Activity of antifungal drugs and Brazilian red and green propolis extracted with different methodologies against oral isolates of Candida spp.

Background Oral candidiasis is an opportunistic disease caused by fungi of the Candida genus. The occurrence of Candida spp. resistance to the commercial antifungal drugs points to the search for alternative treatments. Propolis has been successfully used in the treatment of infectious diseases for centuries. It has been proposed that an ultrasound pretreatment in the propolis extraction protocol can enhance the concentrations of molecules with antimicrobial activities in the final extract. Thus, this study aimed to compare the antifungal activity against oral Candida spp. isolates of green and red propolis extracts submitted or not to an ultrasound pretreatment before the extraction procedure. Methods Candida spp. were isolated from denture stomatitis lesions and identified by sequencing. Oral Candida spp. isolates and reference strains were submitted to broth microdilution assays using commercial antifungals and Brazilian green and red propolis extracts submitted or not to an ultrasound pretreatment. Minimal Inhibitory Concentrations (MIC) and Minimal Fungicide Concentrations (MFC) were determined and biofilm formation interference was evaluated for resistant isolates. Results C. albicans, Candida tropicalis and Candida dubliniensis were isolated from denture stomatitis lesions. Growth inhibition was observed in all Candida isolates incubated with all green and red propolis extracts. At lower doses, red propolis extracts presented significant antifungal activity. The ultrasound pretreatment did not promote an increase in the antifungal activity of green or red propolis. Three isolates, which were highly resistant to fluconazole and itraconazole, were susceptible to low doses of red propolis extracts. These same three specimens had their biofilm formation inhibted by red propolis ethanolic extract. Conclusions Thus, red propolis can be faced as a promising natural product to be used in the auxiliary antifungal therapy of denture stomatitis. Supplementary Information The online version contains supplementary material available at 10.1186/s12906-021-03445-5.

oral infections [3,4], with Candida spp. being the most important etiologic agent of oral diseases [4]. Oral candidiasis is an opportunistic infection that develops in the presence of several predisposing conditions, such as immunodeficiencies, endocrine disorders and poor oral hygiene [5]. C. albicans is the most frequent agent found in oral candidiasis, but other Candida species, such as C. parapsilosis, C. tropicalis, C. glabrata, C. krusei, C. pseudotropicalis and C. guilliermondii, known as non-Candida albicans species (NCA), have been isolated from several clinical cases of the disease [4]. Oral candidiasis associated with denture stomatitis are difficult to treat, and high rates of recurrence are reported [6,7]. The most common treatment for denture stomatitis includes denture hygiene associated with the use of synthetic antifungal drugs [8]. However, the currently available drugs are not fully effective [9], as shown by the occurrence of resistant strains, infection relapse due to the unappropriated use of drugs and the persistence of the fungal infection even after treatment [10]. Furthermore, Candida spp. is known to be a competent biofilm-forming microorganism, and this situation is correlated with an enhanced resistance to antifungals [11][12][13]. The formation of biofilm by Candida spp. is considered a factor that contributes to the recurrence of oral candidiasis and development of chronic infections [14]. In this context, the antimicrobial activity of natural derivatives, such as propolis, has been seen as a promising new therapeutic strategy [7].
Propolis has antibacterial [15], anti-caries [16], antiinflammatory [17], antioxidant [4], antifungal [18], immunomodulatory [19], anticancer and antiproliferative properties [20]. More than 300 different components were identified in samples of propolis from different origins [21]. In Brazil, 14 different types of propolis were classified based on their geographic origin, color, and physicochemical properties [22][23][24]. The traditional methods for obtaining or fractioning propolis extracts are distillation (with or without vacuum), extraction by liquid solvents, chromatography, adsorption, and membrane-selective processes [25]. Among these, ethanol is commonly used as a consequence of its chemical affinity to several propolis compounds with important biological activities [26]. With the objective to improve the concentration of bioactive compounds in propolis extracts and to enhance the reproducibility of the extraction method, the use of ultrasound as a pretreatment of the extraction process has been proposed [27]. Both et al. [28] found that the use of an ultrasound-assisted technology for the extraction of polyphenols from black tea enhanced the yield of this compound by 15%.
Considering the need for more accurate treatments for oral candidiasis, the increasing resistance profile of Candida spp. strains and the promising use of ultrasound-assisted extraction technologies, the present study aimed to compare the antifungal activity of green and red propolis extracts obtained with and without ultrasound as an extraction pretreatment against oral Candida isolates.

Fungal samples and ethical aspects
The clinical isolates used in this study were obtained from patients screened by dentistry professionals at the Dentistry Ambulatory of the UNIME University, Salvador, Brazil. The samples were collected from palate denture stomatitis lesions suggestive of oral candidiasis using sterile swabs, inoculated on Sabouraud dextrose agar (SDA) (HIMEDIA, Mumbai, India) supplemented with 0.2% chloramphenicol, and incubated at 37 °C for 24-48 h [29]. The colonies were then isolated and maintained by weekly reinoculations in the same media. As references, we used four C. albicans strains kindly supplied by the Fundação Oswaldo Cruz (FIOCRUZ -IOC 2508, IOC 2517, IOC 3703 and IOC 3704).
The Ethics Committee of the Institute of Health Sciences, Federal University of Bahia, Salvador, Brazil (protocol number 2.118.563) approved this research.

Identification of the fungal isolates
All clinical isolates were inoculated in CHROMagar ™ Candida ® for presumptive differentiation of the Candida species as described by Madhavan et al. [30]. To confirm the identifications, DNA sequencing of ITS and nuclear large subunit rDNA (LSU) regions was carried out. Briefly, the extraction of the genomic DNAs of the Candida spp. clinical isolates was performed using the FastDNA Spin Kit (Mp Biomedicals, Solon, OH, USA). Polymerase chain reactions (PCR) were performed using the primers ITS4 and ITS5 for the amplification of the complete internal transcribed region 26, and LROR and LR7 for the amplification of the LSU region (Mycology Lab -Duke University | Duke Mycology, 2019) [31]. All PCR reactions were performed using Quatro G Taq DNA polymerase (Porto Alegre, RS, Brazil) in a final volume of 50 μL, containing 10 μl of Quatro G buffer, 3.0 μl MgCl 2 (50 mM), 1 μl DNTP (10 mM), 1 μl forward primer (10 pmol), 1 μl reverse primer (10 pmol), 1 μl DMSO, 1.5 μl BSA (1 μg/μL), 5 μL betaine (5 M), 0.2 μL Taq 5 U/ μL), 24.8 μL sterile water and 1 μL DNA template. The reactions were carried out in thermocycler as following: 2 min at 94 °C, 35 cycles of 1 min at 94 °C, 1 min at 55 °C, 1 min at 72 °C, and a final extension of 5 min at 72 °C.
Subsequently, the ethanol/EDTA 125 mM precipitation protocol was used to obtain the purified PCR products. The DNA sequencing was executed on 3130xl automated

Propolis samples
The red propolis was collected in the state of Bahia, Brazil, and the green propolis was collected in the state of Minas Gerais, Brazil. Four different propolis extracts were used in this experiment. The ethanolic green and red propolis extracts were obtained using conventional methods [20,22]; briefly, frozen red and green propolis were crushed and sieved (60 mesh), with a final particle size of approximately 0.250 nm, and homogenized 2 g samples of each propolis were extracted with ethanol (15 mL, 80%) by mixing the samples for 30 min under constant agitation in an incubation shaker (MA 420/ MARCONI-Brazil) at 70 °C and 710 rpm. The extract was recovered by centrifugation for 11 min at 8800 rpm and 5 °C. Then, an additional centrifugation step was performed with 10 mL of ethanol (80%). The supernatant was collected, homogenized, and kept at 50 °C until completely dry. Afterwards, the extracts were stored in tubes, wrapped in aluminum foil at inert atmospheric conditions (N 2 ) to avoid degradation. All extracts were kept at 5 °C until use. The other two extracts were obtained using the same method, but with the inclusion of a pretreatment with ultrasound at 50 °C for 20 min, according to Reis and collaborators [32]. The quantification of p-coumaric acid, artepilin C, formononetin and kaempferol in the extracts was performed using a high-performance liquid chromatography (HPLC) system equipped with an automatic injector and diode array detector (DAD). The content of total phenolic compounds was achieved based on the reaction with the Folin-Ciocalteau reagent, followed by spectrophotometry analysis at 765 nm. The content of total flavonoid compounds was determined using a method based on the reaction with a 2% methanol solution of aluminum chloride, along with a quercetin standard curve (5 to 105 μg/mL), followed by spectrophotometry analysis at 415 nm. These results, as defined by previous studies [20,32], are shown at the Supplementary Table 1. All the propolis samples were obtained from the Apis mellifera bee species.

Broth microdilution assay
The fungicidal activity of commercial antifungal drugs and Brazilian green and red propolis extracts were evaluated using the broth microdilution assay, as described by the M27-A3 protocol from the Clinical Laboratory Standards Institute [33]. Briefly, the yeasts were resuspended in sterile 0.9% saline solution and adjusted by spectrophotometry to a 0.8-1.0 optical density (530 nm), which corresponds to 0.5 on the McFarland scale. Subsequently, the yeasts were diluted 1:50 in sterile saline solution and then diluted 1:20 in RPMI 1640 culture medium (Sigma Aldrich, St Louis, MO, USA) supplemented with 2% glucose [34] for the inoculum obtaining [33]. The four different propolis extracts used in this study, green propolis extract without ultrasound pretreatment (GP_EtOH), green propolis extract with ultrasound pretreatment (GP_US), red propolis extract without ultrasound pretreatment (RP_EtOH) and red propolis extract with ultrasound pretreatment (RP_US), were dissolved in dimethyl sulfoxide (DMSO) and then serially diluted in 1% DMSO (concentration ranging from 0.015625 to 8 mg/mL). Commercial antifungal agents were used as reference drugs, as follows: fluconazole, with concentrations ranging from 0.125 to 64 μg/mL; itraconazole, ketoconazole, nystatin and amphotericin B, with concentrations ranging from 0.0313 to 16 μg/ mL.
The inoculum was added to 96-well sterile culture plates at 100 μL per well, followed by the addition of the commercial fungicides and propolis extracts in different concentrations. As a negative control, it was used pure RPMI 1640 media with propolis extracts and antifungals at different concentrations, but without the inoculum. As a positive control, culture media with the fungal inoculum and without any treatment was used. The plates were then incubated for 48 h at 37 °C. Then, the Candida spp. growth was assessed using a spectrophotometer (Thermo Scientific, USA) at 625 nm. Each combination of inoculum and propolis or fungicide treatment was performed in triplicate, and the entire procedure was repeated twice.
The minimum inhibitory concentration (MIC) value, which represented the lowest concentration that inhibited 100% of the fungal growth, was determined by the minimum concentration of propolis or antifungal that resulted in no optic densitometry (OD) reading above the negative control OD value. For the determination of the minimum fungicide concentration (MFC -minimal drug concentration able to kill 100% of the yeasts), aliquots from each well of the broth microdilution assay were plated in SDA and then incubated at 37 °C for additional 48 h. Thus, the lowest concentration that revealed no visible fungal growth was determined as the MFC. According to the M27-S4 document [35], the breakpoint for fluconazole was considered for the classification of all isolates according to the corresponding MIC (at μg/mL) as follows: resistant (R) ≥ 8; dose-dependent susceptible (SDD) = 4; susceptible (S) ≤ 2. For itraconazole, it was considered the M27-A3 document classification [33].

Biofilm interference assay
To evaluate the propolis activity against biofilm formation, the ethanolic red propolis extract (RP_EtOH) was tested against three clinical isolates: C. albicans PAC 8, C. dubliniensis PAC 1, and C. tropicalis PAC 15. These species were chosen because of its different susceptibility to RP_EtOH, as determined by the microdilution test. The ability of RP_EtOH to inhibit biofilm formation was evaluated as previously described [36]. Briefly, the strains were incubated in Sabouraud Dextrose Broth at 37 °C in a shaker at 250 rpm for 15 h. The cultures had the cell density adjusted to an OD 600 of 0.38-0.5 with RPMI media and then added to a 96 well plate. The plate was incubated in a shaker at 250 rpm at 37 °C for 90 min. After that time, the media was aspirated, the wells were washed with PBS and RP_EtOH diluted in RPMI (concentrations ranging from 1 to 16 mg/ mL) was added. 24 h later, the optical density was read in a spectrophotometer at 570 nm. The interference rates in % were obtained using the following formula [37]:

Identification of the Candida species isolated from denture stomatitis cases
The twelve oral cavity Candida clinical isolates were phenotypically identified using CHROMagar ™ Candida (data not shown) and Sanger sequencing (Tables 1 and 2). Our results showed the presence of three Candida species with different frequencies. C. albicans was the most frequent isolated species (58.33% -7/12), followed by C. tropicalis (33.33% -4/12) and C. dubliniensis being the less abundant of the isolated species (8.33% -1/12).

Susceptibility to the commercial antifungal drugs
The results showed that the reference strains and the clinical isolates present distinct resistance patterns for the three azoles antifungal drugs used in this study (fluconazole, ketoconazole and itraconazole). The four C. albicans reference strains were all susceptible to fluconazole, with MIC ranging from 0.125 to 0.5 μg/mL. The results of C. albicans susceptibility to fluconazole showed that four (4/7) isolates were susceptible (MIC ≤2 μg/mL), two (2/7) were dose-dependently susceptible (S-DD) (MIC = 4 μg/mL) and one (1/7) was resistant (MIC = 16 μg/mL) (Table 1). Thus, 14.28% of the C. albicans clinical isolates presented resistance to fluconazole. The evaluation of all Candida spp. oral isolates and reference strains used in this study showed that C. albicans had a great variability for fluconazole susceptibility, with MIC values ranging from 0.25 to 16 μg/mL. Eight C. albicans and three C. tropicalis were susceptible to fluconazole (MIC ≤2 μg/mL); one C. albicans and one C. tropicalis were resistant (MIC ≥8 μg/mL); and two C. albicans (MIC = 4 μg/mL) and one C. dubliniensis (MIC = 32 μg/mL) were fluconazole S-DD. The fluconazole MFCs for C. albicans were markedly variable, ranging from 0.125 to undetermined values (> 64 μg/mL). Eight of all C. albicans (8/11) showed MFC ranging from 0.125 to 16 μg/mL, and three (3/11) could not be determined (> 64 μg/mL). Considering the NCA species, the MFC for the C. dubliniensis isolate could not be determined (MFC > 64 μg/mL). Finally, C. tropicalis fluconazole MFC was determined for two isolates (MFC = 0.5 and 16 μg/mL) and undetermined for the other two isolates (> 64 μg/mL) ( Table 1).
The fungistatic effect of ketoconazole was observed at low concentrations for all C. albicans strains (MIC = 0.03125 to 0.125 μg/mL). However, the MFC was undetermined for four (4/11) of the C. albicans specimens (MFC > 64 μg/mL), being all of them clinical isolates. Regarding the other seven (7/11) C. albicans, four reference strains and three clinical isolates presented MFC ranging from 0.03125 to 1 μg/mL. C. dubliniensis presented ketoconazole MIC of 0.03125 μg/mL and MFC of 0.125 μg/mL. Regarding the C. tropicalis isolates, the MIC were 0.03125 and 0.25 μg/mL, and MFC were found for two isolates, with values of 0.03125 and 0.125 μg/mL, but it could not be determined (MFC > 16 μg/mL) for the other two isolates (Table 1). Since the M27-S4 document [35] do not provide a ketoconazole susceptibility classification, this specific classification could not be performed to this drug in this study.
Of the three azoles drugs tested in this study, itraconazole was the one that it could be observed a higher number of resistant isolates. Of the 16 C. albicans specimens, four reference strains and 12 clinical isolates, only seven (7/16) were susceptible (MIC ≤0.125 μg/mL), and one of the four C. tropicalis was susceptible to this drug (MIC = 0.03125 μg/mL). Of the remaining Candida specimens tested, six isolates were resistant (MIC ≥1 μg/mL): three C. albicans, two C. tropicalis and the C. dubliniensis isolate; two (2/16) C. albicans (MIC = 0.25 μg/mL) and one C. tropicalis (MIC = 0.25 μg/mL) were S-DD to itraconazole ( Table 1). The fungicidal concentration (MFC) of itraconazole could be determinate for eight (8/11) of the C. albicans tested, ranging from 0.03125 to 8 μg/mL. For the other three C. albicans (3/11) isolates, the MFC could not be determined (MFC > 16 μg/mL), and these  Table 1). The AmB MIC for C. albicans ranged from 0.25 to 4 μg/mL and the MFC ranged from 0.5 to 4 μg/mL. For C. dubliniensis, the MIC and the MFC were the same (2 μg/mL). Finally, for C. tropicalis isolates, the MFC ranged from 0.25 to 2 μg/mL, and the MFC ranged 0.5 and 4 μg/mL.
When analyzing each isolate alone, it can be seen that the clinical isolate C. albicans PAC 17 presented the highest MIC and MFC concentrations for all commercial drugs tested. Thus, this isolate can be considered a potential multidrug resistant organism (Table 1).

Susceptibility to Brazilian green and red Propolis
All green and red propolis extracts evaluated herein induced growth inhibition in all Candida spp. specimens tested in this study (Fig. 1). However, both red propolis extracts were more effective in lower concentrations when compared to the green propolis extracts  Figs. 1 and 2). The dose-response curves showed that all Candida spp. had a higher tolerance to both green propolis extracts in lower concentrations; however, above 0.125 mg/mL, the growth inhibition was ≥50%, and it was above 95% at 8 mg/mL (Fig. 2 -a and b). In contrast, both red propolis extracts, with and without ultrasound pretreatment, presented a higher activity even in the lowest concentration tested (0.015 mg/mL) (Fig. 2 -c and d).
The green propolis extracted with different pretreatments were tested against the reference strains 2508, 2517, 3703 and 3704, and the MIC values of GP_EtOH and GP_US ranged from 2 to 4 mg/mL ( Table 2). However, when we evaluated the MFC for GP_EtOH (4 to 8 mg/mL), it could be determined for all reference strains, while for GP_US the MFC (2 to > 8 mg/mL) could not be determined for two (2508 and 3703) of the four reference strains ( Table 2). Concerning the clinical isolates, the MIC and MFC were undetermined for most of the Candida spp. treated with both green propolis extracts.
For the red propolis extracts (Table 2), the MIC values were 0.5 and 1 mg/mL for RP_EtOH and 0.25 to 2 mg/ mL for RP_US, when considering the reference strains. For these same strains, the MFC ranged from 1 to 4 mg/

Biofilm formation interference by red Propolis
The red propolis ethanolic extract was able to interfere in the biofilm formation by all the three strains tested in this specific assay (Fig. 3a, b and c). In all cases, the concentration of 8 mg/mL induced the highest interference values, reaching the maximum of 89.8%. The concentration of 16 mg/mL caused interferences in the readings by the spectrophotometer since it condensed in the bottom of the wells. C. dubliniensis (PAC 1) (Fig. 3 -b) was the least susceptible isolate when compared to the other ones tested in the assay, but even so, its biofilm formation was reduced by 78% at the concentration of 2 mg/mL. Regarding the other two strains -C. albicans (Fig. 3a) and C. tropicalis (Fig. 3c), the treatment with the red propolis ethanolic extract at the lowest concentration tested (4 mg/mL) presented interference results of 65.4 and 72.9%, respectively.

Discussion
Denture stomatitis is an infection of the oral cavity characterized by inflammation and erythema, being the fungi from the Candida genus important etiologic agents of the disease [38]. The presence of Candida albicans and Candida-non albicans in these infections have been already described [39,40]. The identification of these species is usually based on the use of a chromogenic medium which has been described as a good accurate method [30,41]. However, the molecular approach represents a more accurate method for species identification [42]. Our results showed the presence of three Candida species isolated from denture stomatitis lesions, being C. albicans the most frequent species and, in a lower abundance, C. tropicalis and C. dubliniensis. C. albicans and C. tropicalis species were found with a similar frequency causing chronic periodontitis at Alagoas [3], a Brazilian state close to Bahia state, where this study was conducted. In Pará state, Brazil, C. albicans was also the most frequent species associated with this oral disease, with a 78% frequency; additionally, C. tropicalis, C. famata and C. parapsilosis were also found, but in association with C. albicans at the infection site [39]. In Spain, C. albicans corresponded to 70% of Candida specimens isolated from oral candidiasis cases, followed by C. glabrata (8.6%), C. parapsilosis (7.4%) and C. tropicalis (3.3%) [41]. The main drugs used to treat denture stomatitis are miconazole, fluconazole, itraconazole, nystatin, amphotericin B, ketoconazole, clotrimazole and chlorhexidine [43]. In this way, this study used five important commercial drugs that are routinely used for oral candidiasis treatment. Our results showed a great variability in C. albicans susceptibility to fluconazole. Besides this, most of the isolates were susceptible to this drug. Siqueira and collaborators [3] showed that some clinical oral isolates of C. albicans presented a great variability in fluconazole susceptibility, and the occurrence of resistance in 36.8% of the Candida spp. isolates tested. In yeasts isolated from bloodstream infections, fluconazole susceptibility tests have demonstrated that 19% of the isolates were resistant to fluconazole [44]. In addition, the susceptibility test to antifungal agents against C. albicans and NCA isolates, when using the criteria of the CLSI document M27-S4, has shown an 2.4-fold increase in the number of species not susceptible to these drugs, when compared to the criteria of document M27-A3 [45].
Regarding the NCA species studied herein and its susceptibility profiles to fluconazole and itraconazole, different patterns were observed. Fluconazole resistance is more common in NCA species than in C. albicans [44,46]. Despite this, no correlation between the phylogenetic distribution of Candida isolates and the susceptibility profile fluconazole has been found [47]. Omran et al. [40] observed that C. albicans, C. glabrata and C. tropicalis isolated from Iranian patients with denture stomatitis presented 15.5, 17.4 and 12.5% of in vitro resistance to fluconazole, respectively. As other NCA species, C. dubliniensis was associate to infections in immunocompromised patients [48,49]. C. dubliniensis azoles resistance cases were already demonstrated in few studies [50,51]; an in vitro induction of fluconazole-resistance showed that C. dubliniensis can easily develop resistance to this drug [50]. Nevertheless, the increase in fluconazole-less susceptible or resistant Candida strains indicates that the use of alternative drugs in the treatment of oral candidiasis is urgently needed [9,10].
The results obtained in ketoconazole susceptibility assays showed lower MIC values. However, the MFC evaluation showed that this drug failed to present a fungicidal effect against the Candida spp. clinical isolates. Ernst and collaborators [52] showed that fluconazole was fungistatic against C. albicans and Candida neoformans, but without measurable fungicidal activity. For Trichosporon asahii, azoles showed a fungistatic effect, but without fungicidal activity [53]. The absence of a fungicidal effect can lead to the selection of resistant lines and consequent treatment failure, increasing the number of relapse cases [54].
Our data showed a great variability in the nystatin MIC and MFC values. However, different to what were observed for the azoles, all MIC values for nystatin and Amphotericin B (AmB) could be determined and, except for one C. albicans isolate for nystatin, all isolates had their MFC values determined. Nystatin and AmB are antifungals from the polyene class and its activity is associated to the membrane ergosterol content [54]. Miranda-Cadena et al. [41] showed that nystatin presented an excellent antifungal activity against all tested isolates, in opposition to fluconazole and itraconazole. Regarding AmB, our results still showed similar MIC and MFC values in almost all Candida isolates. In a previous study, C. albicans presented 71% agreement between MIC and MFC values after exposure to AmB [55]. This situation suggests that the same concentration of the drug is able to inhibit the growth and to kill the fungi, as well as facilitate the treatment of the disease, since the drug concentration can be more effective in the control of the fungal spread.
The evaluation of the Brazilian green and red propolis extracts dose-response curves indicated that both red propolis extracts are potent agents against Candida spp. when compared to the green propolis extracts. The antimicrobial activity of the propolis extracts is attributed to the phenolic and flavonoid content [17,20]. Several researchers have reported that different propolis extracts present antifungal [18], antimicrobial [18,56], antitumoral [57], antioxidant [58], anti-inflammatory and immunomodulatory properties [17,59]. Green propolis extract activities have been associated in part to artepillin C, a cinnamic acid derivative present in high amounts in Brazilian green propolis samples [20,58]. Artepillin C is considered one of the main active components of the green propolis extracts and exhibits antitumor [60], antiinflammatory [61] and antimicrobial activities [62].
The chemical composition of the green and red propolis ethanolic extracts [20,32] and of the red and green propolis ultrasound-assisted extracts used in this study [32], focused in compounds that have already been cited as having antimicrobial activities, are shown at the Supplementary Table 1. The isoflavonoid formononetin was found in the in Brazilian green propolis ethanolic extract at the concentration of 4 mg/g [20], while red propolis ethanolic extracts presented 8.68 mg/g of this compound [32]. For propolis extracts submitted to ultrasound pretreatment, formononetin was present at concentrations of 7.77 mg/g in green propolis and 8.40 mg/g in red propolis [32]. Red propolis ethanolic extracts showed almost two times more formononetin when compared to green propolis ethanolic extracts, while the green and red propolis ultrasound-treated extracts present almost the same formononetin concentration. This compound is considered one of the major active components of the red propolis extract due to its diverse biological activities [18,20,[63][64][65]. Thus, the high formononetin content of the red propolis can be associated to the best Candida growth inhibition by red propolis when compared to green propolis, and this situation can be supported by the results presented by the treatment with the ultrasoundtreated extracts.
Regarding total phenolic compounds and flavonoids, both green and red propolis ethanolic extracts demonstrated to have high contents of these compounds. Green propolis ethanolic extracts showed a concentration of 181.71 mg EAG/g for total phenolic compounds [20] and 46.80 mg EQ/g of flavonoids [20], while ultrasound treated green propolis extracts had 342.09 mg EAG/g and 22.68 mg EQ/g of total phenolic compounds and flavonoids, respectively. Red propolis ethanolic extracts presented concentration of 308.49 mg GAE/g of total phenolic compounds and 82.87 mg EQ/g of flavonoids, while ultrasound-treated red propolis extracts presented 314.75 mg GAE/g of total phenolic compounds and 90.38 mg EQ/g of flavonoids [32]. The green propolis extract obtained using ultrasound-assisted extraction has more total phenolic compounds than the ethanolic extract. Red propolis do not showed any differences in total phenolic compound concentration in the extracts obtained by both extraction methodologies. However, when the flavonoids content was compared, the green propolis ethanolic extract has more of this compound than the ultrasound-assisted extract. Red propolis, independently of the extraction method used, had higher contents of flavonoids when compared to green propolis, presenting two and four-fold more flavonoids than green propolis ethanolic and ultrasound extracts, respectively. Propolis antifungal activity have been associated to the high flavonoid content found in different propolis extracts [66,67]. Considering that the antifungal activity of red propolis extracts were superior to the green propolis extracts in our study, we can associate this better result to the higher contents of flavonoids and formononetin of red propolis.
The comparison between green and red propolis extract activity against Candida spp. showed that both red propolis extracts had more fungistatic and fungicidal activity than the green propolis extracts. The scientific literature reports a strong red and green propolis antifungal activity against Candida spp. [18]. A variability in antimicrobial and antitumoral activities of colored propolis extracts have already been described, but the green and red propolis showed better results when compared to the yellow or brown propolis extracts [20,68].
The ultrasound pretreatment in red propolis was able to enhance the final concentration of the isoflavones formononetin (60% increase) and kaempferol (undetectable in red propolis extracts without ultrasound pretreatment), when compared to extracts without this pretreatment [32]. However, even with these higher concentrations of antimicrobial compounds in RP_US, our results showed that just one strain that had MIC > 8 mg / mL when treated with RP_EtOH presented a MIC of 2 mg / mL after exposure to RP_US, and the MFC values of the RP_US presented two fewer undetermined results. This situation can be explained based on the study by Neves et al. [69], where it was found that the acetate fraction of red propolis extracts presented an increased concentration of formononetin, but these fractions did not show a higher fungicidal activity against Candida spp. These authors suggested that the antimicrobial activity of red propolis extracts cannot be attributed to just one compound, but is a result of a synergistic effect of them [70].
Our data also showed fungistatic and fungicidal activities of the red propolis extracts against Candida spp. that were resistant to fluconazole and itraconazole. In another study, some NCA strains from Rio Grande do Sul, Brazil, selected in a fluconazole resistance step-by-step assay, showed a cross-resistance to itraconazole and an increase in the MIC for ketoconazole, but no change in the susceptibility to red propolis extracts [69]. A clinical trial using a gel and a mouthwash containing green propolis in the treatment of denture stomatitis showed that these treatments had the same effectiveness against Candida spp.-associated denture stomatitis as a miconazole gel [8]. Pippi and collaborators [71] demonstrated that the red propolis from Minas Gerais State, Brazil, had a synergic effect with fluconazole, increasing up to 16-fold the fluconazole susceptibility of C. parapsilosis, C. glabrata and a polymicrobial culture formed by a mix of C. parapsilosis, C. glabrata, C. krusei and C. tropicalis; all of them resistant to fluconazole.
The ability of propolis to interfere with the consolidation and adherence of the biofilm of Candida species is poorly investigated. However, Tobaldini-Valerio and collaborators [72] observed that green propolis effectively affected the biofilm formation by C. albicans, C. parapsilosis and C. tropicalis, but in a species-and strain-dependent manner. In another study, similar results were obtained for green propolis, and the biofilm