Biochemical analysis of microbiotas obtained from healthy, prediabetic, type 2 diabetes, and obese individuals

Objectives: In this study, we aimed to evaluate the intestinal and urinary microbiota diversity of obese, pre-diabetic, diabetic, and healthy subjects together with their food consumption frequency and investigate the effect on glucose metabolism. Methods: DNA was isolated from stool and urinary samples of fifteen obese, fifteen prediabetics, fifteen type 2 diabetic, and fifteen lean participants by using the quantitative real-time polymerase chain reaction (qPCR) method. The amounts of Bi ﬁ dobacterium, Bacteroides , and Firmicutes were measured and food consumption frequency was answered by all participants through a questionnaire. Results: The levels of Bi ﬁ dobacterium in fecal microbiota weresigni ﬁ cantlyhigherintype2diabeticpatientscompared with lean (p=0.034), prediabetic (p=0.009), and obese participants (p=0.012). However, the levels of Bi ﬁ dobacterium in urinary microbiota were decreased in obese, prediabetic, and type 2 diabetic subjects as controls (p=0.048; p=0.038; p=0.015 respectively). Additionally, Bacteroides / Firmicutes ratio decreased in type two diabetic patients compared with lean subjects and had a negative correlation with BMI in prediabetic subjects. Food consumption frequency illustrates that lean subjects have unhealthy eating habits. Conclusions: Urinary microbiota could be considered in the future context of a potential biomarker in the progress of type 2 diabetes and obesity. patientsandcontrolgroupswerenotuniformlydistributed,theKruskal-WallisH test wasused todetermine the differences between the groups. Pairwise comparison of parameters between the groups was done with the Mann-Whitney U test. Linear correlation between the parameters was evaluated by Spearman ’ s correlation test. A p-value of <0.05 was considered statistically signi ﬁ cant.


Introduction
Type 2 diabetes and obesity are metabolic disorders has become major public health concerns. Consumption of energy-dense foods, lack of physical activity, and genetic predisposition might lead to diabetes and obesity [1,2]. Additionally, the microbiota has also been identified as a factor that contributes to obesity and type 2 diabetes [3,4]. Components of microbiota are mostly bacteria, with a minority of viruses, fungi, and eukaryotic cells. The most abundant phyla in both humans and mice are Firmicutes, Bacteroidetes (particularly including Bacteroides), and Actinobacteria (with a predominance of the genus Bifidobacterium) [5]. Dysbiosis of gut microbiota is the cause of the low-grade inflammation that plays a major role in the onset of T2D [6,7]. Recent studies on humans have shown that a higher proportion of Firmicutes and a lower proportion of Bacteroidetes are related to obesity and diabetes [4]. Lower Firmicutes: Bacteroidetes ratio correlated with lower BMI in obesity groups compared to lean groups [4,8]. On the other hand, the ratio of Bacteroidetes to Firmicutes in diabetic compared to non-diabetic subjects is not correlated with BMI [1]. Microbiota composition is considerably influenced by factors such as lifestyle, diet, age, seasonal variations, and geography [9,10]. Food components are the substrates for intestinal microbial metabolism. The end products of bacterial metabolism are predominantly vitamins and short-chain fatty acids that have various functions in lipid, glucose, and cholesterol metabolism [11]. The World Health Organization's definition of probiotics is "live microorganisms which, when administered in adequate amounts, confer a health benefit on the host" [12]. Several recent reviews and meta-analyses, including meta-analyses of randomized controlled trials (RCTs), have suggested the use of probiotics for glycemic benefits in T2D and obesity [13]. Bifidobacteria are one of the most numerous probiotics found in the mammalian gut and are a type of lactic acid bacteria. Studies of T2D with Bifidobacterium have reported that these microbes are potentially protective against T2D and obesity [14]. The studies of urinary microbiota are mostly carried out on women with urgent urinary incontinence (UUI) [15,18]. These studies demonstrated, that urinary microbiota diversity changes with age, body mass index, fasting blood glucose (FBG), and urine glucose (UGLU) [15]. Although urinary microbiota studies are mostly on female subjects, we want to demonstrate urinary microbiota dysbiosis independent of gender, identifying the relationship between obesity and diabetes.
According to the American Diabetes Association (ADA), Diabetes may be diagnosed based on plasma glucose criteria, either the fasting plasma glucose (FPG≥126 mg/dL (7.0 mmol/L)) value or the 2-h plasma glucose (2-h PG≥200 mg/dL (11.1 mmol/L)) value during a 75-g oral glucose tolerance test (OGTT), or A1C (A1C≥6.5% (48 mmol/mol)) criteria. People with prediabetes are defined by the presence of impaired fasting glucose (IFG) and/or impaired glucose tolerance (IGT) and/or A1C 5.7-6.4% (39-47 mmol/mol) [16]. Moreover, obesity is a BMI greater than or equal to 30 [17]. Microbiota may be the alternative diagnostic biomarker of these criteria in the progress of type 2 diabetes and obesity.
This investigation aimed to characterize fecal and urinary microbiota in obese, prediabetic, and diabetic patients, as well as healthy subjects. Bifidobacterium, Bacteroides, and Firmicutes were tested by real-time PCR to understand changes in gut microbiota in all groups. Food frequency questionnaire results have shown that dysbiosis of microbiota can affect nutrition and glucose metabolism.

Collection of fecal and urine samples and DNA isolation
The study included newly diagnosed type 2 diabetes (n:15), prediabetes (n:15), obese (n:15), and lean subjects (n:15). The subjects were selected based on the following criteria: female or male aged between 18 and 65 years, body mass index (BMI) 18.5-24.9 for healthy participants, and ≥30 for obese patients, and subjects who were newly diagnosed type 2 diabetes. Exclusion criteria are ınflammatory bowel disease (such as Crohn's disease, ulcerative colitis) and colorectal carcinoma, other chronic diseases other than obesity and diabetes (hypertension, chronic kidney failure, chronic liver disease, hypo/ hyperthyroidism, coronary artery disease, polycystic ovarian syndrome causing insulin resistance, acanthosis nigricans, lipoatrophy/ lipodystrophy syndromes, etc.), medical treatment including antibiotics and oral contraceptive treatment in the last 3 months, alcohol consumption and the absence of gastrointestinal disease and bowelrelated operations in the last 3 months. Moreover, those who were pregnant and breastfeeding were excluded.
Fecal samples were collected by fecal collectors and transferred into sterile tubes. In the same manner, urine samples were collected into sterile tubes. Until analysis, all samples were kept at −20°C.
For DNA isolation, 150 mg of each fecal sample and 500 µL of each urine sample were taken in tubes separately. Then, total bacterial DNA was isolated from all fecal and urine samples using the Higher Purity Bacterial DNA Extraction kit (Canwax Biotech, Spain) according to the manufacturer's instructions. The DNA concentration in the extracts was determined by a fluorometer (Qubit ® 3.0, Thermo Fisher Scientific, USA) with Qubit™ dsDNA HS Assay Kit. Extracted DNAs were stored at −40°C until real-time qPCR analysis.

Primer and probe designing
Primers and probes for the amplification of Firmicutes, Bacteroides, and Bifidobacterium used in the present study target the 16S rRNA gene. Primer and probes were designed by using the websites ncbi.nlm.nih. gov and https://eu.idtdna.com/calc/analyzer. The specific sequences of primers and probes are shown in Table 1.

Real-time qPCR
Bacterial abundance in fecal and urine samples was quantified by realtime qPCR using the real-time PCR system (Light Cycler 480, Roche,

Statistical analysis
The statistical analyzes of the findings obtained in the study were performed by the IBM SPSS Statistics Version 22.0 program. Results were expressed as mean value ± standard deviation. As the parameters (abundance of gut microbiota and clinical parameters) belonging to the patients and control groups were not uniformly distributed, the Kruskal-Wallis H test was used to determine the differences between the groups. Pairwise comparison of parameters between the groups was done with the Mann-Whitney U test. Linear correlation between the parameters was evaluated by Spearman's correlation test. A p-value of <0.05 was considered statistically significant.

Characteristics of volunteers
The baseline characteristic of the 60 volunteers is shown in Fasting blood glucose progressively increased in healthy, obese, prediabetic, and type 2 diabetic groups, respectively. Pairwise comparisons of all groups were statistically significant. The levels of HDL-cholesterol decreased in the prediabetic group compared to the healthy group (p=0.026). The same situation was observed in type 2 diabetic patients when healthy and type 2 diabetic groups were compared (p=0.029). Triglyceride levels were significantly higher in prediabetic and type 2 diabetic groups than in the healthy group (p=0.009; p=0.007 respectively) while there were no significant differences in cholesterol and LDL-cholesterol levels between all groups. Furthermore, the levels of HbA1c were statistically significant found in pairwise comparisons of all groups except the state that was compared with healthy and obese groups. BMI was statistically higher in three groups than in the healthy group while ıt was lower in prediabetic and type 2 diabetic groups in comparison to the obese group.

Quantitative PCR analysis of bacteria
The levels of Firmicutes, Bacteroides, and Bifidobacterium were determined to evaluate the differences in the composition of fecal and urine samples of obese, patients, and healthy groups.
The abundance of Firmicutes and Bacteroides in fecal and urine was not significantly different between all groups instead for type 2 diabetic and prediabetic groups. The abundance of Firmicutes in type 2 diabetic subjects was higher than in prediabetics (p=0.038). However, the levels of Bifidobacterium in gut microbiota were significantly higher (p=0.020) in type 2 diabetic group compared with the healthy group (p=0.034), and the same situation was observed in the type 2 diabetic group compared with both the obese group and prediabetic group (p=0.012; p=0.009, respectively) ( Table 3). Additionally, the levels of Bifidobacterium in urinary microbiota were significantly decreased in the obese, prediabetic, and type 2 diabetic groups compared to the healthy group (p=0.048; p=0.038; p=0.021 respectively) ( Table 3). Also, Bacteroides/Firmicutes ratio was compared with BMI in all groups but there was only a significantly negative correlation in the prediabetic group ( Figure 1).

Food consumption frequency
We questioned the food consumption frequency of the subjects by considering the foods such as dairy products, meat, eggs, legumes, vegetables, fruits, bread, cereals, desserts, and drinks frequently consumed in Turkish society. Seven options were given to the subjects for the frequency of consumption. These are daily, 5-6 times a week, 3-4 times a week, once or twice a week, every 15 days, once a month, and never consumed. One bread is accepted as 200 g according to the Turkish food codex (2021). Additionally, yogurt is known according to the Turkish food codex (2021) [19], specifically Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus fermented milk product in which symbiotic cultures are used. On the other hand, there was no standardization of homemade yogurt and homemade yogurt consumption has become increasingly prevalent in Turkish society. Because of that, we want to research if there was any effect on the microbiota diversity. When the food frequency consumption of obese, prediabetic, type 2 diabetic and healthy individuals in Turkey were examined. It was seen that 31 of the 60 volunteers included in the study consumed homemade yogurt and the frequency of consumption was highest in healthy individuals (57.1% of healthy individuals consume it daily). While the number of individuals consuming industrial yogurt was close to the number of subjects consuming homemade yoğurt. However, industrial yogurt was consumed most frequently by obese subjects (33.3% of obese subjects consume it daily). Cheese consumption frequency was very high in all groups and 86.7% of obese persons, 93.3% of prediabetics, 93.3% of type 2 diabetics, and all healthy people consumed cheese every day. We observed that the frequency of consumption of potatoes, bulghur, rice, and pasta among subjects with obese, prediabetic, and type 2 diabetes was concentrated 1-2 times a week in consumption. The same situation was observed in the frequency of red and white meat consumption as well. Kefir is known as probiotic food consumed by only six people and has a low frequency of consumption. Additionally, physical activity and smoking knowledge are obtained from the participants (Supplementary Figures S1 and S2 respectively). When Supplementary Figure S1 is examined, It is seen that type 2 diabetic participants are the most sportive group and healthy participants are the least sportive group. Evaluation of smoking data concluded that healthy individuals are the most frequent cigarette users in comparison to all groups. All subjects consume bread every day. Interestingly, it was observed that type 2 diabetic patients consume one bread every day, and daily bread consumption was highest in type 2 diabetic, obese, prediabetic, and healthy individuals, respectively (Supplementary Figure S3).

Discussion
This study focused on the diversity of intestinal and urinary microbiota on obese, prediabetic, type 2 diabetic, and healthy subjects. Bacterial phylum Firmicutes and two bacterial groups including Bacteroides and Bifidobacterium are analyzed in obese, prediabetic, and type 2 diabetes patients which were then compared to healthy groups as well as each other.
In this study, there was no significant difference in abundances of Firmicutes and Bacteroides in fecal samples of the type 2 diabetic, obese and prediabetic groups compared to the healthy group. Only, Firmicutes was  slightly increased in type 2 diabetic patients compared to prediabetics. In addition, we found a significant negative correlation between Bacteroides/Firmicutes ratio and BMI in the prediabetic group. Similar to previous a study identifying five phyla including Bacteroidetes, Firmicutes, Proteobacteria, Verrucomicrobia, and Actinobacteria in type 2 diabetic and prediabetic subjects, no differences in the abundance of Bacteriodetes and Firmicutes were observed [20]. On the other hand, Remely et al. compare type 2 diabetes and healthy individuals and demonstrate type 2 diabetic individuals became more abundant in terms of Firmicutes, especially the subgroup Clostridiales, Bacilli, and Lactobacillales while the abundance of Bacteroidetes was less more [4]. Also, few studies reported higher concentrations of Bacteroidetes and a lower rate of Firmicutes and Bifidobacterium in the fecal microbiota of patients with diabetes compared with the healthy subject [1,6]. These studies hypothesize that there is a connection between metabolic diseases and gram-negative bacteria in the gut. It has been indicated that gram-negative bacteria specifically those belonging to the phylum Bacteroidetes and Proteobacteria are the source of lipopolysaccharide (LPS) which causes metabolic endotoxemia and insulin resistance in obese and type 2 diabetic subjects [1,20].
Bifidobacterium is a group of bacteria that have a beneficial effect on health, and a decrease in the microbiota composition of beneficial bacteria such as Bifidobacterium has been associated with diabetes [6]. The remarkable and interesting result is that the level of Bifidobacterium has been increased in type 2 diabetic groups. We observed that the type 2 diabetic group had higher levels of Bifidobacterium compared to healthy, obese, and prediabetic groups. Our results are not consistent with the reports that find a reduced abundance of Bifidobacteria in type 2 diabetics [6,21].
In our study, we did not observe any significant differences in abundances of Firmicutes, Bacteroides, and Bifidobacterium in fecal samples of obese patients compared to the healthy and other groups. Our results following the report by Duncan et al. were no differences between obese and nonobese subjects in the number of Bacteroides measured in fecal samples. Despite applying weight-loss diets, there was no significant change in the percentage of Bacteroides in feces, while Firmicutes decreased [22]. Our observations differ from previous studies which support the hypothesis that the proportions of Bacteroidetes and Firmicutes are different between obese and lean subjects.
Many studies that investigate urinary microbiota have shown that urinary microbiota composition is complex and it significantly differs among individuals, in both males and females [23,24]. The literature has limited studies researching the relationship between urinary microbiota and metabolic diseases such as type 2 diabetes. This present study is the first report analyzing the urinary microbiota in obese subjects, prediabetic and type 2 diabetes simultaneously. According to our results, Bifidobacterium abundance was significantly decreased in all groups when compared to healthy controls, while the levels of Firmicutes and Bacteroides were not significantly different. There are two reports which compared the urinary bacterial diversity profile in women with type 2 diabetes and healthy subjects, demonstrating that urinary microbiota with type 2 diabetes was dominated by Proteobacteria, Firmicutes, Bacteroidetes, Actinobacteria. In these studies, They have stated that Flavobacteriales, Actinobacteria, and Flavobacteria could be used as potential distinguishing biomarkers, and Akkermansia, and Bifidobacterium notably was detected in all four cohorts [15,25]. Bifidobacterium is considered a probiotic bacteria due to its health-promoting beneficial effects in humans [26,27]. The presence of Bifidobacterium in urine may be explained by the idea that Bifidobacterium has a protective effect in urine and may play a role in the healthy urine microbiome. However, it has been reported that Bifidobacterium may be the pathogenic agent in urinary tract infection, sepsis, and necrotizing pancreatitis [25]. In our study, it is shown that urinary microbiota may affect glucose metabolism with a decrease of Bifidobacterium in type 2 diabetes. Moreover, for urinary microbiota dysbiosis, more studies including larger bacterial diversity and parameters such as urinary glucose level, LPS, etc., are required.
The microbial composition is influenced by various factors such as genetics, age, diet, geographic origin, lifestyle, regular medications, and using antibiotics [26]. Growing evidence shows that especially diet impacts microbiota composition and could be related to obesity and metabolic diseases such as diabetes [28,29]. Carbohydrates, especially non-digestible carbohydrates (such as dietary fiber and resistant starch), have a great influence on the composition and diversity of the gut microbiota [30]. In addition, since these carbohydrates may have prebiotic effects, they can regulate the activity of bacteria such as Bifidobacterium and Lactic acid bacteria [31]. When analyzing the yogurt consumption of participants, we observed that 40% of obese individuals, 66% of prediabetic patients, 66% of type 2 diabetic patients, and 46% of healthy individuals consumed homemade yogurt. Considering the results of food consumption frequency, increasing the consumption of probiotic foods such as yogurt and kefir may have a prevention effect on the progression of diabetes and obesity. In addition, the consumption of foods containing carbohydrates such as bread, pasta, potatoes, and rice can be reduced under the control of a dietitian. They may convert their microbial compositions in favor of beneficial bacteria. On the other hand, healthy participants demonstrate that if their less physical activity and consumption of potatoes, rice, pasta, and desserts like the type 2 diabetic patient could not change, diabetes risk would increase with increasing age. When healthy, obese, and prediabetic participants are compared, with increasing BMI and age, alteration of the Bacteroides/ Firmicutes ratio is observed. Healthy participants had a highrisk factor for type 2 diabetes with less physical activity, smoking, and unhealthy eating habits.

Conclusions
In conclusion, the results of our study demonstrated that Firmicutes and Bacteroides did not differ in fecal and urinary microbiota of three different study groups composed of obese, prediabetic, and healthy individuals. The abundance of Bacteroides did not change but the abundance of Firmicutes increased in type 2 diabetes compared with prediabetic subjects. However, Bifidobacterium has undergone different alterations in both fecal and urinary microbiota. Results of food consumption frequency illustrate that obese, prediabetic, and type 2 diabetic individuals have similar dietary habits. Also, healthy subjects have an unhealthy lifestyle and food consumption. If a new treatment is identified by microbiota analysis for prediabetic or obese patients, these results may help develop new strategies to prevent type 2 diabetes in the future.
Research funding: This study was supported by the YTU, Scientific Research Project Coordination with the project no: FBA-2018-3410. The authors thank the YTU Scientific Research Project Coordination and Okmeydanı Training and Research Hospital. Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission. Competing interests: The authors declare that there is no conflict of interest. Informed consent: Informed consent was obtained from all individuals included in this study. Ethics Committee Approval: The study protocol was approved by the Ethical Committee (Decisions numbers: 48670771-514.10) and performed according to the 1964 Helsinki declaration. Both written and verbal consent was obtained from all subjects of the study.