Iron Metabolism and Its Disorders

John W. Harvey

Department of Physiological Sciences
College of Veterinary Medicine
University of Florida
Gainesville, Florida

Iron is an essential nutrient required in a wide variety of metabolic processes. In solution, iron exists in two oxidation states, ferrous (Fe⁺²) and ferric (Fe⁺³), which can donate or accept electrons, respectively. Iron is typically in the Fe⁺² state for transport across membranes and in the Fe⁺³ state when bound to transport and storage proteins. Iron may be in either oxidation state when present in heme and iron-sulfur (Fe-S) cluster-containing proteins. The ability of iron to vary in oxidation state and redox potential, depending on the ligands it forms, enables iron to serve multiple functions. Iron-containing proteins are critical for oxygen transport and storage, respiration, DNA synthesis, citric acid cycle function, and various enzymatic reactions. Unfortunately, the same physical characteristics that allow iron to function as a cofactor in controlled redox biochemistry also makes iron potentially toxic to cells, because of its ability to catalyze the formation of reactive oxygen species. The binding of iron to coordinating ligands, such as O, N, and S, in proteins and the tight regulation of iron uptake, transport, and storage tends to shield iron from reactions with molecular oxygen that generate free radicals and injure tissues (Ryan and Aust, 1992).

FIGURE 9-1 Iron cycle. Iron (Fe) is highly conserved in the body. Iron in plasma is bound to transferrin, a transport protein that is synthesized in the liver. Iron is transported to all tissues, but most iron is utilized to synthesize hemoglobin in developing erythroid cells. Aged blood erythrocytes are phagocytized by macrophages, and hemoglobin is degraded. Released iron is either returned to plasma or stored in macrophages as ferritin and hemosiderin. Nearly all of the iron in plasma under normal conditions comes from the release of iron by macrophages that have phagocytized and degraded erythrocytes (large arrow). Only about 3% of the iron in plasma results from gastrointestinal (GI) enterocyte absorption in normal individuals (small arrow).

FIGURE 9-2 Overview of iron metabolism. Ferric iron (Fe⁺³) ions present in the diet are reduced to ferrous iron (Fe⁺²) ions and absorbed by duodenal enterocytes. Once inside enterocytes, Fe⁺² may be oxidized and stored as ferritin or exported to plasma where it is oxidized and bound to transferrin (Tf) for transport to the tissues. Iron stored as ferritin is returned to the small intestine lumen when enterocytes are sloughed at the tip of the villus. Iron is incorporated into many proteins in all tissues, but most iron is utilized to synthesize hemoglobin in developing erythroid cells, where Fe⁺³ must be reduced to Fe⁺² during heme synthesis. Aged blood erythrocytes are phagocytized by macrophages. Hemoglobin is degraded and iron is either returned to plasma or stored in macrophages as ferritin and hemosiderin. Modification of Kaneko, 1964, with permission.

Although iron is generally stored in cells as ferritin, iron overloaded cells (macrophages in normal animals) contain another storage form of iron called hemosiderin, which represents partially degraded ferritin that forms after uptake by lysosomes (Anderson and Frazer, 2005; Ponka et al., 1998). Surprisingly little is known about how ferritin releases its iron in cells. It is possible that intracellular iron release requires catabolism in lysosomes (Ponka et al., 1998).

Free cytoplasmic ferritin molecules are water soluble and visible by electron microscopy, but they are not visible by light microscopy unless present in large aggregates. Hemosiderin is insoluble and visible by light microscopy. Hemosiderin appears gray to black in aspirate smears stained with routine bloodstains, golden-brown in H & E-stained tissue sections, and blue when stained with Prussian blue stain. There is a fairly good correlation between Prussian blue staining of tissues and the measurement of storage iron chemically, but staining is less sensitive in detecting storage iron (Blum and Zuber, 1975; Franken et al., 1981). Storage iron can be present as ferritin that is not identified by Prussian blue staining (Blum and Zuber, 1975).

Major sites of iron storage include the liver, spleen, and bone marrow. The amount stored in a given tissue may vary by species. The liver appears to be the major site of iron storage in humans, but the spleen is reported to be more important in horses and cattle (Blum and Zuber, 1975; Franken et al., 1981; Kolb, 1963). The relative contributions of ferritin and hemosiderin to the storage iron pool vary with species, organ, and amount of iron stored. In normal adults, ferritin concentration generally equals or exceeds hemosiderin concentration in the liver, but hemosiderin concentration typically exceeds ferritin concentration in the spleen. When iron stores are high, hemosiderin tends to accumulate more than ferritin, and when iron stores are low, ferritin tends to be higher than hemosiderin (Kolb, 1963; Underwood, 1977). Total body iron stores vary at birth between species, with considerable iron stores in humans, dogs, cattle, goats, horses, and rabbits; intermediate iron stores in cats and rodents; and minimal iron stores in pigs (Keen et al., 1981; Kolb, 1963; Underwood, 1977). Iron stores decrease in nursing animals when demands for growth exceed iron absorption from milk (Blum and Zuber, 1975; Harvey et al., 1987a; Keen et al., 1981; Kolb, 1963).

Myoglobin is an oxygen-binding protein located primarily in muscles. It contains one heme group per molecule and has a structure similar to that of hemoglobin monomers. Myoglobin serves as a local oxygen reservoir that can temporarily provide oxygen when blood oxygen delivery is insufficient during periods of intense muscular activity. Iron within the heme group must be in the Fe⁺² state to bind oxygen. If iron is oxidized to the Fe⁺³ state, metmyoglobin is formed. The total amount of myoglobin in an animal depends on body weight, degree of muscle development, and the myoglobin concentration in muscle, which varies between muscle types (red muscle is rich in myoglobin and white muscle is myoglobin poor). The myoglobin concentration in muscle varies with species. For example, the myoglobin concentration in muscles of racehorses is about 6 times that of human muscles (Kolb, 1963). Because much of the iron in muscle is in myoglobin, the percentage of total body iron present in muscle is higher in horses than in humans.

In addition to the need for iron in hemoglobin and myoglobin, iron is vital in various metabolic processes. Nearly half of the enzymes in the citric acid cycle either contain iron or need it as a cofactor. Heme-containing proteins include catalase, peroxidase, and cytochromes. Cytochromes a, b, and c are required for oxidative phosphorylation in mitochondria. Other cytochromes, such as cytochrome P450 located in the endoplasmic reticulum, function in oxidative degradation of endogenous compounds and drugs. Nonheme iron, in Fe-S compounds and metalloflavoproteins, is required for enzymes including succinate dehydrogenase, cytochrome c reductase, nicotinamide adenine dinucleotide dehydrogenase, xanthine oxidase, and aconitase (Smith, 1997).

FIGURE 9-3 Apotransferrin is a bilobar protein with two binding sites for Fe⁺³ ions. When one or two Fe⁺³ ions are bound, the protein is referred to as monoferric or diferric transferrin, respectively. There is a random distribution of iron on binding sites, with apotransferrin predominant at low plasma iron concentrations and diferric transferrin predominant at high plasma iron concentrations. Monoferric transferrin is predominant at 50% saturation of transferrin with iron, with lesser amounts of apotransferrin and diferric transferrin present.

FIGURE 9-4 Mechanisms of iron absorption. Ferrous iron (Fe⁺²) ions are transported into enterocytes in the duodenum by the divalent metaltransporter-1 (DMT1) after reduction of ferric iron (Fe⁺³) ions using a duodenal cytochrome b (DctyB). Heme is transported into enterocytes using heme carrier protein-1 (HCP1). Once inside, inorganic iron is released from heme by the action of the heme oxygenase (HO) reaction. Fe⁺² ions are exported from enterocytes using ferroportin, oxidized to Fe⁺³ using hephaestin, and bound by apotransferrin (aTf) to form monoferric transferrin (mTf) and diferric transferrin (not shown). Hepcidin in plasma inhibits iron export to plasma by interacting directly with ferroportin, leading to ferroportin’s internalization and lysosomal degradation. Fe⁺² not transported to plasma is stored as ferritin following oxidation to Fe⁺³. Iron stored as ferritin is returned to the small intestine lumen when enterocytes are sloughed at the tip of the villus.

Ferritin is another iron-binding protein that can be measured in plasma. Although ferritin is released in small amounts from the liver and macrophages and can be taken up by cells (especially hepatocytes), it is of little or no importance in iron transport under normal conditions. Ferritin is normally present in very low concentrations in plasma and has low iron content; consequently, it contributes little to the plasma iron pool (Ponka et al., 1998).

Nontransferrin bound iron (NTBI) has been measured in people with severe iron overload, when transferrin is fully saturated, and in mice with hereditary hypotransferrinemia. The nature of NTBI is unclear, but it may be complexed with low-molecular-weight molecules such as citrate, sugars, and amino acids or nonspecifically bound to albumin and other plasma proteins. NTBI has the potential to promote the formation of oxygen-free radicals and cause tissue injury (Ponka et al., 1998).

Heme is released from dietary myoglobin and hemoglobin by the action of digestive enzymes. Dietary heme iron is generally more bioavailable than is nonheme iron and is an important nutritional source of iron in carnivores and omnivores. Heme enters duodenal enterocytes as an intact metalloporphyrin using a membrane protein named heme carrier protein 1 (HCP1) that has homology to bacterial metal-tetracycline transporters (Shayeghi et al., 2005). The expression of HCP1 is regulated pre- and posttranslationally in hypoxic and iron-deficient mice, respectively (Latunde-Dada et al., 2006b). Once inside the enterocyte, inorganic iron is apparently released from heme by the action of the heme oxygenase reaction, and this heme-splitting reaction appears to be the rate-limiting step in the absorption of iron contained within hemoglobin and myoglobin (Wheby and Spyker, 1981).

The intracellular transport of iron within the enterocyte is poorly understood. Iron ions in this labile iron pool (LIP) are presumably bound to chaperone molecules to keep them soluble. The nature of these chaperone molecules is yet to be defined. Iron ions can catalyze oxidative reactions that would injure the cell. The enterocyte protects itself against the toxic effects of excess iron by increasing apoferritin synthesis and incorporating the excess iron into ferritin. Consequently, iron uptake into enterocytes in excess of that needed for metabolic purposes or transferred to plasma is stored as ferritin. Iron stored as ferritin is returned to the small intestine when enterocytes are sloughed at the tip of the villus after 1 to 2 days (Steele et al., 2005).

The transfer of iron atoms from enterocytes to transferrin in plasma is mediated by ferroportin (also known as Ireg1), an iron transport protein located on the basolateral surface of mature enterocytes. In addition to ferroportin, the efflux of iron from enterocytes requires a copper-containing protein called hephaestin that is also located on the basolateral membranes of mature enterocytes. Hephaestin is a membrane-bound ferroxidase that has significant homology to the plasma protein ceruloplasmin. Hephaestin’s function may relate to its ability to oxidize Fe⁺² ions to Fe⁺³ ions for binding to transferrin in plasma (Steele et al., 2005). Sex-linked anemia (sla) mice have a block in mucosal iron transfer to plasma, with resultant microcytic hypochromic anemia because these animals have an inherited hephaestin defect (Wessling-Resnick, 2006).

Components of brush border iron uptake, including DMT1 and DcytB, are strongly influenced by the iron concentration within enterocytes, with increased components expressed when intracellular iron content is low and decreased components expressed when iron content is high (Frazer and Anderson, 2005). These locally responsive changes in brush border transport components help buffer the body against the absorption of excessive iron, but it is the control of the basolateral transport of iron from enterocytes to plasma that represents the primary site at which iron absorption is controlled (Steele et al., 2005). Hepcidin, a peptide produced by hepatocytes and secreted into plasma, inhibits iron transfer from enterocytes to plasma by interacting directly with ferroportin, leading to the internalization and lysosomal degradation of this iron export protein. Hepcidin production is increased in disorders such as iron overload and inflammation that result in decreased intestinal iron absorption. Hepcidin production is decreased in iron deficiency and disorders with increased erythropoiesis that result in increased iron absorption by enterocytes. Hephaestin expression is minimally affected by iron status (Vokurka et al., 2006).

At physiological pH and oxygen tension, Fe⁺² in solution is readily oxidized to Fe⁺³, which is prone to hydrolysis and precipitation. In addition, unless appropriately chelated, iron in plasma can promote harmful oxygen radical formation and peroxidative tissue damage because of its catalytic action in one-electron redox reactions (Ponka et al., 1998). Fortunately, most iron in plasma is bound to the protein apotransferrin to form transferrin. The binding of iron to apotransferrin keeps iron molecules soluble and prevents iron catalyzed oxidative reactions. In addition, the vast majority of iron is transported to cells within the body following binding to apotransferrin in plasma (Ponka et al., 1998). At normal iron saturation levels, apotransferrin is most abundant in plasma, followed by monoferric transferrin and diferric transferrin (Ponka et al., 1998). Although diferric transferrin is normally much lower in concentration (about 10% of total transferrin) than monoferric transferrin, it accounts for most of the iron delivery to cells because diferric transferrin binds with 8- to 10-fold higher affinity to transferrin receptor-1 (TfR1) receptors on cells than does monoferric transferrin (Anderson and Frazer, 2005; Huebers et al., 1985). Apotransferrin has very low affinity for TfR1 receptors on cells (Ponka et al., 1998).

1. Cytoplasmic Regulation of Iron

The synthesis of a number of proteins important in iron metabolism, including apotransferrin, TfR1, apoferritin, DMT1, DcytB, ferroportin, iron responsive protein-1 (IRP1), and erythroid-specific 5-aminolevulinic acid synthase (eALAS or ALAS2), is regulated posttranscriptionally depending on intracellular iron content (Beutler, 2006b; Koury and Ponka, 2004; Latunde-Dada et al., 2006a). The mRNA for each protein contains one or more iron responsive elements (IREs), each consisting of a stem-loop structure. The IREs located at the 5′ end of mRNAs regulate translation, and IREs located at the 3′ end regulate mRNA stability. IRP1 contains an Fe-S (4Fe-4S) cluster and exhibits aconitase activity when cells are iron replete, but it lacks the Fe-S cluster and aconitase activity when cytoplasmic iron is scarce. IRP1 binds tightly to IREs when cytoplasmic iron content is low. IRP2 is closely related to IRP1, but it lacks aconitase activity. The regulation of IRP2 is mediated by proteosomal degradation when cellular iron is adequate, through binding to iron and possibly heme (Wingert et al., 2005). The binding of IRPs to IREs at the 5′ end of mRNAs (including apoferritin and eALAS) inhibits translation and protein synthesis from these mRNAs, but the binding of IRPs to IREs at the 3′ end of mRNAs (including TfR1 and intestinal DMT1) promotes mRNA stability, thereby enhancing protein synthesis from these mRNAs. When cytoplasmic iron content is high, IRPs are displaced from IREs, resulting in opposite effects on protein synthesis (Starzynski et al., 2004). Because of these controlling factors, TfR1 synthesis is higher and apoferritin synthesis is lower when cytoplasmic iron content is low, and TfR1 synthesis is lower and apoferritin synthesis is higher when cytoplasmic iron content is high (Napier et al., 2005).

2. Mitochondrial Iron Metabolism and Regulation

FIGURE 9-5 Iron metabolism in erythroid cells. Diferric transferrin (dTf) binds to transferrin receptor-1 (TfR1) on the surface of developing erythroid cells. The dTf-TfR1 complexes are internalized within an endosome and a proton pump decreases the pH in the endosome, resulting in release of iron ions (dark spheres). Iron is exported from the endosome using the divalent metal transporter-1 (DMT1). The resultant apotransferrin (aTf)-TfR1 complex is recycled to the cell membrane, where aTf is released from the cell, and the TfR1 is available for binding additional dTf. A direct interorganelle transfer of iron occurs between endosomes and mitochondria for insertion into protoporphyrin IX (PP) to form heme and for synthesis of iron-sulfur clusters (not shown). Some iron presumably exists in a cytoplasmic labile iron pool (LIP), with excess iron stored as ferritin.

A direct interorganelle transfer of iron occurs between endosomes and mitochondria where iron is used for heme and iron-sulfur cluster synthesis (Sheftel et al., 2007). Some iron is presumably released from endosomes into a cytoplasmic labile iron pool (LIP), with excess cytoplasmic iron stored as ferritin. Various chaperone molecules have been proposed, but the nature of the iron in the LIP remains enigmatic (Beutler, 2006b). Iron stored as ferritin in the cytoplasm is not available for hemoglobin synthesis in erythroid cells (Ponka et al., 1998).

Even though three different pathways are required for hemoglobin synthesis in erythrocyte precursors and reticulocytes, virtually no intermediates (iron, globin chains, or heme) accumulate in the cytoplasm of these cells. Several positive and negative feedback mechanisms account for the balanced production of these hemoglobin components. As already discussed, an increase in the LIP limits the uptake of additional iron by decreasing TfR1 synthesis. The availability of iron also limits, and thereby controls, heme synthesis. Free “uncommitted” heme inhibits iron uptake by erythroid cells and consequently heme synthesis. In addition, free heme is essential for the synthesis of globin chains at both the transcriptional and translational levels (Koury and Ponka, 2004). Consequently, globin synthesis does not occur in the absence of heme.

A heme exporter termed FLVCR is up-regulated on colony-forming units-erythroid (CFU-E) progenitor cells. It may provide a safety mechanism to prevent the accumulation of toxic amounts of cytoplasmic heme before globin synthesis is initiated. FLVCR is the cell surface receptor for feline leukemia virus, subgroup C (FeLV-C). Cats with FeLV-C infections develop erythroid aplasia because of a block at the CFU-E stage of erythroid development. It appears that the binding of FeLV-C to receptors on CFU-E progenitor cells inhibits heme export and results in apoptosis of these cells (Quigley et al., 2004).

In addition to the entry of iron into mitochondria for incorporation into protoporphyrin IX to form heme, iron is critical for Fe-S cluster biogenesis within mitochondria. Fe-S clusters are important prosthetic groups for numerous proteins involved in electron transfer, metabolic, and regulatory processes. Although Fe-S clusters are formed within mitochondria, Fe-S proteins are located in the nucleus and cytoplasm of cells, as well as in mitochondria (Lill et al., 2006). Recent studies in zebra fish provide evidence for a regulatory link between Fe-S cluster formation and heme synthesis. In the absence of Fe-S clusters, IRP1 binds to eALAS mRNA, which inhibits eALAS synthesis and heme production (Wingert et al., 2005).

FIGURE 9-6 Iron metabolism in macrophages. Nearly all iron enters macrophages by the phagocytosis of aged or prematurely damaged erythrocytes. Following phagocytosis, erythrocytes are lysed, and hemoglobin is degraded to heme and globin. The microsomal heme oxygenase reaction within macrophages degrades heme and releases iron to the labile iron pool (LIP). Most of the released iron is exported from the macrophage by ferroportin as ferrous iron, oxidized to ferric iron by ceruloplasmin (Cp) in plasma, and bound to apotransferrin (aTf) to form monoferric transferrin (mTf) or diferric transferrin (not shown). Hepcidin in plasma inhibits iron export by interacting directly with ferroportin, leading to ferroportin’s internalization and lysosomal degradation. Iron not rapidly released to plasma is stored within macrophages as ferritin, which may be degraded to hemosiderin within lysosomes.

The liver is essential for normal iron homeostasis in the body. It regulates iron movement into and around the body through hepcidin synthesis, accounts for about 40% of the body iron stores, and is the site of synthesis of apotransferrin, ceruloplasmin, haptoglobin, and hemopexin plasma proteins (Anderson and Frazer, 2005).

Hepatocytes synthesize hepcidin in an endocrine manner to regulate iron metabolism in the body. Hepcidin is a small cationic peptide (25 amino acids in humans) that forms from a larger prohepcidin peptide (84 amino acids) (Park et al., 2001). Hepcidin has been cloned, expressed, and sequenced in several species, including dogs (Fry et al., 2004; Verga Falzacappa and Muckenthaler, 2005). Because of its small size, hepcidin is rapidly cleared by the kidney and is measurable in urine. Using immunohistochemistry techniques, prohepcidin is localized to organelles of the secretory pathway, particularly the Golgi apparatus, of hepatocytes suggesting that molecules may accumulate before receiving a signal for secretion (Wallace et al., 2005). Prohepcidin circulates in plasma, but its plasma concentration does not correlate well with urinary hepcidin concentrations or other iron parameters (Kemna et al., 2005). The site, nature, and potential regulation of the conversion of prohepcidin to hepcidin remain to be identified.

A number of systemic stimuli, including body iron stores, anemia, hypoxia, degree of erythropoiesis, and inflammation have been reported to modulate hepcidin levels; however, the pathways involved in hepcidin production remain to be elucidated (Anderson and Frazer, 2005; Steele et al., 2005). Anemia, hypoxia, and erythropoietin themselves do not seem to be of primary importance (Pak et al., 2006; Vokurka et al., 2006). Hepcidin production is modulated by body iron requirements, which are largely influenced by the magnitude of erythropoiesis present (Wilkins et al., 2006). A signal arising from erythropoietic activity in bone marrow is proposed to regulate hepcidin production, but its nature is unknown at the time of this manuscript preparation (Pak et al., 2006). The content of diferric transferrin in plasma may be a key indicator of body iron requirement (Wilkins et al., 2006). Many questions remain concerning how hepcidin expression is modulated in response to body iron requirements; however, three molecules (hemochromatosis [HFE] protein, TfR2, and hemojuvelin) have been identified that are involved in the regulation of this pathway(s). Defects in each of these molecules in humans or mice have resulted in inappropriately low hepcidin levels and increased iron absorption leading to hemochromatosis. It is hypothesized that hepcidin is regulated by HFE and TfR2 on hepatocyte surfaces in response to plasma transferrin saturation. TfR2 has a 25-fold lower affinity for diferric transferrin than does TfR1 and would presumably become active when diferric transferrin concentrations are high (Anderson and Frazer, 2005). The pathway involving plasma diferric transferrin concentration may not be the only way hepcidin production is regulated. Hemojuvelin may modulate this pathway or affect hepcidin synthesis independently (Steele et al., 2005). Studies indicate that hepcidin synthesis is stimulated by certain bone morphogenetic proteins (BMPs) that bind to hemojuvelin, regulating hepcidin expression through a BMP signal transduction pathway (Truksa et al., 2006). In addition, hepcidin expression increases in response to inflammation, and this increase precedes any alteration in transferrin saturation (Steele et al., 2005).

Most of the iron stored within the liver is found in hepatocytes stored as ferritin. Hepatocytes secrete a glycosylated form of ferritin into plasma (Ghosh et al., 2004). Ferritin in human plasma is primarily composed of L subunits, but H subunits predominate in ferritin in dog plasma (Watanabe et al., 2000). Plasma ferritin concentrations generally correlate with body iron stores; however, ferritin is an acute phase protein that increases in plasma during inflammation (Torti and Torti, 2002).

Hepatocyte membranes contain ferritin receptors. However, serum ferritin is present in low concentrations and contains little iron, so it is unlikely that hepatocytes take up significant amounts of iron from plasma by this mechanism under normal conditions (Anderson and Frazer, 2005; Ponka et al., 1998). Like other cells, hepatocytes utilize TfR1 to transport iron into the cells. NTBI is thought to be chelated by small organic acids, such as citrate, but some may be bound loosely to proteins such as albumin (Anderson and Frazer, 2005). The liver rapidly clears NTBI. In experimental studies using mice, the half-time of clearance of NTBI in plasma was 30 sec, compared to a half-time clearance for transferrin-bound iron of 50 min (Craven et al., 1987). Like macrophages and enterocytes, hepatocytes utilize ferroportin to export iron to plasma. Ferroportin expression is lower in hepatocytes than in enterocytes and macrophages, which may help explain why these cells preferentially accumulate iron in most iron-overload conditions (Rivera et al., 2005).

Ceruloplasmin is a glycoprotein with a molecular weight of 100 to 155 kDa depending on species. It migrates in the α₂ region in humans but in the α₁-region in horses on protein electrophoresis (Okumura et al., 1991). During biosynthesis, six atoms of copper are incorporated into ceruloplasmin late in the secretory pathway (Hellman and Gitlin, 2002). It contains most of the copper in the circulation in all species studied, except for the dog in which ceruloplasmin accounts for only about 40% of the plasma copper content (Montaser et al., 1992). It plays no role in copper transport or delivery to tissues (Hellman and Gitlin, 2002). Ceruloplasmin has ferroxidase enzyme activity that facilitates the oxidation of Fe⁺² to Fe⁺³, a process involved in iron mobilization from liver and other tissue, but not from enterocytes (Osaki et al., 1971; Wessling-Resnick, 2006). It also appears to function as an antioxidant in plasma. Ceruloplasmin is a mild to moderate acute phase protein that increases in concentration in association with inflammation (Ceron et al., 2005; Smith and Cipriano, 1987). Its synthesis may also be enhanced by iron deficiency, estrogen, and progesterone. In contrast to iron, hepatocytes can excrete copper in the bile by a process that is dependent on the intracellular concentration of copper (Hellman and Gitlin, 2002).

When the hemoglobin concentration reaches a certain level in developing erythroid cells, it appears to signal the cessation of cell division. Abnormalities in heme or globin synthesis result in deficient hemoglobin synthesis and a delay in the signal for cell division to cease. When this happens, one or more extra cell divisions occur during erythroid cell development, resulting in the formation of microcytic erythrocytes (Stohlman et al., 1963). A deficiency in hemoglobin synthesis can also result in the formation of hypochromic erythrocytes with decreased hemoglobin concentration. Iron is essential for heme synthesis; consequently, iron deficiency results in deficient heme synthesis and the formation of microcytic, hypochromic erythrocytes. In addition to true iron deficiency, disorders that result in functional iron deficiency or defective iron incorporation into heme in mitochondria may result in the formation of microcytic, and possibly hypochromic, erythrocytes. These disorders include the anemia of inflammatory disease, copper deficiency, myelodysplastic disorders, drug or chemical toxicities, and possibly portosystemic shunts (Harvey, 2000).

The mean cell volume (MCV) represents the average volume of a single erythrocyte in femtoliters (fl = 10^–15 l). The MCV is determined most accurately with appropriately calibrated electronic cell counters that determine the size of individual cells and compute the MCV. The MCV varies greatly depending on species and, in some cases, depending on a breed within a species. Some dogs from Japanese breeds (Akita and Shiba) normally have MCV values below the reference intervals established for other breeds of dogs (Gookin et al., 1998), but these dogs are not anemic. The MCV is a fairly insensitive indicator of the formation of microcytes because a relatively high percentage of microcytes must generally be present in blood for the MCV to decrease below the reference interval.

The mean cell hemoglobin (MCH) represents the average amount of hemoglobin in a single erythrocyte. It is calculated by dividing the Hb value (in g/dl) by the RBC count (in millions per μl) and multiplying by 10. The MCH generally provides little added information beyond that obtained from the MCV and MCHC because the MCH depends on both the size and hemoglobin concentration of erythrocytes. It usually correlates directly with the MCV, except in animals with macrocytic hypochromic erythrocytes. Exceptionally low MCH values strongly suggest true iron deficiency is present.

Although quite useful when abnormal, MCV and MCHC values are relatively insensitive in identifying the presence of erythrocytes with abnormal volumes or hemoglobin concentrations. Many microcytic or macrocytic erythrocytes are required to move the MCV below or above the reference interval, respectively, and many hypochromic erythrocytes are needed to move the MCHC below the reference interval. In addition to counting cells, electronic cell counters can determine and plot the volume of individual erythrocytes, and examination of these erythrocyte volume histograms can reveal the presence of increased numbers of microcytes or macrocytes even when the MCV is within the reference interval (Weiser and Kociba, 1983). Some electronic cell counters such as the Advia 120 (Siemens Medical Solutions Diagnostics, Tarrytown, NY) can also determine the hemoglobin concentration of individual erythrocytes from the deflection of light that occurs when a laser beam strikes individual cells. Inspection of hemoglobin concentration histograms can reveal the presence of increased numbers of hypochromic erythrocytes, even when MCHC has not decreased below the reference interval. Individual erythrocytes can be further characterized by creating a cytogram in which the erythrocyte volumes of individual cells are plotted against their respective hemoglobin concentrations. In addition, the percentages of microcytes, macrocytes, hypochromic erythrocytes, hyperchromic erythrocytes, erythrocytes with low hemoglobin content, and erythrocytes with high hemoglobin content can be determined, and these same parameters can also be determined for reticulocytes in a blood sample. Because reticulocytes are recently formed immature erythrocytes, reticulocyte parameters may best reflect the current state of iron sufficiency in an animal. Alterations in several reticulocyte indices—including reticulocyte MCV (MCV_retic), reticulocyte hemoglobin content (CH_retic), percentage of hypochromic reticulocytes (% Hypo_retic), percentage of reticulocytes with low hemoglobin content (% Low CH_retic), percentage of reticulocytes with high hemoglobin content (% High CH_retic), and percentage of macrocytic reticulocytes (% Macro_retic)—appear to be of value in the diagnosis of iron deficiency in dogs (Fry and Kirk, 2006; Steinberg and Olver, 2005).