KPC-producing K. pneumoniae’s ability to form biofilms is a significant concern as it can promote the spread of antibiotic resistance and prolong infections in healthcare facilities. A total of 45 K. pneumoniae strains were isolated from human infections. Antibiograms were performed for 17 antibiotics, ESBL production was tested by Etest ESBL PM/PML, a rapid test was used to detect KPC carbapenemases, and resistance genes were detected by PCR. Biofilm production was detected by the microtiter plate method. A total of 73% of multidrug resistance was found, with the highest resistance rates to ampicillin, trimethoprim–sulfamethoxazole, cefotaxime, amoxicillin-clavulanic acid, and aztreonam. Simultaneously, the most effective antibiotics were tetracycline and amikacin. blaCTX-M, blaTEM, blaSHV, aac(3)-II, aadA1, tetA, cmlA, catA, gyrA, gyrB, parC, sul1, sul2, sul3, blaKPC, blaOXA, and blaPER genes were detected. Biofilm production showed that 80% of K. pneumoniae strains were biofilm producers. Most ESBL- and KPC-producing isolates were weak biofilm producers (40.0% and 60.0%, respectively). There was no correlation between the ability to form stronger biofilms and the presence of ESBL and KPC enzymes in K. pneumoniae isolates.


Introduction
Klebsiella pneumoniae is a Gram-negative bacterium that is commonly found in the human gut and can cause severe infections in different parts of the body, especially in people with debilitated immune systems [1]. The emergence of extended-spectrum βlactamase (ESBL) and Klebsiella pneumoniae carbapenemase (KPC) producing strains of K. pneumoniae has become a major public health concern, as these strains are resistant to many antibiotics and pose a significant challenge in the treatment of infections [2].
ESBLs are enzymes produced by bacteria that break down the β-lactam antibiotics, such as penicillins, cephalosporins, and carbapenems. Since these antibiotics are commonly prescribed to treat these bacterial infections, the treatment of ESBL-producing K. pneumoniae becomes a challenge [3]. KPCs are a type of carbapenemase enzyme that can break down carbapenem antibiotics, which are often used as a last-resort treatment for antibioticresistant infections. This means that KPC-producing K. pneumoniae can be resistant to almost all available antibiotics [4].
One of the important virulence factors of K. pneumoniae is its capacity to form biofilms [5]. Biofilms are communities of microorganisms that adhere to a surface and produce a matrix of extracellular polymeric substances (EPS). The EPS matrix protects the microorganisms from environmental stress, such as antibiotic treatment, making biofilm-associated infections difficult to treat [6]. K. pneumoniae is known to form biofilms on various surfaces in healthcare settings, such as medical devices and surfaces in hospitals [7,8].
Studies have shown that ESBL-producing K. pneumoniae strains have a higher ability to form biofilms compared to non-ESBL-producing strains [8,9]. This is because ESBLs provide a survival advantage to the bacteria, allowing them to resist the effects of antibiotics and persist in the host [10]. The EPS matrix also provides an environment for horizontal gene transfer, which can facilitate the spread of antibiotic resistance [11,12].
KPC-producing K. pneumoniae strains can produce biofilms [13], and this ability has been associated with high levels of antibiotic resistance. This can result in persistent infections that are difficult to treat and can lead to increased morbidity and mortality [14].
There are several methods for evaluating the ability of bacteria to form biofilms. One commonly used method is the microtiter plate assay, in which bacteria are grown in a microtiter plate with a medium that promotes biofilm formation. After a period of incubation, the wells are stained with crystal violet, which binds to the biofilm matrix. The stained biofilm can then be visualized and quantified using microscopy or spectrophotometry [15].
In addition to evaluating biofilm formation, it is also important to understand the genetic mechanisms underlying biofilm formation in K. pneumoniae. Several genes and regulatory pathways have been implicated in biofilm formation in K. pneumoniae, including the type 1 fimbriae and curli systems, as well as the quorum sensing and cyclic dimeric guanosine monophosphate signaling pathways. Understanding these mechanisms may provide insights into new targets for the development of therapies to prevent or disrupt biofilm formation in K. pneumoniae [16,17].
The formation of biofilms by ESBL-and KPC-producing K. pneumoniae is a major concern, as it can increase the spread of antibiotic resistance and lead to the persistence of infections in healthcare settings. It is important to understand the mechanisms of biofilm formation and to develop approaches to control and prevent these infections [13].
In conclusion, ESBL-and KPC-producing K. pneumoniae are a main concern in the healthcare setting due to their ability to form biofilms and resist antibiotics. The formation of biofilms by these strains can increase the spread of antibiotic resistance and make it difficult to treat infections. Therefore, understanding the mechanisms of biofilm formation will help to reduce the spread of antibiotic resistance and improve patient outcomes [18].

Biofilm Formation and Biomass Quantification
The bacterial adhesion of all isolates was assessed using a microtitre plate-based assay as previously described with some modifications [15]. Briefly, a few colonies of each isolate were transferred from fresh cultures to tubes with 3 mL of Tryptic Soy Broth (TSB) and incubated at 37 • C for 24 h. Following incubation, the number of cells in each culture was quantified and adjusted to 0.5 McFarland (1.5 × 10 8 CFU/mL), and 100 µL of each bacterial suspension was transported to a 96-well microtiter plate. Pseudomonas aeruginosa ATCC ® 27,853 is a recognized biofilm-forming strain used as a positive control in biofilm assay. Sterile TSB was incorporated as a negative control. The microplates were incubated for 24 h at 37 • C. After incubation, bacterial cells in suspension were removed by turning the microplates over, and they were washed twice with distilled water. This step helps remove stray cells and media components that may be stained in the next step, significantly reducing background staining. The plates were then allowed to dry at room temperature for 15 min. Then, 125 µL of methanol (Scharlau, Barcelona, Spain) was added to each well and incubated for 15 min to fix the biofilm. Methanol was removed, the plates were allowed to dry at room temperature for 10-15 min, and 125 µL of Crystal Violet (CV) at 1% (v/v) (Liofilchem, Roseto degli Abruzzi, Italy) was added to each well. After incubation, the CV solution was removed, and the microplates were washed 3-4 times with distilled water. Subsequently, the plates were vigorously dried on a stack of paper towels to remove all excess cells and stains and were left to dry overnight.
To quantify the biofilm biomass, 125 µL of acetic acid 30% (v/v) was added to each well of the microtiter plate to solubilize the CV. After incubation at room temperature for 10-15 min, optical density was read at 630 nm (OD630 nm) [13] using a microplate reader BioTek ELx808U (BioTek, Winooski, VT, USA). The results were interpreted as weak, moderate, and strong biofilm producers. The optical density cut-off value (ODc) was determined by arithmetically averaging the OD of the negative control wells and adding a standard deviation of +3. Samples with an OD higher than the ODc were considered positive, whereas those with a lower optical density than the cut-off value were considered negative. Strains were classified using the following criteria: OD ≤ ODc non-biofilm producer; ODc < OD ≤ 2 × ODc, weak biofilm producer; 2 × ODc < OD ≤ 4 × ODc, moderate biofilm producer; OD > 4 × ODc, strong biofilm producer [29].

Bacterial Isolates and Identification
This study was conducted on 45 Klebsiella pneumoniae strains, including 15 ESBL producers, 15 KPC producers, and 15 non-β-lactamase-producers. Based on the type of specimen, we have 77.8% isolates from urinary infections, 11.1% from bacteremia episodes, 6.7% from pulmonary infections, and 4.4% from wounds. The distribution of isolates is presented in Figure 1.
significantly reducing background staining. The plates were then allowed to dry at room temperature for 15 min. Then, 125 µL of methanol (Scharlau, Barcelona, Spain) was added to each well and incubated for 15 min to fix the biofilm. Methanol was removed, the plates were allowed to dry at room temperature for 10-15 min, and 125 µL of Crystal Violet (CV) at 1% (v/v) (Liofilchem, Roseto degli Abruzzi, Italy) was added to each well. After incubation, the CV solution was removed, and the microplates were washed 3-4 times with distilled water. Subsequently, the plates were vigorously dried on a stack of paper towels to remove all excess cells and stains and were left to dry overnight.
To quantify the biofilm biomass, 125 µL of acetic acid 30% (v/v) was added to each well of the microtiter plate to solubilize the CV. After incubation at room temperature for 10-15 min, optical density was read at 630 nm (OD630 nm) [13] using a microplate reader BioTek ELx808U (BioTek, Winooski, VT, USA). The results were interpreted as weak, moderate, and strong biofilm producers. The optical density cut-off value (ODc) was determined by arithmetically averaging the OD of the negative control wells and adding a standard deviation of +3. Samples with an OD higher than the ODc were considered positive, whereas those with a lower optical density than the cut-off value were considered negative. Strains were classified using the following criteria: OD ≤ ODc non-biofilm producer; ODc < OD ≤ 2 × ODc, weak biofilm producer; 2 × ODc < OD ≤ 4 × ODc, moderate biofilm producer; OD > 4 × ODc, strong biofilm producer [29].

Bacterial Isolates and Identification
This study was conducted on 45 Klebsiella pneumoniae strains, including 15 ESBL producers, 15 KPC producers, and 15 non-β-lactamase-producers. Based on the type of specimen, we have 77.8% isolates from urinary infections, 11.1% from bacteremia episodes, 6.7% from pulmonary infections, and 4.4% from wounds. The distribution of isolates is presented in Figure 1.
None of the isolates was susceptible to all antibiotics tested, but two of them only showed resistance to ampicillin (HS39 and HS63). The most resistant strains presented resistance to 15 different antibiotics (13.3%; n = 6), and the most susceptible ones showed resistance to only one antibiotic, ampicillin (4.4%; n = 2). Among the most resistant strains, we verified that all of them were KPC-producing K. pneumoniae. This suggests that high rates of resistance to commonly used antimicrobial agents are speculated to be associated with KPC production. Another result that strengthens this suggestion is the absence of either ESBL or KPC enzymes in the strains that showed resistance to only one antibiotic (Table 1). We also verified that some K. pneumoniae isolates had the same phenotype profile. Thus, the most common was AMP-CN with 15.6% (n = 7), followed by AMP-AUG-  (Table 1).

Detection of Resistance Genes
In our study, the bla CTX-M , bla TEM , and bla SHV genes were screened against the isolates resistant to penicillins and cephalosporins (n = 45). In all ESBL strains, we detected at least one of the β-lactamase genes. The bla CTX-M was detected in 30 isolates (66.7%), bla TEM in 24 isolates (53.3%), and bla SHV in 43 isolates (95.6%). In the majority of the isolates, these genes were present in combinations between them, with bla CTX-M + bla SHV being the most frequently found among all samples (n = 17, 37.8%), followed by bla TEM + bla SHV (n = 12, 26.7%) and bla CTX-M + bla TEM + bla SHV (n = 12, 26.7%). The bla TEM alone was not detected, bla CTX-M alone was verified in one isolate, and bla SHV alone was present in two samples. Only in one isolate (HS39) was the amplification of these genes was not verified (Table 1).
Regarding aminoglycoside resistance genes, we tested three different genes: aac(3)-II, aac(3)-IV, and aadA1. The aac(3)-IV was not detected. The aac(3)-II gene was detected in 27 strains, and the aadA1 gene was detected in 32 strains. We verified that in the 27 strains where we detected the aac(3)-II gene, the aadA1 gene was also present; however, we verified in 5 strains the presence of the aadA1 gene alone. In only one strain (HS125), none of the genes tested was detected ( Table 1).
The tetA and tetB genes were tested in the K. pneumoniae samples that were resistant to tetracycline. The tetA gene was detected in four isolates, and tetB was not detected in the tetracycline-resistant samples. In four tetracycline-resistant samples, neither of the genes was detected ( Table 1).
The amplification of the genes conferring resistance to chloramphenicol tested in this study was cmlA, floR, and catA. A total of 20 of the isolates showed resistance to this antibiotic, and we could only detect the presence of the cmlA and floR in one strain (HS97). In the rest of them, none of the genes were amplified ( Table 1).
The sul1, sul2, sul3, and dfrA genes were tested in the trimethoprim-sulfamethoxazoleresistant strains. The sul1 was detected in 5 isolates, the sul2 gene was detected in 24 isolates, and the sul3 gene was detected only in 1 isolate. Only two strains showed combinations of Antibiotics 2023, 12, 1143 7 of 13 these genes: sul1 and sul2 were verified in the isolate HS154, and sul2 and sul3 were present in the isolate HS97. The dfrA gene was not detected in any isolate, and in three of them (HS102, HS72, and HS85), none of these genes were detected ( Table 1).
The gyrA, gyrB, and parC genes, target genes for mutations in the quinolone resistancedetermining regions, were detected in some of the isolates. The gyrA gene was verified in 9 strains, the gyrB gene in 22 strains, and the parC in 8 strains. Further investigation is needed to identify the specific mutations that confer resistance to the quinolones antibiotics (Table 1).
Regarding the genes responsible for conferring resistance to carbapenem antibiotics, we tested the bla KPC , bla NDM , bla OXA , bla OXA-48 , bla IMP , bla VIM , bla VIM-2 , bla SPM , and bla PER genes only in the KPC-producing strains. So, the presence of the bla KPC gene in all KPCproducing isolates confirms the phenotypic detection test. Concerning the other genes, we only detected the bla OXA gene in 11 isolates, and the bla PER gene was present in one isolate (HS160) ( Table 1).

Biofilm Producers
Non

Biofilm Production among Clinical Specimen
Among the different clinical specimen isolates, we could verify that 80% of the isolates from urinary infections were able to produce biofilms, the majority being weak producers (n = 21). The same percentage was demonstrated in the bacteremia isolates with the same amount of weak (n = 2) as moderate biofilm producers (n = 2). All of the pulmonary infection isolates were able to produce biofilms (n = 3), being classified as weak biofilm producers. One of the two wound isolates was capable of producing biofilm, this being a moderate biofilm producer (Table 3).

Discussion
This study was carried out to evaluate the ability of ESBL-producing and KPCproducing Klebsiella pneumoniae strains, obtained at the hospital center of Trás-os-Montes and Alto Douro, to form stronger biofilms than K. pneumoniae strains without these enzymes. It is important to note that due to privacy and ethical constraints, patient data, including clinical information, were not accessible for this study. As a result, an analysis investigating the relationship between the bacteriological characteristics of the K. pneumoniae strains and the patients was not feasible. Therefore, the comprehensive understanding of the findings may be restricted by the absence of patient data. Due to their capacity to generate β-lactamase enzymes and biofilms, Klebsiella pneumoniae shows resistance to numerous antibiotics [30,31]. Since many β-lactamase genes are located on mobile genetic elements controlled by plasmids, the resistant strains are spreading quickly, causing elevated death rates, illness, and healthcare expenses [32]. The simultaneous expression of multiple β-lactamase genes in an organism can worsen the drug-resistance problem, reducing available treatment options [33]. Hence, identifying these factors and their connection to drug resistance is crucial in diagnostic labs.
Klebsiella spp. are microorganisms that can cause multiple infections, such as urinary tract infections, pneumonia, and blood and wound infections. In this work, we isolated 45 K. pneumoniae, the majority from urinary infections (77.8%), followed by bacteremia episodes (11.1%), pulmonary infections (6.7%), and wounds (4.4%). This disparity could be attributed to the larger number of urine samples collected in comparison to other clinical specimens. Other studies have also been able to isolate most of their microorganism from urine samples [13,34].
Among the seventeen antibiotics used to test antibiotic susceptibility of K. pneumoniae, ampicillin had the highest percentage of resistance (100%). These findings agree with the work of Lagha et al. [35], who found K. pneumoniae isolates 100% resistant to ampicillin. Followed by ampicillin, we found trimethoprim-sulfamethoxazole, and cefotaxime with 68.9% and 66.7% resistance, respectively. The work of Shadkan et al. [36], in which most of the isolates were resistant to trimethoprim-sulfamethoxazole (52%), and cefotaxime (51%), corroborates our findings. Among the highest rates of resistance, we also reported 66.7% resistance to aztreonam as well as to amoxicillin-clavulanic acid. Not many studies reported increased resistance to aztreonam, and a systematic review and meta-analysis research conducted by Heidary et al. [37] showed that, in Iran, there is a high prevalence of drugresistant K. pneumoniae isolates, with the highest rate of resistance against ampicillin (82.2%), aztreonam (55.4%), and nitrofurantoin (54.5%). Some studies conducted on K. pneumoniae have also detected a high prevalence of resistance to amoxicillin-clavulanic acid, such as Kuinkel et al. [13] in 2021, which reported 59.6% of resistance, and Pishtiwan et al. [38], which detected a similar percentage to our own (65%). The latter also reported 100% of susceptibility to amikacin, which is in line with our results for amikacin, which has 84.4% of susceptibility. Among the antibiotics used, tetracycline was also found to be effective (75.6%); however, in a study conducted in Iran, strains isolated from children showed the highest resistance to tetracycline (71.5%), whereas the lowest rate was associated with cefepime (12.7%), imipenem (6%), and gentamicin (6%) [39].
Concerning the detection of resistance genes, ESBL-producing K. pneumoniae carrying bla CTX-M , bla TEM , and bla SHV genes have been found in clinical samples [40,41]. Similar percentages of combinations between these genes were also found in a study conducted in Egypt [42]. The HS39 isolate only showed resistance to ampicillin, but none of the genes tested was amplified. There are other mechanisms of resistance to ampicillin in K. pneumoniae, one possibility is the overexpression of the chromosomal ampC gene [43], and another is the presence of non-β-lactam mechanisms of resistance, such as efflux pumps [44] or modifications of penicillin-binding proteins [45].
Regarding aminoglycoside resistance genes, we only detected aac(3)-II and aadA1 genes in our aminoglycoside-resistant strains. Similar results were reported by Mbelle et al. [40], where, besides these two genes, they also detected aac (6 )−Ib, aacA4, aadA2, aadA5, aadA16, aph(3 )−Ia, strA, and strB genes. Concerning the sample HS125, which was negative for the genes tested, we could find some possibilities why the isolate was negative. For example, the resistance mechanism may involve other genes or mechanisms that we did not test for. There are many other genes that can confer resistance to aminoglycosides, such as aac(6 )-Ib [40], ant(2 )-Ia [46], and aph(3 )-IIIa [47], which encode for aminoglycoside-modifying enzymes that can modify the structure of aminoglycoside antibiotics, such as amikacin and gentamicin, leading to resistance. Another possibility is the strain may have mutations in the target site of the aminoglycosides, such as 16S rRNA or ribosomal proteins, which can also confer resistance, or also the isolate may have acquired the resistance through a non-genetic mechanism, such as through the production of biofilm or the presence of efflux pumps, which can prevent the drug from reaching its target site [48]. Thus, further research will need to be conducted in order to find out the mechanism underlying aminoglycoside resistance.
Relative to tetracycline resistance genes, we only detected the tetA gene in four isolates, and tetB was not detected. This result was unexpected, since both genes are commonly detected in clinical samples of K. pneumoniae [49][50][51]. As mentioned earlier, in some tetracycline-resistant isolates, we did not detect either tetA or tetB genes. There are some possible reasons for that, including the acquisition of other tetracycline resistance genes, such as tetC, tetD, or tetG [52], or the developed tetracycline resistance due to mutations in the bacterial ribosome or other cellular components without specific resistance genes [53].
It was unexpected that most of the chloramphenicol-resistant strains were negative for the cmlA, floR, and catA genes, since these are commonly found in clinical strains of K. pneumoniae [54]. However, the presence of other chloramphenicol-resistance genes, such as catB [55], responsible for enzymatic inactivation, and cml, responsible for encoding a chloramphenicol efflux pump that expels chloramphenicol from the cell, or the presence of mutations in the target site of the drug [56], are two of the possible reasons to justify chloramphenicol resistance present in our samples.
To demonstrate the trimethoprim-sulfamethoxazole resistance, we screened the isolates for sul1, sul2, sul3, and dfrA genes. The sul2 gene was the most frequently detected. The same was also reported by Mbelle et al. [40], where they identified sul2 in 86% of the isolates and sul1 in 78% of the isolates. In addition to dfrA being commonly detected in K. pneumoniae strains [40], in our study, it was not detected in any isolate. This could be due to the fact that the strains may have acquired resistance to trimethoprim-sulfamethoxazole through other mechanisms, such as mutations in other genes involved in the folate biosynthesis pathway or through efflux pumps [57]. Given the three isolates that were negative for the trimethoprim-sulfamethoxazole resistance genes evaluated in this work, it is possible that other genes could be conferring resistance to trimethoprim-sulfamethoxazole such as dfrD, dfrG, dfrK, or sul4 [58][59][60]. Other mechanisms of resistance to trimethoprimsulfamethoxazole that do not involve the genes listed above may also confer antibiotic resistance, such as alterations in drug uptake or metabolism.
Not detecting gyrA, gyrB, and parC in some quinolone-resistant strains was unforeseen. One possible reason for this is that the primers used for PCR may not have been annealing properly to the target genes due to genetic variation or mutations in the gene sequence, or some strains may have deletions or other genetic alterations in these genes, leading to their absence in the genome.
In this study, we detected the presence of the bla KPC gene in all KPC-producing isolates, confirming the phenotypic detection test [61][62][63]. The bla PER gene is not commonly detected in K. pneumoniae strains. However, there have been reports of bla PER -producing K. pneumoniae strains in some countries, such as Iran and Brazil [64,65]. Nevertheless, as far as we know, this is the first report of the bla PER in Klebsiella pneumoniae. Sequencing this isolate is the next step to fully confirm the gene's presence.
Regarding biofilm production by K. pneumoniae strains, we verified that 80% of them were confirmed as biofilm producers. Other studies also reported high rates of biofilm production, such as Türkel et al. [66], who reported that 99% of K. pneumoniae isolates were biofilm producers; Seifi et al. [34], who reported 93.6%; and Shadkam et al. [36], who reported 75%. Among the ESBL-producing and KPC-producing K. pneumoniae, we verified higher rates of weak biofilm producers, 40.0% and 60.0%, respectively. These results were surprising since the majority of the literature reported more strong biofilm producers among β-lactamase producers [67,68]. However, other studies also reported similar findings to ours [13]. Poovendran et al. [69] found that ESBL-producing strains highly form a biofilm compared with non-ESBL producers. This study did not find any correlation, as the majority of non-ESBL-producing strains were capable of forming a biofilm. This is in accordance with Hasan et al. [70], who reported no correlation between ESBL-and non-ESBL-producing bacteria and their capacity to form biofilms.

Conclusions
The emergence of Klebsiella pneumoniae strains that produce ESBL and KPC has become a significant public health concern. The ability of ESBL-and KPC-producing K. pneumoniae to form biofilms is worrisome, as it can facilitate the transmission of antibiotic resistance and prolong infections in healthcare settings. In this study, most of the bacterial strains were found to be multidrug-resistant, with 100% resistance to ampicillin. The most effective antibiotics were also tetracycline and amikacin.
In all ESBL-producing strains, we detected at least one β-lactamase resistance gene, and all KPC-producing isolates had the bla KPC gene, which confirms the phenotypic detection test. We also detected the bla PER gene in one KPC-producing isolate. To the best of the author's knowledge, this is the first time that this gene has been reported in Portugal.
Most of the K. pneumoniae strains isolated from hospitalized patients have the capacity for biofilm production. Still, we did not verify an ability to form stronger biofilms by ESBL-producing and KPC-producing Klebsiella pneumoniae strains.
Further research is needed to be conducted to better understand the mechanisms involved in the biofilm formation of these isolates by sequencing the strain's genome and detecting the resistance genes and the virulence genes involved in biofilm production.