Biochimica et Biophysica Acta, 1053 (1990) 1-12 Elsevier
]
BBAMCR 12693
The uptake of iron and transferrin by the human malignant melanoma cell D.R. Richardson and E. Baker Department of Physiology, Unioersity of Western Australia, Nedlands (Australia) (Received 26 September 1989) (Revised manuscript received 19 January 1990)
Key words: Melanotransferrin; Iron uptake; Transferrin; Iron; (Human melanoma cell)
The role of the transferrin homologue, melanotransferrin (p97), in iron metabolism has been studied using the human melanoma cell line, SK-MEL-28, which expresses this antigen in high concentrations. The mechanisms of iron and transferrin uptake were investigated using human transferrin labelled with iodine-125 and iron-59. Internalised and membrane-bound iron and transferrin were separated using the proteinase, pronase. The uptake of iron from transferrin occurred by at least two processes. The first process was saturable and consistent with receptor-mediated endocytosis, involving internalisation of transferrin bound to specific binding sites. Uptake of iron also occurred by a second process which was non-saturable up to 0.06 m g / m l (0.75 /tM) and was of higher efficiency than the saturable process. This process of iron uptake may be the dominant one at physiological serum transferrin concentrations. A membrane-bound, pronase-sensitive, temperature-dependent, iron-binding component was also identified. The number of binding sites was estimated to he approx. 340000 per cell (assuming 2 atoms of iron per site) and it is suggested that this binding component may be melanotransferrin.
Introduction All cells need iron (Fe) and neoplastic cells have a high requirement related to their rapid rate of proliferation [1]. This increased requirement is reflected by an increase in the expression of the transferrin receptor [2-4]. The metabolic pathways for transferrin and Fe uptake and release by neoplastic ceils may be different from normal cells [5-8]. The human malignant melanoma cell is of particular interest in this regard as these cells express high concentrations of a transferrin homologue, p97, or melanotransferrin, on their membrane as well as the transferrin receptor [9-11]. Melanotransferrin has been shown to have several characteristics in common with serum transferrin. These are: (i) it is a sialoglycoprotein (13% CHO); (ii) it has 37-39% sequence homology with human serum trans-
Abbreviations: FCS, fetal calf serum; MEM, minimal Eagle's medium; DFO, desferrioxamine; PIH, pyridoxal isonicotinoyl hydrazone; BSS, balanced salt solution; NTA, nitrilotriacetate; BSA, bovine serum albumin; gPR, gram protein; Tf, transferrin. Correspondence: E. Baker, Department of Physiology, University of Western Australia, Nedlands, Western Australia 6009, Australia.
ferrin, human lactoferrin and chicken transferrin; (iii) the melanotransferrin gene is on chromosome 3, as are those for transferrin and the transferrin receptor; (iv) it has a high interdomain sequence homology (46%), which is greater than that found for other transferrins (33417o); and (v) it can bind Fe from Fe(III) citrate complexes [12-16]. These similarities to transferrin have prompted Brown and colleagues to suggest that melanotransferrin may have a role in the Fe metabolism of the melanoma cell, although its functional significance in Fe transport has yet to be demonstrated. An increased supply of Fe via melanoma transferrin may increase the proliferation of melanoma cells, as Fe may be a rate-limiting nutrient for growth [1,17,18]. In the present work the Fe metabolism of the melanoma cell has been studied to determine the existence and relative importance of the two putative pathways of Fe uptake in melanoma cells: (i) uptake of Fe from transferrin via receptor-mediated endocytosis, and (ii) Fe uptake mediated by melanotransferrin. The mechanisms of Fe and transferrin uptake were determined by standard procedures using the nonspecific proteinase, pronase, to separate internalised (pronase-resistant) and membrane-bound (pronase-sensitive) Fe and transferrin [19,20]. The results indicate that these cells can obtain Fe from transferrin by a
0167-4889/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
saturable process consistent with receptor-mediated endocytosis. However, the amount of Fe internalised by the cell cannot be entirely accounted for by this process, and it has been concluded that the cell can remove Fe from transferrin by another mechanism. In addition, a membrane-bound, pronase-sensitive, temperature-dependent, Fe-binding component has been identified, the properties of which are consistent with those of melanotransferrin. Materials and Methods
Reagents Iron-59 (as ferric chloride in 0.1 M HCI) and iodine125 (as sodium iodide) were purchased from Amersham International, Amersham, U.K. Pronase was purchased from Boehringer Mannheim, Mt. Waverley, Australia. Eagle's modified minimum essential medium (MEM) as Autopow and fetal calf serum (FCS) were supplied by Flow Laboratories, Annandale, Australia. Penicillin (Crystapen-benzylpenicillin sodium B.P.) was obtained from Glaxo Aus. Pty Ltd., Boronia, Australia Bovine serum albumin (BSA; 98% pure, fatty acid free), human apotransferrin, L-glutamine and Hepes were obtained from Sigma Chemical Co., St. Louis, MO, U.S.A. Nonessential amino acids (100 x concentrate) and trypsinversene solution (1 x ) were obtained from Commonwealth Serum Laboratories, Melbourne, Australia. Balanced salt solution (BSS) was prepared by the method of Hanks and Wallace [21]. All other chemicals were of analytical reagent quality. Protein purification and labelling Human apotransferrin was dialysed extensively before use and was shown to be free of other proteins by SDS-polyacrylamide gel electrophoresis. Apotransferrin was saturated with iron-59 using the ferric nitrilotriacetate (NTA) complex [22] in a molar ratio of 1 F e : 1 0 NTA. This procedure gave a specific activity of approx. 0.3 p,Ci iron-59 per nmole of Tf. Radioiodine labelling of transferrin was then performed by the iodine monochloride method [23] to give a specific activity of approx. 0.8 ~tCi 1-125 per nmole of transferrin. Unbound iodine-125 was removed by gel filtration through Sephadex G-25, followed by vacuum dialysis against at least four changes of 0.15 M NaCI adjusted to pH 7.4 with bicarbonate solution. Protein-free iodine was usually less than 1% of the total iodine present, as determined by precipitation in 7.5% trichlorotriacetic acid. BSA was dialysed extensively against 0.15 M NaCI before use. Cell culture The human melanoma cell line SK-MEL-28 (American Type Culture Collection, 12301 Parklawn Drive, Rockville, MD 20852) was used as these cells have the
highest concentration of melanotransferrin of all cells studied (300000-380000 sites/cell; Refs. 12, 24). Optimal growth was obtained in Eagle's MEM containing 30% FCS, 1% non-essential amino acids and by using a high seeding density. Cells were subcultured usirfg 1 mM EDTA in calcium/magnesium-free PBS. A large excess of 30% FCS was then added to the cell suspension to inactivate EDTA prior to it being centrifuged (2000 r p m / 5 min). The supernatant was then completely removed by aspiration and the cell pellet disaggregated by repeated pipetting in 30% FCS. The cell suspension was then added to culture flasks and incubated at 3 7 ° C in an atmosphere of 5% CO 2, 95% air in a Steri-Cult incubator with automatic CO 2 monitoring (Forma Scientific, Mallinckrodt Inc., Marietta, OH, U.S.A.). Cell viability was assessed by daily phase contrast microscopy and by the release of the cytoplasmic enzyme, lactate dehydrogenase (EC. 1.1.1.27), using a Cobas Bio automatic analyser (Roche Inc.). The possibility of dedifferentiation with increasing passage number [25] was addressed by monitoring the cells for the presence of organelles of the melanosome lineage [26] and for their tumorigenicity in nude mice [27]. For all cell passage numbers studied (43-86), organelles of the melanosome lineage and tumorigenicity in nude mice were always observed. Karyotype of the cells was also examined [28] as chromosomal abberations are characteristic of malignancy and melanoma [29], and were found to be abnormal. In addition, the Fe metabolism of two distant passages (42 and 82) was compared. There was no significant difference in all Fe and transferrin uptake parameters between these cells.
Iron metabolism - experimental procedure Uptake of iron and transferrin. To measure the uptake of transferrin and transferrin-bound Fe by cultured melanoma cells, the medium was replaced at 24-48 h of culture with prewarmed/pregassed (95% air; 5% CO2) MEM containing BSA (5 mg/ml), 20 mM Hepes (pH 7.4) and doubly labelled transferrin. Unless stated otherwise, experiments were performed with diferric human transferrin at a concentration of 1.25 ~M (0.1 mg/ml). Fetal calf serum was not added to the medium during incubation with radiolabeUed transferrin. All experiments were performed on totally confluent plates. At the end of the incubation with radioactive transferrin, the plates were placed on ice, the medium was decanted and the cell monolayer was washed four times with ice-cold BSS. The amount of radioactivity internalised by the cells was then measured by incubation with pronase (1 m g / m l ) for 30 min at 4°C, to remove membrane-bound iron and transferrin [19,20]. Increasing the incubation time with pronase to at least 2 h did not increase the release of Fe or transferrin, suggesting negligible effect of the proteinase on the integrity of the
cell membrane. The cells were removed from the plates in the pronase solution using a Teflon spatula, transferred to ice-cold microfuge tubes and centrifuged at 1 4 5 5 0 × g for 1 min in a Hereus Christ Biofuge A microcentrifuge (Hereus Christ Inc.) to separate internalised radioactivity (in the cell pellet) from formerly membrane-bound radioactivity in the supernatant (this fraction will henceforth be referred to as the membrane compartment). Radioactivity was measured in the cell pellet and membrane compartment in a three channel, well-type, -/-scintillation counter (Packard Tri-Carb model 5360, Packard Instrument Co., Inc., Downers Grove, IL) with appropriate correction for overlap of iron-59 activity into the iodine-125 channel. The cell pellet was resuspended in 1 ml of BSS and sonicated for 15 s. An aliquot of 0.1 ml was used to determine protein by the BCA method (Pierce Chemical Co., Rockford, IL, 61105 USA) as a measure of cell number. SeraChem clinical chemistry control sera (human), assayed, level 1 (Fisher Diagnostics, Orangeburg, N.Y. 10962) were used for protein assay standards. Samples were assayed by a Cobas Bio automatic analyser (Roche Inc.). Cell counts showed that one milligram of protein was equivalent to 11.80 + 0.02 • 106 (n = 132) cells.
Effect of transferrin concentration on iron and transferrin uptake. Plates were incubated for 2 h at 4 or 3 7 ° C with MEM containing 20 mM Hepes (pH 7.4), BSA (5 m g / m l ) and doubly labelled transferrin in the concentration range of 0.001 m g / m l to 0.06 m g / m l with or without a 50-fold excess of non-labelled diferric transferrin. A 2 h incubation period was used to ensure that a steady state of transferrin uptake and release had been reached. The presence of endogenous transferrin occupying transferrin receptors on cultured cells has been suggested to be a cause of error in the estimation of transferrin receptor numbers [30]. In the present study, human melanoma cells were grown in the presence of 30% FCS which contained bovine transferrin, which can bind to the human transferrin receptor [31]. Bovine transferrin was depleted from the cells before the start of the experiment by 3 separate 30 min incubations in MEM at 37°C. Failure to follow this washing regime resulted in a significant decrease in both Fe and transferrin uptake. This washing regime was also followed in an effort to remove any iron-56 from p97 sites which may have taken up in the 30% FCS medium. Following this protocol, Fe uptake into the membrane compartment was far more rapid than in those cell plates which had not been washed. Two methods of calculating nonspecific binding and hence specific binding, were employed. The first method involved calculation of the nonspecific binding from the linear component of the total binding plot [32]. The second method involved calculation of the nonspecific
component by using a 50-fold excess of unlabelled diferric transferrin in the presence of labelled transferfin. Due to the uncertainty in the literature as to which method is valid [32], both were used and the results were compared by statistical analysis. In the present study there was no significant difference in the data obtained using either method and the results were combined. Uptake of Fe and transferrin is expressed as moles of Fe or transferrin per gram of protein (gPR). The data are expressed as mean + S.E. (number of experiments) with 2-8 replicates in each experiment. Experimental data were compared using the Student t-test. Results were considered statistically significant when P < 0.05. Results
(1) Kinetics of iron and transferrin uptake (A) Transferrin uptake. The uptake of internalised, membrane-bound and hence total transferrin was biphasic with incubation time (Fig. la). Membrane-bound uptake initially increased with time and then plateaued after 5 min, there being no significant difference between transferrin uptake at 5 min and 2 h. Internalisation of transferrin increased more slowly, plateauing only after 30 min of incubation. Again no significant difference was found between 30 min and 2 h (Fig. la).
. . . . . .
. . . . .
. . . .
,
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
c C
~O
I%
0
30
0
0
'~5
~
0
~5
~)
165
120
• TotoI 0 Internolised
-b
c
• Membrane A
I0C
0
20
40 Time
60 (rn,n)
80
100
120
Fig. i. Uptake of nSl-59Fe-transferrin with time in melanoma cells. cells were incubated for up to 2 h at 37°C with 1251-59Fe-transferrin (0.1 mg/ml) and treated with pronase (1 mg/ml) for 30 rain at 4 °C. (a) Total transferrin uptake (O); membrane transferrin uptake (A); internalised transferrin uptake (o). (b) Total iron uptake (O); membrane iron uptake (A); internalised iron uptake (o). Results are mean + S.E. (8 separate experiments). The
4 TABLE I
Iron and transferrin uptake parameters of human melanoma cells (SKMEL-28) Results are mean ± S.E. (number of experiments in parentheses). Parameter Rate of Fe uptake (atoms Fe/cell per rain) Rate of internalised Tf uptake (molecules Tf/cell per min) Molar ratio of total Fe:Tfafter 1 h Molar ratio of internalised Fe : Tf after I h Rate of increase in the molar ratio of internalised Fe : Tf (per min) Rate of increase in the molar ratio of total Fe:Tf (per rain)
Melanoma cells 80000
+ 7000 (9) a
44000
+ 3000 (5) h
9.5 +
0.5 (8)
29.0 +
1.5 (8)
(C) Uptake of iron into the membrane-bound compartment. The u p t a k e of transferrin a n d Fe into the memb r a n e - b o u n d c o m p a r t m e n t was biphasic with time (Fig. la, 2a). However, while transferrin b i n d i n g reached a plateau, Fe u p t a k e c o n t i n u e d to decrease at a slow rate (Fig. 2a). T h e first rapid c o m p o n e n t could be almost entirely a c c o u n t e d for by the uptake of transferrinb o u n d Fe, which plateaued off within 5 rnin (as does the u p t a k e of transferrin; Fig. la). The second process 30
• Observed O Calculated Obs-Calc
20 E~ 0% L-
0.46 +
0.04 (8) a
x.
0.14+
0.02(8) a
~c
0
a Calculated from measurements between 15 min and 2 h. b Calculated from 1 min uptake.
30 T,me
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
~o
60 (m~n)
.
.
40
b c 0 ~
After 2 h, 24-1-1% [8] of the total transferrin was i n t e r n a l i s e d at this t r a n s f e r r i n c o n c e n t r a t i o n (0.1 m g / m l ) . I n contrast, when nonspecific b i n d i n g was accounted for (see section D), 81% of specific b i n d i n g sites were internalised. S a t u r a t i o n occurred at a transferrin c o n c e n t r a t i o n of approx. 0.0125 m g / m l (0.16 ~M). The total uptake of transferrin at 4 ° C was approx. 80% of that seen at 3 7 ° C (once the plateau phase of transferrin uptake had been reached at 30 rain). In contrast, i n t e r n a l i s a t i o n of transferrin was m a r k e d l y t e m p e r a t u r e - d e p e n d e n t with o n l y 1 - 3 % of the total transferrin internalised at 4 ° C. (B) Iron uptake. In contrast to transferrin uptake, total and internalised Fe uptake increased linearly with i n c u b a t i o n time (Fig. lb), for at least 24 h. I n t e r n a l i s a tion of Fe shows a lag period up to 15 min, until receptor o c c u p a n c y reached a steady state; thereafter Fe u p t a k e was linear (Fig. lb). The rate of Fe u p t a k e was calculated for both the total Fe a n d internalised Fe c o m p o n e n t s over 2 h. However, there was n o significant difference between them a n d the results were c o m b i n e d . T a b l e I lists the p a r a m e t e r s of Fe a n d transferrin u p t a k e o b t a i n e d for m e l a n o m a cells. I n t e r n a l i s a t i o n of Fe was negligible at 4 ° C a n d all of the Fe f o u n d in the m e m b r a n e c o m p a r t m e n t could be accounted for by that b o u n d to transferrin ( a s s u m i n g that each transferrin molecule b i n d s 2 Fe atoms). I n addition, the m o l a r ratio of Fe to transferrin r e m a i n e d at 2.0-2.2 for up to at least 5 h of i n c u b a t i o n at 4 ° C (i.e., the same as the i n c u b a t i o n medium).
16[" [
., o,
• Woshed D Not washed
i
/
/
/ ,
-°
0
28
1
/
.o..-'
.
2
3 Time
(h)
Time
(mln)
.
.
4,
.
5
6
c
-26 io
So_ ES I~_~ 2 2 o
E
20 "F 0
Fig. 2. Uptake of iron into the membrane compartment of melanoma cells. The cells were incubated for up to 6 h at 37°C with t251-~gFetransferrin and treated with pronase (1 mg/ml) for 30 rain at 4 ° C. (a) The observed (total) membrane iron uptake (O); calculated membrane iron uptake (estimated by assuming that each transferrin molecule in the membrane compartment has two iron atoms bound) (o); observed - calculated membrane iron uptake (zx). Results are Mean±SEM (8 separate experiments). (b) Membrane iron uptake after correction for iron bound to transferrin (see text) using cells which had been washed (i), or not washed (O), prior to the addition of labelled transferrin (results are means of 2 experiments). (c) Increase in the membrane molar ratio of iron to transferrin with time. Results are mean ± S.E. (8 separate experiments).
consisted of a linear increase in Fe uptake (at least initially; Fig. 2b), at the rate of 2900 + 400 [8] atoms of Fe per cell per min. The observed amount of Fe in the membrane-bound compartment at 2 h is significantly greater ( P < 0.005) than that calculated, assuming that each transferrin molecule in the membrane-bound compartment has two Fe atoms bound (Fig. 2a). Saturation of the membrane compartment occurred after approx. 4 h incubation with 0.1 m g / m l transferrin (Fig. 2b). The amount of Fe present in excess of transferrin-bound Fe was calculated to be equivalent to 345 000 p97 sites/cell, assuming that the Fe-binding component was p97 and has two binding sites per melanotransferrin molecule [161. Net Fe uptake into the membrane-bound compartment was also reflected in an increase in the molar ratio of Fe to transferrin (Fig. 2c). Up to 30 rain of incubation there was no significant change in this ratio, which corresponds to the binding of diferric transferrin to the membrane. After 30 min there was a significant linear increase in the molar ratio of Fe to transferrin in the membrane-bound compartment from 2.28 +_ 0.05 [8] at 45 min ( P < 0.025) to 2.66 + 0.05 [8] at 2 h ( P < 0.0005) (Fig. 2c) and 4.70 + 0.20 [3] at 24 h. There was no increase in the molar ratio of Fe to transferrin at 4 ° C, remaining at 2.0-2.2 for all time intervals examined up to 5 h reflecting the temperature dependence of uptake into this compartment. Four control experiments demonstrated that the excess membrane-bound Fe was not artefactual: (i) the effect of pronase on the integrity of the cell membrane was investigated, as leakage of Fe from damaged cells could possibly account for the results observed. Cells were radiolabelled for either 15 rain or 2 h and then incubated with pronase at 4 ° C for up to 2 h. There was no increase in the release of Fe or transferrin from the cells between 30 min and 2 h of incubation with pronase, suggesting the integrity of the cell membrane was maintained. (ii) The excess membrane non-transferrinbound Fe could not be due to passive adsorption of effluxed Fe on to the membrane, as once Fe was bound by the Fe-binding component it could not be removed by the chelators desferrioxamine (DFO; 1 mM) or pyridoxal isonicotinoyl hydrazone (PIH; 0.8 mM). (iii) Increasing the number of times the dishes were washed (immediately prior to pronase treatment) from 4 to 10 times did not reduce the amount of non-transferrinbound Fe present on the membrane. In fact, the molar ratio of Fe to transferrin increased, from 2.61 + 0.05 [2] (4 washes) to 6.88 + 0.28 [2] (10 washes), due to the removal of transferrin. (iv) Experiments with DFO (0.5 raM) demonstrated that uptake of non-transferrinbound Fe could be completely inhibited, while internalised Fe uptake was only reduced to 75% of the control. Conversely, bathophenanthroline sulphonate (0.5 mM) could reduce internalised Fe uptake to 50% of
the control, while membrane Fe uptake was not affected. These experiments suggested that non-transferfin membrane-bound Fe was independent of the amount of Fe internalised. These data suggest that Fe uptake from transferrin occurred by a process consistent with temperature-dependent, receptor-mediated endocytosis of transferrin. In addition, a membrane-bound, pronase-sensitive, temperature-dependent, saturable Fe-binding component was identified.
(II) The effect of transferrin concentration on transferrin and iron uptake (A) Uptake of transferrin. The involvement of specific transferrin receptors in Fe and transferrin uptake was determined by investigating the effect of transferrin concentration. Both total and internalised uptake of transferrin at 37 ° C were biphasic with increasing transferrin concentration (Fig. 3 a,b,c). The first component of these curves was interpreted to be due to the binding of transferrin to specific transferrin receptors and the second component to nonspecific adsorption of transferrin to the cell membrane. In contrast to both the total and internalised transferrin uptake curves, uptake of membrane-bound transferrin was almost linear (Fig. 3a). The nonspecific and specific components of total transferrin uptake (Fig. 3b) and internalised transferrin uptake (Fig. 3c) were derived as described in Materials and Methods. The specific component of the total and internalised transferrin uptake curves increased rapidly to saturate at a transferrin concentration of about 0.0125 m g / m l (0.16 #M). Scatchard analysis [331 gave a total of 116000 binding sites, of which 81 + 2% [7] were internalised (Table II). The apparent association constants ( K a) for the interaction of both internalised and total transferrin with the receptor were not significantly different, and the average value is given in Table II. When transferrin uptake was measured at 4 ° C (surface binding sites only; Ref. 19), Scatchard analysis showed 31 000 binding sites/cell. This is similar to the value for membrane binding estimated at 3 7 ° C using pronase, (22000 binding sites per cell). The K a value obtained at 4 ° C , was not appreciably different from that obtained at 3 7 ° C (Table II). The total transferrin uptake at 4 ° C [5] as a proportion of total transferrin uptake at 3 7 ° C [6] varied from 43% ([transferrin] = 0.001 m g / m l ) to 87% ([transferrin] = 0.06 m g / m l ) due to the large amount of nonspecific binding to the external cell membrane at higher transferrin concentrations. (B) Uptake of iron. As observed for the uptake of transferrin (Fig. 3), the internalised and total uptake of Fe at 3 7 ° C were biphasic with increasing transferrin concentration (Fig. 4a). However, the internalisation of Fe is much greater than that of transferrin after saturation of the binding site has been achieved at 0.0125
c~
c,~-
o
.
.
--°'""
~c
Or
~
~
~-2
#
_I
...4,~ "~
~.-4
.
*..-."
It should be noted from the linear part of the total transferrin and Fe uptake curves (Figs. 3b and 4b, respectively; i.e., transferrin concentration range of 0.0125 m g / m l to 0.055 mg/mi), that the amount of Fe taken up by the cells increased by 1.55 + 0.19.10 -6 mol F e / g P R per mg per ml transferrin [6], while transferrin uptake increased by 9.81 _+0.69.10 -~ mole transferrin/gPR per mg per ml transferrin [6]. Hence there was a 16-fold net accumulation of Fe over transferrin, 8 times that which could be accounted for by diferric transferrin. Similarly, for total internalised transferrin and Fe uptake curves (Figs. 3c and 4c, respectively) when the transferrin concentration was raised from 0.015 to 0.055 mg/ml, the amount of Fe taken up by the cells increased by 1 . 2 2 + 0 . 1 4 . 1 0 -6 mol F e / g P R per mg per ml transferrin [6], while trans15C
, ..... .....
o
~jo
12C
0
001
002 003 004 Tf concentration (mg/ml)
005
\
006
~/O/Y
9C
g~, 6c ~-
2
5F c
/@'~'~'~'~
3C
-0-0-0
-it- --- -~
001
Bc
5~7
/
|.~
~
_..--
c
-
-
.--/~. . . . •
150
--•'-'"
_ . & . • .... I ....
"
g~-2--
oo,~
~k . . . . .
• .....
• .....
003
~
.........
004
" &. . . . .
•
005
o
12C
2-
oo3
....
002
0;5--
5~6
Tt concentration ( m g / m l )
9o1
Fig. 3. Effect of transferrin concentration on transferrin uptake by melanoma cells. The cells were incubated with ~25I-5~Fe-transferrin (0.001-0.06 m g / m l ) for 2 h at 3 7 ° C and then treated with pronase (1 m g / m l ) for 30 rain at 4 ° C . (a) Total transferrin uptake (~); membrane transferrin uptake (~); internalised transferrin uptake (
]
BBAMCR 12693
The uptake of iron and transferrin by the human malignant melanoma cell D.R. Richardson and E. Baker Department of Physiology, Unioersity of Western Australia, Nedlands (Australia) (Received 26 September 1989) (Revised manuscript received 19 January 1990)
Key words: Melanotransferrin; Iron uptake; Transferrin; Iron; (Human melanoma cell)
The role of the transferrin homologue, melanotransferrin (p97), in iron metabolism has been studied using the human melanoma cell line, SK-MEL-28, which expresses this antigen in high concentrations. The mechanisms of iron and transferrin uptake were investigated using human transferrin labelled with iodine-125 and iron-59. Internalised and membrane-bound iron and transferrin were separated using the proteinase, pronase. The uptake of iron from transferrin occurred by at least two processes. The first process was saturable and consistent with receptor-mediated endocytosis, involving internalisation of transferrin bound to specific binding sites. Uptake of iron also occurred by a second process which was non-saturable up to 0.06 m g / m l (0.75 /tM) and was of higher efficiency than the saturable process. This process of iron uptake may be the dominant one at physiological serum transferrin concentrations. A membrane-bound, pronase-sensitive, temperature-dependent, iron-binding component was also identified. The number of binding sites was estimated to he approx. 340000 per cell (assuming 2 atoms of iron per site) and it is suggested that this binding component may be melanotransferrin.
Introduction All cells need iron (Fe) and neoplastic cells have a high requirement related to their rapid rate of proliferation [1]. This increased requirement is reflected by an increase in the expression of the transferrin receptor [2-4]. The metabolic pathways for transferrin and Fe uptake and release by neoplastic ceils may be different from normal cells [5-8]. The human malignant melanoma cell is of particular interest in this regard as these cells express high concentrations of a transferrin homologue, p97, or melanotransferrin, on their membrane as well as the transferrin receptor [9-11]. Melanotransferrin has been shown to have several characteristics in common with serum transferrin. These are: (i) it is a sialoglycoprotein (13% CHO); (ii) it has 37-39% sequence homology with human serum trans-
Abbreviations: FCS, fetal calf serum; MEM, minimal Eagle's medium; DFO, desferrioxamine; PIH, pyridoxal isonicotinoyl hydrazone; BSS, balanced salt solution; NTA, nitrilotriacetate; BSA, bovine serum albumin; gPR, gram protein; Tf, transferrin. Correspondence: E. Baker, Department of Physiology, University of Western Australia, Nedlands, Western Australia 6009, Australia.
ferrin, human lactoferrin and chicken transferrin; (iii) the melanotransferrin gene is on chromosome 3, as are those for transferrin and the transferrin receptor; (iv) it has a high interdomain sequence homology (46%), which is greater than that found for other transferrins (33417o); and (v) it can bind Fe from Fe(III) citrate complexes [12-16]. These similarities to transferrin have prompted Brown and colleagues to suggest that melanotransferrin may have a role in the Fe metabolism of the melanoma cell, although its functional significance in Fe transport has yet to be demonstrated. An increased supply of Fe via melanoma transferrin may increase the proliferation of melanoma cells, as Fe may be a rate-limiting nutrient for growth [1,17,18]. In the present work the Fe metabolism of the melanoma cell has been studied to determine the existence and relative importance of the two putative pathways of Fe uptake in melanoma cells: (i) uptake of Fe from transferrin via receptor-mediated endocytosis, and (ii) Fe uptake mediated by melanotransferrin. The mechanisms of Fe and transferrin uptake were determined by standard procedures using the nonspecific proteinase, pronase, to separate internalised (pronase-resistant) and membrane-bound (pronase-sensitive) Fe and transferrin [19,20]. The results indicate that these cells can obtain Fe from transferrin by a
0167-4889/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
saturable process consistent with receptor-mediated endocytosis. However, the amount of Fe internalised by the cell cannot be entirely accounted for by this process, and it has been concluded that the cell can remove Fe from transferrin by another mechanism. In addition, a membrane-bound, pronase-sensitive, temperature-dependent, Fe-binding component has been identified, the properties of which are consistent with those of melanotransferrin. Materials and Methods
Reagents Iron-59 (as ferric chloride in 0.1 M HCI) and iodine125 (as sodium iodide) were purchased from Amersham International, Amersham, U.K. Pronase was purchased from Boehringer Mannheim, Mt. Waverley, Australia. Eagle's modified minimum essential medium (MEM) as Autopow and fetal calf serum (FCS) were supplied by Flow Laboratories, Annandale, Australia. Penicillin (Crystapen-benzylpenicillin sodium B.P.) was obtained from Glaxo Aus. Pty Ltd., Boronia, Australia Bovine serum albumin (BSA; 98% pure, fatty acid free), human apotransferrin, L-glutamine and Hepes were obtained from Sigma Chemical Co., St. Louis, MO, U.S.A. Nonessential amino acids (100 x concentrate) and trypsinversene solution (1 x ) were obtained from Commonwealth Serum Laboratories, Melbourne, Australia. Balanced salt solution (BSS) was prepared by the method of Hanks and Wallace [21]. All other chemicals were of analytical reagent quality. Protein purification and labelling Human apotransferrin was dialysed extensively before use and was shown to be free of other proteins by SDS-polyacrylamide gel electrophoresis. Apotransferrin was saturated with iron-59 using the ferric nitrilotriacetate (NTA) complex [22] in a molar ratio of 1 F e : 1 0 NTA. This procedure gave a specific activity of approx. 0.3 p,Ci iron-59 per nmole of Tf. Radioiodine labelling of transferrin was then performed by the iodine monochloride method [23] to give a specific activity of approx. 0.8 ~tCi 1-125 per nmole of transferrin. Unbound iodine-125 was removed by gel filtration through Sephadex G-25, followed by vacuum dialysis against at least four changes of 0.15 M NaCI adjusted to pH 7.4 with bicarbonate solution. Protein-free iodine was usually less than 1% of the total iodine present, as determined by precipitation in 7.5% trichlorotriacetic acid. BSA was dialysed extensively against 0.15 M NaCI before use. Cell culture The human melanoma cell line SK-MEL-28 (American Type Culture Collection, 12301 Parklawn Drive, Rockville, MD 20852) was used as these cells have the
highest concentration of melanotransferrin of all cells studied (300000-380000 sites/cell; Refs. 12, 24). Optimal growth was obtained in Eagle's MEM containing 30% FCS, 1% non-essential amino acids and by using a high seeding density. Cells were subcultured usirfg 1 mM EDTA in calcium/magnesium-free PBS. A large excess of 30% FCS was then added to the cell suspension to inactivate EDTA prior to it being centrifuged (2000 r p m / 5 min). The supernatant was then completely removed by aspiration and the cell pellet disaggregated by repeated pipetting in 30% FCS. The cell suspension was then added to culture flasks and incubated at 3 7 ° C in an atmosphere of 5% CO 2, 95% air in a Steri-Cult incubator with automatic CO 2 monitoring (Forma Scientific, Mallinckrodt Inc., Marietta, OH, U.S.A.). Cell viability was assessed by daily phase contrast microscopy and by the release of the cytoplasmic enzyme, lactate dehydrogenase (EC. 1.1.1.27), using a Cobas Bio automatic analyser (Roche Inc.). The possibility of dedifferentiation with increasing passage number [25] was addressed by monitoring the cells for the presence of organelles of the melanosome lineage [26] and for their tumorigenicity in nude mice [27]. For all cell passage numbers studied (43-86), organelles of the melanosome lineage and tumorigenicity in nude mice were always observed. Karyotype of the cells was also examined [28] as chromosomal abberations are characteristic of malignancy and melanoma [29], and were found to be abnormal. In addition, the Fe metabolism of two distant passages (42 and 82) was compared. There was no significant difference in all Fe and transferrin uptake parameters between these cells.
Iron metabolism - experimental procedure Uptake of iron and transferrin. To measure the uptake of transferrin and transferrin-bound Fe by cultured melanoma cells, the medium was replaced at 24-48 h of culture with prewarmed/pregassed (95% air; 5% CO2) MEM containing BSA (5 mg/ml), 20 mM Hepes (pH 7.4) and doubly labelled transferrin. Unless stated otherwise, experiments were performed with diferric human transferrin at a concentration of 1.25 ~M (0.1 mg/ml). Fetal calf serum was not added to the medium during incubation with radiolabeUed transferrin. All experiments were performed on totally confluent plates. At the end of the incubation with radioactive transferrin, the plates were placed on ice, the medium was decanted and the cell monolayer was washed four times with ice-cold BSS. The amount of radioactivity internalised by the cells was then measured by incubation with pronase (1 m g / m l ) for 30 min at 4°C, to remove membrane-bound iron and transferrin [19,20]. Increasing the incubation time with pronase to at least 2 h did not increase the release of Fe or transferrin, suggesting negligible effect of the proteinase on the integrity of the
cell membrane. The cells were removed from the plates in the pronase solution using a Teflon spatula, transferred to ice-cold microfuge tubes and centrifuged at 1 4 5 5 0 × g for 1 min in a Hereus Christ Biofuge A microcentrifuge (Hereus Christ Inc.) to separate internalised radioactivity (in the cell pellet) from formerly membrane-bound radioactivity in the supernatant (this fraction will henceforth be referred to as the membrane compartment). Radioactivity was measured in the cell pellet and membrane compartment in a three channel, well-type, -/-scintillation counter (Packard Tri-Carb model 5360, Packard Instrument Co., Inc., Downers Grove, IL) with appropriate correction for overlap of iron-59 activity into the iodine-125 channel. The cell pellet was resuspended in 1 ml of BSS and sonicated for 15 s. An aliquot of 0.1 ml was used to determine protein by the BCA method (Pierce Chemical Co., Rockford, IL, 61105 USA) as a measure of cell number. SeraChem clinical chemistry control sera (human), assayed, level 1 (Fisher Diagnostics, Orangeburg, N.Y. 10962) were used for protein assay standards. Samples were assayed by a Cobas Bio automatic analyser (Roche Inc.). Cell counts showed that one milligram of protein was equivalent to 11.80 + 0.02 • 106 (n = 132) cells.
Effect of transferrin concentration on iron and transferrin uptake. Plates were incubated for 2 h at 4 or 3 7 ° C with MEM containing 20 mM Hepes (pH 7.4), BSA (5 m g / m l ) and doubly labelled transferrin in the concentration range of 0.001 m g / m l to 0.06 m g / m l with or without a 50-fold excess of non-labelled diferric transferrin. A 2 h incubation period was used to ensure that a steady state of transferrin uptake and release had been reached. The presence of endogenous transferrin occupying transferrin receptors on cultured cells has been suggested to be a cause of error in the estimation of transferrin receptor numbers [30]. In the present study, human melanoma cells were grown in the presence of 30% FCS which contained bovine transferrin, which can bind to the human transferrin receptor [31]. Bovine transferrin was depleted from the cells before the start of the experiment by 3 separate 30 min incubations in MEM at 37°C. Failure to follow this washing regime resulted in a significant decrease in both Fe and transferrin uptake. This washing regime was also followed in an effort to remove any iron-56 from p97 sites which may have taken up in the 30% FCS medium. Following this protocol, Fe uptake into the membrane compartment was far more rapid than in those cell plates which had not been washed. Two methods of calculating nonspecific binding and hence specific binding, were employed. The first method involved calculation of the nonspecific binding from the linear component of the total binding plot [32]. The second method involved calculation of the nonspecific
component by using a 50-fold excess of unlabelled diferric transferrin in the presence of labelled transferfin. Due to the uncertainty in the literature as to which method is valid [32], both were used and the results were compared by statistical analysis. In the present study there was no significant difference in the data obtained using either method and the results were combined. Uptake of Fe and transferrin is expressed as moles of Fe or transferrin per gram of protein (gPR). The data are expressed as mean + S.E. (number of experiments) with 2-8 replicates in each experiment. Experimental data were compared using the Student t-test. Results were considered statistically significant when P < 0.05. Results
(1) Kinetics of iron and transferrin uptake (A) Transferrin uptake. The uptake of internalised, membrane-bound and hence total transferrin was biphasic with incubation time (Fig. la). Membrane-bound uptake initially increased with time and then plateaued after 5 min, there being no significant difference between transferrin uptake at 5 min and 2 h. Internalisation of transferrin increased more slowly, plateauing only after 30 min of incubation. Again no significant difference was found between 30 min and 2 h (Fig. la).
. . . . . .
. . . . .
. . . .
,
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
c C
~O
I%
0
30
0
0
'~5
~
0
~5
~)
165
120
• TotoI 0 Internolised
-b
c
• Membrane A
I0C
0
20
40 Time
60 (rn,n)
80
100
120
Fig. i. Uptake of nSl-59Fe-transferrin with time in melanoma cells. cells were incubated for up to 2 h at 37°C with 1251-59Fe-transferrin (0.1 mg/ml) and treated with pronase (1 mg/ml) for 30 rain at 4 °C. (a) Total transferrin uptake (O); membrane transferrin uptake (A); internalised transferrin uptake (o). (b) Total iron uptake (O); membrane iron uptake (A); internalised iron uptake (o). Results are mean + S.E. (8 separate experiments). The
4 TABLE I
Iron and transferrin uptake parameters of human melanoma cells (SKMEL-28) Results are mean ± S.E. (number of experiments in parentheses). Parameter Rate of Fe uptake (atoms Fe/cell per rain) Rate of internalised Tf uptake (molecules Tf/cell per min) Molar ratio of total Fe:Tfafter 1 h Molar ratio of internalised Fe : Tf after I h Rate of increase in the molar ratio of internalised Fe : Tf (per min) Rate of increase in the molar ratio of total Fe:Tf (per rain)
Melanoma cells 80000
+ 7000 (9) a
44000
+ 3000 (5) h
9.5 +
0.5 (8)
29.0 +
1.5 (8)
(C) Uptake of iron into the membrane-bound compartment. The u p t a k e of transferrin a n d Fe into the memb r a n e - b o u n d c o m p a r t m e n t was biphasic with time (Fig. la, 2a). However, while transferrin b i n d i n g reached a plateau, Fe u p t a k e c o n t i n u e d to decrease at a slow rate (Fig. 2a). T h e first rapid c o m p o n e n t could be almost entirely a c c o u n t e d for by the uptake of transferrinb o u n d Fe, which plateaued off within 5 rnin (as does the u p t a k e of transferrin; Fig. la). The second process 30
• Observed O Calculated Obs-Calc
20 E~ 0% L-
0.46 +
0.04 (8) a
x.
0.14+
0.02(8) a
~c
0
a Calculated from measurements between 15 min and 2 h. b Calculated from 1 min uptake.
30 T,me
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
~o
60 (m~n)
.
.
40
b c 0 ~
After 2 h, 24-1-1% [8] of the total transferrin was i n t e r n a l i s e d at this t r a n s f e r r i n c o n c e n t r a t i o n (0.1 m g / m l ) . I n contrast, when nonspecific b i n d i n g was accounted for (see section D), 81% of specific b i n d i n g sites were internalised. S a t u r a t i o n occurred at a transferrin c o n c e n t r a t i o n of approx. 0.0125 m g / m l (0.16 ~M). The total uptake of transferrin at 4 ° C was approx. 80% of that seen at 3 7 ° C (once the plateau phase of transferrin uptake had been reached at 30 rain). In contrast, i n t e r n a l i s a t i o n of transferrin was m a r k e d l y t e m p e r a t u r e - d e p e n d e n t with o n l y 1 - 3 % of the total transferrin internalised at 4 ° C. (B) Iron uptake. In contrast to transferrin uptake, total and internalised Fe uptake increased linearly with i n c u b a t i o n time (Fig. lb), for at least 24 h. I n t e r n a l i s a tion of Fe shows a lag period up to 15 min, until receptor o c c u p a n c y reached a steady state; thereafter Fe u p t a k e was linear (Fig. lb). The rate of Fe u p t a k e was calculated for both the total Fe a n d internalised Fe c o m p o n e n t s over 2 h. However, there was n o significant difference between them a n d the results were c o m b i n e d . T a b l e I lists the p a r a m e t e r s of Fe a n d transferrin u p t a k e o b t a i n e d for m e l a n o m a cells. I n t e r n a l i s a t i o n of Fe was negligible at 4 ° C a n d all of the Fe f o u n d in the m e m b r a n e c o m p a r t m e n t could be accounted for by that b o u n d to transferrin ( a s s u m i n g that each transferrin molecule b i n d s 2 Fe atoms). I n addition, the m o l a r ratio of Fe to transferrin r e m a i n e d at 2.0-2.2 for up to at least 5 h of i n c u b a t i o n at 4 ° C (i.e., the same as the i n c u b a t i o n medium).
16[" [
., o,
• Woshed D Not washed
i
/
/
/ ,
-°
0
28
1
/
.o..-'
.
2
3 Time
(h)
Time
(mln)
.
.
4,
.
5
6
c
-26 io
So_ ES I~_~ 2 2 o
E
20 "F 0
Fig. 2. Uptake of iron into the membrane compartment of melanoma cells. The cells were incubated for up to 6 h at 37°C with t251-~gFetransferrin and treated with pronase (1 mg/ml) for 30 rain at 4 ° C. (a) The observed (total) membrane iron uptake (O); calculated membrane iron uptake (estimated by assuming that each transferrin molecule in the membrane compartment has two iron atoms bound) (o); observed - calculated membrane iron uptake (zx). Results are Mean±SEM (8 separate experiments). (b) Membrane iron uptake after correction for iron bound to transferrin (see text) using cells which had been washed (i), or not washed (O), prior to the addition of labelled transferrin (results are means of 2 experiments). (c) Increase in the membrane molar ratio of iron to transferrin with time. Results are mean ± S.E. (8 separate experiments).
consisted of a linear increase in Fe uptake (at least initially; Fig. 2b), at the rate of 2900 + 400 [8] atoms of Fe per cell per min. The observed amount of Fe in the membrane-bound compartment at 2 h is significantly greater ( P < 0.005) than that calculated, assuming that each transferrin molecule in the membrane-bound compartment has two Fe atoms bound (Fig. 2a). Saturation of the membrane compartment occurred after approx. 4 h incubation with 0.1 m g / m l transferrin (Fig. 2b). The amount of Fe present in excess of transferrin-bound Fe was calculated to be equivalent to 345 000 p97 sites/cell, assuming that the Fe-binding component was p97 and has two binding sites per melanotransferrin molecule [161. Net Fe uptake into the membrane-bound compartment was also reflected in an increase in the molar ratio of Fe to transferrin (Fig. 2c). Up to 30 rain of incubation there was no significant change in this ratio, which corresponds to the binding of diferric transferrin to the membrane. After 30 min there was a significant linear increase in the molar ratio of Fe to transferrin in the membrane-bound compartment from 2.28 +_ 0.05 [8] at 45 min ( P < 0.025) to 2.66 + 0.05 [8] at 2 h ( P < 0.0005) (Fig. 2c) and 4.70 + 0.20 [3] at 24 h. There was no increase in the molar ratio of Fe to transferrin at 4 ° C, remaining at 2.0-2.2 for all time intervals examined up to 5 h reflecting the temperature dependence of uptake into this compartment. Four control experiments demonstrated that the excess membrane-bound Fe was not artefactual: (i) the effect of pronase on the integrity of the cell membrane was investigated, as leakage of Fe from damaged cells could possibly account for the results observed. Cells were radiolabelled for either 15 rain or 2 h and then incubated with pronase at 4 ° C for up to 2 h. There was no increase in the release of Fe or transferrin from the cells between 30 min and 2 h of incubation with pronase, suggesting the integrity of the cell membrane was maintained. (ii) The excess membrane non-transferrinbound Fe could not be due to passive adsorption of effluxed Fe on to the membrane, as once Fe was bound by the Fe-binding component it could not be removed by the chelators desferrioxamine (DFO; 1 mM) or pyridoxal isonicotinoyl hydrazone (PIH; 0.8 mM). (iii) Increasing the number of times the dishes were washed (immediately prior to pronase treatment) from 4 to 10 times did not reduce the amount of non-transferrinbound Fe present on the membrane. In fact, the molar ratio of Fe to transferrin increased, from 2.61 + 0.05 [2] (4 washes) to 6.88 + 0.28 [2] (10 washes), due to the removal of transferrin. (iv) Experiments with DFO (0.5 raM) demonstrated that uptake of non-transferrinbound Fe could be completely inhibited, while internalised Fe uptake was only reduced to 75% of the control. Conversely, bathophenanthroline sulphonate (0.5 mM) could reduce internalised Fe uptake to 50% of
the control, while membrane Fe uptake was not affected. These experiments suggested that non-transferfin membrane-bound Fe was independent of the amount of Fe internalised. These data suggest that Fe uptake from transferrin occurred by a process consistent with temperature-dependent, receptor-mediated endocytosis of transferrin. In addition, a membrane-bound, pronase-sensitive, temperature-dependent, saturable Fe-binding component was identified.
(II) The effect of transferrin concentration on transferrin and iron uptake (A) Uptake of transferrin. The involvement of specific transferrin receptors in Fe and transferrin uptake was determined by investigating the effect of transferrin concentration. Both total and internalised uptake of transferrin at 37 ° C were biphasic with increasing transferrin concentration (Fig. 3 a,b,c). The first component of these curves was interpreted to be due to the binding of transferrin to specific transferrin receptors and the second component to nonspecific adsorption of transferrin to the cell membrane. In contrast to both the total and internalised transferrin uptake curves, uptake of membrane-bound transferrin was almost linear (Fig. 3a). The nonspecific and specific components of total transferrin uptake (Fig. 3b) and internalised transferrin uptake (Fig. 3c) were derived as described in Materials and Methods. The specific component of the total and internalised transferrin uptake curves increased rapidly to saturate at a transferrin concentration of about 0.0125 m g / m l (0.16 #M). Scatchard analysis [331 gave a total of 116000 binding sites, of which 81 + 2% [7] were internalised (Table II). The apparent association constants ( K a) for the interaction of both internalised and total transferrin with the receptor were not significantly different, and the average value is given in Table II. When transferrin uptake was measured at 4 ° C (surface binding sites only; Ref. 19), Scatchard analysis showed 31 000 binding sites/cell. This is similar to the value for membrane binding estimated at 3 7 ° C using pronase, (22000 binding sites per cell). The K a value obtained at 4 ° C , was not appreciably different from that obtained at 3 7 ° C (Table II). The total transferrin uptake at 4 ° C [5] as a proportion of total transferrin uptake at 3 7 ° C [6] varied from 43% ([transferrin] = 0.001 m g / m l ) to 87% ([transferrin] = 0.06 m g / m l ) due to the large amount of nonspecific binding to the external cell membrane at higher transferrin concentrations. (B) Uptake of iron. As observed for the uptake of transferrin (Fig. 3), the internalised and total uptake of Fe at 3 7 ° C were biphasic with increasing transferrin concentration (Fig. 4a). However, the internalisation of Fe is much greater than that of transferrin after saturation of the binding site has been achieved at 0.0125
c~
c,~-
o
.
.
--°'""
~c
Or
~
~
~-2
#
_I
...4,~ "~
~.-4
.
*..-."
It should be noted from the linear part of the total transferrin and Fe uptake curves (Figs. 3b and 4b, respectively; i.e., transferrin concentration range of 0.0125 m g / m l to 0.055 mg/mi), that the amount of Fe taken up by the cells increased by 1.55 + 0.19.10 -6 mol F e / g P R per mg per ml transferrin [6], while transferrin uptake increased by 9.81 _+0.69.10 -~ mole transferrin/gPR per mg per ml transferrin [6]. Hence there was a 16-fold net accumulation of Fe over transferrin, 8 times that which could be accounted for by diferric transferrin. Similarly, for total internalised transferrin and Fe uptake curves (Figs. 3c and 4c, respectively) when the transferrin concentration was raised from 0.015 to 0.055 mg/ml, the amount of Fe taken up by the cells increased by 1 . 2 2 + 0 . 1 4 . 1 0 -6 mol F e / g P R per mg per ml transferrin [6], while trans15C
, ..... .....
o
~jo
12C
0
001
002 003 004 Tf concentration (mg/ml)
005
\
006
~/O/Y
9C
g~, 6c ~-
2
5F c
/@'~'~'~'~
3C
-0-0-0
-it- --- -~
001
Bc
5~7
/
|.~
~
_..--
c
-
-
.--/~. . . . •
150
--•'-'"
_ . & . • .... I ....
"
g~-2--
oo,~
~k . . . . .
• .....
• .....
003
~
.........
004
" &. . . . .
•
005
o
12C
2-
oo3
....
002
0;5--
5~6
Tt concentration ( m g / m l )
9o1
Fig. 3. Effect of transferrin concentration on transferrin uptake by melanoma cells. The cells were incubated with ~25I-5~Fe-transferrin (0.001-0.06 m g / m l ) for 2 h at 3 7 ° C and then treated with pronase (1 m g / m l ) for 30 rain at 4 ° C . (a) Total transferrin uptake (~); membrane transferrin uptake (~); internalised transferrin uptake (
