- Email: [email protected]
Differential Changes in Transforming Growth Factor-β Isoform Expression during Postnatal Cardiac Growth
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
245, 923–927 (1998)
RC988528
Differential Changes in Transforming Growth Factor-b Isoform Expression during Postnatal Cardiac Growth Robert S. Haworth,*,† Gavin Brooks,* Peter Cummins,† Karen Dobie,† Derrick C. Chilton,† and Metin Avkiran* *Cardiovascular Research, The Rayne Institute, St. Thomas’ Hospital, London SE1 7EH; and †Department of Physiology, University of Birmingham, Birmingham, United Kingdom
Received March 23, 1998
Transforming growth factor-b (TGF-b) is synthesised as an inactive precursor protein; this is cleaved to produce the mature peptide and a latency associated protein (LAP), which remains associated with the mature peptide until activation by LAP degradation. Isoform specific antibodies raised against the LAPs for TGF-b2 and -b3 were used to determine the myocardial levels of LAP (activatable TGF-b) and full length precursor (inactive TGF-b) forms during post-natal development in the rat. TGF-b2 was present predominantly as the precursor in 2 day old myocardium. There was an age-dependent shift from precursor protein to LAP between 2 and 28 days. A corresponding increase in the level of mature (activatable) TGF-b2 was found. TGFb3 was detected in significant quantities only as LAP. However, a four-fold increase in the expression of TGF-b3 LAP was observed between 2 and 28 days. The substantial increases in activatable forms of TGF-b2 and -b3 that occur in myocardium during the first 28 days of life in the rat support a role for these proteins in post-natal cardiac development. q 1998 Academic Press
The TGF-b family belongs to a superfamily of related proteins which affect cell growth, differentiation and extracellular matrix protein (ECM) production (reviewed by Massague (1)). Members of the TGF-b family are found in a wide variety of species and tissues (1), although the specific biological functions of TGF-b isoforms depend upon the cell type and, in vitro, culture conditions (2). In most biochemical assays, the three mammalian isoforms of TGF-b appear to have interchangeable function (3, 4). However, a number of differential effects have been observed, particularly on inhibition of DNA replication in several cell types (3, 5-7). Three human forms of TGF-b (TGF-b1 , -b2 and -b3) have been cloned and sequenced (8–10). These proteins are synthesized as inactive precursor proteins consisting of either 391 (-b1), 415 (-b2) or 410 (-b3) amino
acids (9) (Figure 1, upper panel). At some point during the secretion of TGF-b, homodimers are formed between two mature proteins (each consisting of the carboxyl-terminal 112 amino acids of the full length precursor; Figure 1, lower panel), and this mature dimer is cleaved proteolytically from the rest of the precursor protein (known as latency associated protein; LAP). LAP remains associated with the mature dimer to form an inactive complex (11), which is associated with the ECM. Active TGF-b in the form of the mature homodimer can be released from the ECM by protease-mediated degradation of LAP (12). A number of reports have suggested that TGF-b plays an important role in cardiac development (13, 14), maintenance of myocyte function (15) and formation of scar tissue in heart disease (16, 17). However, there is little information in the literature regarding the post-natal expression of the precursor, LAP and mature forms of TGF-b isoforms in the rat heart. Here we describe the use of isoform-specific antibodies directed against both the LAP and mature forms of TGFb2 and the LAP form of TGF-b3 to determine post-natal changes in TGF-b precursor, LAP and mature protein expression in the rat ventricular myocardium. METHODS Animals and materials. Adult male and time-mated late term pregnant Wistar rats were purchased from B & K Universal (Hull, U.K). All animals were killed by an approved method, in accordance with the U.K. Home Office Animals (Scientific Procedures) Act, 1986. Antibody directed against mature human TGF-b2 (sc-090) was purchased from Santa Cruz Biotechnology, Inc. (California, USA). Protein A Sepharose was from Pharmacia Biotech (St. Albans, U.K.), and synthetic peptides were a kind gift from Dr. Johannes Waltenberger (Ludwig Institute for Cancer Research, Uppsala, Sweden). All other chemicals were of the purest grade commercially available. Antibody production and purification. Polyclonal rabbit antisera were raised against synthetic peptides corresponding to variable regions in each LAP immediately preceding the mature human protein sequence (TGF-b2 aa 285-300 (10), sequence LLPSYRLESQQTNRRK; TGF-b3 aa 282-298 (9), sequence MIPPHRLDNPGQ-
923
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC
Vol. 245, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS In the case of ventricular tissue from very young rats, hearts were pooled from more than one animal.
FIG. 1. Diagrammatic illustration of TGF-b precursor protein showing the location of the peptides used to raise anti-LAP and antimature TGF-b antibodies (upper panel) and (lower panel) TGF-b dimerisation, cleavage of mature peptide and LAP (a) and release of mature peptide by LAP degradation (b).
SDS–PAGE and Western blotting. Proteins were separated by size under reducing conditions using SDS-PAGE. Gels containing 12 % (w/v) acrylamide were prepared essentially as described by Laemmli (21). Protein samples (50-100 mg/lane) were solubilised in Laemmli sample buffer and run at 100 V/gel. Proteins were transferred to PVDF membrane (Amersham, Bucks, U.K.), using a Pharmacia LKB Multiphor II transfer apparatus. After blocking overnight in 10 % (w/v) skimmed milk powder in PBS, membranes were incubated with antibodies (concentration range of 3-10 mg/ml) in 3 % (w/v) skimmed milk powder/PBS for 2 hours, washed twice for 5 min with PBS containing polyoxyethylenesorbitan monolaurate (Tween 20; 0.05 % (v/v)) and once with PBS. Donkey anti-rabbit horseradish peroxidase-conjugated antibody (Amersham) was used at a dilution of 1:2000 to label bound antibody, and the antibody complex was detected using the Amersham Enhanced Chemiluminescence (ECL) system, as recommended by the manufacturer. For quantitation, bands were scanned using an LKB Ultrascan XL Laser Densitometer (Milton Keynes, U.K). Statistics. Data are expressed as mean{SEM. Statistical significance was assessed by repeated measures ANOVA, followed by Dunnett’s test. A value of Põ 0.05 was considered to be significant.
RESULTS GGQRK), as described by Waltenberger et al. (16). In the rat, the sequence of the corresponding region is known only for TGF-b2 (18) (sequence LLPSYRLESQQSSRRK), with rat TGF-b2 exhibiting only two amino acid differences from the human sequence. Polyclonal antisera were raised in New Zealand rabbits using the method of Cummins et al. (19). The isoform specificity of the antisera raised against each of the TGF-b peptides was confirmed by enzymelinked immunoassay using peptide coated plates (data not shown). Immunoglobulin was purified from the serum by ammonium sulphate precipitation followed by protein A-Sepharose adsorption. Briefly, serum (1 ml) was mixed with an equal volume of saturated ammonium sulphate in phosphate buffered saline (PBS) and incubated at 47C for 5 min with constant mixing. The precipitated samples were sedimented by centrifugation at 47C for 30 min at 6500 1 g. The pellet was dissolved in 1 ml of PBS, and dialysed overnight against PBS at 47C. The partially purified serum samples were incubated in the presence of 80 mg protein A Sepharose (pre-swollen for 30 min in PBS) for 30 min at 47C. The protein A Sepharose/ immunoglobulin complexes were isolated by centrifugation, and the antibodies released by the addition of guanidine HCl to a final concentration of 1 M. After mixing for 20 min, the samples were centrifuged at 15000 1 g, and the supernatant dialysed against PBS as described above. Tissue preparation. In a preliminary study to look at overall changes in myocardial TGF-b expression from the foetal to the adult stages, ventricular samples were taken from rats at 18 day gestation (foetal), 2 day post-natal (neonatal) and 42 days post-natal (adult). To follow changes in greater detail during post-natal development, ventricular tissue was obtained from rats aged 2, 5, 7, 10, 14, 21, 28, and 42 days. All samples were frozen in liquid nitrogen immediately after excision, ground to powder using a pestle and mortar pre-cooled with liquid nitrogen, and stored in liquid nitrogen until required. Protein was prepared from 100-300 mg of powdered tissue by mechanical homogenisation on ice in a buffer containing sucrose (250 mM), Tris (25 mM, pH 7.5), EDTA (1 mM), EGTA (1 mM), bmercaptoethanol (0.1 %), leupeptin (50 mM), benzamidine (1 mM) and PMSF (1 mM). Following homogenisation, samples were centrifuged at 10,000 rpm for 10 min at 47C. The supernatant was recovered, and the pellet subjected to a second round of homogenisation and centrifugation. Protein concentration was determined by the method of Bradford (20), using bovine serum albumin as standard.
The TGF-b2 and TGF-b3 LAP antibodies were raised against isoform-specific peptides; such antibodies have been characterised and used previously by Olofsson et al. (22) and Waltenberger et al. (16, 23), as well as in this laboratory in an earlier study (24). Age-related changes in the expression of TGF-b isoforms were initially investigated using ventricular samples obtained from animals killed at 18 days gestation (foetal), 2 days after birth (neonatal) and 42 days after birth (adult). A representative Western blot is presented in Figure 2, using antibodies directed against TGF-b2 mature peptide and TGF-b3 LAP. TGFb2 protein was detected in ventricular tissue (Figure 2, upper panel) in precursor form at a molecular weight of 45 kDa. There was a clear age-dependent change in the quantity of precursor protein, with more protein present in foetal tissue. Precursor protein was undetectable in adult tissue, although lower molecular weight species were also difficult to detect. TGF-b2 immunising peptide competed successfully for antibody (data not shown). To detect TGF-b3 proteins in rat ventricular tissue, antibody raised against TGF-b3 LAP peptide was used (Figure 2, lower panel). TGF-b3 precursor protein was not detected in any ventricular tissue sample tested. However, TGF-b3 LAP with a molecular weight of 28 kDa was detected in foetal, neonatal and adult tissue. A substantial age-dependent increase in the expression of TGF-b3 LAP was observed. To investigate these changes in more detail, the expression of TGF-b isoforms during post-natal development was investigated using ventricular samples obtained from animals aged from 2 to 28 days. Using antibody directed at the LAP region, TGF-b2 was de-
924
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC
Vol. 245, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Detection of TGF-b in foetal (F; 18 days gestation), neonatal (N; 2 days after birth) and adult (A; 42 days after birth) rat ventricular tissue. Ventricular samples from a total of 8-15 foetal or neonatal animals were pooled. Protein samples (100 mg) were separated by 12% SDS-PAGE and transferred to PVDF membranes as described in Materials and Methods. Western blots were probed for the presence of TGF-b2 using mature peptide antibody (upper panel) and TGF-b3 using LAP antibody (lower panel). Approximate molecular sizes were estimated from Rainbow molecular weight markers (Amersham). The Western blot presented here is representative of three similar experiments using tissue derived from different groups of animals.
tectable not only as full length precursor (45 kDa) but also in LAP form (30 kDa; Figure 3, upper panel). Consistent with earlier findings with mature peptide antibody (Figure 2), there was significant expression of TGF-b2 precursor protein in 2 day old myocardial tissue, with very little LAP expression. Again consistent
FIG. 3. Detection of TGF-b2 and TGF-b3 in rat ventricular tissue during post-natal development (2-28 days after birth). Ventricular tissue from a total of 3-10 animals was pooled in each age group. Protein samples (100 mg) were separated by 12% SDS-PAGE and transferred to PVDF membranes as described in Materials and Methods. Western blots were probed for the presence of TGF-b2 (upper panel) and TGF-b3 (lower panel) using isoform-specific LAP antibodies. Approximate molecular sizes were estimated from prestained molecular weight markers (Bio-Rad). The Western blots presented here are representative of at least three experiments using tissue derived from different groups of animals.
FIG. 4. Detection of TGF-b2 in rat ventricular tissue during postnatal development (2-21 days after birth). Ventricular tissue from a total of 3-10 animals was pooled in each age group. Protein samples (40 mg) were separated by 12% SDS-PAGE and transferred to PVDF membranes as described in Materials and Methods. Western blots were probed for the presence of TGF-b2 using mature peptide antibody. Approximate molecular sizes were estimated from pre-stained molecular weight markers (Bio-Rad). The Western blot presented here is representative of three similar experiments using tissue derived from different groups of animals.
with earlier findings, the quantity of TGF-b2 precursor protein decreased with increasing age, but this time we were also able to demonstrate a concomitant rise in TGF-b2 LAP. The change was very marked between 7 and 14 days after birth, and from 21 days of age, TGF-b2 was present predominantly as LAP. In contrast to the observations with TGF-b2 , experiments using TGF-b3 LAP antibody demonstrated that this isoform was detectable almost exclusively as LAP throughout the time period studied (Figure 3, lower panel) with only minor expression of the 45 kDa precursor protein at 21 and 28 days. Moreover, the quantity of TGF-b3 LAP showed an age-dependent increase from 2 days of age to 28 days. Commercial antibody directed against TGF-b2 mature peptide was used to confirm the findings with the TGF-b2 LAP antibody (Figure 4). Both precursor (45 kDa) and mature peptide (12 kDa) were detected. An age-dependent shift from precursor form to mature peptide was apparent, consistent with the earlier findings. Detection of the mature 12 kDa TGF-b2 peptide by Western blotting required long exposure times, resulting in over-exposure of precursor protein bands and the appearance of additional bands of intermediate molecular weight. All bands disappeared when immunising peptide was present in excess during incubation with primary antibody (data not shown). The relative expression of total TGF-b protein (LAP plus precursor proteins) was estimated by densitometric scanning of immunoreactive species in blots such as that shown in Figure 3. No significant differences in the total quantity of TGF-b2 was observed during the time period studied (Figure 5, upper panel). However, TGF-b3 showed a significant (Põ0.05) four-fold increase in expression over the same period (22.5 { 3.8
925
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC
Vol. 245, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
DISCUSSION
FIG. 5. Relative quantities of TGF-b isoforms in neonatal rat ventricular tissue. Data obtained from triplicate Western blots similar to those presented in Figure 3 were quantified by densitometric scanning, from a total of three separate experiments at each age. Total TGF-b (LAP plus precursor) was determined for each isoform. The quantities at each age were normalised relative to the values obtained at 28 days. The means{S.E.M. are plotted for TGF-b2 (upper panel) and TGF-b3 (lower panel). Significant differences were observed only for TGF-b3 at 21 and 28 days (* Põ0.05 vs 2 day values).
v 91.5 { 8.5 arbitary units for 2 day and 28 day, respectively; Figure 5, lower panel). Although no significant change in the total amount (LAP plus precursor) of TGF-b2 was observed during postnatal myocardial growth, this isoform was found to undergo a novel transition between the precursor protein and LAP. Figure 6 shows the relative contribution of LAP and precursor to total immunoreactive species. TGF-b2 exhibited a precursor to LAP transition over the period studied, with equal quantities of LAP and precursor at 7-10 days after birth. Corresponding estimations from Western blots with the mature TGFb2 antibody were not attempted, as the long exposure times required for detection of mature protein resulted in over-exposure of the precursor protein bands in the 2-14 day samples.
Very little is known about the function of TGF-b isoforms in the neonatal heart, although Engelmann et al. (25) have suggested recently that TGF-b1 may be important in regulating differentiation and proliferation of the cardiac myocyte. In almost all cases where TGF-b isoform expression has been investigated in the heart using specific antibodies, immunohistochemical methods have been employed, generally using antibodies raised against the mature part of the protein. However, such antibodies would react with both the mature peptide and full length precursor, and immunohistochemical studies provide no indication of the molecular size of the protein present. To date, there has been no indication of the relative contributions of full length precursor and LAP to the total TGF-b signal observed. This is important since precursor protein would be expected to be inactive and probably localised intracellularly at the site of synthesis, whereas LAP would be predicted to be extracellular (in a latent complex with mature TGF-b) and available for enzymatic activation. In order to determine the levels of TGF-b isoform precursor and LAP in the post-natal rat heart, we used antibodies directed either against heterogeneous regions of LAP from human TGF-b2 and -b3 sequences, or commercial antibody directed against the mature peptide of TGF-b2 (from the published cDNA sequences for TGF-b isoforms, the approximate predicted sizes of the various proteins are 45 kDa for precursor protein, 28-30 kDa for LAP and 15 kDa for mature peptide; Figure 1, upper panel). It should be noted that, in the cleaved mature form, the TGF-b molecule is very difficult to study, and there are reports that it binds poorly
FIG. 6. Relative quantities of full length precursor and LAP for TGF-b2 . Data obtained from Western blots similar to that presented in Figure 3 (upper panel) were quantified by densitometric scanning, from a total of three separate experiments. The relative contributions of LAP and full length precursor were estimated, and the means{ S.E.M. are plotted.
926
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC
Vol. 245, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
to nitrocellulose membranes (26) and is rapidly adsorbed to plastic surfaces in solution (27). From the deduced amino acid sequences, all three TGF-b isoforms are predicted to have one or more potential N-linked glycosylation sites in the LAP region (9). TGF-b1 has been demonstrated to be highly glycosylated, with a significant mobility shift in SDS-PAGE (28). In the present study, TGF-b2 precursor protein and LAP and TGF-b3 LAP migrate close to their predicted molecular sizes in SDS-PAGE, suggesting that neither TGF-b2 nor TGF-b3 are as highly glycosylated in the rat heart as TGF-b1 . This is in agreement with over-expression studies with TGF-b2 in Chinese hamster ovary cells, where a relatively modest shift in mobility was observed (29). The primary observation in this study is the novel progressive shift in TGF-b2 expression from precursor protein to LAP, indicating an increase in the availability of activatable TGF-b2 with increasing age. Although no corresponding age-dependent shift from precursor to LAP was observed for TGF-b3 , there was a significant increase in the expression of LAP over the developmental period investigated, indicating that availability of activatable TGF-b3 is also increased with increasing age. The scarcity of TGF-b3 precursor in ventricular tissue suggests that, relative to TGF-b2 , TGF-b3 synthesis, export and cleavage to LAP and mature protein is quite rapid. This raises the question of whether TGFb2 and -b3 processing are regulated differentially, possibly involving intracellular storage of TGF-b2 precursor protein prior to cleavage and export. The present study does not allow delineation of the cellular source (myocytes, fibroblasts, endothelial cells and/or smooth muscle cells) of TGF-b2 and -b3 in ventricular myocardium. Such delineation may be achieved in future studies through the application of the antibodies used here in conjunction with an immunohistochemical approach. Regardless of the identity of their cellular source(s), however, the marked increases in activatable TGF-b2 and -b3 that have been observed in rat myocardium during the first 28 days after birth suggest that these growth factors may play an important role in post-natal cardiac development.
3. 4. 5. 6. 7. 8.
9.
10.
11. 12. 13. 14. 15. 16.
17. 18. 19. 20. 21. 22. 23. 24.
ACKNOWLEDGMENTS
25.
This work was supported by a grant awarded to P.C. and M.A. from the British Heart Foundation (BHF). G.B. was the recipient of a BHF Intermediate Fellowship and M.A. is a BHF (Basic Science) Senior Lecturer.
26. 27. 28.
REFERENCES 1. Massague, J. (1990) Ann. Rev. Cell Biol. 6, 597–641. 2. Roberts, A. B., Anzano, M. A., Wakefield, L. M., Roche, N. S.,
29.
Stern, D. F., and Sporn, M. B. (1985) Proc. Natl. Acad. Sci. 82, 119–123. Cheifetz, S., Hernandez, H., Laiho, M., ten Dijke, P., Iwata, K. K., and Massague, J. (1990) J.Biol. Chem. 265, 20533–20538. Graycar, J. L., Miller, D. A., Arrick, B. A., Lyons, R. M., Moses, H. L., and Derynck, R. (1989) Mol. Endocrinol. 3, 1977–1986. Rosa, F., Roberts, A. B., Danielpour, D., Dart, L. L., Sporn, M. B., and David, I. B. (1988) Science 236, 783–786. Jennings, J. C., Mohan, S., Linkhart, T. A., Widstorm, R., and Baylink, D. J. (1988) J. Cell Physiol. 137, 167–172. Letterio, J. J., Geiser, A. G., Kulkarni, A. B., Roche, N. S., Sporn, M. B., and Roberts, A. B. (1994) Science 264, 1936–1938. Derynck, R., Jarrett, J. A., Chen, E. Y., Eaton, D. H., Bell, J. R., Assoian, R. K., Roberts, A. B., Sporn, M. B., and Goeddel, D. V. (1985) Nature 316, 701–705. Derynck, R., Lindquist, P. B., Lee, A., Wen, D., Tamm, J., Graycar, J. L., Rhee, L., Mason, A. L., Miller, D. A., Coffey, R. J., Moses, H. L., and Chen, E. Y. (1988) EMBO J. 7, 3737–3743. de Martin, R., Haendler, B., Hofer-Warbinek, R., Gaugitsch, H., Wrann, M., Schlesener, J., Seifert, J. M., Bodmer, S., Fontana, A., and Hofer, E. (1987) EMBO J. 6, 3673–3677. Gentry, L. E., Lioubin, M. N., Purchio, A. F., and Marquardt, H. (1988) Mol. Cell Biol. 8, 4162–4168. Lyons, R. M., Gentry, L. E., Purchio, A. F., and Moses, H. L. (1990) J. Cell Biol. 110, 1361–1367. Akhurst, R. J., Lehnert, S. A., Faissner, A., and Duffie, E. (1990) Development 108, 645–656. Dickson, M. C., Slager, H. G., Duffie, E., Mummery, C. L., and Akhurst, R. J. (1993) Development 117, 625–639. Roberts, A. B., Roche, N. S., Winokur, T. S., Burmester, J. K., and Sporn, M. B. (1992) J. Clin. Invest. 90, 2056–2062. Waltenberger, J., Lundin, L., Oberg, K., Wilander, E., Miyazono, K., Heldin, C.-H., and Funa, K. (1993) Am. J. Pathol. 142, 71– 78. Villarreal, F. J., and Dillman, W. H. (1992) Am. J. Physiol. 262, H1861–1866. McKinnon, R. D., Piras, G., Ida, J. A., and Dubois-Dalcq, M. (1993) J. Cell Biol. 121, 1397–1407. Cummins, B., Auckland, M. L., and Cummins, P. (1987) Amer. Heart J. 113, 1333–1344. Bradford, M. M. (1976) Anal. Biochem. 72, 248–254. Laemmli, U. K. (1970) Nature 227, 680–685. Olofsson, A., Miyazono, K., Kanzaki, T., Colosetti, P., Engstrom, U., and Heldin, C.-H. (1992) J. Biol. Chem. 267, 19482–19488. Waltenberger, J., Wanders, A., Fellstrom, B., Miyazono, K., Heldin, C.-H., and Funa, K. (1993) J. Immunol. 151, 1147–1157. Li, J. M., and Brooks, G. (1997) J. Mol. Cell. Cardiol. 29, 2213– 2225. Engelmann, G. L., Boehm, K. D., Birchenall-Roberts, M. C., and Ruscetti, F. W. (1992) Mech. Develop. 38, 85–98. Potts, J. D., and Runyan, R. B. (1989) Develop. Biol. 134, 392– 401. Brown, P. D., Wakefield, L. M., Levinson, A. D., and Sporn, M. D. (1990) Growth Factors 3, 35–43. Purchio, A. F., A., C. J., Brunner, A. M., Lioubin, M. N., Gentry, L. E., Kovacina, K. S., Roth, R. A., and Marquardt, H. (1988) J. Biol. Chem. 263, 14211–14215. Madisen, L., Lioubin, M. N., Marquardt, H., and Purchio, A. F. (1990) Growth Factors 3, 129–138.
927
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC
245, 923–927 (1998)
RC988528
Differential Changes in Transforming Growth Factor-b Isoform Expression during Postnatal Cardiac Growth Robert S. Haworth,*,† Gavin Brooks,* Peter Cummins,† Karen Dobie,† Derrick C. Chilton,† and Metin Avkiran* *Cardiovascular Research, The Rayne Institute, St. Thomas’ Hospital, London SE1 7EH; and †Department of Physiology, University of Birmingham, Birmingham, United Kingdom
Received March 23, 1998
Transforming growth factor-b (TGF-b) is synthesised as an inactive precursor protein; this is cleaved to produce the mature peptide and a latency associated protein (LAP), which remains associated with the mature peptide until activation by LAP degradation. Isoform specific antibodies raised against the LAPs for TGF-b2 and -b3 were used to determine the myocardial levels of LAP (activatable TGF-b) and full length precursor (inactive TGF-b) forms during post-natal development in the rat. TGF-b2 was present predominantly as the precursor in 2 day old myocardium. There was an age-dependent shift from precursor protein to LAP between 2 and 28 days. A corresponding increase in the level of mature (activatable) TGF-b2 was found. TGFb3 was detected in significant quantities only as LAP. However, a four-fold increase in the expression of TGF-b3 LAP was observed between 2 and 28 days. The substantial increases in activatable forms of TGF-b2 and -b3 that occur in myocardium during the first 28 days of life in the rat support a role for these proteins in post-natal cardiac development. q 1998 Academic Press
The TGF-b family belongs to a superfamily of related proteins which affect cell growth, differentiation and extracellular matrix protein (ECM) production (reviewed by Massague (1)). Members of the TGF-b family are found in a wide variety of species and tissues (1), although the specific biological functions of TGF-b isoforms depend upon the cell type and, in vitro, culture conditions (2). In most biochemical assays, the three mammalian isoforms of TGF-b appear to have interchangeable function (3, 4). However, a number of differential effects have been observed, particularly on inhibition of DNA replication in several cell types (3, 5-7). Three human forms of TGF-b (TGF-b1 , -b2 and -b3) have been cloned and sequenced (8–10). These proteins are synthesized as inactive precursor proteins consisting of either 391 (-b1), 415 (-b2) or 410 (-b3) amino
acids (9) (Figure 1, upper panel). At some point during the secretion of TGF-b, homodimers are formed between two mature proteins (each consisting of the carboxyl-terminal 112 amino acids of the full length precursor; Figure 1, lower panel), and this mature dimer is cleaved proteolytically from the rest of the precursor protein (known as latency associated protein; LAP). LAP remains associated with the mature dimer to form an inactive complex (11), which is associated with the ECM. Active TGF-b in the form of the mature homodimer can be released from the ECM by protease-mediated degradation of LAP (12). A number of reports have suggested that TGF-b plays an important role in cardiac development (13, 14), maintenance of myocyte function (15) and formation of scar tissue in heart disease (16, 17). However, there is little information in the literature regarding the post-natal expression of the precursor, LAP and mature forms of TGF-b isoforms in the rat heart. Here we describe the use of isoform-specific antibodies directed against both the LAP and mature forms of TGFb2 and the LAP form of TGF-b3 to determine post-natal changes in TGF-b precursor, LAP and mature protein expression in the rat ventricular myocardium. METHODS Animals and materials. Adult male and time-mated late term pregnant Wistar rats were purchased from B & K Universal (Hull, U.K). All animals were killed by an approved method, in accordance with the U.K. Home Office Animals (Scientific Procedures) Act, 1986. Antibody directed against mature human TGF-b2 (sc-090) was purchased from Santa Cruz Biotechnology, Inc. (California, USA). Protein A Sepharose was from Pharmacia Biotech (St. Albans, U.K.), and synthetic peptides were a kind gift from Dr. Johannes Waltenberger (Ludwig Institute for Cancer Research, Uppsala, Sweden). All other chemicals were of the purest grade commercially available. Antibody production and purification. Polyclonal rabbit antisera were raised against synthetic peptides corresponding to variable regions in each LAP immediately preceding the mature human protein sequence (TGF-b2 aa 285-300 (10), sequence LLPSYRLESQQTNRRK; TGF-b3 aa 282-298 (9), sequence MIPPHRLDNPGQ-
923
0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC
Vol. 245, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS In the case of ventricular tissue from very young rats, hearts were pooled from more than one animal.
FIG. 1. Diagrammatic illustration of TGF-b precursor protein showing the location of the peptides used to raise anti-LAP and antimature TGF-b antibodies (upper panel) and (lower panel) TGF-b dimerisation, cleavage of mature peptide and LAP (a) and release of mature peptide by LAP degradation (b).
SDS–PAGE and Western blotting. Proteins were separated by size under reducing conditions using SDS-PAGE. Gels containing 12 % (w/v) acrylamide were prepared essentially as described by Laemmli (21). Protein samples (50-100 mg/lane) were solubilised in Laemmli sample buffer and run at 100 V/gel. Proteins were transferred to PVDF membrane (Amersham, Bucks, U.K.), using a Pharmacia LKB Multiphor II transfer apparatus. After blocking overnight in 10 % (w/v) skimmed milk powder in PBS, membranes were incubated with antibodies (concentration range of 3-10 mg/ml) in 3 % (w/v) skimmed milk powder/PBS for 2 hours, washed twice for 5 min with PBS containing polyoxyethylenesorbitan monolaurate (Tween 20; 0.05 % (v/v)) and once with PBS. Donkey anti-rabbit horseradish peroxidase-conjugated antibody (Amersham) was used at a dilution of 1:2000 to label bound antibody, and the antibody complex was detected using the Amersham Enhanced Chemiluminescence (ECL) system, as recommended by the manufacturer. For quantitation, bands were scanned using an LKB Ultrascan XL Laser Densitometer (Milton Keynes, U.K). Statistics. Data are expressed as mean{SEM. Statistical significance was assessed by repeated measures ANOVA, followed by Dunnett’s test. A value of Põ 0.05 was considered to be significant.
RESULTS GGQRK), as described by Waltenberger et al. (16). In the rat, the sequence of the corresponding region is known only for TGF-b2 (18) (sequence LLPSYRLESQQSSRRK), with rat TGF-b2 exhibiting only two amino acid differences from the human sequence. Polyclonal antisera were raised in New Zealand rabbits using the method of Cummins et al. (19). The isoform specificity of the antisera raised against each of the TGF-b peptides was confirmed by enzymelinked immunoassay using peptide coated plates (data not shown). Immunoglobulin was purified from the serum by ammonium sulphate precipitation followed by protein A-Sepharose adsorption. Briefly, serum (1 ml) was mixed with an equal volume of saturated ammonium sulphate in phosphate buffered saline (PBS) and incubated at 47C for 5 min with constant mixing. The precipitated samples were sedimented by centrifugation at 47C for 30 min at 6500 1 g. The pellet was dissolved in 1 ml of PBS, and dialysed overnight against PBS at 47C. The partially purified serum samples were incubated in the presence of 80 mg protein A Sepharose (pre-swollen for 30 min in PBS) for 30 min at 47C. The protein A Sepharose/ immunoglobulin complexes were isolated by centrifugation, and the antibodies released by the addition of guanidine HCl to a final concentration of 1 M. After mixing for 20 min, the samples were centrifuged at 15000 1 g, and the supernatant dialysed against PBS as described above. Tissue preparation. In a preliminary study to look at overall changes in myocardial TGF-b expression from the foetal to the adult stages, ventricular samples were taken from rats at 18 day gestation (foetal), 2 day post-natal (neonatal) and 42 days post-natal (adult). To follow changes in greater detail during post-natal development, ventricular tissue was obtained from rats aged 2, 5, 7, 10, 14, 21, 28, and 42 days. All samples were frozen in liquid nitrogen immediately after excision, ground to powder using a pestle and mortar pre-cooled with liquid nitrogen, and stored in liquid nitrogen until required. Protein was prepared from 100-300 mg of powdered tissue by mechanical homogenisation on ice in a buffer containing sucrose (250 mM), Tris (25 mM, pH 7.5), EDTA (1 mM), EGTA (1 mM), bmercaptoethanol (0.1 %), leupeptin (50 mM), benzamidine (1 mM) and PMSF (1 mM). Following homogenisation, samples were centrifuged at 10,000 rpm for 10 min at 47C. The supernatant was recovered, and the pellet subjected to a second round of homogenisation and centrifugation. Protein concentration was determined by the method of Bradford (20), using bovine serum albumin as standard.
The TGF-b2 and TGF-b3 LAP antibodies were raised against isoform-specific peptides; such antibodies have been characterised and used previously by Olofsson et al. (22) and Waltenberger et al. (16, 23), as well as in this laboratory in an earlier study (24). Age-related changes in the expression of TGF-b isoforms were initially investigated using ventricular samples obtained from animals killed at 18 days gestation (foetal), 2 days after birth (neonatal) and 42 days after birth (adult). A representative Western blot is presented in Figure 2, using antibodies directed against TGF-b2 mature peptide and TGF-b3 LAP. TGFb2 protein was detected in ventricular tissue (Figure 2, upper panel) in precursor form at a molecular weight of 45 kDa. There was a clear age-dependent change in the quantity of precursor protein, with more protein present in foetal tissue. Precursor protein was undetectable in adult tissue, although lower molecular weight species were also difficult to detect. TGF-b2 immunising peptide competed successfully for antibody (data not shown). To detect TGF-b3 proteins in rat ventricular tissue, antibody raised against TGF-b3 LAP peptide was used (Figure 2, lower panel). TGF-b3 precursor protein was not detected in any ventricular tissue sample tested. However, TGF-b3 LAP with a molecular weight of 28 kDa was detected in foetal, neonatal and adult tissue. A substantial age-dependent increase in the expression of TGF-b3 LAP was observed. To investigate these changes in more detail, the expression of TGF-b isoforms during post-natal development was investigated using ventricular samples obtained from animals aged from 2 to 28 days. Using antibody directed at the LAP region, TGF-b2 was de-
924
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC
Vol. 245, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Detection of TGF-b in foetal (F; 18 days gestation), neonatal (N; 2 days after birth) and adult (A; 42 days after birth) rat ventricular tissue. Ventricular samples from a total of 8-15 foetal or neonatal animals were pooled. Protein samples (100 mg) were separated by 12% SDS-PAGE and transferred to PVDF membranes as described in Materials and Methods. Western blots were probed for the presence of TGF-b2 using mature peptide antibody (upper panel) and TGF-b3 using LAP antibody (lower panel). Approximate molecular sizes were estimated from Rainbow molecular weight markers (Amersham). The Western blot presented here is representative of three similar experiments using tissue derived from different groups of animals.
tectable not only as full length precursor (45 kDa) but also in LAP form (30 kDa; Figure 3, upper panel). Consistent with earlier findings with mature peptide antibody (Figure 2), there was significant expression of TGF-b2 precursor protein in 2 day old myocardial tissue, with very little LAP expression. Again consistent
FIG. 3. Detection of TGF-b2 and TGF-b3 in rat ventricular tissue during post-natal development (2-28 days after birth). Ventricular tissue from a total of 3-10 animals was pooled in each age group. Protein samples (100 mg) were separated by 12% SDS-PAGE and transferred to PVDF membranes as described in Materials and Methods. Western blots were probed for the presence of TGF-b2 (upper panel) and TGF-b3 (lower panel) using isoform-specific LAP antibodies. Approximate molecular sizes were estimated from prestained molecular weight markers (Bio-Rad). The Western blots presented here are representative of at least three experiments using tissue derived from different groups of animals.
FIG. 4. Detection of TGF-b2 in rat ventricular tissue during postnatal development (2-21 days after birth). Ventricular tissue from a total of 3-10 animals was pooled in each age group. Protein samples (40 mg) were separated by 12% SDS-PAGE and transferred to PVDF membranes as described in Materials and Methods. Western blots were probed for the presence of TGF-b2 using mature peptide antibody. Approximate molecular sizes were estimated from pre-stained molecular weight markers (Bio-Rad). The Western blot presented here is representative of three similar experiments using tissue derived from different groups of animals.
with earlier findings, the quantity of TGF-b2 precursor protein decreased with increasing age, but this time we were also able to demonstrate a concomitant rise in TGF-b2 LAP. The change was very marked between 7 and 14 days after birth, and from 21 days of age, TGF-b2 was present predominantly as LAP. In contrast to the observations with TGF-b2 , experiments using TGF-b3 LAP antibody demonstrated that this isoform was detectable almost exclusively as LAP throughout the time period studied (Figure 3, lower panel) with only minor expression of the 45 kDa precursor protein at 21 and 28 days. Moreover, the quantity of TGF-b3 LAP showed an age-dependent increase from 2 days of age to 28 days. Commercial antibody directed against TGF-b2 mature peptide was used to confirm the findings with the TGF-b2 LAP antibody (Figure 4). Both precursor (45 kDa) and mature peptide (12 kDa) were detected. An age-dependent shift from precursor form to mature peptide was apparent, consistent with the earlier findings. Detection of the mature 12 kDa TGF-b2 peptide by Western blotting required long exposure times, resulting in over-exposure of precursor protein bands and the appearance of additional bands of intermediate molecular weight. All bands disappeared when immunising peptide was present in excess during incubation with primary antibody (data not shown). The relative expression of total TGF-b protein (LAP plus precursor proteins) was estimated by densitometric scanning of immunoreactive species in blots such as that shown in Figure 3. No significant differences in the total quantity of TGF-b2 was observed during the time period studied (Figure 5, upper panel). However, TGF-b3 showed a significant (Põ0.05) four-fold increase in expression over the same period (22.5 { 3.8
925
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC
Vol. 245, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
DISCUSSION
FIG. 5. Relative quantities of TGF-b isoforms in neonatal rat ventricular tissue. Data obtained from triplicate Western blots similar to those presented in Figure 3 were quantified by densitometric scanning, from a total of three separate experiments at each age. Total TGF-b (LAP plus precursor) was determined for each isoform. The quantities at each age were normalised relative to the values obtained at 28 days. The means{S.E.M. are plotted for TGF-b2 (upper panel) and TGF-b3 (lower panel). Significant differences were observed only for TGF-b3 at 21 and 28 days (* Põ0.05 vs 2 day values).
v 91.5 { 8.5 arbitary units for 2 day and 28 day, respectively; Figure 5, lower panel). Although no significant change in the total amount (LAP plus precursor) of TGF-b2 was observed during postnatal myocardial growth, this isoform was found to undergo a novel transition between the precursor protein and LAP. Figure 6 shows the relative contribution of LAP and precursor to total immunoreactive species. TGF-b2 exhibited a precursor to LAP transition over the period studied, with equal quantities of LAP and precursor at 7-10 days after birth. Corresponding estimations from Western blots with the mature TGFb2 antibody were not attempted, as the long exposure times required for detection of mature protein resulted in over-exposure of the precursor protein bands in the 2-14 day samples.
Very little is known about the function of TGF-b isoforms in the neonatal heart, although Engelmann et al. (25) have suggested recently that TGF-b1 may be important in regulating differentiation and proliferation of the cardiac myocyte. In almost all cases where TGF-b isoform expression has been investigated in the heart using specific antibodies, immunohistochemical methods have been employed, generally using antibodies raised against the mature part of the protein. However, such antibodies would react with both the mature peptide and full length precursor, and immunohistochemical studies provide no indication of the molecular size of the protein present. To date, there has been no indication of the relative contributions of full length precursor and LAP to the total TGF-b signal observed. This is important since precursor protein would be expected to be inactive and probably localised intracellularly at the site of synthesis, whereas LAP would be predicted to be extracellular (in a latent complex with mature TGF-b) and available for enzymatic activation. In order to determine the levels of TGF-b isoform precursor and LAP in the post-natal rat heart, we used antibodies directed either against heterogeneous regions of LAP from human TGF-b2 and -b3 sequences, or commercial antibody directed against the mature peptide of TGF-b2 (from the published cDNA sequences for TGF-b isoforms, the approximate predicted sizes of the various proteins are 45 kDa for precursor protein, 28-30 kDa for LAP and 15 kDa for mature peptide; Figure 1, upper panel). It should be noted that, in the cleaved mature form, the TGF-b molecule is very difficult to study, and there are reports that it binds poorly
FIG. 6. Relative quantities of full length precursor and LAP for TGF-b2 . Data obtained from Western blots similar to that presented in Figure 3 (upper panel) were quantified by densitometric scanning, from a total of three separate experiments. The relative contributions of LAP and full length precursor were estimated, and the means{ S.E.M. are plotted.
926
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC
Vol. 245, No. 3, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
to nitrocellulose membranes (26) and is rapidly adsorbed to plastic surfaces in solution (27). From the deduced amino acid sequences, all three TGF-b isoforms are predicted to have one or more potential N-linked glycosylation sites in the LAP region (9). TGF-b1 has been demonstrated to be highly glycosylated, with a significant mobility shift in SDS-PAGE (28). In the present study, TGF-b2 precursor protein and LAP and TGF-b3 LAP migrate close to their predicted molecular sizes in SDS-PAGE, suggesting that neither TGF-b2 nor TGF-b3 are as highly glycosylated in the rat heart as TGF-b1 . This is in agreement with over-expression studies with TGF-b2 in Chinese hamster ovary cells, where a relatively modest shift in mobility was observed (29). The primary observation in this study is the novel progressive shift in TGF-b2 expression from precursor protein to LAP, indicating an increase in the availability of activatable TGF-b2 with increasing age. Although no corresponding age-dependent shift from precursor to LAP was observed for TGF-b3 , there was a significant increase in the expression of LAP over the developmental period investigated, indicating that availability of activatable TGF-b3 is also increased with increasing age. The scarcity of TGF-b3 precursor in ventricular tissue suggests that, relative to TGF-b2 , TGF-b3 synthesis, export and cleavage to LAP and mature protein is quite rapid. This raises the question of whether TGFb2 and -b3 processing are regulated differentially, possibly involving intracellular storage of TGF-b2 precursor protein prior to cleavage and export. The present study does not allow delineation of the cellular source (myocytes, fibroblasts, endothelial cells and/or smooth muscle cells) of TGF-b2 and -b3 in ventricular myocardium. Such delineation may be achieved in future studies through the application of the antibodies used here in conjunction with an immunohistochemical approach. Regardless of the identity of their cellular source(s), however, the marked increases in activatable TGF-b2 and -b3 that have been observed in rat myocardium during the first 28 days after birth suggest that these growth factors may play an important role in post-natal cardiac development.
3. 4. 5. 6. 7. 8.
9.
10.
11. 12. 13. 14. 15. 16.
17. 18. 19. 20. 21. 22. 23. 24.
ACKNOWLEDGMENTS
25.
This work was supported by a grant awarded to P.C. and M.A. from the British Heart Foundation (BHF). G.B. was the recipient of a BHF Intermediate Fellowship and M.A. is a BHF (Basic Science) Senior Lecturer.
26. 27. 28.
REFERENCES 1. Massague, J. (1990) Ann. Rev. Cell Biol. 6, 597–641. 2. Roberts, A. B., Anzano, M. A., Wakefield, L. M., Roche, N. S.,
29.
Stern, D. F., and Sporn, M. B. (1985) Proc. Natl. Acad. Sci. 82, 119–123. Cheifetz, S., Hernandez, H., Laiho, M., ten Dijke, P., Iwata, K. K., and Massague, J. (1990) J.Biol. Chem. 265, 20533–20538. Graycar, J. L., Miller, D. A., Arrick, B. A., Lyons, R. M., Moses, H. L., and Derynck, R. (1989) Mol. Endocrinol. 3, 1977–1986. Rosa, F., Roberts, A. B., Danielpour, D., Dart, L. L., Sporn, M. B., and David, I. B. (1988) Science 236, 783–786. Jennings, J. C., Mohan, S., Linkhart, T. A., Widstorm, R., and Baylink, D. J. (1988) J. Cell Physiol. 137, 167–172. Letterio, J. J., Geiser, A. G., Kulkarni, A. B., Roche, N. S., Sporn, M. B., and Roberts, A. B. (1994) Science 264, 1936–1938. Derynck, R., Jarrett, J. A., Chen, E. Y., Eaton, D. H., Bell, J. R., Assoian, R. K., Roberts, A. B., Sporn, M. B., and Goeddel, D. V. (1985) Nature 316, 701–705. Derynck, R., Lindquist, P. B., Lee, A., Wen, D., Tamm, J., Graycar, J. L., Rhee, L., Mason, A. L., Miller, D. A., Coffey, R. J., Moses, H. L., and Chen, E. Y. (1988) EMBO J. 7, 3737–3743. de Martin, R., Haendler, B., Hofer-Warbinek, R., Gaugitsch, H., Wrann, M., Schlesener, J., Seifert, J. M., Bodmer, S., Fontana, A., and Hofer, E. (1987) EMBO J. 6, 3673–3677. Gentry, L. E., Lioubin, M. N., Purchio, A. F., and Marquardt, H. (1988) Mol. Cell Biol. 8, 4162–4168. Lyons, R. M., Gentry, L. E., Purchio, A. F., and Moses, H. L. (1990) J. Cell Biol. 110, 1361–1367. Akhurst, R. J., Lehnert, S. A., Faissner, A., and Duffie, E. (1990) Development 108, 645–656. Dickson, M. C., Slager, H. G., Duffie, E., Mummery, C. L., and Akhurst, R. J. (1993) Development 117, 625–639. Roberts, A. B., Roche, N. S., Winokur, T. S., Burmester, J. K., and Sporn, M. B. (1992) J. Clin. Invest. 90, 2056–2062. Waltenberger, J., Lundin, L., Oberg, K., Wilander, E., Miyazono, K., Heldin, C.-H., and Funa, K. (1993) Am. J. Pathol. 142, 71– 78. Villarreal, F. J., and Dillman, W. H. (1992) Am. J. Physiol. 262, H1861–1866. McKinnon, R. D., Piras, G., Ida, J. A., and Dubois-Dalcq, M. (1993) J. Cell Biol. 121, 1397–1407. Cummins, B., Auckland, M. L., and Cummins, P. (1987) Amer. Heart J. 113, 1333–1344. Bradford, M. M. (1976) Anal. Biochem. 72, 248–254. Laemmli, U. K. (1970) Nature 227, 680–685. Olofsson, A., Miyazono, K., Kanzaki, T., Colosetti, P., Engstrom, U., and Heldin, C.-H. (1992) J. Biol. Chem. 267, 19482–19488. Waltenberger, J., Wanders, A., Fellstrom, B., Miyazono, K., Heldin, C.-H., and Funa, K. (1993) J. Immunol. 151, 1147–1157. Li, J. M., and Brooks, G. (1997) J. Mol. Cell. Cardiol. 29, 2213– 2225. Engelmann, G. L., Boehm, K. D., Birchenall-Roberts, M. C., and Ruscetti, F. W. (1992) Mech. Develop. 38, 85–98. Potts, J. D., and Runyan, R. B. (1989) Develop. Biol. 134, 392– 401. Brown, P. D., Wakefield, L. M., Levinson, A. D., and Sporn, M. D. (1990) Growth Factors 3, 35–43. Purchio, A. F., A., C. J., Brunner, A. M., Lioubin, M. N., Gentry, L. E., Kovacina, K. S., Roth, R. A., and Marquardt, H. (1988) J. Biol. Chem. 263, 14211–14215. Madisen, L., Lioubin, M. N., Marquardt, H., and Purchio, A. F. (1990) Growth Factors 3, 129–138.
927
AID
BBRC 8528
/
6950$$1001
04-11-98 10:50:55
bbrcg
AP: BBRC