The samples taken from the patients were transferred to the Department of Medical Mycology, School of Medicine, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran. Then, the samples were cultured on CHROMagar^TM Candida media (CHROMagar^TM, Pioneer, Paris, France) and incubated at 35°C for 48–72 h. In the next step, the grown clinical isolates were compared and identified in terms of colony color based on the desired culture medium brochure and standard samples. Then, the culture medium was examined for colony diversity and the colonies were counted based on colony-forming unit/swab. After that, the isolates were purified on Sabouraud Dextrose Agar + Chloramphenicol (SC) (Liofilchem, Italy). The medium and the isolates were transferred to microtubes containing sterile distilled water for long-term storage (6 months) at two temperatures: 30°C and room temperature.

The genomic DNA was directly extracted by boiling method (Yamada et al., 2002; Silva et al., 2012).

To confirm the isolates as C. albicans, PCR-RFLP, sequencing, and duplex PCR were performed. The primers for PCR-RFLP are listed in Table 1, and the enzyme for PCR-RFLP was MspI. Sequencing was done after PCR of V9g and LS266 area as shown in Table 1.

Biofilm production was performed on RPMI 1640 medium. All isolates were cultured in Sabouraud Dextrose Broth and incubated in a shaker incubator at 30°C for 24 h. After 24 h, the grown colonies were washed twice with sterile phosphate buffer solution (1× PBS) (pH 7.4) (Sigma-Aldrich) and centrifuged at 6063 rpm for 3 min at 4°C (SIGMA 1-15PK). Then, a yeast suspension was prepared from colonies with a concentration of 10⁶ cfu/ml in RPMI 1640 medium. Subsequently, in each well of flat bottom 96 microplates, 200 μl of suspension was poured in the wells. Culture medium was considered as the negative control and culture medium plus yeast suspension was the positive control. The microplates were placed at 37°C for 48 h. After this period, the culture medium in each well was emptied, and the wells were washed three times with sterile PBS. The microplates were then inverted at room temperature for an hour to dry. One hundred microliters of 0.1% crystal violet solution was added to each well and the microplate was placed at 37°C for 15 min without moving. Then, it was removed from the wells and washed again with sterile PBS, and the microplates were dried at room temperature. In the next step, 100 μl of 96% ethanol was poured into each well and the microplate was gently shaken in a circle by hand to extract the crystal violet color bound to the yeasts forming the biofilm faster. Then, the optical density (OD) of this solution was measured at 595 nm with a microplate reader (BioTek, Elx808). Finally, the biofilms were classified using the following formula (Abraham et al., 2020; El-Baz et al., 2021):

No biofilm: Absorbance ≤ Absorbance Control

Weak biofilm: Ac < A ≤ (2 × Ac)

Moderate biofilm: (2 × Ac) < A ≤ (4 × Ac)

Strong biofilm: (4 × Ac) < A

The isolates were grouped based on OD value. The isolates with low biofilm formation (LBF) were classified as first quarter (Q1); those with greater OD values were defined in the third quarter, which had high biofilm formation (HBF) (Q3), and the isolates with OD values between the first and third quarters were classified as second quarter and possessed intermediate biofilm formation (IBF) (Q2) (Li et al., 2018).

All the isolates were cultured on Sabouraud Dextrose Broth and incubated in a shaker incubator at 30°C for 24 h. After this time, the grown colonies were washed twice with sterile PBS and centrifuged at 6063 rpm for 3 min at 4°C. A yeast suspension with a concentration of 10⁶ cfu/ml was subsequently prepared from colonies in RPMI 1640 medium. One hundred microliters of the prepared yeast suspension was added to each well of 96-well flat bottom microplate, which was placed at 37°C for 4 h. After this period, the culture medium in each well was discarded and washed with PBS, 100 μl of RPMI1640 culture medium was added to the wells and incubated for 24 h at 37°C. The culture medium in each well was emptied, and 200 μl of amphotericin B (Sigma Aldrich-USA) diluted with RPMI1640 with a concentration of 100 μg/ml (Moazeni et al., 2014; De Brucker et al., 2016; Alonso et al., 2018) was added to the wells and placed at 37°C for 48 h, after which the contents of each well were discarded and washed once with 100 μl of PBS. After washing, 100 μl of PBS was added to each well and homogenized using a sterile pipette. Twenty microliters from each well was serially diluted in 1:10 ratio in 10 dilutions of PBS. After dilution, 50 μl of each well was harvested and cultured on yeast peptone dextrose agar or YPD medium (Liofilchem, Italy). The plates were kept in an incubator at 48°C for 48–72 h. The plates were examined for colony count after 48 h and the samples were reported in three categories: high persister, low persister, and non-persister (LaFleur et al., 2006, 2010). Finally, samples containing persister cells were examined for MIC determination of amphotericin B according to CLSI M27-S4 guidelines, and if they were sensitive to the population with MIC < 2 μg/ml, the samples were confirmed as persister isolates (Clinical and Laboratory Standards Institute [CLSI], 2008; Maheronnaghsh et al., 2020).

To determine the maximum and minimum survival rates, the isolates were ranked from highest to lowest based on their survival rates against amphotericin and then the first 15 isolates with the highest survival rates were specified as the persister group and the last 15 isolates not showing survival were chosen as those without persister or non-persister group. There were a total of 30 isolates in these two groups. Evaluation of BCR1, EFG1, and CAT1 gene expression was performed in 30 C. albicans isolates by the real-time qPCR method. Real-time qPCR was done on isolates that were under inducing and promoting conditions.

Inducing conditions were applied for isolates in the state of biofilm production while promoting conditions were meant for isolates in the normal condition, and RNA was extracted from these isolates. In other words, the promoting conditions are the conditions before the formation of biofilm, and the inducing conditions are those after the formation of biofilm. Promoting conditions mean the alteration in expression of genes that cause biofilms and inducing conditions are changing the expression of genes causing biofilm formation to give specific properties to cells.

For this purpose, RNA was first extracted (RNX-Plus) and then the purity and integrity of extracted RNA were investigated under both conditions by nanodrop and loading of RNA on agarose gel, after which cDNA was synthesized from RNA (BioFact, South Korea). Finally, real-time qPCR was performed using Real-Time PCR Roche lightcycler^® 96, BioFACT^TM 2X Real-Time PCR Master Mix, and the primers listed in Table 2. Real-time qPCR was carried out with the following reaction conditions: 40 cycles of denaturation: 15 s at 95°C; annealing: 60 s at 60°C; and extension: 60 s at 72°C.

Primer efficiency was determined for all the gene expression assays using a standard curve and LinRegPCR software. The absence of dimer primers and contamination was controlled by observing the melting curves and loading products on the agarose gel. The absence of DNA was also checked with no RT control (no reverse transcriptase control). The expression level of EFG1, BCR1, and CAT1 target genes in the isolates relative to that of the reference ACT1 gene was first calculated manually by Excel and then confirmed by REST2009 software to ensure the accuracy of results. Finally, statistical analysis and plotting with SPSS 22 and GraphPad prism were done.

After sequencing of both groups, a software was used to draw the phylogenetic tree¹, and the phylogenetic tree of C. albicans isolates was used using Maximum Likelihood and Boot Strap 100 to ensure the validity and reproducibility of the drawn trees. The table of genetic distances was drawn using MEGA 7 program of Pairwise Distances model. Then, cluster and clade were determined in the phylogenetic tree.

In this study, t-test and one-way ANOVA, Fisher’s exact test, and Mann–Whitney test with a significance level of <0.05 were used. The normality of data was checked using Shapiro–Wilk test. All tests were analyzed by SPSS software version 22.

Our analysis demonstrated that two isolates had a weak biofilm formation while 102 C. albicans isolates showed strong biofilm production. Another classification was also performed based on optical observation of the isolates that were grouped to three quarters as shown in Figure 1.

In this study, 104 C. albicans isolates were tested to separate persister cells from biofilm-forming isolates. This stage of persister cell isolation was done in triplicate and the results of each sample were reported as average. Out of 104 isolates under study that were exposed to 100 μg/ml amphotericin B, 83 had a survival rate in the range of 0.0003–9% and 21 had no survival ability in the face of the drug. Those isolates that were able to survive in the presence of amphotericin B were examined for MIC. To determine the sensitivity of the population, MICs of the isolates were read by microdilution method in the presence of amphotericin B in the range of 0.3–2 μg/ml. According to available guidelines, namely, CLSI M27-S4, out of 83 isolates, 77 with MIC < 2 μg/ml were sensitive and six isolates with MIC ≥ 2 μg/ml were classified as the resistant population and were excluded from the study. Therefore, 77 C. albicans were considered as persister cell isolates (Table 3 and Figure 2).

According to a study by LaFleur et al. (2010), the isolates with a survival rate of >6% were defined as high persister, and those showing survival <6% were classified as low-persister cells. Besides, the isolates that had no survival were defined as non-persistent (Cohen et al., 2013).

Out of 77 C. albicans persister cells, two isolates with 9% and 8% survival were classified in the high-persister cell group, 75 isolates with the ability to survive in 0.75–0.0003% were classified in the low-persister group, and the 21 remaining isolates were classified in the non-persister group. Two high-persister isolates were taken from patients as follows: one was related to a patient undergoing chemotherapy who was hospitalized for approximately 2 months and did not have clear oral candidiasis symptoms. The only symptoms were dry mouth and redness of the tongue. Another isolate was taken from an HIV-infected patient recently discharged from the infectious ward of Razi Hospital who had been hospitalized for approximately 45 days. He had redness, inflammation, dry mouth, change in taste, and burning sensation, and of course, was in the end stage of the disease or so-called AIDS. According to the case file in Razi Hospital, only fluconazole was used for the patient whose disease was chronic and recurrent (Table 4).

The mean production of low persister based on survival rate in HIV-infected patients with the ability of 0.04 ± 0.14% (survival rate ± standard deviation) was almost 5.5 times higher than patients undergoing chemotherapy with the ability of 0.008 ± 0.13%. The mean of low persister production in all patients was 0.01 ± 0.0013%. Since the survival rate and the number of high persisters in patients under study were almost the same, the same mean value was observed for both groups of patients (Figure 3).

In this part of the research, we first deal with the relationship between different variables related to HIV patients with those related to patients undergoing chemotherapy with the production of persister cells, and then this variable is examined in all patients.

In this study, because there was only one isolate from HIV patients with a survival rate of 9%, statistical analysis was performed only in low and non-persister categories. Statistical results showed no significant correlation between qualitative variables of gender, tuberculosis, pneumocystosis, hepatitis B, hepatitis C, receiving antiviral drugs, tuberculosis prophylaxis drugs, HIV transmission from mother to child or through occupational exposure, blood transfusion, CD4 count, homosexuality, common syringe use for injection, and history of drug injection in non- and low-persister cells (Supplementary Table 1).

Data related to the age of subjects were normal, and no significant difference was observed between the two groups (low- and non-persister cells). There was no significant difference between viral and Candida loads between the two groups of non- and low-persister cells. However, the results of statistical tests showed that the biofilm variable is significantly different in the two groups of non- and low-persister cells (Supplementary Table 2) (p = 0.021).

Considering that there was only one isolate in this group with a survival rate of 8% among the high-persister group, the statistical analysis was performed for two groups: low- and non-persister cells. Accordingly, the calculations showed that there was no significant relationship between gender and persister values in the non- and low persister cell groups.

Data related to the age of subjects were normal and no significant difference was observed between the two groups. Statistical results also showed no significant difference in Candida load and biofilm variables between the two groups (Supplementary Table 3).

Statistical analysis was performed using t-test in the two groups of non- and low-persister cells, but no significant difference was observed between the two groups (Figure 4).

Real-time qPCR was performed in promoting and inducing conditions. The purity of RNA on the A260/A280 and A260/A230 ratios was 1.8–2.0 and 1.9–2.0, respectively. Then, for the synthesis of cDNA, 500 ng of RNA entered the reaction.

After normalizing the data with reference or control gene (ACT1), our findings showed an increase in the expression of EFG1, BCR1, and CAT1 genes in persister isolates, among which the highest expression was related to CAT1 gene, followed by EFG1 and BCR1, respectively.

The mean expression of EFG1, BCR1, and CAT1 genes was statistically evaluated using logarithmic data between the two groups of persister and non-persister isolates using t-test. Among the genes, only the expression of the CAT1 gene was significant between persister and non-persister isolates (Supplementary Figure 1).

In inducing conditions, the expression of EFG1, BCR1, and CAT1 genes was evident in all isolates of both persister and non-persister groups. After normalizing the data with reference or control gene of ACT1, our results showed an increase in the expression of EFG1, BCR1, and CAT1 genes in persister isolates under inducing conditions, among which the highest expression was related to CAT1 gene, then BCR1 and EFG1, respectively. The mean expression of EFG1, BCR1, and CAT1 genes was statistically evaluated through logarithmic data between the two groups of persister and non-persister using t-test. The increasing expression of BCR1 and CAT1 was significant between the persister and non-persister groups (p < 0.05) (Supplementary Figure 1).

The mean increase in gene expression under promoting and inducing conditions is described as fold change (i.e., 2^–ΔΔCT). The highest amount of fold change in both promoting and inducing conditions was related to the CAT1 gene with 15.34 and 5.54, respectively. The lowest fold change in promoting and inducing conditions was related to EFG1 gene and its amount was 2.18 and 1.14, respectively (Table 5 and Figure 5).

The DNA sequencing analysis of 20 isolates confirmed the presence of C. albicans and was recorded on DNA Data Bank of Japan (Accession No. LC612887-612906). After blasting, the similarity of sequences was reported to be 94–99.88%. The bootstrapping value was repeated 100 times to draw the phylogenetic tree to ensure the accuracy of drawing the phylogenetic tree. In our study, bootstrapping equal to 100 was reported, indicating that the phylogenetic tree of the desired nucleotides had a high degree of reliability. The similarity and intra-species difference of the sequences is calculated through pairwise distance. The average pairwise distance in sequences with a distance scale of 0.01 was 0.029. In these isolates, the biggest difference was related to 1A–low persister and 5A-normal isolates, which was equal to 8%. Then, the difference was related to 1A–low persister with normal 3A, 7A, and 8A isolates with a value of 7.8%. The isolate 6A–high persister, 5A-normal, and then 7A, 8A, and 3A had the highest difference (approximately 7%). The 6A–high-persister isolate had a genetic distance of 5.7% with the 7C–high-persister isolate.

Based on the mean pairwise distance of 0.029, the sequences were divided into four clusters that were more genetically similar. The largest cluster belongs to cluster IV, which contains persister cell isolates and non-persister ones. Both high-persistent isolates were genetically separated by 5.7% in two separate clusters, and their offspring were completely different. 7c–high persister had no significant difference from the normal population, including 8A, 7A, 5A, 3A, 4A, 2A, and 6C by about 1–2% and was probably derived from the normal population. Most low-persister isolates differ by about 1% from the population of normal isolates, indicating that the nucleotide changes were negligible (Figure 6).