Trace Metal Absorption and Transport

Abstract

The extent of digestion of the complete diet is a major determinant of how well trace metal (elements) are absorbed by the gastrointestinal tract. Malabsorption tends to reduce trace metal uptake and absorption. Copper must be reduced to the cuprous ion (Cu ⁺) prior to uptake mediated by the apically located CRT1 transporter. CRT1-mediated transport increases in dietary copper deficiency. Cu ⁺is transported across the trans-Golgi network by ATP7A. This transporter is also located at the basolateral membrane where an ATP-requiring mechanism pumps Cu ⁺into the plasma. ZIP4 is the major zinc transporter located at the apical membrane. Zip4 gene transcription increases in zinc deficiency and contributes to homeostatic upregulation of zinc transport at the apical surface. Zn ^{2 +}is transported into the trans-Golgi network by the transporter ZnT7 and/or could be bound to cytoplasmic binding proteins such as metallothionein. This protein acts as a metal buffer and the metallothionein gene is regulated by copper and zinc. Elevated synthesis of the protein may scavenge copper ions and limit copper availability for transport by ATP7A. Zinc associated with vesicles is transported from enterocytes by ZnT1 and perhaps other efflux transporters. Zinc transporter ZIP14 is believed to contribute to intestinal barrier function. Other ZIP and ZnT proteins may contribute to trafficking of zinc and other trace metals by enterocytes. Manganese absorption does not appear to require the metal transporter DMT1. Both hepatic ZIP14 and ZnT10 are necessary for effective secretion of manganese into the bile to prevent manganese accumulation by tissues. Selenocysteine and selenomethionine are well-absorbed sources of selenium and absorption occurs via various amino acid transporters, but without homeostatic regulation. Dietary deficiency of each of these trace metals has been shown to produce deficiency signs in humans and rodents. Defects in transport of zinc, copper, and manganese have been linked to specific medical disorders.

Keywords

Zinc absorption, Copper absorption, Selenium absorption, Manganese absorption, Chromium absorption

Acknowledgments

Research from the author’s laboratory as cited here is the result of support from the National Institutes of Health. Figs. 61.2–61.4 were produced by AASarts.com .

61.1 General Properties of Trace Metal Absorption

The trace metals are roughly divided into the “trace” metals and the “ultratrace” metals. The latter term comprises a group of elements, which includes chromium, for which there are various degrees of evidence to support nutritional essentiality. Evidence may be limited to minimal effects such as growth reduction during a dietary restriction under controlled conditions of husbandry including ultraclean environments. In some cases, the ultratrace metals have no clearly established biochemical function. In contrast, the trace metals such as copper, manganese, selenium, and zinc have well-established, biochemically based, deficiency syndromes that define nutritional essentially. These metals have a well-characterized basis for essentiality derived from the animal nutrition studies and support essentiality for humans. Along with iron, copper and zinc are the metals upon which most of our knowledge about metal transport into and from the intestinal epithelial cell resides. Efficiency of utilization (bioavailability) includes factors such as diet composition which influences the extent of digestion and endogenous secretion as well as medical disorders leading to infection and inflammation.

61.2 Diet Composition

Absorption of all trace metals from food and fluids consumed is influenced by their physical and biochemical properties. The extent to which these components are degraded into smaller components influences absorption of individual constituents. Solubility is a factor that influences trace metal absorption. Trace metals tend to be widely dispersed among foods. For example, zinc has important functions in cell replication and gene expression. Consequently in food ingredients, zinc is abundant where nucleic acids are concentrated, that is, grains. The high zinc content of wheat germ is an example. Zinc is abundant in muscle and provides a major food source. Selenium, manganese, and copper, having enzymatic roles, are concentrated in foods where metabolic activity is very high, that is, grains, and in organ meats and muscle. Animal- and plant-derived foods provide selenium as selenocysteine and selenomethionine, respectively. The trace metals are presented to the intestinal lumen initially as protein/peptide complexes.

The extent of digestion and transit times are determining factors for absorption ( Fig. 61.1 ). Trace metals tightly bound to high molecular mass compounds, for example, large peptides, may not interact with transporters responsible for uptake by enterocytes. Unless these dietary constituents are well digested these constitute an unavailable pool of a particular trace metal. The pool of a trace metal available for absorption most likely is composed of low-molecular mass compounds, for example, amino acid chelates, peptides, or “free” ions of the element. The binding affinities of various molecules for a specific metal dictate how much “free” ion exists.

61.3 Mechanisms of Trace Metal Uptake, Transcellular Movement, and Efflux by the Gastrointestinal Tract

Unlike nutrients whose solubility is high, kinetic analysis of trace metal uptake is quite challenging. These micronutrients tend to be soluble in an acidic environment, but less so in a neutral milieu. Acidic gastric secretions tend to enhance solubility, but the stomach is not usually considered as a site of appreciable trace metal absorption. Recently, zinc uptake by the secretory compartment of the acid-secreting parietal cells of the stomach has been shown in vitro and may be coupled to acid secretion. Some zinc transporters (e.g., ZIP11) are highly expressed in the stomach and are downregulated with dietary zinc depletion. Such findings may cause further consideration of the stomach as a factor in trace metal absorption or may reflect functional roles in the gastrointestinal tract. How the acidic microenvironment of the intestinal apical surface influences trace metal uptake has not been studied. As the gastric contents enter the duodenum, neutralization is rapid and yields a concomitant reduction in solubility. Despite the neutral pH environment, the small intestine is the major site of trace metal absorption. Uptake for copper, manganese, and zinc follows saturable kinetics. Since the extent of absorption is a function of interaction of the metal ion with transporters at the apical surface of enterocytes, the more the interaction that exists with metal-binding dietary constituents of low affinity, the greater should be the uptake rate. Saturable kinetics (mediated component of uptake) implies that one or more transporter molecules are involved. As luminal concentrations of a trace metal increase, nonmediated uptake tends to be predominant. This is believed to represent paracellular transfer of metals and metal chelates, that is, those metals bound to small peptides and potentially trace metals presented in the form of nanoparticles. Transepithelial movement of trace metals occurs by poorly understood mechanisms. The demonstration that many transporters are localized to intracellular vesicles, endosomes, the ER, and Golgi of enterocytes suggests multiple transporters contribute to transcellular movement of metal ions. Mechanisms for transport across the basolateral membrane and the endothelium are also being defined. Efflux transporter genes for copper and zinc have been identified. The intracellular pools from which absorbed trace metals are drawn for cellular efflux are not well understood. These trace metals sequestered in endocytotic vesicles may be the substrates for efflux transporters at the basolateral membrane.

Kinetic analysis of trace metal absorption has employed many techniques, ranging from in vivo perfusion studies with humans to in vivo intestinal cell preparations and even membrane vesicles from experimental animals and immortalized cell lines. Unlike in vitro systems with relatively purified components, in vivo approaches designed to mimic the actual luminal environment can only approximate the true concentration of the metal in question. Nevertheless, measured kinetic parameters from these methods are frequently within the expected range, based on dietary consumption of these micronutrients and eventual concentrations in peripheral blood.

Evidence to support mediated transport for copper, manganese, selenium, and zinc absorption is substantial, but there is a fair degree of controversy regarding the components involved. For example, divalent metal ion transporter 1 (DMT1, SLC11A2) is believed to be the iron transporter responsible for Fe ^{2 +}uptake by enterocytes. In vitro experiments have shown that when Xenopus oocytes are transfected with Dmt1 mRNA, a number of other trace elements, such as copper, zinc, or manganese are also transported. Based on that and other in vitro evidence, it has been suggested that this transporter is involved in the intestinal uptake of multiple elements. However, the transport of copper by DMT1 has been challenged in a more recent in vitro study. In addition, experiments with an intestine-specific Dmt1 knockout mouse model show that DMT1 is specific for iron absorption, but is not required for the absorption of copper, zinc, or manganese. Consequently, in vitro and in vivo experiments can yield different results and interpretations. Aside from a few in vitro competition experiments with transporters, the interactions of trace metals have not been thoroughly evaluated at the molecular level. Despite the lack of evidence, it does seem reasonable to suggest that transport activities for some trace metal transporter proteins are leaky and, thus, may transport both multiple essential trace metals. This situation may occur when the dietary availability of the natural metal substrate is low. Such circumstances may lead to the absorption of these metals with no known biological functions, for example, cadmium, which are considered toxic substances. Few studies have evaluated the influence of dysfunctional human transporter genes on trace metal absorption. A classic example is the malabsorption of ⁶⁵Zn in patients diagnosed with acrodermatitis enteropathica (AE) produced by mutated Zip4 (reviewed in Ref. ). Another example is mutated Zip14 that produces a form of ZIP14 which does not transport manganese properly and may prevent the normal hepatic transfer of absorbed manganese into the bile for fecal excretion.

Numerous metal transporter proteins documented to be expressed in various cell types may contribute to trace metal absorption or excretion, but have not yet been localized within cells of the gastrointestinal tract. Consequently, their individual roles in pathways of metal homeostasis have not been defined, but may influence interpretation of existing data. Further research, particularly studies using knockout rodents which do not express a specific transporter gene in the gastrointestinal tract may shed light on trace metal absorption mechanisms.

61.4 Copper Absorption

Copper accumulation mechanisms have been studied for decades. Early studies with ⁶⁴Cu showed that appreciable amounts of copper were taken up via the stomach. Gastric acidity was believed to enhance copper solubility. The stomach may be the site of some interaction of copper with other ions for uptake. After gastric secretions enter the duodenum, pH change may influence such interactions. As reviewed in detail earlier, dietary protein, fiber, phytate, ascorbic acid, and L-amino acids all have been shown to affect copper uptake by enterocytes. Protein digestibility may influence the formation of these complexes and high-protein diets generally improve copper absorption.

Numerous constituents found in luminal contents of the small intestine, of both dietary and endogenous origin, have been shown to have positive or negative effects on copper absorption, presumably via influences at the brush border surface of enterocytes. Zinc has been shown to inhibit copper absorption. The site of the interaction may be a transporter molecule or a zinc-copper interaction with enterocytes, as discussed. Both saturable and nonsaturable kinetics for copper absorption have been demonstrated. The latter may represent paracellular movement of copper complexes. The fractional absorption of copper in humans ranges from 75% at low copper intakes down to 12% at high intakes. The recommended dietary allowance (RDA) for copper of 900 μg/day for adult humans was based on dietary levels needed to maintain normal plasma copper concentrations, plasma ceruloplasmin activity, and erythrocyte SOD activity.

The homeostatic changes in copper absorption/retention are partially explained by a direct correlation between dietary copper intake and endogenous losses from slow and fast kinetic pools. Copper absorption by the intestine is saturable, and some of the molecules responsible for such kinetics have been identified in the last decade. Luminal copper is reduced to the cuprous (Cu ⁺) form by a reductase prior to transport. The enzymes potentially responsible include cytochrome B reductase 1 (CYBRD1) or STEAP reductase (reviewed in Ref. ). Copper uptake by enterocytes involves primarily the high-affinity transporter, CTR1 (SLC31A1). hCtr1 mRNA is found in human cell types at varying amounts of one of three transcript sizes. These differences may have functional significance. Homozygous Ctr1 -null embryos die at day 8.5 of the development. Intestinal CTR1 has been localized to the apical membrane of intestinal epithelial cells of the mouse and other animals. Protein stabilization is responsible for its regulation by dietary copper. An intestinal-specific Ctr1 knockout results in copper accumulation within enterocytes and the mice develop a copper-deficient phenotype. Liver, pancreas, and heart express more CTR1 than the intestine. The chemotherapeutic agent cisplatin is transported by CTR1. DMT1 (SLC11A2) is an iron transporter (ferrous) that upregulates in intestines of iron-deficient animals. DMT1 is not believed to be a major factor in copper absorption under normal physiological conditions. The transporter could provide a site for interactions among dietary trace metals, particularly when one or more is not abundant in/or available from the diet (reviewed in Ref. ).

Copper is a strong Lewis acid and because of this property it has a high propensity to bind to proteins. Cells are believed to use copper chaperones to help in trafficking of Cu ⁺. These include COX17 and CCS, which deliver copper to cytochrome C oxidase and Cu,Zn-superoxide dismutase, respectively. ATOX1 (which may also act as a transcription factor that undergoes nuclear localization upon copper binding) is a chaperone for copper delivery to ATP7A.

A breakthrough in understanding how copper is released from enterocytes was made through the cloning of the Menkes gene ( Atp7A ). ATP7A is a P-type ATPase (reviewed in Ref. ). Menkes disease is a terminal genetic disorder, with patients presenting with progressive neurological disease. Copper accumulates in intestinal cells of Menkes patients. This suggests that, since ATP7A is expressed in intestinal cells and the protein localizes to the basolateral membrane, it is the copper transporter that is most likely responsible for saturable kinetics and the rate-limiting factor in copper absorption. It appears to be responsible for copper efflux by pumping Cu ⁺out of enterocytes to the extracellular fluid (ECF). Key components of copper transport by enterocytes are shown in Fig. 61.1 .

Release of copper from enterocytes to the ECF under normal conditions results in binding primarily to plasma albumin. Release from enterocytes does not appear to require albumin, as analbuminemic rats show normal copper absorption characteristics. This suggests that another plasma protein, for example, a2-macroglobulin, can contribute to copper transport in the portal circulation in that situation. Small amounts of copper may circulate in plasma bound to low-molecular weight constituents.

High oral intake of zinc has been shown to reduce copper transfer to the plasma. This relationship has been used as a FDA-approved therapy to lower copper absorption in patients with Wilson’s disease following copper depletion therapy. One mechanism proposed to explain this influence on limitation of copper loading is the induction of the copper-/zinc-binding protein, metallothionein (MT). With this model, copper would be retained in enterocytes when induced to high levels of synthesis by zinc. Presumably, the copper would be lost along with desquamation of enterocytes. This mechanism has also been shown in animal studies. Alternatively, a Cu/Zn competition for a common transporter within enterocytes or dysregulation of expression of a copper transporter/chaperone could be a factor that would explain the influence of large amounts of oral zinc on body copper accumulation.

Sheep experience a myelopathy that produces ataxia when a copper deficient diet is fed. A similar clinical scenario with myelopathy is observed in humans with an idiopathic copper deficiency. A role for copper is supported by extremely low serum copper concentrations and improvement with supplemental oral copper in these patients. Abnormal magnetic resonance imaging (MRI) confirmed myelopolyneuropathy in other patients with ataxic-like symptoms. These patients presented with extremely low serum copper concentrations and low ceruloplasmin activity coupled with extremely high serum zinc levels. The putative cause of these clinical findings was the excessive use of zinc-rich denture creams that allowed excess zinc to be leached into the gastrointestinal tract and absorbed. It has been proposed that these conditions created an acute copper malabsorption leading to the clinical findings that are similar to nutritional copper deficiency. The mechanism responsible was not established. One proposed mechanism was that excess zinc in the gastrointestinal tract led to the induction of MT in enterocytes that would bind copper preferentially to zinc and limit copper transport the portal supply. This mechanism is similar to the zinc therapy proposed for Wilson’s disease. Alternatively, symptomatic patients may have defects in genes coding for copper transporters or chaperones and/or zinc may interfere with the expression or function of these proteins in enterocytes ( Fig. 61.2 ).

61.5 Zinc Absorption

The availability of radioactive isotopes of zinc (namely ⁶⁵Zn) during World War II produced research that has led to a firm understanding of zinc metabolism. Particularly relevant milestones include the demonstration of homeostatic regulation of zinc absorption and early application of compartmental analysis to understand zinc flux from specific intracellular pools and the ECF.

Zinc is an excellent example of how availability of binding ligands influences solubility and the extent of absorption ( Fig. 61.1 ). Body zinc status also has an important influence on zinc absorption. This relationship was used to calculate the RDA for zinc. There is no direct evidence suggesting appreciable zinc absorption occurs in the stomach. Newer evidence suggests the transporter ZIP11 is highly expressed in the stomach and upregulated in zinc restriction. Zinc transport into parietal cells influences acid secretions. Zinc treatment increases acid output by the stomach which may explain the emetic response to consumption of high doses of zinc by humans. However, inhibition of gastric secretions lowers zinc absorption, presumably through a reduction in solubility. Grain products can provide a substantial amount of dietary zinc. Much of this is tightly bound as a complex with phytic acid (hexaphosphoinositol), which makes it less available. Zinc bioavailability can vary from 55% (red meat) to 15% (high-fiber cereal). Zinc bioavailability from human milk is greater than cow’s milk, most likely because of the greater extent of gastrointestinal hydrolysis of human milk proteins. Numerous dietary constituents have been reported to influence zinc absorption.

Zinc appears to be absorbed along the entire intestinal tract, including the colon. Perfusion experiments with humans suggest that the jejunum has the highest absorption rate. Regional differences may reflect differences in zinc transporter expression and transit time in that anatomical segment. The absorption rate in humans is saturable above a luminal concentration of 1.8 mM. After a meal, the zinc concentration of the lumen may reach only 100 μM, indicating that most dietary zinc is absorbed by a saturable (mediated) process. This also suggests that, except for ingestion of zinc supplements, little absorption occurs via the paracellular route. Where it has been studied experimentally, absorption rates increase in response to low dietary zinc intake (reviewed in Refs. ). Numerous zinc transporter genes are expressed in the small intestine of mice. ZIP4 is believed to be the major zinc transporter in enterocytes. The mechanism of upregulation during zinc restriction is both transcriptional regulation of Zip4 and stabilization of the Zip4 transcript. The transcription factor, KLF4, has been shown to influence Zip14 mRNA levels in response to zinc intake. Studies using siRNA knockdown, transfections with Zip4 cDNA, and other techniques have shown the important transporting capacity of ZIP4. Some regulation of Zip4 may occur through posttranscriptional regulation. The autosomal recessive trait, AE, is a zinc-responsive, zinc malabsorption disorder. It presents as skin lesions, mental problems, and immune dysfunction that start when children are weaned from breast milk. The decreased bioavailability of zinc from food compared to human milk has been proposed as the cause for the appearance of disease signs at weaning. The AE patients show remission of skin lesions upon provision of supplemental oral zinc therapy. The effectiveness of oral zinc in the alleviation of symptoms of AE suggests that the ZIP4 protein produced by the mutated Zip4 can either still transport zinc or other zinc transporters expressed in enterocytes, perhaps of lower affinity, participate in the absorption process. ZIP10 and ZIP11 would be possible candidates. Zip4 mutations do not produce apparent deficiencies of other trace metals suggesting that in vivo ZIP4 transports only zinc.

The Paneth cells of the crypts of the small intestine are rich in zinc and are the site of synthesis of antimicrobial peptides, particularly the defensins. These cells are one of the four cell lineages associated with intestinal cell proliferation and are the only cells that migrate in a downward route along the villus. ZIP7 has been reported to have an important zinc transport role in Paneth cells as related to intestinal proliferation and limitation of intestinal ER stress. The substantial amount of zinc contained within Paneth cells could be related to peptide stabilization prior to secretion, but the contribution of these cells to endogenous zinc secretion, particularly during gastrointestinal infection has not been investigated.

A considerable body of evidence suggests that zinc within intestinal cells is localized within vesicles. Some transcellular zinc movement within enterocytes may occur via a vesicular pathway. ZnT2 and ZnT4 have vesicular/secretory granule localizations in other cells and are likely involved in zinc trafficking in enterocytes (reviewed in Ref. ). ZnT7 has been shown to participate in zinc absorption as accumulation of ⁶⁵Zn by multiple organs is markedly reduced in ZnT7 knockout mice. Interestingly, these mice do not exhibit a zinc-deficient phenotype, suggesting ZnT7 may have a different function beyond trafficking through the Golgi. The cytosolic protein MT has been extensively studied within the context of intestinal trace element absorption in wild type and knockout mouse models and other species. The consensus is that the protein acts as a buffer to help regulate intracellular zinc levels. The two metal-binding clusters of MT are not equivalent. The beta-cluster has more facile zinc-release properties, particularly under conditions of physiologic adaptations. In rat enterocytes, the zinc exporter protein ZnT-1 has been localized primarily to the basolateral membrane. The expression of ZnT-1 in enterocytes is refractory to dietary zinc restriction, but since it is regulated by the transcription factor MTF-1 expression increases with high zinc intakes in the diet. The basolateral orientation of ZnT1 in enterocytes, particularly in the proximal small intestine, and its constitutive expression in low and adequate zinc intake levels make it a well-suited contributor for the maintenance of the zinc supply. Evidence for zinc transport in the portal blood supply is as a complex with albumin. Some plasma zinc is in the form of low-molecular weight complexes.

There is evidence suggesting that some endogenous zinc is transported into the intestinal lumen by enterocytes. In this regard, the localization of ZIP5 at the basolateral membrane of both enterocytes and pancreatic acinar cells, suggests this transporter may function as a route for the release of endogenous zinc into the gastrointestinal tract and/or serve as a zinc-monitoring mechanism for cells where it is expressed. More recently, the transporter ZIP14 has been localized to the basolateral membrane of mouse enterocytes. Because of this orientation, ZIP14 is believed to play a functional role in providing zinc to the enterocytes. Zip14 knockout mice have defective barrier function related to tight junction protein deficiencies. In these mice, zinc is sequestered in endosomal vesicles and is unavailable. Hence ZIP14 may also serve as a route of endogenous zinc release. The expression of ZIP11 in the stomach and through to the colon have not yet been fully evaluated as a factor in enteric zinc absorption. The transporter ZnT5 is reported to produce bidirectional zinc transport in transfected intestinal cells.

Pancreatic secretions contain large amounts of zinc that enter into the intestinal lumen. This release is homeostatically regulated, and this endogenous zinc may be available for reuptake into enterocytes. Experiments with mice suggest downregulation of ZnT-1 and ZnT-2 in pancreatic acinar cells may account for much of this metabolic adaptation. ZnT1 and ZnT2 are localized to the apical plasma membrane and zymogen granules, respectively. Regulation of ZnT1 is via MTF-1 and ZnT2 is through both MTF-1 and STAT5. The latter transcription factor is activated by glucocorticoid hormone, which also regulates amylase synthesis and secretion.

The amount of endogenous zinc that enters the intestinal lumen is influenced by diseases of the GI tract, particularly those causing changes in transit time or fluid secretion. Since pancreatic secretions are believed to provide the bulk of the endogenous zinc losses (reviewed in Ref. ), pancreatic diseases leading to malabsorption may influence zinc bioavailability. In humans, intestinal endogenous losses reach 5 mg/day, but are markedly reduced with zinc restriction. The wealth of data with stable isotopes of zinc was used to measure endogenous losses in humans and to calculate the estimated average requirement (EAR) to establish the RDA for zinc. The RDA for adult humans is 11 mg/day for males and 8 mg/day for females. Nutrition intake data indicate that the overall risk of inadequate zinc intake in North America and Europe is low, but globally, over half of the world’s population does not get enough zinc.

Intestinal disease can influence zinc absorption and produce signs of zinc deficiency. Crohn’s disease is an inflammatory bowel disease that causes increased excretion, concomitant decreased absorption, and body zinc redistribution. Signs of enteritis are reduced with supplemental zinc (25 mg/day). Abnormalities of zinc metabolism, usually including low plasma zinc concentration, have been reported in patients with celiac sprue, short bowel syndrome, and gastric bypass surgery. This may be a reflection of depressed zinc absorption. Secretory diarrhea associated with intestinal tract infection or AIDS can lead to zinc loss, which further depresses the immune system. Remarkable reductions in childhood morbidity due to diarrheal disease have been made through zinc supplementation of such patients. These efforts may point to a beneficial effect of zinc on mucosal immunity, particularly for the intestinal epithelium. The influence of supplemental zinc on taxa of the intestinal microbiota is an area that needs to be explored. In support of that need is the influence of zinc status on intestinal barrier function and metabolic endotoxemia.

Neutrophil-produced calprotectin (S 100 protein) is a heterodimeric, antimicrobial, protein that sequesters enteric zinc and manganese in a 1:1 stoichiometry to prevent acquisition of these metals by microorganisms and thus limits their colonization. If this zinc-chelating role is overwhelmed, enteric microorganisms, such as the food-borne pathogen Salmonella , thrive in the gastrointestinal tract. Fecal calprotectin is a measure of intestinal inflammation, but the contribution of calprotectin-bound zinc-to‑zinc excretion and loss has not been studied. However, the contribution of endogenous zinc to intestinal calprotectin and enteric microbial growth are beginning to be studied ( Fig. 61.3 ).