Partial purification and characterization of ferritin from the liver and intestinal mucosa of chickens, turtledoves and mynahs

Abstract

Ferritin is the iron-storage protein responsible for sequestering excess iron, to be stored in a safe way in the liver or to be shed with the intestinal epithelial cells. The properties of ferritin in iron-overload-susceptible birds have not been elucidated. Furthermore, there is only scarce information on mucosal ferritin, with no information at all in avian species. Here we have studied the liver and proximal intestine ferritins of iron-overload-susceptible (Indian hill mynahs, common mynahs) and non-susceptible (turtledoves, chicken) bird species. A brief purification process preceded native polyacrylamide gel electrophoresis and staining the gels for protein and iron. Protein amounts and iron-binding characteristics of ferritin were measured and ferritin saturation levels were calculated. Although ferritin protein amounts did not differ significantly, liver and mucosal ferritins of sensitive bird species incorporated much more iron, leading to high saturation levels. Significantly higher ferritin iron content and saturation were observed in the liver of both mynah species and in the intestinal ferritin of Indian hill mynahs when compared with the non-susceptible species. Ferritin appears not to play a major role in the regulation of iron absorption, implicating other phases in iron transport to be more important in the onset and process of iron overload in birds.

Purification partielle et caractérisation de la ferritine du foie et de la muqueuse intestinale chez le poulet, la tourterelle (Streptopelia d decaocto) et le mainate (Acridotheres t. Tristis et Gracula relidiosa)

La ferritine est la protéine de stockage du fer, permettant de piéger le fer en excès, afin de le stocker sans risque dans le foie ou de le répandre au niveau des cellules épithéliales intestinales. Les propriétés de la ferritine chez les poulets présentant une surcharge en fer n'ont pas été élucidées. De plus, il n'existe que quelques informations sur la ferritine muqueuse, et aucune information sur les espèces aviaires. Nous avons étudié les ferritines du foie et de l'intestin proximal chez les espèces aviaires sensibles à une surcharge en fer (mainates religieux indiens, martins tristes) et les espèces non sensibles (tourterelles, poulets). Après une purification par une méthode rapide, une électrophorèse en gel de polyacrylamide a été réalisée suivie d'une coloration des gels pour la protéine et le fer. Les quantités de protéine et les caractéristiques du fer de la ferritine ont été mesurées et les niveaux de saturation de la ferritine ont été calculés. Bien que les quantités de la protéine ferritine n'eussent pas différées significativement, les ferritines du foie et des muqueuses des espèces aviaires sensibles ont incorporé beaucoup plus de fer conduisant à des niveaux élevés de saturation. Le contenu en fer de la ferritine et sa saturation, ont été significativement supérieurs dans le foie des deux espèces de mainates, de même pour la ferritine intestinale des mainates religieux indiens comparés aux espèces non sensibles. La ferritine paraît ne pas jouer un rôle majeur dans la régulation de l'absorption du fer, impliquant d'autres phases dans le transport du fer, pour être plus importante au début et transformer la surcharge de fer chez les oiseaux.

Partielle Reinigung und Charakterisierung von Ferritin aus der Leber und Darmschleimhaut von Huhn, Turteltaube(Streptopelia d. decaocto) und Maina (Acridotheres t. Tristis und Gracula religiosa)

Ferritin ist ein Eisen-Speicherprotein, das für die Aussonderung von überschüssigem Eisen verantwortlich ist, um es sicher in der Leber zu speichern oder es über die Darmepithelzellen auszuscheiden. Die Eigenschaften von Ferritin in für eine Eisenüberladung empfänglichen Vögeln sind bislang nicht geklärt. Darüber hinaus gibt es nur spärliche beziehungsweise beim Vogel überhaupt keine Informationen über Mukosa-Ferritin. Hier haben wir Ferritine aus Leber und proximalem Darm von für Eisenüberladung empfängliche (indischer Hügelmaina (Gracuala religiosa intermedia), Hirtenmaina (Acridotheres tristis)) und unempfängliche (Turteltaube (Streptopelia turtur) und Huhn) Vogelspezies untersucht. Auf einen kurzen Reinigungsprozess folgte eine native Polyacrylamidgelelektrophorese und die Färbung des Gels auf Protein und Eisen. Die Proteinmengen und Eisenbindungscharakteristika des Ferritins wurden gemessen und die Ferritinsättigungswerte wurden berechnet. Obwohl sich die Ferritin-Proteinmengen nicht signifikant voneinander unterschieden, inkorporierten die Leber- und Mukosa-Ferritine der empfänglichen Vogelspezies deutlich mehr Eisen, was zu hohen Sättigungswerten führte. Im Vergleich zu den nicht empfänglichen Vogelspezies wurden in den Lebern beider Maina-Spezies und im Darm-Ferritin der indischen Hügelmainas signifikant höhere Ferrtin-Eisengehalte und -sättigungsgrade beobachtet. Ferritin scheint keine Hauptrolle in der Regulierung der Eisenabsorption zu spielen, was bedeutet, dass andere Phasen des Eisentransports wichtiger für die Auslösung und den nachfolgenden Prozess der Eisenüberladung in Vögeln sind.

Purificación parcial y caracterización de ferritina del hígado y mucosa intestinal de pollos), tórtolas comunes (Streptopelia d. decaocto)y pájaros minah (Acridotheres t. Tristis und Gracula religiosa)

La ferritina es una proteína almacenadora de hierro responsable de secuestrar el exceso de hierro, para ser almacenado de una forma segura en el hígado o para ser excretado en las células epiteliales intestinales. Las propiedades de la ferritina en aves susceptibles a padecer problemas de exceso de hierro se desconocen. Además, hay muy poca información de la ferritina que se encuentra en las mucosas, y no existe información alguna sobre este tema en la especie aviar. En este estudio se ha evaluado la ferritina del hígado e intestino proximal de especies susceptibles (pájaros minah Indian Hill, pájaros minah comunes) y no susceptibles (pollos tórtolas comunes) a problemas de exceso de hierro. Tras una breve purificación, se realizó una electroforesis en geles de poliacrilamida y se tiñeron para visualizar proteína y hierro. Las cantidades de proteína y las uniones a hierro características de la ferritina fueron medidas y se calcularon los niveles de saturación de ferritina. Aunque la cantidad de proteína ferritina no difiere significativamente, las ferritinas del hígado y de la mucosa intestinal de especies de aves sensibles incorporaron mucho más hierro, hasta niveles de saturación elevados. Contenidos de hierro y niveles de saturación significativamente mayores fueron observados en el hígado tanto de especies de aves minah y en la ferritina intestinal de pájaros minahs cuando se compararon con especies no susceptibles. La ferritina parece que no juega un papel importante en la regulación de la absorción de hierro, lo que implica que otras fases del transporte de hierro son más importantes que el inicio del proceso del exceso de hierro en aves.

Introduction

Ferritin is a cytosolic iron-binding protein present in all cells, which can store up to 4500 iron atoms per molecule. There are several functions assigned to ferritin in iron metabolism. In the case of the liver, most important is the safe storage of the potentially toxic excess of iron in an accessible form for later use. In the case of the intestinal cells, excess iron absorbed by the mucosal cell is eliminated from the body by desquamation (Harrison & Arosio, 1996). Thus, ferritin plays a crucial role in providing protection and iron homeostasis for the organism. The pathway of ferritin synthesis regulation lies in the iron-responsive element (IRE)/iron regulatory proteins 1 and 2 (IRP1 and IRP2) system, which is a function of the amount of labile cytosolic iron in mucosal cells. This system provides a translational control of all IRE-containing genes, of which ferritin is a part, in concert (Rolfs et al., 2002; Torti & Torti, 2002). These properties of ferritin have led to critical analyses regarding its function in and effects on iron metabolism disorders, especially in haemochromatosis. In human genetic haemochromatosis where iron absorption is increased inappropriately with respect to body iron status, ferritin does not prevent the transfer of high amounts of iron into the body.

Iron overload occurs in captivity in several avian species, among which mynah birds (Gracula sp.) are the most commonly encountered and most studied susceptible species. Much investigation has been carried out in order to elucidate the aetiological factor(s) leading to this susceptibility. It was previously shown that iron uptake from the intestinal tract and transfer into the body is significantly higher in these birds compared with non-susceptible species (turtledoves and chickens), and that the features of iron metabolism observed in mynahs is similar to the mechanism in hereditary haemochromatosis in humans (Mete et al., 2001 2003). However, the characteristics and the role of ferritin in iron metabolism of iron-overload-susceptible species are unknown. Furthermore, information on mucosal ferritin in birds or in mammals is hardly available.

The few studies conducted on avian ferritins describe methods of purification and characterization of the protein from the liver, muscle or spleen of domestic fowl, ducks and doves in order to determine the molecular weight or amino acid composition (Benito & Martin Mateo, 1983; Gonzalez del Barrio & Martin Mateo, 1983; Diez et al., 1985; Martin Mateo & Alonso-Arribal, 1988; Passaniti & Roth, 1989). In the genetic aspect, the ferritin H gene and its 5′ IRE in the mRNA and the presence of the IRP1 have been fully described in chickens (Rothenberger et al., 1990). The chicken ferritin H-subunit gene structure and amino acid sequence has high (93%) identity with the human orthologue (Stevens et al., 1987). Concerning the regulation of ferritin synthesis in chickens, it has been reported that protein concentrations in the liver and in the intestine were modified according to the dietary iron levels, but the observations on the ferritin subunits were inconclusive (Han et al., 2000).

In this study we have applied a practical technique primarily based on the electrophoretic separation of ferritin in its native state, and to estimate its yield and its iron content, to gain further insight on the role of ferritin in the iron cycle in iron-overload-susceptible and non-susceptible birds. For this purpose, we have compared the amount of ferritin and the amount of iron bound to ferritin in liver and proximal intestines in chickens, turtledoves and mynah birds (common mynahs and Indian hill mynahs), using rats as a mammalian reference species.

Materials and Methods

All procedures were approved by the animal experiments committee of the Faculty of Veterinary Medicine, Utrecht University, The Netherlands.

Animals

The adult animals used were eight chickens (Gallus gallus), eight turtledoves (Streptopelia d. decaocto), four common mynahs (Acridotheres t. tristis), four Indian hill mynahs (Gracula religiosa) and eight rats (Rattus rattus) U:WU (cpb). These were as many birds as were available at the time of the study. The iron content of the diets that the animals received was in the range of 60 to 90 parts/10⁶. They were euthanized using halothane (1 to 2%; Ceva Sante Animale B.V.) or injected intraperitoneally, intravenously or intramuscularly with 0.5 ml Euthesate® (containing 200 mg pentobarbitalnatrium/ml; Ceva Sante Animale B.V.). The proximal part of the intestines and the livers were collected within 15 min of death. After rinsing the lumen of the intestines twice with ice-cold physiological saline and weighing the livers, all organs were placed at −20°C until the time of the analyses and tests. All chemicals were obtained from either Sigma Aldrich Chemicals or Bio-Rad Laboratories.

Sample preparation

The frozen tissues were thawed on ice for approximately 30 min. For the intestinal mucosa preparation, the proximal parts of the intestines were cut open longitudinally and the mucosal layers, except for the muscularis mucosae, were scraped off using a glass slide. One gram of mucosal scraping was diluted into 1 ml of 50 mM Hepes buffer, pH 7.4, and homogenized on ice using an Ultra-Turrax homogenizer at maximum speed (5000 r.p.m.) for 2 min. Liver tissues were treated in the same way except that a piece of 1 g was diluted in 4 ml Hepes buffer, and homogenized for 4 min.

One millitre of each homogenate was subjected to heat treatment for 10 min at 75°C to aid the isolation of ferritin since other proteins are not stable at that temperature (Frenkel et al., 1983). After heat treatment the samples were immediately cooled down on ice for 30 min. Thereafter, the samples were re-centrifuged at 16 000 g for 30 min at 4°C until a clear supernatant was obtained and the pellet containing most of the insoluble denatured proteins was discarded. All tests were conducted in duplicate for each animal.

Electrophoresis and staining of gels

Native polyacrylamide gel electrophoresis was conducted using a 6% running gel and a 5% stacking gel. Samples were run at a constant voltage of 100 V.

After electrophoresis, the gels were treated with either of the two stains as described earlier (Leong et al., 1992); Coomassie blue G-250 stain, specific for proteins, or potassium ferricyanide (K₃Fe(CN)₆) stain, specific for iron. The corresponding band found in the protein and iron stained gel was considered to be ferritin.

Measurements

The gels were scanned with a Bio-Rad densitometer using a red filter and a final resolution of 450 dots per inch after achievement of colour development. Measurements of the bands were conducted using the Molecular Analyst (v. 3.1) program. The local background was subtracted from each sample.

Horse spleen ferritin (Sigma Aldrich) was used as a standard for calibrating ferritin protein and iron concentrations of the samples. Dilutions (ranging from 0.1 to 0.001 µg/µl) of the horse spleen ferritin were made and treated similarly to the liver and intestinal supernatant samples in order to create a reference line for both protein and iron-stained gels. Iron amounts were determined using the same calibration since approximately 20% of the weight of horse spleen ferritin is iron (Cetinkaya et al., 1985).

Saturation levels of ferritin with iron were calculated as the percentage of the iron present in the protein to the maximum amount of iron atoms (4500 iron atoms/ferritin molecule) ferritin can incorporate.

Statistical analysis

Differences in the amount of ferritin protein and of its iron content per gram of tissue were tested for significance using the F test for analysing variance between all groups at once (F values > 2.96 are significant). F=sd_b ²/sd_w ², in which sd_b represents the standard deviation between groups and sd_w the standard deviation within groups. The t test was used for the comparison between the two groups for tissue ferritin content (α=0.05 and P<0.05 used to detect significant differences). Analysis of variance (one-way), followed by the Bonferroni's multiple comparisons test, was used to compare species-dependent ferritin iron incorporation levels (expressed as µmol Fe/µmol ferritin and as the percentage saturation of ferritin). Measurements from all animals were used and P < 0.05 was considered statistically significant.

Results

A representative gel stained for protein and iron is shown in Figure 1a,b, where only iron-containing ferritin bands can be observed in the iron staining. The avian ferritins correspond to a weight of approximately 470 to 500 kDa.

Figure 1.  Native polyacrylamide gel electrophoresis. (1a) protein (Coomassie blue), (1b) iron (potassium ferricyanide) staining of liver samples. Lane A, rat; lane B, chicken; lane C, dove; lane D, Indian hill mynah; lane E, common mynah; lane H, horse spleen ferritin.

Ferritin and iron in the liver and intestines

The mean values of ferritin protein and the amount of iron present in the ferritin of the liver and intestinal samples of all animals are presented in Table 1. The number of iron atoms per ferritin molecule is presented as µmol Fe/µmol ferritin.

Table 1.  Ferritin protein and the iron incorporation levels

	Ferritin (µg/g wet weight)	Iron (µg Fe/g wet weight)	Iron (µmol Fe/µmol ferritin)
Liver ferritin
Rat	969±173	63±24	550±173^a
Chicken	914±382	52±16	515±262^b
Turtledove	634±205	31±8	423±101^c
Indian hill mynah	1207±223	179±30	1252±258^a,b,c
Common mynah	346±233	78±53	1541±374^a,b,c
Intestinal ferritin
Rat	73±24	5±2	567±145^a
Chicken	443±112	34±14	616±185^b
Turtledove	98±18	5±2	412±131^c
Indian hill mynah	116±28	20±9	1401±576^a,b,c
Common mynah	57±15	6±2	958±283
Values presented as the mean±standard deviation. Superscript letters represent significant differences (one-way analysis of variance, P<0.05).

In rats and chickens, liver ferritin seems to be present in similar amounts. It is least in doves and highest in Indian hill mynahs. The ferritin and iron contents per wet weight liver of Indian hill mynahs are significantly higher than in the other species (F value = 16.13). In the intestines, a similar amount of ferritin is evident in rats and doves, whereas it is highest in chickens. In both tissues, the common mynahs have the lowest amount of ferritin present. Comparing the amount of ferritin present in the liver and intestine of all species showed significantly more ferritin in the liver (t-test values varying from 4.06 in chicken to 10.09 in rats, P<0.05).

When looking at the iron contents, the number of iron atoms bound to ferritin is similar among species between the two organs except for the common mynahs. In this species, the liver ferritin iron incorporation is much higher than in the intestine. As shown in Table 1, the doves have the lowest amount of iron per ferritin molecule in both the liver and intestines (423±101 and 412±131 µmol Fe/µmol ferritin, respectively). Both species of mynahs have the highest amount of iron per ferritin molecule: 1252±258 and 1541±374 µmol Fe/µmol ferritin in the liver, and 1401±576 and 958±283 µmol Fe/µmol ferritin in the intestines. The rats and chickens have about the same amounts of iron bound to ferritin per organ, being closer to the dove values than to mynahs. Significant differences in the iron content of ferritin were found in both mynah species compared with the rat, chicken and doves in the liver (P<0.05). In the intestines, the Indian hill mynahs showed significantly high amounts of iron incorporation (P<0.05) compared with the other species.

Saturation of ferritin

The saturation of ferritin in the liver and intestine of the animals is shown in Figure 2. Liver ferritins showed saturation degrees of 12±4 and 11±6% in rats and chickens, respectively, of 9±2% in doves, of 28±6% in Indian hill mynahs and of 34±8% in common mynahs. Similar values are seen in the intestines: 13±3% in rats, 14±4% in chickens, 9±3% in doves, 31±13% in Indian hill mynahs and 21±6% in the common mynahs. Both Indian hill mynahs and common mynahs were found to have significantly higher saturation levels compared with the other species in the liver, and only the Indian hill mynahs had significantly higher levels in the intestinal mucosa (P < 0.05).

Figure 2.  Ferritin iron saturation. Iron saturation values calculated as the percentage of µmol Fe/µmol ferritin protein in liver and mucosal samples of rats (n=8), chickens (n=8), doves (n=8), Indian hill mynahs (n=4) and common mynahs (n=4). Bars represent standard deviation.

Discussion

Here we show for the first time an estimate quantification of in situ ferritin amounts in the major iron metabolism regulatory sites, the liver and proximal intestines, of iron-overload-susceptible (mynahs) and non-susceptible birds (chickens and doves). In addition, studying this protein in its native state has enabled us to determine the iron bound to its core. We have added rats to our study as a mammalian reference species. Comparisons of the present results with earlier studies provide corresponding values for the yield of ferritin protein in rat or mouse liver and mucosa (Cham et al., 1985; Ehtechami et al., 1989; Gérard et al., 1996). In birds and some other species, only liver ferritin has been studied previously with a much lower yield than ours (10-fold to 100-fold) (Gonzalez del Barrio & Martin Mateo, 1983; Suryakala & Deshpande, 1999). This may be due to the purification process. The high yield we have obtained may be due to the lower loss of protein achieved by our procedure since multiple steps used in the isolation and purification of ferritin cause high losses of the protein (Frenkel et al., 1983; Cham et al., 1985 1986).

The ferritin amount in the livers was found to be much higher when compared with the intestines in all species. This is due to the long-term exposure of liver cells to iron, contrary to the intestinal cells with a life span of approximately 2 days. The very high yield of ferritin in the chicken mucosa and the low yields in both the liver and mucosa of the common mynahs were not predicted. For all species, however, the iron content in both the liver and intestinal ferritins showed parallel values between the organs. Indeed, the saturation levels representing the number of iron atoms bound as a percentage of the maximum iron incorporation capacity of ferritin produced the same picture in all species among the two organs. As expected, among the studied species, iron atoms bound per ferritin molecule in both organs were found to be the lowest in doves; since these animals do not accumulate iron in the body due to adequately down-regulated iron absorption (Mete et al., 2001). The same situation holds true for rats and chickens, and the values for ferritin and its iron content in the liver appeared to be similar, while mucosal ferritin and iron in chickens was found to be high. Nevertheless, the iron content of this high ferritin is equivalent to the rat, showing that the protein yield may be high but it has a relatively low iron content. Ferritin, together with haemosiderin, represents the iron status of tissues and the body. The two iron-overload-susceptible mynah species showed considerably and significantly higher iron levels in the ferritins. Thus, the high iron saturation of liver ferritin is consistent with the iron-overload state seen in these species. Intestinal ferritin is rather low if related to the general iron overload. This is in accordance with a study of duodenal iron proteins in human idiopathic haemochromatosis, finding that mucosal ferritin failed to increase in parallel with serum ferritin representing body iron stores, but was appropriate for the level of iron absorption (Whittaker et al., 1989). It is remarkable in our study that the mynah species exhibit such high ferritin iron saturation in intestinal cells. This can be explained by appropriate and effective ferritin iron capture; of the increased amounts of iron taken up by the microvillus membrane during its passage towards the basolateral membrane. Of the animals studied, only the mynahs have an increased absorption of iron (Mete et al., 2001 2003).

The IRE structures and IRPs are highly conserved among chickens and mammals (Stevens et al., 1987; Rothenberger et al., 1990). Regulation by the IRE/IRP system is by translational control of the protein in cells, where the ferritin protein levels change according to the cellular iron pool. Normal ferritin transcription and translation is preserved in genetic haemochromatosis (Pietrangelo et al., 1995). This seems to be the case in mynahs as well.

In the present study we report the high iron binding characteristics of ferritins in iron-overload-susceptible species in both the liver and proximal intestines, and that ferritin amounts are not directly correlated to their iron content. The very high iron saturation of intestinal ferritin of mynahs, resembling human hereditary haemochromatosis, may be due to the increased and rapid cellular iron passage due to the excessively high absorption of iron (Pietrangelo, 2004).

Translations of the abstract in French, Germany and Spanish are available on the Avian Pathology website.

The authors would like to thank Fons van Asten, Henno Hendriks and Hilda Toussaint for their help with the technical part of the research.