The potential relationship between chromium excretion and insulin signaling raises questions about potential associations between chromium action and diabetes. Over 120 million people across the planet have been estimated to have diabetes mellitus, with approximately 16 million of these in the United States. Between 90 and 95% of the US cases are of type 2 diabetes (also called adult-onset or non-insulin dependent diabetes). Type 2 diabetes is the form responsible for the rapid increase in diabetes cases over the last few decades, and the number of cases are increasing rapidly, especially in third-world countries. Obesity is a risk factor for adult-onset diabetes, and the occurrence of the syndrome increases with age; both the average age and rate of obesity are increasing in the United States. African Americans, American Indians, Hispanics, and Pacific Islanders are especially susceptible to the disease. Unlike juvenile diabetes, which is an autoimmune disorder, type 2 diabetes results from insulin resistance, the body produces and releases insulin normally; however, the insulin signal is not properly transmitted into cells. With time, however, the beta cells of the pancreas break down resulting in reductions in insulin production. The cause of the disease at a molecular level has only been elucidated in a tiny fraction of cases. Consequently, any mechanism by which insulin signaling could be stimulated has possible value in the treatment of the symptoms of type 2 diabetes.
Several studies attempted to test the existence of a link between type 2 diabetes and chromium. Studies examining serum and urine Cr concentrations of healthy and diabetic subjects (using analytical techniques developed after c. 1980 and reporting serum or urine Cr levels of healthy subjects less than approximately 0.5 ^g/L)  have found Cr levels of diabetics that are distinctly different from those of healthy individuals. Morris and coworkers have reported that serum levels of type 2 diabetics are approximately one-third lower than those of healthy subjects while urine Cr concentrations are about twice as high [194-196]. In the first 2 years of the onset of diabetes, plasma Cr levels inversely correlated with plasma glucose concentrations, but this trend disappeared for patients with the disease for longer duration. Increased urinary output in type 2 diabetics would be consistent with studies with healthy individuals that indicate increased urinary Cr output and decreased serum Cr levels in response to increases in serum glucose or insulin [156-161]. It should be noted that Anderson and coworkers have shown in a double-blind crossover, placebo-controlled study that serum Cr levels reflect Cr intake but reported no effect by a glucose challenge . However, they only used serum glucose values 90 minutes after glucose treatment; in studies with multiple time points from 0 to 180 minutes after glucose treatment, serum chromium decreases but is restored to near original concentrations by 90 minutes [154, 160, 161].
In his 1998 review article, Anderson  identified 16 studies of chromium supplementation of type 2 diabetic subjects [68, 120, 122, 126, 139, 141, 142, 198-206]. However, only six of these studies were placebo-controlled, double-blind crossover in design. The first in 1968 by Sherman and coworkers giving seven adult males 150 ^g Cr/day as CrCl3 for 16 weeks observed no effect on fasting glucose levels and no effects on glucose tolerance tests (0-180 minutes after glucose administration) . Rabinowitz and coworkers in a larger study in 1983 gave 43 adult men the same amount of CrCl3 for the same length of time . No effects were observed on fasting blood levels of cholesterol, triglycerides, or glucose or observed in glucose levels 2 hours after subjects consumed a meal. Similar results were obtained in 1983 by Uusitupa and coworkers working with ten adults that were supplemented for 6 weeks with 200 ^g of Cr as CrCl3 per day . No effects were observed on total cholesterol, triglycerides, HDL, LDL, or VLDL cholesterol, or fasting glucose or insulin levels. No effects were observed either in glucose levels 1 or 2 hours or in insulin levels 2 hours after an oral glucose challenge; however, insulin levels 1 hour after the challenge were significantly lower. No effects on total cholesterol, HDL or LDL cholesterol, triglycerides, or glucose and insulin levels 90 minutes after a glucose challenge were observed by Thomas and Gropper in 1997 . These researchers used a pool of five adults which were given 200 ^g of Cr as CrNic per day for 8 weeks. In 1994, Lee and Reasner  examined 28 adult diabetic subjects given 200 ^g/day Cr(pic)3 for 8 weeks. (It is possible that the authors meant to indicate that subjects were given 200 ^g/day Cr, not Cr(pic)3, but this is not how the experimental description reads.) No effect was observed on HDL or LDL cholesterol levels or fasting glucose levels. Triglyceride levels were lower significantly. Contrasting results were obtained by Gary Evans in 1989 . Using 11 adults given 200 ^g Cr as Cr(pic)3 for 6 weeks, Evans noted significantly lower fasting glucose, total cholesterol, and LDL cholesterol concentrations. Evans used paired i-tests to determine the significance of the difference of the means (as previously described). This gives results such as initial and final concentrations of LDL cholesterol (mean ± SE) of 148 ± 12 vs. 140 ± 11 and of 142 ± 11 vs. 134 ± 12 to be significant (P < 0.05). The reportedly lower glucose and cholesterol concentrations disappear when standard statistical treatment is used.
While type 2 diabetes is associated with abnormally low serum Cr and high urine Cr levels possibly related to elevated serum glucose and insulin levels, the consensus of these studies appears to be that Cr supplementation for 6-16 weeks has no effect on diabetic subjects. This begs the question of the significance of the increased Cr movement. Do diabetics improperly transport Cr? Is Cr movement the result of some other phenomenon such as altered iron metabolism? Do diabetics slowly deplete Cr stores? However, since 1995, many studies examining the effects of Cr on type 2 diabetics have been reported. These studies are the focus of Chapter 8.
Other conditions resulting in increased urinary Cr loss
Three other conditions reported to result in increased urinary Cr loss are trauma, exercise, and pregnancy. In the case of trauma [207, 208], urinary Cr excretion is very high but appears to decrease rapidly. Effects of Cr intake are difficult to analyze as intake varies dramatically depending on patient treatment. The appropriate level of Cr in TPN solutions has been an issue of debate, but that debate is beyond the scope of this review. Acute exercise-induced changes associated with increased glucose utilization have been found to result in increased urinary chromium excretion in several human studies [71, 157, 209-211], but not all [69, 70]. Clarkson has determined that insufficient evidence of any beneficial effects existed to recommend chromium supplementation for athletes .
Unfortunately, data on any potential relationship between Cr and pregnancy and especially gestational diabetes are sparse, especially after 1980 and the use of reliable analytical techniques to determine tissue and fluid Cr concentrations. Patients in the first half of pregnancy have been reported to have higher Cr excretion . Patients in the second half of pregnancy had urinary Cr levels 30% higher than controls, but the difference was not significant. The concentration of Cr in hair from women with gestational diabetes appears to be lower than that of controls . Jovanovic has presented results of chromium supplementation of women with gestational diabetes . Chromium was reported to lower glucose and insulin levels compared to controls. If the results of this study are reproduced by additional studies, this could have significant implications for treatment of this condition.
If chromium is an essential element, then a specific transport mechanism should exist for the element. However, the existence of such a mechanism is not proof that the element is essential, as toxic metals are also recognized, transported, and excreted.
The mechanisms of absorption and transport of chromic ions are still uncertain. Little is known of the fate of Cr3+ that is taken orally. As a start, essentially no data exists on the forms of Cr(III) in food as a result of its very low concentration. Presumably, some small percentage of chromium in mammalian tissues is LMWCr. However, the fate of this or other forms of dietary chromium in the digestive tract (exposure to proteases and other hydrolases, highly acidic pH of the stomach, etc.) is also unknown. As described earlier, only a small percentage (<2%) of dietary Cr is absorbed, while the remainder is excreted in the feces. Chromium supplementation of the diet results in an increase in urinary chromium loss, and most absorbed chromium is rapidly excreted . Dowling and coworkers have examined the absorption of Cr3+ from an intestinal perfusate with added Cr as CrCl3 . Cr absorption was found to be a nonsaturable process, leading to the conclusion that the Cr3+ was "absorbed by the nonmediated process of passive diffusion in the small intestine of rats fed a Cr-adequate diet." This study was methodologically superior to previous studies which have yielded conflicting results [11, 217, 218]. Yet, the fate of dietary Cr(III) organic complexes could be different than that of the form(s) of chromium formed in the perfusate. For example, the presence of added amino acids, phyate (high levels), and oxalate in the diet reportedly alter Cr uptake [219, 220], as does ascorbic acid . Low levels of phyate appear to have no effect on absorption . Absorption of dietary Cr may be more complicated than simple passive absorption. The perfusate studies also would appear to contradict the inverse relationship between dietary chromium intake and degree of absorption observed in human studies ; the authors of the human study have also examined absorption of chromium by rats . They proposed that Cr homeostasis was maintained at the level of excretion, not absorption, and suggested Cr uptake by rats may be different from that in humans.
The fate of chromium in the bloodstream is somewhat better elucidated. In vivo administration of chromic ions to mammals by injection results in the appearance of chromic ions in the iron-transport protein transferrin. In 1964, Hopkins and Schwarz established 51CrCl3 given by stomach tube to rats resulted in >99% of the chromium in blood being associated with non-cellular components . Ninety percent of the Cr in blood serum was associated with the ^-globulin fractions; 80% immunoprecipitated with transferrin . In vitro studies of the addition of chromium sources to blood or blood plasma also result in the loading of transferrin with Cr(III), although under these conditions albumin and some degradation products also bind chromium [224, 225]. In vitro studies suggest transferrin may be important for transport from the intestines . Transferrin is an 80,000 Da blood serum protein that tightly binds two equivalents of ferric iron at neutral and slightly basic pH's. The protein exhibits amazing selectively for Fe(III) in a biological environment because the metal sites are adapted to bind ions with large charge-to-size ratios. In humans, transferrin is maintained only approximately 30% loaded with iron on average and consequently has been proposed to potentially carry other metal ions . The similar charge and ionic radii of chromic ions to ferric ions suggest that chromic ions should bind relatively tightly to the protein. In vitro studies of the addition of chromic ions to isolated transferrin reveal that Cr(III) readily binds to the two metal-binding sites, resulting in intense changes in the protein's ultraviolet spectrum [228-234]. The two chromium-binding sites can be distinguished by EPR, and only chromic ions at one site can be displaced by iron at near neutral pH [229, 231]. Below pH 6, only one site binds chromium .
As a result of the in vivo and in vitro studies, it has been assumed reasonably that transferrin was involved in chromium transport, although by 1995 transport had not conclusively demonstrated in vivo. Could increased loading of transferrin with iron prevent adequate chromium binding and transfer by transferrin, resulting in insulin resistance and diabetes? The hemochromatotic diabetic condition is certainly exacerbated by reduced chromium retention , as observed in adult-onset diabetic patients. Recent advances on elucidating the mechanism transport of Cr are reviewed in Chapter 6, while association between the transport of Cr and potential roles in insulin signaling are described in Chapter 7.
Only minor amounts of chromium are lost through the bile (for example, [236, 237]). In bile, Cr3+ occurs as part of a low-molecular-weight organic complex ; the molecular weight of this species was not determined. The authors postulated that this complex might be involved in passage of chromium from the liver to the bile.
Exactly how chromium is handled by the kidneys is difficult to determine from the literature. Some useful tracer studies with 51Cr have been reported (for example, see [239-241]); yet, much of the literature looking at absolute Cr levels dates before 1980 and, thus, suffers from analytical problems. The tracer studies appear to be influenced by the form of chromium used and perhaps by the species of animal utilized. Similar problems with chromium form are associated with studies examining Cr distribution in mammals. Differences may exist between studies that give Cr orally or intravenously as different Cr(III) species may be introduced into the blood. Similarly, tracer studies may not accurately reflect the fate of dietary chromium; for example, inorganic chromium added to a diet may not bind to the same chelates and other ligands binding to the chromium naturally in the diet. Yet, some consensus does appear. 51Cr in tracer studies accumulates in the bone, kidney, spleen, and liver (for example, [11,107, 236,241-245]). A three-compartment model has been proposed to examine the kinetics of chromium tissue exchange and distribution for studies with rats and humans[11, 242, 243]. Plasma chromium is in equilibrium with the three pools: a small pool with rapid exchange (T1/2 < 1 day), a medium pool with a medium rate of exchange (days), and a large, slowly exchanging pool (months). Jain and coworkers have suggested from these studies that chromium has specific transporters that regulate its movement . Aging appears to affect Cr distribution and transport .
Was this article helpful?