Iron Absorption In Mammals Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Iron is essential to life because of its unique ability to maintain the function of key metabolic enzymes and the production of red blood cells Johnson et al., 2005. However, the excess accumulation of iron can be toxic as it has the ability to donate and accept electrons. Iron can catalyse the formation of hydroxyl free radicals, via the Fenton reaction. The generation of free radicals can cause a wide spread of damage to various cellular components and thus induce apoptosis. Therefore the body iron levels are regulated at various stages of iron absorption, in order to prevent such disastrous events from occurring.

Iron absorption in mammals occurs within the duodenal enterocytes. Dietary iron, ferric iron (Fe3+) is reduced by a ferric reductase called duodenal cytochrome b (Dcytb) to ferrous iron (Fe2+). Dcytb is found highly expressed in the duodenal apical membrane (Mckie et al., 2008). The ferrous iron is transported from the intestinal lumen and into the cytoplasm of the enterocytes by a divalent metal transporter 1 (DMT1). DMT1 is a Fe2+ - H+ co-transporter, which is also found highly expressed at the apical membrane (Núñez et al., 2009). Once Fe2+ is inside the cytoplasm, it coordinates with a liable iron pool (LIP) (Núñez et al., 2009). LIP is a pool of weakly bound iron but the nature of LIP is still unknown. From LIP iron can either be sequestered within ferritin or transported out of the enterocyte by ferroportin (FPN). Under normal iron levels FPN is found highly expressed at the basolateral membrane of enterocytes, where is meditates the exportation of iron from the cells. Iron (Fe2+) is then oxidised by hephaestin, thus allowing Fe3+ to bind to transferrin within the circulation where it is distributed through out the body (Collin et al., 2005). Hephaestin is highly homologous to ceruloplasmin, which is a plasma copper containing protein (Anderson et al., 2002). However, unlike ceruloplasmin, hephaestin is an "integral membrane protein with a single a single membrane spanning domain" (Anderson et al., 2002). The expression of hephaestin is limited to a certain number of tissues. It is weakly expressed in tissues such as the placenta, lung, spleen and brain (Anderson et al., 2002). Where as hephaestin is highly present through out the small intestine and the colon. Interestingly hephaestin expression is restricted to differentiated enterocytes of the villus (Frazer et al., 2001), as its function is limited to differentiated enterocytes this emphasis its role in intestinal iron transport. However, unlike other iron transport proteins hephaestin expression is not restricted to the proximal region of the small intestine, where iron absorption occurs (Anderson et al., 2002), but it is found expressed through out the duodenum.

The absorption of dietary iron is regulated by hepcidin. This is a peptide hormone which is secreted from the liver and controls the plasma iron levels. Secretion of hepcidin from the liver is decreased during anaemia and conversely increased when iron levels are high. Nemeth et al. (2004) revealed that hepcidin prevents the exportation of iron by directly binding to FPN and consequently removing FPN from the basolateral membrane. Thus leading to its internalisation and degradation and hence eradicating iron transport (Nemeth et al., 2004). Hepcidin was also found to decrease iron uptake by down regulating DMT1 mRNA and protein expression (Mena et al., 2007).

Previous work has found FPN to be located at both the apical and basolateral membrane (Thomas et al., 2004). Thomas et al. (2004) proposed that FPN possibly modulates the activity and the expression of DMT1 at the apical membrane, but itself does not function as an uptake transporter as this is the role of DMT1. In contrast recent work has found DMT1 present at both the apical and basolateral membranes (Núñez et al., 2009). During iron deprived conditions DMT1 is primarily located at the apical membrane, however during iron overloaded conditions DMT1 is internalised within the cytoplasm (Ma et al., 2002). This consequently limits the uptake of iron.

Previous studies have identified novel genes involved in iron absorption by depriving rat duodenum of dietary iron (Collin et al., 2005). The following genes were significantly up-regulated during iron-deprivation; DMT1, Dcytb, heme oxygenase 1 (HO1) and transferrin receptor 1 (TFR1). DMT1 contains two 3' splice variants, either with or without iron responsive element (IRE). During iron-deprived conditions DMT1+IRE was strongly induced. However genes encoding the basolateral proteins, IREG (iron regulated gene 1, also known as ferroportin) and hephaestin were not significantly induced. Interestingly other genes such as Menkes copper ATPase (ATP7a) and sodium-dependent vitamin C transporter were induced in anaemic rats ands iron-deprived rats. Thus, suggesting its possible role in intestinal iron absorption (Collin et al., 2005).

The intestinal epithelium cells contain a unique steep oxygen gradient compared to other cells within the body (Taylor et al., 2007). This hypoxic state of the epithelium cells activates Hypoxia-Inducible transcription factors (HIF). Intestinal HIF have recently been found to play an essential role in iron absorption during iron deficiency. HIF is a nuclear transcription factors which consist of 2 helix-loop-helix proteins; α and β subunits. There are three members of the HIF family (HIF-1, HIF-2 and HIF-3) and are all αβ heterodimers (Benita et al., 2009). The α subunit is the oxygen sensitive component and the β subunit is ubiquitously expressed, also known as the aryl hydrocarbon receptor nuclear translocator (ARNT) (Mastrogiannaki et al., 2009). So far only three α subunits have been characterised; HIF-1α, HIF-2α and HIF-3α (Mastrogiannaki et al., 2009). HIF-2α is up-regulated in specific tissues, whereas HIF-1α is extensively expressed in vivo (Wiesener et al., 2002). Recently Mastrogiannaki et al. (2009) discovered that HIF-2α but not HIF-1α is involved in "maintaining iron balance" by regulating the transcription of DMT1 gene (Mastrogiannaki et al., 2009). Under normoxia conditions the α subunits are hydroxylated by an iron-dependent prolyl hyroxylase (PHD) (Shah et al., 2009). The α subunit is then ubiquitinated by von Hippel-Lindau (VHL) tumour suppressor protein (E3 ligase) and thus degraded via the proteosomal pathway (Shah et al., 2009). In contrast, under hypoxia (iron depletion) hydroxylation of the α subunit is inhibited and therefore forms a stabilised subunit. This leads to the heterodimerization of the α subunit with ARNT and thus forms the active HIF. This active HIF translocates to the nuclease and activates HIF target genes (Mastrogiannaki et al., 2009). The genes which HIF activates are involved in iron metabolism, apoptosis and angiogenesis (Peyssonnaux et al., 2007).

Peyssonnaux et al. (2007) demonstrated that HIF regulates hepcidin expression. They deleted hepatic VHL gene which consequently led to an increase in HIF-1α activity. This resulted in the suppression of hepcidin, increased levels of ferroportin and increased levels of iron in serum. Peyssonnaux and colleagues also showed that the human hepcidin promoter sequence contained 3 candidate consensus HREs (Hypoxia response elements) and these were targets of repression by HIF-1α (Peyssonnaux et al., 2007). Thus establishing the role of HIF-1α to represses the expression of the hepcidin gene.

There are three ankyrin isoforms found in humans; ankyrin-1 (found in erythrocytes), ankyrin-2 (predominately expressed in the brain) and ankyrin-3 (widely expressed). There are also many other alternative splice variants of ankyrin, which are found to be tissue specific. Ankyrin is composed of 4 domains; the N-terminal which is contains the ankyrin repeats, the spectrin-binding domain, the death domain and the C-terminal regulatory domain. A sub-domain called ZU5-ANK is present within the spectrin-binding domain and it is essential for the interaction between spectrin and ankyrin. ZU5-ANK has found to recognise repeats 14 and 15 of β-spectrin (Ipsaro et al., 2010). The function of death domain is unknown. The regulatory domain is responsible for modulating the affinities of molecules. There are a range of membrane proteins which ankyrin associates with, for example, E-cadherin, Na+/K+ ATPase, CD44, Na+/Ca2+ exchangers, etc (Bennett et al., 2007).

Spectrin is composed of α-subunits and β-subunits, forming α/β heterodimer. Two of these heterodimers are assembled into a tetramer (Bennett et al., 2007). In humans there are 2 α-subunits and 5 β-subunits of spectrin. Both subunits predominately contain spectrin repeats.

The binding of ankyrin and spectrin was first discovered in erythrocytes. Spectrin was found to be associated with the erythrocyte membrane through its association with ankyrin via the ZU5-ANK sub-domain. Ankyrin through its ankyrin repeat domain linked to the cytoplasmic domain of the membrane protein. In this case the membrane protein was a chloride-bicarbonate exchanger (Bennett et al., 2007).

Ankyrin and spectrin have been linked to a disease called Hereditary Spherocytosis (HS). HS is an inherited hemolytic anaemia in humans and in mice (Bennett et al., 2007). HS is a result of an abnormal expression of one of the following erythroid cytoskeletal proteins; α-spectrin, β-spectrin, ankyrin, band 3 and protein 4.2 (Rank et al., 2009). Mutations in ankyrin-1 are the most common mutations present in the majority of human HS cases (Bennett et al., 2007). The mutation in ankyrin-1 results in the RBC to become spherical in shape rather than their normal "donut-shape." This in turn results in the fragmentation of these cells.

Recent experimental data has revealed ANKRD37 as a "novel HIF-1-target gene" (Benita et al., 2009). ANKRD37 was shown to be robustly induced in hypoxia. ANKRD37 is a small ankyrin repeat protein consisting of 158 residues (appendix figure 2) and a molecular weight of 16.872 kDa (kilo-daltons). However ANKRD37 function is unknown and is found to be conserved in mammals and zebrafish (Benita et al., 2009). Benita et al. identified four HIF-1 binding sites on the ANKRD37 gene (477 base pairs; see appendix figure 1). ANKRD37 protein contains four conserved ankyrin repeat domains (Jixi Li et al., 2005). Ankyrin repeats are one of the most common protein motifs and have been discovered in a number of proteins, such as cyclin dependent kinase inhibitors, transcriptional complexes and many more (Jixi Li et al., 2005). These domains mediate protein-protein interactions and are also found to be involved in signal transduction, signal regulation pathways and protein sorting (Jixi Li et al., 2005). Ankyrin repeats approximately consist of 30 to 34 residues (Junan Li et al., 2006) of which 15 residues are highly conserved (Jixi Li et al., 2005). Each motif contains a helix-turn-helix conformation and this allows the formation of "hairpin like β-sheets with neighbouring loops" (Junan Li et al., 2006). Even though ankyrin repeat proteins contain sequence similarity their proteins targets are specific (Junan Li et al., 2006). Previous work has identified that the ANKRD37 gene is mainly expressed in the small intestines, colon, human testis and blood leukocytes (Jixi Li et al., 2005).

In this study I worked with ANKRD37 and RIK genes (another iron regulated gene). Both ANKRD37 and RIK were tagged with GFP at their C-terminal. The encoded GFP protein consists of 238 amino acids and contains a molecular weight of 27kDa (Stearns et al., 1995). GFP comes from jelly fish Aequorea Victoria and displays a bright green fluorescence when exposed to blue light. GFP has been used as a reporter molecule to monitor the localisation of secretary proteins (Kaether et al., 1995), locomotor proteins (Gerisch et al., 1995) and cell transfections (Stearns et al., 1995). Previously GFP has been targeted to multiple organelles and been used to supervise the diffusion and location of membrane proteins (Barak et al., 1997).

. Previous work has shown that during low iron concentrations, intestinal iron absorption involves the co-ordinated activities of DMT1 predominantly present at the apical membrane and FPN present at the basolateral membrane (Nunez et al., 2009). However, during high extracellular iron concentrations a relocation of DMT1 and FPN is observed. DMT1 during these conditions was found to be relocated to the cytoplasm and basolateral membrane and a significant reduction of apical membrane DMT1 was found.

These results show that the expression of ANKRD37 correlates with the expression patterns of DMT1, Dcytb and FPN during +Fe and -Fe conditions. As during -Fe conditions DMT1 and Dcytb are highly expressed around the apical membrane and FPN is highly present at the basolateral membrane. In contrast, during +Fe conditions DMT1 is re-localised to the cytoplasm but FPN expression does not significantly change at the basolateral membrane (Anderson et al., 2002).

A nuclear envelope surrounds the nuclease, which consists of two membranes; the inner nuclear membrane and the outer nuclear membrane. The outer nuclear membrane continues with the membrane of the rough endoplasmic reticulum (RER) (Imreh et al., 2003). RER is the site of protein synthesis and it also facilitates protein folding, so when the transfected cells begin to perform gene expression, ANKRD37 and Rik protein synthesis and folding into their native state occurs in the RER.

Generally, correctly folded proteins diffuse through the RER lumen and are selectively targeted to their proper destinations, for example the golgi complex (Imreh et al., 2003). However, since ANKRD37 and RIK were tagged with GFP, this greatly slows down the diffusion of the protein through the ER lumen and to their functional sites within the cell. Recent studies have shown that diffusion of GFP through the ER lumen is slow (Dayel et al., 1999). Dayel et al. (1999) showed that the GFP translational diffusion was 9 to 18 times slower in the ER lumen when compared to water and was also found to be 3 to 6 times slower in the cytoplasm.

Barisani et al. (2004) performed immunostaining on biopsies from the distal duodenum of patients with Celiac disease (CD) and iron deficient anaemia. This study showed that in iron deficient patients DMT1 staining was present through out the entire villus and the staining was stronger at the apical membrane. Also in CD patients DMT1 was "hyper-expressed" on the surface epithelium of the villi (Barisani et al., 2004). . It is possible that ANKRD37 through its ankyrin repeat domains may interact with an apical iron transporters like DMT1 or Dcytb however further work will be required to investigate this.

The duodenal tissue was not the only tissue found to express ANKRD37. Colon tissue sections subjected to -Fe treatment also showed a high ANKRD37 protein expression at the apical membrane. Where as in iron sufficient colon tissue sections, ANKRD37 was found widely distributed through out the tissue. Cecum tissue sections which also experienced iron adequate treatment showed similar results to the colon and duodenal tissue sections (Figure 5). DMT1 and other iron transporter such as Dcytb have been found at different levels of the lower gastrointestinal tract. Primarily they are highly expressed at the proximal regions of the small intestine and the colon (Anderson et al., 2002). As mentioned before ankyrin proteins are involved in binding to membrane proteins through their ankyrin repeat domains, so there is a possibility that ANKRD37 is associated with DMT1 or Dcytb.

Recent work has identified HIF-2α to up-regulate the expression of DMT1 during hypoxic conditions (Mastrogiannaki et al., 2009). Both RIK and ANKRD37 were found highly expressed through out the villlus, but were strongly expressed at the apical membrane of duodenal tissue sections (Figure 6), which had experienced 6 and 24 hours of hypoxia. As both DMT1 and Dcytb are highly localised at the apical membrane during hypoxia, this could suggest the association of RIK and ANKRD37 with these iron transporter molecules. Especially the association of RIK with Dcytb as mentioned before RIK is the proximal gene to Dcytb on chromosome 2. However, to some extent the proteins were expressed at the basolateral membrane, suggesting their possible involvement with FPN, which is found highly expressed at the basolateral membrane during hypoxic conditions (Núñez et al., 2009).

As mentioned before both the proteins are tagged with GFP and the diffusion of GFP within the ER lumen is very slow (Dayel et al., 1999), this could explain why there was such an intense GFP fluorescence received from around the nuclease in both sets of transfected cells. To confirm whether the GFP-tagged proteins are present within the ER lumen after 48 hours of gene-expression, an ER stain could be used. An example of a stain that could be used is ER-Tracker Blue-White DPX (Invitrogen). This is a highly photo-stable and selective stain for ER in live cells. So this stain could help identify the location of the GFP-tagged proteins within the ER. However, as the transfection images (Figure 8 and 9) show the proteins to be highly associated around the nuclear membrane, there is a possibility that these proteins are retained within the ER (Imreh et al., 2003) as they may be nuclear membrane proteins. Or these proteins, especially ANKRD37 as it contains ankyrin repeat domains may be associated with hephaestin, as it was found to be highly localised in a perinuclear distribution (Anderson et al., 2002).

. Previous studies have used Sex - linked anaemic (sla) mice which contain in-frame deletions in the hephaestin gene. This consequently reduces hephaestin's ferroxidase activity. These affected animal models were able to absorb dietary iron from the intestinal lumen across the brush border apical membrane normally, but the exportation of iron into the circulation was impaired (Anderson et al., 2002). As sla mice have difficulties in exporting iron, this suggests hephaestin is involved in transporting iron across the basolateral membrane. One might expect hephaestin to be present at the basolateral membrane. However, most hephaestin appears to be located in a "supranuclear site" (Anderson et al., 2002). This expression of hephaestin correlates with the expression of RIK-GFP and ANKRD37-GFP, which was observed in both iron deprived and overloaded conditions. It is possible that these proteins bind to hephaestin and assist hephaestin to ferrous iron. However, overall these results did not show that iron per se had a marked effect on the sub-cellular location of ANKRD37 and RIK proteins. As during both iron deficient and iron sufficient conditions the locations of the proteins were similar, there was no drastic change in expression observed. This could be due to the GFP tag on the proteins effecting its mobilisation within the cells. In future studies untagged ANKRD37 and RIK should be investigated or they should be tagged with another detectable protein.

In a recent study Rank et al., (2009) characterised the first null mutant mouse lines lacking the functional ankyrin-1 protein in erythrocytes. The mutation was induced by N-ethyl-N-nitrosourea (ENU). These mouse lines carried a mutation in exon 41 and this lead to the homozygous mutant mice lacking anykrin-1. This resulted in a reduced number of RBC and the mice were found to die immediately in their perinatal period. In contrast, the mice which survived developed sever Hereditary Spherocytosis (HS) and the level of spectrin in these mutant models was ~ 25% of wild type mice (Rank et al., 2009). The HS observed in these mice was more sever than the HS present in humans as the approximate levels of spectrin in the membranes of HS patients is ~ 40 to 50% of normal. Overall this study demonstrated that ankyrin-1 plays an essential role in the development of erythroid.

In order to confirm the role of these proteins in iron metabolism, a series of further experiments need to be conducted. In the future a densitometer should be used to quantify the concentration of the protein in either the transfected cells or the duodenal tissues subjected to different iron environments. This will allow us to confirm the difference in protein expression during iron adequate and deficient conditions. Another experiment which could be conducted could use antisense oligonucleotides for ANKRD37 and RIK in cells followed by measurement of radio-iron (59Fe) uptake. This would determine whether either protein plays a role in iron absorption. In addition further immunoprecipitation studies with anti RIK and ANKRD37 antibodies could be attempted using duodenal lysates form iron adequate and iron deficient animals, followed by Western blot for the iron transporters DMT1, Dcytb and FPN. This would identify whether either protein associates with iron transporters.