Occupational exposures to hazardous conditions are important risk factors for non-communicable diseases, including cardiovascular disease, hearing disorders, respiratory disorders, and cancers (1, 2). Although the development of workplace protection devices, such as protective clothing and masks, has dramatically reduced environmentally associated occupational hazards, individuals working in particular environments, such as coal mines and steel mills, are exposed to the health conditions that are less than optimal (3, 4). Moreover, no biomarkers have yet been identified to efficiently evaluate the health conditions of those workers prior to their development of diseases with obvious symptoms.

Proteins in the circulatory system are indicative of the physiology of individuals. Plasma protein level reflects allele differences, inflammatory status, and metabolic status (5). The variations of plasma protein levels in a number of diseases like hepatic and renal disease, acute myocardial infarction, cancer, diabetes, hyperlipidemia, and inflammatory diseases have been reported decades ago (6). More than 100 plasma proteins are cleared or approved by US Food and Drug Administration (FDA) as biomarkers clinically (e.g., C-reactive protein increase is used to predict coronary disease risk; thyroglobulin level is used to detect the recurrence of metastatic thyroid cancer after thyroid removal) and more and more plasma protein biomarkers are developed (5, 7, 8). Identifying plasma biomarkers that can monitor the health status of individuals working in extreme environments would be advantageous. Although the technology of plasma proteome profiling is rapidly developing (9), profile plasma proteins for a large cohort of workers working in a poor working environment, however, would be both costly and labor-intensive.

Mendelian randomization (MR) is a genetic epidemiological method that use genetic variants as instrumental variables to assess the causal relationship between exposures and outcomes (10). MR analysis has been used to investigate the causal association between fibroblast growth factor 23 level and bone mineral density of heel and femoral neck (11). The causal associations between educational attainment and 14 urological and reproductive health outcomes are also assessed via MR analysis (12). In this study, we performed two-sample MR to screen and evaluate the relationships between working in harmful environments (including very cold, hot, dusty, and noisy workplaces, as well as exposure in the workplace with chemicals and other fumes, cigarette smoke, and diesel exhaust, based on workers' self-reports) and the plasma concentrations of 2,992 proteins in European population. This study also assessed the relationships between plasma proteins potentially affected by workplace environments and the risks of disease by screening all 182 diseases related Genome Wide Association Study (GWAS) datesets collected by MRBASE (13) to estimate whether the variant of these plasma protein levels have potential effect on disease risks.

This study evaluated the roles of the workplace environment on plasma protein levels. These analyses showed that distinct plasma proteins are affected by different environments. Cold, hot, and dusty environments had less of an effect on plasma protein levels than noise, chemical fumes, passive smoking, and diesel exhaust.

COL1A1 is likely to be decreased when working in hot environment and GJA8 is likely to be increased when working in dusty environments based on present study. COL1A1 encodes the pro-alpha1 chain of type I collagen, a fibril-forming collagen found in most connective tissues. COL1A1 may be a biomarker and therapeutic target for multiple cancers (20, 21). A recent study showed that a plasma COL1A1 concentration >256.5 ng/mL was associated with poor survival from heart failure in heart transplant recipients (22). The decrease in plasma COL1A1 induced by hot workplace environments may positively affect health. GJA8 is a connexin protein necessary for lens growth, with mutations in the gene encoding this protein associated with multiple eye diseases (23, 24). There is no evidence that an increased GJA8 level has adverse effects.

Thermoregulatory systems in mammals are robust and reliable after millions of years of evolution. Although cold (winter season) has been reported to increase cardiovascular disease morbidity and mortality rates (25, 26), the effect of exposure to cold is unclear. The causes of these increases in morbidity and mortality are complex, and the winter season cannot fully represent exposure to cold. Exposure to cold has been found to activate brown adipose tissue (BAT), which is believed helpful for physical fitness (27–30). Sauna bathing is a type of exposure to heat associated with a reduced risk of sudden cardiac death (31). However, sauna bathing is always coupled with cooling periods in cold environments, suggesting the need for studies of physiological adaptations during the shift from hot to cold environments (29). Based on these findings and the results of the present study, temperature in the workplace is not likely to affect workers' health wearing proper protection. Dusty environments have been associated with respiratory diseases in the absence of proper protection (32). Dust particles, however, are rather large, with most face masks preventing exposure to dust. Indeed, the incidence of pneumoconiosis has decreased in recent years (33).

The present study found that exposure to noise, diesel exhaust fumes, second-hand cigarette smoke, and chemical or other fumes have distinct effects on plasma protein concentrations. Further MR analysis of the plasma proteins affected by workplace environments and disease risks have shown results consistent with previous findings. For example, many studies have shown that overexposure to noise is associated with tinnitus (34, 35). The present study showed that noisy workplaces decreased plasma concentrations of ST3GAL6 and SCG3, proteins thought to protect against tinnitus. This study also showed that workplace exposure to second-hand smoke was associated with increased plasma levels of CA10, a risk factor for lung cancer. These results suggest that the relationships between workplace environments and the identified plasma proteins were both reliable and significant.

MR analyses in the present study have found associations of environmentally affected plasma proteins with oral cavity and oropharyngeal cancers. Household air pollution from the utilization of solid fuels was found to be an independent risk factor for oral cancer (36), and exposure to petroleum-based and oxygenated solvents was associated with the risks of oral and oropharyngeal cancers, although no exposure-response trend was observed (37). The findings in the present study are consistent with these earlier results, providing new evidence for the association between environmental pollution and oral and oropharyngeal cancers.

Most plasma proteins identified in this study are not secreted proteins. They most likely come from exosomes secreted by cells in a particular tissue or organ. Although MR analysis is robust and reliable, further study is needed to validate these biomarkers. The function of these proteins in particular occupational hazards is also worth further investigation.

This study suggested that harsh environments may affect the concentrations of multiple plasma proteins and the change of those plasma proteins concentration have great potential to be used as biomarkers to predict occupational hazards risk. Twenty plasma proteins are recommended as biomarkers to monitor the health condition of workers working in noisy workplaces, workplaces full of diesel exhaust, workplaces full of smoke from other smokers, and workplaces full of chemical or other fumes. More epidemiologic studies are needed to test the sensitivity, specificity, and overall accuracy of these candidate biomarkers.

The source data for Figure 2 are included in Supplementary Data 2. The source data for Figure 3 are included in Supplementary Data 3. Further inquiries can be directed to the corresponding authors.

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

AL: conceptualization, methodology, software, and writing—original draft preparation. WL, JX, and LS: writing—reviewing and editing. XZ: supervision. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.