American Journal of Hematology 399-14 (1992)
Uptake of Transferrin-Bound and Nontransferrin-Bound Iron by Reticulocytes From the Belgrade Laboratory Rat: Comparison With Wistar Rat Transferrin and Reticulocytes Elizabeth A. Farcich and Evan H. Morgan Department of Physiology, University of Western Australia, Nedlands, Western Australia, Australia
The mechanism underlying the impaired uptake of iron from transferrin by reticulocytes from the Belgrade laboratory rat was investigated using lZ5land 59Fe-labeledtransferrin isolated from homozygous Belgrade rats and from Wistar rats, nontransferrin-bound Fe(ll) in an isotonic sucrose solution, and reticulocytes from Belgrade and Wistar rats. The Belgrade rat transferrin had the same molecular weight and net charge as Wistar rat transferrin, donated iron equally well to both types of reticulocytes, and competed equally for transferrin binding sites on the cells. Hence, the defect in iron uptake by Belgrade rat reticulocytes could not be attributed to an abnormality of the transferrin molecule. The rate of uptake of Fe(ll) from sucrose into the cytosolic and stromal fractions of Belgrade rat reticulocytes was only about 35% as great as that by Wistar rat reticulocytes. With both types of cells, the uptake process was saturable, suggesting the presence of a carriermediated process. It was therefore concluded that the defect in iron uptake by Belgrade rat erythroid cells is probably the consequenceof a deficiency in a membrane carrier for iron. Key words: iron transport, Fe(ll), anemia
INTRODUCTION
of the transferrin molecule, which could result in altered delivery of iron to the transport system. The present The anemia of the Belgrade laboratory rat is inherited experiments were aimed at investigating this question. as an autosomal recessive trait. In homozygous (b/b) Two types of experiments were performed. In the first, animals it is expressed as a severe microcytic, hypochrotransferrins isolated from homozygous Belgrade rats and mic anemia associated with reticulocytosis and elevated from non-Belgrade (Wistar) rats were compared with plasma iron and transferrin saturation [ 1,2]. Heterozyrespect to some chemical properties and their ability to gous (b/-) rats are nonanemic. The anemia is associated donate iron to reticulocytes. The second set of experiwith, and is probably due to, impaired utilization of ments involved a direct study of the iron-transporting transferrin-bound iron by developing erythroid cells capacity of Belgrade rat reticulocyte cell membranes [1,31. using a nontransferrin-boundFe(I1) transport system that In a previous study using reticulocytes from homozyhas been described recently IS]. gous Belgrade rats, it was shown that transferrinrreceptor interaction, endocytosis, and recycling were not altered when compared with reticulocytes from iron-deficient, MATERIALS AND METHODS non-Belgrade animals [4].Hence it was concluded that Animals the abnormality in the Belgrade cells probably resided in Homozygous Belgrade (b/b) rats were bred from the process of iron transfer across the lining membrane of heterozygotes (b/-) kindly provided by Dr John A. endocytotic vesicles into the cytosol of the cells. However, these experiments and all earlier published experiments on iron uptake by Belgrade rat reticulocytes were Received for publication March 19, 1991; accepted June 20, 1991 performed using transferrin isolated from normal Wistar Address reprint requests to Dr. E.H. Morgan, Department of Physirats. Hence they do not exclude the possibility that the ology, University of Western Australia, Nedlands, Western Australia defect in iron transport in vivo is due to an abnormality 6009, Australia. 0 1992 Wiley-Liss, Inc.
10
Farcich and Morgan
Edwards (Dept. of Medicine, State University of New unlabeled transferrin. Tubes were incubated at 37°C for York, Buffalo). The anemic animals were kept alive by 20 min, then washed and processed as above. frequent intraperitoneal injections of imferon (Fisons) during the first 3 months of life. These animals display Nontransferrin-Bound Iron Uptake Studies spontaneous reticulocytosis, whereas reticulocytosis was To study the uptake of nontransferrin-bound iron, a induced in control Wistar rats by intraperitoneal injection Fe(I1) solution containing 62.5 KM iron as 59FeC1, and of phenylhydrazine as described previously [4]. Iron 56FeS0, in a 1:lO molar ratio, with a 50-fold molar deficiency was induced in otherwise normal Wistar rats excess of P-mercaptoethanol in 0.27 M sucrose was used by feeding a low-iron diet and bleeding by heart punc- [ 5 ] .The iron was shown to be in the ferrous state by the ture, weekly, as in earlier work [4]. In addition, het- formation of the characteristic colored product on reacerozygous Belgrade rat reticulocytes from phenylhydra- tion with 2,2’-bipyridine. Reticulocytes were suspended zine-treated animals were used in initial studies and in isotonic sucrose to a PCV of approximately 15%. displayed uptake identical to that of Wistar cells in all Samples of this suspension (100 pl) were added to 4.9 ml experiments. Heterozygote reticulocyte data is presented of incubation solution containing the above Fe(I1) solution diluted to the desired final Fe concentration in along with the “Wistar” results. Reticulocyte-rich blood was obtained from all rats by isotonic sucrose buffered to pH 6.5 with 4 mM PIPES. Cells were incubated either for varing time periods in heart puncture and transfer to heparinized tubes, and the cells were washed three times with ice-cold phosphate 1 p M Fe(I1) solution or for 15 min in varying concenbuffered saline (PBS) prior to a 30 minute, 2,OOOg trations of Fe(I1) solution. Following the 37°C incubacentrifugation at 4°C. The reticulocyte-enriched upper tion, a tenfold volume of ice-cold 4:l (v/v) isotonic portion of the washed cells was used for all experiments sucrose:PBS was added and the cells recovered by and invariably contained between 40% and 50% reticu- centrifugation at 8OOg for 10 min at 4°C. For the studies locytes. For brevity the reticulocyte-rich cell suspensions on Fe(I1) concentration effects, aliquots of the supernafrom homologous Belgrade rats, iron-deficient rats, and tant were retained and counted to determine Fe(I1) phenylhydrazine-treated rats will be referred to as Bel- concentration in individual tubes at the time of incubagrade, iron-deficient, and control reticulocytes, respec- tion. The pelletted cells were lysed by addition of 10 mM TRIS, pH 8.2, and separated into cytosolic and stromal tively. fractions as in earlier work [ 5 ] . Both fractions were Reagents and Buffer Solutions counted for radioactivity. Iron-59 (59FeC1,) and iodine- 125 (NaIz5I) were purchased from Amersham international (Amersham, Analytical Methods Reticulocyte numbers were determined by staining U.K.). All other reagents were of analytical grade. Hanks balanced salt solution 161 was used for incubations in with new methylene blue, PCV by the microhematocrit transferrin uptake experiments. Transferrin was isolated procedure and radioactivity in an LKB Wallac 1282 from plasma of Wistar and Belgrade rats and was labeled Compugamma Universal Gamma Counter. PAGE was with 59Fe and 12’1 as previously described for chicken performed as described by Laemmli [S] using 9.7% acrylamide in the resolving gel and reducing samples in transferrin [7]. 10% 2-mercaptoethanol, 4% sodium dodecyl sulfate Measurement of Transferrin and Iron Uptake (SDS) at 95°C. Proteins were visualised by Coomassie Washed reticulocytes were suspended in Hanks solu- blue. Western blot onto Hybond C (Amersham) was tion to a PCV of approximately 15%. The reticulocytes performed using a continuous buffer in an LKB Mulwere incubated in a shaking 37°C water bath in the tiphore I1 Novablot Electrophoresis Unit and staining presence of either labeled Belgrade or labeled Wistar with rabbit anti-Wistar transferrin, followed by reaction transferrin. At various time points, aliquots of the to goat antirabbit IgG conjugated to horseradish peroxisuspension were removed and the cells washed three dase (Pasteur). Transferrin electrophoretic mobility on times with ice-cold Hanks solution and transferred to cellulose acetate was determined using a University Graham Instruments model MR4169 Electrophoresis fresh test tubes before being counted for radioactivity. For competitive analyses, samples of both control and unit, visualised with Ponceau S. Belgrade reticulocytes were incubated in a fixed concentration (2 yM) of labeled heterologous or homologous transferrin. Unlabeled transferrin of either type was RESULTS added to each sample in varying molar ratios of labeled Comparison of Belgrade and Wistar Transferrins No difference between the two proteins was observed to unlabeled protein. This protocol accounts for separate incubation of both cell types with both types of labeled with respect to their mobility on cellulose acetate or transferrin competing for uptake with both types of SDS-PAGE. Their molecular weights as determined by
Iron Uptake by Belgrade Rat Reticulocytes
11
I
4 6 12 Competitor (Molar Ratio)
0
loo\
16
10 15 Incubation Time (min)
5
0
20
B lo0l
0
4 6 12 Competitor (Molar Ratio)
16
0
B
7 0
5 10 15 Incubation Time (mln)
20
Fig. 1. Transferrin (A) and iron (6) uptake by homozygous Belgrade rat reticulocytes from 59Fe-‘251-labeled Wistar transferrin competing with unlabeled Wistar transferrin (m) and with unlabeled Belgrade transferrin).( and for labeled Belgrade transferrin competing for uptake with unlabeled Wistar transferrin (0) and with unlabeled Belgrade transferrin (0).Results are mean 2 S.E.M. for four trials.
Fig. 2. Transferrin (A) and iron (B) uptake by control rat reticulocytes from Wistar transferrin (m) and Belgrade transferrin (0) and for homozygous Belgrade rat reticulocytes from Wistar transferrin (0) and Belgrade transferrin (0). Results are mean f S.E.M. for four trials.
the latter technique were 80,000 2 500. In addition, they stained with equal intensity after Western blot probed with rabbit antiserum to Wistar rat transferrin. To verify the homology between the two transferrins, competitive uptake experiments were performed with results being expressed as percentage of uptake in the absence of the competing unlabeled transferrin. There was found to be no significant difference between transferrin and iron uptake from either transfenin by the individual cell types. Results shown are for Belgrade reticulocytes in Figure 1; however, control reticulocytes showed a similar pattern. These results indicate that unlabeled Belgrade transferrin inhibits uptake of iron and transferrin from labeled Wistar transferrin as effectively as it does from labeled Belgrade transferrin. Similarly, unlabeled Wistar transferrin inhibits iron and transferrin uptake equally from both types of labeled transferrin. The uptakes of transferrin and iron by Belgrade rat reticulocytes from either Wistar or Belgrade transferrin were found to be identical, as was the uptake by the control reticulocytes (Fig. 2A and B). However, Bel-
grade reticulocytes displayed a greatly diminished ability to internalize iron, being only 12% as efficient as the Wistar reticulocytes. Iron uptake by both cell types was linear throughout the 20 min incubation, whereas transferrin uptake was initially rapid, slowing markedly before plateauing after approximately 8 min. Again, control reticulocytes showed higher levels of transferrin uptake, approximately 2.4 times that of Belgrade reticulocytes. Thus transferrin endocytosis was slower in Belgrade than in control reticulocytes, but the relative decrease was only about 30% as great as for iron uptake. A similar difference in transferrin uptake by Belgrade and control reticulocytes, was noted previously and was attributed to an effect of iron deficiency, not the inherited defect, on the cells [4].Non-Belgrade iron-deficient reticulocytes were found to take up transferrin at the same rate as Belgrade reticulocytes, but accumulated iron at three times the rate [4].Hence iron-deficient reticulocytes are probably a better control for studies with Belgrade reticulocytes than are “control” reticulocytes. Iron-deficient reticulocytes were therefore used in the experiments described below.
12
Farcich and Morgan
f:
1
/
200i
30
-
150
l 50
-
0
0
5
:
10 Time (min)
E
15
20 0
1
2
3
4
Iron Concentration (pM)
I 150
5
10 Time (min)
15
20
Fig. 3. Uptake of nontransferrin-bound Fe(ll) by rat reticulocytes into stoma1 (A) and cytosolic (B) fractions of irondeficient rat reticulocytes ()., control rat reticulocytes ( A ) and homozygous Belgrade rat reticulocytes (0). Results are mean & S.E.M. for five trials.
-
l5I 10
5 0
0
1
2
3
4
Iron Concentration (pM)
Uptake of Nontransferrin-Bound Iron
Fe(I1) uptake into stromal and cytosolic fractions for all cell types occurred in a linear manner for at least 20 min (Fig. 3A and B). Iron-deficient reticulocytes had significantly enhanced Fe(I1) uptake into both fractions compared with control cells. Conversely, Belgrade reticulocytes after 20 min had only about 18% as much iron in the stromal fraction as the control reticulocytes and 8.9% as much as the iron-deficient cells. The cytosolic uptake was also reduced in the Belgrade cells, the rate being about 13.5% that of the control and 7.6% that of iron-deficient reticulocytes . When reticulocytes were incubated in medium containing increasing concentrations of iron the rate of Fe(I1) uptake into both stromal and cytosolic fractions for all cell types showed evidence of saturation at about 1 p M iron concentration, but continued to increase in a linear fashion when iron concentration was further raised (Fig. 4A and B). This suggests that both a specific saturable uptake process and a nonspecific nonsaturable process are involved. These two components of the uptake curve were derived assuming an equation for the curve of ax f hxlc x. The nonspecific component, a (the slope of the linear portion of the uptake curve), was then subtracted from the total uptake curve to give the specific component. The specific portion of each curve was further analysed by Eadie-Hofstee plots to determine the
+
Fig. 4. Effect of varying the Fe(ll) concentration in the incubation medium on the rate of iron uptake into stromal (A) and cytosolic (B) fractions of iron-deficient rat reticulocytes (m), control rat reticulocytes ( ), and homorygous Belgrade rat reticulocytes (0). Results are mean r S.E.M. for five trials.
maximum rate of Fe(I1) uptake (V,,,) and the Michaelis constant (Km). A summary of these results is given in Table I. They indicate that Belgrade reticulocytes have a lower ability to take up Fe(I1) into both cytosol and stroma than either the control or iron-deficient cells. Maximum uptake into either fraction was less than 40% of that by the other two cell types. If it is assumed that for the cytosolic uptake the V,, values represent the activity of a membrane carrier for iron, then the Belgrade rat reticulocytes have lower functional activity of this carrier than either the control or the iron-deficient reticulocytes. DISCUSSION
The results indicate that the anemia of the Belgrade laboratory rat is unlikely to be due to defective transfemn function. The transferrin purified from the plasma of homozygous Belgrade rats was of the same M, as transferrin from Wistar plasma and also displayed identical electrophoretic mobility on the cellulose acetate, implying that the two proteins have the same net charge.
iron Uptake by Belgrade Rat Reticulocytes
13
TABLE 1. Vmaxand K, Values for Specific Fe(ll) Uptake*
Reticulocytes Belgrade Control Iron deficient
V,,, (nmol/ml reticulocytes per min) cytosol Stroma 1.23 f 0.17 3.08 0.12 3.38 f 0.13
*
5.02 f 0.36 14.7 f 0.71 16.5 f 0.83
K m (PM) Cytosol
Stroma
0.16 f 0.03 0.08 k 0.01 0.12 f 0.01
0.31 f 0.01 0.33 0.05 0.17 f 0.02
*
*The results show the maximal rates (V,,,,,) and Michaelis constants (K,) for the specific (saturable) component of Fe(I1) uptake into cytosolic and stromal fractions of Belgrade, control, and iron-deficient reticulocytes. The results are the means (fSEM) of five trials.
Furthermore, the two transferrins were shown to be equally favored with respect to both iron donation and transferrin binding to rat reticulocytes as revealed by the competition and direct uptake experiments. Hence, as assessed by these techniques, the transferrins are functionally identical. Although the transferrin from homozygous Belgrade rats appears to be functionally normal, the reticulocytes from such animals were found to have decreased ability to take up iron in either its free state or bound to transferrin. Previous studies showed that the transferrin endocytosis cycle is qualitatively and quantitatively normal in Belgrade reticulocytes compared with iron-deficient reticulocytes but iron fails to accumulate within the cells, implying that there is impaired passage of iron across the endosomal membrane [4]. The present results with transferrin-bound iron support this conclusion, and those with nontransferrin-bound Fe(I1) provide strong direct evidence for the presence of a defective irontransport mechanism in the membrane of Belgrade reticulocytes. Earlier studies with rabbit reticulocytes detailed the ability of such cells to take up nontransferrin-bound Fe(I1) in vitro. The mechanism for this was found to be saturable, pH and temperature dependent, and inhibited by ions of other transition metals. It was concluded that the iron was transported into the cells by a carriermediated process [ 5 ] .The results presented here indicate the presence of a similar iron-uptake system in rat reticulocytes, saturating when the iron concentration is about 0.5 pM. As with rabbit reticulocytes, this was accompanied by a nonsaturable process, which was apparent at the higher iron concentrations. The uptake of nontransferrin-bound iron by Belgrade reticulocytes was found to be less than that by both the iron-deficient and the control reticulocytes. The uptake into the cytosol by Belgrade cells was only 13.5% and 7.6%and the uptake into the stromal fraction of the cells (which consists of both intracellular organelles and outer cell membrane) was about 18% and 8.9% of that by control and iron-deficient reticulocytes, respectively.
These results indicate both impaired transport of ferrous ions into the reticulocytes of the homozygous Belgrade rats in vitro and lower cell surface binding of the Fe(1I) by these cells compared with the other reticulocytes. The amount of iron incorporated into the stromal and cytosolic fractions of the control and iron-deficient reticulocytes was similar at all time points. However, the cytosolic uptake of Fe(I1) into the Belgrade rat reticulocytes was only 70% that into the stromal fraction, suggesting a lower ability of iron passage across the reticulocyte cell membrane of the homozygous Belgrade rat compared with both iron-deficient and control cells. Such evidence further supports the view that the basic defect in the iron metabolism in the Belgrade rat is one of impaired membrane transport of iron. The V,,, values for the specific (saturable) uptakes of Fe(I1) by Belgrade reticulocytes into the cytosol and stroma individually were both about 30% and 35% of the values for the iron-deficient cells and control cells, respectively. Thus the specific uptake of ferrous ions into the cytosol of Belgrade reticulocytes is no more impaired than the specific binding of iron to the cell membranes. The specific site of binding of the Fe(1I) is as yet unknown, but, if a protein carrier is involved in the uptake of Fe(I1) by rat reticulocytes, then the above result implies that the protein itself is deficient in binding iron at the cell suface as well as in allowing the passage of the iron into the cells of the Belgrade rat. In fact, of the total maximal specific uptake of Fe(II), about 20% was found in the cytosol in all three cell types. This suggests that the specific process of iron uptake is qualitatively similar in the three reticulocyte populations and that it is the quantitative parameters that distinguish them, Belgrade reticulocytes having far lower levels of Fe(I1) transport across the cell membrane than the control and irondeficient cells. ACKNOWLEDGMENTS
This work was supported by a grant from the Australian Research Council.
14
Farcich and Morgan
REFERENCES I . Sladic-Simic D, Martinovitch PN, Zwkovic N , Pavic D. Martinovic J : A thalassemic-like disorder in Belgrade laboratory rats. Ann NY Acad Sci 165:93-99, 1969. 2. Edwards JA, Sullivan AL. Hoke JE: Defective delivery of iron to the developing red cells of the Belgrade laboratory rat. Blood 55:645-648, 1980. 3. Edwards JA, Huebers H . Kunzler C, Finch CA: Iron metabolism in the Belgrade rat. Blood 67:623-628, 1986. 4. Bowen B, Morgan EH: Anemia of the Belgrade rat: Evidence for defective membrane transport of iron. Blood 70:3844. 1987.
5. Morgan EH: Membrane transport of non-transfernin-bound iron by reticulocytes. Biochim Biophys Acta 943:428-439, 1988. 6. Hanks JH, Wallace RE: Relation of oxygen and temperature in the preservation of tissues by refrigeration. Proc Sac Exp Biol Med 7 1 :196200. 1949. 7. Lim BC, Morgan EH: Transfenin endocytosis and iron uptake by developing erythroid cells in the chicken (Gaflusdomesticus). J Camp Physiol B155:201-210, 1985. 8. Laemmli UK: Cleavage of structural proteins during the assembly of the head bacteriophage T4. Nature 227:68G685. 1970.
Uptake of Transferrin-Bound and Nontransferrin-Bound Iron by Reticulocytes From the Belgrade Laboratory Rat: Comparison With Wistar Rat Transferrin and Reticulocytes Elizabeth A. Farcich and Evan H. Morgan Department of Physiology, University of Western Australia, Nedlands, Western Australia, Australia
The mechanism underlying the impaired uptake of iron from transferrin by reticulocytes from the Belgrade laboratory rat was investigated using lZ5land 59Fe-labeledtransferrin isolated from homozygous Belgrade rats and from Wistar rats, nontransferrin-bound Fe(ll) in an isotonic sucrose solution, and reticulocytes from Belgrade and Wistar rats. The Belgrade rat transferrin had the same molecular weight and net charge as Wistar rat transferrin, donated iron equally well to both types of reticulocytes, and competed equally for transferrin binding sites on the cells. Hence, the defect in iron uptake by Belgrade rat reticulocytes could not be attributed to an abnormality of the transferrin molecule. The rate of uptake of Fe(ll) from sucrose into the cytosolic and stromal fractions of Belgrade rat reticulocytes was only about 35% as great as that by Wistar rat reticulocytes. With both types of cells, the uptake process was saturable, suggesting the presence of a carriermediated process. It was therefore concluded that the defect in iron uptake by Belgrade rat erythroid cells is probably the consequenceof a deficiency in a membrane carrier for iron. Key words: iron transport, Fe(ll), anemia
INTRODUCTION
of the transferrin molecule, which could result in altered delivery of iron to the transport system. The present The anemia of the Belgrade laboratory rat is inherited experiments were aimed at investigating this question. as an autosomal recessive trait. In homozygous (b/b) Two types of experiments were performed. In the first, animals it is expressed as a severe microcytic, hypochrotransferrins isolated from homozygous Belgrade rats and mic anemia associated with reticulocytosis and elevated from non-Belgrade (Wistar) rats were compared with plasma iron and transferrin saturation [ 1,2]. Heterozyrespect to some chemical properties and their ability to gous (b/-) rats are nonanemic. The anemia is associated donate iron to reticulocytes. The second set of experiwith, and is probably due to, impaired utilization of ments involved a direct study of the iron-transporting transferrin-bound iron by developing erythroid cells capacity of Belgrade rat reticulocyte cell membranes [1,31. using a nontransferrin-boundFe(I1) transport system that In a previous study using reticulocytes from homozyhas been described recently IS]. gous Belgrade rats, it was shown that transferrinrreceptor interaction, endocytosis, and recycling were not altered when compared with reticulocytes from iron-deficient, MATERIALS AND METHODS non-Belgrade animals [4].Hence it was concluded that Animals the abnormality in the Belgrade cells probably resided in Homozygous Belgrade (b/b) rats were bred from the process of iron transfer across the lining membrane of heterozygotes (b/-) kindly provided by Dr John A. endocytotic vesicles into the cytosol of the cells. However, these experiments and all earlier published experiments on iron uptake by Belgrade rat reticulocytes were Received for publication March 19, 1991; accepted June 20, 1991 performed using transferrin isolated from normal Wistar Address reprint requests to Dr. E.H. Morgan, Department of Physirats. Hence they do not exclude the possibility that the ology, University of Western Australia, Nedlands, Western Australia defect in iron transport in vivo is due to an abnormality 6009, Australia. 0 1992 Wiley-Liss, Inc.
10
Farcich and Morgan
Edwards (Dept. of Medicine, State University of New unlabeled transferrin. Tubes were incubated at 37°C for York, Buffalo). The anemic animals were kept alive by 20 min, then washed and processed as above. frequent intraperitoneal injections of imferon (Fisons) during the first 3 months of life. These animals display Nontransferrin-Bound Iron Uptake Studies spontaneous reticulocytosis, whereas reticulocytosis was To study the uptake of nontransferrin-bound iron, a induced in control Wistar rats by intraperitoneal injection Fe(I1) solution containing 62.5 KM iron as 59FeC1, and of phenylhydrazine as described previously [4]. Iron 56FeS0, in a 1:lO molar ratio, with a 50-fold molar deficiency was induced in otherwise normal Wistar rats excess of P-mercaptoethanol in 0.27 M sucrose was used by feeding a low-iron diet and bleeding by heart punc- [ 5 ] .The iron was shown to be in the ferrous state by the ture, weekly, as in earlier work [4]. In addition, het- formation of the characteristic colored product on reacerozygous Belgrade rat reticulocytes from phenylhydra- tion with 2,2’-bipyridine. Reticulocytes were suspended zine-treated animals were used in initial studies and in isotonic sucrose to a PCV of approximately 15%. displayed uptake identical to that of Wistar cells in all Samples of this suspension (100 pl) were added to 4.9 ml experiments. Heterozygote reticulocyte data is presented of incubation solution containing the above Fe(I1) solution diluted to the desired final Fe concentration in along with the “Wistar” results. Reticulocyte-rich blood was obtained from all rats by isotonic sucrose buffered to pH 6.5 with 4 mM PIPES. Cells were incubated either for varing time periods in heart puncture and transfer to heparinized tubes, and the cells were washed three times with ice-cold phosphate 1 p M Fe(I1) solution or for 15 min in varying concenbuffered saline (PBS) prior to a 30 minute, 2,OOOg trations of Fe(I1) solution. Following the 37°C incubacentrifugation at 4°C. The reticulocyte-enriched upper tion, a tenfold volume of ice-cold 4:l (v/v) isotonic portion of the washed cells was used for all experiments sucrose:PBS was added and the cells recovered by and invariably contained between 40% and 50% reticu- centrifugation at 8OOg for 10 min at 4°C. For the studies locytes. For brevity the reticulocyte-rich cell suspensions on Fe(I1) concentration effects, aliquots of the supernafrom homologous Belgrade rats, iron-deficient rats, and tant were retained and counted to determine Fe(I1) phenylhydrazine-treated rats will be referred to as Bel- concentration in individual tubes at the time of incubagrade, iron-deficient, and control reticulocytes, respec- tion. The pelletted cells were lysed by addition of 10 mM TRIS, pH 8.2, and separated into cytosolic and stromal tively. fractions as in earlier work [ 5 ] . Both fractions were Reagents and Buffer Solutions counted for radioactivity. Iron-59 (59FeC1,) and iodine- 125 (NaIz5I) were purchased from Amersham international (Amersham, Analytical Methods Reticulocyte numbers were determined by staining U.K.). All other reagents were of analytical grade. Hanks balanced salt solution 161 was used for incubations in with new methylene blue, PCV by the microhematocrit transferrin uptake experiments. Transferrin was isolated procedure and radioactivity in an LKB Wallac 1282 from plasma of Wistar and Belgrade rats and was labeled Compugamma Universal Gamma Counter. PAGE was with 59Fe and 12’1 as previously described for chicken performed as described by Laemmli [S] using 9.7% acrylamide in the resolving gel and reducing samples in transferrin [7]. 10% 2-mercaptoethanol, 4% sodium dodecyl sulfate Measurement of Transferrin and Iron Uptake (SDS) at 95°C. Proteins were visualised by Coomassie Washed reticulocytes were suspended in Hanks solu- blue. Western blot onto Hybond C (Amersham) was tion to a PCV of approximately 15%. The reticulocytes performed using a continuous buffer in an LKB Mulwere incubated in a shaking 37°C water bath in the tiphore I1 Novablot Electrophoresis Unit and staining presence of either labeled Belgrade or labeled Wistar with rabbit anti-Wistar transferrin, followed by reaction transferrin. At various time points, aliquots of the to goat antirabbit IgG conjugated to horseradish peroxisuspension were removed and the cells washed three dase (Pasteur). Transferrin electrophoretic mobility on times with ice-cold Hanks solution and transferred to cellulose acetate was determined using a University Graham Instruments model MR4169 Electrophoresis fresh test tubes before being counted for radioactivity. For competitive analyses, samples of both control and unit, visualised with Ponceau S. Belgrade reticulocytes were incubated in a fixed concentration (2 yM) of labeled heterologous or homologous transferrin. Unlabeled transferrin of either type was RESULTS added to each sample in varying molar ratios of labeled Comparison of Belgrade and Wistar Transferrins No difference between the two proteins was observed to unlabeled protein. This protocol accounts for separate incubation of both cell types with both types of labeled with respect to their mobility on cellulose acetate or transferrin competing for uptake with both types of SDS-PAGE. Their molecular weights as determined by
Iron Uptake by Belgrade Rat Reticulocytes
11
I
4 6 12 Competitor (Molar Ratio)
0
loo\
16
10 15 Incubation Time (min)
5
0
20
B lo0l
0
4 6 12 Competitor (Molar Ratio)
16
0
B
7 0
5 10 15 Incubation Time (mln)
20
Fig. 1. Transferrin (A) and iron (6) uptake by homozygous Belgrade rat reticulocytes from 59Fe-‘251-labeled Wistar transferrin competing with unlabeled Wistar transferrin (m) and with unlabeled Belgrade transferrin).( and for labeled Belgrade transferrin competing for uptake with unlabeled Wistar transferrin (0) and with unlabeled Belgrade transferrin (0).Results are mean 2 S.E.M. for four trials.
Fig. 2. Transferrin (A) and iron (B) uptake by control rat reticulocytes from Wistar transferrin (m) and Belgrade transferrin (0) and for homozygous Belgrade rat reticulocytes from Wistar transferrin (0) and Belgrade transferrin (0). Results are mean f S.E.M. for four trials.
the latter technique were 80,000 2 500. In addition, they stained with equal intensity after Western blot probed with rabbit antiserum to Wistar rat transferrin. To verify the homology between the two transferrins, competitive uptake experiments were performed with results being expressed as percentage of uptake in the absence of the competing unlabeled transferrin. There was found to be no significant difference between transferrin and iron uptake from either transfenin by the individual cell types. Results shown are for Belgrade reticulocytes in Figure 1; however, control reticulocytes showed a similar pattern. These results indicate that unlabeled Belgrade transferrin inhibits uptake of iron and transferrin from labeled Wistar transferrin as effectively as it does from labeled Belgrade transferrin. Similarly, unlabeled Wistar transferrin inhibits iron and transferrin uptake equally from both types of labeled transferrin. The uptakes of transferrin and iron by Belgrade rat reticulocytes from either Wistar or Belgrade transferrin were found to be identical, as was the uptake by the control reticulocytes (Fig. 2A and B). However, Bel-
grade reticulocytes displayed a greatly diminished ability to internalize iron, being only 12% as efficient as the Wistar reticulocytes. Iron uptake by both cell types was linear throughout the 20 min incubation, whereas transferrin uptake was initially rapid, slowing markedly before plateauing after approximately 8 min. Again, control reticulocytes showed higher levels of transferrin uptake, approximately 2.4 times that of Belgrade reticulocytes. Thus transferrin endocytosis was slower in Belgrade than in control reticulocytes, but the relative decrease was only about 30% as great as for iron uptake. A similar difference in transferrin uptake by Belgrade and control reticulocytes, was noted previously and was attributed to an effect of iron deficiency, not the inherited defect, on the cells [4].Non-Belgrade iron-deficient reticulocytes were found to take up transferrin at the same rate as Belgrade reticulocytes, but accumulated iron at three times the rate [4].Hence iron-deficient reticulocytes are probably a better control for studies with Belgrade reticulocytes than are “control” reticulocytes. Iron-deficient reticulocytes were therefore used in the experiments described below.
12
Farcich and Morgan
f:
1
/
200i
30
-
150
l 50
-
0
0
5
:
10 Time (min)
E
15
20 0
1
2
3
4
Iron Concentration (pM)
I 150
5
10 Time (min)
15
20
Fig. 3. Uptake of nontransferrin-bound Fe(ll) by rat reticulocytes into stoma1 (A) and cytosolic (B) fractions of irondeficient rat reticulocytes ()., control rat reticulocytes ( A ) and homozygous Belgrade rat reticulocytes (0). Results are mean & S.E.M. for five trials.
-
l5I 10
5 0
0
1
2
3
4
Iron Concentration (pM)
Uptake of Nontransferrin-Bound Iron
Fe(I1) uptake into stromal and cytosolic fractions for all cell types occurred in a linear manner for at least 20 min (Fig. 3A and B). Iron-deficient reticulocytes had significantly enhanced Fe(I1) uptake into both fractions compared with control cells. Conversely, Belgrade reticulocytes after 20 min had only about 18% as much iron in the stromal fraction as the control reticulocytes and 8.9% as much as the iron-deficient cells. The cytosolic uptake was also reduced in the Belgrade cells, the rate being about 13.5% that of the control and 7.6% that of iron-deficient reticulocytes . When reticulocytes were incubated in medium containing increasing concentrations of iron the rate of Fe(I1) uptake into both stromal and cytosolic fractions for all cell types showed evidence of saturation at about 1 p M iron concentration, but continued to increase in a linear fashion when iron concentration was further raised (Fig. 4A and B). This suggests that both a specific saturable uptake process and a nonspecific nonsaturable process are involved. These two components of the uptake curve were derived assuming an equation for the curve of ax f hxlc x. The nonspecific component, a (the slope of the linear portion of the uptake curve), was then subtracted from the total uptake curve to give the specific component. The specific portion of each curve was further analysed by Eadie-Hofstee plots to determine the
+
Fig. 4. Effect of varying the Fe(ll) concentration in the incubation medium on the rate of iron uptake into stromal (A) and cytosolic (B) fractions of iron-deficient rat reticulocytes (m), control rat reticulocytes ( ), and homorygous Belgrade rat reticulocytes (0). Results are mean r S.E.M. for five trials.
maximum rate of Fe(I1) uptake (V,,,) and the Michaelis constant (Km). A summary of these results is given in Table I. They indicate that Belgrade reticulocytes have a lower ability to take up Fe(I1) into both cytosol and stroma than either the control or iron-deficient cells. Maximum uptake into either fraction was less than 40% of that by the other two cell types. If it is assumed that for the cytosolic uptake the V,, values represent the activity of a membrane carrier for iron, then the Belgrade rat reticulocytes have lower functional activity of this carrier than either the control or the iron-deficient reticulocytes. DISCUSSION
The results indicate that the anemia of the Belgrade laboratory rat is unlikely to be due to defective transfemn function. The transferrin purified from the plasma of homozygous Belgrade rats was of the same M, as transferrin from Wistar plasma and also displayed identical electrophoretic mobility on the cellulose acetate, implying that the two proteins have the same net charge.
iron Uptake by Belgrade Rat Reticulocytes
13
TABLE 1. Vmaxand K, Values for Specific Fe(ll) Uptake*
Reticulocytes Belgrade Control Iron deficient
V,,, (nmol/ml reticulocytes per min) cytosol Stroma 1.23 f 0.17 3.08 0.12 3.38 f 0.13
*
5.02 f 0.36 14.7 f 0.71 16.5 f 0.83
K m (PM) Cytosol
Stroma
0.16 f 0.03 0.08 k 0.01 0.12 f 0.01
0.31 f 0.01 0.33 0.05 0.17 f 0.02
*
*The results show the maximal rates (V,,,,,) and Michaelis constants (K,) for the specific (saturable) component of Fe(I1) uptake into cytosolic and stromal fractions of Belgrade, control, and iron-deficient reticulocytes. The results are the means (fSEM) of five trials.
Furthermore, the two transferrins were shown to be equally favored with respect to both iron donation and transferrin binding to rat reticulocytes as revealed by the competition and direct uptake experiments. Hence, as assessed by these techniques, the transferrins are functionally identical. Although the transferrin from homozygous Belgrade rats appears to be functionally normal, the reticulocytes from such animals were found to have decreased ability to take up iron in either its free state or bound to transferrin. Previous studies showed that the transferrin endocytosis cycle is qualitatively and quantitatively normal in Belgrade reticulocytes compared with iron-deficient reticulocytes but iron fails to accumulate within the cells, implying that there is impaired passage of iron across the endosomal membrane [4]. The present results with transferrin-bound iron support this conclusion, and those with nontransferrin-bound Fe(I1) provide strong direct evidence for the presence of a defective irontransport mechanism in the membrane of Belgrade reticulocytes. Earlier studies with rabbit reticulocytes detailed the ability of such cells to take up nontransferrin-bound Fe(I1) in vitro. The mechanism for this was found to be saturable, pH and temperature dependent, and inhibited by ions of other transition metals. It was concluded that the iron was transported into the cells by a carriermediated process [ 5 ] .The results presented here indicate the presence of a similar iron-uptake system in rat reticulocytes, saturating when the iron concentration is about 0.5 pM. As with rabbit reticulocytes, this was accompanied by a nonsaturable process, which was apparent at the higher iron concentrations. The uptake of nontransferrin-bound iron by Belgrade reticulocytes was found to be less than that by both the iron-deficient and the control reticulocytes. The uptake into the cytosol by Belgrade cells was only 13.5% and 7.6%and the uptake into the stromal fraction of the cells (which consists of both intracellular organelles and outer cell membrane) was about 18% and 8.9% of that by control and iron-deficient reticulocytes, respectively.
These results indicate both impaired transport of ferrous ions into the reticulocytes of the homozygous Belgrade rats in vitro and lower cell surface binding of the Fe(1I) by these cells compared with the other reticulocytes. The amount of iron incorporated into the stromal and cytosolic fractions of the control and iron-deficient reticulocytes was similar at all time points. However, the cytosolic uptake of Fe(I1) into the Belgrade rat reticulocytes was only 70% that into the stromal fraction, suggesting a lower ability of iron passage across the reticulocyte cell membrane of the homozygous Belgrade rat compared with both iron-deficient and control cells. Such evidence further supports the view that the basic defect in the iron metabolism in the Belgrade rat is one of impaired membrane transport of iron. The V,,, values for the specific (saturable) uptakes of Fe(I1) by Belgrade reticulocytes into the cytosol and stroma individually were both about 30% and 35% of the values for the iron-deficient cells and control cells, respectively. Thus the specific uptake of ferrous ions into the cytosol of Belgrade reticulocytes is no more impaired than the specific binding of iron to the cell membranes. The specific site of binding of the Fe(1I) is as yet unknown, but, if a protein carrier is involved in the uptake of Fe(I1) by rat reticulocytes, then the above result implies that the protein itself is deficient in binding iron at the cell suface as well as in allowing the passage of the iron into the cells of the Belgrade rat. In fact, of the total maximal specific uptake of Fe(II), about 20% was found in the cytosol in all three cell types. This suggests that the specific process of iron uptake is qualitatively similar in the three reticulocyte populations and that it is the quantitative parameters that distinguish them, Belgrade reticulocytes having far lower levels of Fe(I1) transport across the cell membrane than the control and irondeficient cells. ACKNOWLEDGMENTS
This work was supported by a grant from the Australian Research Council.
14
Farcich and Morgan
REFERENCES I . Sladic-Simic D, Martinovitch PN, Zwkovic N , Pavic D. Martinovic J : A thalassemic-like disorder in Belgrade laboratory rats. Ann NY Acad Sci 165:93-99, 1969. 2. Edwards JA, Sullivan AL. Hoke JE: Defective delivery of iron to the developing red cells of the Belgrade laboratory rat. Blood 55:645-648, 1980. 3. Edwards JA, Huebers H . Kunzler C, Finch CA: Iron metabolism in the Belgrade rat. Blood 67:623-628, 1986. 4. Bowen B, Morgan EH: Anemia of the Belgrade rat: Evidence for defective membrane transport of iron. Blood 70:3844. 1987.
5. Morgan EH: Membrane transport of non-transfernin-bound iron by reticulocytes. Biochim Biophys Acta 943:428-439, 1988. 6. Hanks JH, Wallace RE: Relation of oxygen and temperature in the preservation of tissues by refrigeration. Proc Sac Exp Biol Med 7 1 :196200. 1949. 7. Lim BC, Morgan EH: Transfenin endocytosis and iron uptake by developing erythroid cells in the chicken (Gaflusdomesticus). J Camp Physiol B155:201-210, 1985. 8. Laemmli UK: Cleavage of structural proteins during the assembly of the head bacteriophage T4. Nature 227:68G685. 1970.
