Manganese (Mn) toxicity has been associated with neurotoxic effects including decreased motor function (Martin et al., 2020; O’Neal and Zheng, 2015). Occupational exposure to the metal has resulted in lower psychomotor function (Lucchini et al., 1997), general motor function (Ma et al., 2018), and manual dexterity (Cowan et al., 2009; Wells et al., 2018). Chronic manganese exposure has been associated with increased reaction time, impaired finger-tapping and digit-span performance scores (Wennberg et al., 1991), as well as grooved pegboard scores (Wastensson et al., 2012). The development of manganism, a disease like idiopathic Parkinson’s disease, can occur in cases of high or chronic manganese exposure (Lucchini et al., 2009). Additionally, recent evidence suggests that manganese exposure may be associated with Parkinson’s disease severity (Racette et al., 2021).

One of the most widely used biomarkers of Mn exposure is whole blood Mn (BMn) (Ge et al., 2018; Lucchini et al., 1997; Mirmohammadi et al., 2017). Despite its frequent use, previous research has suggested that the relationship between BMn and Mn exposure can be complex (Baker et al., 2014; Smith et al., 2007). BMn has been shown to reflect recent exposure, generally exposure occurring within days, due to the short half-life of Mn in the blood (Laohaudomchok et al., 2011; O’Neal and Zheng, 2015; Zheng et al., 2011). Idiopathic Mn-induced parkinsonism, also referred to as manganism, is a chronic disease which likely results from cumulative Mn exposure over varying lengths of time (Cersosimo and Koller, 2006). A biomarker reflecting recent exposure may not reflect long-term exposure. Until recently, biomarkers reflecting long-term Mn exposure were not available. Thus, the relationship between chronic Mn exposure and motor dysfunction is still unclear due to the lack of a biomarker for cumulative Mn exposure (Chen et al., 2016).

Bone manganese (BnMn) has previously been suggested as a viable biomarker of cumulative Mn exposure due to ~40% of Mn’s body burden being found in bone and a relatively long half-life of the metal in bone, estimated at 8.6 years in human bone (Arnold et al., 2002; O’Neal et al., 2014). BnMn was also associated with Mn concentrations found in the brain, suggesting that BnMn could be reflective of Mn in target tissues of neurologic function (O’Neal et al., 2014). More recently, a lifetime Mn exposure study in rats reported that bone Mn was reflective of recent ongoing exposure but not lifetime exposure (Conley et al., 2021).

Limited research has assessed the relationship between BnMn and Mn exposure in human populations. When analyzed in a group of welders versus controls, BnMn was higher in the occupationally Mn-exposed welders compared to the non-occupationally exposed controls (Pejović-Milić et al., 2009). To continue assessing BnMn at different field sites, our research team developed a transportable in vivo Neutron Activation Analysis (IVNAA) (Liu et al., 2017, 2014, 2013). This transportable IVNAA system was used to assess BnMn in a group of 19 adult males, where BnMn was found to be significantly associated with decreased manual dexterity (Wells et al., 2018).

As part of a continued effort to develop a feasible noninvasive biomarker for assessing the relationship between chronic Mn exposure and motor dysfunction, we applied the transportable IVNAA technology to a larger cross-sectional study. In this study, BnMn was significantly associated with a Mn cumulative exposure index reflecting cumulative occupational Mn exposure over the previous 16 years (MnCEI₁₆), but BMn was not associated with the cumulative exposure index (Rolle-McFarland et al., 2018). Here, we assess the relationship between BnMn or BMn with motor dysfunction in the same population with high, chronic occupational Mn exposure.

This study assessed whether BMn or BnMn were associated with decreasing motor function in a group of Chinese workers. In adjusted models, there were significant associations of BMn with postural tremor (both hands) and fine motor dexterity (Purdue Pegboard), suggesting an association of higher BMn with better performance on these motor assessments. These were unexpected findings. Another unexpected finding was significant associations in adjusted models of BnMn with better performance in postural stability (sway). However, we did observe BnMn was associated with worse performance in postural tremor (dominant hand tremor) and motor coordination (finger-tapping), as hypothesized. Interestingly, while higher BMn was significantly associated with better performance in the fine motor dexterity task, higher BnMn was associated with worse performance in the fine motor dexterity task, although this did not reach statistical significance.

BMn concentrations in this population (geometric mean=14.0 μg/L) is within the range of BMn concentrations reported in prior studies of populations with occupational Mn exposure from a variety of industries (Baker et al., 2014). Average BnMn concentrations in this study are higher than prior studies among welders and controls in the United States (Wells et al., 2018) or Canada (Pejović-Milić et al., 2009) as well as environmentally exposed older adults from Italy (Conley et al., 2021). The higher BnMn in this population is likely in part due to our inclusion of ferroalloy factory workers; differences with Conley et al may also reflect differences in BnMn assessment methods.

Several other studies have reported that Mn exposure, in particular chronic Mn exposure, is associated with lower performance on motor coordination assessments. Blond and Netterstrom report a decrease in manual coordination over an 8 year period among 60 steel workers; specifically, seniority was associated with a decrease in precision of the protonation/supination test (Blond and Netterstrom, 2007). Ellingsen et al. reported an association of air Mn concentrations with lower scores on finger-tapping assessments among Russian workers (Ellingsen et al., 2008). A longitudinal assessment of this cohort also found this association, but only among welders with elevated carbohydrate-deficient transferrin (CDT), a measure of chronic alcohol exposure (Ellingsen et al., 2015). Two studies using models to estimate chronic manganese exposure also reported a significant association of higher manganese exposure with lower scores in finger-tapping assessments among retired German workers (Pesch et al., 2017) and adults from Ohio, USA (Bowler et al., 2016). In contrast, long-term manganese exposure, determined by a cumulative exposure index, was not associated with performance on finger-tapping assessments in a study of former ship welders (Wastensson et al., 2012). The finding in this study that BnMn (also a measure of chronic exposure) is associated with worse finger-tapping performance is consistent with these prior reports.

In our study, we observed a significant association between increased BnMn and increased postural stability, specifically decreased sway measured when using foam with closed eyes. This is in contrast to two cross-sectional studies of environmental (Kim et al., 2011) and occupational manganese exposure (Bowler et al., 2007), both of which report associations of manganese with increased sway. However, our results are somewhat consistent with the report by Ellingsen et al. where a longitudinal study finds a lower, but not significant, mean sway among welders compared to controls (Ellingsen et al., 2015).

In this study, we found significant associations of BMn with lower tremor intensity and center frequency; however, we found significant associations of BnMn with higher center frequency. The literature related to measurements related to postural hand tremor is also mixed. Associations have been reported between air manganese and increased tremor among welders (Bowler et al., 2007), chronic air manganese exposure with increased harmonic index among US adults (Bowler et al., 2016), and increased frequency dispersion among Russian manganese alloy plant workers (Bast-Pettersen et al., 2004). While no significant difference in CATSYS tremor scores based on occupation was reported in Ellingsen’s longitudinal study of Russian welders, they did report that welders had worse performance in a Static Steadiness test which also evaluates hand tremor (Ellingsen et al., 2015). In contrast, active and retired German welders (Lotz et al., 2021) as well as former ship welders (Wastensson et al., 2012) were reported to perform better in assessments requiring hand steadiness compared to the reference groups.

Our results for fine motor coordination were also mixed: we found improved Purdue Pegboard scores among those with higher BMn and worse scores (but not significant) among those with higher BnMn. Prior studies from members of our team found that poorer performance on the Purdue Pegboard was significantly associated with a different measure of BMn not utilized in this study; a blood manganese-iron ratio (Cowan et al., 2009). Our group has also previously seen poorer performance on the Purdue Pegboard significantly associated with BnMn (Wells et al., 2018). Other work has used a similar assessment, the Grooved Pegboard test. Ellingson et al’s study of Russian welders (Ellingsen et al., 2015) as well as a recent study by Racette et al among adults in South Africa (Racette et al., 2021), both reported that manganese exposure was associated with lower performance in the grooved pegboard assessment. Thus, our results in this study for BnMn (chronic exposure) are similar to what has been reported previously (which mostly use estimates of chronic exposure), while our results for BMn (recent exposure) is not.

It is notable that despite the substantial evidence that manganese is a neurotoxin that affects motor function (Aschner, 2000; Martin et al., 2020; Meyer-Baron et al., 2009; O’Neal and Zheng, 2015), several occupational studies, including this report, present at least some data suggesting that workers with higher manganese exposure have somewhat better performance on some motor function tasks compared to controls (Dlamini et al., 2020; Lotz et al., 2021; Pesch et al., 2017). There are a few possibilities for these results. It is possible that the healthy worker effect, where participants who remain in positions with high manganese exposure may be, on average, healthier than others (Dlamini et al., 2020; Lotz et al., 2021). Another possibility is that workers with high manganese exposure are more likely to engage in motor-related tasks through their job responsibilities, and thus have more practice with this type of assessment (Lotz et al., 2021; Wastensson et al., 2012). It is also possible that differences between the different study populations such as genetics, other factors such as iron (Cowan et al., 2009; Pesch et al., 2017), or exposure characteristics (concentration and assessment of chronic versus recent exposure) might explain these divergent results. More investigation would be needed to definitively address this question.

A limitation of our study was the possibly reduced statistical power due to the relatively small sample size. On the other hand, this population had relatively high levels of Mn exposure; thus, a smaller sample size is sufficient to identify significant associations. This study population was limited to Chinese men with higher occupational Mn exposures that are typically reported within the United States; thus, another potential limitation is that our study population somewhat limits the generalizability of our results. As this is a cross-sectional study, there are limits to our ability to establish temporality. However, this is somewhat allayed by the fact that BnMn reflects exposure over multiple years (O’Neal et al., 2014; Rolle-McFarland et al., 2018), although does not appear to represent lifelong cumulative exposure (Conley et al., 2021; Rolle-McFarland et al., 2018). Thus, although the data were collected at one point in time, we are confident that our BnMn measurements represent long-term exposure.

Strengths of this study include the utilization of well-established motor tests that have been successfully used before in non-English speaking populations (Cowan et al., 2009; Iwata et al., 2007). The design was also strengthened by utilizing bilingual English and Mandarin research assistants. The use of the IVNAA system to measure BnMn was also a strength of this study. This is a novel technology only available to a few research groups; this allowed us to obtain quantitative estimate of BnMn, which reflects cumulative exposure. In contrast, alternative methods to assess long-term Mn exposure typically rely at least in part on qualitative and/or self-reported data.

In summary, we describe the relationship between BMn and BnMn with selected measures of motor function from a cross-sectional study of 60 male Chinese workers. The results presented in this study suggest increased concentrations of BnMn are associated with decreases in select motor outcomes. This study not only verifies the practical utility of the newly developed transportable IVNAA system, but more importantly provides critical information on whether BnMn is associated with subtle motor dysfunction as a result of chronic Mn exposure among smelters. When considered along with prior research, our results highlight the importance of incorporating a useful measure of chronic Mn exposure when assessing chronic outcome measures. It is recommended that future work in this area continue to explore the utility of BnMn as a biomarker and should include longitudinal studies of more diverse populations.