Characterization of bla_TEM and bla_CTX-M beta-lactam resistance genes in chronic rhinosinusitis

Abstract

Background

Chronic sinusitis is one of the most challenging health problems of contemporary society. Although several treatment methods have been defined, a comprehensive understanding of the underlying causes (e.g., antibiotic resistance) is still elusive. The aim of this study was to characterize two of the main extended-spectrum beta-lactamase genes—i.e., bla_TEM and bla_CTX-M genes—and investigate antimicrobial resistance in bacteria isolated from chronic sinusitis. Samples from 70 chronic sinusitis patients and 20 healthy individuals (controls) were analyzed for the presence of bla_TEM and bla_CTX-M resistance genes using the polymerase chain reaction (PCR) test, followed by gene sequence analysis.

Results

Phenotypic and genotypic beta-lactam resistance was observed in 58.7% and 61.54% of the gram-negative isolates, respectively, with 38.46% carrying the bla_TEM gene and 34.62% harboring the bla_CTX-M gene. Sequencing data indicated high heterogeneity in bla_CTX-M genes (69–100% similarity to reported sequences) and lower heterogeneity in bla_TEM genes (93–99%).

Conclusion

Broad-spectrum beta-lactam resistance is a major pathogenesis factor in chronic rhinosinusitis, and careful consideration is required for antimicrobial therapy. High bla_CTX-M heterogeneity could mean high horizontal transfer rate of this gene and warrant a surveillance program.

Background

Sinusitis is the inflammation of the mucosal tissue lining the sinuses (hollow cavities in the facial bones), and chronic bacterial sinusitis is one of the most common health problems in the contemporary society. Microbiology of chronic bacterial sinusitis has been explained extensively, encompassing wide variations in terms of geographical areas, patients, laboratory methodologies and sampling methods [1,2,3].

The nasopharynx is colonized by normal bacterial flora, including potential pathogens. They can spread from the nasopharynx to the sinus cavity during viral respiratory infections and cause various respiratory tract infections, including sinusitis [4]. Emergence of antimicrobial resistance in organisms responsible for chronic sinusitis has called into question the efficacy of antibiotic treatment of these infections [5, 6]. Furthermore, the study of recent trends in antimicrobial resistance of bacterial species involved in rhinosinusitis has revealed that the prevalence of extended-spectrum beta-lactamase (ESBL)-producing strains in these infections is on the rise [7]. Therefore, careful consideration of the bacteria involved prior to adopting treatment regimen is paramount [8], especially in the case of beta-lactam and extended-spectrum beta-lactam resistant bacteria in chronic sinusitis, which have further complicated the antimicrobial therapy process [9].

Beta-lactamases are enzymes produced by bacteria, which create resistance to beta-lactam antibiotics, such as penicillins, cephamycins and carbapenems. From a functional viewpoint, three beta-lactamase enzyme groups (groups 1, 2 and 3) have been defined, based on their inhibitor profiles and substrates [10]. In another classification scheme, beta-lactamase proteins have been classified into four classes (classes A, B, C and D) based on amino acid sequences [11]. ESBLs are enzymes that can neutralize extended-spectrum antibiotics such as extended-spectrum penicillins and third-generation cephalosporins as well as aztreonam [12]. They can, however, be overcome by some beta-lactamase inhibitors such as clavulanic acid, tazobactam and sulbactam. ESBL genes are believed to have originated from narrow-spectrum beta-lactamase genes through mutation, and they have managed to spread globally by means of mobile genetic elements, while constantly taking new forms. Two of the most important ESBL genes are the classic bla_TEM gene and the more recently spreading bla_CTX-M gene (both class A beta-lactamases) [13, 14].

Like most respiratory tract infections, sinusitis occurs in stages. The first stage is the viral phase, which ends in complete recovery of most patients after one to two weeks [15]. However, in some cases, the viral phase is followed by acute bacterial infection, and delayed or unsuccessful treatment of acute sinusitis can lead to chronic sinusitis [4, 15].

Although the mechanism of inflammation associated with chronic sinusitis is still controversial, the presence of bacteria within the sinuses has been well documented [2].

Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis and Streptococcus pyogenes have been most commonly isolated in cases of acute bacterial sinusitis. In case of chronic rhinosinusitis, the most common bacterial isolates are Staphylococcus aureus and anaerobic organisms such as Fusobacterium, Prevotella, Porphyromonas and Peptostreptococcus spp. Pseudomonas aeruginosa and other aerobic and facultative gram-negative rods have also been recovered from patients with underlying health problems. Unfortunately, antibiotic resistance is prevalent among these aerobic and anaerobic bacteria, and a large proportion of them (over a third) produce beta-lactamases. Moreover, most Staphylococcus aureus isolates involved in these infections are of the Methicillin-resistant (MRSA) type [2, 4, 8, 16].

Factors affecting the microbiology of sinusitis may include vaccination, history of antimicrobial therapy and inhibiting effect of the normal flora [17]. Understanding the characteristics and antimicrobial resistance/susceptibilities of the organisms responsible for chronic sinusitis is crucial to the success of antimicrobial therapy and must be considered when selecting the antimicrobial regimen [2].

Having established the significant role of gram-negative bacteria in sinus infections, and the important role of extended-spectrum beta-lactam resistance in the pathogenesis of gram-negative bacteria, we aimed this study to investigate the beta-lactam resistance profile of bacterial strains isolated from patients with chronic sinusitis, by determining the prevalence and characterizing two important ESBL genes, namely bla_TEM and bla_CTX-M [11, 18].

Materials and methods

This prospective study was conducted during the period 2018–2019, on samples collected from 90 individuals, out of which 70 were admitted to a large public hospital in Tehran, Iran, suffering from chronic sinusitis. The Ethical Code for this study is IR.IUMS.REC.1395.28467, and all participants were aware of and consented to entering this study and publication of its data. Sinusitis was diagnosed by specialist doctors, based on clinical findings, radiographic imaging (Rx) and CT scan. The other 20 participants had no sinusitis symptoms.

Samples were collected during endoscopy, and the specimens were sealed and immediately transferred to the laboratory under prescribed conditions [19]. Individuals who had received antibiotics in the three weeks preceding sample collection were excluded from the study. Specimens were cultured on blood agar, on enriched Schaedler agar and in thioglycolate broth (enriched with yeast extract, vitamin K and hemin) media. After incubation, we carried out gram staining, standard identification tests and antimicrobial susceptibility tests on isolates as previously described [2]. The antibiotics used in the antimicrobial susceptibility tests were obtained from Mast, England (MASTDISCS™ AST), and included the following disks as previously described [20]: ciprofloxacin (5ug), gentamicin (10ug), penicillin (10U), ampicillin (10ug), clindamycin (2ug), vancomycin (30ug), tetracycline (30ug), erythromycin (15ug), cotrimoxazole (1.25/23.75ug), amikacin (30ug), amoxicillin (20ug), oxacillin (1ug), imipenem (10ug), ceftazidime (30ug), meropenem (10ug), cefotaxime (30ug), piperacillin–tazobactam (30/6ug), colistin (10ug) and ampicillin–sulbactam (10/10ug). Antibiotic susceptibility test results were interpreted and reported using CSLI guidelines (CSLI M021, 2020).

DNA was extracted using the boiling method from all the gram-negative bacterial strains isolated in the previous steps, then stored in double distilled water at − 4 °C. To check the quality of the extracted DNA, the 260/280 nm absorbance ratio was determined (using a Thermo Scientific™ NanoDrop™ 2000 spectrophotometer) prior to storage [21].

All 26 gram-negative isolates were checked for the presence of bla_TEM and bla_CTX-M genes using the PCR technique. Staphylococcus aureus ATCC 29213 was used as the blank control strain. Primer sets used in this study as described previously are provided in Table 1. PCR procedures and mixtures were used as previously described [22, 23]. PCR products were separated using a standard 1.5% agarose gel electrophoresis procedure.

Table 1 Primers sets used for detection of bla_TEM and bla_CTX-M genes

From the PCR products positive for either bla_TEM or bla_CTX-M, 25–50 µl was sent to Microsynth Company, Switzerland, for DNA sequencing (via Topaz Gene Kavosh Company, Karaj, Alborz Province, Iran). DNA sequencing data was checked for the most similar sequences in online databases, using the online NCBI Blast tool. All data were analyzed by SPSS 23.

Sequences obtained were submitted to GenBank, with the following accession numbers: OP374105, OP374106, OP374107, OP374108, OP374109, OP374110, OP374111, OP424993, OP424994 and OP424995.

Results

Bacteriology

Of all the samples collected in this study from 90 people (70 patients with sinusitis and 20 healthy individuals), 74 yielded positive cultures. Fifty-seven (77.03%) of the cultures had a single organism, while 17 (22.97%) contained multiple organisms. A total of 97 different isolates were obtained from these cultures.

Samples collected from individuals with no sinusitis symptoms were mostly sterile (80%), and in 4 cases, Staphylococcus saprophyticus and Staphylococcus aureus were isolated. The most significant bacteria isolated from sinusitis patients are provided in Table 2.

Table 2 Most significant bacteria isolated from sinusitis patients and the number of isolates in 70 patients with sinusitis

Antimicrobial susceptibility test

The highlights of antibiotic susceptibility tests results are as follows:

All group D Streptococci isolates were resistant to clindamycin, and all were susceptible to gentamicin. Penicillin and vancomycin were observed to be equally effective against these isolates, with 42.8% of the isolates showing resistance to each of these antibiotics.

Corynebacteium isolates demonstrated the most resistance to ciprofloxacin (69.2% were resistant) and the most susceptibility to clindamycin and tetracycline (0% were resistant).

Staphylococcus aureus isolates were observed to be 100% resistant to penicillin. The most effective antibiotic against these isolates was ciprofloxacin, with 16.6% of the isolates showing resistance.

Escherichia coli isolates demonstrated mild resistance to some beta-lactam antibiotics (18.1% were resistant to ceftazidime and 27.3% to meropenem), but no resistance to other carbapenems or aminoglycosides.

In case of Staphylococcus saprophyticus, 100% of the isolates were resistant to penicillin and 90.9% were resistant to erythromycin. Ciprofloxacin, ampicillin and cotrimoxazole were observed to be fully (100%) effective against these isolates.

Klebsiella isolates showed the most resistance to beta-lactam antibiotics among all the isolates (87.5% resistant to ampicillin, 75% resistant to cephalosporins and 50% resistant to both imipenem and meropenem), but no resistance to beta-lactams in combination with beta-lactamase inhibitors, such as piperacillin–tazobactam and ampicillin–sulbactam (0% resistant).

Pseudomonas isolates demonstrated some resistance to gentamicin, ceftazidime, colistin and meropenem in equal measure (all 42.8%), but they were all susceptible to other beta-lactam antibiotics.

Enterococcus isolates were observed to be most resistant to erythromycin and clindamycin (both 100%) and least resistant to penicillin and gentamicin (both 0%). Isolates from other bacteria showed varied resistance patterns (results not reported).

Overall, the most drug resistance was observed in Pseudomonas isolates (with 100% of the isolates showing multi-drug resistance), and the most beta-lactam resistance was observed in Klebsiella isolates.

PCR

Out of the 26 isolates selected for molecular assessment of beta-lactamase genes, 9 contained the bla_CTX-M gene, 10 carried the bla_TEM resistance gene, and 3 harbored both the bla_CTX-M and the bla_TEM genes. The results are illustrated in Table 3. PCR products were separated using 1.5% agarose gel electrophoresis, as depicted in Fig. 1.

Table 3 Prevalence of the bla_TEM and bla_CTX-M resistance genes in gram-negative bacterial isolates from 70 patients with sinusitis

Fig. 1

bla_TEM (left image) and bla_CTX-M (right image) genes PCR products. Image on the left: bla_TEM gene PCR products; lane 0 is the ladder; lane 1 is the positive controls; lane 6 is the negative control; lanes 2–12 (except lane 6) are sample results. Image on the right: bla_CTX-M gene PCR products; lane 0 is the ladder; lane 1 is the positive control; lane 2 is the negative control; lanes 3–12 are sample results

DNA sequencing

Nineteen isolates containing at least one of the two beta-lactam resistance genes were selected for DNA sequencing. These included 10 isolates with the bla_TEM gene and 9 isolates with the bla_CTX-M gene.

Our bla_TEM gene sequencing data analysis demonstrated a similarity range of 93% to 99% with sequences reported from India, China, Japan, Switzerland and the USA. The bla_TEM gene obtained from Escherichia coli showed the most variation from bla_TEM gene sequences present in GenBank database (93% similarity), while sequences from other isolates all showed the same 99% similarity. For the bla_CTX-M, the similarity range was observed to be between 69 and 100% with sequences reported from Spain, the USA, China, the UK, Germany, Colombia, Brazil, Singapore, Norway, Sweden, South Korea, Australia, Switzerland, Canada and France.

The bla_CTX-M gene sequence obtained from our Klebsiella isolates showed the least similarity (69%) with sequences reported from other countries, while the sequences from Escherichia coli had 100% similarity with sequences reported from the UK.

Discussion

The study of the molecular mechanisms of antimicrobial resistance in bacterial rhinosinusitis has not received due attention in recent years, even though it is of significant clinical importance. To our knowledge, this study is one of the very few of its kind in recent years. However, findings from various phenotypical studies suggest great diversity in bacterial causative agents of sinusitis, in terms of both genus and species variation, and frequency of isolation. Bacterial agents most commonly reported in association with sinusitis include Haemophilus influenzae, Streptococcus pneumonia, Staphylococcus species, Streptococcus species and Corynebacterium species [3, 6, 8, 15, 24].

Other reported pathogens include Moraxella catarrhalis, Escherichia coli, Klebsiella Pneumonia and anaerobic bacteria, to a lesser extent [3, 9, 16]. In our study, the most commonly isolated bacteria were Group D Streptococci, Corynebacterium spp., Staphylococcus aureus, Escherichia coli, Staphylococcus saprophyticus, Kebsiella spp. and Pseudomonas genus. These results are in line with some previous studies [2, 3, 8], but not so with a number of other studies [16, 25]. This could be largely attributed to varied bacterial populations in different geographical areas, as well as differences in sample sizes, and sample collection and culturing methods [2].

Resistance to beta-lactam antibiotics was observed to be very significant, with 58.7% of all the isolates showing resistance to at least one beta-lactam antibiotic in the disk diffusion test. Similar results have been reported by other research groups in Iran [16, 25]. While phenotypic resistance to beta-lactam antibiotics in gram-negative bacteria was also similar to the above figure (57.69% resistant to at least one beta-lactam antibiotic), the molecular assessment (i.e., beta-lactamase genes PCR test) indicated that beta-lactam resistance genes are slightly more prevalent than detectable by phenotypic methods (61.54% of the isolates harboring at least one beta-lactamase gene). Other factors might also affect the expression of resistance genes, which bear further investigation [18]. No resistance was observed to amikacin, piperacillin–tazobactam and ampicillin–sulbactam, which confirms the accepted practice of prescribing beta-lactam antibiotics in conjunction with beta-lactamase inhibitors as the first-line therapy for bacterial sinusitis [15].

Of the two beta-lactamase genes investigated in this study, the bla_TEM gene had a slightly higher prevalence (38.46%) than the bla_CTX-M gene (34.62%). This may help explain the poor resistance of our isolates to cefotaxime [26]. The relative prevalence of these two genes is very important from an epidemiological view point; Reports from different parts of the world indicate that the prevalence of the bla_CTX-M gene, which was once markedly less than the bla_TEM gene, is now on the rise [27,28,29]. If a gradual increase be observed in the relative prevalence of the more recent ESBL gene (i.e., the bla_CTX-M gene) in future studies, we could conclude that the bacterial infections might be originating from source outside our region. As mentioned earlier, however, although many research teams have investigated the bacteriology and antimicrobial resistance patterns in chronic and acute sinusitis, their studies have mainly been concerned with the phenotypic manifestation of beta-lactam resistance, and up-to-date literature on the prevalence of beta-lactamase genes among bacterial isolates from sinusitis patients is very limited.

Regardless of the species of bacteria harboring beta-lactamase genes, resistance to beta-lactam antibiotics could occur in multi-bacterial infections through the shielding phenomenon, if even only one species carries a beta-lactamase gene [2]. Therefore, it is prudent to consider using beta-lactamase inhibition in all chronic bacterial sinus infections.

Although a number of research groups have investigated the prevalence of bla_TEM beta-lactamase genes in H. influenzae and other sinusitis-related bacteria isolated from various infections, no recent studies have been conducted to determine the prevalence of this gene in other causative agents of chronic sinusitis at a molecular level [30,31,32,33,34]. These studies reported the prevalence of bla_TEM genes in beta-lactamase-producing isolates from various sources to be from 53.7 to 93.7% in different countries, which is higher than our results (38.46%). Hara et al. suggested that beta-lactamase genes could be transferred from one sinusitis agent to another [30], therefore, it is imperative to investigate the presence of these genes in all bacterial agents of sinusitis, especially in multi-bacterial chronic sinusitis.

Similarly, contemporary research on the prevalence of bla_CTX-M gene in bacterial agents of sinusitis is even more limited than the bla_TEM gene. This could be attributed to the fewer number of sinusitis-related isolates reported to harbor this beta-lactamase gene. Bajpai et al. reported the prevalence of bla_CTX-M beta-lactamase gene in isolates from urinary tract infections to be 7.6% [13]. In contrast, another study by Rajivgandhi et al. reported bla_CTX-M to be the predominant beta-lactamase gene (53%) in gram-negative isolates from urinary tract samples in India [22], while in our study, 34.62% of the beta-lactamase-producing bacterial isolates carried the bla_CTX-M gene. These varied results could be due to different geographical distribution of beta-lactamase genes, as well as different infection sources.

Our sequencing data analysis indicates that the bla_CTX-M is the beta-lactamase gene with the most variation (69% to 100% similarity with other reported sequences), and it is most similar to bla_CTX-M sequences reported from the UK. This heterogeneity may suggest a greater rate of horizontal transfer as compared to the more conserved bla_TEM beta-lactamase gene (93–99% similarity with sequences reported from other countries). As a growing global concern in terms of extended-spectrum beta-lactamase, prevalence and variations of bla_CTX-M genes need to be constantly monitored [13, 35]. The heterogeneity that we have observed in this gene’s sequence could be a point of concern, as it might lead to more potent mutant beta-lactamases, and emerging new variations of the gene may escape detection by existing molecular methods.

Conclusion

The presence of antibiotic resistant bacteria in sinuses of patients with chronic sinusitis is indicative of the importance of effective antimicrobial therapy to overcome this major health problem. Our findings indicate that beta-lactam resistance patterns in our samples are in line with existing resistance patterns in other countries, and indeed have the same origin. Although the accepted combination of a beta-lactam antibiotic and a beta-lactamase inhibitor (such as amoxicillin–clavulanic acid) is evidently still effective as the first-line therapy, vigilance programs need to be put in place in all Iranian hospitals to monitor its continued effectiveness, as new resistance mechanisms are emerging in other parts of the world, which will inevitably affect us.