The pattern of inflammatory/anti-inflammatory cytokines and chemokines in type 1 diabetic patients over time

Figures & data

Table I. Characteristics of the healthy subjects and type 1 diabetic patients.

	Healthy subjects (n = 22)	T1DM patients (n = 44)	P
Gender (M/F)	9/13	17/27
Age (years)	10.7 ± 0.78	11.47 ± 0.66	NS
Body mass index (kg/m^2)	21.15 ± 1.17	20.48 ± 1.03	NS
Diabetes duration (months)	—	50.76 ± 11.61 (0–184)
Total cholesterol (mg/dL)	153.25 ± 9.79	165.22 ± 7.14	NS
HDL (mg/dL)	64.57 ± 2.80	58.14 ± 1.81	NS
LDL (mg/dL)	80.12 ± 4.02	91.14 ± 4.28	NS
Total triglycerides (mg/dL)	50.86 ± 4.14	82.08 ± 6.58	<0.001
Lipoprotein (a) (mg/dL)	12.42 ± 0.98	19.58 ± 1.09	NS
Apolipoprotein A (mg/dL)	146.23 ± 6.04	131.19 ± 5.47	NS
Apolipoprotein B (mg/dL)	77.52 ± 4.49	84.04 ± 3.71	NS
Fibrinogen g/L	3.18 ± 0.23	3.35 ± 0.68	NS
Urine microalbumin (mg/L)	4.18 ± 0.97	6.10 ± 0.99	NS
Serum creatinine (mg/dL)	0.72 ± 0.06	0.78 ± 0.05	NS
Serum urea (mg/dL)	28.52 ± 2.69	32.86 ± 2.88	NS
CRP (mg/dL)	0.88 ± 0.06	6.22 ± 0.25	<0.001
HbA1c (%)	4.20 ± 0.42	8.45 ± 0.43	<0.001

The results are expressed as mean ± standard error of mean, while disease duration is expressed as mean ± standard error of mean (min–max).

T1DM = type 1 diabetes mellitus; HDL = high-density lipoprotein; LDL = low density lipoprotein; CRP = C-reactive protein; IL-6 = interleukin-6; MMP-9 = matrix metalloproteinase-9; n = number of subjects; NS = non-significant; P ≤ 0.05 was considered significant (Student's t test.).

Table II. Plasma levels of inflammatory and anti-inflammatory cytokines and chemokines.

	Healthy subjects (n = 22)	T1DM patients (n = 44)	P
CD40 (pg/mL)	54.92 ± 3.09	109.98 ± 17.39	0.003
IL-1a (pg/mL)	384.98 ± 200.52	1370.41 ± 422.64	0.041
IL-1b (pg/mL)	523.81 ± 293.68	1207.59 ± 525.13	NS
IL-2 (pg/mL)	91.23 ± 20.06	283.23 ± 66.12	0.009
IL-4 (pg/mL)	465.11 ± 137.90	1368.29 ± 386.44	0.034
IL-5 (pg/mL)	24.24 ± 4.72	133.46 ± 39.29	0.01
IL-6 (pg/mL)	77.20 ± 21.30	262.67 ± 91.81	0.057^a
IL-8 (pg/mL)	158.67 ± 62.97	229.11 ± 48.77	NS
IL-10 (pg/mL)	118.59 ± 39.00	344.11 ±101.62	0.045
IL-12 (pg/mL)	222.48 ± 95.98	286.49 ± 66.14	NS
IL-13 (pg/mL)	29.00 ± 11.65	81.34 ± 29.86	NS
GM-CSF (pg/mL)	42.54 ± 10.90	108.55 ± 24.26	0.017
GRO (pg/mL)	5412.61 ± 1920.58	9630.00 ± 1799.01	NS
IFN-γ (pg/mL)	205.85 ± 58.72	767.38 ± 542.81	NS
MCP-1 (pg/mL)	73.39 ± 8.51	69.67 ± 5.95	NS
MIP-1a (pg/mL)	405.19 ± 149.35	988.51 ± 236.80	0.042
MIP-1b (pg/mL)	126.29 ± 21.41	432.37 ± 80.86	0.001
RANTES (pg/mL)	11293.77 ± 2476.74	23514.13 ± 2174.60	0.001
TNF-α (pg/mL)	476.79 ± 171.30	1022.79 ± 348.20	NS
MMP-9 (pg/mL)	6121.63 ± 1415.73	12556.80 ± 1653.92	0.005
VEGF (pg/mL)	1076.13 ± 710.04	546.55 ± 117.25	NS

The values are in pg/mL, and data are expressed as mean ± standard error of mean.

^aT1DM patients showed a trend to higher IL-6 (P = 0.057).

NS = non significant, P ≤ 0.05 was considered significant (Student's t test.)

Figure 1. Percentage alteration of cytokine levels in type 1 diabetic patients compared to healthy controls. The figure includes only cytokines that significantly differ between diabetic patients and healthy controls, where P ≤ 0.05 was considered significant.

Figure 2. Plasma soluble CD40 (sCD40) and RANTES in healthy controls (HCs) and type 1 diabetes mellitus (T1DM) patients. Soluble CD40 and RANTES were identified as the significant contributors to the discrimination between diabetic and healthy subjects. Data are presented as bars, where bar length represents mean value for each group, with error bars depicting standard error of mean. P ≤ 0.05 was considered significant. *P = 0.003 versus HCs. **P = 0.001 versus HCs.

Figure 3. Plasma MCP-1 levels between diabetic patients with disease duration of more than and less than 6 months. MCP-1 was found as the most discriminant factor between patients with disease duration of more than and less than 6 months. Data are presented as bars, where bar length represents mean value for each group, with error bars depicting standard error of mean. P ≤ 0.05 was considered significant. *P < 0.001 versus patients with disease duration of more than 6 months.

Table III. Effects of time passed from the diagnosis of type 1 diabetes on the levels of soluble CD40, MCP-1, TNF-α, and RANTES.

			Group 1 (Diabetes duration ≤6 months)
Parameter tested	F (2,44)	P	Subgroup 1a (Diabetes duration <1 month)	Subgroup 1b (Diabetes duration 1–6 months)	Group 2 (Diabetes duration >6 months)	P (SbG-1a versus SbG-1b)	P (SbG-1a versus Group 2)	P (SbG-1b versus Group 2)
CD40 (pg/mL)	4.199	0.024	75.10 ± 17.30	189.96 ± 51.35	87.62 ± 14.45	0.050	NS	0.036
MCP-1 (pg/mL)	11.197	<0.001	72.27 ± 10.88	103.43 ± 11.74	49.16 ± 5.148	NS (0.076)	NS	<0.001
RANTES (pg/mL)	8.497	0.001	34014.92 ± 1965.48	12451.01 ± 4201.30	25417.60 ± 2891.94	0.001	NS	0.019
TNF-α (pg/mL)	5.152	0.012	366.00 ± 125.31	2604.13 ± 1028.97	385.89 ± 143.33	0.039	NS	0.016

The values are in pg/mL, and data are expressed as mean ± standard error of mean (one-way ANOVA, Tukey's s post hoc test).

SbG = subgroup.

Figure 4. Hierarchical clustering comparing T1DM patients and healthy controls. A heat map representation of the correlational coefficients is graphed. Cytokines with positive correlations are represented in graded shades of red and negative correlations in graded shades of blue. Representation of the level of correlations is listed on the side of the graph. The dendrograms on the left are presented as the result of hierarchical cluster analysis with the use of Pearson's correlation coefficients, using the UPGMA method (unweighted pair-group method arithmetic average). The actual distances between clusters are rescaled to numbers between 0 and 25.

Figure 5. Hierarchical clustering comparing T1DM patients with disease duration of more than and less than 6 months. A heat map representation of the correlational coefficients is graphed. Cytokines with positive correlations are represented in graded shades of red and negative correlations in graded shades of blue. Representation of the level of correlations is listed on the side of the graph. The dendrograms on the left are presented as the result of hierarchical cluster analysis with the use of Pearson's correlation coefficients, using the UPGMA method (unweighted pair-group method arithmetic average). The actual distances between clusters are rescaled to numbers between 0 and 25.