- Email: [email protected]
Translocation of cadmium from cytosol to membrane fraction in cadmium-loaded red blood cells of rats
Toxic. in Vitro Vol. 2, No. 1, pp. 57-63, 1988 Printed in Great Britain.All rights reserved
0887-2333/88 $3.00+0.00 Copyright © 1988PergamonJournals Ltd
TRANSLOCATION OF C A D M I U M FROM CYTOSOL TO M E M B R A N E FRACTION IN C A D M I U M - L O A D E D RED BLOOD CELLS OF RATS M. KUNIMOTOand T. MIURA Basic Medical Sciences Division, The National Institute for Environmental Studies, Yatabe, Tsukuba, Ibaraki 305, Japan (Received 23 March 1987; revisions received 2 September 1987)
Abstract--When rat red blood cells were incubated in a cadmium (Cd)-free medium for up to 14 hr following a 1-hr treatment with 0.5 mM-CdC12, the incorporated Cd, which was predominantly in the cytosol at the beginning of the incubation, progressivelyaccumulated in the membrane fraction. In parallel with the Cd-accumulation, several cytosolic proteins including haemoglobin increased in the membrane fraction, resulting in an increase in the protein to phospholipid ratio of the membrane fraction. The membrane fraction was solubilized with sodium deoxycholate and analysed by gel-filtration chromatography. Cd was detected in the high-molecular-weightfraction containing membrane proteins at the beginning of incubation, and the Cd content of this fraction did not alter appreciably during the incubation. On the other hand, Cd in the low-molecular-weight fraction, where haemoglobin is most abundant, increased progressively during the incubation. The Cd-binding capacity of proteins in the membrane fraction was assessed by t°gCd-autoradiography using the western blotting technique. In addition to membrane proteins, haemoglobin and other cytosolic proteins, which increased in the membrane fraction during the incubation, exhibited a significant ability to bind Cd. These results indicate that Cd, once incorporated into red cell cytosol, progressively accumulates in the membrane fraction together with cytosolic proteins with Cd-binding capacity. This accumulation of Cd-cytosolicproteins may affect the membrane properties of red blood cells.
INTRODUCTION
of cellular Cd in the membrane fraction, although intracellular ATP was maintained at the control level (Kunimoto et al. 1986). At higher' Cd concentrations (l.0-2.0mM) a significafft decrease in intracellular ATP was observed, in addition to the age-related changes, during incubation in a Cd-free medium. It is, therefore, important for the assessment of Cd toxicity into red blood cells to clarify the fate of the Cd incorporated into red blood cells. In this report, we present evidence that, during incubation in a Cd-free medium, Cd in cytosol translocates to the membrane fraction together with cytosolic proteins with Cd-binding capacity.
Cadmium (Cd) is a well known industrial and environmental toxicant. Anaemia is a common finding in workers exposed to Cd and in mammals given Cd (Friberg et aL 1974; Samarawickrama, 1979). One of the mechanisms responsible for Cd-induced anaemia is decreased survival of red blood cells in Cd-treated animals (Berlin & Friberg, 1960; Kunimoto & Miura, 1986). After oral or parenteral administration Cd appears in the blood (Klaassen & Kotsonis, 1977; Samarawickrama, 1979). A significant part of the Cd in the blood is located in the red blood cells (Carlson & Friberg, 1957; Friberg, 1952; Garty et al. 1981; Nordberg et al. 1971) and the non-dialysable part of red-cell Cd is apparently predominantly bound to haemoglobin (Carlson & Friberg, 1957). Recently, we showed that the in vitro incubation of rat red blood cells with 0.5-2.0 mM-Cd at 37°C for 1 hr accelerated age-related changes in red blood cells such as shape changes, increased density, decreased flexibility and shortened in vivo survival (Kunimoto et al. 1985). However, the age-related changes in red blood cells were slight upon l-hr treatment with 0.5mM-Cd, compared with the effect on in vivo survival, and only 5% of the incorporated Cd was located in the membrane fraction. During subsequent incubation in a Cd-free medium, age-related changes developed in parallel with progressive accumulation
MATERIALS AND METHODS
Preparation and incubation o f red blood cells. Red blood cells were obtained from male Jcl:Wistar rats (body weight, 450-500g) as described before (Kunimoto et al. 1984). Washed red cells were mixed with 3 vols 50 mM-glycylglycine, pH 7.4, containing 5 mM-KCI, 116 mM-NaC1, 11.1 mM-glucose, 0.54 mMadenine, 12.7 mM-inosine, 0.2 mg streptomycin/ml and 200 IU penicillin G/ml (buffer A). After incubation at 37°C for 1 hr in the presence of 0.5 mMCdCI2, the incubation mixture was centrifuged at 2000 rev/min for 5 rain at room temperature and the pelleted cells were washed once with 10 vols buffer A, resuspended in 3 vols buffer A (Cd-free medium) and incubated at 37°C for up to 14 hr.
Abbreviation: SDS = sodium dodecyl sulphate.
57 T.I.V. 2/I--D
58
M. KUNIMOTOand T. MIURA
Preparation of red cell membranes. After incubation in the Cd-free medium for the times indicated elsewhere, red blood cells were pelleted by centrifugation at 2000 rev/min for 5 min, washed once with 10 vols saline and haemolysed by 1:20 dilution with 10 mM-Tris HCI, pH 7.4. The membrane fraction was pelleted at 15,000rev/min for 20rain at 4°C and washed four times with the same buffer. Sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis. Electrophoreses in 7.5 or 12.5% polyacrylamide containing 0.1% SDS were performed according to the method of Laemmli (1970). Gels stained with Coomassie brilliant blue R-250 were scanned in a densitometer (Auto Scanner Flur-Vis, Helena Laboratory Co., TX, USA) and the area of each peak was integrated with a digitizer (MOP-10, Kontron, Munich, FRG). Solubilization and gel-filtration chromatography of the membrane fractions. Membrane fractions were mixed with 3vols 5% sodium deoxyeholate containing 0.1 M-KCI, 10 mM-Tris HC1, pH 7.4 and 20 #g phenylmethylsulphonyl fluoride/ml (solution I), incubated at 30°C for 20min and centrifuged at 15,000 rev/min for 20 min at 30°C. Two millilitres of the supernatant were applied to a column (1.5 crn x 45 cm) of Ultrogel AcA 34 (LKB, Bromma, Sweden) equilibrated with 10mM-sodium deoxycholate containing 0.1M-KC1, 10mM-Tris HCI, pH 7.4 and 20 pg phenylmethylsulphonyl fluoride/ml (solution II) at room temperature and 1.7-ml fractions were collected (9 ml/hr). Each fraction was subjected to determinations of protein and Cd and SDS-polyacrylamide gel electrophoresis. Red blood cells incubated with 0.5 mM-CdC12 for 1 hr at 37°C were washed once with 10 vols saline and mixed with 9 vols solution I. After 20-min incubation at 30°C the mixture was centrifuged at 15,000 rev/min for 20 min at 30°C and 0.8 ml of the supernatant was also analysed as described above. Detection of Cd-binding proteins in the membrane fraction by 1°9Cd autoradiography on a Durapore membrane using western blotting technique. Proteins in the membrane fraction separated by SDSpolyacrylamide gel electrophoresis were electrophoreticaily transferred to a Durapore membrane (GVHP, Millipore Corp., MA, USA) in a Trans-blot cell (Bio-Rad, CA, USA) according to the method of Towbin et aL (1979). Transfer was performed at a constant voltage of 30 V for 30 min, followed by 150 V for 90 min at 4°C, using a solution containing 20% methanol, 25 mM-Tris and 129 mM-glycine, pH 8.5. After the transfer, the membrane was soaked in 10 mM-Tris HCI, pH 7.4 for 12 hr at 4°C and then incubated in 10mM-Tris HCI, pH7.4 containing 1 pCi 1°gCdCl2/ml (carrier free, New England Nuclear, MA, USA), 1 mM-MgCI 2 and 0.1 M-KC1 for 10min at room temperature. The membrane was washed three times with 10 mM-Tris HCI, pH 7.4 at 5-min intervals. The addition of 0.1 M-KCI is essential for reducing non-specific bindings of m9Cd to the membrane (Aoki et al. 1986). Autoradiography of the membrane was performed as described by Aoki et al. (1986). Other methods. The amount of Cd in red blood cells, membrane fractions and column fractions was determined after digestion with HNO3-60% HCIO 4
(5:l,v/v) using an atomic absorption spectrophotometer (Hitachi Model 170-50, Hitachi Co., Tokyo) as described by Suzuki & Yamamura (1979). Possible interference by sodium ions was corrected using a deuterium lamp. The protein content was estimated by the method of Lowry et al. (1951). The amount of haemoglobin was determined spectrophotometrically after conversion to cyanomethaemoglobin, according to the method of van Kampen & Zijlstra (1961).
RESULTS
Changes in the membrane composition of Cd-loaded red cells during incubation in the Cd-free medium When red blood cells were incubated with 0.5 mM-CdCI2 for 1 hr at 37°C, 35% of the exogenous Cd was incorporated into the cell. The Cd-loaded red cells were then transferred to the Cd-free medium and incubated at 37°C for 14 hr. The incorporated Cd was not released from the cell during the 14-hr incubation in the Cd-free medium (Kunimoto et aL 1986). However, the amount of Cd in the membrane fraction increased progressively (Fig. 1) and reached approximately 20% of the cellular Cd after incubation for 14 hr (data not shown). In parallel with the Cd accumulation, the protein to phospholipid ratio of the membrane fraction increased and reached 1.48 times that of control cells after the 14-hr incubation (Fig. 1). The protein composition of the membrane fraction was also determi'ned by SDS-polyacrylamide gel electrophoresis. As shown in Fig. 2, several proteins (CI-C7 and haemoglobin), which are cytosolic because they exist both in the haemolysate (lane 1) and its supernatant (data not shown), increased as the incubation time increased (lanes 3-6). Increases in haemoglobin and C7 were particularly noticeable.
3.0
~ 2- 7"~5 ~
.-~
. .
e
n
~.
2.0
1.0
//
I
0.5 F z j / / oil
0
I 2
/
"-
2
o
If'I
I 4
I 6
,],I 8
I 10
I t2
I 14
E
4
"a
L)
T i m e (hr)
Fig. 1. Changes in the protein to phospholipid ratio and Cd content of membrane fractions prepared from Cd-treated red cells during incubation in the Cd-free medium. Rat red blood cells were pre-incubated with (O) or without (O) 0.5-mM CdC12 for I hr and the protein to phospholipid ratio (Q O) and the Cd content (A) of their membrane fractions were determined after incubation in the Cd-free medium. Each point represents the mean of three experiments and error bars represent standard deviations.
1,2.1
4.1a,b 4.2
•
Ct IIm,-.Cz
-4--4.5
C3 ID"
-.~--5 t,.-
C5 ~
~
C6
C7 m,,,--
m
0
/
m
Hb
a
"1
2
:3
4
5
-4
6
Fig. 2. SDS-polyacrylamide gel electrophoresis of membrane fractions prepared from Cd-treated red cells. Membrane fractions were prepared from Cd-treated red cells 0 (lane 3), 4 (lane 4), 8 (lane 5) and 14 hr (lane 6) after incubation in the Cd-free medium and from untreated red cells (lane 2) and were analysed by SDS-polyacrylamide gel electophoresis. Whole red blood cells were also analysed (lane I). Hb = haemoglobin and Ct-C 7 = cytosolic proteins which increasingly associated with the membrane fraction. (For Figs 3 & 4, see p. 61)
59
0)
Hb 4.2
>,
p
E 0
D
(b)
8 o
Fig. 5. Detection of Cd-binding proteins in the membrane fraction by ~°9Cd-autoradiography on a Durapore membrane. Membrane fractions were prepared from Cd-treated red cells 8 hr after incubation in the Cd-free medium. Proteins of the membrane fraction were separated by SDS-polyacrylamide gel electrophoresis (Fig. 2, lane 5) and were electrophoretically transferred to a Durapore membrane; (a) proteins on the membrane were stained with amido black; (b) the membrane was incubated with ~°9Cd and its autoradiography was obtained. Both were analysed with a densitometer.
60
!leo
•E
0.8
f 0.6
8
0
~oo,(b)
f"~.z
=50 t.
0 : -±
~_
O.,
.j
.
.
.
.
.
=
L
-~
0
2
~oo1 (c)
A
0.'I B
511... . -
~oo
.
.
_%
--
L_L".
--
, 0
o.z
(d)
0
10
20
61
Accumulation of Cd in red ceil membranes
30 Froction
40
50
60
70
80
number
Fig. 3. Gel filtration of solubilized membrane fractions prepared from Cd-treated red cells. Cd-treated red cells (a) and membrane fractions prepared from Cd-treated red cells 0 (b), 7 (c) and 14hr (d) after incubation in the Cd-free medium were solubilized by 5% sodium deoxycholate. Solubilized proteins were subjected to a column (1.5 cm x 45 cm) of Ultrogel AcA 34; 1.7-ml fractions were collected and protein (Q) and Cd (O) contents (haemoglobin (&) contents, for (a) only) of each fraction were determined.
Gel-filtration chromatography of the membrane fraction of Cd-loaded red ceils after solubilization with sodium deoxycholate To examine Cd-binding components in the membrane fraction, both proteins and Cd of the membrane fraction were solubilized and analysed by gel-filtration chromatography. When the membrane fraction of red blood cells incubated with 0.5 mu-Cd at 37°C for 1 hr (Cd-loaded red cells) was incubated with 3 vols 5% sodium deoxycholate (solution I) at 30°C for 20 rain, 100 and 98% of proteins and Cd, respectively, were recovered in the supernatant. The supernatant was subjected to gel-filtration chromatography on an Ultrogel AcA 34 column. Cd was eluted exclusively in the high-molecular-weight fraction (Fig. 3b, fractions 24-32). On the other hand, both proteins and Cd of the haemolysate prepared from Cd-loaded red cells were found in the lowmolecular-weight fraction (Fig. 3a, fractions 31-39).
When Cd alone in solution I was applied to the same column, it was eluted in fractions 46-80 (data not shown). Therefore, incorporated Cd neither existed in a free form in red blood cells nor was liberated from Cd-protein complexes during the solubilization procedure. During subsequent incubation of Cd-loaded red cells in the Cd-free medium, proteins in the membrane fraction gradually became resistant to solubilization and 5 and 25% of the membranefraction protein remained unsolubilized at 7 and 14 hr of incubation, respectively. Major components of the unsolubilized proteins were haemoglobin and C7 (data not shown). As shown in Fig. 3b--d, proteins and Cd in the low-molecular-weight fraction increased progressively and preferentially as the incubation time increased, whereas those in the highmolecular-weight fraction changed little. Each protein species in the eluted fractions was quantified by densitometry following separation by SDSpolyacrylamide gel electrophoresis (Fig. 4). A major component of the low-molecular-weight fraction was haemoglobin. The high-molecular-weight fraction contained bands 3, 4.1, 4.2, 4.5 and 5. Cd was scarcely detected in the peaks of spectrin (bands I and 2) and band 2.1 (fractions 20-23).
Binding of wgCd to proteins in the membrane fraction on a Durapore membrane To determine whether the cytosolic proteins which increased in the membrane fraction can bind Cd or not, the membrane fraction was prepared from Cdloaded red cells incubated in the Cd-free medium for 8 hr (Fig. 2, lane 5) and Cd-binding abilities of the proteins in the membrane fraction were examined by ~°gCd-autoradiography on a Durapore membrane using western blotting technique. Most proteins except bands 2.1 and 2 were efficiently transferred to a Durapor¢ membrane (Fig. 5a). Haemoglobin and several cytosolic proteins (C2, C3, C6 and C7), in addition to membrane proteins, showed a significant Cd-binding ability (Fig. 5b).
DISCUSSION
These results indicate that when Cd-loaded red cells are incubated in the Cd-free medium, Cd and 5
4
~o==
O~
3
o.~_
2
J~ <~ 20
30 Froction
40
50
number
Fig. 4. Elution profile of solubilized proteins of membrane fractions prepared from Cd-treated red cells. The protein composition of each fraction in Fig. 3c was determined by SDS-polyacrylamide gel electrophoresis and densitometry. Bands I and 2 (0), band 2.1 (A), band 3 (O), band 4.1 (ll), band 4.2 (ZX), band 5 ( 0 ) and haemoglobin (I-l).
62
M. KUNIMOTO and T. MIURA
cytosolic proteins associate increasingly with red cell membranes, resulting in an increase in the protein to phospholipid ratio of the membrane fraction (Figs 1 & 2). It is unlikely that the increasing association was produced by simple denaturation and precipitation of cytosolic proteins because the incubation of red cell cytosoi prepared by hypotonic lysis with Cd did not produce any precipitation (data not shown). Among cytosolic proteins, haemoglobin associated with red cell membranes to the largest extent (Figs 1 & 2). In in vivo administration experiments, Cd--once incorporated into red blood cells---exists predominantly as complexes with cytosolic proteins, especially with haemoglobin (Carlson & Friberg, 1957). This was confirmed by gel filtration of Cd-loaded red cells solubilized with deoxycholate in the present in vitro experiments (Fig. 3a). The part of the Cd that accumulated in the membrane fraction during incubation in the Cd-free medium, appears to be recovered exclusively in the low-molecular-weight fraction of gel-filtration chromatography, where haemoglobin is most abundant (Figs 3 & 4). On the other hand, the amount of Cd in the high-molecular-weight fraction did not alter appreciably over the entire incubation period (Fig. 3 ~ d ) . Thus the haemoglobin-Cd complex in cytosol progressively translocated to the membrane fraction during incubation in the Cd-free medium. The Cd-binding capacity of proteins in the membrane fraction was assessed by l°gCd autoradiography on a Durapore membrane, using western blotting technique (Fig. 5). In addition to membrane proteins, haemoglobin and several cytosolic proteins (C2, C3, C6 and C7), which associated progressively with red cell membranes, showed a significant Cd-binding ability (Fig. 5b). On the basis of these results, it is strongly suggested that the cytosolic proteins with Cd-binding ability associated progressively with red cell membranes during incubation in the Cd-free medium. The accumulation of Cd in the membrane fraction causes an acceleration of age-related changes of red blood cells, such as shape changes from biconcave disc to echinocyte and decreased filterability, without affecting intracellular ATP levels during incubation in the Cd-free medium (Kunimoto et al. 1986). The altered shape is maintained in the membrane fraction isolated from the Cd-loaded cells, suggesting that the shape changes primarily result from the alterations in the organization of red cell membranes. The incubation of membrane fractions with Cd (50-200/~M) induced severe alterations in the organization of the red cell membrane, such as shape changes and strengthened interactions between the components of spectrin-actin network and between transmembrane proteins and the spectrin-actin network (Kunimoto & Miura, 1985). Thus the shape changes are produced, in part, by Cd bound directly to membrane proteins such as bands 3, 4.2 and 5 of the highmolecular-weight fraction. It also seems likely that the membrane properties of red blood cells are modified by increased association of Cd-cytosolic proteins with the red-cell membrane, because haemoglobin and several cytosolic proteins are shown to interact or associate with membrane proteins such as band 3 and modulate membrane organization (Liu &
Palek, 1984; Sailhany & Gaines, 1981). There are several possible mechanisms for the increased association of cytosolic proteins with the red-cell membrane: it might be a consequence of an increase in their affinity for membrane components through the binding of Cd, or a change in the properties of the membrane, which results either from the presence of Cd in the external medium or from the cellular uptake process. Whatever the mechanism, such an increasing association of Cd-protein complexes with cellular organelles must be involved in functional alterations at the cellular level. Acknowledgements--The authors thank Ms Y. Kataoka
and K. Yuzawa for their technical assistance and Ms E. Nemoto and M. Mori for their secretarial help. This work was supported in part by Grants-in-Aid (Nos 59771729 and 60771958) for Scientific Research from the Ministry of Education, Science and Culture of Japan.
REFERENCES
Aoki Y., Kunimoto M., Shibata Y. & Suzuki K. T. (1986). Detection of metallothionein on nitrocellulose membrane using western blotting technique and its application to identification of cadmium-binding proteins. Ana/yt. Biochem. 157, 117-122. Berlin M. & Friberg L. (1960). Bone-marrow activity and erythrocyte destruction in chronic cadmium poisoning. Archs envir. Hlth 1, 478-486. Carlson L. A. & Friberg L. (1957). The distribution of Cd in blood after repeated exposure. Scand. J. clin. Lab. Invest. 9, 67-70. Friberg L. (1952). Further investigations on chronic cadmium poisoning: study on rabbits with radioactive cadmium. AMA Archs ind. Hyg. 5, 30-36. Friberg L., Piscator M., Nordberg G. F. & Kjellstrom T. (1974). Cadmium in the Environment. 2nd Ed. CRC Press, Cleveland, OH. Garty M., Wong K.-L. & Klaassen C. D. (1981). Redistribution of cadmium to blood of rats. Toxic. appL Pharmac. 59, 548-554. Klaassen C. D. & Kotsonis F. N. (1977). Biliary excretion of Cd in the rat, rabbit and dog. Toxic. appl. Pharmac. 41, 101-112. Kunimoto M. & Miura T. (1985). Vesicle release from rat red cell ghosts and increased association of cell membrane proteins with cytoskeletons induced by cadmium. Biochim. biophys. Acta 816, 37-45. Kunimoto M. & Miura T. (1986). Density increment and decreased survival of rat red blood cells induced by cadmium. Envir. Res. 39, 86-95. Kunimoto M., Miura T. & Kubota K. (1985). An apparent acceleration of age-related changes of rat red blood cells induced by cadmium. Toxic. appl. Pharmac. 77, 451-457. Kunimoto M., Miyasaka K. & Miura T. (1986). Changes in membrane properties of rat red blood cells induced by cadmium accumulating in the membrane fraction. J. Biochem., Tokyo 99, 397-406. Kunimoto M., Tsubone H., Tsujii N., Mochitate K., Kaya K., Shimojo N. & Miura T. (1984). Effects of nitrate and nitrite, chemical intermediates of inhaled nitrogen dioxide, on membrane components of red blood cells of rats. Toxic. appl. Pharmac. 74, 10-16. Laemmli U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, Lond. 227, 68(~685. Liu S. C. & Palek J. (1984). Hemoglobin enhances the self-association of spectrin heterodimers in human erythrocytes. J. biol. Chem. 259, 11556-11562.
Accumulation of Cd in red cell membranes Lowry O. H., Rosebrough N. J., Farr A. L. & Randall R. J. (1951). Protein measurement with the Folin phenol reagent. J. bioL Chem. 193, 265-275. Nordberg G. F., Piscator M, & Nordberg M. (1971). On the distribution of Cd in blood. Acta pharmac, tox. 30, 289-295. Sailhany J. M. & Gaines K. C. (1981). Connection between cytoplasmic proteins and the erythrocyte membrane. Trends biochem. Sci. 6, 13-15. Samarowickrama G. P. (1979). Biological effects of cadmium in mammals. In The Chemistry, Biochemistry and Biology o f Cadmium. Edited by M. Webb. pp. 341-421. Elsevier, Amsterdam.
63
Suzuki K. T. & Yamamura M. (1979), Dose dependent formation of zinc-thionein in liver and kidneys of rats and mice by zinc injection. Biochem. Pharmac. 28, 2852-2864. Towbin T., Staehelin T. & Gordon J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. natn. Acad. Sci. U.S.A. 76, 4350--4254. van Kampen E. J. & Zijlstra W. G. (1961). Standardization of hemoglobinometry. II. The hemoglobincyanide method. Clinica. Chim. Acta 6, 538-544.
0887-2333/88 $3.00+0.00 Copyright © 1988PergamonJournals Ltd
TRANSLOCATION OF C A D M I U M FROM CYTOSOL TO M E M B R A N E FRACTION IN C A D M I U M - L O A D E D RED BLOOD CELLS OF RATS M. KUNIMOTOand T. MIURA Basic Medical Sciences Division, The National Institute for Environmental Studies, Yatabe, Tsukuba, Ibaraki 305, Japan (Received 23 March 1987; revisions received 2 September 1987)
Abstract--When rat red blood cells were incubated in a cadmium (Cd)-free medium for up to 14 hr following a 1-hr treatment with 0.5 mM-CdC12, the incorporated Cd, which was predominantly in the cytosol at the beginning of the incubation, progressivelyaccumulated in the membrane fraction. In parallel with the Cd-accumulation, several cytosolic proteins including haemoglobin increased in the membrane fraction, resulting in an increase in the protein to phospholipid ratio of the membrane fraction. The membrane fraction was solubilized with sodium deoxycholate and analysed by gel-filtration chromatography. Cd was detected in the high-molecular-weightfraction containing membrane proteins at the beginning of incubation, and the Cd content of this fraction did not alter appreciably during the incubation. On the other hand, Cd in the low-molecular-weight fraction, where haemoglobin is most abundant, increased progressively during the incubation. The Cd-binding capacity of proteins in the membrane fraction was assessed by t°gCd-autoradiography using the western blotting technique. In addition to membrane proteins, haemoglobin and other cytosolic proteins, which increased in the membrane fraction during the incubation, exhibited a significant ability to bind Cd. These results indicate that Cd, once incorporated into red cell cytosol, progressively accumulates in the membrane fraction together with cytosolic proteins with Cd-binding capacity. This accumulation of Cd-cytosolicproteins may affect the membrane properties of red blood cells.
INTRODUCTION
of cellular Cd in the membrane fraction, although intracellular ATP was maintained at the control level (Kunimoto et al. 1986). At higher' Cd concentrations (l.0-2.0mM) a significafft decrease in intracellular ATP was observed, in addition to the age-related changes, during incubation in a Cd-free medium. It is, therefore, important for the assessment of Cd toxicity into red blood cells to clarify the fate of the Cd incorporated into red blood cells. In this report, we present evidence that, during incubation in a Cd-free medium, Cd in cytosol translocates to the membrane fraction together with cytosolic proteins with Cd-binding capacity.
Cadmium (Cd) is a well known industrial and environmental toxicant. Anaemia is a common finding in workers exposed to Cd and in mammals given Cd (Friberg et aL 1974; Samarawickrama, 1979). One of the mechanisms responsible for Cd-induced anaemia is decreased survival of red blood cells in Cd-treated animals (Berlin & Friberg, 1960; Kunimoto & Miura, 1986). After oral or parenteral administration Cd appears in the blood (Klaassen & Kotsonis, 1977; Samarawickrama, 1979). A significant part of the Cd in the blood is located in the red blood cells (Carlson & Friberg, 1957; Friberg, 1952; Garty et al. 1981; Nordberg et al. 1971) and the non-dialysable part of red-cell Cd is apparently predominantly bound to haemoglobin (Carlson & Friberg, 1957). Recently, we showed that the in vitro incubation of rat red blood cells with 0.5-2.0 mM-Cd at 37°C for 1 hr accelerated age-related changes in red blood cells such as shape changes, increased density, decreased flexibility and shortened in vivo survival (Kunimoto et al. 1985). However, the age-related changes in red blood cells were slight upon l-hr treatment with 0.5mM-Cd, compared with the effect on in vivo survival, and only 5% of the incorporated Cd was located in the membrane fraction. During subsequent incubation in a Cd-free medium, age-related changes developed in parallel with progressive accumulation
MATERIALS AND METHODS
Preparation and incubation o f red blood cells. Red blood cells were obtained from male Jcl:Wistar rats (body weight, 450-500g) as described before (Kunimoto et al. 1984). Washed red cells were mixed with 3 vols 50 mM-glycylglycine, pH 7.4, containing 5 mM-KCI, 116 mM-NaC1, 11.1 mM-glucose, 0.54 mMadenine, 12.7 mM-inosine, 0.2 mg streptomycin/ml and 200 IU penicillin G/ml (buffer A). After incubation at 37°C for 1 hr in the presence of 0.5 mMCdCI2, the incubation mixture was centrifuged at 2000 rev/min for 5 rain at room temperature and the pelleted cells were washed once with 10 vols buffer A, resuspended in 3 vols buffer A (Cd-free medium) and incubated at 37°C for up to 14 hr.
Abbreviation: SDS = sodium dodecyl sulphate.
57 T.I.V. 2/I--D
58
M. KUNIMOTOand T. MIURA
Preparation of red cell membranes. After incubation in the Cd-free medium for the times indicated elsewhere, red blood cells were pelleted by centrifugation at 2000 rev/min for 5 min, washed once with 10 vols saline and haemolysed by 1:20 dilution with 10 mM-Tris HCI, pH 7.4. The membrane fraction was pelleted at 15,000rev/min for 20rain at 4°C and washed four times with the same buffer. Sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis. Electrophoreses in 7.5 or 12.5% polyacrylamide containing 0.1% SDS were performed according to the method of Laemmli (1970). Gels stained with Coomassie brilliant blue R-250 were scanned in a densitometer (Auto Scanner Flur-Vis, Helena Laboratory Co., TX, USA) and the area of each peak was integrated with a digitizer (MOP-10, Kontron, Munich, FRG). Solubilization and gel-filtration chromatography of the membrane fractions. Membrane fractions were mixed with 3vols 5% sodium deoxyeholate containing 0.1 M-KCI, 10 mM-Tris HC1, pH 7.4 and 20 #g phenylmethylsulphonyl fluoride/ml (solution I), incubated at 30°C for 20min and centrifuged at 15,000 rev/min for 20 min at 30°C. Two millilitres of the supernatant were applied to a column (1.5 crn x 45 cm) of Ultrogel AcA 34 (LKB, Bromma, Sweden) equilibrated with 10mM-sodium deoxycholate containing 0.1M-KC1, 10mM-Tris HCI, pH 7.4 and 20 pg phenylmethylsulphonyl fluoride/ml (solution II) at room temperature and 1.7-ml fractions were collected (9 ml/hr). Each fraction was subjected to determinations of protein and Cd and SDS-polyacrylamide gel electrophoresis. Red blood cells incubated with 0.5 mM-CdC12 for 1 hr at 37°C were washed once with 10 vols saline and mixed with 9 vols solution I. After 20-min incubation at 30°C the mixture was centrifuged at 15,000 rev/min for 20 min at 30°C and 0.8 ml of the supernatant was also analysed as described above. Detection of Cd-binding proteins in the membrane fraction by 1°9Cd autoradiography on a Durapore membrane using western blotting technique. Proteins in the membrane fraction separated by SDSpolyacrylamide gel electrophoresis were electrophoreticaily transferred to a Durapore membrane (GVHP, Millipore Corp., MA, USA) in a Trans-blot cell (Bio-Rad, CA, USA) according to the method of Towbin et aL (1979). Transfer was performed at a constant voltage of 30 V for 30 min, followed by 150 V for 90 min at 4°C, using a solution containing 20% methanol, 25 mM-Tris and 129 mM-glycine, pH 8.5. After the transfer, the membrane was soaked in 10 mM-Tris HCI, pH 7.4 for 12 hr at 4°C and then incubated in 10mM-Tris HCI, pH7.4 containing 1 pCi 1°gCdCl2/ml (carrier free, New England Nuclear, MA, USA), 1 mM-MgCI 2 and 0.1 M-KC1 for 10min at room temperature. The membrane was washed three times with 10 mM-Tris HCI, pH 7.4 at 5-min intervals. The addition of 0.1 M-KCI is essential for reducing non-specific bindings of m9Cd to the membrane (Aoki et al. 1986). Autoradiography of the membrane was performed as described by Aoki et al. (1986). Other methods. The amount of Cd in red blood cells, membrane fractions and column fractions was determined after digestion with HNO3-60% HCIO 4
(5:l,v/v) using an atomic absorption spectrophotometer (Hitachi Model 170-50, Hitachi Co., Tokyo) as described by Suzuki & Yamamura (1979). Possible interference by sodium ions was corrected using a deuterium lamp. The protein content was estimated by the method of Lowry et al. (1951). The amount of haemoglobin was determined spectrophotometrically after conversion to cyanomethaemoglobin, according to the method of van Kampen & Zijlstra (1961).
RESULTS
Changes in the membrane composition of Cd-loaded red cells during incubation in the Cd-free medium When red blood cells were incubated with 0.5 mM-CdCI2 for 1 hr at 37°C, 35% of the exogenous Cd was incorporated into the cell. The Cd-loaded red cells were then transferred to the Cd-free medium and incubated at 37°C for 14 hr. The incorporated Cd was not released from the cell during the 14-hr incubation in the Cd-free medium (Kunimoto et aL 1986). However, the amount of Cd in the membrane fraction increased progressively (Fig. 1) and reached approximately 20% of the cellular Cd after incubation for 14 hr (data not shown). In parallel with the Cd accumulation, the protein to phospholipid ratio of the membrane fraction increased and reached 1.48 times that of control cells after the 14-hr incubation (Fig. 1). The protein composition of the membrane fraction was also determi'ned by SDS-polyacrylamide gel electrophoresis. As shown in Fig. 2, several proteins (CI-C7 and haemoglobin), which are cytosolic because they exist both in the haemolysate (lane 1) and its supernatant (data not shown), increased as the incubation time increased (lanes 3-6). Increases in haemoglobin and C7 were particularly noticeable.
3.0
~ 2- 7"~5 ~
.-~
. .
e
n
~.
2.0
1.0
//
I
0.5 F z j / / oil
0
I 2
/
"-
2
o
If'I
I 4
I 6
,],I 8
I 10
I t2
I 14
E
4
"a
L)
T i m e (hr)
Fig. 1. Changes in the protein to phospholipid ratio and Cd content of membrane fractions prepared from Cd-treated red cells during incubation in the Cd-free medium. Rat red blood cells were pre-incubated with (O) or without (O) 0.5-mM CdC12 for I hr and the protein to phospholipid ratio (Q O) and the Cd content (A) of their membrane fractions were determined after incubation in the Cd-free medium. Each point represents the mean of three experiments and error bars represent standard deviations.
1,2.1
4.1a,b 4.2
•
Ct IIm,-.Cz
-4--4.5
C3 ID"
-.~--5 t,.-
C5 ~
~
C6
C7 m,,,--
m
0
/
m
Hb
a
"1
2
:3
4
5
-4
6
Fig. 2. SDS-polyacrylamide gel electrophoresis of membrane fractions prepared from Cd-treated red cells. Membrane fractions were prepared from Cd-treated red cells 0 (lane 3), 4 (lane 4), 8 (lane 5) and 14 hr (lane 6) after incubation in the Cd-free medium and from untreated red cells (lane 2) and were analysed by SDS-polyacrylamide gel electophoresis. Whole red blood cells were also analysed (lane I). Hb = haemoglobin and Ct-C 7 = cytosolic proteins which increasingly associated with the membrane fraction. (For Figs 3 & 4, see p. 61)
59
0)
Hb 4.2
>,
p
E 0
D
(b)
8 o
Fig. 5. Detection of Cd-binding proteins in the membrane fraction by ~°9Cd-autoradiography on a Durapore membrane. Membrane fractions were prepared from Cd-treated red cells 8 hr after incubation in the Cd-free medium. Proteins of the membrane fraction were separated by SDS-polyacrylamide gel electrophoresis (Fig. 2, lane 5) and were electrophoretically transferred to a Durapore membrane; (a) proteins on the membrane were stained with amido black; (b) the membrane was incubated with ~°9Cd and its autoradiography was obtained. Both were analysed with a densitometer.
60
!leo
•E
0.8
f 0.6
8
0
~oo,(b)
f"~.z
=50 t.
0 : -±
~_
O.,
.j
.
.
.
.
.
=
L
-~
0
2
~oo1 (c)
A
0.'I B
511... . -
~oo
.
.
_%
--
L_L".
--
, 0
o.z
(d)
0
10
20
61
Accumulation of Cd in red ceil membranes
30 Froction
40
50
60
70
80
number
Fig. 3. Gel filtration of solubilized membrane fractions prepared from Cd-treated red cells. Cd-treated red cells (a) and membrane fractions prepared from Cd-treated red cells 0 (b), 7 (c) and 14hr (d) after incubation in the Cd-free medium were solubilized by 5% sodium deoxycholate. Solubilized proteins were subjected to a column (1.5 cm x 45 cm) of Ultrogel AcA 34; 1.7-ml fractions were collected and protein (Q) and Cd (O) contents (haemoglobin (&) contents, for (a) only) of each fraction were determined.
Gel-filtration chromatography of the membrane fraction of Cd-loaded red ceils after solubilization with sodium deoxycholate To examine Cd-binding components in the membrane fraction, both proteins and Cd of the membrane fraction were solubilized and analysed by gel-filtration chromatography. When the membrane fraction of red blood cells incubated with 0.5 mu-Cd at 37°C for 1 hr (Cd-loaded red cells) was incubated with 3 vols 5% sodium deoxycholate (solution I) at 30°C for 20 rain, 100 and 98% of proteins and Cd, respectively, were recovered in the supernatant. The supernatant was subjected to gel-filtration chromatography on an Ultrogel AcA 34 column. Cd was eluted exclusively in the high-molecular-weight fraction (Fig. 3b, fractions 24-32). On the other hand, both proteins and Cd of the haemolysate prepared from Cd-loaded red cells were found in the lowmolecular-weight fraction (Fig. 3a, fractions 31-39).
When Cd alone in solution I was applied to the same column, it was eluted in fractions 46-80 (data not shown). Therefore, incorporated Cd neither existed in a free form in red blood cells nor was liberated from Cd-protein complexes during the solubilization procedure. During subsequent incubation of Cd-loaded red cells in the Cd-free medium, proteins in the membrane fraction gradually became resistant to solubilization and 5 and 25% of the membranefraction protein remained unsolubilized at 7 and 14 hr of incubation, respectively. Major components of the unsolubilized proteins were haemoglobin and C7 (data not shown). As shown in Fig. 3b--d, proteins and Cd in the low-molecular-weight fraction increased progressively and preferentially as the incubation time increased, whereas those in the highmolecular-weight fraction changed little. Each protein species in the eluted fractions was quantified by densitometry following separation by SDSpolyacrylamide gel electrophoresis (Fig. 4). A major component of the low-molecular-weight fraction was haemoglobin. The high-molecular-weight fraction contained bands 3, 4.1, 4.2, 4.5 and 5. Cd was scarcely detected in the peaks of spectrin (bands I and 2) and band 2.1 (fractions 20-23).
Binding of wgCd to proteins in the membrane fraction on a Durapore membrane To determine whether the cytosolic proteins which increased in the membrane fraction can bind Cd or not, the membrane fraction was prepared from Cdloaded red cells incubated in the Cd-free medium for 8 hr (Fig. 2, lane 5) and Cd-binding abilities of the proteins in the membrane fraction were examined by ~°gCd-autoradiography on a Durapore membrane using western blotting technique. Most proteins except bands 2.1 and 2 were efficiently transferred to a Durapor¢ membrane (Fig. 5a). Haemoglobin and several cytosolic proteins (C2, C3, C6 and C7), in addition to membrane proteins, showed a significant Cd-binding ability (Fig. 5b).
DISCUSSION
These results indicate that when Cd-loaded red cells are incubated in the Cd-free medium, Cd and 5
4
~o==
O~
3
o.~_
2
J~ <~ 20
30 Froction
40
50
number
Fig. 4. Elution profile of solubilized proteins of membrane fractions prepared from Cd-treated red cells. The protein composition of each fraction in Fig. 3c was determined by SDS-polyacrylamide gel electrophoresis and densitometry. Bands I and 2 (0), band 2.1 (A), band 3 (O), band 4.1 (ll), band 4.2 (ZX), band 5 ( 0 ) and haemoglobin (I-l).
62
M. KUNIMOTO and T. MIURA
cytosolic proteins associate increasingly with red cell membranes, resulting in an increase in the protein to phospholipid ratio of the membrane fraction (Figs 1 & 2). It is unlikely that the increasing association was produced by simple denaturation and precipitation of cytosolic proteins because the incubation of red cell cytosoi prepared by hypotonic lysis with Cd did not produce any precipitation (data not shown). Among cytosolic proteins, haemoglobin associated with red cell membranes to the largest extent (Figs 1 & 2). In in vivo administration experiments, Cd--once incorporated into red blood cells---exists predominantly as complexes with cytosolic proteins, especially with haemoglobin (Carlson & Friberg, 1957). This was confirmed by gel filtration of Cd-loaded red cells solubilized with deoxycholate in the present in vitro experiments (Fig. 3a). The part of the Cd that accumulated in the membrane fraction during incubation in the Cd-free medium, appears to be recovered exclusively in the low-molecular-weight fraction of gel-filtration chromatography, where haemoglobin is most abundant (Figs 3 & 4). On the other hand, the amount of Cd in the high-molecular-weight fraction did not alter appreciably over the entire incubation period (Fig. 3 ~ d ) . Thus the haemoglobin-Cd complex in cytosol progressively translocated to the membrane fraction during incubation in the Cd-free medium. The Cd-binding capacity of proteins in the membrane fraction was assessed by l°gCd autoradiography on a Durapore membrane, using western blotting technique (Fig. 5). In addition to membrane proteins, haemoglobin and several cytosolic proteins (C2, C3, C6 and C7), which associated progressively with red cell membranes, showed a significant Cd-binding ability (Fig. 5b). On the basis of these results, it is strongly suggested that the cytosolic proteins with Cd-binding ability associated progressively with red cell membranes during incubation in the Cd-free medium. The accumulation of Cd in the membrane fraction causes an acceleration of age-related changes of red blood cells, such as shape changes from biconcave disc to echinocyte and decreased filterability, without affecting intracellular ATP levels during incubation in the Cd-free medium (Kunimoto et al. 1986). The altered shape is maintained in the membrane fraction isolated from the Cd-loaded cells, suggesting that the shape changes primarily result from the alterations in the organization of red cell membranes. The incubation of membrane fractions with Cd (50-200/~M) induced severe alterations in the organization of the red cell membrane, such as shape changes and strengthened interactions between the components of spectrin-actin network and between transmembrane proteins and the spectrin-actin network (Kunimoto & Miura, 1985). Thus the shape changes are produced, in part, by Cd bound directly to membrane proteins such as bands 3, 4.2 and 5 of the highmolecular-weight fraction. It also seems likely that the membrane properties of red blood cells are modified by increased association of Cd-cytosolic proteins with the red-cell membrane, because haemoglobin and several cytosolic proteins are shown to interact or associate with membrane proteins such as band 3 and modulate membrane organization (Liu &
Palek, 1984; Sailhany & Gaines, 1981). There are several possible mechanisms for the increased association of cytosolic proteins with the red-cell membrane: it might be a consequence of an increase in their affinity for membrane components through the binding of Cd, or a change in the properties of the membrane, which results either from the presence of Cd in the external medium or from the cellular uptake process. Whatever the mechanism, such an increasing association of Cd-protein complexes with cellular organelles must be involved in functional alterations at the cellular level. Acknowledgements--The authors thank Ms Y. Kataoka
and K. Yuzawa for their technical assistance and Ms E. Nemoto and M. Mori for their secretarial help. This work was supported in part by Grants-in-Aid (Nos 59771729 and 60771958) for Scientific Research from the Ministry of Education, Science and Culture of Japan.
REFERENCES
Aoki Y., Kunimoto M., Shibata Y. & Suzuki K. T. (1986). Detection of metallothionein on nitrocellulose membrane using western blotting technique and its application to identification of cadmium-binding proteins. Ana/yt. Biochem. 157, 117-122. Berlin M. & Friberg L. (1960). Bone-marrow activity and erythrocyte destruction in chronic cadmium poisoning. Archs envir. Hlth 1, 478-486. Carlson L. A. & Friberg L. (1957). The distribution of Cd in blood after repeated exposure. Scand. J. clin. Lab. Invest. 9, 67-70. Friberg L. (1952). Further investigations on chronic cadmium poisoning: study on rabbits with radioactive cadmium. AMA Archs ind. Hyg. 5, 30-36. Friberg L., Piscator M., Nordberg G. F. & Kjellstrom T. (1974). Cadmium in the Environment. 2nd Ed. CRC Press, Cleveland, OH. Garty M., Wong K.-L. & Klaassen C. D. (1981). Redistribution of cadmium to blood of rats. Toxic. appL Pharmac. 59, 548-554. Klaassen C. D. & Kotsonis F. N. (1977). Biliary excretion of Cd in the rat, rabbit and dog. Toxic. appl. Pharmac. 41, 101-112. Kunimoto M. & Miura T. (1985). Vesicle release from rat red cell ghosts and increased association of cell membrane proteins with cytoskeletons induced by cadmium. Biochim. biophys. Acta 816, 37-45. Kunimoto M. & Miura T. (1986). Density increment and decreased survival of rat red blood cells induced by cadmium. Envir. Res. 39, 86-95. Kunimoto M., Miura T. & Kubota K. (1985). An apparent acceleration of age-related changes of rat red blood cells induced by cadmium. Toxic. appl. Pharmac. 77, 451-457. Kunimoto M., Miyasaka K. & Miura T. (1986). Changes in membrane properties of rat red blood cells induced by cadmium accumulating in the membrane fraction. J. Biochem., Tokyo 99, 397-406. Kunimoto M., Tsubone H., Tsujii N., Mochitate K., Kaya K., Shimojo N. & Miura T. (1984). Effects of nitrate and nitrite, chemical intermediates of inhaled nitrogen dioxide, on membrane components of red blood cells of rats. Toxic. appl. Pharmac. 74, 10-16. Laemmli U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, Lond. 227, 68(~685. Liu S. C. & Palek J. (1984). Hemoglobin enhances the self-association of spectrin heterodimers in human erythrocytes. J. biol. Chem. 259, 11556-11562.
Accumulation of Cd in red cell membranes Lowry O. H., Rosebrough N. J., Farr A. L. & Randall R. J. (1951). Protein measurement with the Folin phenol reagent. J. bioL Chem. 193, 265-275. Nordberg G. F., Piscator M, & Nordberg M. (1971). On the distribution of Cd in blood. Acta pharmac, tox. 30, 289-295. Sailhany J. M. & Gaines K. C. (1981). Connection between cytoplasmic proteins and the erythrocyte membrane. Trends biochem. Sci. 6, 13-15. Samarowickrama G. P. (1979). Biological effects of cadmium in mammals. In The Chemistry, Biochemistry and Biology o f Cadmium. Edited by M. Webb. pp. 341-421. Elsevier, Amsterdam.
63
Suzuki K. T. & Yamamura M. (1979), Dose dependent formation of zinc-thionein in liver and kidneys of rats and mice by zinc injection. Biochem. Pharmac. 28, 2852-2864. Towbin T., Staehelin T. & Gordon J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. natn. Acad. Sci. U.S.A. 76, 4350--4254. van Kampen E. J. & Zijlstra W. G. (1961). Standardization of hemoglobinometry. II. The hemoglobincyanide method. Clinica. Chim. Acta 6, 538-544.