Functional Characterization of a Putative Serine Carboxypeptidase in Vascular Smooth Muscle Cells
Rationale: We previously identified a novel serine carboxypeptidase, SCPEP1, that undergoes cleavage across all tissues where it is expressed. SCPEP1 bears the signature catalytic triad found in all serine carboxypeptidases, but its biological function is completely unknown.
Objective: To begin elucidating the functions of SCPEP1 in vitro and in the vessel wall after injury.
Methods and Results: Cultured smooth muscle cells were transduced with adenovirus carrying wild-type Scpep1, a short hairpin RNA to Scpep1, or variants of Scpep1 with mutations that disrupt the catalytic triad domain or SCPEP1 cleavage. Western blotting of key growth regulators or growth and migratory responses were assessed following SCPEP1 gain- or loss-of-function in smooth muscle cells. Vascular injury-induced remodeling and cell proliferation were evaluated in wild-type or newly created Scpep1 knockout mice. Overexpression of wild-type or cleavage-defective SCPEP1, but not a catalytic triad mutant SCPEP1, promotes smooth muscle cell proliferation and migration in vitro. A short hairpin RNA to Scpep1 blunts endogenous growth, which is rescued on concurrent expression of Scpep1 carrying silent mutations that evade knockdown. SCPEP1 protein is highly expressed in the neointima of 2 models of vascular remodeling. Scpep1-null mice show decreases in medial and intimal cell proliferation as well as vessel remodeling following arterial injury.
Conclusions: SCPEP1 promotes smooth muscle cell proliferation and migration in a catalytic triad-dependent, cleavage-independent manner. SCPEP1 represents a new mediator of vascular remodeling and a potential therapeutic target for the treatment of vascular occlusive diseases.
Smooth muscle cells (SMCs) are critical for blood vessel homeostasis, but they also contribute to the pathogenesis of several vasculopathies. In response to arterial injury, SMCs shift from a quiescent, contractile phenotype to a proliferative, synthetic state that undermines normal arterial function leading to neointimal formation.1,2 Myriad factors and cytokines, as well as proteases and their associated substrates, have been implicated in vascular pathology,3–5 but additional proteases are likely involved in this process. For example, the serine carboxypeptidase cathepsin A (CTSA) cleaves a number of substrates (eg, endothelin-1) that effect pathological changes in the vessel wall.6–8 Recently, a mutant CTSA allele defective for enzyme activity was knocked into the wild-type (WT) locus of mice and shown to confer a decrease in the inactivation of endothelin-1, elevated arterial blood pressure, and altered elastogenesis.9
Serine carboxypeptidases belong to the family of serine proteases and are most prevalent in the plant kingdom, where they function in numerous processes related to growth and development.10 Three serine carboxypeptidases are found in mammals and each shares the same catalytic triad (serine, aspartic acid, and histidine) found in plant homologs.10,11 We previously reported a novel serine carboxypeptidase from cultured SMCs in a screen for retinoid-induced genes.12 We call this protease serine carboxypeptidase (SCPEP)1 because it contains several conserved domains common to all members of the serine carboxypeptidase family, including a substrate-binding domain and the catalytic triad. Northern blotting and in situ hybridization studies demonstrated Scpep1 mRNA in SMCs of the aorta and proximal convoluted tubular epithelium of the kidney.12 More recently, we developed an antibody to SCPEP1 and showed its cleavage from a mature 55-kDa isoform to a 35-kDa isoform in all adult mouse tissues studied, including vascular SMCs and renal proximal convoluted tubular epithelium.13 The biological substrates for SCPEP1, however, remain a mystery. We therefore consider SCPEP1 an orphan protease. Here, we have performed gain- and loss-of-function studies in vitro and in vivo to provide the first biological insight into SCPEP1 function.
An expanded Materials and Methods section is available in the Online Data Supplement at http://circres.ahajournals.org.
Cultured SMCs transduced by adenovirus were used for growth curve studies, migration assays, and immunoblotting. Total RNA was extracted from cultured SMCs and tissues with TRIzol reagent (Invitrogen). RT-PCR was performed using the ProSTAR System (Stratagene). Short hairpin RNAs were generated as described.14 SCPEP1 mutagenesis was performed using a QuikChange site-directed mutagenesis kit (Stratagene). All constructs were then incorporated into adenovirus (Invitrogen). The animal protocol was approved by the Institutional Animal Care and Use Committee at the University of Rochester. Scpep1 knockout (KO) mice were generated through the University of Rochester Transgenic Core and back-crossed to C57BL/6 mice. Twelve-week-old Scpep1 KO mice or WT littermates were subjected to complete common carotid artery ligation as described.15
Ectopic SCPEP1 Enhances SMC Growth and Migration
Although SCPEP1 bears the catalytic triad common to all serine carboxypeptidases, we and others have been unable to demonstrate enzymatic cleavage activity. We therefore adopted different approaches to begin understanding the function of SCPEP1 in vascular SMC biology. As a first step, we transduced rat aortic SMCs (RASMCs) with adenovirus harboring full-length SCPEP1 (Ad-Scpep1) and show both the 55-kDa and 35-kDa SCPEP1 products are expressed in a dose-dependent manner (Figure 1A). Similar expression is seen in a pulmonary artery SMC line (data not shown). Adenoviral-mediated expression of SCPEP1 in RASMCs and pulmonary artery SMCs increases cell number (Figure 1B and data not shown). Importantly, such levels of SCPEP1 are comparable to those present in vivo after arterial injury (see below). To further elucidate the mechanisms underlying the growth-promoting effect of SCPEP1, we screened several signaling pathways in serum-stimulated SMCs and found that cyclin D1 is upregulated with Ad-Scpep1 overexpression (Figure 1C and 1D). In addition, downregulation of the negative growth regulator p27kip1 is exaggerated in Ad-Scpep1-transduced SMCs (Figure 1C and 1D).
To determine whether SCPEP1 is involved in SMC migration, a scratch wound assay was performed in SMCs stimulated with platelet-derived growth factor (PDGF)-BB. As shown in Figure 2A, cells migrate faster in the presence of SCPEP1 than in controls as early as 0.5 days post-PDGF stimulation. A quantitative analysis shows Ad-Scpep1 significantly increases SMC migration over controls at every time point examined (Figure 2B). To exclude a cell growth effect in the wound assay, we evaluated cell migration in Ad-Scpep1-transduced SMCs by a modified Boyden chamber assay. SCPEP1 significantly elevates SMC migration in this assay as well (Figure 2C).
Knockdown of Endogenous Scpep1 Reduces SMC Growth
To further assess the effects of SCPEP1 on SMC growth and migration, we generated adenovirus containing either a short hairpin RNA to Scpep1 or a rescue Scpep1 designed to evade knockdown through the introduction of silent mutations that preserve primary amino acid sequence (Online Figure I, A). Because endogenous SCPEP1 is very low in rat SMCs (Figure 1A and data not shown), we turned to a mouse aortic SMC line (MOVAS)16 that expresses higher levels of endogenous SCPEP1 (Figure 3A versus Figure 1A). Immunoblot analysis reveals more than 50% knockdown of endogenous SCPEP1 with our short hairpin RNA to Scpep1; coexpression of the rescue Scpep1 results in full reconstitution of SCPEP1 expression (Figure 3A). Cells transduced with the short hairpin RNA to Scpep1 show dramatic decreases in serum-stimulated growth (Figure 3B). Importantly, this finding does not appear to be attributable to off-target effects because restoring SCPEP1 levels with the rescue transgene normalizes SMC growth (Figure 3B). Similar changes in SMC growth and migration are observed in RASMCs when ectopically expressed SCPEP1 is knocked down (data not shown). Taken together with the above results, we conclude that elevated SCPEP1 induces vascular SMC growth and migration in vitro.
SCPEP1-Mediated SMC Growth and Migration Is Catalytic Triad-Dependent
To date, enzymatic activity of SCPEP1 has not been demonstrated despite the presence of a conserved catalytic triad, which is highly homologous to the catalytic triad of other serine carboxypeptidases (Online Figure I, B). To determine whether the growth and migratory effects of SCPEP1 require an intact catalytic triad, we generated SCPEP1S167A and SCPEP1H431Y point mutants (Figure 4A). Because SCPEP1S167A is expressed and cleaved properly (Figure 4A), we proceeded to study its effect on SMC growth and migration. Adenoviral delivery of SCPEP1S167A fails to increase cyclin D1 and decrease p27kip1 expression (Figure 4B). In addition, overexpression of SCPEP1S167A does not elicit increases in SMC growth (Figure 4C) or migration (Figure 4D). These results suggest that the stimulatory effects of SCPEP1 on SMC growth and migration require an intact catalytic triad providing the first evidence to support enzymatic activity for this novel protease.
SCPEP1 Cleavage Is Dispensable for SMC Growth and Migration
To determine whether SCPEP1 cleavage is important for SMC growth and migration, we first analyzed expression of several random SCPEP1 point mutants. We used the SCPEP1T103A mutant for growth and migration studies because, among mutants displaying a lack of cleavage, its amino acid change was least aggressive (Figure 5A). SCPEP1T103A is shown to be as effective as WT SCPEP1 in accentuating serum-mediated reductions in p27kip1 (Figure 5B). In contrast, SCPEP1S167A did not lower p27kip1 levels (Figure 5B). We next generated stable SMC lines expressing WT SCPEP1, SCPEP1T103A, or empty vector. As with WT SCPEP1, SCPEP1T103A exhibits stimulatory effects on SMC growth and migration (Figure 5C and 5D). These findings suggest that cleavage of SCPEP1 is not necessary for mediating SMC growth and migration.
SCPEP1 Expression Is Increased in the Neointima of Vascular Lesions
As a first step toward understanding SCPEP1 function in an in vivo setting, we investigated its expression by immunohistochemistry of mouse and rat carotid arteries subjected to different modes of arterial injury.15,17,18 We note intense SCPEP1 expression in the neointima of the mouse carotid artery following ligation injury (Figure 6A, a and d). Blocking studies establish the specificity of SCPEP1 staining (Figure 6A, b and e) as previously reported.13 That SCPEP1 is expressed in phenotypically modulated SMCs is evident from immunostaining studies using one of the most specific markers for the SMC lineage, the smooth muscle isoform of myosin heavy chain (MYH11) (Figure 6A, c and f).19 We also found that SCPEP1 expression is enriched at the luminal border of the balloon-injured rat carotid artery where SMCs display strong immunostaining for growth factors (Figure 6B, a and c).3 Importantly, Western blotting reveals a clear increase in SCPEP1 protein 28 days after injury (Figure 6C), with similar increases seen at 7 days (data not shown). Neither platelets nor plasma appears to express SCPEP1 (data not shown), suggesting the major source of elevated SCPEP1 stems from vascular cells. These data demonstrate injury-induced SCPEP1 expression in the vessel wall where modulated SMCs are known to be proliferative and migratory.
Development and Initial Characterization of Scpep1 KO Mice
We generated a Scpep1 KO mouse by replacing exons 1 and 2, encoding the N-terminal signal peptide and a portion of the putative substrate-binding domain, with a nuclear LacZ-floxed Neo cassette (Online Figure II, A). Targeting of the Scpep1 allele was validated by Southern blotting, as well as PCR genotyping, RT-PCR of the endogenous mRNA, and Western blotting of adult tissues (Online Figures II and III and data not shown). Scpep1 KO mice are born in the expected Mendelian ratios (eg, +/+, n=57; +/−, n=114: −/−, n=61). Scpep1-null mice appear normal, are fertile, and exhibit no histopathology across all organ systems analyzed (Online Figure IV and data not shown); similar findings were reported by an independent group during manuscript submission of this study.20 Scpep1 KO mice show no detectable SCPEP1 protein expression in aorta and kidney (Online Figure II, C and D). No obvious differences are seen between KO and WT littermates with respect to blood pressure, heart rate, and body weight (Online Figure V). Given 2 additional serine carboxypeptidases in mammals,11 we considered that one or both paralogs may compensate for loss in Scpep1. However, we found no differences in mRNA expression of the 2 serine carboxypeptidase paralogs between KO and WT mice (Online Figure VI). Collectively, these results reveal that genetic deletion of Scpep1 is compatible with embryonic development and does not elicit an overt phenotype under normal conditions in the adult mouse.
Vascular Remodeling and Intimal/Medial Cell Growth Are Attenuated in Scpep1 KO Mice
To directly test the role of SCPEP1 in a proliferative/migratory model of neointimal formation, we evaluated the extent of neointimal formation in the presence or absence of Scpep1 following ligation injury.15 Medial and intimal areas are similar in the contralateral, uninjured vessels of WT and Scpep1-null mice (Figure 7A and 7B). The ligated arteries of WT mice reveal significant neointimal formation (Figure 7C). Scpep1 KO mice, however, show a dramatic attenuation in injury-induced neointimal formation (Figure 7D and Online Figure VII). Morphometric analyses reveal significantly reduced intimal and medial areas and the intimal/medial ratios in Scpep1 KO mice as compared to WT control littermates (Figure 7E through 7G). Interestingly, we see a significant decrease in the area bound by the external elastic lamina (EEL) in KO vessels (Figure 7H). To determine whether such altered remodeling is associated with changes in matrix protein expression, we performed immunostaining for collagen I but found no obvious differences (Online Figure VIII).
Consistent with in vitro data (Figure 3B), loss of SCPEP1 is associated with significant decreases in intimal and medial cell proliferation 7 days after ligation injury (Figure 8A through 8F). There were no measurable differences in apoptosis rates between WT and KO injured vessels (Online Figure IX). Of note, we observe a stunning increase in p27kip1 protein expression from injured KO vessels (Figure 8H). Taken in aggregate, our in vivo findings complement in vitro gain- and loss-of-function studies and support an important role for SCPEP1 in vascular remodeling after injury.
Since we first identified Scpep1 8 years ago,12 virtually nothing has been published on this presumptive serine carboxypeptidase. The primary amino acid sequence of SCPEP1 contains the classic catalytic triad common to all members of this protease family,21 yet we have been unable to identify substrates for SCPEP1, even though many of the substrates tested can be cleaved by the serine carboxypeptidase, CTSA.6,7 A recent report also failed to reveal intrinsic protease activity for SCPEP1.20 Nevertheless, we hypothesize that SCPEP1-induced increases in SMC growth and migration require catalytic activity because SCPEP1S167A was ineffective in mediating these processes. Loss-of-function studies in vitro demonstrate that SCPEP1 is necessary for SMC growth and migration, a finding substantiated in vivo following vascular injury in Scpep1 KO mice. These results, together with the demonstrated increase in SCPEP1 following acute arterial injury, strongly support an important role for SCPEP1 in vascular remodeling accompanying damage to the vessel wall.
There is evidence indicating that neointimal cells originate, in part, from medial SMCs that switch from a quiescent and sessile phenotype (contractile) to a proliferative and migratory state (synthetic) following various perturbations to the vessel wall.2 Our in vivo data show that SCPEP1 is highly expressed in the neointima, where SMCs display reduced expression of MYH11, the gold standard marker for SMC lineages.19 Western blotting data suggest that cells of the vessel wall, most likely modulated SMCs of the media and intima, account for the increase in SCPEP1 following injury, although we cannot exclude circulating cells or plasma itself as additional sources of SCPEP1. Interestingly, levels of SRF, which is a major transcription factor for differentiation markers such as Myh11, do not appreciably change in the neointima22 or in normal medial SMCs from Scpep1 KO mice (Online Figure II, D). Thus, knockout of SCPEP1 does not appear to alter the SMC differentiation program. Loss of SCPEP1 does, however, have an effect on vascular SMC proliferation, as evidenced by decreases in Ki-67 immunostaining and significant increases in the negative growth regulator p27kip1. Whether the elevation of p27kip1 is a direct or indirect consequence of loss in SCPEP1 awaits further study.
Our previous data showed that full-length SCPEP1 undergoes cleavage to a 35-kDa isoform.13 Whereas another serine carboxypeptidase enriched in macrophages does not appear to undergo cleavage,23 CTSA is proteolytically processed to 32- and 20-kDa fragments, each of which harbors a portion of the catalytic triad.8 Although SCPEP1 is cleaved into a 35-kDa protein, the precise boundaries of this cleavage product are currently unknown. Nevertheless, our data suggest that SCPEP1 cleavage is not necessary for SCPEP1 to mediate SMC growth and migration responses. This result is somewhat unexpected because many proteases involved in vascular remodeling, such as matrix metalloproteinases, are synthesized and secreted as inactive proenzymes that subsequently undergo cleavage and activation either through an autocatalytic process or via other extracellular proteases.4,24 In this context, secreted SCPEP1 exists as a mature 55-kDa protein with no detectable 35-kDa species (data not shown), suggesting that this protease primarily undergoes intracellular proteolytic cleavage, perhaps within the lysosomal compartment of the cell. Evidence supporting the latter include colocalization of SCPEP1 with a lysosomal marker,13,25 as well as loss in SCPEP1 cleavage on treatment with the lysosomotropic agent chloroquine (data not shown).
Mutation of the active site residue (serine) is known to extinguish catalytic activity in other serine carboxypeptidases.9 Here, we show that mutation of serine 167 of the catalytic triad of SCPEP1 confers an inability for SCPEP1 to stimulate SMC growth and migration. Thus, although the substrate(s) for SCPEP1 remains undefined, disruption of the catalytic triad results in loss of SCPEP1 function, suggesting that SCPEP1 cleaves and activates proteins that may be of relevance for SMC growth and migration. Alternatively, SCPEP1 may cleave and inactivate substrates that normally function to maintain a quiescent, sessile SMC phenotype. It will be interesting to determine whether SCPEP1S167A is able to rescue the vascular phenotype in Scpep1 KO mice. Moreover, cells expressing SCPEP1S167A or Scpep1 KO cells will be useful tools to identify SCPEP1 substrates using a variety of biochemical and genetic approaches.26
There are several highly homologous domains between SCPEP1 and CTSA (Online Figure I, B).11 Moreover, each has similar patterns of expression in adult tissues, and both are localized to lysosomes.8,13 These findings suggest that SCPEP1 and CTSA may have overlapping functions. Evidence to date, however, suggests otherwise. CTSA has 2 distinct functions: a structural protective function in the lysosomes and extralysosomal catalytic activity.8 CTSA deficiency in humans and mice results in the lysosomal storage disease galactosialidosis, which is related to the loss of its structural protective function for β-galactosidase and neuraminidase.8,11 Apparently, SCPEP1 cannot compensate for this protective function because gene inactivation of CTSA in both humans and mice is incompatible with normal life. Moreover, studies in transgenic mice that carry a catalytically inactive CTSA reveal normal structural protection of the β-galactosidase/neuraminidase lysosomal complex; however, the degradation rate of endothelin-1, a known CTSA substrate, is significantly reduced, resulting in elevated arterial blood pressure.9 If SCPEP1 cleaved endothelin-1, the hypertensive phenotype would not likely be manifest. In this context, our data show that Scpep1 KO mice and WT littermates appear to have similar systolic blood pressure. Although CTSA does not compensate for loss in SCPEP1 with respect to the reduced neointimal phenotype, it is possible that there is compensation in other organ systems (eg, kidney). Pending further investigation, our results suggest that SCPEP1 and CTSA have distinct functions and thus nonoverlapping pools of substrates that function in cardiovascular homeostasis.
The in vivo remodeling and proliferation data from Scpep1 KO mice point to an important role for SCPEP1 in the response to injury of the vessel wall, although we cannot at this time definitively rule out an effect of the mixed genetic background. Further evidence for the physiological activity of SCPEP1 is highlighted by a sharp increase in the negative growth regulator p27kip1 in injured Scpep1 KO vessels. We speculate that one mechanism for the growth stimulatory effects of SCPEP1 may be through direct or indirect downregulation of p27kip1, a notion supported by the SCPEP1S167A data (see Figure 4B). The reduced caliber of injured Scpep1 KO vessels is also suggestive of SCPEP1 acting on substrates that, following proteolytic cleavage, may effect cell-cell and/or cell-matrix changes accompanying outward remodeling of injured blood vessels. The development of new mouse models using recombinant Scpep1 alleles offers a unique opportunity to directly assess the importance of the catalytic triad domain of SCPEP1 in these and other pathological contexts.
In summary, we provide the first documented biological activity of a putative vascular protease (SCPEP1) that we show enhances SMC proliferation and migration in a catalytic triad-dependent and cleavage-independent manner. SCPEP1 is elevated within the injured vessel wall and genetic inactivation of Scpep1 results in a marked reduction of SMC growth and vascular remodeling. Together, these results lay a critical foundation for future identification of SCPEP1 substrates and the further delineation of the mechanisms of action of SCPEP1 in vascular biology.
We thank former and present members of the Miano laboratory and the reviewers for constructive comments. We thank Dr Lin Gan for technical advice in the design of the Scpep1 targeting vector and its targeting to embryonic stem cells, Hou-Yu Chiang for providing training in ligation injury model, Dr Mansoor Husain for providing mouse aortic SMCs, Michael O'Dell for help with blood pressure measurements, Mary Georger for providing sectioning of mouse tissues, and David Meoli for assistance with the Boyden chamber assay.
Sources of Funding
J.M.M. was an American Heart Association Established Investigator, and J.C. was funded by an American Heart Association Postdoctoral Fellowship Award. The knockout studies were generously supported by the Department of Medicine of the University of Rochester School of Medicine and Dentistry.
Original received February 9, 2009; resubmission received April 27, 2009; revised resubmission received June 17, 2009; accepted June 19, 2009.
Owens GK, Kumar MS, Wamhoff BR. Molecular regulation of vascular smooth muscle cell differentiation in development and disease. Physiol Rev. 2004; 84: 767–801.
Majesky MW, Reidy MA, Bowen-Pope DF, Hart CE, Wilcox JN, Schwartz SM. PDGF ligand and receptor gene expression during repair of arterial injury. J Cell Biol. 1990; 111: 2149–2158.
Fay WP, Garg N, Sunkar M. Vascular functions of the plasminogen activation system. Arterioscler Thromb Vasc Biol. 2007; 27: 1231–1237.
Jackman HL, Tan FL, Tamei H, Beurling-Harbury C, Li XY, Skidgel RA, Erdos EG. A peptidase in human platelets that deamidates tachykinins. Probable identity with the lysosomal “protective protein.” J Biol Chem. 1990; 265: 11265–11272.
Jackman HL, Morris PW, Deddish PA, Skidgel RA, Erdos EG. Inactivation of endothelin I by deamidase (lysosomal protective protein). J Biol Chem. 1992; 267: 2872–2875.
Galjart NJ, Gillemans N, Meijer D, d'Azzo A. Mouse. “protective protein.” cDNA cloning, sequence comparison, and expression. J Biol Chem. 1990; 265: 4678–4684.
Seyrantepe V, Hinek A, Peng J, Fedjaev M, Ernest S, Kadota Y, Canuel M, Itoh K, Morales CR, Lavoie J, Tremblay J, Pshezhetsky AV. Enzymatic activity of lysosomal carboxypeptidase (cathepsin) A is required for proper elastic fiber formation and inactivation of endothelin-1. Circulation. 2008; 117: 1973–1981.
Chen J, Streb JW, Maltby KM, Kitchen CM, Miano JM. Cloning of a novel retinoid-inducible serine carboxypeptidase from vascular smooth muscle cells. J Biol Chem. 2001; 276: 34175–34181.
Lee TH, Streb JW, Georger MA, Miano JM. Tissue expression of the novel serine carboxypeptidase Scpep1. J Histochem Cytochem. 2006; 54: 701–711.
Streb JW, Miano JM. AKAP12α, an atypical serum response factor-dependent target gene. J Biol Chem. 2005; 280: 4125–4134.
Kumar A, Lindner V. Remodeling with neointima formation in the mouse carotid artery after cessation of blood flow. Arterioscler Thromb Vasc Biol. 1997; 17: 2238–2244.
Afroze T, Yang LL, Wang C, Gros R, Kalair W, Hoque AN, Mungrue IN, Zhu Z, Husain M. Calcineurin-independent regulation of plasma membrane Ca2+ ATPase-4 in the vascular smooth muscle cell cycle. Am J Physiol Cell Physiol. 2003; 285: C88–C95.
Goldberg ID, Stemerman MB, Schnipper LE, Ransil BJ, Crooks GW, Fuhro RL. Vascular smooth muscle cell kinetics: a new assay for studying patterns of cellular proliferation in vivo. Science. 1979; 205: 920–921.
Miano JM, Cserjesi P, Ligon KL, Periasamy M, Olson EN. Smooth muscle myosin heavy chain marks exclusively the smooth muscle lineage during mouse embryogenesis. Circ Res. 1994; 75: 803–812.
López-Otín C, and Bond JS. Proteases: multifunctional enzymes in life and disease. J Biol Chem. 2008; 283: 30433–30437.