Since preparation of the GRAS determination document in 2002, additional studies have become available to assess the safety of chromium tripicolinate. This addendum to the 2002 GRAS determination reviews the impact of the findings from those studies on the conclusion that chromium tripicolinate is GRAS.

The 2002 GRAS document noted that several in vitro assays demonstrated that gene mutations, DNA damage, and chromosome damage were seen in cultured cells and DNA preparations treated with relatively high concentrations of various chromium salts, including chromium tripicolinate. However, no increases in mutagenic damage were seen in an Ames Salmonella mutation assay conducted with chromium tripicolinate in the presence and absence of a rat liver homogenate metabolic activation system (Juturu and Komorowski 2002). Chromium tripicolinate was reported to produce increases in chromosome damage in Chinese hamster ovary cells within a concentration range of 0-3.0 mM and damage in isolated plasmid DNA after relatively rigorous treatments for periods from 5 to 180 minutes at concentrations from 0.12 uMto 120 uM (Speetjens et al. 1999). In contrast to some reports of mutagenic effects with in vitro test systems, no evidence of in vivo DNA damage was observed in urine samples of 10 human subjects consuming 400 mcg per day of chromium as chromium tripicolinate for approximately 56 days, as determined by measurements of 5-hydroxymethyl-2-deoxyuridine that was used as an indicator molecule for detecting oxidative DNA strand breakage (Kato et al. 1998). In addition, there were no adverse clinical signs of toxicity in rodents dosed with 5-100 ppm of chromium tripicolinate in the diet for 20 weeks (Anderson et al. 1997). The absence of adverse effects in the rodent study was in agreement with lack of clinical effects in human subjects that consumed 400 mcg per day of chromium as chromium tripicolinate. The lack of evidence of DNA damage in the repeat-dose study with human subjects led to the conclusion that in vitro genotoxic activity observed in cellular test systems was not relevant for evaluating potential risk to humans or for determining the GRAS status of chromium tripicolinate.

A review of the literature on chromium tripicolinate published since the compilation of the GRAS determination indicates that additional studies have generally confirmed the original observations of in vitro effects from chromium tripicolinate. Manygoats et al. (2002) noted that ultrastructural damage was produced in Chinese hamster ovary cells after treatments with chromic chloride or chromium tripicolinate. They noted that chromium tripicolinate, administered for an extensive 48-hour culture period at substantial doses of 1 mM (440 mcg/ml) to 3 mM (1340 mcg/ml), was the more active compound at producing mitochondrial damage.

Stearns et al. (2003) reported that chromium tripicolinate at concentrations of 0.75 to 3.0 mM was mutagenic to Chinese hamster ovary cells in vitro and produced increases in the numbers of mutant cells resistant to 6-thioguanine caused by a specific gene mutation. Chromic chloride was also mutagenic in this same test and produced a 10-fold increase in drug-resistant mutants relative to numbers of mutant cells found in control cultures. Although the authors considered chromium tripicolinate to be more highly active than chromic chloride, mutation effects in this in vitro test are apparently related to treatment of the CHO cells at concentrations of chromium (in any form) that far exceed concentrations that could ever be achieved in vivo.

The National Toxicology Program (NTP) has recently completed independent in vitro and in vivo genotoxicity assays of both chromium tripicolinate
(anhydrous) as well as chromium tripicolinate monohydrate (NTP 2003). The NTP reported summary results that showed that chromium tripicolinate
(anhydrous) did not produce chromosome damage in the in vivo mouse micronucleus assay and was also negative in two Salmonella (Ames) bacterial mutation assays. Chromium tripicolinate monohydrate was also observed to lack potential to produce adverse effects on chromosomes in the mouse micronucleus assay with males, with equivocal findings in females (NTP 2003). The absence of genetic toxicity findings in vivo are consistent with results reported by Greenberg et al. (1999) in studies with rats given chromium tripicolinate orally with doses of up to 2,000 mg/kg body weight. No increases in chromosome aberrations were seen following evaluation of chromosomes in bone marrow cells harvested at two time intervals following dosing to determine potential damage. The absence of genetic toxicity effects in vivo in mice in the NTP studies of two different forms of chromium tripicolinate confirm the absence of chromosome damage seen in rats (Greenberg et al. 1999) as reviewed in the GRAS document in 2002. The absence of genetic toxicity findings from in vivo results is also consistent with the lack of genotoxicity in DNA damage studies with human subjects as evaluated in the 2002 GRAS document (Kato et al. 1998). Thus, these additional studies confirm the conclusions in the GRAS review (ENVIRON 2002) that chromium tripicolinate is not genotoxic in vivo and the lack of significant genotoxicity findings contrasts directly to the reported effects detected in the artificial environment of cell culture screening tests. In vitro tests are conducted at relatively high concentrations of chromium tripicolinate, and may magnify the effect of potential chemical reactions with chromium tripicolinate reported by some authors (Bagchi et al. 1997; 2002). The bioavailability of dietary chromium from trivalent chromium salts is very low following ingestion (approximately 0.5 to 2%, with somewhat higher levels of 3% for chromium tripicolinate, as evaluated in animal and human studies (O'Flaherty 1996; Campbell 1999)). Therefore, the amount of chromium tripicolinate distributed in the bloodstream is several orders of magnitude lower than the concentrations tested in vitro. Therefore, in vitro effects appear to have limited (if any) relevance to living animals or to human risk assessment because of the high doses used, and the consistent lack of significant genetic toxicity seen with in vivo animal and human tests.

Additional genotoxicity and in vitro cytotoxicity studies have also been published since the preparation of the 2002 GRAS determination, but these appear to have limited application to safety assessment of chromium tripicolinate. Hepburn et al. (2003) evaluated chromium tripicolinate prepared in their laboratory for mutagenic potential in a wild-type strain of fruit flies (Drosophila melanogaster). Concentrations of chromium tripicolinate from 10.4 to 260 mcg/kg, given as a component of the standard diet, did not produce any adverse effects on viability, fertility or behavior in adult flies. Larvae exposed to a similar concentration range were reported to undergo developmental delays and decreased pupation success, but there is no substantive way to use these findings with an insect larva for assessing potential mammalian (much less human) toxicity. An unspecified concentration of chromium tripicolinate (described only as "dietary concentrations equivalent to those in human Chromium supplementation") was reported to produce increases in X-linked lethal mutations and dominant female sterility. The significance of mutagenic effects in fruit flies has questionable relevance to human risk assessment because of the significant differences in physiology and metabolism between insects and mammals, as well as the impossibility of extrapolating dosage effects from insects to humans. As noted by experts in genetic toxicology (Hoffman 1996), "the means of exposure, measurement of doses, metabolism and gametogenesis in Drosophila differ from those in mammalian toxicology. Mammalian assays therefore provide the best basis for assessing risk to human germ cells and hold a central place in genetic toxicology." The absence of genotoxic effects seen with in vivo assays in rats and mice described previously confirms that increases in mutations in insects have little if any significance in assessing mammalian genotoxicity of chromium tripicolinate.

Hepburn and Vincent (2003) attempted to determine the tissue distribution of chromium picolinate in rats following dosing by intravenous injection into the tail vein with radiolabelled 51Chromium-tripicolinate with sampling at six time intervals from 30 minutes to 24 hours after dosing. The authors noted that "for [chromium(picolinate)3] to have a deleterious effect on DNA via production of reactive oxygen species, the compound needs to enter cells intact and remain intact long enough to produce a quantity of reactive oxygen species." However, attempts by the authors to detect chromium tripicolinate itself in tissues and body fluids failed with the analytical detection methods employed. Because only 51Chromium marker was followed in the study, it is not possible to distinguish the form in which chromium was found in the cells and tissues. In addition, this intravenous administration is not relevant to extrapolations of the fate of chromium picolinate ingested orally by humans. Thus, although the results of this study confirm reports by others on distribution of chromium from chromium tripicolinate in specific body tissues, the report fails to provide information on chromium tripicolinate itself or its fate following ingestion.

Conclusion

Additional studies on in vitro cytotoxicity and genotoxicity of chromium tripicolinate have been published since preparation of a safety assessment document for this food supplement that concluded chromium tripicolinate was GRAS under its intended conditions of use. A review of these new studies show increases in genetic changes in vitro as noted previously. However, new studies conducted by the NTP confirm the absence of in vivo effects in mammalian test systems reported in the GRAS review document, and also do not show increases in mutations in the standardized Ames bacterial mutation test system. The relatively high doses used in the in vitro tests relative to the amounts ingested by humans, and the low degree of absorption from the diet, show that these in vitro test systems have no relevance to determination of human safety. The consistent lack of adverse toxicological or genetic effects in vivo supports the GRAS determination reviewed and agreed to by the GRAS panel convened in July 2002 (ENVIRON 2002).

Literature Cited

Anderson, RA, NA Bryden, and MM Polansky. 1997. Lack oftoxicity of chromium chloride and chromium picolinate in rats. J. Am. Coll. Nutr.
16: 273-279.

Bagchi, D, M Bagchi, J Balmoori, X Ye, and SJ Stohs. 1997. Comparative induction of oxidative stress in cultured J774A.1 macrophage cells by chromium picolinate and chromium nicotinate. Commun. Mol. Path. Pharm. 97:335-346.

Bagchi, D, SJ Stohs, BW Downs, M Bagchi and HG Preuss. 2002. Cytotoxicity and oxidative mechanisms of different forms of chromium. Toxicol 180: 5-22.

Campbell, WW, LJ Joseph, SL Davey, D Cyr-Campbell, RA Anderson and WJ Evans. 1999. Effects of resistance training and chromium picolinate on body composition and skeletal muscle in older men. J. Appl. Physiol. 86: 29-39.

ENVIRON. July 2002. Generally Recognized as Safe (GRAS) Determination for the Use of Chromax® Chromium Picolinate as a Nutrient Supplement in Food.

Greenberg, D, JR Komorowski, and KJ Loveday. 1999. Rat chromosomes are unharmed by orally administered chromium picolinate. Am. Coll. Nutr. 18: 27.

Hepburn, DDD,, J. Xiao, S Bindom, JB Vincent and J O'Donnell. 2003. Nutritional supplement chromium picolinate causes sterility and lethal mutation is Drosophila melanogaster. Proc. Nat. Acad Sci. 100: 3766-3771.

Hepburn, DDD and JB Vincent. 2003. Tissue and subcellular distribution of chromium picolinate with time after entering the blood stream. J Inorg. Biochem.94:86-93.

Hoffman, GR. 1996. Genetic Toxicology. Casarett & Doull's Toxicology: The Basic Science of Poisons. Ed. C.D. Klaassen. Chapter 9, pp 269-300, Fifth Edition, McGraw-Hill, NY.

Juturu, V, JR Komorowski. 2002. Antimutagenic activity of chromium picolinate in the Salmonella assay. XIV. World Congress of Pharmacol. July 7-12.

Kato, I., JH Vogelman, V Dilman, J Karkoszka, K Frenkel, NP Durr, N Orentreich and P Toniolo. 1998. Effect of supplementation with chromium picolinate on antibody titers to 5-hydroxymethyl uracil. Eur. J. Epidemiol. 14:621-626.

Manygoats, KR, M. Yazzie and Stearns, DV, 2002. Ultrastructural damage in chromium picolinate-treated cells: a TEM study. Transmission electron microscopy. J.Biol. Inorg. Chem 7: 791-8.

National Toxicology Program . 2003. Chromium Picolinate. http://ntp-server.niehs.nih.gov (Search: Chromium Picolinate).

O'Flaherty, EJ. 1996. A physiologically-based model of chromium kinetics in the rat. Tox. App. Pharmacol. 138: 54-64.

Speetjens, JK, RA Collins, JB Vincent, and SA Woski. 1999. The nutritional supplement chromium (III) tris(picolinate) cleaves DNA. Chem. Res. Toxicol 12: 483-487.

Stearns, DM, SM Silveira, KK Wolf and AM Luke. 2002. Chromium (III)
tris(picolinate) is mutagenic at the hypoxanthine (guanine) phosphoribosyl-transferase locus in Chinese hamster ovary cells. Mutat. Res. 513:135-142.