OUP user menu

Hypoxia exposure induces the emergence of fibroblasts lacking replication repressor signals of PKCζ in the pulmonary artery adventitia

(CC)
Mita Das, Nana Burns, Shelly J. Wilson, Wojciech M. Zawada, Kurt R. Stenmark
DOI: http://dx.doi.org/10.1093/cvr/cvn014 440-448 First published online: 24 January 2008

Abstract

Aims Cultured fibroblasts of hypoxia-stimulated remodelled pulmonary artery (PA) adventitia proliferate at a greater rate compared with those of normal adventitia. Since protein kinase C (PKC) ζ is a replication repressor of normal adventitial fibroblasts, we hypothesized that loss of the repressor activity of PKCζ might contribute to increased rate of proliferation in adventitial cells of remodelled PA.

Methods and results Isolated PA adventitial fibroblasts of neonatal control (Fib-C) and chronic hypoxia-exposed (Fib-H) calves were used to test our hypothesis. For evaluation of the role of PKCζ in hypoxia-induced vascular adventitial remodelling, expression and activation of PKCζ were also examined in lung sections of Fib-C and Fib-H animals by immunoperoxidase staining. Although constitutively active PKCζ expression attenuated DNA synthesis in Fib-C, it stimulated proliferation in Fib-H. PKCζ-specific myristoylated pseudosubstrate peptide inhibitor (PKCζ-PI) induced replication in Fib-C, whereas the inhibitor blocked DNA synthesis in Fib-H. Hypoxia stimulated PKCζ as well as MAP kinase kinase (MEK)1/2 and extracellular signal-regulated kinase (ERK)1/2 phosphorylation in Fib-H cells. However, ERK1/2 activation was mediated by both MEK1/2-dependent and MEK1/2-independent PKCζ-regulated mechanisms in hypoxia-exposed Fib-H. PKCζ was selectively activated in the adventitial cells of the remodelled vascular wall, as demonstrated by strong immunoreactivity against the anti-phosphoPKCζ antibody in the Fib-H lung sections.

Conclusion PKCζ acts as a replication repressor in Fib-C cells; however, the same isozyme mediates Fib-H proliferation. Thus, chronic exposure to hypoxia leads to the emergence of cells lacking anti-replication activity of PKCζ in the PA adventitia.

  • Hypoxia
  • Pulmonary hypertension
  • Fibroblast proliferation
  • PKCζ

1. Introduction

Enhanced proliferation of fibroblasts or fibroblast-like cells and marked structural remodelling in the pulmonary artery (PA) adventitia have been reported in the experimental models of hypoxia-induced pulmonary hypertension.1 To study the mechanisms contributing to the increased replication rate of fibroblasts, we and others have characterized the cells isolated and cultured from the PA adventitia of control and hypoxia-exposed animals.2,3 Fibroblasts derived from the remodelled PA adventitia exhibit heightened proliferation rate compared with those of normal adventitia.2,3 These observations are consistent with other reports demonstrating that fibroblasts/mesenchymal cells isolated postmortem from patients with persistent acute respiratory distress syndrome (ARDS) have increased proliferation as well as enhanced pro-survival signalling.46 The molecular mechanisms associated with the heightened replication responses of remodelled PA adventitial fibroblasts are not well characterized though both protein kinase C (PKC) and MAP kinases have been implicated.2,3,7

Multiple PKC isozymes are associated with hypoxia-induced signalling pathways.8,9 Hypoxia activates PKCζ, a calcium- and diacylglycerol-independent isozyme, in various cell types.1014 Recently, we have reported that hypoxia induces PKCζ phosphorylation in normal PA adventitial fibroblasts.15 We have also demonstrated that ERK1/2 activation is a critical mediator of the hypoxia-stimulated proliferation of these cells.7,15 PKCζ regulates ERK1/2 activation in other cell types.16 However, PKCζ activation stimulates ERK1/2 de-phosphorylation in the PA adventitial fibroblasts derived from normal animals suggesting that fibroblasts in the normal vessel wall are equipped with both pro- and anti-proliferative pathways.15 The replication repressor signals are activated along with proliferative pathways to attenuate fibroblast proliferation, presumably to limit excessive replication under normal physiological conditions. Whether PKCζ maintains its proliferation suppressor activity in fibroblasts isolated from hypoxia-stimulated remodelled PA adventitia is unknown.

Therefore, in the present study, we tested the hypothesis that the replication repressor activity of PKCζ would be lost in PA adventitial fibroblasts upon exposure to hypoxia. We compared the role of PKCζ in the proliferative responses of fibroblasts isolated from control and hypoxia-exposed animals using strategies to increase PKCζ levels with a constitutively active PKCζ (CAPKCζ) construct as well as to decrease its activity with PKCζ-specific myristoylated pseudosubstrate peptide inhibitor. We also sought to determine whether there are modifications in the downstream targets of PKCζ in fibroblasts isolated from hypoxia-stimulated remodelled PA adventitia. Finally, we evaluated the expression and activation levels of PKCζ in the normal and hypoxia-induced remodelled vascular wall.

2. Materials and methods

2.1 Cell culture

This investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). Adventitial fibroblasts were isolated from main PA adventitia of 15-day-old neonatal control calves (Fib-C) and calves exposed to hypoxia (simulated altitude of 4570 m in a hypobaric chamber) for 14 days beginning 1 day after birth (Fib-H).17,18 Myristoylated pseudosubstrate peptide inhibitors of PKCζ (PKCζ-PI) and PKC (Biomol) were used to evaluate the role of PKCζ isozyme in fibroblast proliferation.15 For the hypoxia experiments, quiescent Fib-C and Fib-H populations were placed in sealed humidified gas chambers and exposed to either normoxia (21% O2) or hypoxia (1% O2).2

2.2 Proliferation assay

Serum, hypoxia, and basic fibroblast growth factor (bFGF)-stimulated fibroblast proliferation was evaluated either by cell counts or [3H]thymidine incorporation or BrdU incorporation.15,19

2.3 Transfection studies

Quiescent fibroblasts were transiently transfected with either PCMV5 (empty vector) or plasmid containing CAPKCζ (a gift from Dr A. Toker, Harvard Medical School).15

2.4 Immunoblotting

PKCζ, PKCι, ERK1/2, MEK1, MEK2, phosphoPKCζ, phosphoMEK1/2, and phosphoERK1/2 levels were evaluated by western blot analysis.7,15 Images of immunoblots were scanned and quantitated using ImageJ program.20

2.5 Immunoprecipitation

PKCζ was immunoprecipitated from control and hypoxia-exposed cell lysates with anti-PKCζ antibody and immunoblotted with antibody against phosphoThreonine (phosphoThr), phosphoPKCζThr410, PKCζ, phosphoERK1/2, and ERK1/2.15

2.6 Immunofluorescent staining

Quiescent Fib-H populations were processed for immunofluorescent staining of PKCζ and phosphoERK1/2.15 Images were captured using an Olympus Infinity microscope coupled to a Photometrics Quantix cooled CCD camera with Deltavision digital de-convolution software.

2.7 Immunoperoxidase staining

Lung sections of control and hypoxia-exposed calves were used for immunoperoxidase staining for expression and activation of PKCζ using anti-PKCζ and anti-phosphoPKCζThr410 antibodies, respectively. Images were taken with Nikon Eclipse E800 microscope equipped with spot digital camera using Image-Pro Plus software.

2.8 Data analysis

All data are expressed as arithmetic means ± SD; n equals the number of replicate wells per test condition in representative experiments. One-way analysis of variance, followed by the Student–Newman–Keuls multiple-comparison tests were conducted within and between groups of data points. Data were considered significantly different if P ≤ 0.05.

3. Results

3.1 Fib-H have a greater replication rate compared with the Fib-C populations

To evaluate the differences between Fib-C and Fib-H proliferative responses, cells were plated at equal densities and allowed to grow for 10 days. Serum-stimulated cell division was greater for Fib-H than the Fib-C population at 7 and 10 days in culture (Figure 1A).

Figure 1

Fib-H populations proliferate faster than Fib-C cells in culture. (A) Serum-stimulated Fib-H populations divide at greater rate than the Fib-C cells. *P < 0.001 compared with Fib-C data. (B) [3H]thymidine incorporation is higher under both normoxic and hypoxic (24 h exposure) conditions in quiescent Fib-H cells compared with the Fib-C populations. *P < 0.05 compared with unstimulated Fib-C data. **P < 0.05 compared with the data of hypoxia-induced Fib-C and quiescent Fib-H populations. (C) Hypoxia (24 h exposure) induces greater [3H]thymidine incorporation in serum-stimulated Fib-H cells than that of Fib-C populations. *P < 0.001 compared with the basal value of Fib-C. **P < 0.001 compared with stimulated Fib-C data. Similar results were reproduced with at least two other cell populations isolated from two different animals; n = 4 replicate wells for each condition.

Quiescent cells were then exposed to hypoxia (24 h exposure) both in the absence and presence of 10% FBS containing medium. Basal DNA synthesis, as measured by [3H]thymidine incorporation, was higher in Fib-H populations compared with the Fib-C cells (Figure 1B). Hypoxia-stimulated DNA synthesis in both cell types (Figure 1B). However, hypoxia-induced [3H]thymidine incorporation was greater in Fib-H than that in Fib-C (Figure 1B). The combined effects of hypoxia and serum on DNA synthesis were also higher in Fib-H (22-fold) than in Fib-C cells (12-fold) (Figure 1C). Together, these data confirm the previous reports demonstrating that Fib-H cells exhibit heightened replication potential compared with Fib-C populations in culture.2,3

3.2 PKCζ has differential functional role in Fib-C and Fib-H proliferation

We have recently reported that PKCζ is a replication repressor for Fib-C populations.15 To evaluate the role of PKCζ in Fib-H proliferation, we used a CAPKCζ construct which has an amino-terminal myristoylation sequence of p60 c-Src, to induce overexpression of active PKCζ.21 PKCζ levels (75 kD) were upregulated by CAPKCζ in both cell types (Figure 2A and D). A peptide of ∼48 kDa, most likely a proteolytic fragment of PKCζ, was also found in the presence of CAPKCζ (Figure 2A and D). In Fib-H cells, immunoreactivity of anti-PKCζ antibody was also detectable at 65 kDa (Figure 2D).

Figure 2

PKCζ has opposite role in Fib-C and Fib-H proliferation. (A and B) CAPKCζ induces PKCζ overexpression but not PKCι in Fib-C populations. (C) CAPKCζ attenuates basal and hypoxia-stimulated DNA synthesis as evaluated by BrdU incorporation in Fib-C cells. *P < 0.001 compared with PCMV5-transfected normoxic results. **P < 0.01 compared with the data of PCMV5-transfected cells under hypoxic conditions. (D and E) Only PKCζ but not PKCι expression is upregulated by CAPKCζ in Fib-H populations. There was also selective immunoreactivity using anti-PKCζ antibody at 65 kDa. (F) CAPKCζ stimulates an increase in BrdU incorporation in both normoxia and hypoxia-exposed Fib-H populations. *P < 0.001 compared with PCMV5-transfected normoxic results. **P < 0.05 compared with the data of PCMV5-transfected cells under hypoxic conditions. Similar results were obtained from three independent experiments using fibroblast populations isolated from three different animals; n=4 replicate wells for the proliferation studies.

Because PKCι is closely related to PKCζ, we also examined the effects of CAPKCζ on PKCι expression. Overexpression of PKCζ had no effect on PKCι levels in either cell type suggesting that the effects of CAPKCζ on PKCζ levels are very selective to this isozyme (Figure 2B and E).

CAPKCζ attenuated BrdU incorporation in Fib-C under both normoxic and hypoxic conditions (Figure 2C). In contrast, CAPKCζ induced an increase in basal as well as hypoxia-stimulated DNA synthesis in Fib-H populations (Figure 2F). These data suggest that PKCζ is a proliferation-suppressing kinase for the Fib-C, whereas the isozyme is a proliferation-mediating kinase for Fib-H cells.

3.3 PKCζ-PI attenuates proliferation of Fib-H populations

To further establish an opposing role of PKCζ in Fib-C and Fib-H proliferation, we used PKCζ-PI to inhibit the isozyme. PKCζ-PI treatment stimulated the BrdU incorporation in Fib-C under both normoxia and hypoxia (Figure 3A). In contrast, a general PKC-specific peptide inhibitor had no effect on basal DNA synthesis and only partially reduced the hypoxia-stimulated BrdU incorporation (Figure 3A). Whereas hypoxia-induced DNA synthesis was entirely attenuated by PKCζ-PI in Fib-H populations, a general PKC inhibitor did not affect Fib-H proliferation (Figure 3B).

Figure 3

Hypoxia-induced DNA synthesis is differentially affected by PKCζ-PI treatment in Fib-C and Fib-H populations. (A) Only PKCζ-PI but not a general PKC inhibitor induces BrdU incorporation in Fib-C populations. *P < 0.001 compared with normoxic control value. **P < 0.05 compared with the data under hypoxia-stimulated conditions. (B) Hypoxia-induced proliferative responses are attenuated by PKCζ-PI in Fib-H populations. *P < 0.05 compared with normoxic basal value. **P < 0.05 compared with the results of hypoxic conditions. Similar results were reproduced with at least two other Fib-C and Fib-H populations isolated from two different animals; n = 4 replicate wells for all the conditions.

Inhibitory effects of PKCζ-PI on Fib-H replication were further evaluated upon stimulation with a mitogenic growth factor, bFGF. BrdU incorporation was increased by 26-fold in the presence of bFGF (Figure 4A), and as seen with the hypoxia (Figure 3B), bFGF-induced DNA synthesis was completely abolished by PKCζ-PI treatment (Figure 4A). In contrast, a general PKC inhibitor only partially blocked the bFGF-stimulated BrdU incorporation in Fib-H populations (Figure 4A).

Figure 4

PKCζ-PI blocks growth responses of Fib-H cells. (A) bFGF-induced BrdU incorporation is completely attenuated by PKCζ-PI. *P < 0.001 compared with basal data. **P < 0.001 compared with bFGF-induced results. (B) Only PKCζ-PI but not a general PKC inhibitor selectively inhibits serum-stimulated growth of Fib-H populations under both normoxic and hypoxic (72 h) conditions. *P < 0.05 compared with Day 0 cell count. **P < 0.05 compared with serum-stimulated cell numbers under normoxic conditions. ***P < 0.001 compared with serum-stimulated data under hypoxic conditions. (C) PKCζ localizes in the nuclear compartment. (D) PKCζ-PI treatment (24 h) induces translocation of PKCζ from the nucleus to the cytoplasm. Magnification: ×100. Similar results were obtained from three independent experiments using fibroblast populations isolated from three different animals; n = 4 replicate wells for the proliferation studies.

Since in vivo many growth factors, including serum, are present in the hypoxic microenvironments, we then compared the effects of PKCζ-PI and a general PKC inhibitor treatment on the hypoxia-induced (72 h exposure) Fib-H growth in the presence of 10% FBS containing medium. Serum stimulation induced cell division in quiescent Fib-H populations and hypoxia further augmented the proliferation (Figure 4B). However, pre-incubation of cells with PKCζ-PI prior to stimulation with serum and hypoxia completely inhibited Fib-H replication (Figure 4B). A general PKC inhibitor failed to affect proliferation under any conditions (Figure 4B).

We then examined the effects of PKCζ-PI treatment on PKCζ localization in Fib-H populations by immunofluorescent staining. PKCζ was detected within the nucleus of quiescent cells (Figure 4C). However, in the presence of PKCζ-PI, PKCζ immunoreactivity was diminished in the nuclear compartment and increased in the cytoplasm of the cells (Figure 4D). Similar patterns of PKCζ localization were also observed in hypoxia-exposed cells (data not shown). Collectively, these results strongly support the role of nuclear PKCζ as a pro-proliferative kinase in Fib-H populations.

3.4 Hypoxia stimulates PKCζ phosphorylation in Fib-H populations

To evaluate the signalling mechanisms involved in PKCζ-mediated Fib-H proliferation, we then focused on hypoxia as the proliferative stimulus. PKCζ activation is regulated by phosphorylation of Thr-410 which is mediated through phosphoinositide-dependent kinase 1 (PDK-1) as well as Thr-560, an autophosphorylation site.22 Immunoprecipitated PKCζ was immunoblotted for total phosphoThr (Thr-410+Thr-560) levels using antiphosphoThr antibody. Thr phosphorylation was increased at both 10 and 60 min of hypoxia exposure (Figure 5A and D). However, Thr-410 phosphorylation which was evaluated using anti-phosphoPKCζThr-410 antibody was not affected by hypoxia (Figure 5B and E). Equal amounts of PKCζ immunoprecipitation was confirmed by immunoblotting the precipitates against PKCζ (Figure 5C). Therefore, by contrasting the total Thr phosphorylation levels with Thr-410 phosphorylation, we conclude that hypoxia induces PKCζ autophosphorylation at the Thr-560 site in Fib-H populations.

Figure 5

Hypoxia upregulates PKCζ phosphorylation in Fib-H populations. (A) Total Thr (Thr-410+Thr-560) phosphorylation of PKCζ is increased by hypoxia exposure. (B) Hypoxia does not affect PKCζ Thr-410 phosphorylation. (C) PKCζ is equally immunoprecipitated from control and hypoxia-exposed cell lysates. (D) Quantitation of total Thr phosphorylation. (E) Quantitation of Thr-410 phosphorylation. *P < 0.05 compared with the normoxic data. Representative immunoblots of three independent experiments are shown.

3.5 Hypoxia induces MEK1/2-dependent and MEK1/2-independent ERK1/2 phosphorylation in Fib-H populations

Since MEK1/2 and ERK1/2 are known downstream targets of PKC,22 we then examined the hypoxia-induced activation of MEK1/2 and ERK1/2. MEK1/2 phosphorylation was maximal at 10 min of hypoxia exposure (Figure 6A and C). Non-phosphorylated MEK1 and MEK2 protein levels were not affected by hypoxia (Figure 6A).

Figure 6

Hypoxia induces MEK1/2 and ERK1/2 activation in Fib-H populations. (A) MEK1/2 phosphorylation is increased in hypoxia-exposed cells. (B) ERK1/2 phosphorytion is markedly up-regulated in hypoxia-exposed Fib-H. (C) Quantitation of MEK1/2 activation. *P < 0.05 compared with normoxic data. (D) Quantitation of ERK1/2 phosphorylation. *P < 0.001 compared with normoxic data. Representative immunoblots of three independent experiments are shown.

Hypoxia also induced ERK1/2 phosphorylation in Fib-H populations. Although hypoxia-stimulated MEK1/2 activation was greatest at the 10 min time point (Figure 6A and C), ERK1/2 phosphorylation was highest at 60 min of hypoxia exposure (Figure 6B and D). ERK1/2 protein levels were not regulated by hypoxia exposure (Figure 6B). The differences in the time-course of phosphorylation patterns between MEK1/2 and their downstream targets, ERK1/2 suggest that ERK1/2 is activated by both traditional MEK1/2-dependent and non-traditional MEK1/2-independent mechanisms in Fib-H.

3.6 PKCζ stimulates ERK1/2 phosphorylation in Fib-H populations

PKCζ has been shown to activate MEK1/2 and ERK1/2 in other cell types.22,23 Since hypoxia-stimulated ERK1/2 activation is partially mediated through a MEK1/2-independent mechanism, we sought to determine the role of PKCζ in ERK1/2 phosphorylation via a non-canonical pathway. Activated ERK1/2 was associated with PKCζ immunoprecipitates in Fib-H (Figure 7A). Importantly, phosphoERK1/2 levels in PKCζ immunoprecipitates were greatest at 60 min of hypoxia exposure (Figure 7A and D), the same time point at which hypoxia maximally phosphorylated ERK1/2 in MEK1/2-independent fashion (Figure 6B). PKCζ precipitates were also evaluated for the levels of non-phosphorylated ERK1/2. The same amounts of ERK1/2 were present at each time point of the hypoxia exposure (Figure 7B). PKCζ was also equally immunoprecipitated from the control and hypoxia-exposed cell lysates (Figure 7C). These data suggest that PKCζ might phosphorylate ERK1/2 in Fib-H cells.

Figure 7

PKCζ stimulates ERK1/2 phosphorylation in MEK1/2-independent manner in Fib-H populations. (A) Activated ERK1/2 is associated with PKCζ immunoprecipitates in hypoxia-exposed cells. (B) Interaction of non-phosphorylated ERK1/2 with PKCζ immunoprecipitates is not regulated by hypoxia. (C) PKCζ is equally immunoprecipitated from control and hypoxia-exposed cell lysates. (D) Quantitation of phosphoERK1/2 and PKCζ co-immunoprecipitation. *P < 0.05 compared with normoxic data. (E) Nuclear localization of phoshoERK1/2 is disrupted by PKCζ-PI treatment under both normoxic and hypoxic (24 h) conditions. Similar results were obtained in two other experiments using different cell populations isolated from two different animals.

Role of PKCζ in ERK1/2 activation in Fib-H cells was further evaluated by using PKCζ-PI. PhosphoERK1/2 localization in the absence and presence of PKCζ-PI was examined by immunoflurorescent staining. Activated ERK1/2 was detected in the nuclear compartment of Fib-H cells under both normoxic and hypoxic (24 h) conditions (Figure 7E), although the staining intensity was brighter in hypoxia-exposed cells (Figure 7E). Nuclear accumulation of phosphoERK1/2 was completely prevented by PKCζ-PI treatment under both conditions (Figure 7E). Taken together, these data strongly suggest that nuclear PKCζ is an upstream kinase for ERK1/2 in Fib-H populations.

3.7 PKCζ is activated in the adventitial compartment of remodelled vascular wall

To evaluate the expression and activation of PKCζ in the vascular wall, lung sections of Fib-C and Fib-H were processed for immunoperoxidase staining of total and phosphoPKCζ. PKCζ expression was markedly upregulated in the vascular wall of Fib-H compared with that of Fib-C (Figure 8A, upper panel). Greatest expression level of PKCζ was detected in the vascular media of Fib-H (Figure 8A). However, PKCζ in the medial layer of Fib-H was not active as there was no reactivity with anti-phosphoPKCζ antibody in this layer (Figure 8A, lower right panel). PKCζ in the adventitial compartment of remodelled vascular wall (Fib-H) was activated as demonstrated by strong immuno-reactivity with anti-phosphoPKCζ antibody in this compartment (Figure 8A, lower right panel). These data strongly suggest that hypoxia-induced vascular adventitial remodelling in the lung is associated with PKCζ activation.

Figure 8

PKCζ is selectively activated in the adventitial compartment of the hypoxia-induced remodelled vascular wall. (A) Immunoreactivities against anti-PKCζ (upper right panel, black arrow) and anti-phosphoPKCζ antibodies (lower right panel, black arrow) are increased in the structurally remodelled adventitial layer of vascular wall in the lung sections of hypoxia-exposed calves. Magnification: ×20. Similar staining patterns were observed in two other experiments using lung sections from two different animals. (B) Schematic representation of signalling that might be responsible for the opposite functional role of PKCζ in hypoxia-induced ERK1/2 activation and proliferation of Fib-C and Fib-H populations.

4. Discussion

The key finding of the present study is the differential functional role of PKCζ in the proliferative responses of Fib-C and Fib-H populations (Figure 8B). Our results demonstrate that PKCζ is a critical mediator of heightened replication rate of the adventitial cells isolated from hypoxia-induced remodelled PA. PKCζ-PI blocks the proliferation, whereas CAPKCζ induces the replication in these cells. Our data also demonstrate that Fib-H populations are composed of traditional MEK1/2-dependent and non-traditional MEK1/2-independent pathways for ERK1/2 activation and that PKCζ stimulates ERK1/2 phosphorylation in MEK1/2-independent fashion. The presence of activated PKCζ in hypoxia-stimulated remodelled vascular adventitial compartment also strongly supports the idea that PKCζ plays a critical role in the adventitial remodelling process during the development of hypoxia-induced pulmonary hypertension (Figure 8A). We have previously reported that hypoxia stimulates normal PA adventitial fibroblast proliferation through Gαi-dependent ERK1/2 activation and that PKCζ functions as a replication repressor by dephosphorylating ERK1/2 in these cells (Figure 8B).7,15 Therefore, our studies strongly suggest that chronic exposure to hypoxia leads to the appearance of adventitial cells in the PA with a distinct phenotype where the downstream targets and functional role of PKCζ is different from that in the PA adventitial cells of normal animals. This phenotypic modulation of cells has significant implications for therapeutic interventions since changes in signalling mechanisms by fibroblasts in the remodelled vessel wall must be taken into account.

The presence of phenotypically modified fibroblasts, characterized by heightened proliferation and altered signalling mechanisms has been reported in a variety of fibrotic diseases including hereditary gingival fibromatosis (HGF), idiopathic pulmonary fibrosis, ARDS, asthma and adventitial remodelling following systemic arterial injury.6,2427 Enhanced androgen-driven fatty acid biosynthesis has been implicated in the proliferative advantage of HGF fibroblasts.24 Impaired cell-to-cell communication between fibrotic fibroblasts induces enhanced cell proliferation during the progression of idiopathic pulmonary fibrosis.25 MAP kinase phosphorylation is different in fibroblasts derived from an asthmatic airway compared with fibroblasts from a healthy airway.26 Production of versican, the extracellular matrix protein, is increased in both cell types in response to mechanical strain. However, the upregulation of versican production can be abrogated by JNK inhibition only in normal cells suggesting that fibroblasts in the asthmatic airway lose the ability to turn off enhanced matrix production in response to mechanical strain. Plekhanova et al. have recently demonstrated that enhanced urokinase plasminogen activator expression augments adventitial cell accumulation, including myofibroblasts, and adventitial growth after carotid artery adventitial injury.27 Mesenchymal or fibroblast-like cells from patients with non-resolving ARDS exhibit an activational profile characterized by enhanced pro-survival signalling and an antiapoptotic phenotype compared with mesenchymal cells derived from patients with resolving ARDS.6 Collectively, these observations, which are consistent with those of the present study, strongly support the idea that fibrotic diseases are associated with the appearance of cells which have distinctly different phenotypes than those of ‘normal’ tissue fibroblasts.

PKCζ has been implicated in the regulation of different diseases. Defective insulin-induced PKCζ activation in skeletal muscle has been reported in type 2 diabetic patients.28 Increased PKCζ activity is linked to athero-susceptibility in regions of aorta normally subjected to flow turbulence.29 PKCζ also plays a critical role in normal and angiotensin II-stimulated neointimal growth of the vascular wall.30 Reduced osteoblast proliferation observed in arthritic patients correlates with decreased PKCζ expression.31 However, despite the importance of PKCζ in different vascular as well as non-vascular diseases, the role of this isozyme has never been explored in hypoxia-induced pulmonary hypertension. Here, we report that PKCζ is a key mediator of PA adventitial cell proliferation which contributes to the structural remodelling of this compartment during the progression of the disease.

Growing evidence indicates that phosphorylation is a crucial event in regulation of PKCζ activity. PKCζ is phosphorylated at Thr-410 by PDK-1 and also at Thr-560, which is an autophosphorylation site.3236 Thr-410 phosphorylation leads to increased activity, whereas Thr-560 auto-phosphorylation promotes PKCζ degradation.36 PKCζ phosphorylation at a particular site may be an intricate mechanism for the selective regulation of its biological function. In a previous report, we demonstrated that hypoxia stimulates PKCζ Thr-410 phosphorylation in Fib-C populations.15 However, in that cell type, PKCζ activation negatively regulates hypoxia-induced ERK1/2 phosphorylation resulting in decreased cell proliferation. The appearance of PKCζ breakdown products at 65 kDa in the Fib-H, but not in Fib-C (Figure 2A and D) suggests that PKCζ phosphorylation in Fib-H populations is likely occurring at Thr-560, the degradative site, rather than in Thr-410. Indeed, in Fib-H cells, phosphorylation might occur on Thr-560 because of the increase in total Thr phosphorylation and lack of upregulation in Thr-410 phosphorylation (Figure 5). Thus, PKCζ might lose its proliferation suppressing activity due to the deficiency of increase in Thr-410 phosphorylation. The differential PKCζ phosphorylation patterns between Fib-C and Fib-H populations could explain, at least in part, the diverse functional role of the isozyme in these two cell types. A recent report which argues that differential phosphorylation of PKCζ at Thr-410 and Thr-560 predicts differences in susceptibility to atherosclerosis in vivo also supports the idea that differential phosphorylation patterns affect the PKCζ functional roles.29 The specific mechanism involved in PKCζ-mediated cell proliferation in the context of its site-specific phosphorylation will be the focus of our future studies.

ERK1/2 activation is known to be mediated by MEK1/2 pathway. PKCζ directly stimulates MEK1/2, resulting in subsequent activation of ERK1/2 in response to different stimuli.22,37,38 The present study, however, demonstrates that ERK1/2 activation occurs by at least two separate mechanisms in Fib-H populations. MEK1/2-mediated ERK1/2 stimulation is clearly present (Figure 6). The existence of a non-canonical route, MEK1/2-independent, is also supported by the fact that the time course of maximal ERK1/2 activation is different than that of MEK1/2 phosphorylation (Figure 6). Disruption of nuclear phosphoERK1/2 localization with a PKCζ-specific inhibitor as well as the association of activated ERK1/2 with PKCζ confirms the role of PKCζ in the MEK1/2-independent ERK1/2 activation. Reports demonstrating PKC isozyme-mediated MEK1/2-independent ERK1/2 activation in other cell types also support our results regarding ERK1/2 activation by PKCζ in Fib-H populations.39,40 Co-localization of activated ERK1/2 and PKCζ in the nuclear compartment of Fib-H populations might reflect a regulatory mechanism of heightened proliferation in these cells as nuclear localization of activated ERK1/2 is a crucial step leading to cell cycle entry.41 Therefore, multiple lines of evidence suggest that cells with heightened replication potential and distinct proliferation regulatory pathways exist in the hypoxia-induced remodelled PA adventitia.

Based on our observations, we conclude that PKCζ is a critical proliferation mediator in the cells of hypoxia-stimulated remodelled PA adventitia. However, the functional role of PKCζ is opposite in fibroblasts from normal PA adventitia where it serves to limit replication responses. In this regard, PKCζ represents a potentially important disease-associated kinase responsible for marked adventitial cell proliferation during the development of hypoxia-induced pulmonary hypertension and is an attractive therapeutic target. Small molecule inhibitors of PKCζ might potentially provide a powerful therapeutic option for the treatment of hypoxia-induced pulmonary hypertension.

Funding

National Institutes of Health (HL64917 to M.D., HL57144-09, HL14985-33 to K.R.S.).

Acknowledgements

The authors would like to thank Steve Hofmeister and Sandi Walchak for harvesting the bovine PA tissue and Dr R. A. Nemenoff for critical review of the manuscript.

Conflict of interest: none declared.

References

View Abstract