- Email: [email protected]
Altered vascular activation due to deficiency of the NADPH oxidase component p22phox
Cardiovascular Pathology 23 (2014) 35–42
Contents lists available at ScienceDirect
Cardiovascular Pathology
Original Article
Altered vascular activation due to deficiency of the NADPH oxidase component p22 phox He Wang a, Hassan Albadawi b, Zakir Siddiquee a, Jillian M. Stone a, Mikhail P. Panchenko a, Michael T. Watkins b, James R. Stone a,⁎ a b
Center for Systems Biology, Massachusetts General Hospital and Department of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA Department of Surgery, Division of Vascular and Endovascular Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA
a r t i c l e
i n f o
Article history: Received 10 July 2013 Received in revised form 4 August 2013 Accepted 5 August 2013 Keywords: Hydrogen peroxide HnRNP-C MMP12 NADPH oxidase Protein kinase CK1α Protein kinase CK1αLS Reactive oxygen species TIMP1 Vascular activation Vascular injury
a b s t r a c t Background: Reactive oxygen species generated by nicotinamide adenine dinucleotide phosphate (NADPH) oxidase play important roles in vascular activation. The p22 phox subunit is necessary for the activity of NADPH oxidase complexes utilizing Nox1, Nox2, Nox3, and Nox4 catalytic subunits. Methods: We assessed p22 phox-deficient mice and human tissue for altered vascular activation. Results: Mice deficient in p22 phox were smaller than their wild-type littermates but showed no alteration in basal blood pressure. The wild-type littermates were relatively resistant to forming intimal hyperplasia following carotid ligation, and the intimal hyperplasia that developed was not altered by p22phox deficiency. However, at the site of carotid artery ligation, the p22 phox-deficient mice showed significantly less vascular elastic fiber loss compared with their wild-type littermates. This preservation of elastic fibers was associated with a reduced matrix metallopeptidase (MMP) 12/tissue inhibitor of metalloproteinase (TIMP) 1 expression ratio. A similar decrease in the relative MMP12/TIMP1 expression ratio occurred in human coronary artery smooth muscle cells upon knockdown of the hydrogen peroxide responsive kinase CK1αLS. In the ligated carotid arteries, the p22 phox-deficient mice showed reduced expression of heterogeneous nuclear ribonucleoprotein C (hnRNP-C), suggesting reduced activity of CK1αLS. In a lung biopsy from a human patient with p22 phox deficiency, there was also reduced vascular hnRNP-C expression. Conclusions: These findings indicate that NADPH oxidase complexes modulate aspects of vascular activation including vascular elastic fiber loss, the MMP12/TIMP1 expression ratio, and the expression of hnRNP-C. Furthermore, these findings suggest that the effects of NADPH oxidase on vascular activation are mediated in part by protein kinase CK1αLS. © 2014 Elsevier Inc. All rights reserved.
Within the vessel wall, reactive oxygen species (ROS) including superoxide and hydrogen peroxide (H2O2) play important roles in regulating blood pressure and in mediating vascular cell activation due to stimuli such as altered flow and vascular injury [1–3]. The nicotinamide adenine dinucleotide phosphate (NADPH) oxidase complexes are a principal source of ROS in the vessel wall [4–6]. In mammals, these complexes contain one of seven distinct catalytic subunits: Nox1–Nox5, Duox1, and Duox2, although only Nox1, Nox2/ gp91 phox, Nox4, and Nox5 are thought to be expressed in the vessel wall. Of these, Nox1, Nox2, and Nox4 all utilize a structural subunit, p22 phox, which is also required for the function of Nox3. Nox1–Nox3 Abbreviations: hnRNP-C, heterogeneous nuclear ribonucleoprotein C; MMP3, matrix metallopeptidase 3; MMP12, matrix metallopeptidase 12; ROS, reactive oxygen species; TIMP1, tissue inhibitor of metalloproteinase 1. Funding: This work was funded by the Department of Pathology and the Department of Surgery, Division of Vascular and Endovascular Surgery, Massachusetts General Hospital. ⁎ Corresponding author. Massachusetts General Hospital, Boston, MA 02114. Tel.: +1 617 726 8303; fax: +1 617 643 3566. E-mail address: [email protected] (J.R. Stone). 1054-8807/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.carpath.2013.08.003
are also regulated by either of two regulatory subunits, p67 phox and NoxA1, and also by either of two structural/organizing subunits, p47 phox and NoxO1. Different types and concentrations of ROS may impact multiple signaling pathways in cells; for example, low concentrations of H2O2 modulate the function of a nuclear pre-mRNA binding protein, heterogeneous nuclear ribonucleoprotein C (hnRNP-C) [7,8]. This modulation occurs through phosphorylation following activation of Table 1 Baseline measurements on p22phox-deficient mice Parameter assessed
Wild type
p22phox deficient
P
Body weight (g)a Heart weight (mg)b Heart weight/body weight (mg/g)b Systolic blood pressure (mmHg)c Pulse (beats/min)c
20.5±0.4 149±7 7.4±0.4 122±4 592±16
18.0±0.4 140±10 7.7±0.5 126±4 599±13
.0002 .51 .61 .53 .72
a b c
n=26 mice (9 males and 17 females) in each group. n=15 mice (4 males and 11 females) in each group. n=12 mice (6 males and 6 females) in each group.
36
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
1. Materials and methods 1.1. Mice
Fig. 1. Intimal hyperplasia in the proximal and mid carotid artery following ligation. Left common carotid artery after ligation in wild-type (A) and p22phox-deficient (B) mice. Scale bars represent 20 μm. (C) The wild-type mice were relatively resistant to the formation of intimal hyperplasia with no effect due to p22phox deficiency. n=6– 8 male mice per group.
the nuclear protein kinase CK1αLS [8,9]. While hnRNP-C is ubiquitously expressed in some tissues such as liver [10], in arteries, hnRNPC is only highly expressed in activated states such as intimal hyperplasia and atherosclerosis [11]. Knockdown of CK1αLS was found to render human coronary artery smooth muscle cells insensitive to the proliferative effects of exogenous H2O2 and to prevent the up-regulation of hnRNP-C in cultured human arteries [12]. This suggested that the expression of hnRNP-C in arteries may be under the regulation of ROS. Recently, the nmf333 mouse strain with a head-tilt phenotype was identified as being deficient in p22 phox due to a point mutation in the corresponding Cyba gene [13]. Given the requirement of p22 phox for the function of vascular Nox1, Nox2, and Nox4, we hypothesized that p22 phox deficiency could have significant effects on vascular activation and vascular hnRNP-C expression in these mice.
B6 Tyr+/−Cyba(nmf333)/J mice carrying inactivating mutations in the Cyba gene [13] encoding p22 phox and wild-type B6 Tyr+/J littermate controls were obtained from The Jackson Laboratory (Bar Harbor, ME, USA) at 8–12 weeks of age. Blood pressures and pulse rates were determined using a noninvasive tail cuff and pulse transducer system (BP2000 Blood Pressure Analysis System, Visitech Systems, Apex, NC, USA). At least eight separate measurements were obtained for each mouse over the period of 1 week, and the values were averaged. Blood pressure measurements were taken at the same time each day between 9:00 and 11:00 a.m. Mice were euthanized and assessed for body weight and heart weight. Tissues from the heart, liver, and aorta were fixed in formalin and embedded in paraffin. Male mice at 8–9 weeks of age were subjected to left common carotid artery ligation under 40–50 mg/kg pentobarbital anesthesia. Mice were placed on a heated pad to maintain a body temperature of 37°C. Once anesthetized, the neck and chest were shaved, and the skin was cleansed with Betadine Solution. A longitudinal neck incision was made, and the left common carotid artery was exposed. The distal aspect of the left common carotid artery was encircled with fine forceps and permanently ligated just prior to the bifurcation using 80 silk suture. The incision was closed, and the mice were allowed to recover. After 4 weeks, the mice were euthanized, and the vessels were harvested following perfusion fixation with 10% buffered formalin through the left ventricle and embedded in paraffin. In other experiments, carotid arteries from male and female mice were harvested fresh, 1 week after ligation, and snap frozen in liquid nitrogen. The mouse studies were approved by the Subcommittee on Research Animal Care at Massachusetts General Hospital. 1.2. Human cells and tissue Paraffin blocks of discarded lung tissue were obtained from one child with p22 phox deficiency and two control subjects matched for age and gender. All subjects were boys at approximately 1 year of age who had undergone pulmonary wedge biopsy. The indications for biopsy were pneumonia in the p22 phox-deficient patient, and
Fig. 2. Elastic fiber loss at the site of ligation in the carotid arteries. Sites of ligation in the left common carotid artery in wild-type mice (A) showing marked loss of vascular elastic fibers. In contrast, the p22phox-deficient mice (B) show preservation of the vascular elastic fibers. (C) Quantitation of vascular elastic fiber loss. *P=.005, n=5 to 6 male mice per group. At higher magnification, the vascular smooth muscle cells show disorganization in the wild-type animals at the sites of elastic fiber loss (D) compared with the smooth muscle cells in the p22phox-deficient mice (E). Scale bars represent 40 μm (A, B) or 20 μm (D, E).
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
37
for the interior luminal area, the internal elastic lamina, and the external elastic lamina. The ratio between the intimal to the medial areas was calculated for each section, and the ratios were averaged for all the sections in each vessel, excluding the site of the ligation itself in the distal portion of the artery. The amount of elastic fiber loss at the sites of ligation was obtained using ImageJ by measuring the circumferential arc lengths of the regions with preserved and absent elastic fibers using the Freehand Line tool and was expressed as percent of circumference. Immunohistochemistry was performed using the peroxidaseconjugated avidin–biotin method (Vector) following antigen retrieval with either Borg decloaker solution (Biocare Medical) or Tris/Borate EDTA cell conditioning solution CC1 (Ventana Medical Inc.). Mouse tissue was stained for hnRNP-C using goat polyclonal antibodies (Santa Cruz) at 1:200 dilution and rabbit polyclonal antibodies (Santa Cruz) at 1:50 dilution. Human tissue was stained for hnRNP-C using mouse monoclonal anti-hnRNP-C antibody (clone 4F4, Santa Cruz) at 1:200 dilution. After blocking with Rodent Block M (Biocare Medical), mouse tissue was stained for the neutrophil marker Ly-6B using rat monoclonal antibodies (Abd Serotec) at 1:1000 dilution, for the macrophage marker F4/80 using rat monoclonal antibodies (Abcam) at 1:300 dilution, and for the oxidative stress marker 4-hydroxy-2nonenal (HNE) using mouse monoclonal antibodies (Abcam) at 1:100 dilution. Negative controls were performed by omitting the primary antibody. 1.4. Real-time polymerase chain reaction (PCR)
Fig. 3. Changes in the expression of MMP12, MMP3, and TIMP1. One week following carotid artery ligation, mRNA levels were assessed in full-length ligated and contralateral nonligated carotid arteries of both male and female mice by real-time PCR. (A) MMP12, (B) MMP3, (C) TIMP1, (D) MMP12/TIMP1 ratio. #Pb.05 versus the nonligated side, *Pb.01 versus the nonligated side, **Pb.01 versus the nonligated side, and Pb.01 versus the ligated p22phox-deficient group. Two-way ANOVA/Tukey, n=4 to 6 mice per group.
interstitial lung disease and benign neoplasia in the control patients. The clinical and genetic details of the patient with p22 phox deficiency have been reported previously [14]. The research with human tissue was approved by the Human Subjects Institutional Review Board at Massachusetts General Hospital. Paired sets of human coronary artery smooth muscle cells infected with retrovirus either containing a control vector or expressing a short hairpin RNA targeting the L-insert of CK1αLS were prepared and analyzed for CK1αLS knockdown efficiency previously [12]. 1.3. Histology and immunohistochemistry Portions of mouse carotid arteries, ~2–3 mm in length, were serial sectioned with four to six 7-μm sections per slide. Every 10th slide (or one slide per ~300 μM) was stained with the modified Verhoeff elastic–Van Gieson stain (Electron Microscopy Sciences, Hatfield, PA, USA) to assess for intimal hyperplasia in the proximal and mid portions of the artery. Intervening slides were utilized for additional modified Verhoeff elastic-Van Gieson stains and for immunohistochemical stains in the distal portion of the arteries. Image analysis software (Diagnostic Instrument, Inc.) was used to calculate the area
Total RNA was extracted from the frozen mouse carotid arteries with TRIzol Reagent (Ambion, Carlsbad, CA, USA) and additionally purified using an Arcturus PicoPure RNA Isolation Kit (Applied Biosystems, Foster City, CA, USA). Total RNA was isolated from human coronary artery smooth muscle cells as described previously [12]. Total RNA (111.2 ng for mouse carotid arteries and 100.75 ng for human smooth muscle cells) was reverse-transcribed in a final volume of 20 μl using mixture of oligo(dT) and random primers with a SuperScript First-Strand Synthesis SuperMix for qRT-PCR (Invitrogen). Real-time PCR (45 cycles) was performed in a final volume of 25 μl containing 12.5 μl of the SYBR Green 2× Master Mix (Applied Biosystems), 5 μl of the 1:10-diluted reverse transcription reaction, and 7.5 μl of primer pairs (1 μM each) using a 7500 Fast Real-Time PCR System (Applied Biosystems). The mouse primers used were matrix metallopeptidase (MMP) 12: forward 5′-CTGCTCCCATGAATGACAGTG3′, reverse 5′-AGTTG-CTTCTAGCCCAAAGAAC-3′; MMP3: forward 5′ACATGGAGACTTTGTCCCTTTTG-3′, reverse 5′-TTGGCTGAGTGGTAGAGTCCC-3′; tissue inhibitor or metalloproteinase (TIMP) 1: forward 5′-GCAACTCGGACCTGGT-CATAA-3′, reverse 5′-CGGCCCGTGATGAGAAACT-3′. The human primers used were MMP12: forward 5′GGAATCCTAGCCCATGCTTTT-3′, reverse 5′-CATTACGGCCTTTGGA-TCACT3′; TIMP1: forward 5′-ACCACCTTATACCAGCGTTATGA-3′, reverse 5′GGTGT-AGACGAACCGGATGTC-3′. 1.5. Statistical methods Data from experiments with two groups were compared using two-tailed t test. For experiments with more than two groups, multiple comparisons were made using either two-way analysis of variance (ANOVA) with post test by Tukey or one-way ANOVA with posttest by Bonferroni. P values less than .05 were considered significant. All results are expressed as mean±standard error. 2. Results The p22 phox-deficient mice were found to be approximately 10% smaller than their wild-type littermates (Table 1). There was no significant difference in the ages of the mice assessed (wild type: 9.9±
38
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
Fig. 4. Adventitial inflammation at the site of ligation in the carotid arteries. Shown are immunohistochemical stains for the neutrophil marker Ly-6B (A–D) and the macrophage marker F4/80 (F–I). The carotid arteries from the nonligated side show no significant inflammatory infiltrate in either wild-type (A, F) or p22phox-deficient mice (C, H). In contrast, at the site of ligation, both wild-type mice (B, G) and p22phox-deficient mice (D, I) display an inflammatory infiltrate in the adventitia of the carotid arteries. Scale bars represent 20 μm. Insets depict negative controls. (E) Quantitation of neutrophils. (J) Quantitation of macrophages. *Pb.01 versus the nonligated side, two-way ANOVA/Tukey, n=3 to 4 male mice per group.
0.2 weeks, p22 phox deficient: 10.0±0.3 weeks; P=.8). The heart weight was not significantly different in the p22 phox-deficient animals. Surprisingly, the p22 phox-deficient mice showed no alteration in basal blood pressure or pulse rate (Table 1). The wild-type mice were found to be relatively resistant to the formation of intimal hyperplasia in the carotid ligation model. In the proximal and mid segments of the common carotid artery, there was only minimal intimal hyperplasia in the wild-type mice, and this minimal intimal hyperplasia was not altered by p22 phox deficiency (Fig. 1). There was also no difference in the absolute medial area (wild type: 0.04±0.01 mm 2, p22 phox deficient: 0.04±0.01 mm 2; P=.7). However, at the site of ligation, the wild-type mice showed extensive loss of the vascular elastic fibers (Fig. 2). In contrast, the p22 phoxdeficient mice displayed significantly less vascular elastic fiber loss at the site of ligation. In the wild-type mice, the vascular elastic fiber loss
was associated with disorganization of the vascular smooth muscle cells (Fig. 2D). The area of elastic fiber loss was limited to a small region, ~100 μm in length, at the site of ligation. The vascular elastic fibers were intact away from the site of ligation in both groups (Fig. 1). By real-time PCR, ligation of the carotid arteries in the wild-type mice was associated with a marked increase in MMP12 mRNA levels and a more modest elevation in MMP3 mRNA levels (Fig. 3A and B). These responses were also seen in the p22 phox-deficient mice, with a nonsignificant trend towards a reduced magnitude compared with the wild-type mice. Both wild-type and p22 phox-deficient mice showed strong induction of TIMP1 in the ligated carotid arteries (Fig. 3C). The MMP12/TIMP1 expression ratio was markedly increased in the ligated arteries of the wild-type mice, and this expression ratio was significantly reduced (Pb.01) in the setting of p22 phox deficiency (Fig. 3D).
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
39
Fig. 5. Oxidative stress in the distal ligated carotid arteries. Immunohistochemical stains for the oxidative stress marker HNE in the distal ligated carotid arteries of wild-type (A) and p22phox-deficient (B) mice, with corresponding negative controls (C, D). Scale bars represent 20 μm. (E) Quantitation of the percentage of smooth muscle cell staining for HNE. *P= .008, n=3 male mice per group.
Both the wild-type mice and the p22 phox-deficient mice showed a reactive adventitial inflammatory infiltrate at the site of ligation (Fig. 4). There was no difference in the inflammatory infiltrate between the two groups based on immunohistochemical staining for the neutrophil marker Ly-6B and the macrophage marker F4/80. Previously, these p22 phox-deficient mice were shown to have impaired NADPH oxidase activity [13]. In the distal portions of the carotid arteries (~200 μm in length) just proximal to the site of ligation, there was extensive intimal hyperplasia that was not altered by p22 phox deficiency (Figs. 5 and 6). In the wild-type mice, the vascular smooth muscle cells in this region showed evidence of oxidative stress as evidenced by staining for HNE (Fig. 5). In the p22 phox-deficient mice, the degree of HNE staining was significantly
reduced, indicating decreased generation of ROS and consistent with impaired NADPH oxidase activity. Since H2O2 has been previously shown to activate CK1αLS [8,9] and hnRNP-C expression during vascular activation has been shown to be dependent on CK1αLS activity [12], segments of distal common carotid artery just proximal to the ligation site were assessed for hnRNP-C expression to ascertain if the reduced levels of ROS in the p22 phox-deficient mice may be associated with reduced CK1αLS activity. In the wild-type mice, the common carotid arteries in this location showed strong expression of hnRNP-C in vascular smooth muscle cells (Fig. 6). However, in the p22 phox-deficient mice, the expression of hnRNP-C in the carotid artery smooth muscle cells was significantly reduced, suggesting decreased CK1αLS activity in these
Fig. 6. Involvement of the CK1αLS/hnRNP-C pathway. In the distal carotid artery just proximal to the site of ligation, the wild-type mice (A) show strong expression of the vascular activation marker hnRNP-C in smooth muscle cells by immunohistochemistry. The p22phox-deficient mice (B) show reduced expression of hnRNP-C. Scale bars represent 20 μm. Insets depict negative controls. (C) Quantitation of vascular hnRNP-C expression. *P=.005, n=4 male mice per group. (D) Assessment of the relative MMP12/TIMP1 mRNA ratios in paired human coronary artery smooth muscle cell preparations with and without knockdown of CK1αLS. *Pb.05, n=3.
40
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
Fig. 7. hnRNP-C expression in other tissues. Both wild-type (A, C, E) and p22phox-deficient (B, D, F) male mice show strong expression of hnRNP-C in the liver (A, B), heart (C, D), and periaortic soft tissue (E, F right side) by immunohistochemistry. Neither group shows strong expression of hnRNP-C in the aorta (E, F left side). Scale bars represent 20 μm. Insets depict negative controls.
mice. To determine if impaired CK1αLS activity could account for the decreased MMP12/TIMP1 expression ratio seen in the p22 phoxdeficient mice, the MMP12/TIMP1 mRNA ratios were assessed in paired human coronary artery smooth muscle cell samples with and without knockdown of CK1αLS. Knockdown of CK1αLS was found to cause the relative MMP12/TIMP1 mRNA ratio to decrease to a similar extent as that seen in the p22 phox-deficient mice (Fig. 6D). The efficiency of CK1αLS knockdown in these samples was reported previously [12]. Although vascular hnRNP-C expression is induced during vascular activation, hnRNP-C is expressed ubiquitously at high levels in other tissues, such as the liver. Interestingly, p22 phox deficiency did not alter the high basal level of expression of hnRNP-C in the liver, heart, or periaortic soft tissue (Fig. 7). In contrast, hnRNP-C was not strongly expressed in the aorta, either in the presence or in the absence of p22 phox deficiency, as expected for a normal nonactivated artery. In humans, deficiency of NADPH oxidase activity results in a condition termed chronic granulomatous disease (CGD). CGD is most often an X-linked disease due to deficiency of Nox2 (gp91 phox). CGD resulting from deficiency of p22 phox is much less common. To determine if the decreased vascular hnRNP-C expression observed
in the p22 phox-deficient mice is relevant to humans, a rare lung biopsy of a case of human p22 phox deficiency was assessed for vascular hnRNP-C expression. In the lung biopsy from the p22 phox-deficient patient, there was markedly reduced expression of hnRNP-C in the vascular smooth muscle cells compared with the vascular smooth muscle cells in the lung biopsies from two age- and gender-matched control patients (Fig. 8). In contrast, the perivascular cells maintained hnRNP-C expression in the p22 phox-deficient patient, suggesting that the effects of p22 phox deficiency on hnRNP-C expression are relatively specific to the vasculature. 3. Discussion Vascular NADPH oxidase complexes may contain any one of four distinct catalytic subunits: Nox1, Nox2, Nox4, or Nox5 [4–6]. Vascular effects of NADPH oxidase subunit deletion have largely been assessed with deletion of the catalytic subunits Nox1 or Nox2, or the structural/ organizing subunit p47 phox. However, deletion of any one catalytic subunit may be compensated for in part by other catalytic subunits. In addition, loss of p47 phox, which facilitates activity of both Nox1 and Nox2, could be compensated for in part by an alternative homologous
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
41
Fig. 8. Vascular hnRNP-C expression in human patients with and without p22phox deficiency. Shown are histologic images of lung biopsies from a p22phox-deficient patient (A, B) and two control patients (C, D and E, F). Hematoxylin & eosin stained sections (A, C, E) show small arteries within the biopsies. By immunohistochemistry for hnRNP-C, the p22phoxdeficient patient shows strong expression of hnRNP-C in periarterial cells but not in the vascular smooth muscle cells (B). In contrast, the two control patients show strong expression for hnRNP-C in both periarterial cells and in vascular smooth muscle cells (D, F). Scale bars represent 40 μm. Insets depict negative controls. (G) Quantitation of vascular smooth muscle cell hnRNP-C staining. *Pb.0001, one-way ANOVA/Bonferroni, n=5 arteries per patient.
organizing subunit NoxO1. However, p22 phox is essential for the function of complexes containing Nox1, Nox2, Nox3, and Nox4. Thus, it was theorized that knockout of p22 phox may have more profound phenotypic changes than typically observed with knockout of other NADPH oxidase subunits. In fact, the p22 phox-deficient mice were observed to be smaller than their wild-type littermates. In contrast, knockout of Nox1 is apparently not associated with a decreased body weight [15–17]. Deletion of Nox2 has been reported to be associated with either an increased body weight [18], a decreased body weight [19], or no alteration in body weight [20]. Deletion of p47 phox was reported to not significantly alter body weight [21]. Thus, the decrease in body weight observed here may indicate more profound systemic effects with growth retardation due to deletion of p22 phox than occur with deletion of other NADPH oxidase components. Given that superoxide generated from NADPH oxidases can inactivate nitric oxide, it is possible that deletion of NADPH oxidase subunit components could result in lower blood pressure. However, in this study, the p22 phox-deficient mice had normal basal blood pressure. This observation is not inconsistent with prior studies. While it has been reported that knockout of Nox1 results in decreased basal blood pressure [16], multiple additional studies showed no change in basal blood pressure in the setting of Nox1 deficiency [15,17]. Multiple studies have reported decreased basal blood pressure in the setting of Nox2 deficiency [19,22,23], but one study reported no difference in basal blood pressure with Nox2 deficiency [24]. Deletion of p47 phox, which could impair the function of both
Nox1 and Nox2 containing NADPH oxidase complexes, appears to have no significant effect on basal blood pressure [25–27]. Furthermore, p22 phox deficiency would also impair the function of Nox4, and Nox4 activity may actually promote lower blood pressures [3]. Thus, loss of Nox4 function could potentially counteract any reduction in basal blood pressure due to loss of Nox1 and Nox2 activity. There was no effect of p22 phox deficiency on the formation of intimal hyperplasia in this model. However, the wild-type mice used in this study displayed only minimal intimal hyperplasia in the proximal and mid portions of the carotid artery after ligation, so the effects of p22 phox deficiency on intimal hyperplasia could not be reliably assessed with this model using this mouse strain. Deficiency of Nox1, Nox2, and p47 phox has each been shown to inhibit intimal hyperplasia in wire injury models [28–30]. However, Nox1 deficiency did not prevent intimal hyperplasia in the carotid ligation model in the setting of apolipoprotein E deficiency [15]. An important but somewhat unexpected observation was that the p22 phox-deficient mice did show markedly impaired elastic fiber loss at the site of the ligation of the carotid artery. Regulation of elastic fiber loss by NADPH oxidase complexes may account in part for some previous observations. Deletion of p47 phox was observed to decrease aneurysm size in an experimentally induced cerebral aneurysm model [25] and to decrease the size of aortic aneurysms induced by angiotensin II infusion in the setting of apolipoprotein E deficiency [31]. Furthermore, deletion of Nox1 was observed to decrease susceptibility to angiotensin-II-induced aortic dissection in mice [32].
42
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
Matrix metallopeptidase (MMP) activity is known to be regulated by ROS [1]. In fact, deletion of p47 phox has been reported to decrease aortic MMP2 activity [31], and deletion of Nox1 has been reported to increase aortic expression of the MMP inhibitor TIMP1 [32]. Recently, it was shown that knockdown of the hydrogen peroxide response kinase CK1αLS rendered human coronary artery smooth muscle cells insensitive to the proliferative effects of H2O2 and decreased MMP3 expression in both cultured smooth muscle cells and cultured human arteries [12]. In addition to regulating MMP activity, ROS may also promote elastic fiber degradation by direct oxidative modification of elastic fibers [33,34]. Here, we found that both p22 phox deficiency in vivo and knockdown of CK1αLS in vitro were associated with a decreased MMP12/TIMP1 expression ratio. Here p22 phox deficiency also resulted in decreased vascular expression of hnRNP-C in both mice and a human patient. It is not certain if this observation will extrapolate to all patients with p22 phox deficiency, but this is in agreement with a previous study showing that knockdown of the H2O2-responsive kinase CK1αLS resulted in decreased hnRNP-C expression in cultured human arteries [12] and suggests that p22 phox deficiency may be associated with impaired CK1αLS activity. Although hnRNP-C appears to have a fundamental role in sorting RNA polymerase II transcripts for export to the cytoplasm [35], hnRNP-C also specifically promotes the translation of transcripts that enhance cell growth and survival, including plateletderived growth factor B chain and c-myc [36,37]. While hnRNP-C is ubiquitously expressed at high levels in some tissues such as liver, in normal arteries, hnRNP-C is only expressed at low levels but is upregulated in activated disease states such as intimal hyperplasia and atherosclerosis [11]. The observations here with p22 phox-deficient mice and human tissue indicate that the up-regulation of hnRNP-C during vascular activation is regulated by ROS generated from NADPH oxidases. Interestingly, NADPH oxidase regulation of hnRNP-C expression appears to be relatively vascular specific, as there was no effect of p22 phox deficiency on the expression of hnRNP-C in other tissues such as heart and liver. In summary, deficiency of p22 phox in mice was associated with decreased body size, impaired vascular elastic fiber loss, and a reduced vascular MMP12/TIMP1 expression ratio after injury. In addition, deficiency of p22 phox resulted in decreased expression of the vascular activation marker hnRNP-C in both mice and a human patient, and knockdown of CK1αLS resulted in a similar decrease in the MMP12/ TIMP1 expression ratio in cultured cells. These studies suggest that vascular hnRNP-C expression is regulated by ROS derived from NADPH oxidases and that the effects of NADPH oxidase on vascular activation are mediated in part by protein kinase CK1αLS.
References [1] Stone JR, Yang S. Hydrogen peroxide: a signaling messenger. Antioxid Redox Signal 2006;8:243–70. [2] Schramm A, Matusik P, Osmenda G, Guzik TJ. Targeting NADPH oxidases in vascular pharmacology. Vasc Pharmacol 2012;56:216–31. [3] Sirker A, Zhang M, Shah AM. NADPH oxidases in cardiovascular disease: insights from in vivo models and clinical studies. Basic Res Cardiol 2011;106: 735–47. [4] Lassegue B, Martin AS, Griendling KK. Biochemistry, physiology, and pathophysiology of the NADPH oxidases in the cardiovascular system. Circ Res 2012;110: 1364–90. [5] Bedard K, Krause K-H. The NOX-family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiol Rev 2007;87:245–313. [6] Brandes RP, Weissmann N, Schroder K. NADPH oxidases in cardiovascular disease. Free Radic Biol Med 2010;49:687–706. [7] Stone JR, Collins T. Rapid phosphorylation of heterogeneous nuclear ribonucleoprotein C1/C2 in response to physiologic levels of hydrogen peroxide in human endothelial cells. J Biol Chem 2002;277:15621–8.
[8] Kattapuram T, Yang S, Maki JL, Stone JR. Protein kinase CK1α regulates mRNA binding by heterogeneous nuclear ribonucleoprotein C in response to physiologic levels of hydrogen peroxide. J Biol Chem 2005;280:15340–7. [9] Bedri S, Cizek SM, Rastarhuyeva I, Stone JR. Regulation of protein kinase CK1αLS by dephosphorylation in response to hydrogen peroxide. Arch Biochem Biophys 2007;466:242–9. [10] Kamma H, Portman DS, Dreyfuss G. Cell type-specific expression of hnRNP proteins. Exp Cell Res 1995;221:187–96. [11] Panchenko MP, Silva N, Stone JR. Up-regulation of a hydrogen peroxideresponsive pre-mRNA binding protein in atherosclerosis and intimal hyperplasia. Cardiovasc Pathol 2009;18:167–72. [12] Panchenko MP, Siddiquee Z, Dombkowski DM, et al. Protein kinase CK1αLS promotes vascular cell proliferation and intimal hyperplasia. Am J Pathol 2010;177:1562–72. [13] Nakano Y, Longo-Guess CM, Bergstrom BE, et al. Mutation of Cyba gene encoding p22phox causes vestibular and immune defects in mice. J Clin Invest 2008;118: 1176–85. [14] Harris JB, Michelow IC, Westra SJ, et al. Case records of the Massachusetts General Hospital. Case 21-2008: an 11-month-old boy with fever and pulmonary infiltrates. N Engl J Med 2008;359:178–87. [15] Sheehan AL, Carrell S, Johnson B, et al. Role for Nox1 NADPH oxidase in atherosclerosis. Atherosclerosis 2011;216:321–6. [16] Gavazzi G, Banfi B, Deffert C, et al. Decreased blood pressure in NOX1-deficient mice. FEBS Lett 2006;580:497–504. [17] Matsuno K, Yamada H, Iwata K, et al. Nox1 is involved in angiotensin II-mediated hypertension: a study in Nox1-deficient mice. Circulation 2005;112:2677–85. [18] Hayashi T, Yamashita C, Matsumoto C, et al. Role of gp91phox-containing NADPH oxidase in left ventricular remodeling induced by intermittent hypoxic stress. Am J Physiol Heart Circ Physiol 2008;294:H2197–203. [19] Wang HD, Xu S, Johns DG, et al. Role of NADPH oxidase in the vascular hypertrophic and oxidative stress response to angiotensin II in mice. Circ Res 2001;88:947–53. [20] Kirk EA, Dinauer MC, Rosen H, et al. Impaired superoxide production due to a deficiency in phagocyte NADPH oxidase fails to inhibit atherosclerosis in mice. Arterioscler Thromb Vasc Biol 2000;20:1529–35. [21] Jackson SH, Gallin JI, Holland SM. The p47phox mouse knock-out model of chronic granulomatous disease. J Exp Med 1995;182:751–8. [22] Byrne JA, Grieve DJ, Bendall JK, et al. Contrasting roles of NADPH oxidase isoforms in pressure-overload versus angiotensin II-induced cardiac hypertrophy. Circ Res 2003;93:802–4. [23] Bendall JK, Cave AC, Heymes C, et al. Pivotal role of a gp91phox-containing NADPH oxidase in angiotensin II–induced cardiac hypertrophy in mice. Circulation 2002;105:293–6. [24] Haque MZ, Majid DSA. Assessment of renal functional phenotype in mice lacking gp91phox subunit of NAD(P)H oxidase. Hypertension 2004;43:335–40. [25] Aoki T, Nishimura M, Kataoka H, et al. Reactive oxygen species modulate growth of cerebral aneurysms: a study using the free radical scavenger edaravone and p47phox−/− mice. Lab Invest 2009;89:730–41. [26] Landmesser U, Cai H, Dikalov S, et al. Role of p47phox in vascular oxidative stress and hypertension caused by angiotensin II. Hypertension 2002;40:511–5. [27] Hsich E, Segal BH, Pagano PJ, et al. Vascular effects following homozygous disruption of p47phox: an essential component of NADPH oxidase. Circulation 2000;101:1234–6. [28] Lee MY, Martin AS, Mehta PK, et al. Mechanisms of vascular smooth muscle NADPH oxidase 1 (Nox1) contribution to injury-induced neointimal formation. Arterioscler Thromb Vasc Biol 2009;29:480–7. [29] Vendrov AE, Hakim ZS, Madamanchi NR, et al. Atherosclerosis is attenuated by limiting superoxide generation in both macrophages and vessel wall cells. Arterioscler Thromb Vasc Biol 2007;27:2714–21. [30] Chen Z, Keaney JF, Schulz E, et al. Decreased neointimal formation in Nox2deficient mice reveals a direct role for NADPH oxidase in the response to arterial injury. Proc Natl Acad Sci USA 2004;101:13014–9. [31] Thomas M, Gavrila D, McCormick ML, et al. Deletion of p47phox attenuates angiotensin II-induced abdominal aortic aneurysm formation in apolipoprotein Edeficient mice. Circulation 2006;114:404–13. [32] Gavazzi G, Deffert C, Trocme C, et al. Nox1 deficiency protects from aortic dissection in response to angiotensin II. Hypertension 2007;50:189–96. [33] Akhtar K, Broekelmann TJ, Miao M, et al. Oxidative and nitrosative modifications of tropoelastin prevent elastic fiber assembly in vitro. J Biol Chem 2010;285: 37396–404. [34] Cantor JO, Shteyngart B, Cerreta JM, Ma S, Turino GM. Synergistic effect of hydrogen peroxide and elastase on elastic fiber injury in vitro. Exp Biol Med 2006;231:107–11. [35] McCloskey A, Taniguchi I, Shinmyozu K, Ohno M. hnRNP-C tetramer measures RNA length to classify RNA polymerase II transcripts for export. Science 2012;335:1643–6. [36] Sella O, Gerlitz G, Le S-Y, Elroy-Stein O. Differentiation-induced internal translation of c-sis mRNA: analysis of the cis elements and their differentiationlinked binding to the hnRNP-C protein. Mol Cell Biol 1999;19:5429–40. [37] Kim JH, Paek KY, Choi K, et al. Heterogeneous nuclear ribonucleoprotein C modulates translation of c-myc mRNA in a cell cycle phase-dependent manner. Mol Cell Biol 2003;23:708–20.
Contents lists available at ScienceDirect
Cardiovascular Pathology
Original Article
Altered vascular activation due to deficiency of the NADPH oxidase component p22 phox He Wang a, Hassan Albadawi b, Zakir Siddiquee a, Jillian M. Stone a, Mikhail P. Panchenko a, Michael T. Watkins b, James R. Stone a,⁎ a b
Center for Systems Biology, Massachusetts General Hospital and Department of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA Department of Surgery, Division of Vascular and Endovascular Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA
a r t i c l e
i n f o
Article history: Received 10 July 2013 Received in revised form 4 August 2013 Accepted 5 August 2013 Keywords: Hydrogen peroxide HnRNP-C MMP12 NADPH oxidase Protein kinase CK1α Protein kinase CK1αLS Reactive oxygen species TIMP1 Vascular activation Vascular injury
a b s t r a c t Background: Reactive oxygen species generated by nicotinamide adenine dinucleotide phosphate (NADPH) oxidase play important roles in vascular activation. The p22 phox subunit is necessary for the activity of NADPH oxidase complexes utilizing Nox1, Nox2, Nox3, and Nox4 catalytic subunits. Methods: We assessed p22 phox-deficient mice and human tissue for altered vascular activation. Results: Mice deficient in p22 phox were smaller than their wild-type littermates but showed no alteration in basal blood pressure. The wild-type littermates were relatively resistant to forming intimal hyperplasia following carotid ligation, and the intimal hyperplasia that developed was not altered by p22phox deficiency. However, at the site of carotid artery ligation, the p22 phox-deficient mice showed significantly less vascular elastic fiber loss compared with their wild-type littermates. This preservation of elastic fibers was associated with a reduced matrix metallopeptidase (MMP) 12/tissue inhibitor of metalloproteinase (TIMP) 1 expression ratio. A similar decrease in the relative MMP12/TIMP1 expression ratio occurred in human coronary artery smooth muscle cells upon knockdown of the hydrogen peroxide responsive kinase CK1αLS. In the ligated carotid arteries, the p22 phox-deficient mice showed reduced expression of heterogeneous nuclear ribonucleoprotein C (hnRNP-C), suggesting reduced activity of CK1αLS. In a lung biopsy from a human patient with p22 phox deficiency, there was also reduced vascular hnRNP-C expression. Conclusions: These findings indicate that NADPH oxidase complexes modulate aspects of vascular activation including vascular elastic fiber loss, the MMP12/TIMP1 expression ratio, and the expression of hnRNP-C. Furthermore, these findings suggest that the effects of NADPH oxidase on vascular activation are mediated in part by protein kinase CK1αLS. © 2014 Elsevier Inc. All rights reserved.
Within the vessel wall, reactive oxygen species (ROS) including superoxide and hydrogen peroxide (H2O2) play important roles in regulating blood pressure and in mediating vascular cell activation due to stimuli such as altered flow and vascular injury [1–3]. The nicotinamide adenine dinucleotide phosphate (NADPH) oxidase complexes are a principal source of ROS in the vessel wall [4–6]. In mammals, these complexes contain one of seven distinct catalytic subunits: Nox1–Nox5, Duox1, and Duox2, although only Nox1, Nox2/ gp91 phox, Nox4, and Nox5 are thought to be expressed in the vessel wall. Of these, Nox1, Nox2, and Nox4 all utilize a structural subunit, p22 phox, which is also required for the function of Nox3. Nox1–Nox3 Abbreviations: hnRNP-C, heterogeneous nuclear ribonucleoprotein C; MMP3, matrix metallopeptidase 3; MMP12, matrix metallopeptidase 12; ROS, reactive oxygen species; TIMP1, tissue inhibitor of metalloproteinase 1. Funding: This work was funded by the Department of Pathology and the Department of Surgery, Division of Vascular and Endovascular Surgery, Massachusetts General Hospital. ⁎ Corresponding author. Massachusetts General Hospital, Boston, MA 02114. Tel.: +1 617 726 8303; fax: +1 617 643 3566. E-mail address: [email protected] (J.R. Stone). 1054-8807/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.carpath.2013.08.003
are also regulated by either of two regulatory subunits, p67 phox and NoxA1, and also by either of two structural/organizing subunits, p47 phox and NoxO1. Different types and concentrations of ROS may impact multiple signaling pathways in cells; for example, low concentrations of H2O2 modulate the function of a nuclear pre-mRNA binding protein, heterogeneous nuclear ribonucleoprotein C (hnRNP-C) [7,8]. This modulation occurs through phosphorylation following activation of Table 1 Baseline measurements on p22phox-deficient mice Parameter assessed
Wild type
p22phox deficient
P
Body weight (g)a Heart weight (mg)b Heart weight/body weight (mg/g)b Systolic blood pressure (mmHg)c Pulse (beats/min)c
20.5±0.4 149±7 7.4±0.4 122±4 592±16
18.0±0.4 140±10 7.7±0.5 126±4 599±13
.0002 .51 .61 .53 .72
a b c
n=26 mice (9 males and 17 females) in each group. n=15 mice (4 males and 11 females) in each group. n=12 mice (6 males and 6 females) in each group.
36
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
1. Materials and methods 1.1. Mice
Fig. 1. Intimal hyperplasia in the proximal and mid carotid artery following ligation. Left common carotid artery after ligation in wild-type (A) and p22phox-deficient (B) mice. Scale bars represent 20 μm. (C) The wild-type mice were relatively resistant to the formation of intimal hyperplasia with no effect due to p22phox deficiency. n=6– 8 male mice per group.
the nuclear protein kinase CK1αLS [8,9]. While hnRNP-C is ubiquitously expressed in some tissues such as liver [10], in arteries, hnRNPC is only highly expressed in activated states such as intimal hyperplasia and atherosclerosis [11]. Knockdown of CK1αLS was found to render human coronary artery smooth muscle cells insensitive to the proliferative effects of exogenous H2O2 and to prevent the up-regulation of hnRNP-C in cultured human arteries [12]. This suggested that the expression of hnRNP-C in arteries may be under the regulation of ROS. Recently, the nmf333 mouse strain with a head-tilt phenotype was identified as being deficient in p22 phox due to a point mutation in the corresponding Cyba gene [13]. Given the requirement of p22 phox for the function of vascular Nox1, Nox2, and Nox4, we hypothesized that p22 phox deficiency could have significant effects on vascular activation and vascular hnRNP-C expression in these mice.
B6 Tyr+/−Cyba(nmf333)/J mice carrying inactivating mutations in the Cyba gene [13] encoding p22 phox and wild-type B6 Tyr+/J littermate controls were obtained from The Jackson Laboratory (Bar Harbor, ME, USA) at 8–12 weeks of age. Blood pressures and pulse rates were determined using a noninvasive tail cuff and pulse transducer system (BP2000 Blood Pressure Analysis System, Visitech Systems, Apex, NC, USA). At least eight separate measurements were obtained for each mouse over the period of 1 week, and the values were averaged. Blood pressure measurements were taken at the same time each day between 9:00 and 11:00 a.m. Mice were euthanized and assessed for body weight and heart weight. Tissues from the heart, liver, and aorta were fixed in formalin and embedded in paraffin. Male mice at 8–9 weeks of age were subjected to left common carotid artery ligation under 40–50 mg/kg pentobarbital anesthesia. Mice were placed on a heated pad to maintain a body temperature of 37°C. Once anesthetized, the neck and chest were shaved, and the skin was cleansed with Betadine Solution. A longitudinal neck incision was made, and the left common carotid artery was exposed. The distal aspect of the left common carotid artery was encircled with fine forceps and permanently ligated just prior to the bifurcation using 80 silk suture. The incision was closed, and the mice were allowed to recover. After 4 weeks, the mice were euthanized, and the vessels were harvested following perfusion fixation with 10% buffered formalin through the left ventricle and embedded in paraffin. In other experiments, carotid arteries from male and female mice were harvested fresh, 1 week after ligation, and snap frozen in liquid nitrogen. The mouse studies were approved by the Subcommittee on Research Animal Care at Massachusetts General Hospital. 1.2. Human cells and tissue Paraffin blocks of discarded lung tissue were obtained from one child with p22 phox deficiency and two control subjects matched for age and gender. All subjects were boys at approximately 1 year of age who had undergone pulmonary wedge biopsy. The indications for biopsy were pneumonia in the p22 phox-deficient patient, and
Fig. 2. Elastic fiber loss at the site of ligation in the carotid arteries. Sites of ligation in the left common carotid artery in wild-type mice (A) showing marked loss of vascular elastic fibers. In contrast, the p22phox-deficient mice (B) show preservation of the vascular elastic fibers. (C) Quantitation of vascular elastic fiber loss. *P=.005, n=5 to 6 male mice per group. At higher magnification, the vascular smooth muscle cells show disorganization in the wild-type animals at the sites of elastic fiber loss (D) compared with the smooth muscle cells in the p22phox-deficient mice (E). Scale bars represent 40 μm (A, B) or 20 μm (D, E).
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
37
for the interior luminal area, the internal elastic lamina, and the external elastic lamina. The ratio between the intimal to the medial areas was calculated for each section, and the ratios were averaged for all the sections in each vessel, excluding the site of the ligation itself in the distal portion of the artery. The amount of elastic fiber loss at the sites of ligation was obtained using ImageJ by measuring the circumferential arc lengths of the regions with preserved and absent elastic fibers using the Freehand Line tool and was expressed as percent of circumference. Immunohistochemistry was performed using the peroxidaseconjugated avidin–biotin method (Vector) following antigen retrieval with either Borg decloaker solution (Biocare Medical) or Tris/Borate EDTA cell conditioning solution CC1 (Ventana Medical Inc.). Mouse tissue was stained for hnRNP-C using goat polyclonal antibodies (Santa Cruz) at 1:200 dilution and rabbit polyclonal antibodies (Santa Cruz) at 1:50 dilution. Human tissue was stained for hnRNP-C using mouse monoclonal anti-hnRNP-C antibody (clone 4F4, Santa Cruz) at 1:200 dilution. After blocking with Rodent Block M (Biocare Medical), mouse tissue was stained for the neutrophil marker Ly-6B using rat monoclonal antibodies (Abd Serotec) at 1:1000 dilution, for the macrophage marker F4/80 using rat monoclonal antibodies (Abcam) at 1:300 dilution, and for the oxidative stress marker 4-hydroxy-2nonenal (HNE) using mouse monoclonal antibodies (Abcam) at 1:100 dilution. Negative controls were performed by omitting the primary antibody. 1.4. Real-time polymerase chain reaction (PCR)
Fig. 3. Changes in the expression of MMP12, MMP3, and TIMP1. One week following carotid artery ligation, mRNA levels were assessed in full-length ligated and contralateral nonligated carotid arteries of both male and female mice by real-time PCR. (A) MMP12, (B) MMP3, (C) TIMP1, (D) MMP12/TIMP1 ratio. #Pb.05 versus the nonligated side, *Pb.01 versus the nonligated side, **Pb.01 versus the nonligated side, and Pb.01 versus the ligated p22phox-deficient group. Two-way ANOVA/Tukey, n=4 to 6 mice per group.
interstitial lung disease and benign neoplasia in the control patients. The clinical and genetic details of the patient with p22 phox deficiency have been reported previously [14]. The research with human tissue was approved by the Human Subjects Institutional Review Board at Massachusetts General Hospital. Paired sets of human coronary artery smooth muscle cells infected with retrovirus either containing a control vector or expressing a short hairpin RNA targeting the L-insert of CK1αLS were prepared and analyzed for CK1αLS knockdown efficiency previously [12]. 1.3. Histology and immunohistochemistry Portions of mouse carotid arteries, ~2–3 mm in length, were serial sectioned with four to six 7-μm sections per slide. Every 10th slide (or one slide per ~300 μM) was stained with the modified Verhoeff elastic–Van Gieson stain (Electron Microscopy Sciences, Hatfield, PA, USA) to assess for intimal hyperplasia in the proximal and mid portions of the artery. Intervening slides were utilized for additional modified Verhoeff elastic-Van Gieson stains and for immunohistochemical stains in the distal portion of the arteries. Image analysis software (Diagnostic Instrument, Inc.) was used to calculate the area
Total RNA was extracted from the frozen mouse carotid arteries with TRIzol Reagent (Ambion, Carlsbad, CA, USA) and additionally purified using an Arcturus PicoPure RNA Isolation Kit (Applied Biosystems, Foster City, CA, USA). Total RNA was isolated from human coronary artery smooth muscle cells as described previously [12]. Total RNA (111.2 ng for mouse carotid arteries and 100.75 ng for human smooth muscle cells) was reverse-transcribed in a final volume of 20 μl using mixture of oligo(dT) and random primers with a SuperScript First-Strand Synthesis SuperMix for qRT-PCR (Invitrogen). Real-time PCR (45 cycles) was performed in a final volume of 25 μl containing 12.5 μl of the SYBR Green 2× Master Mix (Applied Biosystems), 5 μl of the 1:10-diluted reverse transcription reaction, and 7.5 μl of primer pairs (1 μM each) using a 7500 Fast Real-Time PCR System (Applied Biosystems). The mouse primers used were matrix metallopeptidase (MMP) 12: forward 5′-CTGCTCCCATGAATGACAGTG3′, reverse 5′-AGTTG-CTTCTAGCCCAAAGAAC-3′; MMP3: forward 5′ACATGGAGACTTTGTCCCTTTTG-3′, reverse 5′-TTGGCTGAGTGGTAGAGTCCC-3′; tissue inhibitor or metalloproteinase (TIMP) 1: forward 5′-GCAACTCGGACCTGGT-CATAA-3′, reverse 5′-CGGCCCGTGATGAGAAACT-3′. The human primers used were MMP12: forward 5′GGAATCCTAGCCCATGCTTTT-3′, reverse 5′-CATTACGGCCTTTGGA-TCACT3′; TIMP1: forward 5′-ACCACCTTATACCAGCGTTATGA-3′, reverse 5′GGTGT-AGACGAACCGGATGTC-3′. 1.5. Statistical methods Data from experiments with two groups were compared using two-tailed t test. For experiments with more than two groups, multiple comparisons were made using either two-way analysis of variance (ANOVA) with post test by Tukey or one-way ANOVA with posttest by Bonferroni. P values less than .05 were considered significant. All results are expressed as mean±standard error. 2. Results The p22 phox-deficient mice were found to be approximately 10% smaller than their wild-type littermates (Table 1). There was no significant difference in the ages of the mice assessed (wild type: 9.9±
38
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
Fig. 4. Adventitial inflammation at the site of ligation in the carotid arteries. Shown are immunohistochemical stains for the neutrophil marker Ly-6B (A–D) and the macrophage marker F4/80 (F–I). The carotid arteries from the nonligated side show no significant inflammatory infiltrate in either wild-type (A, F) or p22phox-deficient mice (C, H). In contrast, at the site of ligation, both wild-type mice (B, G) and p22phox-deficient mice (D, I) display an inflammatory infiltrate in the adventitia of the carotid arteries. Scale bars represent 20 μm. Insets depict negative controls. (E) Quantitation of neutrophils. (J) Quantitation of macrophages. *Pb.01 versus the nonligated side, two-way ANOVA/Tukey, n=3 to 4 male mice per group.
0.2 weeks, p22 phox deficient: 10.0±0.3 weeks; P=.8). The heart weight was not significantly different in the p22 phox-deficient animals. Surprisingly, the p22 phox-deficient mice showed no alteration in basal blood pressure or pulse rate (Table 1). The wild-type mice were found to be relatively resistant to the formation of intimal hyperplasia in the carotid ligation model. In the proximal and mid segments of the common carotid artery, there was only minimal intimal hyperplasia in the wild-type mice, and this minimal intimal hyperplasia was not altered by p22 phox deficiency (Fig. 1). There was also no difference in the absolute medial area (wild type: 0.04±0.01 mm 2, p22 phox deficient: 0.04±0.01 mm 2; P=.7). However, at the site of ligation, the wild-type mice showed extensive loss of the vascular elastic fibers (Fig. 2). In contrast, the p22 phoxdeficient mice displayed significantly less vascular elastic fiber loss at the site of ligation. In the wild-type mice, the vascular elastic fiber loss
was associated with disorganization of the vascular smooth muscle cells (Fig. 2D). The area of elastic fiber loss was limited to a small region, ~100 μm in length, at the site of ligation. The vascular elastic fibers were intact away from the site of ligation in both groups (Fig. 1). By real-time PCR, ligation of the carotid arteries in the wild-type mice was associated with a marked increase in MMP12 mRNA levels and a more modest elevation in MMP3 mRNA levels (Fig. 3A and B). These responses were also seen in the p22 phox-deficient mice, with a nonsignificant trend towards a reduced magnitude compared with the wild-type mice. Both wild-type and p22 phox-deficient mice showed strong induction of TIMP1 in the ligated carotid arteries (Fig. 3C). The MMP12/TIMP1 expression ratio was markedly increased in the ligated arteries of the wild-type mice, and this expression ratio was significantly reduced (Pb.01) in the setting of p22 phox deficiency (Fig. 3D).
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
39
Fig. 5. Oxidative stress in the distal ligated carotid arteries. Immunohistochemical stains for the oxidative stress marker HNE in the distal ligated carotid arteries of wild-type (A) and p22phox-deficient (B) mice, with corresponding negative controls (C, D). Scale bars represent 20 μm. (E) Quantitation of the percentage of smooth muscle cell staining for HNE. *P= .008, n=3 male mice per group.
Both the wild-type mice and the p22 phox-deficient mice showed a reactive adventitial inflammatory infiltrate at the site of ligation (Fig. 4). There was no difference in the inflammatory infiltrate between the two groups based on immunohistochemical staining for the neutrophil marker Ly-6B and the macrophage marker F4/80. Previously, these p22 phox-deficient mice were shown to have impaired NADPH oxidase activity [13]. In the distal portions of the carotid arteries (~200 μm in length) just proximal to the site of ligation, there was extensive intimal hyperplasia that was not altered by p22 phox deficiency (Figs. 5 and 6). In the wild-type mice, the vascular smooth muscle cells in this region showed evidence of oxidative stress as evidenced by staining for HNE (Fig. 5). In the p22 phox-deficient mice, the degree of HNE staining was significantly
reduced, indicating decreased generation of ROS and consistent with impaired NADPH oxidase activity. Since H2O2 has been previously shown to activate CK1αLS [8,9] and hnRNP-C expression during vascular activation has been shown to be dependent on CK1αLS activity [12], segments of distal common carotid artery just proximal to the ligation site were assessed for hnRNP-C expression to ascertain if the reduced levels of ROS in the p22 phox-deficient mice may be associated with reduced CK1αLS activity. In the wild-type mice, the common carotid arteries in this location showed strong expression of hnRNP-C in vascular smooth muscle cells (Fig. 6). However, in the p22 phox-deficient mice, the expression of hnRNP-C in the carotid artery smooth muscle cells was significantly reduced, suggesting decreased CK1αLS activity in these
Fig. 6. Involvement of the CK1αLS/hnRNP-C pathway. In the distal carotid artery just proximal to the site of ligation, the wild-type mice (A) show strong expression of the vascular activation marker hnRNP-C in smooth muscle cells by immunohistochemistry. The p22phox-deficient mice (B) show reduced expression of hnRNP-C. Scale bars represent 20 μm. Insets depict negative controls. (C) Quantitation of vascular hnRNP-C expression. *P=.005, n=4 male mice per group. (D) Assessment of the relative MMP12/TIMP1 mRNA ratios in paired human coronary artery smooth muscle cell preparations with and without knockdown of CK1αLS. *Pb.05, n=3.
40
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
Fig. 7. hnRNP-C expression in other tissues. Both wild-type (A, C, E) and p22phox-deficient (B, D, F) male mice show strong expression of hnRNP-C in the liver (A, B), heart (C, D), and periaortic soft tissue (E, F right side) by immunohistochemistry. Neither group shows strong expression of hnRNP-C in the aorta (E, F left side). Scale bars represent 20 μm. Insets depict negative controls.
mice. To determine if impaired CK1αLS activity could account for the decreased MMP12/TIMP1 expression ratio seen in the p22 phoxdeficient mice, the MMP12/TIMP1 mRNA ratios were assessed in paired human coronary artery smooth muscle cell samples with and without knockdown of CK1αLS. Knockdown of CK1αLS was found to cause the relative MMP12/TIMP1 mRNA ratio to decrease to a similar extent as that seen in the p22 phox-deficient mice (Fig. 6D). The efficiency of CK1αLS knockdown in these samples was reported previously [12]. Although vascular hnRNP-C expression is induced during vascular activation, hnRNP-C is expressed ubiquitously at high levels in other tissues, such as the liver. Interestingly, p22 phox deficiency did not alter the high basal level of expression of hnRNP-C in the liver, heart, or periaortic soft tissue (Fig. 7). In contrast, hnRNP-C was not strongly expressed in the aorta, either in the presence or in the absence of p22 phox deficiency, as expected for a normal nonactivated artery. In humans, deficiency of NADPH oxidase activity results in a condition termed chronic granulomatous disease (CGD). CGD is most often an X-linked disease due to deficiency of Nox2 (gp91 phox). CGD resulting from deficiency of p22 phox is much less common. To determine if the decreased vascular hnRNP-C expression observed
in the p22 phox-deficient mice is relevant to humans, a rare lung biopsy of a case of human p22 phox deficiency was assessed for vascular hnRNP-C expression. In the lung biopsy from the p22 phox-deficient patient, there was markedly reduced expression of hnRNP-C in the vascular smooth muscle cells compared with the vascular smooth muscle cells in the lung biopsies from two age- and gender-matched control patients (Fig. 8). In contrast, the perivascular cells maintained hnRNP-C expression in the p22 phox-deficient patient, suggesting that the effects of p22 phox deficiency on hnRNP-C expression are relatively specific to the vasculature. 3. Discussion Vascular NADPH oxidase complexes may contain any one of four distinct catalytic subunits: Nox1, Nox2, Nox4, or Nox5 [4–6]. Vascular effects of NADPH oxidase subunit deletion have largely been assessed with deletion of the catalytic subunits Nox1 or Nox2, or the structural/ organizing subunit p47 phox. However, deletion of any one catalytic subunit may be compensated for in part by other catalytic subunits. In addition, loss of p47 phox, which facilitates activity of both Nox1 and Nox2, could be compensated for in part by an alternative homologous
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
41
Fig. 8. Vascular hnRNP-C expression in human patients with and without p22phox deficiency. Shown are histologic images of lung biopsies from a p22phox-deficient patient (A, B) and two control patients (C, D and E, F). Hematoxylin & eosin stained sections (A, C, E) show small arteries within the biopsies. By immunohistochemistry for hnRNP-C, the p22phoxdeficient patient shows strong expression of hnRNP-C in periarterial cells but not in the vascular smooth muscle cells (B). In contrast, the two control patients show strong expression for hnRNP-C in both periarterial cells and in vascular smooth muscle cells (D, F). Scale bars represent 40 μm. Insets depict negative controls. (G) Quantitation of vascular smooth muscle cell hnRNP-C staining. *Pb.0001, one-way ANOVA/Bonferroni, n=5 arteries per patient.
organizing subunit NoxO1. However, p22 phox is essential for the function of complexes containing Nox1, Nox2, Nox3, and Nox4. Thus, it was theorized that knockout of p22 phox may have more profound phenotypic changes than typically observed with knockout of other NADPH oxidase subunits. In fact, the p22 phox-deficient mice were observed to be smaller than their wild-type littermates. In contrast, knockout of Nox1 is apparently not associated with a decreased body weight [15–17]. Deletion of Nox2 has been reported to be associated with either an increased body weight [18], a decreased body weight [19], or no alteration in body weight [20]. Deletion of p47 phox was reported to not significantly alter body weight [21]. Thus, the decrease in body weight observed here may indicate more profound systemic effects with growth retardation due to deletion of p22 phox than occur with deletion of other NADPH oxidase components. Given that superoxide generated from NADPH oxidases can inactivate nitric oxide, it is possible that deletion of NADPH oxidase subunit components could result in lower blood pressure. However, in this study, the p22 phox-deficient mice had normal basal blood pressure. This observation is not inconsistent with prior studies. While it has been reported that knockout of Nox1 results in decreased basal blood pressure [16], multiple additional studies showed no change in basal blood pressure in the setting of Nox1 deficiency [15,17]. Multiple studies have reported decreased basal blood pressure in the setting of Nox2 deficiency [19,22,23], but one study reported no difference in basal blood pressure with Nox2 deficiency [24]. Deletion of p47 phox, which could impair the function of both
Nox1 and Nox2 containing NADPH oxidase complexes, appears to have no significant effect on basal blood pressure [25–27]. Furthermore, p22 phox deficiency would also impair the function of Nox4, and Nox4 activity may actually promote lower blood pressures [3]. Thus, loss of Nox4 function could potentially counteract any reduction in basal blood pressure due to loss of Nox1 and Nox2 activity. There was no effect of p22 phox deficiency on the formation of intimal hyperplasia in this model. However, the wild-type mice used in this study displayed only minimal intimal hyperplasia in the proximal and mid portions of the carotid artery after ligation, so the effects of p22 phox deficiency on intimal hyperplasia could not be reliably assessed with this model using this mouse strain. Deficiency of Nox1, Nox2, and p47 phox has each been shown to inhibit intimal hyperplasia in wire injury models [28–30]. However, Nox1 deficiency did not prevent intimal hyperplasia in the carotid ligation model in the setting of apolipoprotein E deficiency [15]. An important but somewhat unexpected observation was that the p22 phox-deficient mice did show markedly impaired elastic fiber loss at the site of the ligation of the carotid artery. Regulation of elastic fiber loss by NADPH oxidase complexes may account in part for some previous observations. Deletion of p47 phox was observed to decrease aneurysm size in an experimentally induced cerebral aneurysm model [25] and to decrease the size of aortic aneurysms induced by angiotensin II infusion in the setting of apolipoprotein E deficiency [31]. Furthermore, deletion of Nox1 was observed to decrease susceptibility to angiotensin-II-induced aortic dissection in mice [32].
42
H. Wang et al. / Cardiovascular Pathology 23 (2014) 35–42
Matrix metallopeptidase (MMP) activity is known to be regulated by ROS [1]. In fact, deletion of p47 phox has been reported to decrease aortic MMP2 activity [31], and deletion of Nox1 has been reported to increase aortic expression of the MMP inhibitor TIMP1 [32]. Recently, it was shown that knockdown of the hydrogen peroxide response kinase CK1αLS rendered human coronary artery smooth muscle cells insensitive to the proliferative effects of H2O2 and decreased MMP3 expression in both cultured smooth muscle cells and cultured human arteries [12]. In addition to regulating MMP activity, ROS may also promote elastic fiber degradation by direct oxidative modification of elastic fibers [33,34]. Here, we found that both p22 phox deficiency in vivo and knockdown of CK1αLS in vitro were associated with a decreased MMP12/TIMP1 expression ratio. Here p22 phox deficiency also resulted in decreased vascular expression of hnRNP-C in both mice and a human patient. It is not certain if this observation will extrapolate to all patients with p22 phox deficiency, but this is in agreement with a previous study showing that knockdown of the H2O2-responsive kinase CK1αLS resulted in decreased hnRNP-C expression in cultured human arteries [12] and suggests that p22 phox deficiency may be associated with impaired CK1αLS activity. Although hnRNP-C appears to have a fundamental role in sorting RNA polymerase II transcripts for export to the cytoplasm [35], hnRNP-C also specifically promotes the translation of transcripts that enhance cell growth and survival, including plateletderived growth factor B chain and c-myc [36,37]. While hnRNP-C is ubiquitously expressed at high levels in some tissues such as liver, in normal arteries, hnRNP-C is only expressed at low levels but is upregulated in activated disease states such as intimal hyperplasia and atherosclerosis [11]. The observations here with p22 phox-deficient mice and human tissue indicate that the up-regulation of hnRNP-C during vascular activation is regulated by ROS generated from NADPH oxidases. Interestingly, NADPH oxidase regulation of hnRNP-C expression appears to be relatively vascular specific, as there was no effect of p22 phox deficiency on the expression of hnRNP-C in other tissues such as heart and liver. In summary, deficiency of p22 phox in mice was associated with decreased body size, impaired vascular elastic fiber loss, and a reduced vascular MMP12/TIMP1 expression ratio after injury. In addition, deficiency of p22 phox resulted in decreased expression of the vascular activation marker hnRNP-C in both mice and a human patient, and knockdown of CK1αLS resulted in a similar decrease in the MMP12/ TIMP1 expression ratio in cultured cells. These studies suggest that vascular hnRNP-C expression is regulated by ROS derived from NADPH oxidases and that the effects of NADPH oxidase on vascular activation are mediated in part by protein kinase CK1αLS.
References [1] Stone JR, Yang S. Hydrogen peroxide: a signaling messenger. Antioxid Redox Signal 2006;8:243–70. [2] Schramm A, Matusik P, Osmenda G, Guzik TJ. Targeting NADPH oxidases in vascular pharmacology. Vasc Pharmacol 2012;56:216–31. [3] Sirker A, Zhang M, Shah AM. NADPH oxidases in cardiovascular disease: insights from in vivo models and clinical studies. Basic Res Cardiol 2011;106: 735–47. [4] Lassegue B, Martin AS, Griendling KK. Biochemistry, physiology, and pathophysiology of the NADPH oxidases in the cardiovascular system. Circ Res 2012;110: 1364–90. [5] Bedard K, Krause K-H. The NOX-family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiol Rev 2007;87:245–313. [6] Brandes RP, Weissmann N, Schroder K. NADPH oxidases in cardiovascular disease. Free Radic Biol Med 2010;49:687–706. [7] Stone JR, Collins T. Rapid phosphorylation of heterogeneous nuclear ribonucleoprotein C1/C2 in response to physiologic levels of hydrogen peroxide in human endothelial cells. J Biol Chem 2002;277:15621–8.
[8] Kattapuram T, Yang S, Maki JL, Stone JR. Protein kinase CK1α regulates mRNA binding by heterogeneous nuclear ribonucleoprotein C in response to physiologic levels of hydrogen peroxide. J Biol Chem 2005;280:15340–7. [9] Bedri S, Cizek SM, Rastarhuyeva I, Stone JR. Regulation of protein kinase CK1αLS by dephosphorylation in response to hydrogen peroxide. Arch Biochem Biophys 2007;466:242–9. [10] Kamma H, Portman DS, Dreyfuss G. Cell type-specific expression of hnRNP proteins. Exp Cell Res 1995;221:187–96. [11] Panchenko MP, Silva N, Stone JR. Up-regulation of a hydrogen peroxideresponsive pre-mRNA binding protein in atherosclerosis and intimal hyperplasia. Cardiovasc Pathol 2009;18:167–72. [12] Panchenko MP, Siddiquee Z, Dombkowski DM, et al. Protein kinase CK1αLS promotes vascular cell proliferation and intimal hyperplasia. Am J Pathol 2010;177:1562–72. [13] Nakano Y, Longo-Guess CM, Bergstrom BE, et al. Mutation of Cyba gene encoding p22phox causes vestibular and immune defects in mice. J Clin Invest 2008;118: 1176–85. [14] Harris JB, Michelow IC, Westra SJ, et al. Case records of the Massachusetts General Hospital. Case 21-2008: an 11-month-old boy with fever and pulmonary infiltrates. N Engl J Med 2008;359:178–87. [15] Sheehan AL, Carrell S, Johnson B, et al. Role for Nox1 NADPH oxidase in atherosclerosis. Atherosclerosis 2011;216:321–6. [16] Gavazzi G, Banfi B, Deffert C, et al. Decreased blood pressure in NOX1-deficient mice. FEBS Lett 2006;580:497–504. [17] Matsuno K, Yamada H, Iwata K, et al. Nox1 is involved in angiotensin II-mediated hypertension: a study in Nox1-deficient mice. Circulation 2005;112:2677–85. [18] Hayashi T, Yamashita C, Matsumoto C, et al. Role of gp91phox-containing NADPH oxidase in left ventricular remodeling induced by intermittent hypoxic stress. Am J Physiol Heart Circ Physiol 2008;294:H2197–203. [19] Wang HD, Xu S, Johns DG, et al. Role of NADPH oxidase in the vascular hypertrophic and oxidative stress response to angiotensin II in mice. Circ Res 2001;88:947–53. [20] Kirk EA, Dinauer MC, Rosen H, et al. Impaired superoxide production due to a deficiency in phagocyte NADPH oxidase fails to inhibit atherosclerosis in mice. Arterioscler Thromb Vasc Biol 2000;20:1529–35. [21] Jackson SH, Gallin JI, Holland SM. The p47phox mouse knock-out model of chronic granulomatous disease. J Exp Med 1995;182:751–8. [22] Byrne JA, Grieve DJ, Bendall JK, et al. Contrasting roles of NADPH oxidase isoforms in pressure-overload versus angiotensin II-induced cardiac hypertrophy. Circ Res 2003;93:802–4. [23] Bendall JK, Cave AC, Heymes C, et al. Pivotal role of a gp91phox-containing NADPH oxidase in angiotensin II–induced cardiac hypertrophy in mice. Circulation 2002;105:293–6. [24] Haque MZ, Majid DSA. Assessment of renal functional phenotype in mice lacking gp91phox subunit of NAD(P)H oxidase. Hypertension 2004;43:335–40. [25] Aoki T, Nishimura M, Kataoka H, et al. Reactive oxygen species modulate growth of cerebral aneurysms: a study using the free radical scavenger edaravone and p47phox−/− mice. Lab Invest 2009;89:730–41. [26] Landmesser U, Cai H, Dikalov S, et al. Role of p47phox in vascular oxidative stress and hypertension caused by angiotensin II. Hypertension 2002;40:511–5. [27] Hsich E, Segal BH, Pagano PJ, et al. Vascular effects following homozygous disruption of p47phox: an essential component of NADPH oxidase. Circulation 2000;101:1234–6. [28] Lee MY, Martin AS, Mehta PK, et al. Mechanisms of vascular smooth muscle NADPH oxidase 1 (Nox1) contribution to injury-induced neointimal formation. Arterioscler Thromb Vasc Biol 2009;29:480–7. [29] Vendrov AE, Hakim ZS, Madamanchi NR, et al. Atherosclerosis is attenuated by limiting superoxide generation in both macrophages and vessel wall cells. Arterioscler Thromb Vasc Biol 2007;27:2714–21. [30] Chen Z, Keaney JF, Schulz E, et al. Decreased neointimal formation in Nox2deficient mice reveals a direct role for NADPH oxidase in the response to arterial injury. Proc Natl Acad Sci USA 2004;101:13014–9. [31] Thomas M, Gavrila D, McCormick ML, et al. Deletion of p47phox attenuates angiotensin II-induced abdominal aortic aneurysm formation in apolipoprotein Edeficient mice. Circulation 2006;114:404–13. [32] Gavazzi G, Deffert C, Trocme C, et al. Nox1 deficiency protects from aortic dissection in response to angiotensin II. Hypertension 2007;50:189–96. [33] Akhtar K, Broekelmann TJ, Miao M, et al. Oxidative and nitrosative modifications of tropoelastin prevent elastic fiber assembly in vitro. J Biol Chem 2010;285: 37396–404. [34] Cantor JO, Shteyngart B, Cerreta JM, Ma S, Turino GM. Synergistic effect of hydrogen peroxide and elastase on elastic fiber injury in vitro. Exp Biol Med 2006;231:107–11. [35] McCloskey A, Taniguchi I, Shinmyozu K, Ohno M. hnRNP-C tetramer measures RNA length to classify RNA polymerase II transcripts for export. Science 2012;335:1643–6. [36] Sella O, Gerlitz G, Le S-Y, Elroy-Stein O. Differentiation-induced internal translation of c-sis mRNA: analysis of the cis elements and their differentiationlinked binding to the hnRNP-C protein. Mol Cell Biol 1999;19:5429–40. [37] Kim JH, Paek KY, Choi K, et al. Heterogeneous nuclear ribonucleoprotein C modulates translation of c-myc mRNA in a cell cycle phase-dependent manner. Mol Cell Biol 2003;23:708–20.