Change of monocytes/macrophages in ulcerative colitis patients with symptoms of anxiety and depression

Ulcerative colitis patients admitted to West China Hospital, Sichuan University from May 2019 to January 2021 were included in our study. Exclusion criteria were as follows: (1) comorbidities highly associated with anxiety or depression such as carcinoma and cardiovascular disease; (2) concomitant with other chronic or severe psychiatric diseases, like psychosis, psychoactive substance abuse, and dementia; (3) pregnancy; (4) inability to cooperate. In addition, none of our included participants was treated for anxiety or depression. They were divided into two groups, patients with symptoms of anxiety/depression and patients without based on the Hospital Anxiety and Depression Scale (HADS). A HADS score ≥ 8 is indicative of a patient with symptoms of anxiety/depression. Other questionnaires, including the Inflammatory Bowel Disease Questionnaire (IBDQ), the Composite Autonomic Symptom Score (COMPASS)-31, the Fatigue Severity Scale (FSS) and the Pittsburgh Sleep Quality Index (PSQI) were completed at the same time. IBDQ is a disease-specific tool to assess the disease consequences on a patient’s quality of life. The COMPASS-31 is a concise and statistically robust instrument to assess autonomic symptoms that provides clinically relevant scores of autonomic symptom severity. FSS and PSQI were used to evaluate the participants’ fatigue and quality of sleep, respectively. Healthy controls were also involved. Patients’ demographics, clinical disease activity scores (using Mayo score), endoscopic evaluation (using ulcerative colitis endoscopic index of severity, UCEIS) and histological score (Geboes score) were all recorded and compared between the two groups.

Peripheral blood samples were collected on the day of admission. The acquired plasma was stored in aliquots at -80℃ for batched cytokine measurements. Blood samples were processed with modification of a previously described whole-blood technique protocol. After removing erythrocytes, fluorescein isothiocyanate-conjugated anti-CD45 (No. 555482; BD Biosciences), phycoerythrin-conjugated anti-CD16(No. 555407; BD Biosciences), and brilliant violet 510-conjugated anti-CD14(No. 563079; BD Biosciences) were used. The final cell resuspension went through a FACSAria (BD Biosciences) flow cytometer. Monocytes were divided into three subtypes, classical subtype (CD14 + + CD16- cells), intermediate subtype (CD14 + + CD16 + cells) and nonclassical subtype (CD14 + CD16 + + cells). The percentages of monocytes in each subtype were compared. Similar procedures were applied to the measurement of monocyte phagocytosis. Latex beads-rabbit IgG-PE complex was used.

CD4 + T cells isolated from peripheral blood mononuclear cells (PBMCs) of healthy volunteers by fluorescence-activated cell sorting (using PE-conjugated anti-CD4, No.555347, on FACSAria, BD Biosciences) were cocultured with monocytes from healthy volunteers, UC patients with symptoms of anxiety/depression and without, respectively. The concentration ratio of CD4 + T cells to monocytes was 5:1. LPS in a final concentration of 100 ng/mL was used in the coculture system. Cells were harvested after 72 h for flowcytometric analysis of Th1/Th2/Th17/Treg proportions.

PE-conjugated anti-CD4, BV711-conjugated anti-CD25 (No.563159, BD Biosciences) and brilliant blue (BB)515-conjugated anti-CD127 (No.564423, BD Biosciences) were used to label Treg cells. PE-cyanine (Cy)7-conjugated anti-IFN-γ, allophycocyanin (APC)-conjugated anti-IL-4 and BV421-conjugated IL-17A (No.557844, No.562438, NO.562933, all from BD Biosciences) with PE-conjugated anti-CD4 were used to detect Th1/Th2/Th17 cells, respectively.

Plasma concentrations of tumor necrosis factor (TNF)-α, IL-6, IL-10, macrophage-colony stimulating factor (M-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and monocyte chemotactic protein-1(MCP-1) were measured by Luminex kit (LXSAHM-10, R&D Systems) according to the manufacturer’s instructions. The level of transforming growth factor (TGF)-β1 was assayed based on the manufacturer’s instruction from Human TGF-β1 ELISA kit (SEKH-0316, G-CLONE).

Colon biopsy and surgical specimens were achieved and sliced into 4 μm as we described in our previous study. Primary antibodies including anti-CD68 (as a pan-macrophage marker, 1: 100, No. MA5-13324, Invitrogen), anti-CD86 (as a marker for M1 macro-phage, 1: 100, No. MA5-30196, Invitrogen), and anti-CD163 (as a marker for M2 macrophage, 1: 500, No. PA5-78961, Invitrogen) were used.

Proteins were extracted from the colon samples via homogenization in ice-cold lysis buffer. The bicinchoninic acid (BCA) protein assay kit (Thermo) was used to measure the concentrations. Prepared protein samples were separated on 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride membranes. Then the membranes were incubated with the anti-M-CSF antibody (1:1000, No. GB11685, Servicebio) overnight at 4 °C. On the following day, membranes were incubated with the secondary antibody (HRP-conjugated Goat anti-Rabbit IgG, No. abs20040, Absin) at room temperature for 2 h. Antibody binding was detected by chemiluminescence using the ECL system (ChemiDoc MP, Bio-Rad). Densitometry of the blots was analyzed with Image Lab 5.2 software.

For qRT analyses, colon samples were lysed immediately after using the TRIzol Reagent (Life Technologies), and total RNA isolation was performed according to the manufacturer’s protocol. cDNA was generated using EuroScript Reverse Transcriptase (Euroclone Cytogenetics), with random examers and 2.5 µg RNA per reaction. qRT-PCR reactions were then prepared with the PowerUp SYBR Green Mix (Applied Biosystems) and run using a Bio-Rad CFX Maestro system (Applied Biosystems). Primer pairs were designed ex novo with NCBI Primer-BLAST. Verification and location of target gene sequences were performed on Ensembl Genome Browser. Primer sequences of M-CSF are listed as follows: (5'-3') F: CCTGCTGTTGTTGGTCTGTCTC, R: GGTACGAGGTCTCCATCTGA. Results were normalized using GAPDH as housekeeping gene as a reference and evaluated using the 2^−ΔΔCt method.

All data were analyzed with SPSS 22.0 and GraphPad Prism 6.0 software. The Kolmogorov–Smirnov test was performed to demonstrate if the data were in normal distribution. Continuous variables were presented as median values (interquartile range [IQR]), while categorical variables were presented as percentages. Comparisons between the 2 groups were made using the Mann–Whitney U test for continuous variables or with the Pearson chi-square test for categorical variables. Comparisons among multiple groups were done with the Kruskal–Wallis test as well as Dunn’s multiple comparisons test. Correlations between 2 variables were assessed with Spearman’s rank correlation coefficient. Statistical significance was considered achieved for P < 0.05.

139 UC patients were included and were evaluated based on HADS. 37.41% (51/139) of UC patients had the symptoms of anxiety, and 32.37% (45/139) had the symptoms of depression. UC patients with either one of two symptoms, anxiety or depression, or with both were defined as UC patients with symptoms of anxiety/depression, including 15 cases only with symptoms of anxiety, 8 cases only with symptoms of depression and 37 cases with both. In total, 43.17% (60/139) of UC patients had symptoms of anxiety/depression.

Patients were divided into two groups, UC patients with symptoms of anxiety/depression and patients without. UC patients with symptoms of anxiety/depression were older than UC patients without [48(33,55) vs. 32(24,51), P = 0.0012], and had lower education level (primary school and junior school graduation or even below,43.33% vs. 26.58%, P = 0.0164). There were no significant differences in the gender ratio, BMI or disease duration between the two groups (P = 0.5114, 0.7696 and 0.0969, respectively) (Table 1).

IBDQ score was lower, and scores of FSS, PSQI and COMPASS-31 were higher in UC patients with symptoms of anxiety/depression compared to UC patients without symptoms of anxiety/depression, indicating the worse quality of life (P < 0.05, Fig. 1A).

In patients with symptoms of anxiety/depression, mayo score, platelet count, erythrocyte sedimentation rate, and endoscopic score were significantly higher than those in UC patients without symptoms of anxiety/depression (P < 0.05), and UC patients with symptoms of anxiety/depression were prescribed steroids more frequently (58.33% vs. 29.11%, P = 0.0005). Histological score (Geboes score) was also higher in UC patients with symptoms of anxiety/depression [5.1(4.2,5.2) vs. 4.1(3.1,4.2), P < 0.0001] (Table 1).

Moreover, we further analyzed the correlation between the disease activity of the two groups and the scores of hospital anxiety and depression scale. Results showed that mayo score, UCEIS score and Geboes score were positively correlated with anxiety symptom scores in HADS (P < 0.0001, 0.0012, < 0.0001, respectively). Similarly, mayo score, UCEIS score and Geboes score were positively correlated with depressive symptom scores in HADS (P = 0.0003, 0.0001, < 0.0001, respectively) (Fig. 1B).

The plasma levels of proinflammatory cytokines, including IL-1β、IL-6 and TNF-α were significantly higher in UC patients with symptoms of anxiety/depression, while the level of anti-inflammatory cytokine TGF-β1 was evidently lower in UC patients with symptoms of anxiety/depression compared to UC patients without symptoms of anxiety/depression (P < 0.05, Fig. 1C).

The gating strategies for monocytes in peripheral blood was shown in Fig. 2A. The percentages of intermediate monocytes (CD14 +  + CD16 + monocytes) and nonclassical monocytes (CD14 + CD16 +  + monocytes) were higher in UC patients with symptoms of anxiety/depression [17.29(10.00,21.40) vs. 8.73(6.26,11.50), P = 0.0004; 7.46 ± 0.76 vs. 4.18 ± 0.45, P = 0.0010] (Fig. 2B). Phagocytosis was significantly impaired in UC patients with symptoms of anxiety/depression compared to UC patients without (68.88 ± 2.39 vs. 78.79 ± 1.78, P = 0.0027) (Fig. 2C).

Monocytes from UC patients with symptoms of anxiety/depression significantly inhibited CD4 + T cells from healthy volunteers polarized to Treg cells [1.37(0.68,2.34) vs. 2.40(2.15,4.56), P = 0.0124] (Fig. 3A), and induced CD4 + T cells to differentiate into Th1 cells [4.52(3.05,6.37) vs. 3.25(2.07,3.77), P = 0.0409] (Fig. 3B) contrasted to monocytes from UC patients without symptoms of anxiety/depression. However, the effects of monocytes on the polarization of CD4 + T cells to TH2 cells (Fig. 3C) and TH17 cells (Fig. 3D) were comparable between UC patients with and without symptoms of anxiety and depression.

The total number of intestinal mucosal macrophages was increased in UC patients compared to that in healthy controls (P < 0.05). UC patients with symptoms of anxiety/depression had a larger number of CD68 + cells in the intestine mucosal layer compared to UC patients without symptoms of anxiety/depression [43(38,49) vs. 29(26,31), P < 0.0001]. And the ratio of M1/M2 was higher in UC patients with symptoms of anxiety/depression (2.15 ± 0.18 vs. 1.31 ± 0.11, P = 0.0004) (Fig. 4A). Phagocytosis of intestinal lamina propria macrophages in UC patients with symptoms of anxiety/depression was impaired, but the difference showed no statistical significance (29.10 ± 2.70 vs. 36.71 ± 2.63, P = 0.1489) (Fig. 4B).

The renewal of intestinal macrophages depends on the continuously supplemented of peripheral blood monocytes. We compared the cytokines involved in monocyte/macrophage differentiation among groups, including M-CSF, GM-CSF and MCP-1. The levels of M-CSF and GM-CSF in plasma of patients with ulcerative colitis with symptoms of anxiety/depression were significantly higher than those in patients with ulcerative colitis without symptoms of anxiety/depression [M-CSF:118.20(64.85, 154.10) vs. 65.16(47.28, 83.81), P = 0.0057; GM-CSF: 2.49 ± 0.16 vs. 1.69 ± 0.05, P = 0.0001], while there was no significantly difference in MCP-1 level (pg/mL) between the two groups [139.3(98.4, 176.9) vs. 144.5(107.9, 192.8), P = 0.7983] (Fig. 5A). Further, the protein level of intestinal M-CSF was analyzed and showed that the level in patients with ulcerative colitis accompanied by anxiety/depression was significantly higher than that in patients without anxiety/depression (1.72 ± 0.16 vs. 1.22 ± 0.07, P = 0.0295) (Fig. 5B). The expression of M-CSF at the RNA level was also significantly increased (33.88 ± 28.11 vs. 11.55 ± 7.29, P = 0.0303) (Fig. 5C).