5
Toxicity of Selected Slag Constituents

This chapter discusses potential health outcomes related to exposures to two chemicals of potential concern (COPCs) in unencapsulated electric arc furnace (EAF) slag, which are specified in the committee’s statement of task: chromium (Cr) and manganese (Mn). Due to data limitations in the scientific literature, the committee was unable to examine quantitative relationships between human health outcomes directly from the levels of Cr and Mn exposures expected from unencapsulated slag used for residential applications. Therefore, the committee relied on studies of health effects from Cr and Mn exposures in general. This chapter also considers susceptible groups of people who may be at increased risk for health effects following Cr or Mn exposures, due to factors such as life stage, genetics, sex, health status, and disease. The next chapter considers disadvantaged communities who may be at increased health risk from slag exposures in terms of stressors related to race or ethnicity, lifestyle factors, and disproportionate exposures to chemicals from sources other than slag.

The U.S. Environmental Protection Agency (EPA) derives toxicity values for noncancer and cancer effects from exposure to individual hazardous chemicals:¹

• Reference concentration (RfC): an estimate of a continuous inhalation exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime.
• Reference dose (RfD): an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily oral exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime.
• Oral slope factor: an estimate of the increased cancer risk from oral exposure to a dose of 1 mg/kg-day for a lifetime. The slope factor can be multiplied by an estimate of lifetime exposure (in mg/kg-day) to estimate the lifetime cancer risk.
• Inhalation unit risk: an estimate of the increased cancer risk from inhalation exposure to a concentration of 1 µg/m³ for a lifetime. The unit risk can be multiplied by an estimate of lifetime average exposure concentration (in µg/m³) to estimate the lifetime cancer risk.

The committee recognizes that EAF slag is a complex mixture of multiple COPCs and the composition can vary from one steelmaking plant to another, depending on the composition of the scrap steel fed into the furnaces and other factors. In addition, challenges are inherent in assessing health risks from chemical mixtures due to the complexity and variability of the toxicological interactions and overall effects resulting from different COPCs in a mixture (Grover and Farley, 1987). Therefore, the health risk considerations may vary if, for example, slag applied at a site contained significant concentrations of COPCs other than Cr and Mn, such as arsenic or vanadium.

Often there are limited or no data available on the toxicity of particular chemical combinations or the effects of long-term exposure to low levels of chemicals (Elcombe et al., 2022). Furthermore, many chemicals have exposure–response relationships that vary by life stage, and specific effects at various life stages may be unknown (Rumrich et al., 2020).

Responses to chemicals in mixtures may be additive in that they reflect the sum of effects of the individual components or the sum of the exposure concentrations (considering relative toxicities). In contrast, toxicological interactions could cause the effect of the mixture to differ from the effect that would be predicted based on the effects of individual chemicals. The various types of interaction can be synergistic (greater-than-additive) or antagonistic (less-than-additive). For example, one chemical may additively

___________________

¹ See https://www.epa.gov/iris/basic-information-about-integrated-risk-information-system.

enhance the toxicity of another chemical, while a third may reduce it or increase it; thus, the effects can be extremely contextual.

Human exposure to multiple chemicals through different routes, such as inhalation, ingestion, or skin contact, can also impact toxicity. As mentioned, the timing of exposure to chemicals in mixtures can also impact their toxicity.

Although reliance on health effects data for the mixture of concern is the preferred approach, such data are generally lacking. Due to the foregoing factors, there may be a higher level of uncertainty in risk assessments for chemical mixtures than for individual chemicals, which can make it difficult to accurately predict the potential health effects of exposure. A particular issue is whether a mixture of components, each of which is present at concentrations without an appreciable risk of deleterious effects when considered individually, may be more hazardous than the components’ additive effects due to interactions.

In discussing the potential health outcomes of two metals commonly found in slag, the committee notes that other slag components may have additional health impacts, including component-specific effects and effects due to the complex mixture for a particular form of slag.

CHROMIUM

Cr is a ubiquitous element present in soil, water, air, and food that can originate from both natural and anthropogenic sources. Cr was named from the Greek word “chroma,” meaning color, because of the many colorful compounds made from it depending on its oxidation state. It is part of the mineral crocoite (lead chromate). Most naturally occurring chromium is found in the trivalent state (Cr³⁺) in chromite ores that are processed to sodium- or potassium-dichromate, which are hexavalent chromium (Cr⁶⁺) compounds. Cr compounds are used for stainless steel production as well as for welding, chromium plating, ferrochrome alloys, chrome pigment production, and tanning industries (ATSDR, 2012a).

Important sources of Cr emissions to ambient air include ferrochrome production, ore refining, and chemical processing. Workers and individuals living in close proximity to industrial sources are likely to be exposed to Cr⁶⁺. Environmental exposure to Cr can occur through food, water, air, and consumer products. The most common exposure pathway is through contaminated water, particularly in areas with high industrial activity. Widespread industrial uses have increased Cr concentrations on land, water, and sediments. For example, up to 38 percent of drinking water supplies in California have detectable levels of Cr⁶⁺. However, little is known about the relationships between human health effects and various levels of environmental exposures (Sedman et al., 2006; Smith and Steinmaus, 2009; Sun et al., 2015). In addition to being toxic to humans, Cr⁶⁺ compounds are toxic to other organisms, although some microbial organisms and plants have developed tolerances to high chromium concentrations in the environment (Cervantes et al., 2001).

The majority of chromium in EAF slag is in the form of Cr³⁺ with trace amounts of Cr⁶⁺. In the environment, most Cr³⁺ compounds are insoluble in water, immobile in soils, and relatively inert, limiting their oxidation to Cr⁶⁺ (Hausladen and Fendorf, 2017; Fendorf et al., 2000) and toxicity. In contrast, most Cr⁶⁺ compounds are readily soluble in water, highly mobile, and bioaccessible (Fendorf et al., 2000; Hausladen and Fendorf, 2017). However, conversion of Cr⁶⁺ to Cr³⁺ readily occurs in the environment under reducing conditions, especially in the presence of ferrous iron, sulfides, and organic matter (Fendorf et al., 2000; Hausladen and Fendorf, 2017). The rate of Cr⁶⁺ reduction to Cr³⁺ in the environment is highly variable and can range from seconds to days. Oxidation of Cr³⁺ to Cr⁶⁺ may occur under oxidizing conditions in the presence of manganese oxides and low organic carbon (Hausladen and Fendorf, 2017).

Toxicokinetics

Dermal absorption depends on the chemical form, vehicle, and integrity of the skin (ATSDR, 2012a). Concentrated potassium or sodium chromate may cause chemical burns to the skin and facilitate absorption. Inhaled Cr⁶⁺compounds are absorbed in the lung via transfer across alveolar cell membranes. Likewise, ingested Cr⁶⁺ is absorbed across the epithelial lining of the gastrointestinal (GI) tract. However,

the bioavailability of Cr compounds is limited by rapid reduction of ^Cr6+ to Cr³⁺ by extracellular biological fluids in the airway and GI tract lumen. Systemic bioavailability beyond the luminal epithelium is minimal unless the reductive capacity of the lumen and epithelial cells is exceeded (EPA, 2022a). A caveat to the National Toxicology Program (NTP)–conducted oral toxicity studies in rodents was that Cr⁶⁺ absorbed by the GI tract and systemically distributed occurred only at oral ad libitum doses greater than or equal to 1 mg/kg-day (EPA, 2022a). This threshold may be higher in humans, since Cr⁶⁺ would be reduced more efficiently at the lower pH of the human stomach, relative to the rodent (pH 1.3 versus pH 4.0). Cr⁶⁺that is not reduced in the stomach is more stable and available for uptake by the intestine where pH is more neutral (EPA, 2022a). Rodent studies using dosing that exceeded the reductive capacity of the GI tract demonstrated widespread organ distribution. Particles containing Cr can be retained in the lungs for years.

Cr⁶⁺ that is not reduced in the airways or GI tract enters the blood and is readily taken up by erythrocytes, while Cr³⁺ is only loosely associated with erythrocytes. Cr compounds from excessive doses can be distributed to all organs of the body, with higher levels found in the liver, spleen, and kidney. However, evidence—for exposure of systemic tissues to Cr—is inconsistent due to variable dose–response ranges between studies (EPA, 2022a) and due to variable reductive capacities in different tissues. Absorbed chromium is excreted primarily in urine. The half-life for excretion of oral potassium chromate is about 35 to 40 hours (Sedman et al., 2006; ATSDR, 2012a). Given the poor level of systemic distribution, the hazards of Cr⁶⁺ exposures, especially at levels relevant to EAF slag, are largely limited to the airways and GI tract.

Cr⁶⁺ readily crosses epithelial cell membranes via sulfate and phosphate transporters, while Cr³⁺ compounds form octahedral complexes making entry into cells difficult (Standeven and Wetterhahn, 1989; ATSDR, 2012a). This is the main reason that inhaled or ingested Cr³⁺ is not toxic. Once Cr⁶⁺ enters cells, it is reduced intracellularly by ascorbic acid, glutathione, and/or cysteine, ultimately to Cr³⁺ that becomes bound to proteins or nucleic acids. It is thought that the toxicity of Cr⁶⁺ results from damage to cellular components during this reduction process, including the generation of free radicals and the formation of DNA adducts (Zhitkovich, 2005; Reynolds et al., 2012; Sun et al., 2015).

Toxicity

Cr³⁺ is relatively inert and considered to be nontoxic to humans. The essentiality of Cr³⁺ as a nutrient has been controversial, since it has been difficult to find populations that are deficient in the metal. There is evidence suggesting that Cr³⁺ serves as a cofactor for insulin signaling in glucose metabolism (Vincent, 2013; Bailey, 2014). While there are suggestions that tissue levels of Cr³⁺ are reduced among diabetic individuals, meta-analysis of a large body of clinical trials of Cr³⁺ supplementation to improve glucose metabolism in diabetics has shown no beneficial effects (Bailey, 2014). However, Cr³⁺ supplementation is nontoxic with 35 and 25 μg/kg-day considered to be safe and adequate intake for men and women, respectively (IOM, 2001). Given the lack of toxicity of Cr³⁺ and negligible potential for it to be converted to Cr⁶⁺ under conditions of residential EAF slag use, the remaining discussion of potential toxicity from EAF slag will focus on Cr⁶⁺.

It should be noted that Cr³⁺ is frequently found in dietary supplements, and the growing popularity of Cr³⁺ compounds and their consumption in the form of dietary supplements creates some concern as to the safety of its use, although little data exist to show toxicity. Likewise, the literature supporting a nutritional role for Cr³⁺ as an essential nutrient is controversial, and there is no clear benefit to Cr supplementation despite its frequent use (Maret, 2019; Vincent, 2017).

Carcinogenesis is the primary endpoint for the majority of studies in the literature on the human health impact of exposure to Cr⁶⁺ in humans. Human hazards from Cr⁶⁺ exposure have been attributed primarily to airborne Cr⁶⁺ compounds in industrial and occupational settings (Gibb et al., 2000a, 2000b). There is high confidence that inhaled Cr⁶⁺ is a human lung carcinogen, primarily with occupational exposures (EPA, 2022a).

Cr⁶⁺, being a strong acid, is corrosive and may cause chronic ulceration and perforation of the nasal septum, as well as chronic ulceration of skin surfaces (ATSDR, 2012a; Singhal et al., 2015). Occupational exposure to Cr⁶⁺ may be a cause of asthma (Bright et al., 1997; Antonini et al., 2004; Mattila et al., 2021).

Accidental ingestion of high doses of Cr⁶⁺ compounds may cause acute renal failure characterized by proteinuria, hematuria, and anuria, but kidney damage from lower-level chronic exposure is equivocal (ATSDR, 2012a). Other noncancer hazards, primarily in airways and the GI tract, have been identified in rodents. Due to its toxicokinetics, Cr⁶⁺ is viewed as primarily a point of contact toxicant unless exposures are sufficiently high to overwhelm the reductive capacity of biological fluids in the airways or GI tract.

Noncancer Toxicity

Using primarily high-confidence rodent studies and several low-confidence epidemiology studies indicating that Cr⁶⁺ is likely to cause GI tract and hepatic toxicity in humans, EPA derived several organ-specific RfDs. Hyperplasia of the small intestine epithelium in female mice was used to calculate an overall RfD of 9 × 10^-4 mg/kg-day. Human equivalent doses were calculated using physiologically based pharmacokinetic modeling to account for species differences and human variability in detoxification of Cr⁶⁺ in the stomach, and an overall uncertainty factor of 100 was applied. An oral RfD for hepatic inflammation was calculated to be 7 × 10^-4 mg/kg-day based on chronic hepatic inflammation in female mice. The oral RfD for hematological toxicity was 0.01 mg/kg-day based on decreased hemoglobin B in male rats. Using a low-confidence study, an oral RfD of 0.07 mg/kg-day for developmental effects was calculated based on poor weight gain in the F1 generation.

Inhalation of Cr is well recognized as an occupational hazard to the nasal passages and upper respiratory tract. Inhalation of Cr⁶⁺ particles causes irritation and inflammation. Chronic exposure causes damage to the lungs and respiratory tract and in severe cases cancer of the upper and lower airways. However, the studies to support a quantitative RfC are of medium confidence. An RfC of 1 × 10^-5 mg/m³ for upper respiratory tract toxicity was calculated based on a single longitudinal occupational study. Several rodent studies demonstrated lower airway hyperplasia and were used to calculate an RfC of 1 × 10^-4 mg/m³. However EPA defaulted to the more conservative human RfC for respiratory disease of 1 × 10^-5 mg/m³ (EPA, 2022a).

Cancer

Cr⁶⁺ compounds are classified as known human carcinogens by the NTP (2011) and the International Agency for Research on Cancer (IARC, 2012). Occupational exposure to Cr⁶⁺ compounds, particularly in the chrome production and pigment industries, is associated with increased risk of lung cancer (Gibb et al., 2000b). An association of Cr⁶⁺ in the drinking water with stomach cancer has also been suggested (Sedman et al., 2006; Smith and Steinmaus, 2009), but the evidence that Cr causes stomach cancer or other cancers is inconclusive (IARC, 2012; EPA, 2022a). The mode of action of Cr⁶⁺ compounds is genotoxicity (EPA, 2022a). As discussed previously, during the intracellular reduction process, various genetic lesions can be generated, including Cr-DNA adducts, DNA–protein cross-links, DNA-Cr intrastrand cross-links, DNA strand breaks, and oxidized DNA bases (O’Brien et al., 2003; Macfie et al., 2010). Cr⁶⁺ compounds are mutagenic, causing base substitutions, deletions, and transversions in bacterial systems, and hypoxanthine guanine phosphoribosyl transferase and supF mutations in mammalian mutagenesis systems (O’Brien et al., 2003). During the reduction process, Cr intermediates can also react with other cellular constituents, causing reactive oxygen radical generation, inhibiting protein synthesis, and inducing apoptosis (Zhitkovich, 2005; Salnikow and Zhitkovich, 2008). Exposure to Cr⁶⁺ can interfere with DNA damage repair by disrupting the p53 signaling pathway, altering the ATM/ATR cell cycle checkpoints, and arresting DNA replication (Zhitkovich, 2005; Salnikow and Zhitkovich, 2008). All of these effects may play an integrated role in Cr carcinogenesis (O’Brien et al., 2003; Sun et al., 2015).

EPA and international agencies, such as IARC and the National Institute for Public Health and the Environment, Netherlands (RIVM), have indicated that oral exposure to Cr⁶⁺ may cause stomach cancer and possibly laryngeal cancer in humans, but the weight of evidence is low and mostly relies on sex-related observations in high-dose rodent studies (den Braver-Sewradj et al., 2021; EPA, 2022a; IARC, 2012). There is currently insufficient evidence that oral Cr⁶⁺ can cause cancer of the small intestine, oral cavity, or other

organs in humans (den Braver-Sewradj et al., 2021; EPA, 2022a). This is primarily due to Cr⁶⁺ being a point-of-contact carcinogen with limited systemic bioavailability. Under EPA’s Guidelines for Carcinogen Risk Assessment (EPA, 2005b), oral Cr⁶⁺ is likely to be carcinogenic to humans based on the robust evidence of carcinogenicity to the GI tract in the NTP rodent studies but only slight evidence of carcinogenicity from human studies (EPA, 2022a). There is strong supporting mechanistic evidence for Cr⁶⁺ involvement in biological pathways contributing to carcinogenesis. Thus, based on tumors of the small intestine of male and female mice, EPA used the NTP rodent studies to derive a human adult-based oral slope factor for Cr⁶⁺ of 0.3 mg/kg-day (EPA, 2022a). The total lifetime oral slope factor is 0.5 mg/kg-day, and, in accordance with EPA guidelines, age-dependent adjustment factors were applied to provide partial oral slope factors for different age groups (EPA, 2022a).

Inhaled Cr⁶⁺ is well recognized as a human lung carcinogen (den Braver-Sewradj et al., 2021; EPA, 2022a; IARC, 2012). Given the plausibility of the pathways that lead to Cr⁶⁺ genotoxicity and the wealth of primarily occupational studies demonstrating that Cr⁶⁺ is a lung carcinogen, a total lifetime inhalation unit risk for lung cancer has been set at 2 × 10^-2 µg /m³ (EPA, 2022a). Inhaled Cr⁶⁺ is not associated with cancers in other organs or tissues.

Susceptible Groups and Life Stages

Due to bioavailability the airways and GI tract are the primary targets of Cr⁶⁺-promoted cancers and noncancer disease. Thus, oral or inhaled Cr⁶⁺ exposure might exacerbate health conditions in individuals with pre-existing pathologies or disease burdens in the GI, liver, lungs, and blood. Concurrent risk factors, such as smoking, can contribute to the risk of Cr-induced cancer and may act synergistically with Cr. While the bioavailability of the levels of Cr⁶⁺ found in slag may be below the threshold for cancer risk due to rapid reduction of Cr⁶⁺ to Cr³⁺ in saliva and the upper GI (EPA, 2022a), the efficiency of detoxifying reduction of Cr⁶⁺ to Cr³⁺ in the stomach is highest at a low pH. Individuals with low stomach acid (hypochlorhydria) or elevated pH due to medications inhibiting acid release would be at higher risk of GI toxicities due to higher Cr⁶⁺ bioavailability (EPA, 2022a). Neonates and infants less than 30 months old have elevated stomach pH that could increase the bioavailability of oral Cr⁶⁺. Likewise, elderly people with pre-existing conditions and elevated stomach pH due to medications would be at risk due to increased bioavailability.

The mode of action for Cr⁶⁺-promoted cancers is genotoxicity and inhibition of DNA repair. Individuals with genetic polymorphisms conveying deficiencies in DNA repair capacity may have increased risk of GI and/or lung cancers following Cr⁶⁺ exposure. Cr⁶⁺ has been shown to inhibit expression of cystic fibrosis transmembrane conductance regulator (CFTR), a suppressor of intestinal tumorigenesis in mice, which may increase susceptibility to Cr⁶⁺ carcinogenesis (Than et al., 2016). Thus carriers of the mutated CFTR allele, such as those with cystic fibrosis, would be more prone to Cr⁶⁺-induced cancers as well as exacerbation of their pulmonary disease (EPA, 2022a). However, this has not been adequately studied.

Despite the low confidence in the rodent study finding potential developmental effects of reduced weight gain in offspring, the Integrated Risk Information System (IRIS) RfD for developmental effects of Cr⁶⁺ in humans indicates that in utero exposures during pregnancy may pose a risk to children (EPA, 2022a). A small body of literature has shown that exposure to high levels of Cr during pregnancy may result in adverse outcomes, such as low birth weight, preterm birth, and developmental abnormalities. Maternal exposure to high levels of Cr⁶⁺ during pregnancy have been associated with increased risk of delivering low birth weight infants, particularly for female infants (Xia et al., 2016; Peng et al., 2018). Elevated levels of placental Cr have also been related to reduced birth length and somewhat counterintuitively to increased gestational age (Freire et al., 2019). Maternal Cr levels have also been associated with increased risk of delivering preterm infants, especially among male infants (Pan et al., 2017b). However, a systematic review of the literature concluded that any health effects from environmental Cr exposure in pregnancy are small (McDermott et al., 2015).

There is limited information available on the toxicity of Cr in children across different stages of development, from the ages of 2 to 14. Data have been obtained from case studies of children who have ingested lethal doses of Cr⁶⁺-containing compounds, which suggests that children are more vulnerable to lower lethal doses of Cr ingestion compared to adults (Clochesy, 1984; Ellis et al., 1982; Iserson et al., 1983; Kaufman et al., 1970). A systematic review found no significant differences in hair, serum, and urine Cr levels between children with cognitive deficits and healthy control children when all study data were pooled in a meta-analysis but did show a marginally significant association with serum and urine Cr levels cross-sectionally (Islam et al., 2022).

The available evidence on the sex-specific effects of chromium toxicity is limited and inconclusive. Some studies suggest that men may be more susceptible to the toxic effects of Cr⁶⁺ than women, although occupational exposure is dominated by industries in which male employees greatly outnumber women, making sex-specific inferences difficult (EPA, 2022a).

MANGANESE

Mn is one of about 30 elements considered essential to life. Mn serves as a cofactor for multiple enzymes, and for humans, diet is the main source of its intake. Adequate daily intake levels for Mn change with age, and exposures to elevated amounts of this metal can be toxic, primarily to the brain.

Mn is an indispensable trace element that is essential for prokaryote and eukaryote survival (Himeno et al., 2019). It has essential biological functions as an enzyme cofactor (e.g., glutamine synthetase, signaling kinases, glycosylation enzymes, etc.), in immune regulation, and for oxidative stress response (Horning et al., 2015; Wang et al., 2018). Mn is needed for neurotransmitter metabolism and function, with key connections to tyrosine, an amino acid that serves as a precursor to several important neurotransmitters, including dopamine, norepinephrine, and epinephrine. These neurotransmitters are involved in regulating mood, stress response, and energy levels, and their neurons are found in the basal ganglia and ventral striatum in high concentrations (Balachandran et al., 2020; Sanders et al., 2015). Mn activates the enzyme phenylalanine hydroxylase that converts phenylalanine into tyrosine. Therefore, Mn is indirectly involved in the production of these neurotransmitters. Interestingly, because Mn is a nutrient, effects can be nonlinear, with toxic effects seen at high and low ends of the dose response curve (Claus Henn et al., 2010; Roels et al., 2012).

The majority of absorbed Mn via oral intake is derived from water consumption and Mn-enriched foods, such as nuts, soybeans, and rice. Occupational settings may serve as an alternative yet significant exposure route. Orally ingested Mn is tightly regulated with approximately 5 percent of Mn entering the adult circulation, and 0.1 percent of the Mn in blood is found in the brain, which is the target organ for toxicity (Chen et al., 2018). Brain Mn is predominantly concentrated in the globus pallidus, and other basal ganglia nuclei (striatum and substantia nigra), as well as the frontal cortex (Balachandran et al., 2020).

Manganese Forms

Mn exists in multiple inorganic and organic species. Common inorganic species include oxides (dioxide, MnO₂, and tetraoxide, Mn₃O₄), chloride (MnCl₂), sulfate (MnSO₄), manganese phosphate (MnPO₄), carbonate (MnCO₃), and silicate (MnSiO₃), to name a few (Martinez-Finley et al., 2013). Methylcyclopentadienyl Mn tricarbonyl (MMT), as organic Mn-containing gasoline additive, and Maneb and Mancozeb as pesticides/fungicides have been considered as potential health hazards upon overexposure (Michalke and Fernsebner, 2014). In addition to anthropogenic sources of Mn into the environment, natural sources including soil erosion contribute to Mn emissions (Martinez-Finley et al., 2013).

Bioavailability/Bioaccessibility/Absorption

The Manganese REACH Consortium reported on read-across as a strategy to better understand data gaps for the bioavailability of a set of manganese-based substances of most concern as related to industrial

exposures (Read-across Approach: An Application of Group Concept under the Registration, Evaluation and Authorisation of Chemicals, Regulation (EC) No. 1907/2006 (REACH)). Using data for each on (1) physico-chemical properties; (2) toxicokinetic (TK) behavior; (3) for insoluble substances, a bioavailability (test on artificial body fluids) for human health assessments and transformation dissolution for environmental assessment; (4) other in vitro tests (such as skin and eye irritation), used to confirm categories from a health effect standpoint, and acute daphnia, used to confirm categories from an environmental perspective; and (5) in vivo acute oral toxicity, the authors subdivided the substances into seven sub-categories, with Category 1, and specifically MnCl₂, representing the “worst case scenario” toxic potential (Categories 1: Mn(NO₃)₂, MnSO₄, MnCl₂; 2: MnCO₃ and MnO; 3: Mn₃O₄; 4: Manganese ore (reduced); slags FeMn and slags SiMn; 5: MnS; 6: Mn; and 7: MnO₂).

Oral Bioavailability

Mn average concentrations in the cerebral cortex of groups of mice dosed with MnO₂ and MnCO₃ differed from controls (un-dosed), and average concentrations in those groups also differed from the MnCl₂ and Mn(CH₃COO)₂ dosed groups (Komura and Sakamoto, 1993).

EPA has concluded through a series of studies that the bioavailability of ingested Mn from water is essentially the same as the bioavailability of Mn in food (see IRIS: EPA, 1995). Rather than the actual medium of exposure, the total diet was invoked as the determining factor for the absorption of Mn from the GI tract. Furthermore, it was determined that relative bioavailability (RBA) of food-derived Mn was 0.7-fold compared with that from drinking water, and statistically indistinguishable.

Mn absorption from the GI tract can be affected by several factors, including the intake of dietary fiber, oxalic acids, tannins, and phytic acids, all of which tend to decrease Mn absorption (Gibson, 1994; EPA, 2003a). Other factors that influence Mn uptake are iron (Fe) storage status and calcium (Ca) or phosphorus intake (Greger et al., 1990; IOM, 2001; Meltzer et al., 2010). Competition between Mn and Fe at the GI tract has been established, given that transporters such as the divalent metal transporter 1 (DMT1) are nonspecific and able to transport multiple metals (Fitsanakis et al., 2010). Finally, it should be noted that Mn absorption is lower in males versus females (IOM, 2001).

Notably, Mn and Fe compete for several transporters (Fitsanakis et al., 2010). The authors derived RBAs of 48.4 percent (95 percent confidence interval: 39.3–61.1 percent) and 14.0 percent for the liver and lung, respectively. No RBA could be calculated for Mn in the striatum with either chow or EAF slag, with the authors suggesting no systemic delivery of Mn to the striatum. The absence of its accumulation in this brain region may reflect the short exposure duration (up to 15 days) and/or inconsistencies in the isolation of this relatively small brain region, ensuring homogeneity of sampling.

In addition, when addressing bioaccessibility and bioavailability, it should be considered that only a small percentage of dietary Mn is absorbed. Its absorption from a test meal containing 1 mg Mn is in the order 1.35 ± 0.51 and 3.55 ± 2.11 percent for men and women, respectively (Finley et al., 1994). Using radiolabeled Mn (⁵⁴Mn) in a test meal containing 0.3 to 0.34 mg of Mn, Davidsson et al. (1988) estimated Mn retention at 5.0 ± 3.1 percent 10 days after administration in young adult women, with turnover of orally administered ⁵⁴Mn being much higher after oral versus intravenous administration (Davidsson et al., 1989; Sandstrom et al., 1986). In addition, Mn absorption after 30 weeks of supplementation has been shown to be 30 to 50 percent lower than noted in nonsupplemented subjects (Sandstrom et al., 1990). Absorbed Mn is rapidly excreted via the bile into the GI tract (Britton and Cotzias, 1966; Davis et al., 1993).

Bioaccessibility

To assess the rate of ingestion of elements from food, it is critical to consider bioaccessibility, referring to the accessible concentration of an element (O’Neal and Zheng, 2015) or the amount that is released from the food matrix in the GI tract and is available for absorption. Studies have evaluated the bioaccessible and bioavailable fractions of dietary Mn.

Notably, studies performed with volunteers who received 9 mg to 15 mg of Mn daily showed no adverse health effects. The European Union Scientific Committee for the Food Industry (EFSA Panel, 2013) has established a safe daily intake of Mn for an adult human being of 10 mg. The Expert Group, based on the no adverse effect level value, estimated that the dose of manganese in dietary supplements of 4 mg/day is safe for the general population, and the total daily manganese consumption by an adult should not exceed 12.2 mg (European Parliament²; Expert Group on Vitamins and Minerals, 2003), given an estimated daily food intake average of 8.2 mg. The Institute of Medicine (IOM) has established a safe level of Mn intake (UL, upper limit) of 11 mg/day (Greger, 1998; IOM, 2001).

Dietary Standards

Dietary standards and recommended intakes for daily Mn consumption are not uniform for health organizations and government agencies throughout the globe. The European Food Safety Authority of the European Union proposed an adequate intake (AI) for adults, including pregnant and lactating women, of 3 mg/day; for children, the AI of Mn also varies with age, where for infants 7–11 months the AI required is in the range of 0.02–0.5 mg/day (EFSA Panel, 2013). Since insufficient data exist to set an estimated average requirement and a recommended dietary allowance, IOM developed an AI for Mn estimated in healthy individuals based on data from the Food and Drug Administration’s Total Diet Study (1991–1997). The AI for adults (>19 years) is 1.8 mg/day for women and 2.3 mg/day for men, reflecting the lesser absorption of Mn in males versus females. The AI for children varies with age (< 6 months, 0.003 mg/day; 7–12 months, 0.6 μg/day; 1–3 years, 1.2 mg/day; 4–8 years, 1.5 mg/day; girls 9–18 years, 1.6 mg/day; boys 9–13 years, 1.9 mg/day, and 14–18 years, 2.2 mg/day) (IOM, 2001).

Inhalation

The bioavailability of various inhaled forms of Mn is an important factor for consideration. For example, even at almost identical airborne total dust concentrations (geometric means of 0.94 and 0.95 mg/m³, respectively), the bioavailability of various forms of Mn differs (Roels et al., 1992). Geometric mean blood and urinary Mn levels of workers exposed only to MnO₂ were lower (MnB: 0.81 ug/dL; MnU: 0.84 ug/g creatinine) than those of workers exposed to mixed oxides and salts (MnB: 1.22 ug/dL; MnU: 1.59 ug/g creatinine). Yet, no difference between the absorption of 1 um particles of MnCl₂ and manganese sesquioxide (Mn₂O₃) in healthy adults was noted by Mena et al. (1969). In turn, following intratracheal instillation of MnCl₂ and Mn₃O₄ in rats, Drown et al. (1986) found that the soluble chloride cleared four times faster than the insoluble oxide from the respiratory tract. Nonetheless, despite this initial variance in clearance, the amounts of labeled Mn in the respiratory tract after 2 weeks were essentially identical for these two compounds.

RfD: EPA set the RfD value for Mn at 0.14 mg/kg-day from all oral exposures. The value is based on a composite of data from several studies: Freeland-Graves et al. (1987), NRC (1989), and WHO (1973). The agency suggested a modifying factor of 1 for ingested Mn in food and a modifying factor of 3 for ingested Mn in water or soil.³ Also see ATSDR (2012b) for a discussion of the derivation of the RfD.

RfC: The EPA chronic inhalation RfC value for respirable Mn is 5 × 10^-5 mg/m³. This value is derived from the low adverse effect level of 0.15 mg/m³ reported in a study (Roels et al., 1992) of battery workers exposed to manganese dioxide. In addition, EPA notes that given insufficient information by which to determine the relative toxicities of different forms of Mn, for the purpose of deriving an RfC for Mn, no distinction has been made between various Mn compounds.⁴

___________________

² See https://eur-lex.europa.eu/eli/reg/2011/1169/2018-01-01

³ See https://cfpub.epa.gov/ncea/iris2/chemicalLanding.cfm?substance_nmbr=373.

⁴ Ibid.

Toxicity Endpoints

Mn toxicity most commonly occurs from occupational exposures in welders, miners, and steel workers upon long-term high concentrations of airborne particulate matter (PM) containing this metal. As recently summarized by Taylor et al. (2020), Mn toxicity in adults is primarily manifested as a Parkinsonian-like movement disorder, secondary to its accumulation in the basal ganglia. Epidemiological studies are consistent with environmental exposure to elevated Mn (e.g., via drinking water), especially in children and adolescents, resulting in fine motor, emotional, cognitive, and intellectual deficits. As Mn is predominantly excreted in bile, patients with liver dysfunction (e.g., due to cirrhosis) may develop Mn neurotoxicity even in the absence of elevated exposure to Mn.

In addition, when inhaled, Mn can lead to inflammation in the lungs and respiratory symptoms, including cough, bronchitis, pneumonitis, and impaired pulmonary function (Roels et al., 1999).

Mn is not classifiable as to human carcinogenicity, as existing studies are inadequate to assess its carcinogenicity. Quantitative estimates of carcinogenic risk from either oral or inhalation exposures are not available.⁵

Mn toxicity can have various effects on the body. Mn metabolism and physiology overlaps with iron metabolism. Iron deficiency can enhance the absorption and body burden of Mn. Because iron deficiency is more common in women, this may lead to increased body burdens on average in women of childbearing age (Bjorklund et al., 2017). Mnmay dysregulate menstrual cycles, which could affect fertility. Results are mixed for sexual dimorphic effects, and while some studies suggest that men may be more susceptible to the toxic effects of Mn than women with regard to neurotoxicity, others suggest women may be more susceptible, for example with regard to motor effects (Chiu et al., 2017).

The primary toxicity for excess Mn is due to oxidative stress. Mn is a transition element and can catalyze reactions that produce free radicals—typically via the Fenton reaction, a common chemical reaction involving iron (Fe) that oxidizes dopamine and water to create free radicals (Wright and Baccarelli, 2007). Both Fe and Mn are transition elements that can serve as redox catalysts, and both are normally sequestered by cellular proteins to limit the amount of reactive free metal.

In the presence of excess brain Mn, excess Mn may catalyze the Fenton reaction that can accelerate the accumulation of reactive oxygen species locally. Typically, Fe catalyzes this reaction, but Mn can replace iron in many reactions. For these reasons both neuroprotective and neurotoxic effects can be seen for Mn, and both high and low levels of Mn can be associated with toxic effects. The frontal lobes, basal ganglia, and ventral striatum are particularly high in Mn concentration, suggesting that Mn is key to the functions of these regions which are related to attention, decision making, reward seeking, and motor functions.

Hepatic Toxicity as a Modifier of Mn Neurotoxicity

Mn is processed and excreted in the liver and is found in high concentration in hepatic tissues. After absorption in the intestinal tract, 100 percent of oral Mn will circulate through the liver prior to any other organ. Some studies suggest that liver toxicity is associated with excess Mn, particularly when in combination with another liver toxicant, such as alcohol. Cross-sectional studies have linked blood Mn with steatosis (i.e., fatty liver) (Spaur et al., 2022). The relationship between Mn and liver toxicity is complex because Mn is primarily excreted in the liver. Hence, primary liver disease can reduce Mn excretion and elevate Mn body burden, secondarily causing neurotoxicity. Studies on patients with liver cirrhosis identified an association between decreased Mn excretion and the risk of developing Mn-induced Parkinsonism in the absence of elevated Mn exposure (Gurol et al., 2022). Because of its key role as a cofactor for the antioxidant enzyme superoxide dismutase, some investigators have suggested that mildly elevated blood Mn in the setting of nonalcoholic fatty liver disease may be protective (Tijani et al., 2022). Obviously, given the ability of free Mn to generate free radicals, there may be cases in which Mn accelerates

___________________

⁵ See https://cfpub.epa.gov/ncea/iris2/chemicallanding.cfm?substance_nmbr=373.

liver toxicity as well (Liu et al., 2021b). Several experimental studies have shown this as Mn can potentiate inflammation. Mn will accumulate in hepatic mitochondria, and mitochondrial dysfunction in hepatocytes including damage of mitochondrial DNA has been observed in animal studies (Gandhi et al., 2022; Jiao et al., 2008). Finally, it should be noted that patients with liver toxicity failure receive total parenteral nutrition, which may be a source of intravenous Mn that can exacerbate the toxicity of Mn.

Reproductive Effects

The understanding of the effects of Mn exposure during the preconception period is also currently limited. However, one epidemiological study investigating Mn exposure during preconception revealed an association between preconception iron deficiency and high concentrations of Mn (Davies et al., 2021). Studies have found significant positive associations between the concentrations of Mn in maternal blood and/or umbilical cord blood and the incidence of pre-eclampsia or gestational hypertension (Vigeh et al., 2013, 2006).

U-shaped and/or negative associations have been established between Mn exposure and various measures at birth including size, lower birth weight, growth restriction, and preterm birth (Guan et al., 2014; Ashrap et al., 2020; Chen et al., 2014; Eum et al., 2014; Zota et al., 2009; Tsai et al., 2015; Vigeh et al., 2008; Yamamoto et al., 2019).

Neurotoxic Effects of Mn Exposure in Children

During the past 15 years there has been a growing literature on the neurotoxicity of Mn exposure in children. Several studies and reviews have concluded that there exists a U-shaped or negative association between postnatal Mn exposure and cognitive impairment (Vollet et al., 2016; Sanders et al., 2015). Studies have used blood Mn as the primary exposure index, even though it is not a strong correlate of body burden (Shilnikova et al., 2022), perhaps because it is tightly regulated by the body within a relatively narrow blood concentration range. Urine may not be elevated in the setting of high body burden because the liver is the primary organ for excreting Mn. Teeth and hair have also been used and may be better exposure biomarkers as opposed to a biomarker of effect.

Unlike lead or mercury, there are biological reasons to store Mn, and Mn is physiologically stored in many tissues (e.g., liver and bone) and cell compartments (mitochondria) where it serves several critical functions. These stores also help maintain blood concentrations in a physiologic range. Blood Mn may not correlate well with body Mn stores, but it may reflect biologically active Mn. For example, while not a strong correlate of body Mn stores, several studies have shown that blood Mn levels in pregnancy predict infant development and child development (de Water et al., 2022, 2018; Oppenheimer et al., 2022; Wright et al., 2006), including attention and other aspects of executive function. Some studies in postnatal infants/children have shown nonlinear effects as Mn is an essential nutrient critical for brain development (Claus Henn et al., 2010). While blood Mn may not be a strong correlate of body burden with environmental sources, it may still represent the biologically functional fraction of Mn in the body, explaining its ability to predict neurotoxicity. Nonlinear correlations with diet/environmental sources may exist as well.

Studies using tooth Mn have been consistent with regard to the benefits and potential toxicity of Mn in children and generally show a complex pattern that is dependent on life stage. Higher tooth Mn levels in prenatal dentine levels are associated with better performance on neurocognitive testing that switches to an inverse relationship for higher tooth Mn in childhood dentine levels. While this is in contrast to some studies that used maternal blood Mn in pregnancy as a biomarker, it should be noted that Mn is actively transported across the placenta with blood Mn levels 2–4 times higher in the fetus (Guan et al., 2014; Munoz-Rocha et al., 2018; Takser et al., 2004). Unlike lead, which is not actively transported, maternal blood Mn in pregnancy may have little relationship to fetal exposure. It may be that the association between higher blood Mn in pregnant women and lower cognitive test scores in children is related to this active transport process and is not reflective of Mn exposure in pregnancy—explaining the results using teeth,

which are a direct measure of fetal exposure, unlike maternal blood Mn levels, which are indirect and confounded by active Mn transport across the placenta.

Mn exposure during childhood (ages 2 to 14) has been linked to adverse neurocognitive outcomes, reviewed in Iyare (2019). For example, Mn has been negatively associated with cognitive development and has shown a biphasic dose–response relationship in several studies (Vollet et al., 2016; Rodriguez-Barranco et al., 2013; Sanders et al., 2015). Studies have shown that children exposed to high levels of Mn from environmental sources, such as drinking water and air pollution, are at an increased risk of developing intellectual disabilities, attention deficit hyperactivity disorder, and other neurodevelopmental disorders (Claus Henn et al., 2018; de Water et al., 2019; Menezes-Filho et al., 2009; Schildroth et al., 2022). Shaffer et al. (2023) point out the limited information on comparative susceptibility of children and adults to inhaled Mn, reporting that children may be 0.37 to 2.03 times as susceptible to the neurotoxic effects of Mn as adults.

Adult Mn Neurotoxicity

In adults, blood Mn levels can vary depending on dietary intake and exposure to environmental or occupational sources. Blood levels in adults are approximately half the levels found in children. The body tightly regulates Mn levels, and excess Mn is eliminated through the liver primarily, but renal clearance does occur as a secondary pathway. Most of the literature on adult Mn toxicity comes from occupational exposures. The classic manganism case consists of chronic long-term exposure at work. An example might be exposure of a welder to airborne PM containing Mn, which bypasses the liver and increases the dose that can be transported to the brain. With long-term deposition of Mn in the basal ganglia and ventral striatum—areas high in dopaminergic neurons—Parkinsonian symptoms arise. These consist of reduced response speed, irritability, mood changes, and compulsive behaviors, sometimes resembling schizophrenia. With prolonged exposure, symptoms accelerate and begin to resemble those of idiopathic Parkinson’s disease, with which it is often misdiagnosed (Martins et al., 2019). Because of the similarity to Parkinson’s disease, environmental Mn exposure has been studied as a risk factor for idiopathic Parkinson’s disease. Interestingly, welding and an overall increased risk of Parkinson’s was not seen in a review of the literature (Mortimer et al., 2012). Regardless of whether Mn is a cause of Parkinson’s disease, its neurotoxicity is clear and overlaps with the pathophysiology of Parkinson’s disease (Dlamini et al., 2020). While distinct differences exist between manganism and idiopathic Parkinson’s disease, they share multiple analogous symptoms.

Studies that evaluate associations between Mn exposure and adverse health outcomes in elderly populations generally establish increased prevalence of Parkinson’s disease or Parkinson’s-like behaviors (Lucchini et al., 2014, 2007; Willis et al., 2010). Excessive occupational Mn exposure is neurotoxic, but there is unclear evidence whether it causes idiopathic Parkinson’s disease. It may be that the primary etiology of Parkinson’s disease is environmental factors that occur at younger pre-occupational ages, and occupational exposures may only have additive impacts on top of early life risk factors (Dlamini et al., 2020). Age-related changes in the body’s ability to process and eliminate Mn, such as the increased risk of liver disease with aging, can increase the risk of elevated exposure and neurotoxic effects.

Genetics of Mn

Because Mn is an essential nutrient under tight metabolic regulation, several genes are key to its metabolism. Genetic susceptibility to Mn-induced toxicity clearly occurs. Mutations in SLC39A8, SLC30A10, and SLC39A14 can also impact Mn homeostasis and lead to human disease.

Notably, the past few years have seen growing interest in Mn genetics driven by the discovery of hereditary Mn neurotoxicity caused by mutations in SLC30A10 or SLC39A14 (Rodichkin and Guilarte, 2022; Tinkov et al., 2021; Zhang et al., 2022; Zogzas and Mukhopadhyay, 2017). Both genes play key roles in Mn excretion. Likewise, common polymorphisms in the Mn transporter genes SLC30A10 and SLC39A8 impact intracellular Mn levels and can drive neurotoxicity. Human carriers of the missense mutation

(A391T) in SLC39A8 show strong association with schizophrenia, which is characterized by reduced serum Mn, altered plasma glycosylation, and brain magnetic resonance imaging (MRI) changes. Wilcox et al. (2022) showed that age, sex, and the Huntington’s disease-genotype collectively affect Mn homeostasis, and that each of these variables affected both behavior and dopaminergic system function upon excessive dietary Mn intake.

Genetic variation in iron regulatory genes may to a lesser extent also influence Mn levels and toxicity. The HFE gene that regulates iron absorption also influences Mn blood levels. Whether this effect is direct or indirect via iron status is unclear. Mn transporters SLC39A14, SLC39A8, and SLC30A10 are now known to mediate systemic and brain Mn handling. Multiple metabolic pathways that could mediate Mn neurotoxicity are regulated by these genes, including oxidative stress, endoplasmic reticulum stress, apoptosis, neuroinflammation, cell signaling pathways, and interference with neurotransmitter metabolism (Tinkov et al., 2021). Whether there are common polymorphisms in these that would constitute increased vulnerabilities to Mn exposure in a substantial proportion of the general population is unknown at present.

Epidemiology

Since the Agency for Toxic Substance and Disease Registry’s (ATSDR’s) most recent toxicological profile of Mn (ATSDR, 2012b) reviewed the scientific literature for carcinogenic and noncarcinogenic health outcomes, a substantial amount of research on Mn effects has been undertaken. Examples of meta-analyses that focused on various specific health outcomes include the following:

• Gu et al. (2022): relationship between exposure to certain metals and hyperuricemia
• Lambert et al. (2021): dietary intake of adults with inflammatory bowel disease, including macronutrients, micronutrients, and food group data
• Mortimer et al. (2012): associations of welding and Mn exposure with Parkinson’s disease (mentioned previously)
• Sharma et al. (2022): association of serum concentrations of 11 trace elements with polycystic ovary syndrome pathogenesis
• Spritzer et al. (2017): trace element concentrations in the blood of women with polycystic ovary syndrome
• Wong et al. (2022): dietary and environmental Mn exposure and metabolic syndrome

The considerable body of research on relationships between Mn exposure and health outcomes points to the need for an updating of previous assessments by ATSDR and EPA.

Pre-existing Disease

A new area of research on the effects of environmental exposures on pre-existing disease is gaining traction in environmental health. The vast majority of environmental health studies have focused on the role of chemical exposure as causes of disease (Niedzwiecki et al., 2019; Vermeulen et al., 2020). A small twist on this approach is to study the impact of exposure on populations with pre-existing disease. For example, if Mn is neurotoxic, regardless of whether it is a cause of Alzheimer’s disease or Parkinson’s disease, it may exacerbate the symptoms of those disorders (e.g., memory loss, impaired motor function). In this way patients with neurodegenerative disorders represent a vulnerable population to Mn exposure. Consider the potential for significant toxic effects if Mn exposure were to be overlaid upon already damaged neurons from the pre-existing disease. In this scenario, exposure need not have caused the disease; instead, it is studied as a modifier of disease severity or the rapidity of disease progression. Given that the prevalence of neurodegenerative diseases is rising in the United States, the potential impact of exposures on these patients deserves serious study as they may represent a potentially highly vulnerable subpopulation. At present, the literature is very scant in this regard, with the vast majority of research being done to determine

if Mn exposure causes neurodegeneration. It is a very small step to consider what would happen if Mn exposure occurs in the presence of neurodegeneration, and research in this area is sorely needed.

SUMMARY

Consistent with its statement of task, the committee focused on Cr and Mn as examples of COPCs likely to be present in EAF slag at concentrations relevant to human health risk assessment. This focus is intended to support a more comprehensive, site-specific evaluation that would include a broader list of COPCs identified in Chapter 7. Both cancer and noncancer toxicological endpoints are relevant to assessments of EAF slag risks.

Chromium

Estimates of exposure and risk are highly sensitive to assumptions regarding the chemical form of Cr present in EAF slag, and changes in Cr speciation that may occur due to leaching from slag or in dusts generated from slag. There is a significant lack of site-specific measurements of Cr⁶⁺ at sites of slag production or application, which leads to uncertainty whether there is sufficient oral or inhaled bioaccessibility or bioavailability of Cr⁶⁺ to promote disease. A major challenge for regulating hazards like Cr⁶⁺ from EAFs is the need to speciate when quantifying Cr compounds leaching from EAFs or in dusts generated from EAFs as only the Cr⁶⁺ poses a hazard.

Cancer is the primary toxicity of Cr⁶⁺. There are very few studies of environmental Cr⁶⁺ exposures that address noncancer endpoints. In general, the oral slope factors, RfDs, and RfCs are conservative, protective estimates due to application of appropriate uncertainty factors and age adjustment factors in their derivation. They account reasonably well for potential sex differences in susceptibility to potential cancers and noncancer diseases from oral ingestion of Cr⁶⁺. There is some controversy that they may be too conservative based on the limitation of relying primarily on high-dose rodent studies to determine dose–response relationships that may not reflect the toxicokinetics of environmental human exposures. The bioavailability of Cr⁶⁺ is greatly limited by the protective reduction of Cr⁶⁺ to Cr³⁺ in the fluids of the GI and lung airways, and it is not clear whether there is sufficient accessible Cr⁶⁺ in slag or slag dust to overwhelm this reductive capacity. Unless this reductive capacity is overwhelmed, there is minimal risk of Cr⁶⁺-promoted disease burden (De Flora. 2000; Proctor et al., 2002).

There are clear epidemiologic studies, primarily occupational, that demonstrate increased risks of cancer, particularly lung cancer and cancers of the upper airway, from Cr⁶⁺ exposure, suggesting that ambient exposures are common sources. However, data are lacking on susceptible human populations exposed to environmental Cr⁶⁺. Again, there is a significant gap in demonstrating whether there is sufficient bioaccessible respirable Cr⁶⁺ in slag.

Based on limited data, pregnant women, young children, and elderly individuals would likely be more vulnerable to Cr⁶⁺ exposure from slag. Given that the airways and GI tract are the primary targets of Cr⁶⁺-promoted cancers and noncancer disease, Cr⁶⁺ exposure might exacerbate health conditions in individuals with pre-existing pathologies or disease burdens in the GI tract, liver, lungs, and blood. In addition, individuals with particular genetic susceptibilities can increase vulnerability to Cr⁶⁺ exposure. There are insufficient epidemiological studies of the health effects of environmental exposures to Cr⁶⁺ to exclude the risk of noncancer disease endpoints in susceptible populations at the important life stages of development and advanced age.

Manganese

Because Mn is an essential element, and average daily intake is important to a healthy diet, toxicity from environmental exposure is typically evaluated for conditions when exposures exceed dietary intake. The regulatory toxicity values (chronic oral RfD and RfC) developed by EPA were most recently updated

in 2012. The lack of epidemiological studies in susceptible populations (as noted previously for Cr) is also inherent to Mn-exposed populations.

The primary toxicity of Mn is neurological, with some evidence of other health effects such as liver toxicity. Exposures during pregnancy and early childhood have been associated with lower neurodevelopmental test scores including Bayley Scales of Infant Development, tests of spatial memory, tests of motor function (i.e., postural sway), and depressive symptoms, among other effects. In general, the majority of studies are in adults and are occupational, making direct comparisons with children’s studies complicated, as children exposed to environmental levels may be more susceptible due to the vulnerability inherent in a developing child. Many growth and developmental processes highly operant in children are not operant in adults, which is why lower doses of toxins can cause health effects.

Several studies have also shown genetic predisposition to Mn-induced neurotoxicity. For example, human carriers of the missense mutation (A391T) in SLC39A8 show strong association with schizophrenia, which is characterized by reduced serum Mn, altered plasma glycosylation, and brain MRI changes. Genetic susceptibility should be considered in epidemiological studies when addressing the role in Mn metabolism and toxicity. Several genotypes of the Mn transporter genes, such as SLC39A8 and SLC30A10, have been repeatedly shown to regulate Mn homeostasis and susceptibility to Mn neurotoxicity. Therefore, the association between these (and other) common variants of these genes and intracellular Mn concentrations offers one of the strongest gene–environment interactions reported thus far, yet systematic epidemiological studies have yet to be undertaken.

Similar to Cr⁶⁺, pregnant women, early childhood, and elderly individuals would likely be more vulnerable to Mn exposure from slag. Those with chronic neurological diseases (such as autism and Alzheimer’s disease), pre-existing liver disease, or particular genetic susceptibilities can also increase their vulnerability to Mn exposures.