To elucidate the role of cAMP and different cAMP-dependent protein kinases (PKA; A-kinase) in lung cell proliferation, we investigated rat alveolar type 2 cell proliferation in relation to activation or inhibition of PKA and PKA regulatory subunits (RIIα and RIα). Both the number of proliferating type 2 cells and the level of different regulatory subunits varied during 7 days of culture. The cells exhibited a distinct peak of proliferation after 5 days of culture. This proliferation peak was preceded by a rise in RIIα protein level. In contrast, an inverse relationship between RIα and type 2 cell proliferation was noted. Activation of PKA increased type 2 cell proliferation if given at peak RIIα expression. Furthermore, PKA inhibitors lowered the rate of proliferation only when a high RII level was observed. An antibody against the anchoring region of RIIα showed cell cycle-dependent binding in contrast to antibodies against other regions, possibly related to altered binding to A-kinase anchoring protein. Following activation of PKA, relocalization of RIIα was confirmed by immunocytochemistry. In conclusion, it appears that activation of PKA II is important in regulation of alveolar type 2 cell proliferation.
- protein kinase A
- A-kinase anchoring protein
extensive cell proliferation takes place during lung development. In the rat lung, the percentage of proliferating cells decreases to a low level during adolescence. Thereafter, proliferation only occurs to replace damaged or dying cells. The population of proliferating cells in the alveolar epithelium consists mainly of type 2 cells (49). These cells are the progenitor cells of the respiratory type 1 cells and are also able to proliferate in culture (24, 27, 30).
The cAMP-dependent pathway stimulates cell proliferation in some cell types while inhibiting cell growth in others (8, 37). The cAMP-dependent signaling pattern involved in the regulation of lung cell proliferation has been elucidated only to a minor extent. Previous findings using cultured lung tissue have indicated cAMP as a positive regulator in type 2 cell proliferation (55). Furthermore, involvement of PKA in the regulation of type 2 cell differentiation, surfactant production, and secretion has also been indicated. The major receptor for cAMP in eukaryotes is the cAMP-dependent protein kinase (PKA; A-kinase). PKA activity and PKA subunit mRNA expression have been found to be developmentally regulated in fetal lung (1). It has been suggested that this results in an optimal PKA activity at the time of type 2 cell differentiation, presumably in preparation for gas exchange (1).
The PKA holoenzyme consists of two catalytic (C) subunits, for which activity is inhibited by a dimer of regulatory subunits (R2). Upon binding of cAMP, the holoenzyme dissociates into a regulatory subunit dimer and two active C subunits. Active C subunits phosphorylate specific substrates that can be translocated to the nucleus where they activate cAMP-responsive genes (18, 19, 33). Proteins involved in regulation of proliferation, such as cdc2 (47), cyclin A (13), and the Rb protein and D-cyclins (34, 54) have been found to be modulated by PKA activation. However, these proteins did not seem to be phosphorylated directly by PKA.
Two main classes of PKA have been identified, type I (PKA I) and type II (PKA II). The two holoenzymes differ only in their regulatory subunits, but share the same type of C subunits (RI2C2 and RII2C2, respectively). Two isoforms of each subunit have been identified in rat tissues, α and β (14). Whereas the α-isoforms are found in most tissues examined including lung, the β-isoforms seem to be tissue specifically expressed mainly in neural and reproductive organs (6). In addition to inhibition of the C subunit, the regulatory subunits are considered to be involved in the localization of the PKA holoenzyme. The NH2-terminal region of the regulatory subunit of PKA II (RII) contains a binding site for A-kinase anchoring proteins (AKAPs) (32). Both RI and RII can exist in soluble or anchored phases depending either on cell type or phase or state of the cell (2). The RII subunit has been found to bind with high affinity to AKAPs, whereas the RI subunit only binds to some AKAPs and with much lower affinity (21, 28). Different AKAPs are localized preferentially to different membrane structures, such as sarcoplasmatic reticulum (31), centrosome and Golgi region (40), or mitochondria (41), and thereby specify PKA II colocalization with specific substrates (12).
Several studies have shown PKA I to be directly involved in cell proliferation and neoplastic transformation (9, 45). PKA I has also been reported to be required for the transition from the G1 to the S phase of the cell cycle (45) as well as mediating mitogenic signals through different growth factors (10, 11, 46). Cell lines derived from adult mouse lung epithelial cells and displaying low levels of PKA I (RIα) protein have been found to grow faster than cell lines with higher levels indicating an inverse relationship between growth and PKA I (29). PKA II, which is preferentially expressed in normal tissues, seems mostly to be involved in cell growth arrest and differentiation (9, 35, 44). However, in some cell types, an association between PKA II and cell proliferation has been reported (4, 15). Thus PKA I and PKA II may mediate both stimulatory and inhibitory proliferation signals, depending on the cell type.
The aim of this study was to elucidate the role of cAMP and PKA isoforms in regulating proliferation of primary rat alveolar type 2 cells.
MATERIALS AND METHODS
Ethidium bromide, dibutyryl-cAMP (db-cAMP), forskolin, protease (type 1, crude), DNase 1 (type 3), RNase, hydrocortisone, epidermal growth factor (EGF), insulin, and bovine serum albumin (BSA) were purchased from Sigma Chemical (St. Louis, MO). PKA inhibitors H8 and H89 were from Seikagaku (Falmouth, MA), Williams E medium from BioWhittaker (Walkersville, MD), Hoechst 33258 from Calbiochem-Boehringer (La Jolla, CA), and fetal bovine serum (FBS) from Gibco BRL (Paisley, Scotland). All other chemicals were purchased from commercial sources at the highest purity available.
Cell isolation and culture.
Type 2 cells were isolated from male Wistar-Kyoto (WKy/NHsd) rats between 6 and 8 wk of age. The cells were isolated by the sequential use of enzymatic digestion, centrifugal elutriation, and differential attachment, as previously described (27). In brief, development of the isolation procedure included verification of type 2 cell presence by electron microscopy, light microscopy, immunocytochemistry, and Western blotting. Current confirmation of epithelial cell origin in general and type 2 cell presence in specific included expression of epithelial cadherin, pan-keratin, and prosurfactant protein C. The results indicated that cells expressing these proteins dominated throughout culture. The animal experiments have been approved by the local responsible laboratory animal science specialist under the surveillance of the National Experimental Animal Board in Norway and registered by the board.
The epithelial lung cells were cultured in Williams E medium fortified with EGF (10 ng/ml), insulin (5 μg/ml), hydrocortisone (87 ng/ml), and 5% FBS. Cell viability was assayed by trypan blue exclusion. Purity in the type 2 cell fraction was ∼90% (27).
Determining cell number.
Cell culture dishes were fixed for 5 min in methanol and stained with 2% Giemsa solution in neutral distilled water. The stained cells were counted using a Nikon Optiphot light microscope.
Antisera to rat RIα and RIIα were raised against synthetic peptides corresponding to amino acids (AA) 1–18 and 5–21, respectively, as previously described (39). A second antibody to RIIα raised against AA 367–394 was obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Antibody to C subunit (BD Biosciences, Erembodegem, Belgium) was a generous gift from Bjørn Skalhegg (Faculty of Medicine, University of Oslo). As secondary antibodies, horseradish peroxidase (HRP)-conjugated anti-rabbit IgG (Sigma Chemical) and FITC-conjugated anti-rabbit IgG (Dako, Glostrup, Denmark) were used.
Activation and inhibition of PKA in type 2 cells.
To activate PKA, db-cAMP (200 μM) or forskolin (3 μM) was added to the medium at specific time points. To inhibit PKA activity, 50 and 3 μM, respectively, of the PKA inhibitors H8 or H89 were used.
SDS-PAGE and Western analysis.
Cultured type 2 cells were scraped off the dish. Total protein (15 μg) of each sample from the cells were prepared and electrophoresed in 10% SDS-PAGE and blotted onto nitrocellulose filters according to standard procedures (26). Even loading and transfer of proteins to nitrocellulose membrane was confirmed by Ponceau S staining of the blots. Proteins were analyzed according to the Western blotting technique (48). The filters were blocked for 30 min with 5% nonfat dry milk in Tris-buffered saline (TBS)-Tween (0.1% Tween 20) and then incubated with antisera diluted in TBS-Tween containing 1% nonfat dry milk. Working dilutions of 1:10,000 (anti-RIα), 1:5,000 (anti-RIIα 5–21), 1:15,000 (anti-RIIα 367–394), and 1:2,000 (anti-C) were found to be appropriate for immunoblotting techniques. Immunoreactive protein was detected using the chemiluminescence system (ECL; Amersham Laboratories, Buckinghamshire, UK) after incubation with a HRP-conjugated secondary antibody.
Flow cytometric cell cycle analysis.
Freshly isolated cells and cells cultured for up to 7 days were stored in citrate buffer (29) at −70°C. Cellular DNA was stained with Hoechst 33258 (1.2 μg/ml), essentially as previously described (50). Samples were measured using a flow cytometer (Argus 100; Skatron, Tranby, Norway), and the DNA histograms were analyzed using the Multiplus Program (Phoenix Flow Systems, San Diego, CA).
Dual parameter DNA/PKA flow cytometry.
To determine the level of regulatory subunits during different phases of the cell cycle, cultured cells were trypsinized off the culture dish and processed according to a previously described method (36) with small modifications. Isolated cells were fixed in 4°C ethanol for 5 min. After centrifugation, the samples were washed twice in PBS containing 1% BSA and incubated overnight at room temperature with RNase (50 μg/ml in PBS) and primary antibody to RIα or two different RIIα antibodies (diluted 1:1,000 and 1:500, respectively). Samples were washed and incubated for 2 h with secondary antibody (FITC-conjugated goat anti-rabbit, Dako). Finally, the cells were stained with ethidium bromide (2 μg/ml). As control, cells were treated identically but with preimmune rabbit serum at the same concentration as the antibody.
To investigate subcellular localization of RIIα and possible translocation during cell cycle, unstimulated as well as dibutyryl-treated type 2 cells were incubated with MitoTracker according to the protocol provided by the manufacturer. The cells were subsequently washed in PBS and fixed in methanol for 3 min before incubation for 20 h with rabbit anti-RIIα (working dilution 1:200). After washing and incubating for 3 h with an anti-rabbit FITC-conjugated antibody, the preparations were mounted and visualized using a Nikon Eclipse E400 microscope and a SPOT diagnostic instruments digital camera. As controls, the anti-RIIα antibody was omitted.
The data were analyzed using one-way ANOVA with pairwise multiple comparison procedures according to the Holm-Sidak method, Kruskal-Wallis one-way ANOVA on ranks, or Mann-Whitney rank sum test to evaluate the difference between groups. P values <0.05 were considered significant.
Lung cell proliferation.
The proliferative behavior of type 2 cells in primary culture was monitored over a period of 7 days. Figure 1A shows the percentage of type 2 cells in S and G2/M phases of the cell cycle as measured by flow cytometry. The percentage of type 2 cells in these phases increased significantly from day 3 of culture and reached a proliferation peak at day 5 with >25% of the cells in S and G2/M phase. The cell number increased ∼3-fold from day 1 to day 7 (Fig. 1B). The largest increase was from day 5 to day 7.
To study the possible involvement of PKA in type 2 cell proliferation, PKA activity was stimulated with db-cAMP or inhibited with the PKA inhibitor H8 at different time points (Fig. 2A). The inhibitor and the stimulator had no detectable effect on the percentage cells in S and G2/M phase early (day 1) or late (day 7) in the culture. In contrast, db-cAMP exerted a statistically significant stimulatory effect on the number of S and G2/M phase cells just prior to and at the proliferation peak at days 3 and 5. During the same period H8 had a statistically significant inhibitory effect (Fig. 2A).
Figure 2B shows a comparison of the relative effects of different stimulators and inhibitors of PKA on day 3 of the culture. The PKA inhibitor H8 significantly lowered the number of proliferating cells by about 3–4% (20% reduction compared with control). Also, the PKA inhibitor H89 appeared to lower the number of proliferating cells, although we did not find this decrease statistically significant. Cells incubated with db-cAMP, 8-bromo-cAMP, or forskolin showed a significant increase of 3–4% in S/G2/M phase cells (a relative increase of 18–25% compared with control). db-cAMP seemed to exert a slightly stronger effect than forskolin or 8-bromo-cAMP.
Furthermore, the number of cells in culture increased significantly 48 h after PKA activation, whereas PKA inhibition significantly reduced that number relative to untreated control (Fig. 2C). Cells simultaneously exposed to PKA stimulators and inhibitors exhibited the same reduction in proliferating cells as with the inhibitors alone (data not shown). After prolonged incubation (48 h), the percentage of S/G2/M cells was similar in control cells and cells with stimulated PKA (data not shown).
PKA regulatory subunit protein levels.
To investigate if the differential effects of PKA activity modulators resulted from changes in PKA regulatory subunit levels, we measured their levels in type 2 cells during 7 days of culture by immunoblotting. Immunoblotting of PKA RIα detected a protein at 49 kDa (Fig. 3A). Freshly isolated type 2 cells exhibited a relatively high level of RIα that declined after the first 24 h in culture. The significantly lowest levels were observed at days 3 to 5. Subsequently, RIα increased approximately to the same level as found in 24-h cultured type 2 cells (Fig. 3A). Immunoblotting of RIIα detected a protein at 52 kDa (Fig. 3B). The RIIα level in type 2 cells increased from very low levels at day 0 to significantly higher levels between days 1 and 5 of culture and with an apparent peak at day 3. Thereafter, the level of RIIα declined to levels similar to the level at day 1 (Fig. 3B). Thus the time curve for stimulation with cAMP analogs and appearance of increased RIIα levels correlated with the proliferation curve, whereas the time curve for RIα levels showed an inverse correlation.
PKA subunit levels after PKA activation.
PKA may regulate the levels of its subunits as a “feedback” inhibition. Immunoblotting was used to monitor the PKA subunit level after activation with either db-cAMP or forskolin. After PKA activation, the level of C was significantly reduced (Fig. 4A), whereas RIα was significantly upregulated after treatment with db-cAMP (Fig. 4B). However, forskolin also appeared to cause an increase in RI subunit level, although we did not find this statistically significant. The level of RIIα was not affected (data not shown).
Cell cycle-dependent level of the regulatory subunits.
Dual parameter flow cytometry was used to measure the binding of antibodies to RIα or RIIα during the cell cycle. Analysis with RIα antibody and ethidium bromide showed that the antibody binding was proportional to the DNA content of the cells (Fig. 5A). There was twice as much antibody binding in G2/M phase cells as in G1 phase cells, indicating that the increase in the RIα level followed average protein doubling. The antibody raised against the NH2-terminal region of RIIα (AA 5–21), which also is the AKAP binding region of the RIIα protein, was bound to the cells in a cell cycle-specific manner. Antibody binding was relatively low in G0/G1 phase, but increased severalfold in the G1-to-S phase transition. Thereafter, the level increased proportionally to the DNA content through S and G2/M phase (Fig. 5B). The antibody raised against AA 367–394 of RIIα was used to see if RIIα levels varied abnormally or if the changes were due to differences in antigen availability during a certain period of the cell cycle. This antibody did not show any cell cycle-specific binding (Fig. 5C), suggesting the latter possibility.
Our observation that the antibody against the anchoring region of RIIα showed cell cycle-dependent binding could imply altered binding to AKAP and a change in RIIα localization during cell cycle. Such a possibility was confirmed immunohistochemically where mitochondrial location of RIIα was observed in unstimulated cells (Fig. 6A), whereas stimulation by db-cAMP for 1 h caused a more diffuse cytoplasmatic localization of RIIα (Fig. 6B).
cAMP has the dual ability to act both as a negative and a positive regulator of cell proliferation. By using stimulators of cAMP, it has been shown that cAMP acts as a positive regulator in proliferation of lung tissues (55). These latter results further indicated that cAMP-induced proliferation mainly occurred in alveolar regions, where the type 2 cells are the main proliferating cell population. Most likely, the effect of cAMP was mediated through activation of PKA since no other cAMP receptors have been reported in epithelial cells.
Studies exploring the role of PKA isozymes have mainly related PKA I to proliferation and PKA II to growth arrest and differentiation (8). However, we found an inverse pattern between the wave of type 2 cell proliferation in vitro and changes in RIα protein levels. This is in agreement with the results of Lange-Carter and coworkers (29) reporting an inverse correlation with the RIα subunit and proliferation of cell lines derived from mouse lung type 2 cells. Also, in other cell types a similar correlation between increased levels of RIα and decreased proliferation rate has been reported (38). Some studies have also suggested a correlation between RIIα and proliferation (4, 15). Interestingly, we find that the wave of type 2 cell proliferation in vitro and the RIIα protein levels of these cells mainly display a corresponding pattern.
In their study of RIα expression in tumor cell lines of lung epithelial origin, Lange-Carter and coworkers (29) did not observe any changes in RIIα protein levels. We find that peak RIIα protein levels preceded the peak of proliferating type 2 cells. This may indicate an involvement of the kinase early in the cell cycle of type 2 cells. The data suggest that PKA may be involved in the G1 phase or early S phase of the type 2 cell cycle, consistent with what has been reported in thyroid cells and liver cells (5, 15). This would also be in agreement with previous findings indicating a role for cAMP and probably PKA in regulating phosphorylation status of Rb protein and D-type cyclins (54). These proteins appear to have central roles during the G1 phase and the entrance into the S phase of the cell cycle. In addition, cyclin A exerts its effect during the G1-S phase of the cell cycle, and PKA is shown to have a role in the regulation of cyclin A2 level in hepatocytes (13). The involvement of PKA II early in the cell cycle is supported by our results showing that the level of RIIα antibody binding in G2/M phase is severalfold higher than in the G1 phase and not two times as expected for the average of proteins. Activation of PKA II is a result of conformational changes in the NH2-terminal region of RII dimer after cAMP binding. This region is also known to bind to PKA AKAPs, which have been suggested to dissociate from RII upon cAMP binding. Only the antibody raised against this region of RIIα showed cell cycle-specific binding. This further suggests that the increase in antibody binding in the G1/S is due to activation of PKA II in this period.
The subcellular localization of signaling molecules is suggested to be an important mechanism for regulating specificity of signal transduction pathways. Most associations between AKAPs and regulatory (R) subunits as well as other signaling proteins are likely to be conditional. In this context, phosphorylation of RIIα and its association with the centrosome has been shown to change with the phase of the cell cycle (25). In thyroid cells, cAMP-dependent regulation of cell cycle genes has been linked to intracellular PKA II translocation (15). We find that cAMP stimulation of PKA II resulted in transient proliferation and relocation of RII from mitochondria to the cytoplasm in alveolar type 2 cells. AKAPs direct the subcellular localization of PKA by binding to the R subunits. Two AKAPs (DAKAP1 and 2) localized to mitochondria of several tissues including lungs and having affinity for both RI and RII have been identified (41). Although the physiological relevance of PKA and AKAPs within mitochondria is not fully understood, it has been recognized that PKA plays a critical role in mitochondrial physiology. In this context, it has been suggested that mitochondrial AKAP create a junction at which cAMP and tyrosine kinase signal transduction converge and interact (16, 51). Other proposed physiological responses by mitochondria-anchored PKA include involvement in the apoptotic process through phosphorylation and inactivation of the proapoptotic protein BAD (20). Interestingly, we found that the antibody against the anchoring region of RIIα showed cell cycle-dependent binding. This could indicate altered binding to AKAP and a change in RIIα localization during cell cycle. We confirmed this immunohistochemically as mitochondrial location of RIIα was observed in unstimulated cells, whereas stimulation by db-cAMP caused diffuse cytoplasmatic localization of RIIα.
Although the initial events associated with type 2 cell proliferation are likely to involve complex signaling pathways in addition to cAMP and PKA, our results support that PKA II is an important regulator in the process. In this regard, the observed increase in RIIα level just prior to proliferation may indicate a prerequisite for PKA II before proliferation may be initiated. For subsequent entry into the cell cycle, we suggest that the already increased level of PKA II requires activation. This is supported by our results showing that stimulation of PKA led to an increased percentage of proliferating cells as well as an increased number of cells. However, this effect could only be achieved when PKA II was already the dominating isoform. Similarly, inhibition of PKA reduces the percentage of proliferating cells as well as lowers the increase in cell number, but only when the PKA II isoform dominates. Moreover, stimulation of PKA at high PKA II levels appears to induce increased RIα levels, whereas the RIIα levels remained stable. Upregulation of cytoplasmatic RIα may function as a negative feedback regulating mechanism on proliferative signals by PKA II. This upregulation, together with a cytoplasmatic translocation of RIIα similar to what we observe, and which most likely localize the PKA holoenzyme to other substrate proteins, could then have an inhibitory influence on proliferation.
cAMP regulates PKA activity through altering the binding affinity between C and R subunits. However, PKA may also regulate its own activity by altering the R:C ratio (23, 42). In normal adult tissues, the ratio between R and C is 1:1 (22). We find reduced level of C and increased level of R after PKA stimulation, thus increasing the R:C ratio. This implies that higher cAMP threshold levels would be needed for PKA activation. Such a possibility is suggested by our observation that prolonged PKA activation results in proliferative activity similar to unstimulated levels.
Overall, PKA isotypes most likely play a partial role only for regulation of lung cell growth and differentiation (7, 52, 53). In line with this, we find that inhibitors and stimulators of PKA modulate type 2 cell proliferation only to a certain extent and only at specific time points. Other signaling pathways suggested to be of relevance in regulating type 2 cell proliferation in vitro are those in the extracellular signal-regulated kinase (ERK) cascade (43). The ERK pathway could be controlled in both PKA-dependent and PKA-independent ways (3, 17), although the relevance of this in lung cell proliferation has not been elucidated.
In conclusion, our results indicate that cAMP and the activation of PKA II participate in the stimulation of rat alveolar type 2 cell proliferation in vitro, whereas RIα may have other functions in this cell type. The activation of PKA II seems to be specific to the G1-to-S transition of the cell cycle.
The financial support by The Research Council of Norway is gratefully acknowledged.
The expert technical assistance of Edel M. Lilleaas is greatly appreciated. We are also thankful to Dr. Richard Wiger for useful discussion and comments on the manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society