The Intracellular Growth of M. tuberculosis Is More Associated with High Glucose Levels Than with Impaired Responses of Monocytes from T2D Patients

Diabetes mellitus, a metabolic disease characterized by hyperglycemia and poor glucose control, is a risk factor for Mycobacterium tuberculosis (M. tuberculosis) infection and the development of active tuberculosis. To evaluate whether M. tuberculosis infection susceptibility is associated with an intrinsic factor in monocytes from type 2 diabetes (T2D) patients or it is associated with hyperglycemia per se, we analyzed TLR-2 and TLR-4 expression by flow cytometry and the cytokines IL-1β, IL-6, IL-8, IL-10, and TNF-α by cytometric bead array assays, either stimulated with TLR-2 and TLR-4 ligands or infected with M. tuberculosis in the whole blood from T2D patients (n = 43) and healthy subjects (n = 26) or in CD14+ monocytes from healthy subjects cultured in high glucose (HG) (30 mM). The intracellular growth of M. tuberculosis was evaluated by CFU counts at 0, 1, and 3 days in both monocytes from T2D patients and monocytes from healthy subjects cultured in HG. We did not find significant differences in TLR expression, cytokine production, or growth of M. tuberculosis in monocytes from T2D patients compared with those in monocytes from healthy subjects. Despite these results, in vitro assays of monocytes cultured with 30 mM glucose led to significantly increased TLR-2 and TLR-4 basal expression compared to those of monocytes cultured with 11 mM glucose (P < 0.05). Conversely, the production of IL-6 by TLR-2 ligand stimulation, of IL-1β, IL-6, and IL-8 by TLR-4 ligand stimulation, and of IL-8 by M. tuberculosis infection significantly decreased in monocytes cultured in HG (P < 0.05). Additionally, the intracellular survival of M. tuberculosis increased in monocytes in HG after day 3 of culture (P < 0.05). In conclusion, HG decreased IL-8 production and the intracellular growth control of M. tuberculosis by monocytes, supporting the hypothesis that hyperglycemia plays an important role in the impaired immune responses to M. tuberculosis in patients with T2D.


Introduction
Diabetes mellitus (DM) represents a risk factor for Mycobacterium tuberculosis (M. tuberculosis) infection and the development of active tuberculosis (TB) [1]. Meta-analysis studies have shown that DM patients have an approximately 3-fold higher risk of developing TB disease [2], and TB control treatment is difficult in these patients [3]. The risk of developing active TB among diabetic patients is particularly high in Hispanic people, perhaps because latent TB infection is more common in these populations [4]. The frequency of DM among patients with pulmonary TB is 29.63%, and these patients have more severe clinical manifestations (i.e., a higher frequency of cavities on chest X-rays) and higher risks of recurrence and relapse than non-DM individuals with pulmonary TB [4], with poor glycemic control being a risk factor for TB [5].
DM is a metabolic disease characterized by hyperglycemia or high glucose (HG) concentrations in blood resulting from defects in insulin secretion, insulin action, or both. Hyperglycemia induces a chronic inflammatory state through the secretion of proinflammatory cytokines in type 2 diabetes (T2D) patients [2,6]. It is noteworthy that in spite of the activation state characteristic in these individuals, they still possess a higher susceptibility to developing infections by many pathogens, including M. tuberculosis [2] than non-T2D individuals, which makes it difficult to dissect and identify the participant mechanisms associated with the predisposition to infections seen in diabetic patients. Nevertheless, hyperglycemia has been suggested as an important factor in M. tuberculosis infection susceptibility [6].
In physiological conditions, the mechanism of inflammation through the activation of monocytes and macrophages results after pathogen recognition by protein receptors, such as the toll-like receptor (TLR) family. HG concentrations are able to upregulate TLR-2 and TLR-4 and activate nuclear factor-κB (NF-κB) to induce proinflammatory cytokine production in THP1 monocytes in the absence of pathogens [7]. TLRs play a critical role in both innate immune responses and the initiation of adaptive immunity to M. tuberculosis, specifically TLR-2 and TLR-4. TLR activation upregulates the transcription of proinflammatory cytokines, including IL-1β, IL-6, and TNF-α, which are essential for the recruitment of immune cells to the site of infection and the control of M. tuberculosis infection [8].
In monocytes from T2D patients, TLR-2 and TLR-4 expression is increased [9,10], and these patients present chronic low-grade inflammation with secretion of TNF-α and IL-8. The proinflammatory cytokines are induced by different mechanisms, such as TLR-2/4 activation (TLR-4 activation is induced by elevated exogenous ligands, such as dietary fatty acids and enteric lipopolysaccharide, and endogenous ligands, such as free fatty acids, which are elevated in obese states), reactive oxygen species (ROS) generation, or CD33 downregulation [9,[11][12][13]. Those mechanisms are important for pathogen control; however, the ROS and proinflammatory cytokines induced by HG do not protect against infections. In contrast, T2D patients are more susceptible than non-T2D individuals to infections, including M. tuberculosis infection, suggesting an immune susceptibility induced by HG concentrations.
The immunologic mechanisms of susceptibility to M. tuberculosis in T2D patients are still not fully understood. To investigate whether susceptibility to tuberculosis is due to an intrinsic alteration of monocytes from T2D patients, we used a whole-blood assay to preserve the physiological environment. To address whether the enhanced susceptibility to tuberculosis is due to HG concentrations per se, we used an in vitro model of hyperglycemia (using 30 mM glucose). In both assays, TLR-2 and TLR-4 expression, the cytokines released after TLR activation, and the growth control of M. tuberculosis were evaluated.

Study Population.
Forty-three T2D patients were recruited from the Metabolic Syndrome Clinic of the National Institute of Respiratory Diseases in Mexico City according to the diagnosis criteria of the American Diabetes Association (Diagnosis and Classification of DM) [14]. Twenty-six healthy subjects (based on clinical laboratory tests) were invited to participate in this study. The Ethics Committee of the National Institute of Respiratory Diseases (INER) approved this study, and all participants provided written informed consent. Routine laboratory tests on blood as well as chest X-rays and Mantoux tuberculin skin tests (TSTs) were conducted (Table 1). Additionally, buffy coats from healthy adult donors were kindly provided by the blood banks at the INER.
2.2. High-Glucose Medium. Unless otherwise specified, all cells were cultured in RPMI 1640 medium according to formula specification with 11 mM glucose (Cambrex, Walkersville, MD, USA) that was supplemented with 50 μg/ml gentamicin sulfate, 2 mM L-glutamine, and 10% heatinactivated pooled human serum (Gemini Bioproducts, Co., MD) on an orbital shaking incubator at 37°C. Mycobacteria were harvested and frozen at -80°C until use ("stock bacteria"). For the infection assay, "stock bacteria" were thawed, and to remove any remaining large clumps, bacterial pellets were homogenized by passing them through a 26-gauge needle and centrifuged. The bacterial concentration was determined by serial dilution, and colony-forming units (CFU) were counted on Difco Middlebrook 7H10 agar plates (Becton Dickinson Co, MD) after disruption of mycobacterial clumps. M. tuberculosis was dyed with fluorescein diacetate (Fluka, Chemical, Switzerland) in 1 ml of phosphatebuffered saline (pH 9.1) for 30 min at 4°C.

TLR Surface Expression by Flow Cytometry.
The whole blood from T2D and healthy subjects or CD14+ monocytes from healthy subjects cultivated in vitro with HG were stained with phycoerythrin-(PE-) labeled anti-human TLR-2, TLR-4, and CD14-FITC antibodies (eBioscience, San Diego, CA) or isotype controls. Cells were fixed with 1% paraformaldehyde, and TLR-2 and TLR-4 surface expression was analyzed in the monocyte region by flow cytometry. The results were analyzed as the percentage of positive cells and by mean fluorescence intensity (MFI) based on 10,000 events acquired in a FACSCanto II cytometer (BD). FlowJo software version 10.4.1 Mac OS X was used for analysis.
2.6. Cytokine Production in Response to TLR Ligands. One milliliter of whole blood from T2D and healthy subjects or 0:5 × 10 6 CD14+ monocytes/well placed in 24-well plates were cultured in HG (30 mM) for 24 h and infected with M. tuberculosis (MOI of 10) or stimulated with 1 ng/ml TLR-2 ligand, synthetic lipoprotein palmitylated N-acyl-S-diacylglyceryl cysteine (Pam 3 Cys; EMC Microcollections, Tuebingen, Germany), or 100 ng/ml TLR-4 ligand, Escherichia coli lipopolysaccharide (LPS; Sigma), and were further incubated for 4 h for T2D and healthy subjects or 24 h for monocytes in HG. Supernatants were collected, and interleukin-1β (IL-1β), interleukin-6 (IL-6), interleukin-8 (IL-8), interleukin-10 (IL-10), and tumor necrosis factor-α (TNF-α) levels were measured by a cytometric bead array (BD Bioscience) following the manufacturer's recommendation. CellQuest software was used for sample acquisition, and data were formatted using BD Bioscience CBA software. The results were based on a standard concentration curve, and the cytokine concentration was expressed as pg/ml.

Patients and Healthy Subjects.
The clinical and demographic characteristics of the T2D patients and healthy subjects are summarized in Table 1. No significant differences in sex, age, or levels of creatinine or HDL or LDL cholesterol between T2D patients and healthy subjects were observed; however, both groups were overweight. The levels of glucose, HbA1c, and triglycerides were significantly higher among T2D patients than among healthy subjects (P < 0:05).

The Expression of TLR-2 and TLR-4 in Monocytes from
T2D Patients Is Similar to That in Monocytes from Healthy Subjects. TLRs are essential for pathogen recognition. To analyze the expression of TLR-2 and TLR-4 cell surface receptors in monocytes from T2D patients, we evaluated the TLR-2 and TLR-4 expression in the monocyte region in the whole blood from T2D patients by flow cytometry. We do not find any differences in the T2D group vs. healthy subjects; because of this, we analyzed a subgroup with an HbA1c of 9% and above (poor glycemic control). As shown in Figures 1(a) and 1(b), TLR expression in T2D and healthy subjects was assessed on monocytes, and the cell surface expression of TLR-2 (median, 28.7 for T2D and 34.3 for healthy) and TLR-4 (median, 24.8 for T2D and 27.9 for healthy) was similar in monocytes from T2D and healthy subjects (Figures 1(c) and 1(d)). The percentage of monocytes expressing TLR-2 and TLR-4 in healthy and T2D patients was similar in both groups (data not shown).

High Glucose Concentrations Upregulate TLR-2 and TLR-4 Expression in Monocytes from Healthy Subjects.
Because we observed that the TLR expression in T2D patients was not affected, for a next hypothesis, we evaluated whether the glucose per se affects the main innate immune mechanisms for M. tuberculosis recognition. We analyzed TLR cell surface expression using an in vitro model of HG (30 mM or 540 mg/dl glucose). For this experiment, monocytes were cultured in HG for 24 h, and TLR-2 and TLR-4 expression was subsequently determined by flow cytometry. As shown in Figures 3(a)

High Glucose Reduces Intracellular M. tuberculosis
Growth in Monocytes. After phagocytosis, macrophages induce antibacterial mechanisms, including cytokine production, that inhibit the intracellular growth of M. tuberculosis in monocytes. We evaluated the intracellular M. tuberculosis growth in monocytes cultured in HG. Figure 4(d) shows an increase in intracellular M. tuberculosis growth in monocytes after 3 days of culture with HG with respect to that in monocytes with RPMI medium (P < 0:05) (median, RPMI medium: 1:2 × 10 3 -D0, 1:7 × 10 3 -D1, and 3:0 × 10 3 -D3 and HG: 1:2 × 10 3 -D0, 3:4 × 10 3 -D1, and 1:1 × 10 3 -D3). TLR expression, proinflammatory cytokine production, and intracellular growth of M. tuberculosis in monocytes from T2D patients to assess the intrinsic impairment of monocytes in M. tuberculosis recognition and control. We also used an in vitro hyperglycemia model (30 mM glucose) to assess whether high glucose impaired the control of M. tuberculosis.

Discussion
In this population, we did not find a significant difference in TLR-2 and TLR-4 expression, cytokine production, or intracellular growth of M. tuberculosis in T2D patients compared to healthy subjects. However, using the in vitro model of hyperglycemia at 30 mM glucose, we found that cytokine production was diminished after TLR-2 and TLR-4 activation and that the intracellular growth of M. tuberculosis in human monocytes was not controlled. TLRs are an important mechanism for the recognition of M. tuberculosis; TLR-2 and TLR-4 play critical roles in TB infection and are the immediate mechanism for the recognition of pathogens [15,16]. Also, TLRs are important regulators of the immune and metabolic systems, including in DM [17]. However, the role of TLRs in diabetes is controversial; these apparent discrepancies could be explained by the population, experimental design, and type of cell. Dasu et al. report an increased TLR-2 and TLR-4 expression in monocytes from recently diagnosed type 2 diabetic patients [9]. Ahmad et al. report an increased expression of TLR-2 and TLR-4 in PBMCs from obese and overweight individuals with type 2 diabetes [18]. Gupta et al. report no differences in TLR-2, TLR-4, and TLR-6 levels in monocytes from patients with good glycemic control when compared to nondiabetic volunteers [19]. In addition, it was reported that the expression of TLRs did not differ significantly between T2D patients and healthy voluntaries, although their monocytes are less responsive to TLR ligands [20]. Using the whole blood assay to preserve physiological environment and avoid possible artificial stimulation of monocytes, Khondkaryan et al. evaluate the expression of TLR-2 and TLR-4 on the total monocyte surface in culture from the whole blood and report significantly reduced TLR-4 expression in unstimulated T2DM monocytes compared to respective cells from healthy donors [21]. In concordance with Khondkaryan et al., we observed that T2D patients express TLR-2 and TLR-4 in monocytes at the same level as healthy subjects; although the healthy population was overweight, a previous report demonstrated that subjects with T2D had significantly elevated mRNA levels of TLR-2 and TLR-4 compared with nondiabetic obese subjects [18]. These contradictory results show that the levels of TLR-2 and TLR-4 are not a hallmark of T2D patients and the levels of TLR expression are rather associated with factors such as glucose levels.
Additionally, in in vitro studies, HG increased the expression of TLR-2 and TLR-4 in THP1 cells and primary monocytes from healthy subjects via protein kinase C (PKC) [7]. We also observed the in vitro effect of HG on the upregulated expression of TLR-2 and TLR-4 on the cell surface of primary monocytes from healthy subjects, similar to recently diagnosed type 2 diabetic patients [9], supporting the upregulation of TLRs in the presence of HG concentrations.
After pathogen recognition, TLRs activate signaling cascades via NF-κB to induce proinflammatory responses. The role of hyperglycemia as a proinflammatory factor was demonstrated by in vitro models. In THP-1 monocytes, HG concentrations (15 mM) induce the production of TNF-α, interleukin-(IL-) 1β, chemokines, such as monocyte chemoattractant protein-1 (MCP-1), and interferon gamma inducible protein-10 (IP-10), many of which are NF-κB regulated [22]. Previous reports on models of diabetic rats infected with M. tuberculosis showed a reduced production of TNF-α, IFN-γ, and IL-12 by alveolar macrophages [23]. In contrast, in patients with diabetes who had poorly controlled blood glucose levels, TNF-α, IFN-γ, IL-2, and IL-1β cytokine levels were increased in response to PPD [24]. However, the response was different when we used the live M. tuberculosis H37Ra strain. We observed similar levels of TNF-α, IL-1β, IL-6, and IL-8 cytokine production, whereas in response to Pam3Cys, a synthetic ligand of TLR-2, only IL-10 production was reduced in T2D patients. A similar response was reported in diabetic mice, with low levels of IL-10 expression in macrophages from diabetic mice [25]. Additionally, HG decreases the release of IL-1β, IL-6, and IL-8 in LPS-stimulated monocytes cultured in the presence of HG. These results are partially consistent with Beitland et al. who found that IL-1β production was significantly reduced in LPS-stimulated blood samples in the presence of 20 mmol/l glucose [26]. However, the TLR responses differ in the type and magnitude of the innate immune responses they elicit when interacting with different ligands. We used live M. tuberculosis to evaluate the response by monocytes in HG and observed that IL-8 production was significantly reduced. Interleukin-8 (IL-8) displays two major biological activities: chemoattraction and activation of several types of white blood cells. IL-8 plays a central role in the normal immune response to M. tuberculosis and has been shown to be absolutely required for granuloma formation. Monocytes and macrophages infected with M. tuberculosis are the primary producers of IL-8 during the course of TB disease, an inability to stimulate the production of IL-8 corre-lated with a poor prognosis in patients with TB [27], and the reduction in IL-8 production by HG could participate in tuberculosis susceptibility.
It has been well established that recognition and phagocytosis of M. tuberculosis induce the production of antimicrobial peptides, which are important for the control of the intracellular growth of pathogens, including M. tuberculosis [28]. We observed that the growth of M. tuberculosis increased in monocytes cultured in HG concentrations. The reduced cytokine production observed in this study suggests inadequate activation through these receptors, which may be a putative mechanism of the reduction in TLR-mediated M. tuberculosis control. These findings are consistent with previous studies in animal models of diabetes, in which mice infected with M. tuberculosis did not control the infection and died in a short time [29]. Additionally, diabetic rats (with HG levels in their urine) had a large number of granulomas, with a high Mycobacterium load [23]. Our study presents evidence of the role of hyperglycemia in the mechanism associated with M. tuberculosis intracellular control.

Conclusion
TLR-2 and TLR-4 expression, cytokine production, and the intracellular growth of M. tuberculosis are not associated with an intrinsic factor in monocytes from T2D patients; however, the presence of high glucose concentrations decreased cytokine production and increased the intracellular growth of M. tuberculosis in monocytes, suggesting that hyperglycemia at high levels may play a key role in altering the immune responses to M. tuberculosis in diabetic patients.

Data Availability
The research data used during this study are included in the paper. The clinical and demographic data for each patient or healthy subject analyzed in this study are available from the corresponding author upon request.

Ethical Approval
The protocol was approved by the Ethics Committee in Research of the Instituto Nacional de Enfermedades Respiratorias Ismael Cosío Villegas with reference number B20-12. and wrote the manuscript. All authors reviewed and approved the final manuscript.