Introduction

Vitamin D (VitD) is not a traditional vitamin, but a precursor substance to hormones. VitD is a lipid-soluble open-loop steroid, including animal-derived vitamin D3 [cholecalciferol] and plant-derived vitamin D2 [ergocalciferol] [1]. VitD in the human body is mainly synthesized by the skin and a small amount is obtained from food. The synthesis of 25 hydroxyvitamin D by 25 hydroxylase is the main storage form of VitD in vivo, which reflects the nutritional status of vitamin D in vivo. 25 hydroxyvitamin D [25(OH)D] is hydroxylated to 1,25-dihydroxyvitamin D[1,25(OH)2D] by alpha and is the main active metabolite of VitD in vivo. It binds to VitD receptors widely existing in tissues and plays a hormone-like role, also known as D hormone. Therefore, VitD is also considered a prohormone[2]. The classical function of VitD and its metabolites is to promote the absorption of calcium and phosphorus, inhibit the release of parathyroid hormone, maintain the normal level of blood calcium and phosphorus, thereby ensuring bone health and normal neuromuscular function, and prevent rickets and osteoporosis[3]. Further studies found the non-classical effects of VitD include cardiovascular, metabolic, immune regulation, tumorigenesis, and many other effects[4].

The detection of serum 25(OH)D level has been recognized as the most reasonable indicator of VitD status. The application of VitD and its related preparations (or analogues) has fundamentally curbed the widespread prevalence of rickets/osteomalacia worldwide[5, 6]. So what is the normal range of vitamin D? At present, most international and domestic institutions and experts believe that serum 25(OH)D < 20 ng/ml is VitD deficiency, 20 ~ 30 ng/ml VitD is insufficient, > 30 ng/ml is sufficient VitD, and < 10 ng/ml is severe deficiency. However, VitD deficiency and undernutrition are still prevalent in the population. There are more than 1 billion people in the world whose serum 25(OH)D level does not reach to the recommended level of 30 ng/ml (75 nmol/l)[3].

Research reports in recent years, many chemicals similar to VitD (active vitamin D analogues) have been developed and applied in the clinic, especially in osteoporosis, rickets, chronic kidney diseases, and skin diseases[7, 8]. With the discovery of VitD receptors (VDR) and 1α-hydroxylase (CYP27Bi) in many extraskeletal tissues, the role of VitD is no longer limited to regulating calcium and phosphorus metabolism, and its role in muscle, cardiovascular disease, diabetes, and autoimmune and inflammatory responses has also gradually attracted attention[9,10,11].

Population studies have shown that VitD deficiency is related to the increased incidence of T2DM patients[12]. Many studies have confirmed VitD level was independently correlated with insulin sensitivity and islet β cell function with T2DM patients and metabolic syndrome[13, 14]. The prevalence of VitD deficiency in T2DM patients is more than 80%. Those with higher 25(OH)D concentration had lower blood glucose levels at fasting and 2 h after glucose loading[15]. Longitudinal cohort studies and meta-analysis showed a correlation between higher VitD intake and lower risk of T2DM[16]. Observational studies have also revealed that VitD deficiency is associated with chronic complications of diabetes, such as diabetic retinopathy[17]. However, there are no definite results on whether VitD supplementation can reduce or prevent the occurrence of T2DM[18].

Studies have confirmed that the proportion of M1 macrophages and M2 macrophages in patients with T2DM is imbalanced[19, 20]. Reduced macrophage tolerance to dietary proteins and resident commensal microbiota is thought to contribute to chronic inflammatory diseases. Macrophages are a heterogeneous group of immune cells that exhibit significant plasticity and change their physiological functions under various microenvironmental stimulation. These changes can produce different cell populations with different functions: the classically activated macrophages (M1) and the alternatively activated macrophages (M2). M1 polarized macrophages are characterized by high expression levels of CD86, the ability to produce large amounts of pro-inflammatory cytokines, such as IL-1β, IL-6, and TNF-α. In contrast, M2 polarized macrophages are characterized by increased expression level of CD206, CD163, and CD209. The anti-inflammatory cytokines secreted by M2 macrophages include TGF-β and IL-10[21].

M1/M2 describes the two major and opposing activities of macrophages. M1 activity inhibits cell proliferation and causes tissue damage, whereas M2 activity promotes cell proliferation and tissue repair. Furthermore, 1,25-dihydroxyvitamin D3 has been reported to promote macrophage polarization from a pro-inflammatory phenotype (M1 macrophages) to an anti-inflammatory phenotype (M2 macrophages), and inhibited the expression of pro-inflammatory cytokines by monocytes and macrophages.

However, there is currently no definitive conclusion about the relationship between plasma VitD levels and macrophage typing in patients with T2DM. Why do T2DM patients often have low serum 25(OH)D levels? Whether VitD affects the progress of T2DM by regulating the polarization of macrophages and the specific regulatory mechanism is not very clear. Therefore, we launched the following research.

Materials and methods

Populations and clinical data

Firstly, we collected the peripheral blood (EDTA anticoagulation) of T2DM patients with different vitamin D levels, including VitD deficiency (< 10 ng/ml), VitD insufficiency (10 ~ 30 ng/ml), and VitD sufficiency (> 30 ng/ml) groups for flow cytometry. There are 3 groups in total, and 68 cases were finally enrolled in this study. All participants were selected from center for endocrine metabolism and immune diseases, Beijing Luhe Hospital, Capital Medical University between November 2017 and January 2019. We selected the target population based on the 25-hydroxyvitamin D3 levels in the results of the three examination items of bone metabolism or the five examination items of bone metabolism in T2DM patients. All subjects had background of type 2 diabetes mellitus and had no history of autoimmune diseases, including osteoporosis, rickets, chronic kidney, skin diseases, or other bone metabolic diseases. No active VitD and its analogues were supplemented in all participants (Table 1). Personal information and laboratory data including age, sex, serum 25 (OH)D, glucose (GLU), glycosylated hemoglobin (HbA1c), and white blood cell (WBC) count were recorded by experienced rheumatologists.

Table 1 Clinical characteristics of the populations studied
Full size table

Sample preparation

We enrolled according to 25(OH)D level of T2DM patients. A portion of peripheral blood samples (EDTA anticoagulation) was collected and isolated PBMCs for flow cytometry. The remaining peripheral blood were centrifuged and plasma separated and frozen at − 80℃ for ELISA.

Preparation sample of PBMCs

We first collected the fresh peripheral blood of patients about 2 ml, took 500 μl, then gently diluted in 500 μl PBS and peripheral blood mononuclear cells (PBMCs) were isolated through Ficoll-PaqueTMPREMIUM sterile solution (density 1.077, GE Healthcare, 17–5442-02, Sweden) gradient centrifugation based on the manufacturer’s protocol.

Flow cytometric analysis

All data were acquired using the BD FACSCantoII Flow Cytometer and 100,000 events per sample were recorded. Acquired data were subsequently analyzed with FACSDiva Analysis Software (BD, USA). In addition, titration was performed to determine the concentration of optimal antibody staining. During the analysis, isotype control was used to exclude non-specific antibody binding. PBMC is distinguished from debris by its forward scattering and side scattering. Cell surface CD163 (label: FITC, Lot: 8,043,562, BD Bioscience), CD206 (label: APC, Lot: 8,005,527, BD Bioscience), and CD209 (label: Percp-Cy5.5, Lot: B250672, Biolegend) were used to differentiate M2 macrophage. CD86 (label: PE-Cy7, Lot: 7,202,711, BD Bioscience) was used to label M1 macrophages. CD11b-PE was used to label the mononuclear-macrophage system.

Separation and purification of CD14 + monocytes from peripheral blood samples by immunomagnetic beads

We first isolated PBMC from the peripheral blood of T2DM patients. Next, the PBMCs were washed twice with MACS buffer and counted, and CD14 + cells are magnetic with CD14 MicroBeads (human CD14 MicroBeads, Lot#: 130–050-201, MACS MiltenyiBiotec). Then, the cell suspension is loaded onto a MACS LS Column which is placed in the magnetic field of a MACS Separator. Finally, CD14 + cells can be enriched by using LS Columns. We cultured the CD14 + monocytes after sorting for 7 days in RPMI 1640 (Gibco, USA) medium supplemented with 10% fetal bovine serum (FBS) (Gibco, USA), 100 U/ml recombinant human M-CSF (MCE, HY-P7050A), 100 U/ml penicillin, and 100 µg/ml streptomycin to generate immature macrophage.

Flow cytometry analysis of mononuclear cell separation purity

We took PBMC before sorting, CD14 + monocytes obtained after sorting, 1 × 104 cells each, and added 2.5 µl human CD14-APC antibody (Lot: B262993, Biolegend), and used anti-mouse IgG monoclonal antibody as an isotype control, incubated at 4℃ for 20 min in the dark, after washing, the cells were suspended in the washing solution, and the rate of CD14 positive cells before and after sorting was detected by flow cytometry. The results were analyzed by FlowJo_V10 software.

Plasma sample preparation and enzyme-linked immunosorbent assay (ELISA)

Peripheral venous blood (1.5 ml) was collected into EDTA anticoagulant tubes, and then the upper plasma layer was extracted after centrifuging and stored at − 80℃. The levels of IL-6 and TGF-β in fasting plasma were measured by ELISA following the manuals (IL-6: Human IL-6 Quantikine ELISA Kit, Lot#: D6050, R&D Systems; TGF-β: Human TGF-beta 1 DuoSet ELISA, Lot#: DY240, R&D Systems).

Statistical analyses

Statistical analysis was performed using the GraphPad Prism 5 software and SPSS 18.0. All data were presented as means ± SEM. Unpaired, 2-tailed Student’s test and Fisher’s exact test were used to compare the means and the proportions between high VitD group and low VitD group. The three groups were compared by ANOVA. p < 0.05 was considered statistically significant.

Results

The clinical and metabolic characteristics of the enrolled population

The baseline clinical and biochemical parameters of all patients with T2DM according to serum 25(OH)D are summarized in Table 1.

With the increase of serum 25(OH)D level, the polarized phenotype of peripheral blood macrophages tended to be M2 type, that is, M1/M2 decreased

In order to explore the correlation between vitamin D levels and macrophage types in patients with type 2 diabetes, we first used flow cytometry to detect the proportions of M1 and M2 macrophages in peripheral blood of T2DM patients with different vitamin D levels. For phenotyping, macrophages were stained with anti-CD11b-PE, anti-CD206-APC, anti-CD163-FITC, anti-CD209-PerCP-Cy5.5, and anti-CD86-PE-Cy7 in PBS containing 2% FBS. Fluorescence of the cells was assessed by flow cytometry (Canto II, BD Biosciences). Then, we analyzed the number of M1 and M2 macrophages in the peripheral blood of 3 groups (Fig. 1).

Fig. 1
figure 1

Comparison of M1/M2 macrophages in T2DM patients with different VitD levels. A In peripheal blood PBMC, the ratio of CD86-positive M1 macrophage to CD163-positive M2 macrophage decreased with the increase of VitD level. B The ratio of CD86-positive M1 macrophages to CD209-positive M2 macrophages also decreased with the increase of VitD level

Full size image

With the increase of serum 25(OH)D level, the level of IL-6 in plasma gradually decreased

Furthermore, we detected inflammatory factors related to macrophage differentiation in peripheral blood. The inflammatory factor IL-6 secreted by M1 macrophages in the plasma of the enrolled patients was detected by ELISA, and the correlation between serum IL-6 concentration and VitD level was analyzed (Fig. 2).

Fig. 2
figure 2

The level of IL-6 secreted by M1 macrophages decreased with the increase of VitD. A Inflammatory cytokine IL-6 levels were determined by ELISA, p < 0.05 for the indicated comparison after ANOVA. B Linear regression analysis of the correlation between serum IL-6 concentration and 25(OH)D. The result showed that the level of plasma IL-6 was negatively correlated with VitD

Full size image

With the increase of serum 25(OH)D level, the level of TGF-β in plasma gradually increased

The inflammatory factor TGF-β secreted by M2 macrophages in the plasma of the enrolled patients was detected by ELISA, and the correlation between serum TGF-β concentration and VitD level was analyzed (Fig. 3).

Fig. 3
figure 3

With the increase of VitD level, TGF-β secreted by M2 macrophages gradually increased. A Inflammatory cytokine TGF-β levels were determined by ELISA, p < 0.05 for the indicated comparison after ANOVA. B Linear regression analysis of the correlation between serum TGF-β concentration and 25(OH)D. The result predicted a positive correlation between the level of plasma TGF-β and VitD in peripheral blood of patients with T2DM

Full size image

Separation of CD14 + monocytes in peripheral blood

To further confirm our above experimental results, we sorted CD14 + monocytes for in vitro generation of macrophages. CD14 MicroBeads (human, 130–050-201, MiltenyiBiotec) were used for the positive selection of human monocytes and macrophages from PBMCs (Fig. 4). We achieved successful isolation of monocytes from human PBMCs with > 96% purity of CD14 + monocytes.

Fig. 4
figure 4

Detection of CD14-positive monocytes in peripheral blood by flow cytometry In the figure, CD14-APC was stained on PBMC before sorting and cells after sorting, respectively. The percentage of CD14-positive cells was detected by flow cytometry. A Before sorting, only 0.4% of CD14-positive cells were detected. B After sorting, more than 96% of CD14-positive cells were screened

Full size image

25(OH)D can promote the differentiation of CD14 positive monocytes cultured in vitro into M2 macrophages

Human macrophage cultures: isolation of CD14 + monocytes for in vitro generation of macrophages. Peripheral blood mononuclear cells were prepared by density centrifugation from the blood of healthy adult volunteers. Monocytes were isolated to high purity by magnetic cell sorting using anti-CD14-coated beads (per manufacturer recommendations, MiltenyiBiotec, Auburn, CA). Monocytes were plated in 24-well cell culture plate and subsequently cultured for 7 days in medium (IMDM medium, GIBCO, Invitrogen) with 10% FBS (HyClone) and recombinant human M-CSF. In addition, different concentrations of calcitrol (1,25-dihydroxyvitamin D3, it is the most active vitamin D metabolite and a vitamin D receptor agonist, HY-10002, MCE) were added (0, 0.1 µM, 1 µM) and four replicates per concentration. After 7 days, M1 and M2 macrophages were detected by flow cytometry (Fig. 5).

Fig. 5
figure 5

CD14-positive monocytes were treated with different concentrations of calcitrol, and the proportions of M1 and M2 macrophages were detected by flow cytometry. A Proportion of M1 macrophages (CD86 +) differentiated from monocytes treated with different concentrations of calcitrol. The higher the concentration of calcitrol, the less M1 macrophages differentiated. B Proportion of M2 macrophages (CD206 +) differentiated from monocytes treated with different concentrations of calcitrol. After calcitrol treatment, the differentiation of monocytes into M2 macrophages increased. C After calcitrol treatment, the ratio of M2/M1 macrophages increased. The X-axis of the three panels are all calcitrol concentrations

Full size image

Discussion

Vitamin D and its analogues have been widely used in health promotion, disease prevention, and treatment. Adequate sunlight is the safest, inexpensive, and effective way to prevent VitD deficiency. Vitamin D can be supplemented to those who cannot get enough sunshine or VitD deficiency. Vitamin D is a basic health supplement for the prevention and treatment of osteoporosis[22]. Active VitD and its analogues are also commonly used in the clinic for rickets/osteomalacia, osteoporosis, hypoparathyroidism, and skin diseases. The use of VitD and its analogues should pay attention to its safety, monitor blood and urine calcium levels, and prevent VitD poisoning[23, 24]. Although the effects of VitD on calcium and phosphorus metabolism and extraskeletal functions have been continuously discovered, the dosage and efficacy of VitD in the prevention and treatment of diabetes, cancer, immune diseases, and infectious diseases are still uncertain[25, 26]. With the deepening of future research, more new VitD preparations and new drug indications are expected to be continuously developed and applied.

In this study, we first used flow cytometry to detect the proportion of M1 and M2 macrophages in the peripheral blood of T2DM patients with different VitD levels. The results showed that with the increase of serum 25(OH)D, the polarization phenotype of macrophages in peripheral blood tended to be M2 type (M1/M2 decreased). Furthermore, we used ELISA experiments to detect the inflammatory factors that affect the differentiation of macrophages, including IL-6 secreted by M1 macrophages, The results showed that as the level of 25(OH)D increased, the concentration of IL-6 in plasma gradually decreased, and the TGF-β secreted by M2 macrophages gradually increased.

The polarization of macrophages is a complex process of multi-factor interaction, which is regulated by a variety of signal molecules and their pathways. Currently, there are five signal pathways that are more mature in research: JAK/STAT[27], PI3K/Akt[28], C-Jun N-terminal kinase (JNK)[29], Notch[30], and B7-H3/STAT3[31]. Macrophages are important regulators in the process of systemic metabolism, hematopoietic function, angiogenesis, apoptosis, and tumor formation. The study of macrophage polarization regulation can provide new ideas for the treatment of related diseases involving macrophages.

T2DM is mainly characterized by hyperglycemia, insulin resistance, and impaired insulin secretion. Studies in the past few years have emphasized the adverse effects of T2DM on bone quality and strength, and have made people accept that diabetic bone disease is a long-term serious complication of T2DM. The ratio of M1/M2 macrophages was also significantly higher in peripheral blood of T2DM patients, signifying an imbalance between inflammatory and T2DM[20, 32].

Macrophages comprise a heterogeneous population of cells that belong to the mononuclear phagocyte system. On the other hand, various studies indicate that VitD inhibits M1 macrophage infiltration, and it can promote M2 macrophage activation[33]. Calcitriol treatment notably reduced macrophage infiltration and meanwhile inhibited the pro-inflammatory cytokine production[34]. However, the effect of VitD on the differentiation of monocytes from T2DM patients into M1 or M2 macrophages is not clear. In this study, we confirmed that the ratio of M1/M2 decreased with the increase of VitD level in T2DM patients, and VitD could promote the differentiation of monocytes in PBMC into M2-type macrophages. The effect of VitD on M1/M2 macrophages may be an influencing factor of M1/M2 ratio imbalance in T2DM patients. Our study confirmed that VitD can promote the anti-inflammatory function of macrophages, which is a new complementary evidence that supplementation of VitD can reduce or prevent the occurrence of T2DM.

However, our research also has certain limitations. Due to the limitation of the number of sample cases, it only plays a role of “adding bricks and tiles” to the development of science. Next, we want to identify some biomarkers to explore the potential mechanism of T2DM patients with low serum 25(OH)D levels, which is VitD deficiency, and to study the specific mechanism of VitD regulating the polarization of macrophages. Our current hypothesis is that it modulates the signal pathway that affects polarization by binding to the VitD receptor in monocytes/macrophages. Our future research directions will continue to develop around the above assumptions.

Conclusion

When VitD is deficient in T2DM patients, the proportion of M1 and M2 macrophages in peripheral blood was higher, and the M1/M2 ratio decreased with the increase of serum 25(OH)D level. That is to say, monocytes-macrophages differentiate into M2 macrophages more when vitamin D is sufficient, and differentiate into M1 macrophages when VitD is deficient. At the same time, with the increase of VitD level, the IL-6 secreted by M1-type macrophages in the patient’s plasma gradually decreased, and the TGF-β secreted by M2-type macrophages gradually increased. In vitro culture experiments also confirmed that VitD can promote the differentiation of CD14-positive monocytes into M2-type macrophages. The overall study shows that VitD levels in T2DM patients are related to the type of macrophages. Vitamin D deficiency is more conducive to the differentiation of macrophages into the M1 pro-inflammatory type, thereby promoting the occurrence of inflammation and the progression of T2DM, which may also be the reason for the high risk of T2DM in people with VitD deficiency. We are conducting more and more in-depth studies on the specific mechanism by which VitD may regulate macrophage differentiation.