Long-term exposure to diesel engine exhaust affects cytokine expression among occupational population

Long-term exposure to diesel exhaust is associated with reduced IL-6 and increased CRP levels.

Abstract

Diesel engine exhaust (DEE) is a predominant contributor to urban air pollution. The International Agency for Research on Cancer classified DEE as a group I carcinogen. Inflammatory response is considered to be associated with various health outcomes including carcinogenesis. However, human data linking inflammation with long-term DEE exposure are still lacking. In this study, a total of 137 diesel engine testing workers with an average exposure of 8.2 years and 108 unexposed controls were enrolled. Peripheral blood samples were collected from all subjects, and the association of DEE exposure with inflammatory biomarkers was analyzed. Overall, DEE exposed workers had a significant increase in the C-reactive protein (CRP) and a significant decrease in cytokines including interleukin (IL)-1β, IL-6, IL-8, and macrophage inflammatory protein (MIP)-1β compared to controls after adjusting for age, BMI, smoking status, and alcohol use, and findings were highly consistent when stratified by smoking status. In addition, exposure time dependent patterns for IL-6 and CRP were also found (P_trend = 0.006 and 0.026, respectively); however, the levels of IL-1β and MIP-1β were significantly lower in subjects with a DEE working time of less than 10 years compared with the controls and then recovered to control levels in workers exposed for >10 years. There were no significant differences in blood cell counts and major lymphocyte subsets between exposed workers and the controls. Our results provide epidemiological evidence for the relationship between DEE exposure and immunotoxicity considering the important roles of cytokines in immunological processes.

Introduction

Diesel engine exhaust (DEE) is produced by diesel-powered machines and vehicles used in the construction, transportation, marine shipping, mining, etc. industries. DEE is a complex mixture consisting of gaseous and organic components such as carbon monoxide (CO), nitrogen dioxide (NO₂), sulphur dioxide (SO₂), polycyclic aromatic hydrocarbons (PAHs) etc., which are absorbed onto carbon core particles of different sizes ranging from large particles to ultrafine nanoparticles. As a predominant contributor to urban fine particulate matter,1 the adverse effects of DEE on human health are currently a serious concern. Abundant evidence from epidemiological and experimental studies showed that DEE exposure was associated with cardiovascular health outcomes2–4 and respiratory disease including allergy,5 asthma,6,7 chronic obstructive pulmonary disease,8,9 as well as lung cancer.10,11 In 2012, the International Agency for Research on Cancer (IARC) classified DEE as a group I carcinogen.12

Some human diseases of affluence and extended lifespan such as obesity, cardiovascular and neurodegenerative diseases, and cancer are associated with chronic inflammation,13 which is accompanied by the production and release of a series of signaling molecules including cytokines, chemokines, and adhesion molecules and which operates in a complex network between epithelial cells and immune cells.14,15 There was evidence of the association between particulate matter (PM) exposure and inflammation response.16–18 Some short-term human controlled DEE exposure studies have also provided evidence that DEE can affect the immune system. For example, healthy volunteers who were exposed to DEE under controlled conditions for 1 h 15,19,20 or 2 h 21,22 had a significant increase in the expressions of interleukin (IL)-13 and IL-8 in the bronchial epithelium cells and lavage fluid. The number of neutrophils and B lymphocytes in airway lavages, the neutrophils and platelets in peripheral blood, and neutrophils, mast cells, and CD4⁺ and CD8⁺ T lymphocytes in bronchial tissues significantly increased following DEE exposure.15,21 Y. Xu et al.23 observed that monocyte and total leukocyte counts in peripheral blood were higher in volunteers exposed to DEE for 3 h, and a trend towards increased serum IL-6 concentrations was also found. However, these findings represent the acute inflammatory response after short-term DEE exposure, which is different from the actual situation of long-term DEE exposure in the general population.

To date, the associations between long-term DEE exposure and inflammatory responses have not been adequately defined and have not been well characterized among population in a non-experimental setting. In the present study, we aimed to evaluate whether the distributions of major lymphocyte subsets (T cell, B cells, and NK cells) in the peripheral blood and serum levels of inflammatory markers including IL-1, IL-6, IL-8, tumor necrosis factor (TNF)-α, macrophage inflammatory protein (MIP)-1β, monocyte chemotactic protein (MCP)-1 and C-reactive protein (CRP) were altered in workers with long-term DEE exposure compared to unexposed controls, and whether these associations differed by smoking status and length of DEE exposure.

Methods

Subjects and sample collection

We conducted a cross-sectional molecular epidemiology study in China. Exposed workers consisted of 137 males who were responsible for testing the diesel engines in a diesel engine manufacturing factory with an average production of about 100 000 machines per year in 2011–2013. In the engine testing workshop, the engines are kept running when tested by workers, who stand beside the engine and are exposed to DEE. The control group consists of 108 unexposed male workers, who were selected from a water plant located in the same geographical region as the diesel engine factory. Controls were frequency-matched to the exposed workers by age and smoking status. Exclusion criteria for both DEE-exposed and control subjects included having a history of cancer and autoimmune diseases, acute infection, and previous exposure to other occupational chemicals that are known hematotoxins or carcinogens.

The study was integrated into a regular health examination administered by the local Center for Disease Control. The information on workers regarding age, smoking and drinking habits, occupational history of exposure, and personal medical history was collected by trained interviewers using a detailed questionnaire. On the day of health examination, in the morning, peripheral blood samples were collected from each subject for the complete blood cell count (CBC), lymphocyte subset analysis, and serum separation used for cytokines and CRP detection. A spot urine sample was obtained from each subject at the end of a work shift, maintained at –70 °C and used for the detection of 1-hydroxypyrene (1-OHP).

This study was approved by the Research Ethics Committee of the National Institute for Occupational Health and Poison Control at the Chinese Center for Disease Control and Prevention. Informed consent was obtained from each participant.

Exposure assessment

Airborne samples were collected on 2 consecutive work days (for 8 h on each work day) within one week before blood sample collection. In total, there were 16 and 6 airborne samples for the engine testing workshop and the water plant, respectively. The DEE exposure was assessed by particulate matter (PM)_2.5, PM_2.5 related elemental carbon (EC), NO₂, SO₂, and airborne PAHs. An air particle sampler (TH150-DII, Wuhan, China) was used for PM_2.5 collection on pre-weighed PTFE filters (90 mm, Millipore, USA), and carbon was collected onto 37 mm quartz fiber filters using cyclones with a PM_2.5 cut point. A UMEX-200 passive sampler (SKC Inc., PA, USA) was used for NO₂ and SO₂ collection. The PM_2.5 filters were analyzed for the PAH constituents based on the guidelines of NIOSH No. 58. EC was determined by using a Carbon Analyzer (DRI2001A, Atmoslytic Inc., USA) according to the method of NIOSH5040. The distribution of particles in the atmosphere of the workplace was monitored using an Aerodynamic Particle Sizer Spectrometer 3321 (TSI, USA) and the data were analyzed using TSI software (Aerosol Instrument Manager).

Measurement of urinary 1-OHP

Urinary 1-OHP was measured using the HPLC-MS/MS method according to C. Huang et al.24 Briefly, 2 mL urine from each sample was mixed with 1 mL of sodium acetate buffer (0.5 mol L^–1, pH = 5.0), and then the samples were incubated with 20 μL β-glucuronidases (Sigma-Aldrich, USA) at 37 °C for 8 h in the dark followed by addition of 80 μL naphthol-d8 (100 μg mL^–1, Sigma-Aldrich, USA) as the internal standard. Subsequently, 1-OHP was extracted from urine with 4 mL dichloromethane and was evaporated to dryness by nitrogen before being dissolved in 1.5 mL HPLC solvent. 1-OHP was analyzed using an ultrafast liquid chromatography system (LC-20AD, Shimadzu, Kyoto, Japan) coupled to an integrated triple quadrupole mass spectrometer (ABI3200, Applied Biosystems, USA). 1-OHP was separated using a reverse-phase C18 column (2.1 mm × 150 mm, 5 μm, Waters Xterra MS C18) and was eluted using a mobile phase composed of methanol and water. The identification and quantification of 1-OHP were based on the retention time and peak area measured using a linear regression curve obtained from internal standard solutions. The detection limit of urinary 1-OHP was 0.2 μg L^–1. Creatinine, a clearance protein that adjusts for differences in urinary concentration, was also measured at the same laboratory. The concentration of creatinine-adjusted urinary 1-OHP was expressed as μg g^–1 Creat.

Lymphocyte subset analysis

Lymphocytes in peripheral blood were stained with monoclonal antibodies conjugated to fluorescein isothiocyanate (FITC), allophycocyanin (APC), phycoerythrin (PE), or peridinin-chlorophyll protein (PerCP) (BD Biosciences, San Jose, CA, USA). Briefly, 50 μL of EDTA-preserved whole blood from each sample was added to 2 individual test tubes containing pre-added monoclonal antibodies. The first tube contains an antibody combination of CD45–PerCP, CD3–FITC, CD4–APC, and CD8–PE; the second one contains an antibody combination of CD45–PerCP, CD3–FITC, CD56 + 16–PE, and CD19–APC. Tubes were incubated for 20 min at room temperature in the dark. After adding 450 μL lysing solution, the tubes were vortexed for 30 s and incubated for 15 min. A lymphocyte subset analysis was performed using BD FACS Aria II and FACS DIVA software (BD Biosciences).

Cytokines and CRP measurements

Serum inflammatory markers, including IL-1β, IL-6, IL-8, TNF-α, MIP-1β, and MCP-1, were measured using a BD Cytometric Bead Array Human Soluble Protein Flex Set system (BD Biosciences). A Human Soluble Protein Master Buffer Kit, a Human IL-1β Flex Set (Bead B4), a Human IL-6 Flex Set (Bead A7), a Human IL-8 Flex Set (Bead A9), a Human MCP-1 Flex Set (Bead D8), a Human MIP-1β Flex Set (Bead E4), and a Human TNF Flex Set (Bead C4) were used for cytokine detection according to the manufacturer's protocol (BD Biosciences). Sample measurements were performed on BD Canto and BD LSRFortessa flow cytometers, and data were analyzed using FCAP Array Software (Soft Flow Inc., Pecs, Hungary). If the concentration of cytokine was lower than the detection limit of the analysis, one half of the detection limit value was substituted for the level of cytokine for that sample. CRP in serum was measured by using an immunoturbidimetric assay of latex particles coated with CRP monoclonal antibodies (DiaSys, Germany).

Statistical analysis

Linear regression was used to test for differences in the level of each endpoint between exposed and control workers, as well as the exposure time–response trend. Analyses were also conducted for each endpoint in smokers and non-smokers separately. All statistical models were adjusted for age (as a continuous variable), smoking status (yes or no), current alcohol consumption (yes or no), and body mass index (BMI). Comparisons of baseline demographic characteristics between exposed and unexposed workers were conducted using a Student's t test for continuous variables or Pearson's χ² for categorical variables. Values of P < 0.05 were considered significantly different. All the statistical analyses were performed using SPSS 15.0.

Results

DEE exposure levels and characteristics

A significantly higher average concentration of PM_2.5, EC, NO₂, SO₂, and total PAHs was found in the diesel engine workshop (282.32 μg m^–3, 135.17 μg m^–3, 0.258 ppm, 0.146 ppm, and 4.53 μg m^–3, respectively) compared to the water plant (91.93 μg m^–3, 11.81 μg m^–3, 0.038 ppm, 0.004 ppm, and 0.035 μg m^–3, respectively) (P < 0.001, Table 1). Distributions of particle sizes that were measured in the diesel engine testing workshop and water plant are also shown in Table 1. The distributions of particle sizes were similar in the diesel engine workshop and the water plant, with particle sizes <2.5 μm and <0.523 μm accounting for ∼99% and ∼50% of the total particle distribution in both spots, respectively. However, the number concentrations of total particle were significantly higher in the diesel engine workshop (403.84 ± 107.78 particles per cm³) compared to that in the water plant (92.53 ± 8.85 particles per cm³, P < 0.01).

Table 1. Air concentration of particulate matter at a water plant and a diesel engine testing workshop (mean ± SD).

Exposure	Water plant	Diesel engine testing workshop	P*
PM2.5, μg m–3	91.9 ± 3.4	282.3 ± 111.3	<0.001
Elemental carbon, μg m–3	11.8 ± 0.6	135.2 ± 56.6	<0.001
NO2, ppm	0.038 ± 4.873 × 10–4	0.26 ± 0.15	<0.001
SO2, ppm	0.004 ± 5.56 × 10–5	0.15 ± 0.07	<0.001
Total PAH concentration, μg m–3	0.035 ± 0.001	4.53 ± 1.83	<0.001
Number of particles in the air, particles per cm3 (% in total particles)
Particle size <0.523 μm	49.9 ± 5.8 (54.0)	230.3 ± 74.3 (57.0)	1.57 × 10–11
Particle size of 0.523–1.0 μm	40.5 ± 3.6 (43.8)	163.0 ± 34.0 (40.4)	6.76 × 10–18
Particle size of 1.0–2.5 μm	1.7 ± 0.06 (1.8)	9.0 ± 2.1 (2.2)	6.62 × 10–17
Particle size of 2.5–20 μm	0.37 ± 0.02 (0.4)	1.6 ± 0.5 (0.4)	1.01 × 10–11
Total	92.5 ± 8.8 (100)	403.8 ± 107.8 (100)	3.97 × 10–12

The demographic characteristics of the study population

All of the subjects included in this study are males. The demographic characteristics of the study subjects are summarized in Table 2. The distributions of age, BMI, and the history of smoking and alcohol use were not significantly different between the two groups. The average number of years of employment among the DEE-exposed workers was 8.18 ± 5.14 years.

Table 2. The characteristics of workers in the controls and DEE exposure groups.

Variables	Control (n = 108)	DEE exposed workers (n = 137)	P
Age (years, mean ± SD)	33.37 ± 11.05	31.99 ± 8.60	0.271 a
BMI (kg m–2, mean ± SD)	23.71 ± 4.26	24.58 ± 3.40	0.076 a
Current smoker, yes/no (% yes)	56/52 (51.9)	81/56 (59.1)	0.255 b
Alcohol user, yes/no (% yes)	68/40 (63.0)	89/48 (65.0)	0.895 b
1-Hydroxypyrene (μg g–1 Creat., Median, 5–95%)	0.84 (0.13–3.27)	2.30 (0.63–6.27)	<0.001 a
DEE exposed year (years, mean ± SD)	—	8.18 ± 5.14	—

^a t-test for difference between the two groups.

^b χ ² test for difference between the two groups.

Given the PAHs as predominant organic components of DEE, we measured the urinary 1-OHP level as an internal exposure dose. The results showed that the level of urinary 1-OHP in the exposed group (2.30 μg g^–1 Creat.) was significantly higher than that in the control group (0.84 μg g^–1 Creat.) (P < 0.001, Table 2).

Effects of DEE on inflammatory indices in the peripheral blood

Associations of DEE exposure with peripheral blood cell counts and lymphocyte subsets are shown in Table 3. No significant differences in the total white blood cell count, neutrophils, eosinophils, monocytes, hemoglobin, and platelets were observed between the control and exposure groups. We did not find the effects of DEE exposure on the lymphocyte subsets, including T cells, CD4⁺ T cells, CD8⁺ T cells, B cells, and NK cells.

Table 3. Peripheral blood cell counts and lymphocyte subsets in the controls and DEE exposed workers (mean ± SD).

	Control (n = 108)	DEE exposed workers (n = 137)	P crude  a	P adjust  b
White blood cells (109 per L)	6.61 ± 1.65	6.77 ± 1.67	0.45	0.82
Neutrophils (109 per L)	3.93 ± 1.29	4.10 ± 1.27	0.21	0.61
Eosinophils (109 per L)	0.14 ± 0.12	0.15 ± 0.13	0.34	0.28
Lymphocytes (109 per L)	2.24 ± 0.60	2.24 ± 0.64	0.83	0.29
T cell (109 per L)	1.49 ± 0.48	1.47 ± 0.45	0.80	0.22
CD4+ (109 per L)	0.76 ± 0.23	0.76 ± 0.26	0.88	0.24
CD8+ (109 per L)	0.63 ± 0.31	0.62 ± 0.25	0.89	0.42
B cells (109 per L)	0.25 ± 0.10	0.25 ± 0.12	0.58	0.20
NK cells (109 per L)	0.47 ± 0.23	0.46 ± 0.30	0.52	0.55
Monocytes (109 per L)	0.33 ± 0.15	0.31 ± 0.11	0.33	0.08
Hemoglobin (g per L)	163.6 ± 9.2	164.3 ± 10.1	0.60	0.89
Platelets (109 per L)	240.6 ± 58.7	234.9 ± 49.9	0.58	0.28

^a t-test for difference between the two groups.

^bAdjusted for age, smoking and alcohol status, and BMI.

The associations between DEE exposure and levels of CRP and cytokines in serum are shown in Table 4. The levels of IL-1β, IL-6, IL-8, and MIP-1β were significantly lower in DEE exposed workers compared to controls (P < 0.01). In contrast, the level of CRP in DEE-exposed workers was significantly higher than that in controls (P < 0.001). The cytokines including IL-1β, IL-6, IL-8, TNF-α, and MIP-1β showed significant positive correlations with each other, but not with CRP levels except for MIP-1β. The levels of CRP and IL-8 were significantly correlated with urinary 1-OHP (ESI Table 1†).

Table 4. The levels of cytokines and CRP in serum of the controls and DEE exposed workers (median, 5%–95%).

	Control (n = 108)	DEE exposed workers (n = 137)	P crude  a	P adjust  b
IL-1β (pg ml–1)	3.29 (0.21–16.73)	1.18 (0.09–11.70)	0.002	0.004
IL-6 (pg ml–1)	27.59 (1.70–141.04)	15.34 (0.80–68.34)	<0.001	<0.001
IL-8 (pg ml–1)	608.60 (157.18–1725.36)	414.40 (0.60–1170.84)	<0.001	<0.001
MIP-1β (pg ml–1)	821.03 (235.40–2810.92)	611.29 (0.40–2002.62)	0.001	0.002
TNF-α (pg ml–1)	78.43 (0.60–195.90)	74.83 (0.60–171.37)	0.858	0.745
MCP-1 (pg ml–1)	225.95 (70.12–420.64)	222.21 (0.65–420.09)	0.034	0.071
CRP (mg L–1)	0.10 (0.07–4.48)	0.70 (0.07–4.95)	<0.001	<0.001

^a t-test between the control and DEE-exposed groups.

^bAdjusted for age, smoking and alcohol status, and BMI.

To evaluate the consistency of these findings by smoking status, we conducted analyses among the smokers and non-smokers. As shown in Fig. 1, the above results were highly consistent with that after stratifying by smoking status. We also observed that smokers have higher levels of IL-1β, IL-6, IL-8, MIP-1β, and CRP compared with non-smokers among the control workers (Fig. 1).

Fig. 1. Levels of IL-1β (A), IL-6 (B), IL-8 (C), MIP-1β (D), TNF-α (E), MCP-1 (F), and CRP (G) in serum of DEE-exposed workers and controls by smoking status. The circle and triangle represent DEE-exposed workers and controls, respectively.

In order to study the relationship between exposure duration and inflammatory biomarkers, all subjects were divided into 4 groups by their working years (expressed as 0, <5, 5–10, and >10 years). We observed a statistically significant exposure time-dependent increase in CRP and a decrease in IL-6 (P_trend = 0.026 and 0.006, respectively). The exposure time–response relationship with IL-8 was borderline significant (P_trend = 0.078) (Table 5). The levels of IL-1β and MIP-1β were significantly lower in subjects with a DEE working time of less than 5 years and/or 5–10 years compared with the controls and then recovered to the control level in the subjects with working duration longer than 10 years (Table 5).

Table 5. Levels of cytokines and CRP stratified by DEE exposure duration (median, 5–95%).

Indicator	Controls (n = 108)	Exposure duration (years)			P trend #
		<5 (n = 33)	5–10 (n = 83)	>10 (n = 21)
IL-1β (pg ml–1)	3.29 (0.21–16.73)	1.18 (0.16–7.86)*	1.18 (0.04–11.43)*	2.54 (0.04–15.01)	0.43
IL-6 (pg ml–1)	27.59 (1.70–141.04)	20.58 (1.45–79.97)	16.47 (0.80–67.06)*	13.81 (0.58–66.78)*	0.006
IL-8 (pg ml–1)	608.60 (157.18–1725.36)	468.55 (0.60–1663.90)*	414.11 (0.60–1062.81)*	373.80 (234.41–1077.12)	0.078
MIP-1β (pg ml–1)	821.03 (235.40–2810.92)	663.12 (248.50–1481.72)	512.69 (0.40–2042.21)*	854.70 (230.97–2865.27)	0.54
TNF-α (pg ml–1)	78.43 (0.60–195.90)	97.22 (18.30–188.22)	60.08 (0.60–165.38)	73.67 (17.05–180.32)	0.70
MCP-1 (pg ml–1)	225.95 (70.12–420.64)	192.54 (78.02–457.75)	214.64 (0.65–411.95)*	265.02 (124.95–451.38)	0.74
CRP (mg L–1)	0.10 (0.07–4.48)	0.50 (0.07–3.60)	0.80 (0.07–5.64)*	0.80 (0.07–3.73)*	0.026

Discussion

Although convincing evidence has consistently demonstrated that short-term exposure of DEE was associated with airway inflammatory response,25 the effects of long-term exposure of DEE on the immune system have not been clarified. In the present study, systemic inflammation was investigated by measuring differential blood cell counts, lymphocyte subsets, levels of cytokines and CRP in serum among the workers with long-term exposure of DEE. We observed a significant decrease in IL-1β, IL-6, IL-8, and MIP-1β in DEE exposed workers compared with unexposed controls, suggesting immunotoxic effects caused by DEE exposure. These results were in agreement with an in vivo study that reported a significant suppression of mRNA expression levels of TNF-α, IL-1β, IL-4, and IL-6 in lung tissues of mice treated with DEE for 4 weeks.26 In contrast to the short-term exposure induced inflammation response that might reflect a normal response in healthy individuals rather than adverse health effects, the decrease of cytokine levels induced by long-term exposure of DEE might be associated with immune pathological injury. The exposure time-dependent patterns of DEE effects on the levels of CRP and IL-6 observed in this study highlight the importance of exposure length for the evaluation of adverse effects of DEE.

One previous study evaluated the blood cell counts and lymphocyte subsets among the salt miners exposed to DEE, salt dust and nitrogen oxides. They found evidence of the association between DEE exposure and increased levels of lymphocytes and T cells.27 In our previous pilot study in a diesel engine truck testing facility of 54 workers exposed to DEE and 55 unexposed comparable controls, we also reported that occupational exposure of DEE was associated with an elevation in the lymphocyte, CD4⁺ T cells, CD8⁺ T cells and B cells.28 However, we did not find significant differences in blood cell counts and lymphocyte subsets between DEE exposed workers and unexposed controls in the present study. The inconsistent results might be explained by the difference in exposure length, which is much longer in both prior studies (mean 24.4 years and 19.6 years, respectively) than in the present study (mean 8.2 years).

A multi-ethnic study reported that smokers have a significantly higher level of highly-sensitive CRP and IL-6 compared with non-smokers,29 suggesting that smoking status is a common confounder in inflammatory response. To control for the potential confounding effect of smoking, we adjusted for smoking status in the linear regression analyses. We also conducted analyses in subjects stratified by smoking status, and the findings on the association of DEE exposure with a decrease in cytokine levels were highly consistent across smokers and non-smokers. The associations of smoking with elevated IL-6, IL-8, IL-1β, and MIP-1β in the control group were found to be consistent with the findings of J. W. McEvoy et al.29 As both cigarette smoke and DEE are complex mixtures of chemicals and particles, it is difficult to identify the independent health effects of individual characteristics or components for DEE or smoke. There is also no universally accepted approach to measure the internal dose or body burden from DEE exposure.30 Some studies suggested urinary 1-OHP as a biomarker to assess occupational exposure to PAHs and to assess exposure to traffic-related air pollution.31–33 N. Brucker et al.34 observed linear positive correlations between 1-OHP and cytokines of IL-1β, IL-6 and TNF-α in taxi drivers and indicated that inflammatory response might be the consequence of major exposure to airborne PAHs. In the present study, the correlations of urinary 1-OHP with CRP and IL-8 were also found, suggesting that PAHs might be one of the contributors to the immunotoxic effects of DEE exposure. The identification of causal compositions needs further studies.

An interesting result is that the CRP level was elevated in DEE exposed workers compared with controls, which is in contrast to the finding of a decrease in cytokine caused by DEE exposure, and both changes in CRP and IL-6 showed DEE exposure time-dependent patterns. CRP is a widely used marker for systemic inflammation, which reflects that the host's principal immune response aimed at eliminating foreign substances invading the body and must rapidly terminate in order to prevent overreaction.35 The cytokines also play important roles in immunological processes; the seemingly contradictory results on CRP and cytokines might reflect the complexity of the immune response. On the other hand, cytokines such as IL-6 play important roles in host defense against various infections. The suppressive effect of DEE exposure on these cytokines could increase vulnerability to infections, which might be a possible reason for the elevated CRP in DEE exposed workers. In addition, we observed that the levels of IL-1β and MIP-1β recovered to control levels in workers with working time longer than 10 years. This change might be explained by the healthy-worker survivor effect. Workers with symptoms and diseases may leave or change work and consequently healthy workers exposed to DEE for a longer time increase the proportion of workers within the category of long-term exposed workers. This may explain the inconsistent findings regarding IL-1β and MIP-1β.

The main advantage of the current study was the selection of the study population with a high level of DEE exposure in specific occupational settings where the exposure uncertainties were far less than in studies associated with the general population. The long-term and high level of exposure was also beneficial for elucidating the adverse effects of DEE. However, we recognized that the limitation of the current study is the shortage of data on personal exposure assessment, so we cannot analyze the exposure–response relationship for cytokine expression. Moreover, the subjects were generally healthy male workers, so the results may not be applicable to more sensitive populations or to females.

In summary, our study showed that cytokines including IL-1β, IL-6, IL-8 and MIP-1β were significantly decreased and CRP were significantly increased among the workers exposed to DEE compared to controls, with evidence of an exposure time-dependent pattern for levels of CRP and IL-6. These findings provide epidemiological evidence for the relationship of long-term DEE exposure with immunotoxicity. Further studies are needed to evaluate additional immune markers and the underlying mechanism in occupationally exposed workers in order to illustrate the full extent of the immunotoxicity resulting from DEE exposure.