Oral Colonization by Candida Species and Methicillin-Resistant Staphylococcus aureus in Patients with Malignant and Potentially Malignant Lesions

2.1 Study population

This was a cross-sectional descriptive observational study to evaluate the presence of oral cavity colonization by Candida spp. and the coexistence of co-colonization with Staphylococcus aureus strains, including MRSA, in cancer patients. The population of the investigation group was made up of 105 participants. These patients were treated in cancer centers that were registered at Baghdad Medical College from January 2023 to December 2023. Participants in the study were adults (>24 years) with a confirmed diagnosis of cancer who provided written informed consent and attended the participating oncology centers between January and December 2023. Eligible participants had completed their planned cancer treatment (surgery, chemotherapy, radiotherapy, or combined therapy) before oral sample collection and were enrolled during routine follow-up visits.

2.2 Sample collection for microbiological analysis

Sampling targeted both clinically evident lesions and adjacent oral mucosal surfaces. Sterile swabs were used to sample four intraoral locations, including the hard palate, buccal mucosa, dorsum of the tongue, and floor of the mouth, in a Z-pattern movement. Participants were instructed not to eat, drink, smoke, brush their teeth, or use mouthwash for at least 2 h before sample collection. Swabs were combined with 2 mL of sterile saline per patient and transported at 2–8 ℃ for laboratory analysis within 2 h of collection and processed immediately upon arrival.

To isolate Staphylococcus aureus, aliquots of the combined samples from swabs taken from the hard palate, buccal mucosa, dorsum of the tongue, and floor of the mouth were plated on blood agar and mannitol salt agar (MSA). The bacteria were allowed to grow for a period of time under standard growth conditions for bacteria, and those that resembled Staphylococcus aureus based on their appearance and biochemical characteristics were identified as such.

2.3 Determination of the colony-forming unit of Candida

To measure the colony-forming unit (CFU) of Candida, 500 µL of the combined sample was diluted in 4.5 mL of sterile saline. Serial dilutions (10⁻¹ to 10⁻³) were plated onto Sabouraud dextrose agar (40 g/L dextrose, 10 g/L peptone, and 15 g/L agar) supplemented with chloramphenicol, and the culture was incubated at 37 ℃ for 48 h. Typical colonies of Candida (opaque, creamy, white to beige, 0.5–20 mm) were counted, and the CFU was determined by the following formula [21]:

$C F U=$ number of colonies $\times$ dilution factor $\times 10$

2.4 Identification of Candida spp.

At least 10% of the Candida-like morphology colonies were stained with lactophenol blue and viewed under a microscope. Round-to-oval blastoconidia with budding were some of the diagnostic features used for characterization. Confirmed colonies were subcultured on 2% SDA at 37 ℃, and then the culture was incubated at 37 ℃ for 48 h.

2.5 Species-level characterization of Candida

The RAPIDYeast Plus system (Thermo Scientific) was then used to perform species-level identification, following the instructions of the manufacturer. Candida spp. identification was confirmed using the Electronic RapID Coded Compendium (ERIC) database criteria, including identification level (adequate), frequency (acceptable), bioscore (average 1/100), and probability values of 98% or higher. All results were recorded in duplicate to ensure reliability and reproducibility. Also, isolates carrying mecA and/or mecC genes were classified as MRSA regardless of cefoxitin susceptibility results. Cefoxitin resistance was additionally used as a phenotypic marker for methicillin resistance.

2.6 Characterization of Staphylococcus aureus colonization

Blood agar and MSA were used for the culture, and the plates were incubated at 37 ℃ for 24 h. The colonies were large, circular, creamy, opaque colonies with a white to golden-yellow color and characteristic β-hemolysis on blood agar. Production of yellow colonies on MSA with a yellow halo zone around the colonies confirmed positive fermentation of mannitol. To identify Staphylococcus aureus, a coagulase test was conducted. The colonies from MSA were transferred to brain heart infusion (BHI) broth and allowed to grow at 37 ℃ for 24 h. Citrated plasma (0.5 mL) was added to the culture and incubated. The culture was monitored for clotting at 2 h, 4 h, and 18 h. Colonies on nutrient agar were suspended in sterile physiological saline, and the turbidity was adjusted to the 0.5 McFarland standard (1.5 × 10⁸ CFU/mL) with a nephelometer.

2.7 Antimicrobial susceptibility testing

The disk diffusion test was used to determine the susceptibility of Staphylococcus aureus isolates on Mueller-Hinton agar following the Clinical and Laboratory Standards Institute (CLSI) guidelines. The bacterial suspension was spread, then antibiotic discs were placed in three directions, with erythromycin and clindamycin discs spaced by 16–24 mm. Plates were incubated at 35 ℃ for 16–18 h, and the interpretation of inhibition zones was done according to the CLSI standards (Table 1).

Table 1. List of the antimicrobials used and susceptibility criteria

2.8 Determination of β-lactamase production

A penicillin G disc was used to evaluate the production of β-lactamase according to the CLSI guidelines. Isolates with no inhibition zone or with a sharp-edged halo were categorized as β-lactamase producers. In contrast, those with a diffuse, “beach-shaped” halo were considered negative for β-lactamase production [22].

2.9 Detection of inducible clindamycin resistance in Staphylococcus aureus

Inducible clindamycin resistance was detected by placing erythromycin and clindamycin discs 16–24 mm apart (center-to-center) on Mueller-Hinton agar and incubating the culture for 16–18 h. Isolates exhibiting a flattened, D-shaped zone of inhibition adjacent to the erythromycin disc around the clindamycin disc were classified as having inducible clindamycin resistance [22].

2.10 DNA extraction by heat shock

The heat shock technique was used to extract genomic DNA. Nutrient agar was used to culture Staphylococcus aureus over a period of 20 h. About 3–6 colonies were vortexed in 150 μL of molecular-grade water in a 1.5 mL Eppendorf tube. Heat shock was then applied to the suspension by boiling it in boiling water for 10 minutes and then freezing at -20 ℃ for 10 minutes to lyse the bacterial cells. Samples were centrifuged at 10,000 rpm for 5 minutes at room temperature. A 50 µL aliquot of DNA extract was retained as stock DNA, from which 0.7 µL was used as template in each PCR reaction [23].

2.11 Evaluation of DNA quality

Purity and concentration of the extracted DNA were measured on a Thermo Scientific NanoDrop Lite spectrophotometer. One microliter of molecular-grade water was first used to blank the instrument. Subsequently, 1 µL of each vortexed DNA sample was loaded onto the pedestal, and measurements were initiated using the “Measure” function. Absorbance was read at 260 nm and 280 nm, and the concentration of the DNA was calculated in ng/μL. A sample absorbance ratio between 260 and 280 with a ratio near 1.8 was regarded as being of acceptable purity [24].

2.12 Multiplex polymerase chain reaction

There were two multiplex PCR assays that were designed to amplify staphylococcal virulence and antimicrobial-resistance genes. Multiplex Set A focused on resistance genes (mecA, mecC, and blaZ) and the pvl virulence gene. Multiplex Set B was dedicated to virulence genes (eta, etb, and tsst). Table 2 depicts the hybridization temperatures and anticipated amplicon lengths. The KOD Hot Star DNA Polymerase kit (Novagen, Toyobo, Merck Millipore) was used to perform the PCR according to the manufacturer’s instructions. The PCR mixture contained the following: 2.5 μL of 10× Buffer, 2.5 μL of dNTPs, 0.5 μL of KOD polymerase, 1.5 μL of 25 μM MgSO₄, and 0.4 μL of forward and reverse primers (10 mM) for each gene in multiplex Set A. In multiplex Set B, 0.3 μL of etb and tsst and 0.5 μL of eta were used. Finally, 0.7 μL of DNA template and molecular-grade water were added to arrive at a final reaction volume of 25 μL [24, 25].

Table 2. Attributes of the primers employed for the multiplex polymerase chain reaction (PCR) amplification [26]

2.13 Agarose gel electrophoresis

Agarose gels (2% for PCR Set A, 1.5% for Set B) were prepared by dissolving 0.8 g or 0.6 g of agarose in 40 mL of 0.5× TBE buffer, respectively. To stain DNA, SYBR Safe (0.4 μL) was added, and the molten agarose solution was poured into a tray. The gel was placed in a horizontal electrophoresis unit (EasyCast™, Thermo Scientific) filled with 0.5× TBE, and the combs were removed. A 100 bp DNA ladder (10 μL) was loaded into one lane. PCR products (5 μL) were loaded into different wells with 1 μL of loading buffer (BlueJuice™, Thermo Fisher™). Electrophoresis was performed at 100 V in Set A and 90 V in Set B, and a Bio-Rad PowerPac was used; the duration of the electrophoresis was 30 and 40 minutes, respectively. Bands were visualized with a UV transilluminator [27].

2.14 Statistical analysis

Data were statistically analyzed employing Microsoft Excel (2019) and Statistical Package for the Social Sciences (SPSS; version 25). Univariate analysis was utilized to characterize the research population and important outcome variables. Frequencies and percentages were employed to summarize qualitative variables (age group, socioeconomic status, marital status, sample type, hospital service, and the presence of genes). In the case of quantitative variables, the measures of central tendency (mean and median) and dispersion (standard deviation or interquartile range, as needed) were determined. Associations between categorical variables (e.g., presence of resistance or virulence genes and clinical or demographic factors) were evaluated using the Chi-square test or Fisher’s exact test, depending on expected cell counts. Statistical significance was set at a 99% confidence level (p ≤ 0.01). Where significant, Cramer's V was employed to assess association strength: C = 0.0–0.1 (none), 0.1–0.3 (low), 0.3–0.5 (moderate), 0.5–1.0 (strong). In cases where significant associations were found, the strength of association was determined using Cramer's V coefficient [28]. Tables and graphs were used to present the descriptive statistics.

The present research examined the occurrence of Candida spp. within the oral mucosa of cancer patients and their contribution to oral infections. The research involved 10 patients (aged 24 to 84). The findings showed that there was a high prevalence of C. albicans and C. dubliniensis that were isolated in 63% of the cases. The findings are consistent with earlier research, including that by Whibley et al. [31], who also observed a C. albicans prevalence. C. albicans is the most commonly isolated species, although non-albicans species have increased. It was found that poor oral hygiene was a contributing factor. Mathur et al. [32] concluded that poor dental hygiene or lack of dental care was identified in more than 42% of the study participants with cancer, which directly correlates with Candida infections. C. albicans was also identified in over half of their cases by Sabino et al. [33], which supports the significance of a timely diagnosis and a coordinated medical response to help avoid complications.

Candida infections have also grown, particularly in immunocompromised people, such as those with hematological diseases. C. glabrata was isolated in the present research in a 55-year-old patient with leukemia. This complements a report by Hassan et al. [34], which indicated increased colonization of C. glabrata in patients with low T-cell counts, such as individuals with AIDS, undergoing transplants, and cancer patients. Further, the non-albicans species C. parapsilosis was detected in a 66-year-old female patient who had ovarian cancer. As Branco et al. [35] observed, C. parapsilosis was on the rise throughout Latin America, with a particular focus on patients with intravenous or catheter-related infections and long-term catheter or device usage. Patel [36] suggested a risk of Candida infection is a threshold of 200 CFU/mL. Although counts of 200-400 CFU/mL may be taken as normal, counts over 400 CFU/mL give a higher probability of symptomatic candidiasis. In the present investigation, the majority of the patients had concentrations of 1–200 CFU/mL, and two patients were within the carrier range. A patient with a count of 1,800 CFU/mL had not yet developed any infection symptoms, which means that a high CFU count doesn't mean symptoms of infection are guaranteed to develop. One interesting finding was the significantly higher mean CFU associated with C. glabrata (1800 CFU) than with C. albicans/C. dubliniensis (117.5 CFU). These observations indicate that C. glabrata could establish higher colonization levels in immunocompromised patients and might be a clinically relevant opportunistic species. Increased persistence and antifungal tolerance of C. glabrata isolates have been reported in previous studies.

Hamzavi et al. [37] emphasized the need for a professional approach and timely treatment of Candida infections, especially in conjunction with cancer treatment. The adherence to oral hygiene practices in this study was reflected by 70% reporting brushing more than three times per day. Regular tooth care, such as applying mouthwashes, flossing, and denture hygiene, is vital to lessening microbial colonization and complications. This present research is a preliminary step towards studying Candida colonization in the mouth of cancer patients. While it wasn't particularly convenient to have access to data or patient recruitment as a result of institutional limitations and COVID-19, the study clearly shows a growing trend of non-albicans species, which may have distinct pathogenic profiles or treatment resistance. It employed the traditional method and multiplex PCR to reveal information about resistance genes and virulence factors that were desperately sought nationally.

Among the 105 Staphylococcus aureus isolates, resistance was highest to penicillin (85/105, 81.0%), followed by oxacillin (62/105, 59.0%) and cefoxitin (54/105, 51.4%) (Table 6). These findings are comparable to those reported by Zaid and Mujahid [38], who reported high resistance levels to beta-lactams. This is because of two mechanisms: the production of beta-lactamases that break down the antibiotic and the presence of the mecA gene that codes for the PBP-2a protein, which allows cell wall synthesis to continue even when the antibiotic is inhibiting beta-lactamase. This resistance is due to the frequent use of beta-lactams in clinical practice. Hu et al. [39] and Márquez-Oviedo et al. [40] reported that resistance to penicillin is frequently attributed to the production of plasmid-mediated beta-lactamase (blaZ) in bacteria. In this study, the penicillin disk test was used, and the detection of beta-lactamase was confirmed in 80.0% of isolates, which is in accordance with the CLSI guidelines that this test is a sensitive method for detecting beta-lactamase. Oxacillin and cefoxitin resistance were found in 58.1% and 50.5% of isolates, slightly lower than the results of Idrees et al. [41]. One can explain the difference by the difference in clinical settings of the isolates. The levels of resistance in this study are consistent with the ones provided by Liang et al. [42] except that the current study revealed low-level resistance to vancomycin (6.8)—an anomaly relative to the past. The resistance rates to rifampicin, chloramphenicol, and trimethoprim-sulfamethoxazole were also similar to previous research. Nonetheless, the resistance rates of clindamycin, erythromycin, and gentamicin were less than those found by Andrade et al. [43].

This study identified resistance to linezolid (17/105, 16.2%) and vancomycin (9/105, 8.6%), in contrast to Hakim et al. [44], who reported no resistance to these last-line antibiotics. Resistance to levofloxacin (32/105, 30.5%), tetracycline (14/105, 13.3%), and rifampin (12/105, 11.4%) was higher than that reported in previous studies. Resistance to ceftaroline (10/105, 9.5%) is of particular concern because it is generally considered an effective therapeutic option against MRSA. Linezolid, vancomycin, and ceftaroline resistance are present, and there is an urgent need to understand the underlying resistance mechanisms via molecular surveillance. Resistance to 3 or more antimicrobial classes was seen in 53.3% of the isolates, which is similar to the findings of Liang et al. [42] and Moges et al. [45] but lower than the finding of Quispe et al. [46], which reported an MDR of 82.2%. These rates of MDR underscore the clinical complexities of treating Staphylococcus aureus infections with no detailed resistance profile. Also, 10.5% of isolates had inducible clindamycin resistance (iMLSB phenotype), 63.6% of which were MRSA. This is similar to that of Silvagni et al. [47] but is contrary to that of Saderi et al. [48], which indicates a geographic variation in the expression of iMLSB. Diagnosis of iMLSB is important to prevent treatment failure with clindamycin.

The identification of MRSA is traditionally based on the mecA gene, which codes for PBP-2a. Nonetheless, Vindel and Cercenando [49] discovered MRSA strains that do not have mecA and have mecC, which is 70 percent homologous to mecA. Thus, both genes and cefoxitin resistance were used as markers of MRSA in this study. In 56.2% of the isolates, MRSA was discovered. Even though 42.9% contained mecA, this research established a strong statistical association between MRSA and cefoxitin resistance, mecA, and mecC (p < 0.001). MRSA was detected four times more often in bronchial aspirates, and the sample type demonstrated an intermediate association, but gender and origin showed lower correlations. MRSA prevalence varies globally: 80.0% in Saudi Arabia [50], 55.1% in China [6], and 46.1% in Peru [46]. In Spain, the pediatric isolates contained only 15.4% of the isolates with the study underlining the geographic variability and the need for local surveillance systems. The most frequent resistance pattern (IQ15) was: Penicillin, Cefoxitin, Oxacillin, Clindamycin, Erythromycin, Gentamicin, Levofloxacin [51]. The same patterns were found by Gómez-Gamboa et al. [52] for resistance to beta-lactams, and then for resistance to macrolides and lincosamides. MecA was found in over 80% of the samples, in agreement with previous local reports. These samples had the highest resistance profile in the IQ15 and the blaZ, mecA, and tsst gene combination profiles. Molecular methods of detection, especially multiplex PCR, played a critical role in the detection of both resistance and virulence genes. The most frequently detected resistance gene was blaZ (78/105, 74.3%), followed by mecA (45/105, 42.9%) and mecC (19/105, 18.1%). This distribution is comparable to that reported by Andrzejczuk et al. [53], who also observed that blaZ was the predominant resistance gene, whereas mecC was detected less frequently. The presence of virulence and resistance genes is important. The blaZ, mecA, and tsst genes, which encode the TSST-1 toxin capable of causing severe systemic reactions, were the most common gene combination. The four virulence genes were screened in all isolates: tsst, eta, etb, and pvl. The tsst gene was most common (34.3%), in line with Fijałkowski et al. [54]. Other studies, namely those conducted by Amelia et al. [55] and Ade et al. [56], reported a higher occurrence of pvl, eta, and tsst, respectively. These discrepancies may reflect differences in strain distribution and suggest that virulence gene expression may occur independently of AMR profiles, warranting further investigation into their distribution in both resistant and non-resistant strains.

This study was limited by its cross-sectional design, which precluded causal inferences, and its conduct at a single center with a relatively small sample size, potentially limiting generalizability. The diagnostic accuracy of the identification of Candida spp. was reduced using RAPIDYeast. Furthermore, no antifungal susceptibility testing was done, and molecular sequencing was not available to confirm species identification.

Another limitation in the current study is that the identification was performed using the RAPIDYeast Plus system. While this technique allows for a practical and fast way of identifying yeast, it cannot distinguish C. albicans from C. dubliniensis and cannot further identify specific species within the C. parapsilosis complex. Thus, when C. albicans/C. dubliniensis and C. parapsilosis complex are cited as isolated yeasts, they need to be considered as Candida spp. complexes, not as specific Candida spp. This limitation can affect the Candida spp. distribution presented in the results since it might lead to an underestimation or overestimation of specific Candida spp. within their respective complexes. Future research needs to utilize molecular techniques such as PCR, Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS), and DNA sequencing to precisely identify Candida spp. at the species level. All participants had completed their planned cancer treatment before sample collection; therefore, the findings reflect oral microbial colonization during the post-treatment follow-up period rather than during active cancer therapy.