- Email: [email protected]
Replacement of transferrin in serum-free cultures of mitogen-stimulated mouse lymphocytes by a lipophilic iron chelator
Immunology Letters, 15 (1987) 23-25 Elsevier IML 00865
Replacement of transferrin in serum-free cultures of mitogenstimulated mouse lymphocytes by a lipophilic iron chelator J. H. B r o c k a n d J o a n Stevenson University Department of Bacteriology and Immunology, Western Infirmary, Glasgow, U.K. (Received 29 October 1986) (Revision received and accepted 2 January 1987)
It is now well established that the iron-binding protein transferrin is essential for lymphocyte proliferation [1, 2]. Other forms of iron, including inorganic iron salts and chelates such as Fe-citrate or Fe-nitrilotriacetate (FeNTA) cannot replace transferrin-bound iron [3], and for this reason transferrin is invariably required in serum-free cultures of lymphocytes, and of transformed cell lines
[4]. Although the major function of transferrin is probably to supply the cells with metabolicallyavailable iron, it has been suggested that transferrin might fulfil some additional membrane-signalling function which is required for proliferation [1, 5]. We have, however, presented evidence indicating that the ability of transferrin to promote lymphocyte proliferation is closely related to its ability to donate iron to the cells [6], but unequivocal proof that transferrin does not provide any additional membrane-signalling event would require the demonstration that proliferation can occur in the absence of transferrin. Ponka et al. have identified a lipophilic iron chelator, pyridoxal isonicotinoyl hydrazone (PIH), which, unlike citrate and nitrilotriacetate, was able to donate iron to erythroid precursors for use in haem synthesis [7] and subsequently Landschulz et al. [8] showed that FePIH permitted growth of embryonic kidney cells in vitro in the absence of transferrin. Since PIH bypasses the transferrin"~ransferrin receptor pathway of iron uptake [7], we considered it of interest to investigate whether FePIH would support proliferation of mitogenstimulated lymphocytes, as this could further clarify the role of transferrin in lymphocyte proliferation.
Key words." Iron; Lymphocyte Serum-free medium
3. Materials and Methods
1. Summary
Proliferation of mouse lymph node lymphocytes in response to concanavalin A in serum-free medium is normally dependent upon the presence of transferrin. In the absence of transferrin, little proliferation occurred, but the response was restored by addition of the iron complex of pyridoxal isonicotinoyl hydrazone (FePIH), a lipophilic iron chelator. Since cellular acquisition of PIH-bound iron is known not to involve the transferrin receptor, these results indicate that transferrin promotes lymphocyte proliferation solely because of its irondonating properties, and does not provide any additional signalling event for proliferation.
2. Introduction
proliferation;
Transferrin;
Correspondence to: Dr. J. H. Brock, University Department of Bacteriology and Immunology, Western Infirmary, Glasgow Gll 6NT, U.K.
H u m a n iron-free (apo) transferrin and human serum albumin were obtained from Behringwerke
0165-2478 / 87 / $ 3.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
23
(Hoechst, Hounslow, London). The latter contained no detectable transferrin. Concanavalin A (Con A) was obtained from Miles Laboratories (Slough, England). Pyridoxal isonicotinoyl hydrazone (PIH) was kindly provided by Dr. P. Ponka, McGill University, Montreal, Canada. 3.1. Lymphocyte cultures L y m p h - n o d e cells from adult B A L B / c mice were cultured for 48 h in microtitre plates using serumfree m e d i u m consisting o f R P M I 1640 (without H E P E S ; Flow Laboratories, Irvine, Scotland) supplemented with h u m a n serum albumin (1 m g / m l ) and 2-mercaptoethanol (50/zM), as described previously [9]. Proliferation was induced by C o n A (1 /zg/ml) and assessed by [laC]thymidine incorporation [9]. Transferrin was added at 50 #g/ml, and was 30% saturated with iron as described previously [3]. A stock solution o f P I H (0.5 m M ) was made in deionised water, a few drops o f 0.1 M HC1 being added and the solution warmed to 60 °C to aid dissolution o f the chelator. Iron was then added from a stock solution o f 10 m M FeNTA. Unless otherwise stated, the final concentration o f P I H was 40 I~M, and the F e : P I H molar ratio was 1:2. For each experiment the results are expressed as the mean o f 7 replicate cultures, and controls without mitogen (quadruplicate cultures) were always included.
14C-Thy. incorporation (c.p.m. x 10-3) 5432-
T FeTf
(30~0
No FePII" addition
PIH
FeNTA
sat'n)
Fig. 1. Proliferation of mouse lymph node cells in response to concanavalin A in serum-free medium containing various iron compounds. Tf = transferrin; PIH = pyridoxal isonicotinoyl hydrazone; NTA = nitrilotriacetate.
produced by the lymphocytes. However, this seems highly unlikely as FeNTA was unable to p r o m o t e proliferation, despite the fact that this chelate readily donates its iron to transferrin [11], and furthermore uptake o f iron from F e P I H is k n o w n to proceed by a mechanism independent o f the transferrin receptor [7]. Optimal proliferation occurred at a concentration o f P I H between 10 ~M and 50 #M (Table 1).
4. R e s u l t s a n d D i s c u s s i o n
Mouse lymphocytes proliferated well in serumfree m e d i u m containing transferrin (Fig. 1), but p o o r l y if transferrin was omitted, in agreement with previous findings [3]. However, g o o d proliferation also occurred in m e d i u m supplemented with FePIH. P I H alone produced only a slight enhancement o f proliferation over that seen when neither transferrin nor F e P I H were present and, as reported previously [3], FeNTA was totally ineffective. It thus appeared that iron b o u n d to P I H was able to replace transferrin-bound iron in allowing proliferation to occur. Since it has been reported that lymphocytes synthesise transferrin [10] it is theoretically possible that iron b o u n d to P I H was first transferred to trace a m o u n t s o f transferrin 24
Table 1 Proliferation of mouse lymphocytes in response to Con A in serum-free medium containing different concentrations of Fepyridoxal isonicotinoyl hydrazone (FePIH). FePIH (#M) 0 0.4 2 4 10 20 50
[14C]Thymidine incorporation (cpm) 144+ 152" 144+ 74 305 _+ 59 935 +_ 635 2275 + 446 2457_+ 172 2937 _+1218
100
20 +
2
400
21 _+
2
* Mean + SD; n = 7. For conditions see text.
Table 2 Effect of different proportions of iron and pyridoxal isonicotinoyl hydrazone (PIH) on proliferation of mouse lymphocytes. Cells were cultured in serum-free medium containing 1 ~zg/ml of Con A. Counts in cultures without Con A never exceeded 128 cpm. Different ratios of Fe:PIH were obtained by holding the PIH concentration at 40 #M and varying the iron concentration. Medium containing
Molar ratio Fe:PIH
[14C]Thymidine incorporation (cpm)
PIH FePIH FePIH FePIH FePIH Transferrin (30% Fe-saturated) No addition
0 0.25 0.5 0.75 1.0
669+_ 464* 3112 +_ 1474 3056_+ 521 2396_+ 416 432+- 658 2632 +- 383 502+- 98
* Mean +_ SD; n = 7 .
Higher concentrations were inhibitory. Proliferation also decreased when the Fe:PIH ratio was raised to 1:1 (Table 2). This might be a result of decreased hydrophobicity of the complex, or alternatively it could be due to excessive iron uptake. Since iron bound to P I H bypasses the transferrin receptor route it is possible that its uptake is less well controlled than that of transferrin-bound iron, and any excess over that needed for biosynthetic processes becomes toxic for the cells. The fact that the response of lymphocytes to mitogens in serum-free medium containing FePIH is virtually the same as when Fe-transferrin is present would seem to finally exclude any possibility that transferrin itself promotes proliferation by some membrane-signalling event as well as by supplying iron. This conclusion is in agreement with our earlier studies showing that proliferation depends upon whether the cells acquire iron from transferrin [6], and also with those of Kay and Benzie [12] who found that transferrin was required for
D N A synthesis but not for initiating entry into Sphase. P I H may be a useful alternative to transferrin in serum-free tissue culture medium when elimination of exogenous proteins from the medium is desirable. The failure of FeNTA to promote lymphocyte proliferation, despite the fact that it donates iron to lymphocytes [13], suggests that iron bound to this chelator is handled by cells in a way that fails to make the metal available for metabolic use. The fact that FePIH is lipophilic, whereas FeNTA is hydrophilic may be of relevance, as the latter is less likely to be able to traverse membranes and may therefore remain trapped within endocytotic vesicles. These chelators may be of use in helping to understand intracellular events in iron acquisition.
References [1] Brock, J. H. and Mainou-Fowler, T. (1983) Immunol. Today 4, 347-351. [2] Dillner-Centerlind, M.-L., Hammarstr6m, S. and Perlmann, P. (1979) Eur. J. lmmunol. 9, 942-948. [3] Brock, J. H. (1981) Immunology 43, 387-391. [4] Barnes, D. and Sato, G. (1980) Anal. Biochem. 102, 255 -270. [5] May, W. S. and Cuatrecasas, P. (1985) J. Membrane Biol. 88, 205-216. [6] Brock, J. H., Mainou-Fowler, T. and Webster, L. M. (1986) Immunology 57, 105-110. [7] Ponka, P., Schulman, H. M. and Wilczynska, A. (1982) Biochim. Biophys. Acta 718, 151-156. [8] Landschulz, W., Thesleff, I. and Ekblom, P. (1984) J. Cell Biol. 98, 596-601. [9] Mainou-Fowler, T. and Brock, J. H. (1985) Immunology 54, 325- 332. [10] Lum, J. B., Infante, A. J., Makker, D. M., Yang, E and Bowman, B. H. (1986) J. Clin. Invest. 77, 841-849. [11] Bates, G. W. and Schlabach, M. R. (1973) J. Biol. Chem. 248, 3228-3232. [12] Kay, J. E. and Benzie, C. R. (1986) lmmunol. Lett. 12, 55-58. [13] Brock, J. H. and Rankin, M. C. (1981) Immunology 43, 393 - 398.
25
Replacement of transferrin in serum-free cultures of mitogenstimulated mouse lymphocytes by a lipophilic iron chelator J. H. B r o c k a n d J o a n Stevenson University Department of Bacteriology and Immunology, Western Infirmary, Glasgow, U.K. (Received 29 October 1986) (Revision received and accepted 2 January 1987)
It is now well established that the iron-binding protein transferrin is essential for lymphocyte proliferation [1, 2]. Other forms of iron, including inorganic iron salts and chelates such as Fe-citrate or Fe-nitrilotriacetate (FeNTA) cannot replace transferrin-bound iron [3], and for this reason transferrin is invariably required in serum-free cultures of lymphocytes, and of transformed cell lines
[4]. Although the major function of transferrin is probably to supply the cells with metabolicallyavailable iron, it has been suggested that transferrin might fulfil some additional membrane-signalling function which is required for proliferation [1, 5]. We have, however, presented evidence indicating that the ability of transferrin to promote lymphocyte proliferation is closely related to its ability to donate iron to the cells [6], but unequivocal proof that transferrin does not provide any additional membrane-signalling event would require the demonstration that proliferation can occur in the absence of transferrin. Ponka et al. have identified a lipophilic iron chelator, pyridoxal isonicotinoyl hydrazone (PIH), which, unlike citrate and nitrilotriacetate, was able to donate iron to erythroid precursors for use in haem synthesis [7] and subsequently Landschulz et al. [8] showed that FePIH permitted growth of embryonic kidney cells in vitro in the absence of transferrin. Since PIH bypasses the transferrin"~ransferrin receptor pathway of iron uptake [7], we considered it of interest to investigate whether FePIH would support proliferation of mitogenstimulated lymphocytes, as this could further clarify the role of transferrin in lymphocyte proliferation.
Key words." Iron; Lymphocyte Serum-free medium
3. Materials and Methods
1. Summary
Proliferation of mouse lymph node lymphocytes in response to concanavalin A in serum-free medium is normally dependent upon the presence of transferrin. In the absence of transferrin, little proliferation occurred, but the response was restored by addition of the iron complex of pyridoxal isonicotinoyl hydrazone (FePIH), a lipophilic iron chelator. Since cellular acquisition of PIH-bound iron is known not to involve the transferrin receptor, these results indicate that transferrin promotes lymphocyte proliferation solely because of its irondonating properties, and does not provide any additional signalling event for proliferation.
2. Introduction
proliferation;
Transferrin;
Correspondence to: Dr. J. H. Brock, University Department of Bacteriology and Immunology, Western Infirmary, Glasgow Gll 6NT, U.K.
H u m a n iron-free (apo) transferrin and human serum albumin were obtained from Behringwerke
0165-2478 / 87 / $ 3.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
23
(Hoechst, Hounslow, London). The latter contained no detectable transferrin. Concanavalin A (Con A) was obtained from Miles Laboratories (Slough, England). Pyridoxal isonicotinoyl hydrazone (PIH) was kindly provided by Dr. P. Ponka, McGill University, Montreal, Canada. 3.1. Lymphocyte cultures L y m p h - n o d e cells from adult B A L B / c mice were cultured for 48 h in microtitre plates using serumfree m e d i u m consisting o f R P M I 1640 (without H E P E S ; Flow Laboratories, Irvine, Scotland) supplemented with h u m a n serum albumin (1 m g / m l ) and 2-mercaptoethanol (50/zM), as described previously [9]. Proliferation was induced by C o n A (1 /zg/ml) and assessed by [laC]thymidine incorporation [9]. Transferrin was added at 50 #g/ml, and was 30% saturated with iron as described previously [3]. A stock solution o f P I H (0.5 m M ) was made in deionised water, a few drops o f 0.1 M HC1 being added and the solution warmed to 60 °C to aid dissolution o f the chelator. Iron was then added from a stock solution o f 10 m M FeNTA. Unless otherwise stated, the final concentration o f P I H was 40 I~M, and the F e : P I H molar ratio was 1:2. For each experiment the results are expressed as the mean o f 7 replicate cultures, and controls without mitogen (quadruplicate cultures) were always included.
14C-Thy. incorporation (c.p.m. x 10-3) 5432-
T FeTf
(30~0
No FePII" addition
PIH
FeNTA
sat'n)
Fig. 1. Proliferation of mouse lymph node cells in response to concanavalin A in serum-free medium containing various iron compounds. Tf = transferrin; PIH = pyridoxal isonicotinoyl hydrazone; NTA = nitrilotriacetate.
produced by the lymphocytes. However, this seems highly unlikely as FeNTA was unable to p r o m o t e proliferation, despite the fact that this chelate readily donates its iron to transferrin [11], and furthermore uptake o f iron from F e P I H is k n o w n to proceed by a mechanism independent o f the transferrin receptor [7]. Optimal proliferation occurred at a concentration o f P I H between 10 ~M and 50 #M (Table 1).
4. R e s u l t s a n d D i s c u s s i o n
Mouse lymphocytes proliferated well in serumfree m e d i u m containing transferrin (Fig. 1), but p o o r l y if transferrin was omitted, in agreement with previous findings [3]. However, g o o d proliferation also occurred in m e d i u m supplemented with FePIH. P I H alone produced only a slight enhancement o f proliferation over that seen when neither transferrin nor F e P I H were present and, as reported previously [3], FeNTA was totally ineffective. It thus appeared that iron b o u n d to P I H was able to replace transferrin-bound iron in allowing proliferation to occur. Since it has been reported that lymphocytes synthesise transferrin [10] it is theoretically possible that iron b o u n d to P I H was first transferred to trace a m o u n t s o f transferrin 24
Table 1 Proliferation of mouse lymphocytes in response to Con A in serum-free medium containing different concentrations of Fepyridoxal isonicotinoyl hydrazone (FePIH). FePIH (#M) 0 0.4 2 4 10 20 50
[14C]Thymidine incorporation (cpm) 144+ 152" 144+ 74 305 _+ 59 935 +_ 635 2275 + 446 2457_+ 172 2937 _+1218
100
20 +
2
400
21 _+
2
* Mean + SD; n = 7. For conditions see text.
Table 2 Effect of different proportions of iron and pyridoxal isonicotinoyl hydrazone (PIH) on proliferation of mouse lymphocytes. Cells were cultured in serum-free medium containing 1 ~zg/ml of Con A. Counts in cultures without Con A never exceeded 128 cpm. Different ratios of Fe:PIH were obtained by holding the PIH concentration at 40 #M and varying the iron concentration. Medium containing
Molar ratio Fe:PIH
[14C]Thymidine incorporation (cpm)
PIH FePIH FePIH FePIH FePIH Transferrin (30% Fe-saturated) No addition
0 0.25 0.5 0.75 1.0
669+_ 464* 3112 +_ 1474 3056_+ 521 2396_+ 416 432+- 658 2632 +- 383 502+- 98
* Mean +_ SD; n = 7 .
Higher concentrations were inhibitory. Proliferation also decreased when the Fe:PIH ratio was raised to 1:1 (Table 2). This might be a result of decreased hydrophobicity of the complex, or alternatively it could be due to excessive iron uptake. Since iron bound to P I H bypasses the transferrin receptor route it is possible that its uptake is less well controlled than that of transferrin-bound iron, and any excess over that needed for biosynthetic processes becomes toxic for the cells. The fact that the response of lymphocytes to mitogens in serum-free medium containing FePIH is virtually the same as when Fe-transferrin is present would seem to finally exclude any possibility that transferrin itself promotes proliferation by some membrane-signalling event as well as by supplying iron. This conclusion is in agreement with our earlier studies showing that proliferation depends upon whether the cells acquire iron from transferrin [6], and also with those of Kay and Benzie [12] who found that transferrin was required for
D N A synthesis but not for initiating entry into Sphase. P I H may be a useful alternative to transferrin in serum-free tissue culture medium when elimination of exogenous proteins from the medium is desirable. The failure of FeNTA to promote lymphocyte proliferation, despite the fact that it donates iron to lymphocytes [13], suggests that iron bound to this chelator is handled by cells in a way that fails to make the metal available for metabolic use. The fact that FePIH is lipophilic, whereas FeNTA is hydrophilic may be of relevance, as the latter is less likely to be able to traverse membranes and may therefore remain trapped within endocytotic vesicles. These chelators may be of use in helping to understand intracellular events in iron acquisition.
References [1] Brock, J. H. and Mainou-Fowler, T. (1983) Immunol. Today 4, 347-351. [2] Dillner-Centerlind, M.-L., Hammarstr6m, S. and Perlmann, P. (1979) Eur. J. lmmunol. 9, 942-948. [3] Brock, J. H. (1981) Immunology 43, 387-391. [4] Barnes, D. and Sato, G. (1980) Anal. Biochem. 102, 255 -270. [5] May, W. S. and Cuatrecasas, P. (1985) J. Membrane Biol. 88, 205-216. [6] Brock, J. H., Mainou-Fowler, T. and Webster, L. M. (1986) Immunology 57, 105-110. [7] Ponka, P., Schulman, H. M. and Wilczynska, A. (1982) Biochim. Biophys. Acta 718, 151-156. [8] Landschulz, W., Thesleff, I. and Ekblom, P. (1984) J. Cell Biol. 98, 596-601. [9] Mainou-Fowler, T. and Brock, J. H. (1985) Immunology 54, 325- 332. [10] Lum, J. B., Infante, A. J., Makker, D. M., Yang, E and Bowman, B. H. (1986) J. Clin. Invest. 77, 841-849. [11] Bates, G. W. and Schlabach, M. R. (1973) J. Biol. Chem. 248, 3228-3232. [12] Kay, J. E. and Benzie, C. R. (1986) lmmunol. Lett. 12, 55-58. [13] Brock, J. H. and Rankin, M. C. (1981) Immunology 43, 393 - 398.
25