Phosphodiesterase 6 subunits are expressed and altered in idiopathic pulmonary fibrosis

Background Idiopathic Pulmonary Fibrosis (IPF) is an unresolved clinical issue. Phosphodiesterases (PDEs) are known therapeutic targets for various proliferative lung diseases. Lung PDE6 expression and function has received little or no attention. The present study aimed to characterize (i) PDE6 subunits expression in human lung, (ii) PDE6 subunits expression and alteration in IPF and (iii) functionality of the specific PDE6D subunit in alveolar epithelial cells (AECs). Methodology/Principal Findings PDE6 subunits expression in transplant donor (n = 6) and IPF (n = 6) lungs was demonstrated by real-time quantitative (q)RT-PCR and immunoblotting analysis. PDE6D mRNA and protein levels and PDE6G/H protein levels were significantly down-regulated in the IPF lungs. Immunohistochemical analysis showed alveolar epithelial localization of the PDE6 subunits. This was confirmed by qRT-PCR from human primary alveolar type (AT)II cells, demonstrating the down-regulation pattern of PDE6D in IPF-derived ATII cells. In vitro, PDE6D protein depletion was provoked by transforming growth factor (TGF)-β1 in A549 AECs. PDE6D siRNA-mediated knockdown and an ectopic expression of PDE6D modified the proliferation rate of A549 AECs. These effects were mediated by increased intracellular cGMP levels and decreased ERK phosphorylation. Conclusions/Significance Collectively, we report previously unrecognized PDE6 expression in human lungs, significant alterations of the PDE6D and PDE6G/H subunits in IPF lungs and characterize the functional role of PDE6D in AEC proliferation.


Introduction
IPF is a progressive interstitial lung disease of unknown etiology associated with high morbidity and mortality [1], and further characterized by abnormal alveolar epithelial and fibro-proliferative responses, excessive extra-cellular matrix deposition, patchy inflammatory infiltrations and progressive loss of normal lung structure [2]. At present there is no effective therapy for blocking or reversing the progression of the disease [3]. This situation demands a better understanding of the molecular and cellular mechanisms involved in the pathogenesis of IPF.
PDEs comprise a family of related proteins which can be subdivided into 11 families based on their amino acid sequences, sensitivity to different activators and inhibitors and their ability to preferentially hydrolyze either cAMP or cGMP, or both [4]. Of these, PDE6 is a cGMP-specific PDE family and presents multi-component enzyme complexes [5]. The rod PDE6 enzyme is comprised of two catalytic subunits, PDE6α and PDE6β, encoded by the PDE6A and PDE6B genes respectively, two identical inhibitory subunits PDE6γ, encoded by PDE6G [6,7], and one regulatory subunit PDE6δ, encoded by the PDE6D gene [8]. The cone PDE6 enzyme represents two identical catalytic subunits of PDE6α' and two identical inhibitory subunits PDE6γ', encoded by the PDE6C and PDE6H genes, respectively [9]. Primarily localized in the rod and cone photoreceptive cells of the mammalian retina, PDE6 has been widely studied in the context of visual dysfunctions [10,11]. Until now, the expression and characterization of PDE6 in other organs outside of the retina has received little attention. However, recent reports suggest functionality of PDE6 apart from the classical phototransduction cascade [12][13][14]. PDE6 activity has been coupled to non canonical Wnt5a-Frizzled-2 signaling in non-retinal tissue [12][13][14]. Recently, a significant increase of Wnt signaling in ATII cells derived from IPF patients and its involvement in epithelial cell injury and hyperplasia has been documented [15].
More interestingly, the specific PDE6D subunit has been reported to regulate the membrane association of Ras and Rap GTPases [16]. The striking similarity between PDE6D and Rho guanine nucleotide dissociation inhibitor (GDI) reasons involvement of PDE6D in cytoskeleton reorganization, membrane trafficking, transcriptional regulation and cell growth control [17]. The study of Cook TA et al. demonstrates that PDE6D can modify cGMP hydrolytic activity in preparations of broken rod outer segments [18]. cGMP plays a role in controlling key epithelial cell functions such as ciliary motility, cytokine production and proliferation [19][20][21].
We therefore hypothesized that i) the PDE6 subunits potentiality can be expressed in the lung, ii) the subunits are differentially regulated in IPF and iii) the specific subunit of PDE6, PDE6D, modulates the proliferation rate of AECs. To this end, we achieved our aim to elucidate previously unrecognized PDE6 expression in normal human lungs, significant alterations of the PDE6D and PDE6G/H subunits in IPF-derived lungs and characterize the functional role of PDE6D in AEC proliferation.

Ethics Statement
The study protocol for tissue donation was approved by the Ethics Committee of the Justus-Liebig-University School of Medicine (AZ 31/93). Informed consent was obtained from each individual patient or the patient's next of kin.

Human Tissues
Explanted lung tissues from IPF subjects (n = 6) or donor (n = 6) were obtained during lung transplantation at the Department of Cardiothoracic Surgery, University of Vienna, Austria. Diagnosis was established on the basis of a proof of a usual interstitial pneumonitis (UIP) pattern in the explanted lungs from lung transplant recipients [(n = 6; 4 males, 2 females; mean age = 63.33 ± 1.71 yr, mean forced vital capacity (FVC) = 39.00 ± 2.58 (% of standard); mean forced expiratory volume (FEV) = 44.67 ± 6.39 (% of standard); mean carbon monoxide lung diffusion capacity (DL co ) = 30.5 ± 1.5 (% of predicted)]. Apart from IPF subjects, tissue was also obtained from 6 donor lungs, which could not be utilized due to size limitations between donor and putative recipient (mostly single lobes) or due to incompatibility between donor and recipient.

Isolation of human ATII cells
Primary human ATII cells were isolated, as previously described [22]. Briefly, the lung was digested and minced. The cell-rich fraction was filtered, layered onto a Percoll density gradient, and centrifuged. The cells were then incubated with anti-CD14 antibody-coated magnetic beads. The remaining cell suspension was incubated in human IgG-coated tissue culture dishes at 37°C in a 5% CO 2 , 95% O 2 atmosphere. The purity of isolated human ATII cells was examined by Papanicolaou staining. The purity and viability of ATII cell preparations was consistently between 90 and 95%.
RNA isolation, cDNA synthesis and mRNA quantification by qRT-PCR or semi-quantitative RT-PCR Total RNA was isolated from frozen human lung tissues and cell pellets using Trizol reagent (Invitrogen, Carlsbad, CA, USA). cDNA synthesis was carried out with an ImProm-II-™ reverse transcription system (Promega Corporation, Madison, WI, USA) by incubating 5 μg of RNA, following the manufacturer's protocol.
qRT-PCR was performed with 2 μl cDNA set up with the Platinum SYBRGreen qPCR SuperMix UDG (Invitrogen The PCR products were sequence analyzed.

Blocking with immunizing peptides
Anti-PDE6A and -PDE6B antibodies specificity was validated with PDE6A blocking peptide (M(1)GEVTAEE-VEKFLDSN(16)C, Abcam, Cambridge, UK) and PDE6B blocking peptide (H(20)QYFG(K/R)KLSPENVAGAC (36), Abcam, Cambridge, UK), respectively. The signals were developed with an ECL kit as described above. The signal that disappeared when using the blocking peptide (BP) was considered specific to the antibody. GAPDH was used as a control for equal loading.

PDE activity assay
The A549 cell protein was extracted with RIPA buffer (Santa Cruz, Heidelberg, Germany) and equalized to the same concentration for use. The reactions were performed with 10 μg protein in 100 μl HEPES buffer (40 mM) at pH 7.6 consisting of MgCl 2 (5 mM), BSA (1 mg/ml) and [ 3 H]-cGMP (1 μCi/ml, Amersham Biosciences, Munich, Germany) at 37°C for 15 min. The samples were boiled for 3 min, subsequently cooled for 5 min and incubated with 25 μl Crotalus atrox snake venom (20 mg/ml, Sigma-Aldrich, Munich, Germany) for 15 min at 37°C. After being chilled on ice, the samples were applied to QAE Sephadex A-25 (Amersham Biosciences, Munich, Germany) mini-chromatography columns and eluted with 1 ml ammonium formate (30 mM, pH 7.5). The elutes were collected in 2 ml scintillation solution (Rotiszint®eco plus, Roth, Germany) and counted by a beta-counter with CPM (counts per minute) values. Each assay was performed in triplicate and repeated twice independently. Data are expressed as picomoles of cGMP per minute per milligram of protein. (pmol cGMP/minute/mg protein).

cGMP enzyme immunoassay (EIA)
At the end of culture, cells were washed with PBS twice and lysed in 0.1 M HCl at room temperature for 10 min. After centrifugation the supernatants were equalized to the same protein concentration for use. 50 μl protein samples which were pre-diluted to 0.3 μg/ml and standard solutions were incubated with 50 μl tracer and 50 μl antibody in darkness at 4°C overnight. After washing 5 times, plates were incubated with Ellman's solution for 90-120 min at room temperature with gentle shaking. The plates were read at a wavelength of 405 nm and the concentration was calculated by the readymade Cayman EIA Double workbook. The standard curve was made as a plot of the %B/B0 value (%Bound/ Maximum Bound) vs concentration of a series of known standards using a linear (y) and log (x) axis. Using the 4-parameter logistic equation obtained from the standard curve, the cGMP concentration of samples was determined and is given as nmol/mg protein. Each sample was determined in duplicate and repeated twice.

Statistical analysis
All data were expressed as the means ± S.E. Data were compared using a two-tailed Student's t-test, or a 1-way ANOVA with the Bonferroni's post hoc test for studies with more than 2 groups. Statistical significance was assumed when P < 0.05.

mRNA detection of the PDE6 enzyme subunits
The mRNA expression of each PDE6 subunit in lung tissue homogenates of donors and IPF patients was analyzed by qRT-PCR technique. As illustrated in Figure 1A, PDE6A, PDE6B, PDE6C, PDE6D, PDE6G and PDE6H mRNAs were expressed in the human lung. PDE6A, PDE6B, PDE6C and PDE6G showed no significant alterations in the IPF lungs as compared to donor lungs. In contrast, PDE6D subunit was significantly down-regulated in the IPF lungs as compared to the donor lungs (relative mRNA expression: 2.44 ± 0.28 and 0.30 ± 0.56, respectively) and PDE6H showed a tendency of down-regulation in the IPF lungs as compared to the donor lungs (relative mRNA expression: -7.22 ± 0.34 and -8.98 ± 0.66, respectively). In addition, the resultant PCR products were validated by direct sequencing, followed by BLAST analysis that confirmed the similar sequence alignment for each subunit ( Figure 1B).

Protein expression of the PDE6 enzyme subunits
The protein content of the PDE6 subunits in whole lung tissue homogenates of donors and IPF patients was quantified by immunoblotting. As illustrated in Figure 2A,  immunoreactivity was detected for PDE6A (~105 kDa), PDE6B (~105 kDa), PDE6D (~17 kDa) and PDE6G/H (~11 kDa) subunits. PDE6A and PDE6B blocking peptide studies were carried out to reconfirm the specificity of PDE6A and PDE6B immunoreactivity ( Figure 2C and  2D). Additionally, pig retinal lysate served as a positive control for immunoreactivity and proper protein size ( Figure 2E). Notably, the PDE6D and PDE6G/H subunits were significantly down-regulated in the IPF lungs as compared to donor lungs, whereas PDE6A and PDE6B showed no significant alterations between donor and IPF-derived lung tissues ( Figure 2B).

Cellular localization of the PDE6 enzyme subunits
The cellular localization of the PDE6 subunits was assessed by serial immunohistochemical stainings on tissue sections from donor and IPF lungs. As shown in Figure 3A, PDE6A, PDE6B, PDE6D and PDE6G/H were co-stained with pro-SPC, suggesting the presence of PDE6 subunits in ATII cells. PDE6A immunoreactivity was recognized in the cytoplasm and membrane of ATII cells, PDE6B immunoreactivity was recognized in the nuclei, PDE6D immunoreactivity in the cytoplasm and PDE6G/H immunoreactivity in the membrane of ATII cells.

PDE6 enzyme subunits expression in human AECs
To confirm the AEC localization pattern, the PDE6 subunits were qRT-PCR amplified from primary human donor and IPF-derived ATII cells. All PDE6 subunits (except for PDE6C, no amplicons were detected by qRT-PCR) were found to be expressed by these cells ( Figure 3B). Notably, PDE6D mRNA levels were significantly decreased in IPF-derived ATII cells as compared to donor ATII cells (relative mRNA expression: 1.56 ± 1.05 and -3.80 ± 1.40, respectively). In contrast, PDE6A, PDE6B, PDE6G and PDE6H were not differentially regulated in AECIIs from IPF versus control lungs.

TGF-b1 down-regulates PDE6D in A549 cells
A549 cells were used as an in vitro AEC model. Firstly, the cells were characterized for the expression of PDE6 subunits. mRNAs of all PDE6 subunits (except for PDE6C and PDE6H) and the complete set of PDE6 proteins were found to be expressed by these cells ( Figure  4A and 4B). Next, to explore whether TGF-β1 promotes PDE6D down-regulation in AECs, A549 cells were treated with two different concentrations of TGF-β1 (2 ng/ ml and 5 ng/ml) for 12 and 24 h. Decrease in PDE6D protein expression was clearly evident at concentration as low as 2 ng/ml ( Figure 4C and 4D), with no further decrease at higher concentration (5 ng/ml) ( Figure 4E and 4F). PDE6D down-regulation occurred within 12 h of TGF-β1 stimulation and was sustained up to 24 h ( Figure 4C-F).

Effects of PDE6D modulations on A549 cells proliferation
Further, we studied the functional impact of PDE6D modulations on A549 cells proliferation. siRNA silencing of PDE6D resulted in a significant loss of PDE6D protein expression 24 and 48 h post transfection. Transfection with non-targeting siRNA caused no change in PDE6D protein expression ( Figure 5A). The loss of PDE6D expression was coupled to a significantly decreased cell number

PDE6D knockdown regulates cGMP levels and ERK phosphorylation
We then opted to explore signaling pathways related to PDE6D-mediated proliferative responses. In particular, we studied the effects of PDE6D down-regulation on (i) cGMP hydrolyzing PDE activity, (ii) intracellular cGMP levels and (iii) serum induced phosphorylation of ERK protein in A549 cells. cGMP hydrolyzing PDE activity was decreased in PDE6D siRNA as compared to non-targeting siRNA and mock transfection 24 h post serum stimulation. In corroboration, intracellular cGMP determined by EIA assay was increased 1.6 fold by PDE6D downregulation ( Figure 7A and 7B). ERK phosphorylation was increased 1 h, 12 h and 24 h post serum stimulation as compared to unstimulated cells (0.1% FBS). siRNA mediated loss of PDE6D protein expression was detectable 12 h and 24 h post serum stimulation and this was related to a decrease in ERK phosphorylation as compared to control siRNA treated cells ( Figure 7C-E). However, no apparent changes in the phospho-p38α/β levels were observed by PDE6D down-regulation, suggesting the specificity of PDE6D for ERK signaling ( Figure 7C-E).

Discussion
In the present study, we report previously unrecognized PDE6 expression in the human lung. The members of the PDE family, PDE1, PDE2, PDE3, PDE4 and PDE5 are highly expressed in the lung and have been shown to potentially contribute to the pathogenesis of various lung diseases [27,28]. Nevertheless, to our knowledge this is the first report that has described the expression and characterization of PDE6 subunits in both the physiology and pathophysiology of the lung. Among these, PDE6D (mRNA and protein levels) and PDE6G/H subunit (protein levels) were found significantly downregulated in the IPF lungs as compared to the donor lungs. All PDE6 subunits were detected in ATII cells, with PDE6D significantly down-regulated in IPF-derived ATII cells. PDE6D down-regulation was induced in vitro by TGF-β1 in A549 cells, suggesting a link between the observed PDE6D down-regulation in IPF specimens and the pathogenesis of the disease [29]. Furthermore, using A549 cells as an in vitro AECs model, we were able to show that PDE6D modulates the proliferation rate of these cells (siRNA and ectopic expression studies). More interestingly, we showed that mechanisms accounting for PDE6D effects on AEC proliferation is related to PDE6D increasing the intracellular cGMP levels and suppressing the phosphorylation of ERK.
This finding is further supported by the reports that have demonstrated solitary PDE6 subunit expression in a variety of non-retinal tissues. The rod catalytic PDE6A and PDE6B subunits were found to be weakly expressed in brain [30,31]. Piriev et al. demonstrated that the catalytic core of the rod PDE6 enzyme can be synthesized in human kidney cells with consequent expression of enzymatic activity [32]. The regulatory and the inhibitory PDE6D and PDE6G subunits, respectively, have been reported to be expressed in a variety of heterogeneous tissues, including the lung [8,33]. Identifying the expression of PDE isoforms in organs and cells that had not been reported previously is a subject gaining interest. For example, PDE5, known to express in lung, recently reported to be also expressed in vascular, ganglion and bipolar cell layers of retinal tissue. It was claimed to play a physiological role in the retina and might contribute to PDE5 inhibitor-associated ocular side effects [34].
Although at present the physiological roles of the PDE6 subunits in the lung are unknown and the functionality of the PDE6 enzyme in IPF needs to be explored, the study of Wang et al. [14] does provide evidence for the presence of functional PDE6 enzyme in non-retinal tissues. Based on our findings, all the PDE6 subunits appear to be expressed and localize in human lung alveolar epithelium. Among those, PDE6D and PDE6G/H subunit protein levels were found significantly down-regulated in the IPF lungs as compared to the donor lungs, suggesting a plausible contribution of these PDE6 subunits to the pathogenesis of IPF. Thus, we believe that PDE6 alterations may play a crucial role in epithelial apoptosis, proliferation, surfactant synthesis and reactive oxygen species (ROS) generation abnormalities associated with IPF [35].
In fact, based on the data obtained from human donor and IPF lungs, it is not possible at present to determine whether PDE6 functions as a complex or each PDE6 subunit has a solitary function. However, considering the requirement for multiple subunits assembly to produce functional rod and cone PDE6 enzymes and the difficulties in expressing functionally active rod and cone PDE6 enzymes in various systems [36], we herein explored independent functionality of the specific PDE6D subunit in AEC proliferation. In addition, we assessed the contribution of PDE6D to PDE6 as well as to the presence or absence of cGMP. In our studies, gain of function (overexpression) or loss of function (targeting siRNA) of PDE6D affected AEC proliferation, with increased PDE6D resulting in increased AEC proliferation. The anti-proliferative effects encountered in response to PDE6D knockdown were largely due to a decrease in cGMP hydrolyzing PDE activity that may subsequently stimulate the intracellular levels of cGMP. Of note, we were able to measure only total cGMP hydrolyzing activity ( Figure 7B), but not PDE6 specific cGMP hydrolyzing activity due to less selectivity of PDE6 inhibitors. Several classes of PDE inhibitors inhibit PDE6 equally as well as the PDE family to which they are targeted [37]. Similarly, further studies are needed to explore the role of PDE6 inhibitory subunits (PDE6G and PDE6H) that were found down regulated at the protein level in IPF lungs. Several lines of evidence reported that the inhibitory PDE6G/H subunits of the PDE6 are expressed in non-retinal tissues [33] and are involved in the stimulation of the p42/p44 mitogenactivated protein kinase (MAPK) pathway by growth factors and G-protein-coupled receptor agonists in human embryonic kidney 293 cells [38]. Impaired AECs proliferation is a significant finding in IPF [39]. Multiple studies have reported rapid proliferation of ATII cells following injury [40,41] or reduced proliferative capacity of ATII cells and inability to differentiate into ATI cells in both experimental lung fibrosis [42] and IPF [39]. Herein, we report modulatory effects of the specific PDE6D subunit on AECs proliferation, as deduced from PDE6D siRNA-mediated knockdown and over-expression studies in A549 cells. This functional property of PDE6D is significant, considering its c-Myc/ E2F4 controlled expression (http://www.unleashedinformatics.com). In line with these studies, PDE6D (-/-) mice are consistently smaller in size, indicating a plausible involvement of PDE6D in growth arrest [43]. Thus, it can be imagined that the proliferative phenotype of IPF-derived ATII cells is associated with the observed PDE6D down regulation in IPF lungs.
ERK activation has been shown to be of critical importance for ATII cell proliferation [44]. ERK signaling has also been documented to regulate differentiation of fetal ATII cells [45]. In agreement, our study indicates that ERK is a key mediator of A549 AECs proliferation and that PDE6D mediated proliferative responses are related to ERK signaling. siRNA mediated inhibition of PDE6D decreased the serum induced phosphorylation of ERK in a time response fashion. Thus, we propose PDE6D as a critical regulator of ERK mediated ATII cells proliferation.
In conclusion, these data demonstrate previously unrecognized PDE6 expression in human lung, significant alterations of the PDE6D and PDE6G/H subunits in IPF-derived lungs and characterize the functional role of PDE6D in AEC proliferation. For a further consolidation of the proposed pathomechanistic link between PDE6D content and type II cell proliferation on an in vivo level, transgenic mice with epithelial cell-specific PDE6D knockout would have to be generated. Hence, we can, right now, only postulate that decreased PDE6D expression in IPF might be involved in attenuation of type II cell hyperplasia. Further, it is tempting to speculate that therapeutic prevention of PDE6D down-regulation and/or PDE6D overexpression in animal models of pulmonary fibrosis may be beneficial to boost up alveolar re-epithelization and may represent a therapeutic option in IPF.