Mutational Analysis of the Insulin-like Growth Factor I Prohormone Processing Site

Insulin-like growth factor I (IGF-I) is a mitogenic peptide that is produced in most tissues and cell lines and plays an important role in embryonic development and postnatal growth. IGF-I is initially synthesized as a prohormone precursor that is converted to mature IGF-I by endoproteolytic removal of the carboxyl-terminal E-domain. Regulation of the conversion of proIGF-I to mature IGF-I is a potential mechanism by which the biological activity of this growth factor might be modulated. Endoproteolysis of the IGF-I prohormone occurs at the unique pentabasic motif Lys-X-X-Lys-X-X-Arg71-X-X-Arg-X-X-Arg. Recently, a family of enzymes which cleave prohormone precursors at sites containing multiple basic residues has been discovered. The goals of this study were 1) to determine which basic residues in the pentabasic proIGF-I processing site were necessary for proper cleavage and 2) to examine the role that subtilisin-related proprotein convertase 1 (SPC1/furin) might play in proIGF-I processing. We have shown that an expression vector coding for an epitope-tagged proIGF-I directs synthesis and secretion of mature IGF-I-(1-70), extended IGF-I-(1-76), proIGF-I, and N-glycosylated proIGF-I in human embryonic kidney 293 cells. Extended IGF-I-(1-76) is produced by cleavage at Arg77 and requires both Arg74 (P4) and Arg77 (P1). Cleavage at Arg77 does not occur in the SPC1-deficient cell lines RPE.40 and LoVo, suggesting that processing at this site is mediated by SPC1. Mature IGF-I-(1-70) is produced by cleavage at Arg71 and requires both Lys68 (P4) and Arg71 (P1). Lys65 in the P7 position is important for efficient cleavage. SPC1 is not required for processing at Arg71 since this cleavage occurs in RPE.40 and LoVo cells. These data suggest the existence of a processing enzyme which is specific for the Lys-X-X-Arg motif of proIGF-I. Insulin-like growth factor I (IGF-I) is a mitogenic peptide that is produced in most tissues and cell lines and plays an important role in embryonic development and postnatal growth. IGF-I is initially synthesized as a prohormone precursor that is converted to mature IGF-I by endoproteolytic removal of the carboxyl-terminal E-domain. Regulation of the conversion of proIGF-I to mature IGF-I is a potential mechanism by which the biological activity of this growth factor might be modulated. Endoproteolysis of the IGF-I prohormone occurs at the unique pentabasic motif Lys-X-X-Lys-X-X-Arg71-X-X-Arg-X-X-Arg. Recently, a family of enzymes which cleave prohormone precursors at sites containing multiple basic residues has been discovered. The goals of this study were 1) to determine which basic residues in the pentabasic proIGF-I processing site were necessary for proper cleavage and 2) to examine the role that subtilisin-related proprotein convertase 1 (SPC1/furin) might play in proIGF-I processing. We have shown that an expression vector coding for an epitope-tagged proIGF-I directs synthesis and secretion of mature IGF-I-(1-70), extended IGF-I-(1-76), proIGF-I, and N-glycosylated proIGF-I in human embryonic kidney 293 cells. Extended IGF-I-(1-76) is produced by cleavage at Arg77 and requires both Arg74 (P4) and Arg77 (P1). Cleavage at Arg77 does not occur in the SPC1-deficient cell lines RPE.40 and LoVo, suggesting that processing at this site is mediated by SPC1. Mature IGF-I-(1-70) is produced by cleavage at Arg71 and requires both Lys68 (P4) and Arg71 (P1). Lys65 in the P7 position is important for efficient cleavage. SPC1 is not required for processing at Arg71 since this cleavage occurs in RPE.40 and LoVo cells. These data suggest the existence of a processing enzyme which is specific for the Lys-X-X-Arg motif of proIGF-I. Insulin-like growth factor I (IGF-I)1 1The abbreviations used are: IGF-Iinsulin-like growth factor ISPCsubtilisin-related proprotein convertasePCRpolymerase chain reactionCMVcytomegalovirusDMEMDulbecco's modified Eagle's mediumTricineN-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine. circulates in human serum as a 70-amino-acid peptide consisting of four domains: B, C, A, and D. The physiological importance of IGF-I for normal development and growth is underscored by the severity of the phenotype of mice lacking a functional IGF-I gene. Mice carrying a disrupted IGF-I gene display growth deficiencies, delayed bone development, infertility, and a high mortality rate(1Liu J.-P. Baker J. Perkins A.S. Robertson E.J. Efstratiadis A. Cell. 1993; 75: 59-72Abstract Full Text PDF PubMed Scopus (2595) Google Scholar, 2Baker J. Liu J.-P. Robertson E.J. Efstratiadis A. Cell. 1993; 75: 73-82Abstract Full Text PDF PubMed Scopus (2072) Google Scholar). Liver is the main site of production of circulating IGF-I, although this growth factor is also synthesized and secreted by most tissues. Mature IGF-I can be derived from either of two IGF-I prohormones by removal of the E-domain (see Fig. 2). Alternative splicing of IGF-I mRNA is responsible for generating the two IGF-I prohormones, and the physiological significance of the different forms is unknown(3Daughaday W.H. Rotwein P. Endocr. Rev. 1989; 10: 68-91Crossref PubMed Scopus (1624) Google Scholar). ProIGF-IA contains a 35-amino-acid E-domain, while the E-domain of proIGF-IB is 77 amino acids in length. The sequences of human proIGF-IA and proIGF-IB are identical through the first 16 residues of the E-domain(4Rotwein P. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 77-81Crossref PubMed Scopus (222) Google Scholar), including the unique pentabasic prohormone cleavage motif Lys-X-X-Lys-X-X-Arg71-X-X-Arg-X-X-Arg (Fig. 1). This motif has been conserved in mammals, birds, amphibians, and teleosts.Figure 1:The IGF-I prohormone processing site. The amino acid sequence of the proIGF-I processing site is shown in single-letter code. Basic residues are indicated by boldface type and numbered. The scissile bond connecting the P1 Arg71 and the P1’ Ser72 is indicated by the vertical arrow.View Large Image Figure ViewerDownload Hi-res image Download (PPT) insulin-like growth factor I subtilisin-related proprotein convertase polymerase chain reaction cytomegalovirus Dulbecco's modified Eagle's medium N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine. It is now well established that most peptide hormones and growth factors are initially synthesized as biologically inactive precursors that are converted to active forms by endoproteolysis at specific sites. Proinsulin has served as a model for studies of prohormone processing. From these studies it has been determined that the C-peptide of proinsulin is excised by proteolytic cleavage at paired dibasic residues to produce insulin(5Steiner D.F. Cuatrecasas P. Jacobs S. Handbook of Experimental Pharmacology. Vol. 92. Springer-Verlag, New York1990: 67-92Google Scholar, 6Steiner D.F. Smeekens S.P. Ohagi S. Chan S.J. J. Biol. Chem. 1992; 267: 23435-23438Abstract Full Text PDF PubMed Google Scholar). Cleavage is mediated by two subtilisin-related proprotein converting (SPC) enzymes, SPC2 (PC2) and SPC3 (PC1/3)(7Davidson H.W. Rhodes C.J. Hutton J.C. Nature. 1988; 333: 93-96Crossref PubMed Scopus (301) Google Scholar, 8Smeekens S.P. Montag A.G. Thomas G. Albiges-Rizo C. Carroll R. Benig M. Phillips L.A. Martin S. Ohagi S. Gardner P. Swift H.H. Steiner D.F. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 8822-8826Crossref PubMed Scopus (260) Google Scholar). These enzymes are serine proteases and have been shown to process several prohormones to mature hormones, including proglucagon (9Rouillé Y. Westermark G. Martin S.K. Steiner D.F. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 3242-3246Crossref PubMed Scopus (179) Google Scholar) and the proopiomelanocortin precursor(10Thomas L. Leduc R. Thorne B.A. Smeekens S.P. Steiner D.F. Thomas G. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 5297-5301Crossref PubMed Scopus (276) Google Scholar, 11Benjannet S. Rondeau N. Day R. Chretien M. Seidah N.G. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 3564-3568Crossref PubMed Scopus (539) Google Scholar, 12Zhou A. Bloomquist B.T. Mains R.E. J. Biol. Chem. 1993; 268: 1763-1769Abstract Full Text PDF PubMed Google Scholar). Expression of SPC2 and SPC3 is limited to endocrine and neuroendocrine tissues, and the preferred cleavage site is carboxyl-terminal to Arg-Arg and Lys-Arg doublets(6Steiner D.F. Smeekens S.P. Ohagi S. Chan S.J. J. Biol. Chem. 1992; 267: 23435-23438Abstract Full Text PDF PubMed Google Scholar). Other members of the mammalian subtilisin-related proprotein convertase family include SPC1 (furin), SPC4 (PACE4), SPC5 (PC4), and SPC6 (PC5/6). SPC1 and SPC4 are widely distributed and cleave proprotein precursors at tri- and tetrabasic sites(13Halban P.A. Irminger J.-C. Biochem. J. 1994; 299: 1-18Crossref PubMed Scopus (283) Google Scholar). The preferred cleavage site for SPC1 appears to be Arg-X-Lys/Arg-Arg(14Hosaka M. Nagahama M. Kim W.-S. Watanabe T. Hatsuzawa K. Ikemizu J. Murakami K. Nakayama K. J. Biol. Chem. 1991; 266: 12127-12130Abstract Full Text PDF PubMed Google Scholar), and the Arg-X-X-Arg sequence may serve as a minimal processing site(15Molloy S.S. Bresnahan P.A. Leppla S.H. Klimpel K.R. Thomas G. J. Biol. Chem. 1992; 267: 16396-16402Abstract Full Text PDF PubMed Google Scholar). SPC4 has been reported to have a more strict requirement for a basic residue in the P2 position than does SPC1(16Rehemtulla A. Barr P.J. Rhodes C.J. Kaufman R.J. Biochemistry. 1993; 32: 11586-11590Crossref PubMed Scopus (57) Google Scholar, 17Creemers J.W.M. Kormelink P.J.G. Roebroek A.J.M. Nakayama K. Van de Ven W.J.M. FEBS Lett. 1993; 336: 65-69Crossref PubMed Scopus (40) Google Scholar). The paradigm established for processing of proinsulin suggests that proIGF-I would be converted to mature IGF-I by cleavage at the carboxyl terminus of Arg71, followed by removal of the basic residue by a carboxypeptidase. SPC1 is a candidate proIGF-I convertase since, like IGF-I, it is expressed ubiquitously and it does not have a strict requirement for a basic residue in the P2 position. We have used site-directed mutagenesis to determine the importance of each of the basic residues in the proIGF-I cleavage site for recognition by the processing enzyme. We have also examined the role of SPC1 in proIGF-I processing through the use of cell lines deficient in SPC1 activity. PreproIGF-IA was amplified by PCR from cDNA obtained from the human fibroblast cell line GM 03652C (ATCC). Primers specific for the Met−25 region of the signal peptide (hIGF1-5) and the translation termination codon region of the E-domain (hIGF1-6) were used: 5’-GGGAATTCTTGAAGGTGAAGATGCACAC and 5’-GGGGATCCCCTACATCCTGTAGTTCTTGT. The PCR product was cloned into the EcoRI and BamHI sites of the pCMV6c expression vector, which contains the CMV promotor and the SV40 poly(A) tail, to generate pCMVigf1. pCMVigf1 was then used as a template for recombinant PCR (18Higuchi R. Innis M.A. Gelfand D.H. Sninsky J.J. White T.J. PCR Protocols: A Guide to Methods and Applications. Academic Press, New York1990Google Scholar) to generate pCMVigf1-FLAG. The signal peptide of IGF-I was amplified using primers hIGF1-5 and Flg-2 (5’-CTTGTTCATCGTCGTCCTTGTAGTCAGCCGTGGCAGAGCTGGT). Flg-2 codes for Thr−6 to Ala−1 of the signal peptide and contains the 24-nucleotide antisense FLAG sequence on the 5’ end. The coding region of proIGF-I was amplified using primers hIGF1-6 and Flg-1 (5’-GACTACAAGGACGACGATGACAAGGGACCGGAGACGCTCTGC). Flg-1 codes for Gly1 to Cys6 of the B-domain and contains the 24-nucleotide sense FLAG sequence on the 5’ end. The PCR products were purified by polyacrylamide gel electrophoresis, and 50 ng of each was mixed, and extended PCR and amplified by of PCR hIGF1-5 and The recombinant PCR product was cloned into the EcoRI and BamHI sites of pCMV6c to generate pCMVigf1-FLAG. The of PCR was by were using the in and as sequences and are shown in were from to pCMV6c and by 1 in a embryonic kidney 293 were in serum and and 293 were by the R. R.E. K. in New Scholar). and RPE.40 by Thomas of were in medium serum and and and RPE.40 were using the were in and for in Dulbecco's modified Eagle's medium 1 serum and The medium was then Dulbecco's modified Eagle's medium containing 1 serum and for medium was and to medium was at were and from the in 1 1 and at medium to was acid and a that been the in were in and in of was of of at and were and were The were in containing 1 and then in consisting of 1 were and to cell of was and as were in containing and by at for were on a and a gel of G. Biochem. PubMed Scopus Google Scholar). were in for and then in a for were to an of medium from was on a and as The were in by at for and on 77 of was and were of of of of these enzymes for at of was and epitope-tagged were and on polyacrylamide as to study the processing of proIGF-I were by in proIGF-I from to IGF-I have been using mature IGF-I as and the E-domain of proIGF-I may this a expression vector that an the FLAG Ala−1 of the signal peptide and of the B-domain (Fig. 2). to the FLAG peptide have been The requires for and a FLAG peptide an amino The can be used to determine the signal peptide has been from The the FLAG peptide in and epitope-tagged an for Arg71 of to a This directs synthesis of the 70-amino-acid mature IGF-I the FLAG peptide at the amino terminus (Fig. and 1). it was necessary to determine the of the FLAG peptide would processing of the IGF-I 293 were IGF-I, and expression and were from medium using IGF-I (Fig. of were from of were also from of These have a of and the epitope-tagged of the IGF-I synthesized from the The from the at the position as the which has a of that is at The of the IGF-I and may forms of mature medium from 293 was and a was at (Fig. The by the may be an of medium from 293 was these several were The from the was also in and was by and IGF-I that this mature The from the was also by and may a of mature to the the and of and 16 as well as a from these were by both and not forms of IGF-I containing a signal peptide and the FLAG peptide does not signal activity. The which were not the may be forms of proIGF-I. amino acid was in to determine the high from by were proIGF-I proIGF-IA contains two both of which are in the E-domain. be specific for proIGF-I as to mature Fig. the of and of the expression The as a on this polyacrylamide gel and was were was not and were from both and and the be not The of is in the in The proIGF-IA E-domain contains potential site at and several serine and residues which be used for determine the and were forms of proIGF-I, from medium of 293 were to shown in Fig. and on the of proIGF-I the while the of the and It is that the peptide is N-glycosylated the peptide is and the peptide may be are of growth factors that can be it was of to determine the of conversion of proIGF-I to mature from 293 were the and were on polyacrylamide The as well as and were in cell that conversion can occur (Fig. We have also that are not to mature from 293 serum for to at of occurs not determine the importance of each of the basic residues in the pentabasic proIGF-I processing motif for prohormone conversion each basic residue to an residue 1). and were the and expression indicated that proIGF-I was at Arg71 and a also two codon for use as Arg74 and Arg77 were to generating and several of a basic residue an amino acid the of the of these it was necessary to codon in to the processing site. Fig. the of expression of the in 293 cells. polyacrylamide the as a The as a and the as a This gel is of of and amino acids in length. It is also to that both the and are at Arg71, that as as two residues on the carboxyl terminus of Arg71 are for processing activity. shown Figure is at two sites a peptide that the at as well as a of these the produced by expression of and that the peptide is by cleavage at Arg77 (Fig. several different have that of is at Arg71 and at We have not for cleavage at Arg74 1 in a of Arg71 to processing at this that the basic residue is required at the P1 site. at Arg77 is not (Fig. at either Arg74 Arg77 processing at Arg77 cleavage at Arg71 (Fig. The processing is to that of proIGF-I processing that on (Fig. This may be to of the and and are on it is that processing does not occur at The peptide more than the codon not and is at a site as The residue is required for processing at Lys65 is to the peptide is at two sites (Fig. The the that it is by cleavage at the of the is well it is that it processing at were The and as a the as the (Fig. These are at Arg71 not The has a than the that the two residues are required for cleavage at Arg71 (Fig. It is that cleavage of occurs at Arg77 since the and are it was not to cleavage for the it is that these are at at of the and of the were (Fig. and 2). of the was and it is that cleavage at Arg77 since this site is preferred to Arg74 in Arg74 is by of for Arg71 and Arg77 of the precursor is Arg74 and Arg77 are to residues of the precursor is at Arg71 (Fig. and 2). The data on processing of proIGF-I indicated that processing at Arg77 was in a at either Arg74 Arg77 (Fig. and suggesting that this sequence might be a cleavage site for which has been as S.S. Bresnahan P.A. Leppla S.H. Klimpel K.R. Thomas G. J. Biol. Chem. 1992; 267: 16396-16402Abstract Full Text PDF PubMed Google Scholar). These data also that processing at Arg71 requires both Lys68 (P4) and Arg71 (P1). It is not SPC1 process precursors containing a for in the position. to determine SPC1 is in proIGF-I expressed in and RPE.40 cells. RPE.40 were derived from to A and for T.J. PubMed Google Scholar). RPE.40 to process precursor of several as well as the This processing can be by SPC1 Robertson T.J. J. Biol. Chem. 1993; 268: Full Text PDF PubMed Google Scholar, T.J. J. Biol. Chem. 1993; 268: Full Text PDF PubMed Google Scholar). were the and codon to cleavage products at Arg71 and Arg77 were from (Fig. produced the Arg71 and Arg77 as well as proIGF-I and proIGF-I. Expression of and in RPE.40 in the of as was in cells. was expressed in RPE.40 a of mature IGF-I, to at Arg71, was This that RPE.40 process proIGF-I at Arg71 not We have also a to Arg71 cleavage was into LoVo not LoVo not process the the A. S. Cell. Biol. 1991; PubMed Scopus Google Scholar), and this processing has been to a SPC1 gene S. K. Hatsuzawa K. N. Y. Y. Murakami K. Nakayama K. Biochem. 1993; PubMed Scopus Google Scholar). We have used the expression vector and human embryonic kidney 293 to study proteolytic processing of human proIGF-I. the IGF-I have shown that the pCMVigf1 and expression synthesis and secretion of IGF-I and (Fig. the and have that the FLAG does not removal of the signal peptide (Fig. is to mature and can be used to study processing of proIGF-I. expressed in 293 directs synthesis of multiple forms of high which are converted to by Figure forms of human proIGF-I have not been Van J.J. 1993; PubMed Scopus Google Scholar) have shown that proIGF-I can be N-glycosylated in an in translation from of translation at not were for The expression vector contains translation at and has been proIGF-I has not been in and the significance of this is it has been that can proteolytic processing of precursor of the human precursor at can processing at R. M. J. 1991; PubMed Google Scholar). precursors can be converted to mature forms at the cell have shown that the E-domain of is at several sites through the of these including the cleavage to produce the mature to occur in the and may also occur Liu S.P. J. Biol. Chem. Full Text PDF PubMed Google Scholar). We mature in 293 cell that conversion can occur (Fig. We have not for processing of to mature were from 293 serum from not Expression of and in 293 that the IGF-I prohormone can be at two Arg71 and The Arg77 cleavage site to the minimal SPC1 cleavage site by S.S. Bresnahan P.A. Leppla S.H. Klimpel K.R. Thomas G. J. Biol. Chem. 1992; 267: 16396-16402Abstract Full Text PDF PubMed Google Scholar) Arg74 as the residue and Arg77 as the P1 containing in of Arg74 Arg77 processing at Arg77 containing of residues for Arg71, the of precursor processing at Arg77 is 2). for the processing at Arg77 in these is that the of the peptide may have been modified in a that for more each of the and Arg71 also processing at It is that the processing at Arg77 is a of the of a efficient Arg71 cleavage site. to data obtained from the expression of in SPC1-deficient cell lines also SPC1 as the Arg77 cleavage enzyme. is expressed in cleavage occurs at Arg71 and expressed in RPE.40 a lacking SPC1 cleavage of occurs at Arg71 not Arg77 (Fig. at Arg71, not was also was expressed in the SPC1-deficient LoVo cell line not It appears that SPC1 is required for cleavage of at Arg77 not IGF-I from human serum is amino acids it is that occurs by cleavage of the precursor at The of a cleavage site at which processing at Arg71, it to determine the minimal processing motif necessary for has that Lys68 and Arg71 are both for processing at of for Lys68 for Arg71 processing at Arg71 (Fig. 2). The P7 residue may also be important for efficient processing since of the is at Arg71, as to for proIGF-I. It that the and are more at Arg71 than the and 2). A is that the the and P1 of the SPC1 cleavage more for the Arg71 enzyme. Arg71 is not for processing since the is (Fig. 2). The SPC1 cleavage site has been as K. Nagahama M. S. K. Murakami K. Nakayama K. J. Biol. Chem. 1992; 267: Full Text PDF PubMed Google Scholar). SPC1 also cleave precursors an motif S.S. Bresnahan P.A. Leppla S.H. Klimpel K.R. Thomas G. J. Biol. Chem. 1992; 267: 16396-16402Abstract Full Text PDF PubMed Google Scholar, A. Barr P.J. Rhodes C.J. Kaufman R.J. Biochemistry. 1993; 32: 11586-11590Crossref PubMed Scopus (57) Google Scholar, K. Nagahama M. S. K. Murakami K. Nakayama K. J. Biol. Chem. 1992; 267: Full Text PDF PubMed Google Scholar) while precursors this motif are not Y. Nakayama K. Barr P.J. Thomas G. Y. J. 1992; PubMed Google Scholar). It is SPC1 cleave containing a residue in the position. The precursor of the of contains the cleavage The of was not in Y. Nakayama K. Barr P.J. Thomas G. Y. J. 1992; PubMed Google Scholar). A containing the cleavage motif was by SPC1 in both and in the of this motif was to to of the was and the was not SPC1 in S.S. Thomas G. T. T. Y. J. 1994; PubMed Google Scholar). data on expression of in RPE.40 and LoVo that SPC1 is not necessary for of The of the proIGF-I converting enzyme is SPC4 is a candidate it is expressed in this enzyme has a more strict requirement for a basic residue in the P2 position than does SPC1(16Rehemtulla A. Barr P.J. Rhodes C.J. Kaufman R.J. Biochemistry. 1993; 32: 11586-11590Crossref PubMed Scopus (57) Google Scholar, 17Creemers J.W.M. Kormelink P.J.G. Roebroek A.J.M. Nakayama K. Van de Ven W.J.M. FEBS Lett. 1993; 336: 65-69Crossref PubMed Scopus (40) Google Scholar). proprotein precursors that are at basic the scissile bond is on the carboxyl terminus of the P1 cleavage of the scissile bond by a specific the basic residues are by Recently, a which peptide on the amino terminus of residues was cloned from a cDNA A. T. P. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: PubMed Scopus Google Scholar). the proIGF-I cleavage site is to the motif by enzymes of the it is that a different of enzyme may proIGF-I to mature It is that the and codon are at This that as as two amino acids on the carboxyl terminus of the P1 residue are for precursor processing by the enzyme. of these as well as containing at Arg74 also that cleavage at Arg77 not cleavage at have shown that is at multiple sites in the Liu S.P. J. Biol. Chem. Full Text PDF PubMed Google Scholar). be necessary to determine proIGF-I is at Arg77 this is for the existence of in and the cleavage at Arg77 in may be a of this is in of site and proIGF-I may be a model for this of prohormone processing. of residues are amino SPC1 the motif The motif may also be by it is to be by converting enzyme as The motif does not to be by SPC1 the proIGF-I converting enzyme. this site appears to be to cleavage. to cleavage at Arg74 by the the Arg74 were (Fig. 2). of the role of residues in proIGF-I the of the pentabasic into processing of prohormone We for the pCMV6c expression vector and Thomas for the and RPE.40 cell The was a from of the and We also Gardner for synthesis of We are to and for and on this