Open Access

Commentary

Methods and Results Top

The data described in this article are the byproducts of an investigation we undertook to study the developmental neurotoxicity of mercury vapor in rats. We did not a priori plan the diet assay protocols reported here, and although limited, the results of these evaluations have significance that must be considered in both evaluating past studies and designing future ones. Because surprisingly little is known about the developmental effects of metallic mercury despite its lengthy history in toxicology and its recognized potency as a neurotoxicant (Clarkson 2002), we had planned to examine this aspect of it.

In experiment 1, female Long-Evans rats (Charles River, Wilmington, MA) were bred 3 weeks after receipt from the supplier and then exposed via inhalation to mercury vapor concentrations of 0, 30, 100, or 300 μg/m³ during gestational days (GD)6–20. The 0-ppm control group was held in a separate mercury-free chamber during exposures. The mercury vapor concentration within a chamber was monitored continuously by a continuous mercury vapor analyzer dual-beam ultraviolet photometer in standard flow configuration (model 791.741; EPM Environmental Products Manufacturing, Dalerstraat, the Netherlands), which was capable of measuring concentrations from 2 to 1,999 μg/m³ in air. Mercury in the blood served as a biomarker of exposure. A cold vapor atomic absorption procedure (Lapham et al. 1995; Magos and Clarkson 1972) was used to assay blood samples from the pregnant dams on GD18 and from the pups on postnatal day (PND)4 and PND18.

Control dam (Figure 1) and pup (Table 1) samples showed unexpected, relatively high levels of mercury (particularly as organic mercury). By analyzing the samples for the presence of inorganic mercury specifically, we could estimate the amount of organic mercury (i.e., total – inorganic). As shown in Figure 1 and Table 1, the blood values were predominantly of the organic form.

When we first detected the high levels of mercury in our control subjects, we immediately sought to evaluate, on a probing basis, potential sources. Our sampling procedures were designed and employed to prevent and mitigate recognized potential sources of contamination, as we have done in the past. We did not detect mercury in either the control chamber or the room housing the chambers; in either the atmosphere in the vivarium room assigned to the animals in the experiment or the bedding in the animal cages; in the breath of the investigators who pipetted the blood during the tail-nick procedure used with the dams; or in the heparin that was used for the collection procedures. We did not believe our mercury assay procedures were at fault because they are continually evaluated as part of an international mercury quality control program administered by the Centre de Toxicologie du Québec (Institute National de Santéé Publique, Sainte-Foy, Québec, Canada), which has run the Interlaboratory Comparison Program since 1979.

Together, these results led us to suspect the diet as the source of contamination. Purina Laboratory Rodent Diet 5001 (Scott’s Distribution, Hudson, NH), which has widespread use, was fed to the rats in this research. Sample pellets from the batch in use at that time were ground or milled and then analyzed (we used more than one procedure to systematically replicate our observations and convince ourselves that we had not introduced confounds). For the second procedure, we used a ball mill with zirconium pellets. Between samples, both were washed, and then the jar and Zr pellets were baked at 150°C for several hours to ensure the absence of mercury. Then the samples were individually ground for at least 48 hr. These analyses, as shown by the examples in Table 2, verified that the elevated mercury levels in our control dams and pups were due to the contaminated diet and that they reflected organic mercury.

The Purina 5001 diet is an open diet; that is, its ingredients are subject to change, depending on the source of the raw materials. Fish meal is one of the ingredients, and it is possible that methylmercury present in tuna scraps, for example, may have been the source of the fish meal used in the batch provided by the vivarium. The supplier gives the limit of detection as 0.02 ppm (20 ng/g), so the problem apparently escaped detection. Even so, such levels are excessively high for experiments on mercury, especially those focusing on low-level dose–response outcomes. We were unprepared for the results in the present study because, in an earlier methylmercury study with mice (Stern et al. 2001) also fed the Purina 5001 diet, we detected no mercury in control dams or pups.

To preclude contamination in further experiments, we contacted BioServ (Frenchtown, NJ), a supplier of laboratory animal feed, which recommended the synthetic AIN-93G diet. The protein in this diet is casein. BioServ provided samples of whole pellets as well as the casein incorporated into the diet. Table 3 shows the results of our analysis of the ground pellets and, independently, of the casein. Although the pellets contained mercury, it was 100% inorganic. To determine its effects on blood levels, we fed three females the AIN-93G diet and three males the Purina 5001 diet. We detected no mercury in blood samples from the females fed the AIN-93G diet, but we did find it in the males fed the Purina 5001 diet (28.24, 22.84, and 16.08 ng/mL). (Only total mercury was measured because the focus was on comparing mercury levels in the AIN-93G diet-fed subjects with those fed Purina 5001.) Because inorganic mercury is poorly absorbed after ingestion, these findings are not surprising. These results also confirmed that the rats did not carry a significant mercury burden when they were received from the supplier.

More recently, in our ongoing attempts to find a suitable, mercury-free diet, we analyzed samples of the Teklad 2018 diet (Harlan Teklad, Madison, WI), which does not contain fish meal. We ground four pellets in a mortar to obtain a fine powder, which was then digested with sulfuric acid. No mercury was detected.

Discussion Top

Figure 1 shows why the possibility of methylmercury contamination in laboratory animal diets cannot be ignored. The levels in control dams were close to the 58 ng/g determined by the National Academy of Sciences committee on methylmercury, on the basis of developmental neurotoxicity, as the benchmark dose lower bound for cord blood in human populations (National Research Council 2000). Although not measured here, we would certainly expect fetal levels in our rats to be even higher (Watanabe et al. 1999), especially in brain, because levels in rodent neonates fall rapidly after birth (Newland and Reile 1999; Stern et al. 2001).

It is impossible to know how much of the published experimental data, as well as ongoing research, may be distorted by contaminated diets. Although the “certified” diets provided by manufacturers may prove useful to investigators, independent confirmation of ingredients should be encouraged. Biomarkers of exposure, that is, tissue indices, are the key to interpreting exposure data. Such direct measures in experimental subjects (including controls) provide assurances that the investigator’s protocols are properly conducted. We uncovered our problem only because we include blood and tissue assays in our standard operating procedures when conducting research with mercury. We strongly urge all researchers to do likewise. In a brief survey of recent literature, we have been surprised by how often researchers neglect to mention diet, or describe it in terms such as “standard rat chow.” Infrequently, the authors may provide the name of the supplier and the diet label, which should be the minimum information provided.

Although the results reported here stem from our research focusing on mercury, the issue of diet-based contamination certainly is not limited to one agent. For example, investigators who study endocrine disruptors have become concerned by the presence of agents in laboratory animal diets that may mimic estrogens (Boettger-Tong et al. 1998). Particularly in investigations of low-level, environmentally relevant exposures, diet is an unwelcome confounder.

Figures and Tables Top

Figure 1.

Blood levels of total mercury, inorganic mercury, and percent inorganic mercury in control dams. The inorganic component is the product of the slow conversion of methylmercury, the source of the mercury, to the inorganic form (e.g., Rowland et al. 1984).

Table 1.

Blood mercury in control pups from dams that had been fed the Purina 5001 diet (experiment 1).

Table 2.

Total mercury in rat chow samples.

Table 3.

Mercury content analysis of BioServ AIN-93G diet and casein.

References Top

1. Boettger-Tong H, Murthy L, Chiappetta C, Kirkland JL, Goodwin B, Adlercreutz H, et al. 1998. A case of a laboratory animal feed with high estrogenic activity and its impact on in vivo responses to exogenously administered estrogens Environ Health Perspect 106:369–373.9637793 Find this article online
2. Clarkson TW 2002. The three modern faces of mercury Environ Health Perspect 110(suppl 1):11–23.11834460 Find this article online
3. Davidson PW, Myers GJ, Weiss B 2004. Mercury exposure and child development outcomes Pediatrics 113(4 suppl):1023–1029.15060195 Find this article online
4. Grandjean P, Budtz-Jorgensen E, Steuerwald U, Heinzow B, Needham LL, Jorgensen PJ, et al. 2003. Attenuated growth of breast-fed children exposed to increased concentrations of methylmercury and polychlorinated biphenyls FASEB J 17:699–701.12586743 Find this article online
5. Lapham LW, Cernichiari E, Cox C, Myers GJ, Baggs RB, Brewer R, et al. 1995. An analysis of autopsy brain tissue from infants prenatally exposed to methylmercury Neurotoxicology 16:689–704.8714873 Find this article online
6. Magos L, Clarkson TW 1972. Atomic absorption determination of total, inorganic and organic mercury in blood J Assoc Anal Chem 55:966–971. Find this article online
7. National Research Council 2000. Committee on the Toxicological Effects of Methylmercury. Toxicological Effects of Methylmercury. Washington, DC:National Academy Press
8. Newland MC, Reile PA 1999. Blood and brain mercury levels after chronic gestational exposure to methylmercury in rats Toxicol Sci 50:106–116.10445759 Find this article online
9. Rowland IR, Robinson RD, Doherty RA 1984. Effects of diet on mercury metabolism and excretion in mice given methylmercury: role of gut flora Arch Environ Health 39:401–408.6524959 Find this article online
10. Stern S, Cox C, Cernichiari E, Balys M, Weiss B 2001. Perinatal and lifetime exposure to methylmercury in the mouse: blood and brain concentrations of mercury to 26 months of age Neurotoxicology 22:467–477.11579926 Find this article online
11. Stewart PW, Reihman J, Lonky EI, Darvill TJ, Pagano J 2003. Cognitive development in preschool children prenatally exposed to PCBs and MeHg Neurotoxicol Teratol 25:11–22.12633733 Find this article online
12. Watanabe C, Yoshida K, Kasanuma Y, Kun Y, Satoh H 1999. In utero methylmercury exposure differentially affects the activities of selenoenzymes in the fetal mouse brain Environ Res 800:208–214. Find this article online
13. Widholm JJ, Villareal S, Seegal RF, Schantz SL 2004. Spatial alternation deficits following developmental exposure to Aroclor 1254 and/or methylmercury in rats Toxicol Sci 82:577–589.15456922 Find this article online