Introduction
Kefir, a probiotic fermented milk produced from kefir grains, contains a complex mixture of bacteria, yeasts in a matrix of proteins, and polysaccharides Reference Farnworth1 . Kefir exerts anticancer properties through different mechanisms, including participation in several biological processes, such as antioxidant, apoptosis and proliferation, secretion of cytokines, suppression of the Th2 immune response and activation of the Th1 immune response Reference Bengoa, Iraporda, Garrote and Abraham2–Reference Sharifi, Moridnia, Mortazavi, Salehi, Bagheri and Sheikhi4 . For this reason, fermented dairy foods have gained evidence in the cancer context Reference Zhang, Dai, Liang, Zhang and Deng5 .
Cancer represents a leading cause of death globally and is a public health problem worldwide. According to data published by the International Agency for Research on Cancer, there were an estimated 19.3 million new cases and 10 million cancer deaths worldwide in 2020. Colorectal cancer ranks third in neoplasia incidence and second among causes of mortality for both sexes Reference Sung, Ferlay and Siegel6 . Dietary factors have an impact on cancer risk Reference Steck and Murphy7 . In this sense, clinical and preclinical evidence has shown that probiotics have antimutagenic and immunomodulatory activity and can act in the prevention and treatment of colorectal cancer Reference Eslami, Yousefi and Kokhaei8,Reference de Brasiel P.G., Dutra Luquetti, do Peluzio, Novaes and Gonçalves9 .
In addition to these factors, the gut microbiota influences the development of colon tumors Reference Rinninella, Raoul and Cintoni10 . The host colonization starts in early life, and the development of the microbiota is influenced by diverse conditions, such as the mode of delivery, type of feeding, genetic factors, diet, mother’s age and metabolic status, and antibiotic use Reference Milani, Duranti and Bottacini11 . Breast milk is an essential factor for the development and maturation of an infant’s microbiota Reference Pannaraj, Li and Cerini12 . Bioactive milk components and the breast milk microbiome are transferred by breastfeeding and have an impact on the infant gut microbiota Reference Komatsu, Kumakura, Seto and Izumi13 . Due to the difficulties of follow-up and evaluation, we do not fully know the effects of breastfeeding on the long-term composition of the gut microbiota of the progeny and how this can impact the development of diseases in adulthood. The composition of the offspring’s gut microbiota is dependent on the maternal diet and postweaning infant diet and can be altered by changing the environment at the postnatal stage Reference Edwards14 .
Studies have shown that the way in which certain environmental factors, such as nutritional factors, present themselves in critical periods of development, including lactation and puberty, can influence human biology and health in the long term, reinforcing the concept of “developmental origins of health and disease” (DOHaD) Reference Barker, Bull, Osmond and Simmonds15 . These critical periods are crucial for developmental plasticity, where epigenetic responses to environmental changes may lead to transgenerational transmission Reference Sookoian, Gianotti, Burgueño and Pirola16 . Thus, lactation and puberty are windows of opportunity for intervention to prevent and reduce the risk of several diseases in adult life, including colorectal cancer. In this way, we investigated the long-term effects of kefir consumption during lactation by the mother or at puberty by the offspring on the adult gut microbiota in colorectal cancer-induced rats.
Materials and methods
Experimental design and procedures
Three-month-old Wistar rats (200–250 g) were obtained from the Center of Reproduction Biology of the Federal University of Juiz de Fora, Minas Gerais, Brazil. The study was approved by the Ethics Committee for Animal Handling of the Federal University of Juiz de Fora at Minas Gerais, Brazil (protocol no. 21/2016) and conformed to the guidelines of the Brazilian Council for Control of Animal Experimentation (CONCEA, Brazil). All animals were housed in plastic cages under controlled conditions of humidity (44–65%), light (12-h light/dark cycle), and temperature (22 ± 2°C).
To evaluate the effects of kefir intake during lactation or puberty, a day after the birth of the puppies, the litters were adjusted to ten male pups for each dam. Then, lactating rats with their offspring were randomized using a computer based random order generator, divided into three groups: control (C, n = 7); kefir lactation (KL, n = 8); and kefir puberty (KP, n = 7). To avoid litter effects, male pups from different litters were used in this study. During lactation, dams in the C and KP groups were given 1 mL of filtered water by oral gavage once per day. For the KL group, the dams were given 1 mL of kefir milk (108 colony forming units (CFU)/mL) by gavage once per day. After weaning (21 days), KP pups received kefir treatment by gavage until 60 days old (Fig 1a). The offspring of the KP group received an initial volume of 0.2 mL (at weaning), which was progressively increased to 1 mL on the 40th day of life. In addition to their respective treatments, all the animals were allowed free access to standard rodent chow (Nuvilab®, Paraná, Brazil) and drinking water.
Experimental protocol. On postnatal day (PN) 1, litters from both groups of dams were adjusted to ten pups. The groups of dams were placed on kefir treatment (KL, 108 CFU/mL, green indicated) or water (C) for the 21 days of the lactation period. The Kefir puberty (KP) group litters received the kefir treatment postweaning up to PN60 and were then maintained on commercial chow until PN240. Offspring (C, KL, and KP) received 1,2-dimethylhydrazine (DMH, red arrows) at PN 67, 69, 73, and 75 for colon cancer induction. After 24 weeks, the animals were euthanized, the cecal content was collected, and gastrointestinal tissues were evaluated for the presence of tumors. Colony forming unit (CFU).
At 67 days old, colorectal carcinogenesis was induced. All animals received an intraperitoneal injection of 1,2-dimethylhydrazine (DMH, Sigma–Aldrich, St Louis, MO, USA) at a dose rate of 40 mg/kg body weight twice weekly for 2 consecutive weeks. DMH was freshly prepared and dissolved in 0.9% saline solution containing 1 mM EDTA and 10 mM sodium citrate, pH 8. At 240 days old, equivalent to 24 weeks after the last DMH injection, the animals were euthanized with a lethal dose of ketamine (90 mg/kg) and xylazine (10 mg/kg). After euthanasia, the colon was washed in saline solution and opened along the mesenteric margin. The number, incidence, and location of tumors were assessed, as well as, colon length and weight.
Preparation and analysis of kefir milk
The kefir grains used in the study were obtained from the Department of Nutrition and Health, Federal University of Viçosa, Minas Gerais, Brazil. They were prepared with pasteurized whole milk (Benfica®, Juiz de Fora, Brazil) according to a previous study Reference Kim, Chon and Kim17 . In summary, in a glass container, 10 g of kefir grain was inoculated into 100 mL of milk and incubated at 25°C for 24 h. After that period, the grains of the fermented milk were separated. The procedure was repeated daily to obtain fresh kefir during the treatment period. The fermented beverage was analyzed for microbial composition by next-generation sequencing and showed a high prevalence of Lactococcus and Lactobacillus among the bacterial groups and Aspergillus and Cordyceps from the fungal community, as previously described Reference de Brasiel P.G., Medeiros, Machado, Ferreira, do Peluzio and Luquetti18 . The counts of lactic acid bacteria, yeasts and evaluation of the pH were performed periodically throughout the treatment and were kept at 108 CFU/mL, 106 CFU/mL and 3.9–4.1, respectively, thus meeting the standards proposed by the international body Codex Alimentarius 19 .
Microbiota analysis using 16S rRNA high-throughput sequencing and bioinformatics
Samples were collected directly from the cecum in sterile tubes and randomly chosen using a random number generator from the C (n = 5), KL (n = 5), and KP (n = 5) groups for gut microbiota analysis. DNA was isolated from the samples using the MagaZorb® DNA Mini-Prep Kit (Promega, Madison, WI, USA). 16S rRNA high-throughput sequencing was performed for each sample by GenOne Biotechnologies enterprise (Rio de Janeiro, Brazil).
The V3-V4 region of the 16S rRNA bacterial gene of the DNA samples was amplified using the 341F (5’-CCTAYGGGRBGCASCAG-3’) and 806R (5’-GGACTACNNGGGTATCTAAT-3’) primers. All PCRs were carried out with Phusion® High Fidelity PCR MasterMix (New England Biolabs, USA). The PCR products were purified with a Qiagen Gel Extraction Kit (Qiagen, Germany). Sequencing libraries were generated using the NEBNext® UltraTM DNA Library Prep Kit for Illumina, following the manufacturer’s recommendations, and index codes were added. The quantity and quality of the libraries were assessed on a Qubit® 2.0 Fluorometer (Thermo Scientific, USA) and Agilent Bioanalyzer 2100 system. Sequencing was carried out on the MiSeq (Illumina, USA) platform (2x 250 bp), as suggested by the Earth Microbiome Project Reference Caporaso, Lauber and Walters20 . Amplicon sequence variants (ASVs) were inferred using the DADA2 (Divisive Amplicon Denoising Algorithm 2) method Reference Callahan, McMurdie, Rosen, Han, Johnson and Holmes21 , and taxonomy was assigned using a naive Bayesian classifier method against SILVA database release 132 Reference Pruesse, Quast, Knittel, Fuchs, Glo and Ludwig22 . Diversity and phylogenetic analyses were performed with the aid of several R packages, including phyloseq Reference McMurdie and Holmes23 , vegan Reference Oksanen, Blanchet and Friendly24 , DECIPHER Reference Wright25 , phangorn Reference Schliep26 and ggplot2 Reference Wickham27 . All the plots (taxonomy, alpha and beta diversity, differential abundance) were carried out with ggplot2 implemented on R. We used DESeq2 to identify differentially abundant ASVs in pairwise comparisons between the groups Reference Love, Huber and Anders28 .
The sequence data were submitted to the GenBank database and are accessible under Bioproject PRJNA672176.
Colonic and inflammatory biomarkers and correlation analysis
For the measurement of cytokine levels in colonic tissues, 100 mg of tissue (medial colon) from each animal was homogenized in 1 mL of PBS buffer containing 0.05% Tween 20, 0.5% bovine serum albumin, protease inhibitors (0.01 mM EDTA), and 20 IU aprotinin A, using a tissue homogenizer. The resulting homogenate was centrifuged at 13,500 r.p.m. for 20 min at 4°C, and the supernatant was used for cytokine quantification. Cytokines were measured using ELISA kits (PeproTech Inc., now moved to Thermo Fisher, Waltham, MA, USA) – Interleukin (IL)-1β (Catalog # 900-M91), IL-6 (Catalog # 900-M86), interferon-γ (Catalog # 900-K109K), and tumor necrosis factor (TNF, Catalog # 900-K73K), following the manufacturer’s instructions, and the results were expressed as pg/mL. Total nitric oxide (NO) was evaluated through the Griess reaction Reference Miranda, Espey and Wink29 . Pearson’s correlation coefficient was used to analyze the correlation between microbiota, colonic and inflammatory biomarkers, and visualization was made in the form of a heat map.
Statistical analyses
The calculation of sample size was performed by power analysis using statistical software available online (http://www.3rs-reduction.co.uk/html/6__power_and_sample_size.html), considering a statistical power of 0.8 and a type 1 error rate. The data were analyzed by the statistical program GraphPad Prism version 8 (GraphPad Software, Inc., La Jolla, CA, USA) and are expressed as the mean ± standard error of the mean (SEM). Data normality was evaluated by the Kolmogorov–Smirnov test. For parametric analysis, one-way ANOVA followed by Newman–Keuls posttest determined differences between groups, and the relative risk (RR) and 95% confidence interval (CI) calculations were performed using the Koopman asymptotic score method. Correlations were identified by Pearson’s correlation coefficient. The differences were considered statistically significant at p < 0.05.
Statistical analysis was also performed on the microbiota data. To evaluate alpha diversity differences between groups, the Wilcoxon test was carried out. To test for differences in microbial community composition (PCoA), permutational multivariate analysis of variance on UniFrac distances was used.
Results
Maternal kefir intake prevents the development of tumors in offspring
KL offspring presented an increase in body weight at 21 days old, which did not change their body weight and adipose tissue in adulthood. No significant difference was observed in chow consumption between the groups (not shown). At 240 days old, offspring showed no change in glycemia, colon length, and colon weight (Table 1). Animals whose mothers consumed kefir during the lactation period did not develop colon tumors (Supplementary Fig S1), as evidenced by the significant decrease in tumor number compared to that of the control group (p = 0.027), as shown in Table 1. Maternal kefir intake resulted in a reduced relative risk (RR) of offspring tumor development (RR = 3.333, 95% CI = 1.282 to 7.700; p = 0.018). Photos of the colon (one representative colon from each group) are shown in Figure 2.
Maternal kefir intake prevents colon tumor development. Macroscopic view of the colons of each group, control (C), Kefir lactation (KL), and Kefir puberty (KP).
Nutritional and colon characteristics of pups
The results are expressed as the mean ± SEM (n = 7–8), *p < 0.05 vs. C, ***p < 0.001 vs. C and KP. Control (C), kefir lactation (KL), kefir puberty (KP).
16S rRNA sequencing data
The sequencing of the DNA samples extracted from the cecal contents resulted in 609,011 high-quality sequences with an average of 40,600 sequences per sample. Reads were clustered into 1,716 ASVs. Rarefaction curves indicated that sampling provided sufficient ASV coverage to accurately describe the bacterial composition of each sample, as it could be seen in Supplementary Fig S2.
Maternal kefir intake increases the richness and diversity of the offspring gut microbiota
The richness and diversity of the gut microbiota of the progeny were evaluated. The number of observed ASVs did not differed statistically between control and groups KL and KP. However, the observed ASVs was higher in the KL group, whose dams received kefir during lactation, than in the puberty alone group (p = 0.021) (Fig 3). The Simpson index was also greater in the KL group, although the difference was not significant (Fig 3). In contrast, the overall microbial community structures (beta diversity) differed significantly (r2 = 0.499; p = 0.001 by ADONIS test, cluster centroids are shown in Supplementary Fig S3) between the groups, as shown in PCoA using weighted UniFrac distance (Fig 3). However, it is possible to note that the microbiota from the C and KP groups were more related to each other when compared with that of the KL group.
Maternal kefir intake improves gut microbiota richness in adult offspring with DMH-induced colon carcinogenesis. Analysis of observed ASV, alpha diversity assessed by the Simpson index, and beta diversity principal coordinates analysis (PCoA) plot using weighted uniFrac distance highlighting differences between microbial communities (p = 0.002 by ADONIS test) in the groups. Control (C), Kefir lactation (KL), Kefir puberty (KP).
The critical period of diet intervention affects the gut microbiota in adulthood
The phyla Bacteroidetes, Firmicutes, and Proteobacteria dominated the gut microbiota of the animals (Fig 4). According to the taxonomic data, the microbiota composition of control animals (C) is in some ways more similar to the ones that received kefir during puberty (KP) and differed significantly from the animals whose dams received kefir during lactation (KL). Bacteroidetes were more abundant in the C and KP groups than in the KL group, while the opposite was observed for Firmicutes (Fig 4). The Proteobacteria abundance was higher in the control animals. The results showed that kefir administration in early life modified the microbiota of the animals in adulthood. The overall microbial composition for each sample at the family and genus levels is shown in Supplementary Fig S4 and S5. We observed a reduction in the abundances of the genera Acinetobacter, Prevotella_9, Prevotellaceae-NK3B31, and Prevotellaceae_UCG-001 and an increase in the abundances of Lactobacillus and Lachnospiraceae_NK4A136_group in the intestinal microbiota of KL animals. A rise in the abundance of the genus Lachnospiraceae_NK4A136_group was also observed in the KP group (Supplementary Fig S5). Differential abundance using DESeq2 was carried out to verify significant ASV differences between the groups. Fifteen ASVs were differentially abundant between the C and KL groups, with enrichment in ASVs associated with Lactobacillus, Ruminococcaceae_UCG-14, and Coprococcus_2 in the KL group (Fig 5). Acinetobacter and members of Prevotellaceae were enriched in the control group. We confirmed that the genera Lactobacillus, Romboutsia, Blautia, and Lachnospiraceae_UCG-004 were significantly (p < 0.001) associated with kefir maternal intake during lactation, while members of the Prevotellaceae family and Bacteroides were enriched in the animals that received kefir in puberty (Fig 5).
Kefir intake affects the bacterial abundance in the model of colorectal cancer. Boxplots of the relative abundances of bacterial phyla in the gut microbiota of each group: control (C), Kefir lactation (KL), and Kefir puberty (KP). Data are presented as the median, minimum, and maximum values, n = 5.
The gut microbiota of offspring whose mothers received kefir during lactation is enriched with butyrogenic and lactic bacteria. Plot showing the enrichment of genera with significant differential abundance between the control and KL groups, and the KL and KP groups, as detected and filtered by DESeq2. Control (C), Kefir lactation (KL), Kefir puberty (KP).
Tumor numbers have a strong correlation with IL-1
To better understand the relationship between gut microbiota and colon carcinogenesis, we conducted a correlation analysis between genus-level microorganisms and key biomarkers. As shown in Fig 6, the heat map indicated that the Acinetobacter genus was positively related to the pro-inflammatory cytokine TNF (Supplementary Table S1). Lactobacillus abundance was positively related to colon length, showing its trophic effects on gut mucosa. By contrast, NO which were negatively associated with the colon length. There is a strong correlation (r = 0.911; p = 0.002) between tumor number and IL-1 level, confirming the important relationship between inflammation and the tumor development.
Correlation between microorganisms (genus level), colonic and inflammatory biomarkers. Blue is positive correlation, and red is negative correlation, *p < 0.05.
Discussion
In this study, we show for the first time that kefir intake by mothers during lactation or by offspring at puberty produced distinct changes in the richness and composition of the gut microbiota of adult offspring induced by colorectal cancer. The administration of kefir in dams during lactation was associated with the nondevelopment of tumors and an increased microbial richness and relative abundance of genera associated with gut health in male offspring. However, administration of kefir up to puberty in pups did not have such significant effects on chemically induced tumor development and gut microbiota composition outcomes in offspring. These data together demonstrate the importance of lactation for the establishment of the intestinal microbiota and intestinal health of the progeny.
Maternal kefir intake during lactation prevented tumor development in adult offspring. Kefir anticancer properties were previously reported Reference Sharifi, Moridnia, Mortazavi, Salehi, Bagheri and Sheikhi4 , although in this study, we investigated the effects of its use in early life on the development of diseases in adults. Different environmental exposures can result in epigenetic modifications that act by modulating chromatin structure and gene regulation that do not involve changes to the underlying DNA sequence Reference Miranda, Espey and Wink29 . Distinct epigenetic profiles can be retained throughout the lifespan and may be inherited with transgenerational effects Reference Bianco-Miotto, Craig, Gasser, van Dijk and Ozanne30 . In this sense, epigenetic regulation has been reported to contribute to cancer development and progression Reference Fitz-James and Cavalli31 . Thus, exposure to environmental factors, such as diet during early life, may lead to favorable or unfavorable consequences, impacting the risk and prevention of diseases in the progeny Reference Bailey, Tokheim and Porta-Pardo32 .
Unlike previous studies Reference Zanardi, Grancieri and Silva34,Reference Zeng, Jia and Zhang35 , our group has studied the effects of kefir intake during critical periods of development on adult offspring. Our group previously demonstrated that kefir treatment during lactation and continued during puberty had an impact on colorectal carcinogenesis via a reduction in mucosa inflammation, reversing the effects of neonatal overfeeding Reference de Brasiel P.G., Luquetti and Medeiros36 . Here, we show that the period of lactation is a determinant period of these beneficial effects; that is, the maternal kefir intake during lactation is able to modulate the gut microbiota and program adult offspring to prevent the development of colorectal cancer. It is known that the establishment of the intestinal microbiota occurs early in life Reference Xiao and Zhao37,Reference Tian, Li and Zheng38 and that this initial microbiota will affect the microbial core in adulthood, potentially leading to more pronounced preventive effects; which was confirmed in our study.
High bacterial richness was a marker of cancer protection in our study. A species-rich gut ecosystem is more robust against environmental influence, and these microbial communities are resilient and more resistant to change Reference Sommer, Anderson, Bharti, Raes and Rosenstiel39 , which seems to have occurred in the KL group. These animals did not develop colon tumors during the follow-up. A low bacterial richness was associated with an inflammatory phenotype in obese individuals Reference Le Chatelier, Nielsen and Qin40 , and intestinal dysbiosis and increased intestinal permeability are factors that contribute to inflammation Reference Stone, Mcpherson and Darlington41 , creating a favorable microenvironment for neoplastic development Reference Cani42 . In addition to the change in structure, we also observed long-term alterations in the composition of the microbiota between groups. Alterations in the gut microbiota confer a predisposition to certain malignancies and influence the response to a variety of cancer therapies Reference Sims, El Alam and Karpinets43 . The lower diversity may explain the high incidences of autoimmune and inflammatory disorders Reference Belkaid and Hand44 . It is known that the microbiota is relatively stable after the first years of life Reference Laursen45 ; thus, exposure to modulating factors, especially in early periods, can promote alterations that have repercussions in later life.
We found an increase in the Lachnospiraceae_NK4A136_group in the cecal microbiota of animals that received kefir for one of the periods, lactation or puberty. This genus is a butyrate producer and has been shown to protect against inflammation Reference Borton, Sabag-Daigle and Wu46 . In contrast, control animals presented an increase in the relative abundance of Acinetobacter (Proteobacteria phylum); species in this genus, such as Acinetobacter baumannii, are opportunistic pathogens Reference McConnell, Actis and Pachón47 and are frequently isolated from cancer patients Reference Wu, Zhang and Zhao48 . The association of the genera Lactobacillus, Romboutsia, and Blautia with kefir maternal intake during lactation may be a good indication of the advantage of this period over the postweaning period for the establishment of the adult gut microbiota and colon cancer prevention. Lactobacillus was the second most abundant bacterial genus found when kefir milk was used to treat animals Reference de Brasiel P.G., Medeiros, Machado, Ferreira, do Peluzio and Luquetti18 . Thus, it is possible that these animals received a greater amount of these bacteria via breast milk during lactation. It has already been shown in animal Reference Morales-Ferré, Azagra-Boronat and Massot-Cladera49 and human studies Reference Padilha, Danneskiold-Samsøe and Brejnrod50 that breast milk is capable of modulating the intestinal microbiota of the progeny, with breastfeeding being fundamental for the establishment of the gut microbiota, especially colonization by Lactobacillus Reference Stinson51 .
Enrichment of the Romboutsia and Blautia genera in lactation compared to the puberty period was observed. Romboutsia has already been identified in the human gut and was reported to be depleted in cancerous mucosa, representing a novel microbial biomarker associated with colorectal cancer Reference Mangifesta, Mancabelli and Milani52 . This depletion was also recently observed in pancreatic cancer patients Reference Kartal, Schmidt and Molina-Montes53 . According to these findings, the KL animals did not develop colon tumors, suggesting an association between Romboutsia and a possible protection against cancer. The Blautia genus is widely distributed in mammalian feces and intestines, and some species can use carbohydrates as energy sources with the production of acetic acid and long-chain fatty acids, among others Reference Liu, Mao and Gu54 . The enrichment of Blautia in mice fed fructooligosaccharide Reference Gu, Cui and Tang55 and its reduction in the mucosal adherent microbiota of colorectal cancer patients Reference Chen, Liu, Ling, Tong, Xiang and Moschetta56 have also been observed.
Here, although the Blautia anti-inflammatory effect, we observed a positive correlation with IL-6. Meanwhile, the cytokine IL-1 was significantly correlated with the number of tumors. IL-1 has a significant role in the pathogenesis of colorectal cancer Reference Cheng, Mejia Mohammed, Khong, Mohd Zain, Thavagnanam and Ibrahim57 , and it can be secreted by tumor cells and induce the recruitment of immunosuppressive cells Reference Xie, Zhang and Jiang58 . Furthermore, as shown in a previous study by the group, the combination of the two critical periods (lactation and puberty) enhanced the anti-inflammatory effects of kefir on the colonic mucosa, leading to a significant reduction in pro-inflammatory cytokines Reference de Brasiel P.G., Luquetti and Medeiros36 . The results presented here for treatment during independent periods (lactation or puberty) suggest that critical pathways for the anti-tumor immune response are developed in these phases Reference Hauer, Fischer and Borkhardt59,Reference Drijvers, Sharpe and Haigis60 , with the contribution of several cytokines Reference Yi, Li and Niu61 , and should be investigated in future studies.
In summary, the administration of kefir during the critical periods of development affected the gut microbiota community structure to promote host benefits with a reduction in colon tumor development. Maternal and postweaning kefir intake produced a distinct change in the richness and composition of the gut microbiota of colon cancer-induced offspring, where the lactation period was superior with respect to the programming of the gut microbiota of the adult offspring. These findings suggest that the maternal use of probiotic fermented food during lactation could be a strategy for colon cancer prevention in progeny, especially when there are associated risk factors.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S2040174424000461.
Acknowledgements
The authors are grateful to Ms. Kácia Mateus and Mr. João Pablo Fortes Pereira for technical assistance.
Author contribution
PGAB: Conceptualization, Methodology, Formal analysis, Investigation, Writing – Original Draft. JDM: Formal analysis, Writing – Original Draft. TCA: Investigation, Writing – Review & Editing. CTS: Investigation. GCAA: Investigation. ABFM: Visualization, Writing – Review & Editing. SCPDL: Conceptualization, Funding acquisition, Supervision, Writing – Original Draft.
Financial support
This work was supported by the “Fundação de Amparo à Pesquisa do Estado de Minas Gerais” – FAPEMIG [APQ-02397-21]; “Coordenação de Aperfeiçoamento de Pessoal de Nível Superior” – CAPES; and “Pró-Reitoria de Pós-Graduação e Pesquisa” – Propesq/UFJF.
Competing interests
None.
