British journal oJ Haematology, 1979,42, 255-267,
Ferritin as a Cytosol Iron Transport Intermediate in Human Reticulocytes BARBARA E. SPEYER A N D J. FIELDING Department of Haematology, St Mary’s Hospital, London (Received 29 March 1978; accepted f o r publication
11
October 1978)
SUMMARY. The major iron-bearing cytosol components of human reticulocytes identified after incubation with 5yFe-1251-transferrinhave been studied further. Component C previously found to behave consistently as an intermediate in the iron transport pathway to haem is shown to consist entirely of ferritin. After a short pulse of labelled transferrin incubation, chase experiments showed a fall of ferritin label with time and a corresponding increase in haemoglobin-iron incorporation. There was no loss of ferritin to the culture medium. Restriction of iron uptake by reticulocytes using both p-hydroxymercuribenzoate inhibition of uptake and incubation with progressively lower saturations of iron-transferrin gave linearly related incorporation of 5yFeinto ferritin and haemoglobin at all levels of iron uptake, thus negating the concept of ferritin as an ‘overspill’form of reticulocyte iron. The results suggest that cytosol ferritin is an obligatory intermediate in reticulocyte iron transport.
The location of iron within the hollow spheroidal protein shell of ferritin gives visual emphasis to the well-established view of its function as an iron storage protein (Granick, 195 I ; Crichton, 1973; Harrison el al, 1977). Relatively little exchange takes place between plasma iron and storage iron compared with the rapid movement of plasma iron into other body iron pools (Pollycove & Mortimer, 1961;Finch et al, 1965).In conditions of positive iron balance, iron in excess of metabolic requirement is incorporated into ferritin in liver and other tissues; in conditions of negative iron balance ferritin iron is mobilized and utilized. During intestinal absorption, iron in excess of metabolic requirements is taken up by mucosal cells and held in temporary storage as ferritin, until lost with exfoliated mucosa (Conrad & Crosby, 1963). A more active metabolic role for ferritin has been postulated. Mazur & Carleton (1963) isolated labelled haem and labelled ferritin from rat bone marrow and from peripheral blood reticulocytes after incubation with 5yFe-labelledserum, and suggested that ferritin behaved as an intermediate in iron transport between transferrin and haem. However, later workers, who also identified labelled ferritin in cytosol fractions after incubation of marrow cells (Zail et al, 1964; Primosigh & Thomas, 1968) or reticulocytes (Borovi et all I973), failed to show that it acted as a precursor of haem iron. In these and other studies (Allen & Jandl, 1960; Greenough et all 1962) unidentified iron-containing fractions, some of low molecular weight, which behaved kinetically as intermediates, were also observed in cytosol preparations. Correspondence: Dr J. Fielding, S t Mary’s Hospital, Harrow Road, London W9 3RL. 0007-1048/79/0600-0255$02.0001979 Blackwell Scientific Publications 255
256
Barbara E. Speyer a n d j . Fielding
Recently emphasis has been placed o n the heterogeneity of the ferritin molecule. Isoferritins variously distributed in different tissues have been described (Richter, 1965; Drysdale rt al, 1977); fcrritin of varying iron content varies also in the ease with which iron is taken up or mobilized (Harrison et a l , 1974); Yamada & Gabuzda (1974) have shown that ferritin of low iron content in red cell precursors is metabolically highly active. Mechanisms which fit the ferritin molecule for an autocatalytic and active metabolic role have been postulated from studies of its structure and i M vitro reactivity (Harrison et al, 1977; Crichton rt a!, 1977). Wc havc described techniques for the fractionation of whole reticulocytes after incubation with 59Fe-1251-transferrin (Speyer & Fielding, 1974). Membrane 59Fe was separated into two components, A and B, and the latter furthcr separated into two sub-components. Bz has been recognized as the complex of transferrin and its membrane receptor (Speycr & Fielding, 1974; Fielding & Speyer, 1974; Van der Heul et al, 1977). Two labelled components were found in thc cytosol: haemoglobin and component C. In five chase experiments (Fielding & Speyer, 1974) it was shown that component C represented the major cytosol iron transport intermediate of reticulocytes. It eluted in the void volume of Sephadex G-zoo columns and was thus of relatively high molecular weight. We therefore investigated further the nature of this intermediate. Wc report here further studies of cytosol component C which show it to be ferritin. O u r results indicate that human reticulocyte ferritin acts not merely as an iron storage protein, but plays an essential role in iron transport on thc hacm pathway. A preliminary outline of this work has been reported (Fielding & Speyer, 1977). MATERIALS A N D METHOIIS Purified human transferrin (Hoechst Pharmaceuticals Ltd) and human reticulocytes from patients with‘pernicious anaemia rcsponding to vitamin therapy were used in all experimetits. Methods were as previously described (Speycr & Fielding, 1974; Fielding & Speyer, 1974) except for the modifications described below. Chase experiments. Sodium bcnzylpenicillin BP, 30 pg/ml (Glaxo Laboratories Ltd) was added to the Hanks’ solution. Incubation of rcticulocytes for 10-15 min with labelled transferrin a t 67% iron saturation was followed by a chase with non-radioactive iron-transferrin for 0-0 min. Supernatants from the first centrifugation step following the chase were kept for identification of radioactive compounds eluting from thc cells during the chase. Chase experiments with reticulocytes inhibited by p-hydroxymercuribrnzoate ( P M B ) . Rcticulocyte suspensions were preincubated with 0 . 3 mM PMU, washed with Hanks’ solution, incubated with labelled iron-transferrin, and chased with non-radioactive iron-transfcrrin by the procedures previously described. A control reticulocyte suspension was put through the same procedures as the test except that PMU was omitted from the preincubation mixture. Cytosol fractionation. Cytosol was fractionated cither on Sephadex G-zoo as previously described or on Sepharose 6B. For Sepharose 613 fractionation, 3 ml portions of cytosol were passed by upward flow through columns 150 cm x 1.6 cm diameter previously equilibrated with 50 mM sodium phosphate buffer, pH 7.4, and followed with the same buffer. Fractions were collected in 2.4 ml volumes at a flow rate of 6 ml/h. Membranefractionation. Washed membranes were solubilized in I YO(v/v) Triton X-IOOand
Ferritin as an Iron Transport Intermediate
257 passed through columns of Sepharose t U to separate components A and B as previously described. For thc purposes of these experiments it was not necessary to separate component I3 into its two sub-components Br and B2. lncubation of reticulocytes with transferrin of varying iron saturation. In these experiments, in order to ensure equivalence of the iodinated transferrin a t varying iron saturation, transferrin was first iodinated before iron labelling, in contrast with our usual procedure of adding 59Fe before iodination. An aliquot of the iodinated protein solution was used for a spectrophotometric titration against FeC13 a t 465 nm to establish the amount of iron needed for 100% saturation. 59FeC13was then added to further aliquots to achieve 8%, 17%, 33 % and 67% iron saturation. Non-radioactive transferrin solutions at the same concentrations and at the same levels of iron saturation were also prepared. Reticulocyte-rich erythrocytes were washed, resuspended in an equal volume of Hanks’ solution, and divided into four aliquots. T o stabilize cell kinetics a t varying iron saturations, the aliquots were incubated a t 37°C for 20 min with non-radioactive iron transferrin at 8 % , 17%, 3 3 % and 67% saturation respectively, after which the cells were washed once with Hanks’ solution at 4”C, resuspended in Hanks’ solution, and incubated a t 37°C for 20 min with 59Fe-i251-transferrin at the same varying iron saturations. Final washing, haemolysis and fractionation were as previously described. Antibody precipitation. Rabbit antisera to human liver ferritin and to human transferrin (Hoechst Pharmaceuticals Ltd) were used. To avoid possible exchange of 59Fe from components other than ferritin which might lead to a false positive result, carrier ferritin was not added. T o each t ml of 59Fe containing solution, 0.2 ml antiserum was added. The solutions were incubated for I h at 37”C,left overnight a t 4OC then centrifuged a t 500 g for 1 0 min. Supernatant and antibody-antigen precipitate were separated, and the 59Fc activity of each determined. 59Feand lZ5I activity are shown in the tables and graphs after correction to a standard by which the two activities in the original labelled transferrin are equalized (except where varying saturations of transferrin are compared, in which case the absolute activities are shown).
RESULTS Fractionation of Reticulocyte Cytosol on Sepharose 6B Sepharose 6B fractionation of cytosol prepared from reticulocytes after 10min incubation with 59Fe-i251-transferrin is shown in Fig I . T w o iron-bearing components emerged, component C and haemoglo bin. As on Sephadex G-200, component C emerged as a single peak on Sepharose 6B but now well separated from the void volume. It occupied the same position as marker ferritin. Six separate fractionations through Sepharose 6B all produced a single peak within the column volume in the same position as marker ferritin. When the fractions of these peaks were tested with rabbit anti-human fertitin serum, 8 7 9 6 % of the 59Fe activity was precipitated, whereas none was precipitated by anti-transferrin serum. A small amount of 1251 emerged in the early emergent limb of the haemoglobin peak in the same position as marker transferrin. This 1251 precipitated with anti-transferrin serum and clearly represented small amounts of free transferrin. The amount of 59Fe co-precipitated by
Barbara E. Speyer a n d j . Fielding Hb
0.8 -
0.6
0.4 v)
\ In
:0.2
'
0
V
0
-
O'"030
1 52
t
v.v
74
I18
96
t
Fn
t
140
no.
Tf
FIG I . Fractionation of reticulocyte cytosol on Sepharose 6B. v.v.=void volume; Fn=marker ferritin; Tf= marker transferrin. The two main iron bearing peaks are ferritin and haemoglobin.
the anti-transferrin serum accounted for only 2% of the 5yFein the corresponding fractions, and only a negligible proportion of the 5yFe in the whole haemoglobin peak.Thus cytosol contained labelled ferritin, labelled haemoglobin and a small amount of free transferrin. No other labelled components were detected.
Chase Experiments The fate of the components after reticulocyte incubation for 10 min with 5yFe-1251-transfrrrin was followed after varying times of further incubation with non-radioactive iron-transfcrrin (chase) and subsequent separation into membrane and cytosol fractions. Fig 2 shows cytosol fractionation after times of chase varying from o to 90 min; at all times only one peak of 59Fe in addition to the haemoglobin peak was observed, and its position in evcry case corresponded to that of marker ferritin. The identification of these peaks as ferritin was confirmed with anti-ferritin serum (Table I): 8 3 9 6 % of 5yFefrom three pooled fractions at the apex of these peaks from each of the five columns was precipitated by anti-ferritin serum, and less than 10% by anti-transferrin serum. Fig 3 shows the sum of the 59Fe counts in each cytosol iron-bearing component at varying times of chase. Haemoglobin counts were obtained by summing the counts in fractions 93-120 of Fig 2 and ferritin counts by summing the counts in fractions 7092.Total haemoglobin increased linearly with time; total 5yFe-ferritin increased during the early phase and later decreased reciprocally with the increase of haemoglobin. The 1251 counts in transferrin (obtained by summing the 1251 counts in fractions 93-120) are also plotted on the graph after correction by a factor which equalized the 5yFeand 1251 counts of the original transferrin used in the experiment. Recoveries of l Z 5 I from the Sepharose 6B
Ferritin a s a n Iron Transport Intermediate
259
r
el 128I
O**[ 0
55
35
?
?
f
135
Fraction no.
Tf
Fn
V.V.
I I5
95
75
FIG2. Fractionation of cytosols o n Sepharose 6B at varying times of chase after 10min incubation of reticulocytes with 59Fe-1251-transferrin: zero time, 0;7 min, A; 90 min, O. Ferritin counts increase initially, then fall; Haemoglobin counts increase progressively.
TABLE I. Identification of 59Fepeak from Sepharose 6B fractionation of cytosol at varying times of chase. Sepharose 6B fractions from reticulocyte cytosol prepared at varying times of chase after incubation with 59Fe-'251-transferrin were tested with anti-ferritin and with anti-transferrin sera. The marker ferritin peak emerged in fraction 78.
59Fe precipitated
Time of chase Fraction numbers Anti-frritin serum Anti-transferrin serum (min) 0
7 I7 40
90
59Fepeak
(%I
(%)
78-80 78-80 79-8 I 77-79 79-8 I
87 96 83 84 94
< 10 < I0 < 10
.W. & J A N I N , J.H. (1960)Kinetics of iiitraALLLN, cellular iron in rabbit reticulocytes. Blood, 15, 71-81. BAWI I.I:TT, G.R.(1976)Phosphate compounds in rat erythrocytes and reticulocytes. Biochemical and Biophysical Research Communications, 70, 105~-1062. B o ~ o v bJ., , P O ~ K P. A ,& NEuw1RT.J. (1973)Study of intracellular iron distribution in rabbit reticulocytes with normal and inhibited heme synthesis. Biochimica et Biophysica Acta, 320, 143-156. CAPFI I OS, C. & BILLSKI, B.H.J.(1972)Kinetic System. Wiley-Interscience, London. C O N R A I I , M.E., J R & C R O ~ HW.H. Y , (1963)Intestinal mucosal mechanisms controlling iron absorption. Blood, 22, 406-415. CRi(.muN, R.R. (1973)Structure and function offerritin. Angewandte C h a i i e , I n t e r n a r i o d Edition, 12, 574s. CRI(:IFI ON, R.R., COLLET-CASSAHI., D., PONCF-ORTIZ, Y., WALKERS, M., ROMAN,F. & PAQUES,E. (1977) Ferritin: comparative structural studies, iron depogition and mobilization. Proteins of Iron Metabolisrn (cd. by E. B. Brown, P. Aiscn, J . Fielding and R. R. Crichton), p.13. Grune & Stratton, London. r)RYSDALt,J.W.,Alltl M A N , T.G., AROSIo, P., C A F A R t A I . E , I]., FITZPATRICK l’., HAZARD, J.T. & YOKOTA, M .(1977)Human isoferritins in normal and disease states. Seniinars in Haemafology, 14, 71 8 8 . El)WARm, S.A. & Fir1 I I I N G , J. (1971) Studies of the effect of sylphydryl and other inhibitors on reticulocytc uptake of doubly-labelled transferrin. Brirish Journal qf’Haetwatoluyy, 20, 405-416. F I ~I HI N G ,J . & SPEYER, B.E. (1974)Iron transport intermediates in human reticulocytes and the membrane binding site of iron-transferrin. Biochimica et Biophysics Acta, 363, 387-396. FIELUING,J. & SPFYFR, U.E. (1977)Ferritin as a cytosol transport protein. Proteiris qflron Metabolism (ed. by E.B. Brown, P. Aisen, J . Fie1din.g and R.R. Crichton), p. 3 1 1 . Grune & Stratton, London. FINCH,C.A., HOSAIN, F., MORGAN, E.H., MARSA(:IIA, G., G I H LE.~ &, HILLMAN R. (1965)Theferrokinetic approach to anaemia. Series Haernatologicn, 6, 30-40. G A H U Z D A , T.G. & P ~ A R S OJ. N ,(1969)Metabolic and molecular heterogeneity of marrow ferritin. Binrhimica er Biophysica Acfa,194, 50-54. G I ASS, J., NUNEZ, M.T. & ROBINSON, S.H. (1977)Iron
transport from Sepharose-bound transferrin. Biochemical and Biophysical Research Conitnirnications, 75, 226-232. GRANICK, S. (195I ) Structure and physiological functions of ferritin. Physio/oyical Reviews, 31, 489-51 I. GRFFNOUGH, W.B.. 111, Pt-ILRS,T., J R& THOMAS, E.U. (1962) An intracellular protein intermediate for hemoglobin formation.Joirrna/ qf Clinical Investi‘qation, 41, 1116-1 124. HARRISON, P.M., BANYAN), S.H., HoARr, R.J., RusSELL, S.M. & T R t r r R Y A. (1977) The structure and function of ferritin. Iron Metabolism (ed. by R. Porter and U . W . FitzSimons). Ciba Foirndation Symposium 5 1 (new series), pp. 19-40. Elsevier, Amsterdam. HARRISON, P.M., HOY, T.G., MACARA, I.G. & HOARF, R.J. (1974)Ferritin iron uptake and release. Structure-function relationships. Biochemical Journal, 143,445-451. HEMMAPLARDH, D. & MORGAN, E.H. (1977)The role of endocytosis in transferrin uptake by reticulocytes and bone marrow cells. British ]ourrial qf HaemafoIofy, 36. 8996. JACOBS, A. (1977)Low molecular weight intracellular iron transport compounds. Blood, 50, 433-439. LOH,T.T., YEUNG,Y.G. & YEUNG, D.(1977)Transferrin and iron uptake by rabbit reticulocytes. Biochimica et Biophysica Acta, 471, 118-124. MARTINtL-Mtl)FI [.IN, J. R: SCHULMAN, H.M. (1972) The kinetics of iron and transferrin incorporation into rabbit erythroid cells and the natureof stromalbound iron. Biochirnica et Biophysica Acta, 264, 272-2 84. MAZUR, A . & CARLETON, A . (1963)Relationofferritin iron to heme synthesis in marrow and reticulocytes. Journal ofBiological Chemistry, 238, rX17-r824. MORGAN, E.H. & AIVLLTON, T.C. (1969)Autoradiographic localization of 1251-labelled transferrin in rabbit reticulocytes. Nature, 223, 1371-1372. MORGAN, E.H. & B A K E R E., (1969)Theeffectofmetabolic inhibitors on transferrin and iron uptake and transferrin release from erythrocytes. Biochimica et Biophysica Acta, 184,442-454. POLLYCOVE, M. & MORrIMER, R. (1961)The quantitative determination of iron kinetics and hemoglobin synthesis in human subjects._loirmalqf Clinical Invest i p t i o n , 40.753-782. P O ~ K AP., , NEUWIRT, J., BOROVA, J. & FUCHS,0.
Ferritin as an Iron Transport Intermediate (1977) In: Proteins ofIron Metabolism (ed. by E. B. Brown, P. Aisen, J. Fielding and R . R. Crichton), p.3 19. Grune & Stratton, London. PRIMOSIGH,J.V. & THOMAS, E.D. (1968) Studieson the partition of iron in bone marrow cells. Journal of Clinical Investigation, 47, 1473-1482. RICHTER,G.W. (1965) Comparison of ferritins from neoplastic and non-neoplastic human cells. Narure, 207,616d18. B.E. & FIELDING, J. (1974) Chromatographic SPEYER, fractionation of human reticulocytes after uptake of doubly-labelled [59Fe,'251]transferrin. Biorhimira et Biophysica Acra, 332, 192-200. STORRING, P.L. & FATIH,S. (197s) Erythropoietin effects on iron metabolism in rat bone marrow cells. Biorhimica et Biophysira Acra, 392, 26-38. M.J. & V A N EIJK,H.G. VAN DER HEUL,C., KROOS,
267
(1977) O n the binding sites o f iron-transferrin on rat reticulocytes. Proteins oflron Metabolirm (ed. by E. B. Brown, P. Aisen, J. Fielding and R. R. Crichton), p.299. Grune & Stratton, London. VANSTEVENINCK, J . , WEED,R.I. & ROTHSTEIN, A. (1965) Localization of erythrocyte membrane sulphydryl groups essential for glucose transport.journu1 of General Physiology, 48, 6 1 7 6 3 2 . H. & GARUIDA, T.G. (1974) Erythroblast YAMADA, ferritin: synthesis, structure, and function in developing erythroid cells. Journal of Laboratory and Clinical Medicine, 83, 478-488. R.W., TORRANCE, J.D. & Z A I I ,S.S., CHARLTON, BOTHWELL, T.H. (1964)Studieson the formation of ferritin in red cell precursors. Journal of Clinical Inverfiprion, 4 3 , 6 7 4 8 0 .
Ferritin as a Cytosol Iron Transport Intermediate in Human Reticulocytes BARBARA E. SPEYER A N D J. FIELDING Department of Haematology, St Mary’s Hospital, London (Received 29 March 1978; accepted f o r publication
11
October 1978)
SUMMARY. The major iron-bearing cytosol components of human reticulocytes identified after incubation with 5yFe-1251-transferrinhave been studied further. Component C previously found to behave consistently as an intermediate in the iron transport pathway to haem is shown to consist entirely of ferritin. After a short pulse of labelled transferrin incubation, chase experiments showed a fall of ferritin label with time and a corresponding increase in haemoglobin-iron incorporation. There was no loss of ferritin to the culture medium. Restriction of iron uptake by reticulocytes using both p-hydroxymercuribenzoate inhibition of uptake and incubation with progressively lower saturations of iron-transferrin gave linearly related incorporation of 5yFeinto ferritin and haemoglobin at all levels of iron uptake, thus negating the concept of ferritin as an ‘overspill’form of reticulocyte iron. The results suggest that cytosol ferritin is an obligatory intermediate in reticulocyte iron transport.
The location of iron within the hollow spheroidal protein shell of ferritin gives visual emphasis to the well-established view of its function as an iron storage protein (Granick, 195 I ; Crichton, 1973; Harrison el al, 1977). Relatively little exchange takes place between plasma iron and storage iron compared with the rapid movement of plasma iron into other body iron pools (Pollycove & Mortimer, 1961;Finch et al, 1965).In conditions of positive iron balance, iron in excess of metabolic requirement is incorporated into ferritin in liver and other tissues; in conditions of negative iron balance ferritin iron is mobilized and utilized. During intestinal absorption, iron in excess of metabolic requirements is taken up by mucosal cells and held in temporary storage as ferritin, until lost with exfoliated mucosa (Conrad & Crosby, 1963). A more active metabolic role for ferritin has been postulated. Mazur & Carleton (1963) isolated labelled haem and labelled ferritin from rat bone marrow and from peripheral blood reticulocytes after incubation with 5yFe-labelledserum, and suggested that ferritin behaved as an intermediate in iron transport between transferrin and haem. However, later workers, who also identified labelled ferritin in cytosol fractions after incubation of marrow cells (Zail et al, 1964; Primosigh & Thomas, 1968) or reticulocytes (Borovi et all I973), failed to show that it acted as a precursor of haem iron. In these and other studies (Allen & Jandl, 1960; Greenough et all 1962) unidentified iron-containing fractions, some of low molecular weight, which behaved kinetically as intermediates, were also observed in cytosol preparations. Correspondence: Dr J. Fielding, S t Mary’s Hospital, Harrow Road, London W9 3RL. 0007-1048/79/0600-0255$02.0001979 Blackwell Scientific Publications 255
256
Barbara E. Speyer a n d j . Fielding
Recently emphasis has been placed o n the heterogeneity of the ferritin molecule. Isoferritins variously distributed in different tissues have been described (Richter, 1965; Drysdale rt al, 1977); fcrritin of varying iron content varies also in the ease with which iron is taken up or mobilized (Harrison et a l , 1974); Yamada & Gabuzda (1974) have shown that ferritin of low iron content in red cell precursors is metabolically highly active. Mechanisms which fit the ferritin molecule for an autocatalytic and active metabolic role have been postulated from studies of its structure and i M vitro reactivity (Harrison et al, 1977; Crichton rt a!, 1977). Wc havc described techniques for the fractionation of whole reticulocytes after incubation with 59Fe-1251-transferrin (Speyer & Fielding, 1974). Membrane 59Fe was separated into two components, A and B, and the latter furthcr separated into two sub-components. Bz has been recognized as the complex of transferrin and its membrane receptor (Speycr & Fielding, 1974; Fielding & Speyer, 1974; Van der Heul et al, 1977). Two labelled components were found in thc cytosol: haemoglobin and component C. In five chase experiments (Fielding & Speyer, 1974) it was shown that component C represented the major cytosol iron transport intermediate of reticulocytes. It eluted in the void volume of Sephadex G-zoo columns and was thus of relatively high molecular weight. We therefore investigated further the nature of this intermediate. Wc report here further studies of cytosol component C which show it to be ferritin. O u r results indicate that human reticulocyte ferritin acts not merely as an iron storage protein, but plays an essential role in iron transport on thc hacm pathway. A preliminary outline of this work has been reported (Fielding & Speyer, 1977). MATERIALS A N D METHOIIS Purified human transferrin (Hoechst Pharmaceuticals Ltd) and human reticulocytes from patients with‘pernicious anaemia rcsponding to vitamin therapy were used in all experimetits. Methods were as previously described (Speycr & Fielding, 1974; Fielding & Speyer, 1974) except for the modifications described below. Chase experiments. Sodium bcnzylpenicillin BP, 30 pg/ml (Glaxo Laboratories Ltd) was added to the Hanks’ solution. Incubation of rcticulocytes for 10-15 min with labelled transferrin a t 67% iron saturation was followed by a chase with non-radioactive iron-transferrin for 0-0 min. Supernatants from the first centrifugation step following the chase were kept for identification of radioactive compounds eluting from thc cells during the chase. Chase experiments with reticulocytes inhibited by p-hydroxymercuribrnzoate ( P M B ) . Rcticulocyte suspensions were preincubated with 0 . 3 mM PMU, washed with Hanks’ solution, incubated with labelled iron-transferrin, and chased with non-radioactive iron-transfcrrin by the procedures previously described. A control reticulocyte suspension was put through the same procedures as the test except that PMU was omitted from the preincubation mixture. Cytosol fractionation. Cytosol was fractionated cither on Sephadex G-zoo as previously described or on Sepharose 6B. For Sepharose 613 fractionation, 3 ml portions of cytosol were passed by upward flow through columns 150 cm x 1.6 cm diameter previously equilibrated with 50 mM sodium phosphate buffer, pH 7.4, and followed with the same buffer. Fractions were collected in 2.4 ml volumes at a flow rate of 6 ml/h. Membranefractionation. Washed membranes were solubilized in I YO(v/v) Triton X-IOOand
Ferritin as an Iron Transport Intermediate
257 passed through columns of Sepharose t U to separate components A and B as previously described. For thc purposes of these experiments it was not necessary to separate component I3 into its two sub-components Br and B2. lncubation of reticulocytes with transferrin of varying iron saturation. In these experiments, in order to ensure equivalence of the iodinated transferrin a t varying iron saturation, transferrin was first iodinated before iron labelling, in contrast with our usual procedure of adding 59Fe before iodination. An aliquot of the iodinated protein solution was used for a spectrophotometric titration against FeC13 a t 465 nm to establish the amount of iron needed for 100% saturation. 59FeC13was then added to further aliquots to achieve 8%, 17%, 33 % and 67% iron saturation. Non-radioactive transferrin solutions at the same concentrations and at the same levels of iron saturation were also prepared. Reticulocyte-rich erythrocytes were washed, resuspended in an equal volume of Hanks’ solution, and divided into four aliquots. T o stabilize cell kinetics a t varying iron saturations, the aliquots were incubated a t 37°C for 20 min with non-radioactive iron transferrin at 8 % , 17%, 3 3 % and 67% saturation respectively, after which the cells were washed once with Hanks’ solution at 4”C, resuspended in Hanks’ solution, and incubated a t 37°C for 20 min with 59Fe-i251-transferrin at the same varying iron saturations. Final washing, haemolysis and fractionation were as previously described. Antibody precipitation. Rabbit antisera to human liver ferritin and to human transferrin (Hoechst Pharmaceuticals Ltd) were used. To avoid possible exchange of 59Fe from components other than ferritin which might lead to a false positive result, carrier ferritin was not added. T o each t ml of 59Fe containing solution, 0.2 ml antiserum was added. The solutions were incubated for I h at 37”C,left overnight a t 4OC then centrifuged a t 500 g for 1 0 min. Supernatant and antibody-antigen precipitate were separated, and the 59Fc activity of each determined. 59Feand lZ5I activity are shown in the tables and graphs after correction to a standard by which the two activities in the original labelled transferrin are equalized (except where varying saturations of transferrin are compared, in which case the absolute activities are shown).
RESULTS Fractionation of Reticulocyte Cytosol on Sepharose 6B Sepharose 6B fractionation of cytosol prepared from reticulocytes after 10min incubation with 59Fe-i251-transferrin is shown in Fig I . T w o iron-bearing components emerged, component C and haemoglo bin. As on Sephadex G-200, component C emerged as a single peak on Sepharose 6B but now well separated from the void volume. It occupied the same position as marker ferritin. Six separate fractionations through Sepharose 6B all produced a single peak within the column volume in the same position as marker ferritin. When the fractions of these peaks were tested with rabbit anti-human fertitin serum, 8 7 9 6 % of the 59Fe activity was precipitated, whereas none was precipitated by anti-transferrin serum. A small amount of 1251 emerged in the early emergent limb of the haemoglobin peak in the same position as marker transferrin. This 1251 precipitated with anti-transferrin serum and clearly represented small amounts of free transferrin. The amount of 59Fe co-precipitated by
Barbara E. Speyer a n d j . Fielding Hb
0.8 -
0.6
0.4 v)
\ In
:0.2
'
0
V
0
-
O'"030
1 52
t
v.v
74
I18
96
t
Fn
t
140
no.
Tf
FIG I . Fractionation of reticulocyte cytosol on Sepharose 6B. v.v.=void volume; Fn=marker ferritin; Tf= marker transferrin. The two main iron bearing peaks are ferritin and haemoglobin.
the anti-transferrin serum accounted for only 2% of the 5yFein the corresponding fractions, and only a negligible proportion of the 5yFe in the whole haemoglobin peak.Thus cytosol contained labelled ferritin, labelled haemoglobin and a small amount of free transferrin. No other labelled components were detected.
Chase Experiments The fate of the components after reticulocyte incubation for 10 min with 5yFe-1251-transfrrrin was followed after varying times of further incubation with non-radioactive iron-transfcrrin (chase) and subsequent separation into membrane and cytosol fractions. Fig 2 shows cytosol fractionation after times of chase varying from o to 90 min; at all times only one peak of 59Fe in addition to the haemoglobin peak was observed, and its position in evcry case corresponded to that of marker ferritin. The identification of these peaks as ferritin was confirmed with anti-ferritin serum (Table I): 8 3 9 6 % of 5yFefrom three pooled fractions at the apex of these peaks from each of the five columns was precipitated by anti-ferritin serum, and less than 10% by anti-transferrin serum. Fig 3 shows the sum of the 59Fe counts in each cytosol iron-bearing component at varying times of chase. Haemoglobin counts were obtained by summing the counts in fractions 93-120 of Fig 2 and ferritin counts by summing the counts in fractions 7092.Total haemoglobin increased linearly with time; total 5yFe-ferritin increased during the early phase and later decreased reciprocally with the increase of haemoglobin. The 1251 counts in transferrin (obtained by summing the 1251 counts in fractions 93-120) are also plotted on the graph after correction by a factor which equalized the 5yFeand 1251 counts of the original transferrin used in the experiment. Recoveries of l Z 5 I from the Sepharose 6B
Ferritin a s a n Iron Transport Intermediate
259
r
el 128I
O**[ 0
55
35
?
?
f
135
Fraction no.
Tf
Fn
V.V.
I I5
95
75
FIG2. Fractionation of cytosols o n Sepharose 6B at varying times of chase after 10min incubation of reticulocytes with 59Fe-1251-transferrin: zero time, 0;7 min, A; 90 min, O. Ferritin counts increase initially, then fall; Haemoglobin counts increase progressively.
TABLE I. Identification of 59Fepeak from Sepharose 6B fractionation of cytosol at varying times of chase. Sepharose 6B fractions from reticulocyte cytosol prepared at varying times of chase after incubation with 59Fe-'251-transferrin were tested with anti-ferritin and with anti-transferrin sera. The marker ferritin peak emerged in fraction 78.
59Fe precipitated
Time of chase Fraction numbers Anti-frritin serum Anti-transferrin serum (min) 0
7 I7 40
90
59Fepeak
(%I
(%)
78-80 78-80 79-8 I 77-79 79-8 I
87 96 83 84 94
< 10 < I0 < 10
.W. & J A N I N , J.H. (1960)Kinetics of iiitraALLLN, cellular iron in rabbit reticulocytes. Blood, 15, 71-81. BAWI I.I:TT, G.R.(1976)Phosphate compounds in rat erythrocytes and reticulocytes. Biochemical and Biophysical Research Communications, 70, 105~-1062. B o ~ o v bJ., , P O ~ K P. A ,& NEuw1RT.J. (1973)Study of intracellular iron distribution in rabbit reticulocytes with normal and inhibited heme synthesis. Biochimica et Biophysica Acta, 320, 143-156. CAPFI I OS, C. & BILLSKI, B.H.J.(1972)Kinetic System. Wiley-Interscience, London. C O N R A I I , M.E., J R & C R O ~ HW.H. Y , (1963)Intestinal mucosal mechanisms controlling iron absorption. Blood, 22, 406-415. CRi(.muN, R.R. (1973)Structure and function offerritin. Angewandte C h a i i e , I n t e r n a r i o d Edition, 12, 574s. CRI(:IFI ON, R.R., COLLET-CASSAHI., D., PONCF-ORTIZ, Y., WALKERS, M., ROMAN,F. & PAQUES,E. (1977) Ferritin: comparative structural studies, iron depogition and mobilization. Proteins of Iron Metabolisrn (cd. by E. B. Brown, P. Aiscn, J . Fielding and R. R. Crichton), p.13. Grune & Stratton, London. r)RYSDALt,J.W.,Alltl M A N , T.G., AROSIo, P., C A F A R t A I . E , I]., FITZPATRICK l’., HAZARD, J.T. & YOKOTA, M .(1977)Human isoferritins in normal and disease states. Seniinars in Haemafology, 14, 71 8 8 . El)WARm, S.A. & Fir1 I I I N G , J. (1971) Studies of the effect of sylphydryl and other inhibitors on reticulocytc uptake of doubly-labelled transferrin. Brirish Journal qf’Haetwatoluyy, 20, 405-416. F I ~I HI N G ,J . & SPEYER, B.E. (1974)Iron transport intermediates in human reticulocytes and the membrane binding site of iron-transferrin. Biochimica et Biophysics Acta, 363, 387-396. FIELUING,J. & SPFYFR, U.E. (1977)Ferritin as a cytosol transport protein. Proteiris qflron Metabolism (ed. by E.B. Brown, P. Aisen, J . Fie1din.g and R.R. Crichton), p. 3 1 1 . Grune & Stratton, London. FINCH,C.A., HOSAIN, F., MORGAN, E.H., MARSA(:IIA, G., G I H LE.~ &, HILLMAN R. (1965)Theferrokinetic approach to anaemia. Series Haernatologicn, 6, 30-40. G A H U Z D A , T.G. & P ~ A R S OJ. N ,(1969)Metabolic and molecular heterogeneity of marrow ferritin. Binrhimica er Biophysica Acfa,194, 50-54. G I ASS, J., NUNEZ, M.T. & ROBINSON, S.H. (1977)Iron
transport from Sepharose-bound transferrin. Biochemical and Biophysical Research Conitnirnications, 75, 226-232. GRANICK, S. (195I ) Structure and physiological functions of ferritin. Physio/oyical Reviews, 31, 489-51 I. GRFFNOUGH, W.B.. 111, Pt-ILRS,T., J R& THOMAS, E.U. (1962) An intracellular protein intermediate for hemoglobin formation.Joirrna/ qf Clinical Investi‘qation, 41, 1116-1 124. HARRISON, P.M., BANYAN), S.H., HoARr, R.J., RusSELL, S.M. & T R t r r R Y A. (1977) The structure and function of ferritin. Iron Metabolism (ed. by R. Porter and U . W . FitzSimons). Ciba Foirndation Symposium 5 1 (new series), pp. 19-40. Elsevier, Amsterdam. HARRISON, P.M., HOY, T.G., MACARA, I.G. & HOARF, R.J. (1974)Ferritin iron uptake and release. Structure-function relationships. Biochemical Journal, 143,445-451. HEMMAPLARDH, D. & MORGAN, E.H. (1977)The role of endocytosis in transferrin uptake by reticulocytes and bone marrow cells. British ]ourrial qf HaemafoIofy, 36. 8996. JACOBS, A. (1977)Low molecular weight intracellular iron transport compounds. Blood, 50, 433-439. LOH,T.T., YEUNG,Y.G. & YEUNG, D.(1977)Transferrin and iron uptake by rabbit reticulocytes. Biochimica et Biophysica Acta, 471, 118-124. MARTINtL-Mtl)FI [.IN, J. R: SCHULMAN, H.M. (1972) The kinetics of iron and transferrin incorporation into rabbit erythroid cells and the natureof stromalbound iron. Biochirnica et Biophysica Acta, 264, 272-2 84. MAZUR, A . & CARLETON, A . (1963)Relationofferritin iron to heme synthesis in marrow and reticulocytes. Journal ofBiological Chemistry, 238, rX17-r824. MORGAN, E.H. & AIVLLTON, T.C. (1969)Autoradiographic localization of 1251-labelled transferrin in rabbit reticulocytes. Nature, 223, 1371-1372. MORGAN, E.H. & B A K E R E., (1969)Theeffectofmetabolic inhibitors on transferrin and iron uptake and transferrin release from erythrocytes. Biochimica et Biophysica Acta, 184,442-454. POLLYCOVE, M. & MORrIMER, R. (1961)The quantitative determination of iron kinetics and hemoglobin synthesis in human subjects._loirmalqf Clinical Invest i p t i o n , 40.753-782. P O ~ K AP., , NEUWIRT, J., BOROVA, J. & FUCHS,0.
Ferritin as an Iron Transport Intermediate (1977) In: Proteins ofIron Metabolism (ed. by E. B. Brown, P. Aisen, J. Fielding and R . R. Crichton), p.3 19. Grune & Stratton, London. PRIMOSIGH,J.V. & THOMAS, E.D. (1968) Studieson the partition of iron in bone marrow cells. Journal of Clinical Investigation, 47, 1473-1482. RICHTER,G.W. (1965) Comparison of ferritins from neoplastic and non-neoplastic human cells. Narure, 207,616d18. B.E. & FIELDING, J. (1974) Chromatographic SPEYER, fractionation of human reticulocytes after uptake of doubly-labelled [59Fe,'251]transferrin. Biorhimira et Biophysica Acra, 332, 192-200. STORRING, P.L. & FATIH,S. (197s) Erythropoietin effects on iron metabolism in rat bone marrow cells. Biorhimica et Biophysira Acra, 392, 26-38. M.J. & V A N EIJK,H.G. VAN DER HEUL,C., KROOS,
267
(1977) O n the binding sites o f iron-transferrin on rat reticulocytes. Proteins oflron Metabolirm (ed. by E. B. Brown, P. Aisen, J. Fielding and R. R. Crichton), p.299. Grune & Stratton, London. VANSTEVENINCK, J . , WEED,R.I. & ROTHSTEIN, A. (1965) Localization of erythrocyte membrane sulphydryl groups essential for glucose transport.journu1 of General Physiology, 48, 6 1 7 6 3 2 . H. & GARUIDA, T.G. (1974) Erythroblast YAMADA, ferritin: synthesis, structure, and function in developing erythroid cells. Journal of Laboratory and Clinical Medicine, 83, 478-488. R.W., TORRANCE, J.D. & Z A I I ,S.S., CHARLTON, BOTHWELL, T.H. (1964)Studieson the formation of ferritin in red cell precursors. Journal of Clinical Inverfiprion, 4 3 , 6 7 4 8 0 .
