 |
INTRODUCTION |
In the past, the establishment of nutritional and toxicological standards for minerals has not been coordinated. This has led to inconsistencies between the nutritional standards [e.g., the Recommended Dietary Allowance (RDA)3 and Estimated Safe and Adequate Daily Dietary Intake (ESADDI)] and the toxicological standards [e.g., the reference dose (RfD) and lowest-observable-adverse-effect level (LOAEL)].
In 1989, the Food and Nutrition Board, in the absence of good biomarkers (indicators of nutritional or toxicological status in regard to manganese), based the ESADDI for manganese on typical intakes of manganese and on data from short-term balance studies (NRC 1989). Similarly the Environmental Protection Agency (EPA) used estimates of manganese levels in typical Western and vegetarian diets to calculate the RfD for manganese in food (Velazquez and Du 1994
). Thus the ESSADDI is 2-5 mg Mn/d and the RfD for manganese in food is 0.14 mg Mn/(kg·d) or 10 mg Mn/d for a 70-kg individual.
EPA officials thought that a separate standard was required for manganese in drinking water because of the potentially greater bioavailability of manganese in water than in food (Velazquez and Du 1994). Data from an epidemiological study by Kondakis et al. (1989) were used. Kondakis et al. (1989) observed more neurological symptoms among individuals > 50 y of age living in areas in Greece in which the drinking water contained 1.8-2.3 mg Mn/L than among individuals living in areas in which the drinking water contained <0.015 mg Mn/L. Assuming that a 70-kg individual would consume 2 L of water daily, the EPA estimated the LOAEL for manganese in water to be 4.2 mg Mn/d or 0.06 mg Mn/(kg·d) (Velazquez and Du 1994).
Thus these standards "suggest" on the basis of very limited data that individuals consuming the amount of manganese recommended by the ESADDI (5 mg Mn/d) could consume "excessive" amounts of manganese (>4.2 mg Mn/d) if much of the manganese came from water. Conflicting standards of this sort lead to either alarm and/or cynicism among the public and scientists. Obviously, policy decisions such as toxic clean-up efforts and food fortification should be based on stronger scientific evidence.
The rest of this review will examine the types of data that are available for use in calculating nutritional or toxicological standards. The limitations of the data will be noted.
| |
MANGANESE EXPOSURE |
One reason for the difficulty in recommending optimal intakes of manganese is that scientists really are not sure of typical manganese intakes. Daily food composites prepared on the basis of the FDA's Total Diet Study menus contained on average 2.2 mg Mn for women and 2.7 mg Mn for men (Pennington et al. 1989). The average intakes of adults eating Western-type and vegetarian diets in various surveys ranged from 0.7 to 10.9 mg Mn/d (Freeland-Graves 1994, Gibson 1994).
These variations in manganese intakes reflect variations in food choices. Pennington and Young (1991) calculated that grain products contributed 37%, beverages (particularly tea) contributed 20% and vegetables contributed 18% of the manganese in the FDA's Total Diet Study menu for 30- to 35-y-old males.
| |
BIOMARKERS FOR MANGANESE EXPOSURE |
Another limitation faced by experts trying to establish nutritional and toxicological standards for manganese is that the levels of oral intake of manganese associated with adverse effects (both deficient and toxic) are debatable. This reflects the scarcity of specific biomarkers for manganese.
Manganese balance and excretion.
In general, data from balance studies do not provide useful biomarkers of manganese exposure. Manganese balances during short- and moderate-term balance studies do not appear to be proportional to manganese intakes. Subjects in carefully controlled studies achieved positive balances in regard to manganese when fed diets containing 2.55-15.0 mg Mn/d (Freeland-Graves et al. 1988, Friedman et al. 1987, Greger et al. 1978b, Schwartz et al. 1986) and achieved negative balances in regard to manganese when fed diets containing 0.011-13.9 mg Mn/d (Freeland-Graves et al. 1988, Friedman et al. 1987, Greger et al. 1978a, Greger and Baier 1983, Schwartz et al. 1986).
The wide range of manganese intakes at which positive manganese balances were achieved reflects a number of factors. Past manganese intakes of subjects and contents of iron, calcium, phosphorus and phytate in the diet have all been found to affect manganese retention (Davis et al. 1992, Freeland-Graves 1994, Keen and Zidenberg-Cherr 1996, Lee and Johnson 1988, Schwartz et al. 1986, Wedekind et al. 1991).
Moreover, traditional balance techniques are inadequate for assessing manganese absorption because absorbed manganese is very quickly (within minutes) secreted into the gut as bile (Davidsson et al. 1989, Davis et al. 1993b, Lee and Johnson 1988). Hence separation of unabsorbed manganese from endogenous gut losses of manganese is difficult. However, Davidsson et al. (1989), on the basis of Mn-54 retention from meals, estimated that adult humans absorbed 5.9 ± 4.8% (mean ± SD) of ingested manganese. Davis et al. (1993b) estimated that growing rats absorbed 8.7% of their manganese intake but lost 37% of the absorbed manganese through gut secretions, presumably bile.
Accordingly, biliary manganese secretion might be a biomarker of deficient or excessive intake of manganese. However, biliary manganese excretion was found to be roughly proportional to recent injected or oral doses of aluminum but not to body stores of manganese (Malecki et al. 1996, Thompson and Klaassen 1982).
The usefulness of urinary manganese as an indicator of oral manganese exposure is also doubtful. Friedman et al. (1987) showed that urinary manganese excretion by men tended to decline during severe manganese depletion and to increase during repletion. However, Davis and Greger (1992) could not demonstrate that women given 15 mg Mn/d during a 125-d supplementation trial excreted more manganese in urine than women consuming 1.7 mg Mn/d in food.
In general, the balance data demonstrate that the body is protected against manganese toxicity primarily by low absorption and/or rapid presystemic elimination of manganese by the liver. Accordingly, the potential of excess accumulation of manganese and toxic symptoms would be greater when manganese is delivered parenterally vs. orally or when liver function and biliary secretion are decreased.
Tissue manganese concentrations.
The biomarkers most used for manganese exposure in animal studies are tissue manganese concentrations (Brock et al. 1994, Davis et al. 1990, DeRosa et al. 1980, Hurley and Keen 1989, Keen and Zidenberg-Cherr 1996, Malecki and Greger 1995, Malecki et al. 1996, Paynter 1980). Malecki et al. (1994) noted a fairly uniform loss of 40-50% of the manganese in tissues of rats fed ~1 mg Mn/g diet for 2 mo.
Unfortunately, most tissues are inaccessible in humans without biopsies. Serum manganese is the most easily available biomarker of manganese status in humans. Davis and Greger (1992) found that women consuming 1.7 mg Mn/d had lower serum manganese concentrations than women ingesting 15 mg of supplemental manganese for more than 20 d (Fig. 1).
View larger version (21K):
[in this window]
[in a new window]
| Fig 1.
Changes in serum manganese concentrations in 47 women supplemented with 15 mg Mn/d ( ), 60 mg Fe/d ( ), 15 mg Mn and 60 mg Fe ( ), or placebo ( ). Values are mean ± SEM. *P < 0.05;  P < 0.01; and  P < 0.0001 compared with placebo at the same time point. Reprinted with permission from Davis and Greger (1992).
|
|
Hambidge et al. (1989) reported that plasma manganese levels were elevated in infants with choleostatic liver disease who were given supplemental manganese in total parental nutrition (TPN) solutions. Similarly, others have observed manganese accumulation in blood, brain and liver of patients provided with supplemental manganese in parenteral solutions (Fell et al. 1996, Mehta and Reilly 1990, Mirowitz and West- rich 1992, Ono et al. 1995) and in patients with impaired liver function or biliary secretion (Fell et al. 1996, Hauser et al. 1994, Mehta and Reilly 1990, Versieck et al. 1974).
Symptoms of manganese deficiency.
Most of the gross symptoms (e.g., growth retardation, dermatitis or reproductive failure) and metabolic changes (e.g., reduced HDL cholesterol levels or diabetic-like glucose tolerance curves) observed in severely manganese-deficient animals are not specific enough to use as biomarkers (Baly et al. 1990, Davis et al. 1990, Friedman et al. 1987, Keen and Zidenberg-Cherr 1996, Taylor et al. 1996). The exceptions are the gross skeletal changes that occur during manganese deficiency. They are characterized by enlarged joints, deformed legs with thickened and shortened long bones and lameness in pigs, ruminants and poultry (Hurley and Keen 1989).
Symptoms of manganese excess.
Typical symptoms (i.e., depressed appetite, growth depression, reproductive failure and anemia) of chronic oral exposure to excess manganese are nonspecific (Hurley and Keen 1989). The consequences of chronic inhalation of excess manganese, as occurs among manganese miners, are more specific and include a severe psychiatric disorder resembling schizophrenia and an irreversible neurological disorder resembling Parkinson's disease (Hurley and Keen 1989).
Biomarkers that can indicate excessive manganese exposure before permanent neurological symptoms appear must be identified. Neurological tests have been used in at least two epidemiological studies to gauge the toxicity of manganese in water (Kondakis et al. 1989, Vieregge et al. 1995). Other investigators have reported that altered responses to a battery of neurofunctional tests (e.g., postural instability when visual input is prevented, slowed finger tapping speed, increased reaction times or decreased hand steadiness) were useful in identifying excess manganese exposure among industrial workers who were still asymptomatic for the classic neurological symptoms of manganese toxicity (Chia et al. 1995, Iregren 1994, Mergler et al. 1994). The results of neurofunctional tests and biological indications of excess manganese exposure should be studied together in the future to assess their relative usefulness.
Magnetic resonance imaging (MRI) scans of brains may provide useful biomarkers for excess exposure to manganese. High intensity signals of several regions of the brain, but especially the globus pallidus, have been observed in T-1 weighted MRI scans of patients receiving manganese-supplemented TPN solutions (Fell et al. 1996, Mirowitz and Westrich 1992, Ono et al. 1995) and/or scans of patients who have compromised liver or biliary function (Fell et al. 1996, Hauser et al. 1994). The abnormal imaging was reduced after manganese was removed from the TPN solutions in at least two cases (Mirowitz and Westrich 1992, Ono et al. 1995). Nelson et al. (1993) observed similar high intensity signals from the globus pallidus of the brain of a man who had been an arc welder for 25 y.
Activity of manganese-activated enzymes and metalloenzymes.
Scientists have had difficulty identifying enzymes that could be useful biomarkers for several reasons. Many of the manganese-activated enzymes are activated by other ions, especially magnesium (Keen and Zidenberg-Cherr 1996). Most manganese-activated enzymes are not reduced during manganese deficiency. The exceptions are phosphoenol pyruvate carboxykinase and the glycosyltransferases and xylosyltransferases, which are important in proteoglycan synthesis during bone formation (Baley et al. 1985, Hurley and Keen 1989).
There are three primary manganese metalloenzymes. The activity of one, liver pyruvate carboxylase, is more sensitive to developmental changes than to manganese intake (Baly et al. 1985). The activity of the second, arginase, is depressed in livers of manganese-deficient rats (Paynter 1980). Moreover, manganese-deficient rats had not only depressed hepatic arginase activity but also depressed plasma urea and elevated plasma ammonia concentrations (Brock et al. 1994). Unfortunately, the usefulness of arginase as a biomarker in clinical situations is debatable because a variety of factors, including diabetes, liver disease and ingestion of high protein diets, can affect arginase activity (Morris 1992).
The third manganese metalloenzyme is mitochondrial or manganese-dependent superoxide dismutase (MnSOD). MnSOD activity is usually depressed in heart and often depressed in the livers of manganese-deficient animals (Davis et al. 1990 and 1992, DeRosa et al. 1980, Malecki et al. 1994, Malecki and Greger 1995, Zidenberg-Cherr et al. 1983).
However, little work had been done to assess MnSOD activity in humans. Davis and Greger (1992) demonstrated that lymphocyte MnSOD activity was elevated in women ingesting 15 mg of supplemental Mn/d for >90 d (Fig. 2). Unfortunately, MnSOD is not a perfect biomarker. MnSOD activity can also be affected by a variety of factors that induce oxidative stress, including ethanol (Keen et al. 1985), diets rich in polyunsaturated fatty acids (Davis et al. 1990, Phylactos et al. 1994), non-heme iron (Davis et al. 1993a) and strenuous physical exercise (Ohno et al. 1993).
View larger version (24K):
[in this window]
[in a new window]
| Fig 2.
Changes in lymphocyte manganese-dependent superoxide dismutase (MnSOD) activity in 47 women supplemented with 15 mg Mn/d (), 60 mg Fe/d (), 15 mg Mn and 60 mg Fe (), or placebo (). Values are means ± SEM. *P < 0.05 and P < 0.01 compared with placebo at the same time point. Reprinted with permission from Davis and Greger (1992).
|
|
The reduction of MnSOD activity in manganese deficiency is interesting because it can be related to symptoms. Malecki and Greger (1996) observed that manganese-deficient rats had more oxidative damage (as indicated by increased conjugated diene formation) in the mitochondrial membranes of their hearts than did control rats. The conjugated diene levels were inversely correlated (r = 
0.71, P < 0.005) to MnSOD activity in the heart.
| |
SUMMARY |
There are no perfect biomarkers for assessing manganese exposure. However, serum manganese concentrations in combination with lymphocyte MnSOD activity and perhaps blood arginase activity seem to be the best way to monitor ingestion of insufficient manganese intake. Serum manganese concentrations in combination with brain MRI scans and perhaps a battery of neurofunctional tests seem to be to the best ways to monitor excessive exposure to manganese. A combination of these biomarkers should be used to estimate standards such as the ESSADI and RfD for manganese.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Abstract
Introduction
Abstract
