Parathyroid Hormone–Related Protein as a Regulator of pRb and the Cell Cycle in Arterial Smooth Muscle
Background— Parathyroid hormone–related protein (PTHrP), a normal product of arterial vascular smooth muscle (VSM), contains a nuclear localization signal (NLS) and at least 2 translational initiation sites, one that generates a conventional signal peptide and one that disrupts the signal peptide. These unusual features allow PTHrP either to be secreted in a paracrine/autocrine fashion, and thereby to inhibit arterial smooth muscle proliferation, or to be retained within the cytosol and to translocate into the nucleus, thereby serving as an intracrine stimulator of smooth muscle proliferation.
Methods and Results— Here, we demonstrate 2 important findings. First, PTHrP dramatically increases the percentage of VSM cells in the S and in G2/M phases of the cell cycle. These effects require critical serine and threonine residues at positions Ser119, Ser130, Thr132, and Ser138 in the carboxy-terminus of PTHrP and are associated with the phosphorylation of the key cell cycle checkpoint regulator retinoblastoma protein, pRb. Second, because PTHrP devoid of the NLS serves as an inhibitor of VSM proliferation, we hypothesized that local delivery of NLS-deleted PTHrP to the arterial wall at the time of angioplasty might prevent neointimal hyperplasia. As hypothesized, using a rat carotid angioplasty model, adenoviral delivery of NLS-deleted PTHrP completely abolished the development of the neointima after angioplasty.
Conclusions— PTHrP interacts with key cell cycle regulatory pathways within the arterial wall. Moreover, NLS-deleted PTHrP delivered to the arterial wall at the time of angioplasty seems to have promise as an agent that could reduce or eliminate the neointimal response to angioplasty.
Received December 20, 2002; de novo received September 26, 2003; revision received March 9, 2004; accepted March 22, 2004.
Angioplasty, one of the mainstays of coronary, peripheral vascular, renovascular, and carotid arterial disease treatment, is highly effective but is limited by both early and late failures.1,2 Late failure is commonly a result of restenosis, a phenomenon that results, in part, from the proliferation and migration of arterial smooth muscle cells from the smooth muscle layer of the arterial wall, the media, into the lumen, where they form a new arterial layer, the neointima. The neointima, composed of vascular smooth muscle (VSM) cells and the extracellular matrix they have secreted, ultimately compromises the lumen of the angioplastied artery.
Parathyroid hormone-related protein (PTHrP), originally identified as the cause of humoral hypercalcemia of malignancy,3–5 is now known to be a widely distributed paracrine, autocrine, intracrine, and endocrine factor that has diverse roles in regulating mammalian development and survival.3–6 One of the tissues that produces PTHrP is the arterial VSM cell.3–5 PTHrP is a potent vasodilator and hypotensive agent when injected systemically.3–5,7,8 PTHrP can also serve as an inhibitor of angiogenesis.9 Moreover, overexpression of PTHrP or its receptor in the arterial wall of transgenic mice results in hypotension mediated by nitric oxide and by cAMP.10,11 PTHrP also seems to regulate the rate of arterial smooth muscle proliferation both in vitro and in vivo.10–13 We have shown that overexpression of PTHrP in VSM cells stimulates proliferation, whereas disruption of the PTHrP gene results in deceleration of the cell cycle in the arterial wall of embryonic mice.12
This ability of PTHrP to drive VSM proliferation requires an intact nuclear localization signal (NLS), a classic bipartite sequence of basic amino acids (see Figure 1A) that interact with the components of the nuclear import machinery, including importin-β.4,5,12–16 The PTHrP mRNA also contains 2 alternative translational initiation sites. One is a traditional Kozak sequence directly upstream of a functional signal peptide that directs the PTHrP translation product to the secretory pathway, with resultant exocytosis. Alternative translation initiation sites internal to the signal peptide may also be used.4,14,17 Use of these latter translational initiation sites disrupts the signal peptide and directs the PTHrP translation product to the cytosol, where, in concert with the NLS, it is directed to the nucleus. Thus, PTHrP can cause either mitogenic or antimitogenic effects in VSM cells, depending on whether or not the NLS is present: overexpression of wild-type (WT) PTHrP results in marked increases in VSM cell number and tritiated thymidine incorporation in VSM cultures, associated with nuclear entry of PTHrP.12,13 Conversely, overexpression of PTHrP from which the NLS has been deleted (ΔNLS-PTHrP) results in the opposite: marked slowing of proliferation in VSM cells and failure of PTHrP to gain access to the nucleus.12,13
Because PTHrP has repeatedly been demonstrated to be upregulated in arterial smooth muscle in angioplastied arteries, because it is also upregulated in atherosclerotic human coronary arteries resected at the time of coronary bypass grafting, and because it is apparently able to bidirectionally regulate VSM cell proliferation,3–5,7,8,12,13 it seemed likely that PTHrP might be involved in the arterial injury response to angioplasty. Therefore, we were interested in (1) further understanding the mechanisms through which wild-type (NLS-intact) PTHrP activates VSM cell proliferation and (2) determining whether antimitogenic NLS-deleted mutants of PTHrP administered to the angioplastied arterial wall might reduce the arterial response to injury.
Construction of PTHrP Mutants
The constructs shown in Figures 1 and 2⇓ were generated by in vitro site-directed mutagenesis as described previously,13 using the cDNA for human PTHrP (−36/+139) cloned into plasmid pcDNA-3+ as initial template. Each contains the 3′-UTR of human β-globin for stabilization of the mRNA (to replace the native PTHrP 3′-UTR AUUUA instability motif, which accelerates mRNA degradation) and to provide transcriptional termination, polyadenylation, and splicing signals. The constructs also contain a hemagglutinin (HA) tag, not used in the present study but previously demonstrated to have no effect on the localization or functional effects of PTHrP in A-10 cells.13
Cell Culture, Stable Transfections, and Cell Counting
The VSM cell line A10, derived from embryonic rat thoracic aorta, was purchased from American Type Culture Collection (Rockville, MD). Cells were cultured, stably transfected, selected, and counted as described in detail previously.13 Each growth curve in Figure 2 was performed 3 to 4 times on each of 3 clones derived from each construct, for a total of 7 to 12 growth curves per construct. Although this method assesses the combined effects of PTHrP on cellular proliferation and cell survival, the effects of PTHrP in this system reflect primarily proliferation as determined by use of tritiated thymidine incorporation12 and flow cytometry (see Figures 1 and 4⇓).
PTHrP Immunoradiometric Assay
PTHrP secreted from A10 cells stably transfected with the different PTHrP constructs or infected by the different adenovirus was measured in 24-hour conditioned medium obtained at confluence using a 2-site immunoradiometric assay (IRMA) specific for PTHrP(1–36).12,13 The detection limit of the assay is 0.5 pmol/L. For the measurement of PTHrP in cell extracts, A-10 cell extracts were performed as reported previously.13 Protein was measured according to the method of Bradford, and results are expressed as pmol/mg extract protein.
Cell Cycle Analysis
Cell cycle distribution was analyzed by flow cytometry. Exponentially growing A-10 cells stably transfected with the vector alone, WT-PTHrP, ΔNLS-PTHrP, or carboxy-terminal mutants shown in Figures 2 and 4⇑ were harvested, trypsinized, washed with PBS, and incubated in 70% ethanol at 4°C at least overnight. On the day of flow cytometry analysis, fixed cells were washed with PBS, pelleted, and resuspended in the staining PBS solution containing 50 μg/mL propidium iodide, 100 U/mL RNAse A, and 1 g/L glucose. Stained cells were filtered through a 30-μm nylon mesh, and DNA content was analyzed on a Becton-Dickinson flow cytometer.
Cell extracts were prepared and analyzed by 7.5% SDS-PAGE immunoblotted and transferred to Immobilon-P membranes by standard methods.18 The primary anti-retinoblastoma protein (pRb) antibody (Pharmingen) recognizes both pRb and ppRb.
Adenovirus encoding β-galactosidase (Invitrogen) (Ad-LacZ), wild-type human PTHrP (−36 to 139) (Ad-WT-PTHrP), and human PTHrP with a deleted NLS (Ad-ΔNLS-PTHrP) were prepared as we have reported previously.19,20 Multiplicity of infection (MOI) was determined by OD260 and plaque assay.
Rat Common Carotid Artery Angioplasty
The experimental protocol was approved by the University of Pittsburgh Institutional Animal Care and Use Committee. Balloon injury and adenovirus infection were performed on the left common carotid artery of adult Sprague-Dawley male rats weighing 450 to 600 g anesthetized with intraperitoneal injections of ketamine (150 mg/kg body wt) and xylazine (15 mg/kg body wt). A 2F Fogarty balloon catheter (Baxter) was inserted via arteriotomy into the left common carotid artery, inflated with a calibrated inflation device to a pressure of 2 atm for 5 minutes, and passed back and forth 3 times. A plastic catheter (27.5 gauge) was introduced through the external carotid arteriotomy, and the common carotid artery was flushed with PBS before introduction of adenovirus. Recombinant adenovirus stocks were used within 2 hours of thawing. Fifty microliters of 1010 pfu/mL adenoviral vector (Ad-LacZ, Ad-WT-PTHrP, or Ad-ΔNLS) or DMEM was instilled into the injured isolated common carotid segment. After 15 minutes, the adenovirus or DMEM was aspirated. The proximal external carotid artery was ligated, and blood flow was reestablished. Two weeks after balloon injury, the control uninjured right and the balloon-injured left carotid arteries with no adenovirus treatment (DMEM) or adenovirus treatment (Ad-LacZ, Ad-WT-PTHrP, or Ad-ΔNLS in DMEM) were harvested, subsequently fixed in 4% paraformaldehyde for 48 hours at 4°C, embedded in paraffin blocks, sectioned (5 μm), and stained either with hematoxylin and eosin or by the van Gieson method to reveal the internal and external elastic lamina. Images were acquired and analyzed for the cross-sectional areas of neointima and media using the NIH Image program, and the area ratio was calculated.
Retinoblastoma Protein, pRb, Is Constitutively Phosphorylated in PTHrP-Overexpressing VSM Cells
We were interested to know which cell cycle phases were activated by WT PTHrP overexpression and which were inhibited by overexpression of the NLS-deleted form of PTHrP. As can be seen in Figure 1B, A-10 VSM cells grown under normal, 10% serum-replete conditions display a cell cycle distribution with 81% G0/G1, 12% S, and 6.8% G2/M. A-10 cells overexpressing WT-PTHrP proliferate at a faster rate than control A-10 cells, with marked increases in both S and G2/M and a reduction in G0/G1. A-10 VSM cells overexpressing WT-PTHrP failed to decelerate under conditions of serum starvation (not shown). In marked contrast, A-10 cells overexpressing NLS-deleted PTHrP display even lower rates of S and G2/M than control A-10 cells, and the percentage of cells in G0/G1 is higher.
As can be seen in Figure 1C, representative of 5 experiments, the retinoblastoma protein, pRb, is present in control A-10 cells and is heavily phosphorylated to ppRb in WT-PTHrP overexpressing A-10 cells. In contrast, the majority of pRb is in the hypophosphorylated state in A-10 cells overexpressing NLS-deleted PTHrP. These results independently confirm the effects on proliferation of WT-PTHrP and the NLS-deleted forms of PTHrP derived from tritiated thymidine and cell growth curves.12,13
Ser119, Ser130, Thr132, and Ser138 in the Carboxy-Terminus of PTHrP Are Required for Activation or VSM Proliferation
We have demonstrated that although the NLS is required for nuclear targeting, it is not alone sufficient to stimulate proliferation.12,13 This requires the carboxy-terminal region of PTHrP, with crude mapping defining the PTHrP(107–139) region as important. Thus, PTHrP(88-139), including the NLS and the carboxy-terminus, is all that is required for stimulating VSM proliferation.12,13 We therefore sought to more finely map the carboxy-terminal region. Constructs deleted for bases encoding amino acids 107–111, 112–120, 121–130, and 131–139 (Figure 2A) were stably transfected into the rat arterial smooth muscle cell line A-10. The 107–111 region was selected for deletion because it is extremely highly conserved among mammalian species, in contrast to the 112–139 region, which is less well conserved.3–5,21–25 The other regions (112–120, 121–130, and 131–139) were selected randomly for deletion, because there is no known functional significance of these segments. Surprisingly, as shown in Figure 2B, despite its intense evolutionary conservation, deletion of the 107–111 region had no adverse effect on proliferation. Equally surprisingly, each of the other 3 deletion mutants essentially completely prevented the stimulation of proliferation (Figure 2B).
A NetPhos database search identified Ser119, Ser130, Thr132, Ser133, and Ser138 as potential phosphorylation sites in the PTHrP(112–139) region (Figure 2A). Accordingly, alanine substitution mutants at each of these sites were prepared individually, along with a sixth construct in which all of these serines and threonines were mutated to alanine and stably transfected into A-10 cells. As shown in Figure 2C, Ser133 is not required for the activation of proliferation by PTHrP. Conversely, Ser119, Ser130, Thr132, and Ser138 all seem to be essential for activation of proliferation in VSM by PTHrP. These differences were not because of ineffective expression or underexpression of the constructs, because analysis of the conditioned medium and cell extracts indicated that all were produced at elevated levels compared with the control A-10 cells (Figure 3).
Given the flow cytometric and pRb phosphorylation results shown in Figure 1, one would predict that the individual mutants associated with accelerated growth in Figure 2, B and C, would also be associated with increases in the S and G2/M cell cycle phases and with increased pRb phosphorylation. This proves to be correct (Figure 4, A and B). Each of the experiments shown here was repeated a minimum of 3 times, with 2 different clones representing each construct. WT-PTHrP, Δ107–111, and A133 mutants demonstrated increases in S and G2/M and pRb phosphorylation, whereas the other deletion and substitution mutants failed to stimulate either cell cycle progression or pRb phosphorylation.
Adenoviral Gene Delivery of NLS-Deficient PTHrP Completely Prevents Arterial Restenosis in a Rat Model of Carotid Angioplasty
Because PTHrP devoid of its NLS is a potent inhibitor of VSM proliferation (Figure 1B and References 12 and 13), we wondered whether this might have application to arterial restenosis. To address this question, we prepared adenoviruses containing (1) the WT-PTHrP construct (Ad.WT-PTHrP), (2) the ΔNLS-PTHrP construct (Ad.ΔNLS), and (3) a control adenovirus containing the LacZ gene (Ad.LacZ). We first sought to assess the efficiency of adenovirus-mediated gene transfer in cultured A10 VSM cells. In preliminary experiments, a 2500 MOI of Ad.LacZ for 15 minutes was sufficient to transduce essentially 100% of the cells as assessed by β-galactosidase activity. Under these conditions, Ad.WT-PTHrP and Ad.ΔNLS constructs resulted in easily measurable PTHrP(1–36) (≈40 pmol/L) in the medium. These findings indicate that the Ad.PTHrP constructs can efficiently transduce A-10 cells.
To ask whether overexpression of Ad.ΔNLS-PTHrP might inhibit the neointimal response to arterial injury in vivo, we used a standard rat carotid angioplasty arterial injury model. As can be seen in the example in Figure 5B, angioplasty induced marked neointima formation that was not present in the contralateral control carotid (Figure 5A). Similarly, angioplasty followed by the administration of Ad-lacZ resulted in comparable degrees of neointima formation (Figure 5C). Interestingly, angioplasty accompanied by Ad.WT-PTHrP (Figure 5D) resulted in a mild increase in neointima formation. In dramatic contrast to these several controls, angioplasty followed by the administration of Ad.ΔNLS-PTHrP essentially completely prevented the development of a neointima in this model (Figure 5E). These experiments were repeated multiple times (Figure 6). The neointima was reproducibly enhanced by the delivery of Ad.WT-PTHrP. In contrast, angioplasty followed by treatment with Ad.ΔNLS-PTHrP essentially completely (95% versus angioplasty alone) prevented the formation of a neointima.
The arterial wall in the carotids treated with the Ad.ΔNLS-PTHrP mutant appeared histologically in Figures 5 and 6⇑ to contain fewer cells than the other controls. We therefore performed quantitative histomorphometry in which the number of nuclei in the medial layer per high-power field on the 5 groups were counted. There were no differences between the 4 angioplasty groups (mean±SD, n=5 per group): DMEM alone, 82.6±5.1; Ad.LacZ, 92.7±4.5; Ad.WT-PTHrP, 89.8±3.7; Ad.ΔNLS-PTHrP, 98.4±8.9; P=NS). In contrast, each was higher than the control, untreated carotid (64.3±4.9, P<0.05).
The effects of adenovirus delivery to the arterial wall on the lumen, the external elastic lamina area, the neointima area, the media area, and the neointima-to-media ratio are summarized in the Table. As can be seen from Figure 6 and in the Table, the lumen was larger in the Ad.ΔNLS-PTHrP group. The external elastic lamina area was comparable in all 4 angioplasty groups.
We explored the cellular mechanisms that underlie the mitogenic effects of WT-PTHrP and examined the potential therapeutic usefulness of the antimitogenic actions of ΔNLS-PTHrP in VSM. We report that the PTHrP region required for VSM cell proliferation lies within the PTHrP(112–139) region and involves critical serine and threonine residues at positions 119, 130, 132, and 138. In addition, we demonstrate that overexpression of PTHrP in VSM cells results in activation of the cell cycle and release of the G1/S checkpoint, with a redistribution of the cells into S and G2/M phases of the cell cycle, all in association with hyperphosphorylation of pRb. In contrast, overexpression of the NLS-deleted form of PTHrP in association with reduced pRb phosphorylation almost completely blocks cell cycle progression in VSM cells. Perhaps most importantly, we demonstrate, in an in vivo model, that Ad.ΔNLS-PTHrP dramatically and completely inhibits carotid arterial neointima formation.
Cell Cycle Progression and Phosphorylation of pRb
The tumor suppressor protein pRb is a critical regulator of VSM cell proliferation (Stuart et al18 and references therein). Phosphorylation and inactivation of pRb in response to mitogenic stimulation result in G1/S transition and proliferation. The inhibition of pRb phosphorylation results in cell cycle arrest in VSM cells. Our observation that overexpression of WT-PTHrP induces cell cycle progression at the S and G2/M checkpoints (Figure 1B) in association with hyperphosphorylation of Rb (Figure 1C) is in accord with previous observations by Stuart et al18 and with our previous results showing stimulation of tritiated thymidine incorporation in A-10 cells overexpressing the WT-PTHrP.12 These findings suggest that the nuclear presence of PTHrP leads to pRb phosphorylation.
Conversely, our previous studies had demonstrated that overexpression of the NLS-deleted form of PTHrP in VSM cells was associated with reduced rates of cell growth, as assessed by cell counting and tritiated thymidine. Here, we show that ΔNLS-PTHrP overexpression leads to G0/1 arrest in association with low levels of pRb phosphorylation. Thus, both the stimulatory and inhibitory effects of PTHrP on VSM proliferation are ultimately regulated at the level of pRb and probably via regulatory molecules upstream of pRb, most likely including the cyclin D-cdk4/6 and/or cyclin E-cdk2 pathways.
C-Terminal Mapping of PTHrP
We had previously demonstrated that the 107–139 region of PTHrP, in conjunction with the NLS, was critical to driving proliferation in VSM cells.12,13 The deletion mutants and point mutants described here extend these studies, using 3 independent approaches (cell growth, cell cycle analysis, and pRb phosphorylation status) by identifying several specific serine and threonine residues within the carboxy-terminus of PTHrP that are critical for driving VSM cell proliferation. These serine/threonine residues presumably serve as sites for posttranslational modification, such as phosphorylation or acylation. The location of the key amino acids came as a surprise, for we had anticipated that the PTHrP(107–111) region, which is intensely conserved among species,3–5,21–25 would prove to be essential for proliferation. Presumably, this conserved region plays a role distinct from those explored here.
The precise function of these amino acids, and indeed of this entire region of PTHrP, in driving VSM cell proliferation remains undefined. Some have suggested that PTHrP binds to mRNA within the nucleus.26 Others have suggested that because the NLS in PTHrP is homologous to that in certain transcription factors, such as the glucocorticoid receptor family, the thyroid hormone receptor family, and the AP-1 family, including c-jun and c-fos, PTHrP may bind DNA itself and may be a transcription factor.14 It is also possible that PTHrP may interact with other nuclear or cytosolic proteins and that the carboxy-terminal phosphorylation status of PTHrP may regulate this interaction. In support of this possibility, Conlan et al27 have recently identified β-arrestin as a partner for PTHrP using yeast 2-hybrid strategies in a human placental library. Interestingly, the β-arrestin–PTHrP interaction involved the PTHrP(121–141) region, findings entirely consistent with our observations on the importance of this region. They hypothesize that the PTHrP–β-arrestin interactions in signaling rafts inside the cell surface may coordinate the interaction of G protein–coupled receptors with key signaling pathways related to mitogenesis, such as the Ras/Raf pathway.
From the data available at present, one can construct a model in which PTHrP, using an internal or downstream translational initiation site,5,6,17 is retained in the cytosol, with appropriate Thr85 regulation and β-importin interactions,15,16 and perhaps in the company of additional cytosolic proteins, transits the nuclear pore complex. It then enters the nucleus and in some manner, as yet undefined, drives the cell cycle.
ΔNLS-PTHrP Delivered at Angioplasty Inhibits Neointima Formation
In the rat carotid, PTHrP is markedly increased during neointimal formation after balloon angioplasty.7 In human coronary arteries, VSM cells at sites of coronary atherosclerosis overexpress PTHrP.8 Ishikawa et al28 recently demonstrated that local administration of PTHrP(1–34) inhibits intimal thickening induced by a nonobstructive polyethylene cuff in a rat iliac artery model of arterial injury. These observations imply that PTHrP produced locally within the arterial wall may play a role in the arterial response to injury but do not define what such a role might be.
Our previous observation that PTHrP devoid of the NLS is a potent inhibitor of VSM proliferation prompted the question as to whether ΔNLS-PTHrP delivered adenovirally to the arterial wall at the time of carotid angioplasty might have therapeutic efficacy in preventing neointimal hyperplasia. Here, we demonstrate that, as hypothesized, the delivery of ΔNLS-PTHrP using an adenoviral construct at the time of angioplasty profoundly suppressed the development of neointimal hyperplasia. This inhibitory response to neointimal development was quantitatively large, statistically significant, and highly reproducible. The effect could be attributed only to the ΔNLS-PTHrP, because control administration of either an Ad-LacZ virus or an Ad-WT-PTHrP virus did not inhibit neointima formation. Importantly, the method and timing of ΔNLS-PTHrP delivery are feasible in humans undergoing angioplasty.
The mechanism through which Ad-ΔNLS might inhibit neointima formation is not clear, but 2 general hypotheses seem reasonable. First, deleting the NLS in PTHrP prevents nuclear targeting of PTHrP and thus prevents its ability to drive the cell cycle. Moreover, as documented in Figure 3 and in the Results section, overexpression of ΔNLS-PTHrP also leads to enhanced secretion of PTHrP(1–36) compared with control VSM cells. As noted above, PTHrP(1–36), acting on the G-coupled PTH1-receptor on VSM cells to stimulate adenylyl cyclase, is a potent inhibitor of VSM proliferation.4,5,12,18 Thus, in this scenario, overexpression of ΔNLS-PTHrP would lead to 2 outcomes: ablation of the nuclear-PTHrP stimulus to VSM proliferation, and enhancement of PTHrP(1–36) secretion, resulting in cell surface PTH1-receptor–mediated inhibition of VSM proliferation. Theoretically, a second scenario could also be operative in which ΔNLS-PTHrP overexpression acts in a dominant negative fashion. In such a scenario, ΔNLS-PTHrP could serve to prevent endogenous PTHrP from entering the nucleus and prevent endogenous PTHrP from stimulating cell cycle progression. There is no current evidence to favor either of these possibilities.
One might wonder whether the dramatic inhibitory effect of the Ad.ΔNLS to prevent neointima would persist for longer than the 15 days used in the present study. Others have shown that after injury in the rat, the majority of the media VSM cells complete their proliferation and migration within 7 days, and the neointima shows a dramatic increase from 7 to 14 days. No significant further growth of neointima occurs beyond 14 days after balloon injury.29,30 Thus, it is possible that these results at 15 days predict a durable response.
As noted above, we have previously reported that arterial smooth muscle cell proliferation rates are reduced in the normal developing aorta of PTHrP-null mice, suggesting that nuclear-targeted PTHrP may participate in the normal proliferation and development of the arterial wall in embryonic life, and previous studies have demonstrated that PTHrP is upregulated in the media and neointima of angioplastied and diseased arteries, findings that suggest that PTHrP might actively participate in the neointimal response to injury. However, whether this participation might take the form of actively driving the VSM cell injury responses (perhaps via nuclear-targeted PTHrP) or serving as an attempt to halt or attenuate the injury response (perhaps via secreted PTHrP) has been difficult to ascertain. The observation in the present study that WT-PTHrP overexpression at the time of injury actually accentuates the neointima response to injury is concordant with the observations from the PTHrP null mice and suggests that the former hypothesis may be correct.
In summary, these studies demonstrate that PTHrP, specifically the NLS-deleted form, may have therapeutic benefit in disorders associated with arterial smooth muscle cell proliferation, migration, and matrix secretion. Future studies are necessary to identify the intracellular targets of PTHrP, the mechanisms through which it regulates the cell cycle, and whether the results observed in the rat carotid apply to arterial response to injury in other species.
This work was supported by grant R-01-54308 from the National Institute of Diabetes and Digestive and Kidney Diseases. The authors wish to thank Drs Timothy Billiar, Edith Tzeng, Jean-Jacques Helwig, Melina Kibbe, Joon Lee, Thierry Massfelder, Tom Clemens, and William Stuart for their help in conceptualizing these studies and in teaching us relevant techniques.
Philbrick WM, Wysolmerski JJ, Galbraith S, et al. Defining the roles of parathyroid hormone-related protein in normal physiology. Physiol Rev. 1996; 76: 127–173.
Karaplis AC, Luz A, Glowacki J, et al. Lethal skeletal dysplasia from targeted disruption of the parathyroid hormone-related peptide gene. Genes Dev. 1994; 8: 277–289.
Ozeki S, Ohtsuru A, Seto S, et al. Evidence that implicates the parathyroid hormone-related peptide in vascular stenosis: increased gene expression in the intima of injured carotid arteries and human restenotic coronary lesions. Arterioscler Thromb Vasc Biol. 1996; 16: 565–575.
Massfelder T, Dann P, Wu TL, et al. Opposing mitogenic and anti-mitogenic actions of parathyroid hormone-related protein in vascular smooth muscle cells: a critical role for nuclear targeting. Proc Natl Acad Sci U S A. 1997; 94: 13630–13635.
Lam MHC, House CM, Tiganis T, et al. Phosphorylation at the cyclin-dependent kinases site (Thr85) of parathyroid hormone-related protein negatively regulates its nuclear localization. J Biol Chem. 1999; 274: 18559–18566.
Lam MHC, Briggs LJ, Hu W, et al. Importin β recognizes parathyroid hormone-related protein with high affinity and mediates its nuclear import in the absence of importin α. J Biol Chem. 1999; 274: 7391–7398.
Stuart WD, Maeda S, Khera P, et al. Parathyroid hormone-related protein induces G1 phase growth arrest of vascular smooth muscle cells. Am J Physiol. 2000; 279: E60–E67.
Garcia-Ocana A, Takane KK, Reddy VT, et al. Adenovirus-mediated hepatocyte growth factor expression in mouse islets improves pancreatic islet transplant performance and reduces beta cell death. J Biol Chem. 2003; 278: 343–351.
De Miguel F, Martinez-Fernandez P, Guillen C, et al. Parathyroid hormone-related protein (107–139) stimulates interleukin-6 expression in human osteoblastic cells. J Am Soc Nephrol. 1999; 10: 796–803.
Whitfield JF, Isaacs RJ, Jouishomme H, et al. C-terminal fragment of parathyroid hormone-related protein, PTHrP-(107–111), stimulates membrane-associated protein kinase C activity and modulates the proliferation of human and murine skin keratinocytes. J Cell Physiol. 1996; 166: 1–11.
Aarts MM, Davidson D, Corluka A, et al. Parathyroid hormone-related protein promotes quiescence and survival of serum-deprived chondrocytes by inhibiting rRNA synthesis. J Biol Chem. 2001; 276: 37934–37943.
Liu MW, Roubin GS, King SB III. Restenosis after coronary angioplasty: potential biologic determinants and role of intimal hyperplasia. Circulation. 1989; 79: 1374–1387.
Clowes AW, Schwartz SM. Significance of quiescent smooth muscle migration in the injured rat carotid artery. Circ Res. 1985; 56: 139–145.