The beneficial role of Candida intermedia and Saccharomyces boulardii yeasts on the immune response of mice vaccinated with a SARS-CoV-2 experimental vaccine

Non-Saccharomyces yeasts emerge as possible new probiotics with a beneficial effect equal to or greater than the reference probiotic yeast, Saccharomyces boulardii. In this work, we evaluated the immunomodulation effect caused by Candida intermedia in mice vaccinated with inactivated SARS-CoV-2. We conducted preliminary tests using murine macrophages (RAW 264.7) stimulated with viable and heat-killed yeast cells, culture supernatant, and DNA, using qPCR to detect the mRNA transcription. Next, mice were supplemented with C. intermedia before each dose of the SARS-CoV-2 vaccine, and then antibody production was measured by ELISA. The probiotic strain S. boulardii CNCM I-745 was used as a control. We also explored the differences in fecal microbiomes between the non-supplemented and supplemented groups. Live cells of C. intermedia increased the transcription of IL-4, IL-13, and STAT3 by macrophages RAW 264.7, while heat-killed cells up-regulated TNFα and Bcl6, and the culture supernatant positively impacted TLR2 transcription. Concanavalin, zymosan, and lipopolysaccharide were used to stimulate splenocytes from C. intermedia-supplemented animals, which showed increased transcription of TNFα, IFNγ, IL-4, Bcl6, and STAT3. Sera from these animals showed enhanced levels of anti-SARS-CoV-2 IgG, as well as IgG1 and IgM isotypes, and sIgA in fecal samples. The microbiome of the C. intermedia-supplemented group showed a higher abundance of Bacteroides spp. and Clostridium spp., impacting the Bacteroidetes/Firmicutes balance. We concluded that C. intermedia and S. boulardii could stimulate and impact the gene expression of cells important for innate immunity, influence the composition of the gastrointestinal microbiome, and primarily boost the humoral response after vaccination. Statements and Declarations Funding The present work was carried out with the support of Conselho Nacional de Desenvolvimento Científico (CNPq, Brazil), grant number 150538/2021-9.

Staphylococcus aureus, Pseudomonas aeruginosa, among others. Strains of these three yeasts are generally associated with their presence in the microbiome, fermentative profile, and sensory contribution in fermented products such as wine, beer, and cocoa [17], and they currently arouse interest regarding their antagonism activity to other competitive or contaminant microorganisms [18][19][20].
Some strains of C. intermedia have demonstrated the ability to decrease or suppress the growth of L. monocytogenes [8,21], in addition to being responsible for the production of antimicrobial peptides that affect other yeasts [20]. This yeast had its genome recently sequenced [22,23], but its presence in fermented dairy foods was already described 30 years ago [24]. In a previous study [11], we identified in vitro some potential probiotic characteristics in the isolate C. intermedia ORQ001, such as high levels of autoaggregation, co-aggregation with pathogenic Gram-and Gram+ bacteria, and a low decrease in cell viability after exposure to GIT conditions. Based on these results, we focused on the development of new studies regarding the probiotic potential of this yeast.
Although there are several studies evaluating the probiotic activity of specific strains, information regarding their administration and behavior in vivo is scarce, as well as knowledge of pathways and roles of action in GIT performance. The mechanisms of action of the immunomodulatory effect are still not completely elucidated, mainly for non-Saccharomyces yeasts, however, there is enough evidence that suggests the need for a deep investigation into it [25]. As highlighted for yeasts like Pichia pastoris [26] and by our group for S. cerevisiae and S. boulardii [27][28][29], the evaluation of immune system modulation and ability to control pathogens in vivo represent important and expressive results in the search for probiotic characteristics in these microorganisms. When these yeasts are administered as supplements, it is possible to observe the improvement of specific immune responses to vaccines [29], immunostimulatory activity [30], reduction of pathogen concentration in experimentally infected animals [28], or even the prevention of infections [27,31].
Recently, researchers have linked the immunostimulatory effect of probiotics to the potentiation of the immune response to different vaccines and antigens, suggesting a great relevance in the treatment and prevention of infection by SARS-CoV-2, the virus responsible for the COVID-19 pandemic (Coronavirus disease) [32][33][34]. Previous studies developed by our research group using an experimental vaccine composed of inactivated SARS-CoV-2 [35] demonstrated that the vaccine stimulated humoral and cellular immune responses, which it is believed can be potentiated by the administration of probiotics. Thus, here we evaluated some potential probiotic characteristics through in vitro and in vivo tests, regarding the stimulation of macrophages by live cells of C. intermedia ORQ001 and S. boulardii CNCM I-745 and their derivatives (inactivated cells, culture supernatant, and yeast DNA), the immunomodulatory effect, and the impact on the gastrointestinal microbiome performed by oral supplementation with these yeasts, using as a target the immune response of mice vaccinated with inactivated SARS-CoV-2.

Strains and culture conditions
The wild isolate of C. intermedia ORQ001 was previously characterized regarding its probiotic potential by Piraine et al. [11], and for this study, it was obtained from the microorganism bank of the Microbiology Laboratory in the Federal University of Pelotas, as well as the commercial yeast S. boulardii CNCM I-745 (Floratil ®, a reference probiotic strain), which were cryopreserved in glycerol at -80 °C.
Yeasts were grown overnight in YM (Yeast and Malt Extract) medium (0.3% yeast extract, 0.3% malt extract, 0.5% peptone, and 1% glucose) at 30    After 15 min in the dark at room temperature (~25 °C), it was added 50 µl/well of 2 N sulphuric acid to stop the reaction. Absorbance was measured in a microplate reader (Thermoplate®) with a 492nm filter.
Moreover, for antibody titration, using the same serum samples (D28, D35, and D42), two-fold dilutions were made in a range of 1:100 to 1:6400. The cut-off for antibodies titers we defined as the absorbance of day 0 plus the standard deviation.
The Mouse Monoclonal Antibody Isotyping Reagents Kit (Sigma-Aldrich®) was used for the evaluation of IgG1 and IgM isotypes, following the protocol suggested by the company. In this test, plates were coated with the SARS-CoV-2 virus as described before, and then serum samples of days 28, 35, and 42 from each group were pooled, diluted as described previously, and applied in triplicates. Secondary isotype-specific goat antibodies conjugated to HRP were used in 1:5.000 dilution, and plates were read as described before. For the detection of sIgA (secretory IgA isotype) in fecal samples, we conducted the ELISA test protocol described by Santos et al. [42]. For this purpose, 0.1 g of pooled fecal samples from days 28, 35, and 42 were suspended in 1% PBS with 1 mM Phenylmethylsulfonyl fluoride (PMSF, Sigma-Aldrich®) and 1% Bovine Serum Albumin (BSA), and mixed by vortex until complete homogenization.
Thus, these samples were diluted with PBS-T + 5% powdered milk in a 1:2 ratio, then added (100 µl/well) over a plate coated with the SARS-CoV-2 virus. Specific sIgA antibodies to the SARS-CoV-2 virus were detected with goat anti-mouse IgA alpha-chain + HRP (Abcam®), diluted to 1:1.000. Incubation periods, washing steps, and solutions used for ELISA tests described before were also used for sIgA ELISA.

Gastrointestinal microbiome evaluation
Total DNA present in feces from samples on day 0 (after 5 days of yeast supplementation) was Negative and positive controls were used during all processes. The raw data obtained from sequencing were submitted to quality control, taxonomic classification, visualization, and description of communities using the Knomics-Biota system [43].
Using the R programming language and microbiome package, the OTUs (Operational Taxonomic Units) table was normalized by "clr-transformation" to calculate alpha and beta diversities [44]. Alpha diversity was determined using Observed, Shannon, Simpson, InvSimpson, Fisher, and Evenness indices.
The Kruskal-Wallis test was inferred to verify significant differences among groups (p<0.05). Beta diversity was estimated by Principal Component Analysis (PCA), using the Aitchison matrix [45].

Statistical analysis
Serology data and those related to cellular response were analyzed by analysis of variance (2way ANOVA) with Dunnett's multiple comparison test, and the statistical difference was determined if the pvalue < 0.05. All analyzes were performed mainly in the statistical software GraphPad Prism 11 v. 7.

Immunostimulatory activity of viable yeast cells and derivatives on RAW 264.7 cell culture
Live cells of C. intermedia were responsible for a significant stimulation (p<0.05) of the IL-4 and IL-13 mRNA production, with an expressive increase of 7.6 and 2.4-fold, respectively. On the other hand, down-regulation was observed for IL-2 and TNFα, decreasing their expression (p<0.05) by 2 and 1.6-fold, respectively. Moreover, we observed that TLR2 transcription was down-regulated (2.2-fold decrease) and, more expressively, the transcription factor STAT3 by 3.4-fold. Meanwhile, we found that viable cells of S. boulardii were responsible for an important stimulus, up-regulating the mRNA transcription of cytokines IL-2 (8.6-fold), IL-4 (3.0-fold), IL-13 (8.7-fold), and IL-23 (7.4-fold). Observing the expression profile for the other genes analyzed, there were no significant differences (p>0.05) in mRNA transcription after the stimulus with S. boulardii (Fig. 2).  The supernatant of C. intermedia culture was also used to evaluate the response generated by in vitro cultured macrophages. We observed that C. intermedia culture supernatant induced an increase in TLR2 transcription (a 4.4-fold increase) and a decrease in IL-2 and IL-13 levels (a 2.6 and 2.0-fold decrease, respectively). The supernatant of S. boulardii culture also revealed an impact on the mRNA transcription of macrophages RAW 264.7, down-regulating the transcription of cytokines IL-2 (2.0-fold) and TNFα (3.4fold), and up-regulating Bcl6 transcription factor (1.5-fold). Finally, we found that the DNA extracted from C. intermedia discreetly stimulated an increase in IL-4 production (1.5-fold), but mainly the inhibition of IL-2 by a 4.1-fold decrease in mRNA transcription, with less intensity for IL-13, TNFα, and STAT3 between 1.6 to 2.3-fold. Likewise, S. boulardii DNA has also been shown to impact the mRNA transcription of some genes, positively stimulating IL-2 (1.7-fold) and TLR2 (1.6-fold), and negatively stimulating IL-13 (2.9fold), IL-23 (2.3-fold), Bcl6 and STAT3 (1.6-fold both).

Modulation of immune cells response by a short period of supplementation with yeasts
After five days of supplementation with C. intermedia and S. boulardii, splenocytes from animals from each group were collected and cultured in vitro. These cells were stimulated with ConcA, Zymosan, and LPS, with mRNA transcription of cytokines (TNFα, IFNγ, IL-4, IL-12, IL-23), transcription factors (Bcl6, NFκβ, STAT3), and TLR2 being evaluated by qPCR (Fig. 3). We identified that splenocytes from animals supplemented with C. intermedia showed increased mRNA transcription of TNFα, IFNγ, and IL-4 for all stimuli when compared to non-supplemented animals, in which the biggest difference levels were observed for TNFα in ConcA-stimulated cells (6.7-fold increase) and IFNγ in LPS-stimulated cells (20.4-fold increase). For some gene transcription profiles, the mRNA transcription could even result in a change from down-regulation to up-regulation when animals were supplemented with C. intermedia, as was observed for IL-4 cytokine mRNA transcription in LPS-stimulated cells. For IL-23 cytokine transcription, it was also detected an increase in its level in splenocytes from animals supplemented with the non-Saccharomyces yeast.
The stimulation of splenocytes from the S. boulardii group generated a response of cytokine transcription similar to that observed for the C. intermedia group; however, it was detected in some cases at lower levels than the non-Saccharomyces group or with no statistical difference for the non-supplemented group. The only two cases in which a different behavior was identified were on IL-12 mRNA transcription  "a" means a significant difference (p<0.05) from the non-supplemented group, and "b" represents a significant difference (p<0.05) from the basal expression level

SARS-CoV-2 inactivated virus
Observing the production of antibodies specific to the SARS-CoV-2 virus (total IgG), the C.
intermedia-supplemented group showed greater antibody levels (p<0.05) than the non-supplemented group intermedia-supplemented group starting at day 28, in which higher absorbance values were detected, representing an enhanced antibody level that lasted until day 42. Compared to the non-supplemented group on days 28, 35, and 42, sera from animals that ingested C. intermedia during two rounds of supplementation showed an increase of 41, 28, and 22% (respectively) in the absorbance values of anti-SARS-CoV-2 IgG at these time points (Fig. 4).  The data represent the means ± standard deviation of the absorbance values obtained by indirect ELISA.
The titer was considered the highest pooled sera dilution detected above the cut-off (p<0.05) Possible modulation of the humoral response mediated by the yeasts was also monitored by the evaluation of antibody isotypes. Analyzing the IgG1 isotype (Fig. 6A), it was observed for day 28 that all groups presented the same levels in pooled sera, with no statistical difference (p>0.05) from the non-   (IL-4) cytokines. The relative mRNA transcription of these cytokines was normalized using the β-actin transcription level as a reference. Data are shown as the mean ±SD (Standard Deviation). The letter "a" means a significant difference (p<0.05) from the non-supplemented group, and "b" represents a significant difference (p<0.05) from the basal expression level

Gastrointestinal tract microbiomes of supplemented and non-supplemented animals
The microbiome of fecal samples from treated and non-treated groups revealed some important differences in bacterial composition, which were suggested to be related to supplementation with C.
intermedia and S. boulardii. We observed in the control group (non-supplemented, with a normal diet) that the microbial environment on GTI is composed, at the phylum level, of Bacteroidetes (53.9%) and Firmicutes ( In the betadiversity analysis (Fig. 8D), we observed that the microbiome of animals supplemented with C. intermedia was less altered when compared to non-supplemented animals, while this alteration was more significant for the group supplemented with S. boulardii.

Discussion
Yeasts, especially those belonging to Saccharomyces spp., can participate in the activation of immune cells [38,46,47]. When macrophages (and other immune cells) are exposed to fungi, this interaction initiates intracellular signaling cascades that culminate in opsonization, phagocytosis, and the production of cytokines, chemokines, and antimicrobial peptides, among others [48]. The stimulation of macrophages by viable and non-viable cells depends on the cell structure of each yeast, cellular portions, yeast surface, internal cell components, and actively secreted molecules (by live cells) to the extracellular environment [46]. In addition, the medium supernatant (cell-free), obtained after yeast culture, also may have metabolic byproducts that interact with immune system cells [49].
This interaction still needs to be explored, as researchers have characterized different response patterns mediated by probiotics, pathogenic yeasts, and their components [50,51]. Although Smith et al. [46] did not find differences in cytokine-inducing properties among live, UV-irradiated, and heat-killed cells, in our study, a considerable variation (p<0.05) in relative mRNA transcription was detected between stimuli with viable and non-viable cells. After a heat treatment associated with high pressure, cell inactivation occurs through membrane damage, loss of nutrients and ions, protein denaturation, and essential enzyme inactivation, which can lead to modifications in cell coarseness and roughness [59,60]. Previous studies reported even higher expression levels of cytokines when macrophages were stimulated with heat-killed probiotics than viable cells, as observed by Miyazawa et al. [61]. In our study, the level of TNFα mRNA transcription was potentiated when macrophages were stimulated by heat-killed cells of C. intermedia, a behavior also observed for other Candida species by Navarro-arias et al. [56], who stimulated peripheral blood mononuclear cells (PBMCs) with viable and heat-killed C. albicans, C. tropicalis, C. guilliermondii, C. krusei, and C. auris.
Since Bcl6 plays a fundamental role in the regulation of Th2-type inflammation and is constantly expressed in monocytes [62], we sought to observe its transcriptional levels in stimulated macrophages.
Bcl6 also regulates macrophage function by repressing pro-inflammatory cytokine production and controlling the differentiation of Th17 cells [63]. During active inflammation occasioned by infections, microorganisms can bind to PRRs (Pattern Recognition Receptors) and activate signaling pathways via MAP kinases and NFκβ, resulting in the stimulation of cytokines with pro-inflammatory activity [64].
Increased NFκβ transcription is involved with TLR2 receptor activity, which has its expression level increased by mannose and β-glycans recognition [65]. Toll-like receptors are innate immune response infection sensors that participate in the activation or inhibition of macrophage activity via the Jak-STAT pathway; signaling via STAT3 is activated by several cytokines and their receptors, such as IL-2, IL-6, IL-10, IL-23, and IL-27 [66]. The main role of STAT3 in macrophages is to mediate anti-inflammatory effects, restricting gene transcription of pro-inflammatory cytokines [67] and repressively impacting NFκβ signaling pathways [68]. Molecules present in C. intermedia culture supernatants were also responsible for stimulating significant levels of TLR2, suggesting that metabolites secreted by yeasts are also important in stimulating receptors of immune system cells. Secreted proteins by Candida spp. yeasts are linked with TLR2/TLR4 recognition, as demonstrated by Wang et al. [76], promoting an inflammatory response in DCs and macrophages stimulated in vitro. The most common ligands related to TLR2 are PAMPs (Pathogen-Associated Molecular Patterns) originating from glycolipids, lipopeptides, or GPI-anchored structures [77], thus it is suggested that higher receptor mRNA transcription in these cases is related to secretome products or proteins detached from yeast cell walls that have these structures in their conformations [76,78].
Probiotic microorganisms act in multiple ways, including the activation and stimulation of immune cells (lymphocytes, granulocytes, macrophages, mast cells, epithelial cells, and dendritic cells) to release various cytokines, regulating the innate and adaptive immune responses, and potentiating anti-inflammatory or pro-inflammatory responses, or even maintaining homeostasis [79]. Santos et al. [80] observed Zymosan, ConcA, and LPS are well-known immune response stimulants with different origins [75,82], so comparing splenocyte responses after stimulation with these molecules permits the prediction of and helps identify the influence of yeast administration on animals' immune systems in different scenarios. Based on our findings, C. intermedia was suggested as a microorganism with potent immunomodulation effects in cytokine transcription by splenocytes, even superior to S. boulardii, which was observed in TNFα, IFNγ, IL-4, and IL-12 mRNA transcription. IFNγ is one of the most potent macrophage activators, and together with TNFα and IL-12, it is a pro-inflammatory cytokine that promotes cell-mediated immunity [69]. Besides inducing an up-regulation in cytokine transcription, it was also observed an increase in the mRNA levels of transcription factors, of which STAT3 transcription was consistently up-regulated no matter what the stimulus. Since our work was the first to explore the immunomodulatory activity of C. intermedia and its application in vivo, more data is necessary to better understand all the immunological signals and pathways involved after its administration to animals.
When dealing with the Candida genus, it is always important to highlight that safety is crucial to determining the possibility of yeast usage as a tool to promote health, since even non-albicans species may cause infections [83]. Most of the infections (63-70%) related to Candida spp. are caused by C. albicans, and the rest are associated with other 18-30 species classified in Candida genus (non-albicans) that comprises around 200 species, such as C. glabrata, C. tropicalis, C. parapsilosis, C. krusei, C. auris [83,84]. At this date, C. intermedia is not a common human pathogen; it is described in the literature only in 3 cases associated with infections involving immunosuppressed individuals [85,86]. Nevertheless, a presumption of safety for use of C. intermedia in fermented food products was published in the Bulletin of the International Dairy Federation 495/2018 [87], and, in our study, there were no related deaths or apparent signals of yeast infection in the supplemented animals. Therefore, further studies are needed to confirm the safety of supplementation with C. intermedia.
Modulation of vaccine response by C. intermedia and S. boulardii ingestion was evaluated in our study associated with an experimental vaccine composed of inactivated SARS-CoV-2. The potential of S.
boulardii to improve immune responses after vaccination was identified by our group in previous studies [29,80], and is also related in the literature [88][89][90]. Our results demonstrated for the first time the possibility of supplementation with yeasts to improve the immune response to a SARS-CoV-2 vaccine, as well as showed non-Saccharomyces yeasts as potential (and promising) candidates for this application.
While studies linking bacteria and their probiotic activity are more common, an unknown number of nonconventional yeasts (besides Saccharomyces spp.) are waiting to be discovered in this field [2]. C.
intermedia supplementation suggested a positive impact on the humoral response of mice vaccinated with the anti-SARS-CoV-2 vaccine, an aspect that confirmed the immunomodulatory role of this yeast. Although our results do not reflect confirmation of higher protection against virus infection or organism protection, they provoke interest in new studies using other vaccines (viral, bacterial, recombinant, etc.), animal models (ovine, bovine, humans, etc.), pathogen challenge, and protocols of supplementation (short-term × longterm) associated with the yeast administration.
Some authors described enhanced levels of IgG and sIgA when probiotics were administered in association with immunizations using DNA vectors, recombinant subunits, and viral vaccines [80,88,91], which was also observed in our results for C. intermedia-supplemented animals. These findings are relevant because secretory IgA is predominant in mucosal surfaces, playing an important role in viral immunity, and it is suggested its role in an enhanced capacity to neutralize SARS-CoV-2 [92]. Klingler et al. [93] also diets with S. boulardii is being studied mainly for farm animals as an alternative to antibiotics, in which its activity of shaping the gut microbiome is highlighted [15]. This change in microbiome composition caused by S. boulardii also impacts the decrease of obesity, fat mass, hepatic disorders, and metabolic inflammation, as demonstrated by Everard et al. [95]. While Everard et al. [95] and Yu et al.
[96] described in their results a decrease in Firmicutes and an increase in Bacteroidetes abundance in mice supplemented with this yeast, we observed a similar Firmicutes/Bacteroidetes ratio between the non-supplemented and S.
On the other hand, differences in the Firmicutes/Bacteroidetes ratio were observed in the microbiome of C. intermedia animals, with increased levels of Bacteroidetes and lowered levels of Firmicutes. The increase in Bacteroidetes abundance, especially for the Bacteroides genus, may be related to the utilization of mannan available on Candida cell walls through degrading enzymes (mannanases and mannosidases) expressed by Bacteroides spp., serving as a nutrient (carbon) source for the bacteria [97].
LPS derived from Bacteroides spp. are described as showing impaired or even inhibitory capacity to elicit an inflammatory response, and a decrease in its abundance in the gut microbiome often results in an augmented population of pro-inflammatory bacteria [98,99]. The enriched presence of certain Bacteroidetes (Prevotella spp. and Bacteroides spp.) with anti-inflammatory properties was linked to fewer adverse effects after vaccination and associated with the highest antibody titers in some cases [100], correlating with our results. However, this positive association is not a consensus in the literature.
Bacterial species can take advantage of the biofilm created by Candida and adhere to it to thrive under conditions that would be harsh for them in GIT [97,101]. It was demonstrated for Clostridium perfringens and C. difficile that these bacteria benefit from the microenvironment and microbial interactions when co-cultured with Candida albicans [101], and in our study, this positive interaction between Candida-Clostridium may be the answer to enhanced levels of Clostridium spp. in the microbiome of C. intermediasupplemented animals. The abundance of Lactobacillus spp. was lower in the microbiome of yeastsupplemented animals, which may be due to the competition between yeasts and bacteria for the same metabolic niches throughout the gastrointestinal tract [84]; however, further studies are needed to prove this linkage. Finally, several mechanisms affect the response to vaccines, including vaccine formulation, dose, immunization route, vaccination schedule, the host immune system, and the gut microbiota [90]. The modulation of resident microbiota by probiotics is a factor that may improve vaccine immunogenicity [90,102], thus we suggest that the alterations detected in the microbiomes of the supplemented mice could be one of the key factors associated with the increase in the humoral response detected in these groups.
In summary, we demonstrated in this study that supplementation with the yeasts C. intermedia