Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Mar 1.
Published in final edited form as: Calcif Tissue Int. 2024 Mar 14;114(4):430–443. doi: 10.1007/s00223-024-01191-7

The PDE4 Inhibitors Roflumilast and Rolipram Rescue ADO2 Osteoclast Resorption Dysfunction

Jung Min Hong a, Rita L Gerard-O’Riley b, Dena Acton b, Imranul Alam b, Michael J Econs b,c, Angela Bruzzaniti a,*
PMCID: PMC11239147  NIHMSID: NIHMS1999840  PMID: 38483547

Abstract

Autosomal Dominant Osteopetrosis type II (ADO2) is a rare bone disease of impaired osteoclastic bone resorption caused by heterozygous missense mutations in the chloride channel 7 (CLCN7). Adenylate cyclase, which catalyzes the formation of cAMP, is critical for lysosomal acidification in osteoclasts. We found reduced cAMP levels in ADO2 osteoclasts compared to wild-type (WT) osteoclasts, leading us to examine whether regulating cAMP would improve ADO2 osteoclast activity. Although forskolin, a known activator of adenylate cyclase and cAMP levels, negatively affected osteoclast number, it led to an overall increase in ADO2 and WT osteoclast resorption activity in vitro. Next, we examined cAMP hydrolysis by the phosphodiesterase 4 (PDE4) proteins in ADO2 versus WT osteoclasts. QPCR analysis revealed higher expression of the three major PDE4 subtypes (4a, 4b, 4d) in ADO2 osteoclasts compared in WT, consistent with reduced cAMP levels in ADO2 osteoclasts. In addition, we found that the PDE4 antagonists, rolipram and roflumilast, stimulated ADO2 and WT osteoclast formation in a dose-dependent manner. Importantly, roflumilast and rolipram displayed a concentration-dependent increase in osteoclast resorption activity which was greater in ADO2 than WT osteoclasts. Moreover, treatment with roflumilast rescued cAMP levels in ADO2 OCLs. The key findings from our studies demonstrate that osteoclasts from ADO2 mice exhibit reduced cAMP levels and PDE4 inhibition rescues cAMP levels and ADO2 osteoclast activity dysfunction in vitro. The mechanism of action of PDE4 inhibitors and their ability to reduce the high bone mass of ADO2 mice in vivo are currently under investigation. Importantly, these studies advance the understanding of the mechanisms underlying the ADO2 osteoclast dysfunction which is critical for the development of therapeutic approaches to treat clinically affected ADO2 patients.

Keywords: osteopetrosis, chloride channel 7, osteoclast, phosphodiesterase, forskolin, cAMP

1. Introduction

Osteoclast dysfunction underpins several human diseases leading to either bone loss, as observed in osteoporosis, or alternatively, bone gain associated with osteoclast-rich osteopetrosis. Chloride channel 7 (CLCN7) is a member of a family of 2 chloride-1 proton transporters, which are generally classified as voltage gated and forming transmembrane dimers and have various intracellular functions including in lysosomes and the plasma membrane [1–5]. The CIC channels are involved in a wide range of physiological processes and cell types, including the regulation of resting membrane potential in skeletal muscle, control of transepithelial Cl- reabsorption in kidneys, and regulation of pH and chloride ion concentration in intracellular compartments through coupled Cl-/H+ exchange mechanisms. Of importance to the current study, Clcn7 is required for osteoclast resorption and heterozygous human missense mutations of CLCN7 are associated with Autosomal Dominant Osteopetrosis type II (ADO2), a rare bone disease that results in excessive bone gain [6–11]. However, the overall mechanisms of action of Clcn7 in normal and ADO2 osteoclasts and treatment options for ADO2 patients are limited.

Clcn7, in conjunction with osteopetrosis-associated membrane protein 1 (Ostm1), are transported to the ruffled border membrane for the acidification of the resorption lacunae, an extracellular lysosome that is essential for bone resorption [12–15]. Clcn7 activity also supports late endosomal/lysosomal vesicular transport [3, 16, 17] which is required for the acidification of lysosomes and the accumulation of lysosomal chloride [18]. Published electrophysiology literature suggests that the human disease-causing CLCN7 G215R mutation does not fully abolish Clcn7 chloride exchange activity but increases channel gating, thereby reducing overall chloride release [14, 15, 18, 19] and leading to defective endosomal-lysosomal trafficking in osteoclasts [18].

We previously confirmed the role of Clcn7 as a critical regulator of osteoclast function in vitro [20–22]. Further, we reported that mice with the CLCN7 G213R knock-in mutation (ADO2 mice) analogous to the human G215R mutation [20], exhibit significantly elevated bone mass. Further, osteoclasts from ADO2 human patients and ADO2 G213R mice display accelerated osteoclast formation/differentiation and exhibit defects in actin ring formation (sealing zone formation), leading to significantly reduced bone resorption activity in vitro [18]. Our previous investigations of ADO2 osteoclast mechanisms led us to examine the effects of chloroquine, a pH sensitive lysosome stimulating agent [23, 24]. Although we were unable to demonstrate a decrease in bone mass in chloroquine treated ADO2 mice, the studies demonstrated an important dose-dependent increase in ADO2 osteoclast formation and resorption activity in response to chloroquine in vitro [22]. Based on these findings, we investigated the potential effects of other intracellular modulators of lysosomal acidification in wild-type and ADO2 osteoclasts.

It has long been established that adenylate cyclase is critical for lysosomal acidification in osteoclasts and other cells [25]. Adenylate cyclase catalyzes the formation of cyclic adenosine monophosphate (cAMP), an important second messenger molecule that activates key cellular target proteins, including cAMP-dependent protein kinases (PKA) [26, 27], cyclic nucleotide-gated channels (CNGC) [28, 29] and exchange proteins directly activated by cAMP (Epac) [30]. Intracellular cAMP concentration is tightly regulated at the level of its synthesis by adenylate cyclase. In addition, the cAMP is hydrolyzed to its inactive 5′-monophosphate form by the actions of specific phosphodiesterase (PDEs) [31, 32]. There are numerous known isoforms of human PDE, many of which have specific subtypes contributing to the complexity of the PDE cellular responses. Of the three PDEs that are cAMP-specific, PDE4 has emerged as an important therapeutic target. To this end, rolipram, a potent inhibitor of PDE4, has been investigated for the treatment of chronic lung disease. Roflumilast is a newer generation PDE4 inhibitor and is a more selective competitive inhibitor of PDE4, without effects on PDE1, 2, 3 or 5 isoenzymes, leading to better clinical tolerance, and is clinically approved for moderate to severe chronic obstructive pulmonary disease (COPD) [33, 34]. Moreover, rolipram and roflumilast have been shown to increase the intensity and duration of cAMP-mediated signaling in vitro [35]. In the current study, we demonstrate that osteoclast-like cells (OCLs) generated from ADO2 mice exhibit reduced cAMP levels, which led us to investigate whether PDE4 inhibition by rolipram and roflumilast could rescue ADO2 osteoclast activity dysfunction in vitro. Our demonstrated herein, our in vitro findings support a role for the investigation of cAMP modulation in future therapeutic approaches to treat the ADO2 osteoclast dysfunction.

2. Material and Methods

2.1. Osteoclast Culture and TRAP Staining

Multi-nucleated osteoclast-like cells (OCLs) were prepared from bone marrow-derived macrophage/monocyte cells (BMMs) as previously described [22, 36]. Briefly, BMMs were collected from the marrow of tibias and femurs from 6–8 weeks old Clcn7 G213R+/+ mice (WT) and Clcn7 G213R+/− mice (ADO2). The nonadherent BMMs cells were collected after 24 h in culture with α-MEM containing 10% FBS. An equal number of cells from each mouse were cultured in a 96-well culture plate at a density of 9 × 104 cells/well in α-MEM containing 10% FBS plus 80 ng/ml RANKL (PeproTech, NJ, USA) and 10 ng/ml MCSF (PeproTech, NJ, USA). The cells were cultured for up to 5–6 days and then fixed in 4% formaldehyde for 10 min. Cytological demonstration of tartrate-resistant acid phosphatase (TRAP) activity was performed using the Leukocyte Acid Phosphatase (TRAP) Kit (Sigma-Aldrich, MO, USA). As described by the manufacturer, the kit employs the naphthol AS-Bl phosphate and freshly diazotized fast garnet GBC salt in the presence of L(+)-tartrate to detect tartaric acid-resistant acid phosphatase in cells, which stain purple. Tartrate-sensitive cells are devoid of activity and remain clear. Multinucleated cells containing three or more nuclei that are positive for TRAP activity (purple) were scored as mature OCLs. All images were taken using a Leica DMI 4000B microscope with (Leica Microsystems) ImagePro software. OCL size was measured in the same plates used to determine OCL number in osteoclastogenesis assays. Briefly, OCL assays were performed in 48-well plates using 3–4 replicate wells per condition and replicated 3 times. Each well contained approximately 100–300 OCLs (see Figure 4–5) and the average area of all cells in each well was calculated using ImagePro software. All OCL assays were performed using quadruplicate wells and the findings were replicated 3–5 times, with representative data shown in the figures. In addition, data from 2–5 individual experiments (depending on the drug concentration) were analyzed together (Supplementary Figure 2), which showed that the cumulative data supported the representative findings shown in Figures 4–5.

The PDE4 inhibitors rolipram and roflumilast were purchased from Sigma-Aldrich. As specified in the results and figures, the rolipram concentration used for osteoclastogenesis assays was 0, 100 and 200 nM. For resorption assays, we used 0, 20, 100 and 250 nM rolipram. The roflumilast concentrations used in our studies include 0, 10, 25, 50 and 100 nM depending on the assay as detailed in the figures.

2.2. Resorption Activity and Pit Staining

BMMs were cultured in the MCSF (10 ng/ml) and RANKL (80 ng/ml) for 5–6 days until mature OCLs with three or more nuclei appeared. Cells were treated with Trypsin-EDTA with gentle scraping and reseeded onto pre-wet cortical bone slices (Immunodiagnostic Systems, MD, USA) in 96-well plates. Cells were cultured on the bone slices with MCSF (10 ng/ml) and RANKL (80 ng/ml) for an additional 3 days. To stain osteoclast resorption pits, all OCLs on bone slices were removed using mechanical agitation, and the bone slices were then incubated with peroxidase-conjugated wheat germ agglutinin for 1 h, followed by the staining of the resorption pits (without OCLs) with 3,3′-diaminobenzidine (Sigma-Aldrich, MO, USA). The presence of resorption pits on each bone slice was confirmed microscopically (Leica DMI 4000B, Leica Microsystems). As we previously reported [37], osteoclast bone resorption activity was measured by the release of CTX-1 into conditioned media after culturing mature OCLs on bovine cortical bone slices for 3 days (CrossLaps® for Culture ELISA kit, Immunodiagnostic Systems, MD, USA). In parallel wells, the OCLs were assayed for TRAP activity and the number of TRAP-positive multinucleated cells was enumerated. The level of CTX-1 in the media was normalized by the number of mature TRAP-positive OCLs (CTX/OCL number).

2.3. Real-Time qPCR

Total RNA was isolated from cultured OCLs using RNeasy® Mini Kit (Qiagen, CA, USA). The complementary DNA (cDNA) was synthesized from 1 μg of RNA using High-Capacity cDNA Reverse Transcription Kit according to the manufacturer’s instructions (Applied Biosystems, CA, USA). Quantitative RT-PCR (qPCR) was performed on QuantStudio 7 Flex Real-Time PCR System using SYBR Green dye (Applied Biosystems, CA, USA). The amplification reaction was performed for 40 cycles with denaturation at 95°C for 10 min, followed by annealing at 95°C for 15 s and extension and detection at 60°C for 1 min. RNA and cDNA were prepared from 4 separate mice per genotype and individual cDNA samples were assayed using 2–3 technical replicates per qPCR reaction. GAPDH was used as the housekeeping gene. The threshold cycle (Ct) for each test gene was normalized against the GAPDH. For all graphs the ΔCt values (absolute mRNA values) are shown which were calculated using the equation ΔCt = Ct (target gene) - Ct (housekeeping gene). The following oligonucleotide primers were used to identify mouse PDE4 isoforms.

GAPDH forward 5’-GGTCGGTGTGAACGGATTTGGC-3’

GAPDH reverse 5’-GCAGTGATGGCATGGACTGTGG-3’

TRAP forward 5’-TGGCTGAGGAAGTCATCTGAGTTG-3’

TRAP reverse 5’-GACCACCTTGGCAATGTCTCT-3’

PDE4a forward 5’-AACGATGAGTCTGTGCTTGAG-3’

PDE4a reverse 5’-CTGAGGTTCTGGAAGATGTCG-3’

PDE4b forward 5’-AAGAGGAACACTGCGACATC-3’

PDE4b reverse 5’-TGTTTTAAGGTCTGCCAGGAG-3’

PDE4d forward 5’-GAAATCCCCTCTCCAACTCAG-3’

PDE4d reverse 5’-TCTTCCTGCTCCGTTTTAACC-3’

2.4. cAMP ELISA

BMMs were cultured in 6 well plates at a density of 2 × 106 cells/ml with MCSF (10 ng/ml) and RANKL (80 ng/ml) for 3 days (immature pre-OCLs) or 6 days (mature OCLs). Cells were washed with PBS and incubated in 0.1M HCl for 20 minutes at room temperature to lyse the cells and then centrifuged at 13,000 rpm for 10 min. The supernatant was collected and the protein concentration was determined using a Bicinchoninic Acid Kit (Pierce, IL, USA). An equal amount of each sample was used to determine cAMP concentration using the cAMP Direct Immunoassay Kit (Biovision, CA, USA) according to the manufacturer’s instructions. The absorbance at 450 nm was measured using a microplate reader. cAMP levels (pM) were determined by comparison to a cAMP standard provided in the kit.

For cAMP analysis of roflumilast treated OCL cultures, the BMMs from female WT and ADO2 mice (3 mice per group) were cultured separately in α-MEM with 10% FBS plus RANKL and MCSF for 3 days. The medium was removed and the cells briefly washed and incubated in α-MEM without FBS for 1 h and then replaced with serum free α-MEM containing 100 nM roflumilast for 0, 15, 30 and 60 min in serum free media. Cell lysates were prepared in ice and then assayed for cAMP using technical duplicates. cAMP levels (pM) per μg protein were determined by comparison to a cAMP standard.

2.5. Sample Size and Statistical Analyses.

Unless otherwise stated, BMMS from 2–3 male mice were used for individual experiments. OCLs assays were performed using 3 or 4 experimental replicates per treatment group per mouse. Representative figures from 3–5 independent experiments are shown. In supplementary Figures 2, the data from 2–5 individual experiments (depending on the drug concentration) was analyzed together for statistical significance. The ANOVA framework was used for statistical significance with p<0.05. For comparisons between two groups, the student t-test was used with significance value of p<0.05.

3. Results

3.1. ADO2 osteoclast dysfunction is associated with low cAMP levels.

As has been previously reported, CLCN7 mutation is associated with osteoclast hypofunction in humans [21, 38–40] and mouse models [20, 22], although the mechanism is not yet clear. Our previously reported data [20, 22] and new studies described herein demonstrate increased ADO2 osteoclastogenesis. To further explore the osteoclast resorption dysfunction associated with ADO2 disease, we focused on examining the important intracellular second messenger cAMP, which is produced by adenylate cyclase, and plays critical roles in intracellular signaling pathways in osteoclasts, and in particular with proton transport, lysosomal acidification, and osteoclast resorption activity [41–43].

We first cultured BMMs from ADO2 and WT mice with RANKL and MCSF in vitro and then compared OCL formation, size (area) and resorption activity between genotypes. Similar to our previous report [22], ADO2 cultures produced a higher number of OCLs that were larger in size and contained a larger number of nuclei compared to WT OCLs (Figure 1A–B), which was likely due to accelerated BMM differentiation and/or pre-osteoclast fusion in culture. Despite the increase in OCL number, ADO2 OCLs exhibit significantly reduced bone resorbing activity in vitro as shown by reduced staining of bone resorption pits as well as reduced CTX-1 levels in conditioned media from ADO2 OCLs on cortical bone slices, compared with WT OCLs (Figure 1C–D). Next, we examined whether cAMP levels were altered in ADO2 OCLs, which might provide a mechanistic understanding of the ADO2 OCL resorption dysfunction. We collected OCLs at early and mature stages of differentiation and performed a specific cAMP ELISA. As shown in Figure 1E–F, we established for the first time that both immature and mature ADO2 OCLs exhibit reduced intracellular cAMP levels, compared to WT OCLs. Based on these findings, we investigated known modulators of the cAMP pathway for their effects on ADO2 osteoclast formation, resorption, and survival, compared to WT.

Figure 1. Decreased ADO2 osteoclast activity is associated with decreased cAMP levels.

Figure 1.

Open in a new tab

A-F. WT and ADO2 BMMs were differentiated with MCSF and RANKL until multinucleated OCLs formed. A. Mature OCLs demonstrating TRAP activity (TRAP-positive) (purple) were counted. B. Osteoclast area (μm2) in day 5–6 cultures was measured using ImagePro software. Representative images of OCLs are shown. Scale bars are 20 μm. C-D. Mature OCLs were replated onto cortical bone discs and cultured for 2–3 days. Conditioned media from bone discs were assayed and CTX-1 (nM) was normalized for the number of TRAP-positive OCLs. D. Resorption pits were greater in WT than ADO2 OCLs. Vertical scale bars (top left) are 20 μm. E-F. OCLs were differentiated with MCSF and RANKL for 3 days (early OCs) or 6 days (mature OCs) and then lysed and assayed using a cAMP ELISA. Graphs represent mean ± SEM. Hash marks (#) indicate p<0.05 for ADO2 OCLs compared to WT OCLs.

3.2. Forskolin promotes ADO2 osteoclast activity with reduced osteoclast formation.

Forskolin is a well-established adenylate cyclase agonist which increases cAMP levels, and at low concentrations (10−6–10−7M) has been shown to increase osteoclast resorption activity in vitro [41]. To examine whether forskolin can promote ADO2 OCL formation, we performed WT and ADO2 osteoclastogenesis assays in the presence of 25 μM forskolin or vehicle. Mature multinucleated OCLs that stained positive for TRAP activity were counted (Figure 2A). In parallel studies, mature OCLs were replated onto bone discs for 2–3 days, after which OCLs were enumerated and bone resorption activity was determined using the CTX-1 ELISA (Figure 2B). In contrast to the inhibitory effects of forskolin on osteoclast differentiation, forskolin increased ADO2 osteoclast resorption activity when added to mature osteoclast (2–3 fold increase), although overall ADO2 activity remained below the level of WT OCLs. Given the positive increase in ADO2 resorption activity but unexpected inhibitory effects of forskolin on OCL formation, we proceeded to investigate alternative modulators of the cAMP pathway for their effects on ADO2 OCL formation and resorption.

Figure 2. Forskolin increases ADO2 OCL resorption activity.

Figure 2.

Open in a new tab

BMMs were differentiated with MCSF and RANKL in the presence or absence of 25 μM forskolin (F25), an activator of adenylate cyclase. Mature TRAP-positive OCLs were enumerated showing increased number after forskolin treatment. Osteoclast resorption activity on bone slices was measured by assay of CTX-I in conditioned media and then normalized for the number of TRAP-positive OCLs. Graphs are mean ± SEM. Asterisks (*) indicate p<0.05 for WT and ADO2 OCLs treated with forskolin or vehicle.

3.3. Elevated PDE4 mRNA expression in ADO2 OCLs.

As an alternative to using forskolin to stimulate adenylate cyclase activity, we examined the effects of blocking the cAMP hydrolysis pathway. The cAMP dependent PDEs catalyze the hydrolysis of cAMP to 5′ AMP [22, 44, 45] and inhibitors of PDE4 are known to increases cAMP levels [46, 47]. Given the reduced levels of cAMP in ADO2 OCLs, we examined whether the expression of the PDE4 isoforms is altered in ADO2 OCLs. QPCR analysis was used to examine the expression of the major PDE4 subtypes (PDE4 a, b, d) in WT versus ADO2 OCLs at different stages of OCL formation. As shown in Figure 3, all three PDE4 subtypes were elevated in ADO2 OCLs after 5 days. At this time point, numerous mature multinucleated OCLs are evident (see Figure 4 for representative images of day 5 OCLs). The data also reveal that PDE4b is the most highly expressed isotype in OCLs. Interestingly, TRAP mRNA was also elevated in mature ADO2 OCLs, which is consistent with the elevated TRAP activity seen by cytological staining for TRAP activity (Figure 1A). We speculate the increase in TRAP mRNA and TRAP activity may be compensatory responses to the abnormally low bone resorbing activity of ADO2 OCLs.

Figure 3. Increased PDE4 expression in ADO2 OCLs compared to WT OCLs.

Figure 3.

Open in a new tab

BMMs from WT and ADO2 mice were differentiated into OCLs and stopped at different days of culture, representing BMMs (day 0), immature OCLs (day 3) and mature-multinucleated OCLs (day 5). QPCR analysis was performed as indicated in the methods. Data shown is absolute mRNA. Graphs represent mean ± SEM and asterisks indicate p<0.05 for ADO2 OCLs compared to WT OCLs at each time point.

3.4. Rolipram stimulates ADO2 OCL formation and resorption activity.

Given the elevated levels of PDE4 isoforms in our ADO2 OCLs, we examined whether inhibition of PDE4 activity could promote ADO2 osteoclast formation and/or osteoclast activity. To this end, we first examined the effects the PDE4 inhibitor, rolipram. Published literature has reported the potency of rolipram to be in the nanomolar to subnanomolar range, depending on the cell type investigated [48, 49]. For example, rolipram has an IC50 of 313 +/− 6.7 nM in human monocytes [48] and 30–200 nM in human eosinophils [49], which provided the rationale for the drug concentrations tested in our in vitro OCL studies.

We examined the effects of rolipram of OCL formation using BMMs in ADO2 verses WT mice. In these studies, rolipram was added either at the start of the osteoclastogenesis assays or after mature OCLs had formed. When added at the start of the osteoclast cultures (Figure 4A) 100 nM and 200 nM rolipram led to a comparable increase in OCL number in the WT and ADO2 cultures. Specifically, 100 nM and 200 nM rolipram increased WT OCL formation by 24% and 21% while ADO2 OCL formation was increased 15% and 20%, respectively. Comparison between genotypes showed that ADO2 OCL number was 17% higher than WT at baseline (vehicle), 9% higher in 100 nM rolipram and 9% higher in 200 nM rolipram (Figure 4A). Moreover, the rolipram treated cultures contained significantly larger OCLs than vehicle controls, with the largest OCLs observed in ADO2 OCLs treated with 200 nM rolipram and represented a 5-fold increase in area compared to similarly treated WT OCLs (Supplementary Figure 1A).

Figure 4. Rolipram stimulates osteoclast resorption activity.

Figure 4.

Open in a new tab

A. WT and ADO2 OCLs were differentiated with the PDE4 antagonist, rolipram. After 5–6 days, multinucleated cells appeared and were fixed and assayed for TRAP activity. TRAP-positive multinucleated cells were then counted. Representative images of TRAP-positive OCLs are shown (right). Scale bar indicated is 50 μm. B. WT and ADO2 OCLs were differentiated until mature OCLs were formed, after which rolipram or vehicle were added, and the cells cultured for an additional 2–3 days. TRAP-positive OCLs were counted. Representative images of OCLs are shown (right). C. Mature OCLs were replated onto cortical bone slices and cultured in the presence of rolipram or vehicle for 2–3 days. Conditioned media was collected and assayed for CTX-I and data normalized for the number of TRAP-positive OCLs (nM/OC number). Resorption pits on cortical bone slices were stained and imaged, with representative pit images shown (right panel). Scale bar indicated is 20 μm. Mean ± SEM are shown. Asterisks (*) indicate p<0.05 compared to vehicle within one genotype, whereas hash marks (#) indicate statistical significance p<0.05 between genotypes at the same drug concentration.

We examined whether rolipram would affect OCL number when added at late stages of OCL formation when mature multinucleated OCLs were present (Figure 4B). Overall, rolipram increased OCL formation when added to mature OCLs, leading to a 25% and 15% increase in WT OCLs and a 12% and 5% increase in mature ADO2 OCLs at 100 nM and 200 nM rolipram, respectively. Compared to WT OCLs, ADO2 OCL number was overall higher in vehicle and rolipram treated groups. However, the data indicate a slight decrease in relative differences between ADO2 and WT after rolipram treatment; ADO2 OCL was 32% higher than WT OCLs in vehicle groups but only 18% (100 nm) and 11% (200 nM) in the rolipram treated groups.

We next investigated whether rolipram modulates ADO2 OCL resorption activity. For these studies, WT and ADO2 OCLs were grown until mature and then replated onto cortical bone slices in vitro. The OCLs were then cultured with rolipram or vehicle for up to 3 days. In the vehicle groups, ADO2 resorption activity was lower than WT OCLs. However, in the treatment groups, rolipram stimulated a concentration-dependent increase in bone resorption activity, with greater effects observed in ADO2 OCLs compared to WT OCLs (Figure 4C). This was demonstrated by both an increase in CTX-1/OC present in rolipram-treated ADO2 cultures (Figure 4C, left) and by the number of resorption pits formed on bone slices, compared to WT cultures (Figure 4C, right). Indeed, at the highest 250 nM rolipram concentration, ADO2 OCL resorption activity was 300% higher than WT, demonstrating that rolipram had a greater effect in ADO2 OCLs compared to WT.

3.5. Roflumilast stimulates ADO2 OCL formation and resorption to a greater extent than WT OCLs.

Roflumilast is reported to be more efficacious and a more highly selective inhibitor of PDE4 compared rolipram and is FDA approved to treat human COPD [34]. The potency and selectivity of roflumilast and its active metabolite against human recombinant PDE1–11 enzymes (except PDE6) were previously reported and shown to affect PDE4 in the range from 0.2–4.3 nM, depending on the PDE4 splice variant investigated [50, 51]. No effects on PDE1, 2, 3 or 5 isoenzyme activity were reported [50, 51]. Therefore, we compared the effects of roflumilast at 10, 25, 50, 100 and 150 nM on the osteoclastogenesis and bone resorption activity of WT versus ADO2 OCLs in vitro, compared to vehicle-treated cells.

As shown in Figure 5A, roflumilast added at the start of the osteoclastogenesis assays led to a concentration dependent increase in WT and ADO2 OCL number. WT OCLs were increased 26% and 56% while ADO2 OCL number was increased 14% and 26% with 50 nM and 100 nM roflumilast, respectively. Comparison between genotypes revealed osteoclastogenesis was 38%, 26% and 11% higher in ADO2 cultures compared with WT at 0, 50 and 100 nM roflumilast. The roflumilast treated cultures contained significantly larger OCLs than vehicle treated OCLs, with the largest sizes observed when ADO2 OCLs were cultured in 100 nM roflumilast. Roflumilast OCL size is shown in the supplemental material. Overall, the roflumilast findings were comparable to rolipram and occurred at slightly lower concentrations than rolipram. The larger size of ADO2 OCLs, compared to WT OCLs, suggests roflumilast and rolipram likely stimulated cell fusion without significant negative effects on cell apoptosis, although this was not specifically investigated in the current studies.

Figure 5. Roflumilast increases ADO2 OCL number and bone resorption activity in vitro.

Figure 5.

Open in a new tab

A. WT and ADO2 BMMs were differentiated in the presence of roflumilast, a PDE4 antagonist that increases cAMP. Once mature OCLs were formed, thy were assayed for TRAP activity, and TRAP-positive OCLs were imaged and counted. Representative images of OCLs are shown (right). B. WT and ADO2 BMMS were differentiated in the absence of roflumilast until mature OCLs were formed, after which roflumilast (or vehicle) was added, and the cells cultured for an additional 2–3 days. TRAP-positive multinucleated cells were then counted. Representative images of OCLs are shown (right) Scale bar indicated is 50 μm. C. Mature OCLs were replated onto cortical bone slices and cultured in the presence of roflumilast or vehicle for 2–3 days. Conditioned media was collected and assayed for CTX-I (nM) and the data normalized for the number of TRAP-positive OCLs. Resorption pits on cortical bone slices were stained and imaged, with representative pit images shown (right panel). Scale bar indicated is 20 μm. Mean ± SEM are shown. Asterisks (*) indicate p<0.05 compared to vehicle treated OCLs within a genotype, and hash marks (#) indicate statistical significance p<0.05 between genotypes at the same concentration of roflumilast.

Roflumilast added to mature OCLs also led to an increase in WT and ADO2 OCL number (Figure 5B). Compared to vehicle-treated WT cultures, 50 nM and 100 nM roflumilast led to a 14% and 30% increase in OCL number, respectively. Similarly, ADO2 OCL number was 19% and 30% higher with 50 nM and 100 nM roflumilast, respectively compared to the respective vehicle-treated group. Importantly, roflumilast also led to a robust concentration-dependent increase in ADO2 OCL resorption activity on bone slices, resulting in significantly increased CTX-I/OC levels and a greater number of resorption pits compared to WT OCLs (Figure 5C). At the highest roflumilast concentration (100 nM), ADO2 resorption activity was two-fold higher than WT OCLs.

3.5. Roflumilast rescues cAMP levels in ADO2 OCLs.

Given that antagonists of PDE4 such as roflumilast lead to the accumulation of intracellular cAMP, we investigated the mechanism of roflumilast-driven rescue of ADO2 OCL resorption activity by analyzing cAMP levels in early OCLs treated with roflumilast. As shown in Figure 6, cAMP levels in vehicle-treated WT and ADO2 groups were highest at 15 min and declined to baseline or below by 60 min suggesting that cAMP was short lived. Compared to WT OCLs, ADO2 OCLs treated with vehicle for 0 (−17%) and 15 min (−8%) showed slightly lower cAMP levels at the same time points, although in the 30 min vehicle ADO2 OCL group, cAMP levels were higher (+123%) than WT. Importantly, all cultures treated with roflumilast for 15, 30 and 60 min showed increased cAMP levels compared to their vehicle controls, with roflumilast having overall positive effects on both WT and ADO2 OCLs at all time points: 15 min (WT +87%; ADO2 +87%), 30 min (WT +312%; ADO2 +119%) and 60 min (WT +167%; ADO2 +125%). Overall, roflumilast robustly increased cAMP levels in ADO2 osteoclasts, and could potentially provide a mechanistic explanation for the roflumilast-stimulated increase in the bone resorbing activity of ADO2 OCLs.

Figure 6. Roflumilast increases intracellular cAMP in ADO2 and WT OCLs.

Figure 6.

Open in a new tab

BMMs from WT and ADO2 mice were cultured for 3 days with RANKL and MCSF to generate early OCLs (pre-OCLs) and then treated with 100 nM roflumilast for 0, 15, 30 and 60 min followed by ELISA assay of cAMP levels. The graph represents mean ± SEM. Statistical significance (p<0.05) for vehicle treated groups at different time points, compared to baseline (0 min) is indicated by asterisks (*) within each genotype. Differences between roflumilast-treated groups compared to vehicle at the same time points within a genotype is indicated with ampersand (&), and significance between ADO2 and WT at similar conditions/times is indicated with hash marks (#).

4. Discussion

The disease-causing CLCN7/Clcn7 mutation in humans (G215R) and mice (G213R) leads to osteopetrosis due to dysfunctional osteoclastic activity; bone resorption is significantly reduced in ADO2 osteoclasts [20, 21, 38, 39, 52, 53]. Disease causing mutations of CLCN7 in ADO2 humans have been reported to affect endosomal-lysosomal transport and acidification of osteoclasts, decreasing osteoclasts resorption activity [1, 15, 18, 54, 55]. Consistent with previous data, in vitro assays demonstrate that ADO2 OCLs are larger in size, show accelerated differentiation/fusion and exhibit defects in bone resorption activity (Figure 1). Our previous publication also focused on the osteoclast effects of chloroquine [22], a compound known to affect the intracellular pH of lysosomes and endosomes [56, 57]. We reported that chloroquine increases the resorption activity of ADO2 OCLs in vitro. Unfortunately, in those studies we were unable to demonstrate a reduction in the high bone mass phenotype of ADO2 mice in vivo at the concentration range tested, and further studies are necessary to better understand the mechanism of action of chloroquine in vivo [22].

Another important regulator of osteoclast function is the adenylate cyclase-cAMP pathway which is known to regulate exocytosis in a variety of cells [42, 58, 59]. This pathway is also implicated in the processing and secretion of Cathepsin K, a critical osteoclast resorption enzyme that catabolizes collagen matrix in bone [42]. Despite the larger size of ADO2 OCLs, our current studies demonstrate reduced intracellular cAMP levels in pre-OCLs and mature OCLs differentiated from ADO2 BMMs, compared with WT cells at similar days of culture (Figure 1). This finding and our previous studies led us to examine whether modulating cAMP activity in ADO2 OCLs could potentially restore bone resorption in these cells.

In the current study, we examined the effects of forskolin, a known activator of adenylate cyclase and cAMP levels, on the bone resorbing activity of WT and ADO2 OCLs. We chose low micro molar concentrations of forskolin, which were similar to concentrations tested by other groups using chicken osteoclasts [41]. Forskolin stimulated the bone resorption activity of both ADO2 and WT OCLs, supporting the involvement of the adenylate cyclase/cAMP pathway in the ADO2 osteoclast mechanism. However, forskolin exerted an overall negative effect on osteoclast differentiation, leading to an overall decrease in the number of TRAP-positive multinucleated OCLs, with a greater decrease observed in ADO2 cultures (Figure 2). Although disappointing, these findings were not surprising given the conflicting effects of forskolin in the literature [41]. In early reports, it was shown that forskolin exerts a biphasic effect on embryonal chicken osteoclasts. At low concentrations (10−6–10−7M) forskolin increases cAMP levels leading to an increase in basal osteoclast activity in vitro, whereas at higher concentrations forskolin inhibits osteoclast resorption [41]. Forskolin treatment in vivo enhanced bone formation, presumably through increasing cAMP in osteoblasts [60]. However, another study demonstrated that cAMP inhibits osteogenesis in vitro and bone formation ex vivo in rodent models [61]. It was suggested that cAMP duration, rather than the strength of the cAMP signal may in part explain some of the disparities of studies examining the actions of cAMP [62].

Adding to the complexity of cAMP responses that occur in different cell types, cAMP levels are regulated by enzymatic hydrolysis by PDEs. The family of PDE proteins consists of 11 members which are differentially expressed in different cell types and regulate the levels of the cyclic nucleotides, cAMP and cGMP. Overall, PDE4, PDE7 and PDE8 are cAMP-specific members of the PDE4 family of proteins. Several members of the PDE family also exist as subtypes, including PDE4 which has 4 major subtypes (PDE4 a, b, c, d) [31]. Moreover, compartmentalization of PDE4 to cell microdomains via binding proteins can affect overall cellular responses [31, 63, 64]. Nevertheless, the diversity of PDE4 expression and actions has been exploited for the development of several specific and selective PDE4 antagonists which inhibit PDE4 activity, consequently increasing cAMP levels. One therapeutic application of PDE4 antagonists is for the treatment of COPD [33, 34]. Of relevance to bone, COPD is associated with several negative sequelae, including osteoporosis. Recent studies have suggested that both decreased bone mineral density and impaired bone quality contribute to bone fragility and fracture risk in COPD patients [33].

Consistent with our observation of reduced basal levels of cAMP in ADO2 OCLs, we demonstrated elevated expression of three PDE4 subtypes (PDE4 a, b, d) in ADO2 OCLs compared to WT OCLs (Figure 3). Based on these data, we examined the effects of the PDE4 antagonist, rolipram on WT and ADO2 OCL differentiation and bone resorbing activity in vitro. Rolipram was originally developed as an antidepressant and although it was subsequently abandoned for that disease indication, it is currently being investigated for neurodegenerative diseases such as Alzheimer’s disease, in part because of its ability to accelerate proteosome activity to dispose of neurotoxic protein aggregates [65, 66]. In addition to rolipram, we examined the effects of roflumilast, a selective PDE4 inhibitor, which can inhibit all PDE4 isoforms to some degree. Roflumilast is currently FDA approved to treat patients with COPD [33].

Importantly, in our studies both rolipram (20–250 nM) and roflumilast (10–100 nM) significantly increased ADO2 OCL resorption activity to a greater extent than WT OCLs (Figures 4 and 5). Moreover, no inhibitory effect on OCL formation was observed by either rolipram or roflumilast at the doses tested. In fact, rolipram and roflumilast increased OCL number when added either at the start of osteoclastogenesis or once mature OCLs had formed, resulting in the formation of large OCLs, especially in ADO2 cultures. These data are consistent with a previous report showing rolipram stimulates osteoclast formation in culture [67]. Both rolipram and roflumilast appeared to be more efficacious in increasing the bone resorbing activity of ADO2 OCLs compared to WT OCLs (Figure 4–5), as shown by an increase in CTX-I products in conditioned media. Given the CTX-I data was normalized for OCL number, the increased resorption activity of ADO2 osteoclasts was not simply due to an increase in ADO2 OCL number and may reflect changes in intracellular signaling.

To further examine the potential intracellular mechanism behind the effects of the roflumilast, a PDE4 antagonist, on ADO2 bone resorption activity, we analyzed cAMP levels. As shown in Figure 6, roflumilast led to a robust increase in cAMP levels in both WT and ADO2 OCLs within 15 minutes of treatment. The elevated cAMP levels declined slightly by 60 min but remained significantly higher than vehicle treated samples for all time points investigated. Our data also indicate the kinetics for cAMP decline are restored to normal with inhibition of PDE4 (Figure 6).

Figure 7 summarizes the key findings from our studies demonstrating reduced cAMP in ADO2 OCLs, which was correlated with increased PDE4 expression. Importantly, we found that ADO2 overall osteoclast resorption activity is increased by forskolin, an adenylate cyclase activator, and with the PDE4 antagonists, rolipram and roflumilast, which inhibit cAMP hydrolysis. Moreover, inhibition of PDE4 with roflumilast led to increased cAMP levels, which were correlated with increased bone resorption activity of WT OCLs and to a greater extent in ADO2 OCLs.

Figure 7. Modulation of the cAMP pathway by adenylate cyclase stimulation or inhibition of PDE4 promotes ADO2 OCL activity.

Figure 7.

Open in a new tab

Schematic summary of our findings demonstrating reduced cAMP in ADO2 OCLs, and that the resorption activity of ADO2 OCLs was increased by forskolin, an adenylate cyclase (AC) activator, and with the specific PDE4 antagonists, rolipram and roflumilast, which act by inhibiting cAMP hydrolysis. Restoring cAMP levels in ADO2 OCLs by inhibition of PDE4 may be part of the mechanism to rescue the bone resorption activity of ADO2 osteoclasts.

Taken together, our studies suggest that cAMP pathway and its regulation by PDE4 plays a central role in the mechanism leading to ADO2 osteoclast dysfunction. Based on the positive osteoclastic effect of roflumilast described herein, we performed a subsequent in vivo study to investigate whether the PDE4 inhibitor, roflumilast, is effective in reducing the bone mass of ADO2 mice [68]. The complete findings of the animal study are described in the separate publication in this journal (Alam et al, 2024) [68]. Our future studies will investigate the roflumilast mechanism leading to ADO2 OCL rescue. In addition, approaches that alter cAMP kinetics in ADO2 OCLs by modulating adenylate cyclase cAMP synthesis, PDE4/cAMP activity, intracellular compartmentalization and signaling will be pursued with the goal to identify novel strategies with minimal cytotoxicity and appropriate systemic bioavailability to rescue ADO2 osteoclast dysfunction in vivo. Ongoing efforts to understand the cellular mechanisms underlying ADO2 osteoclast dysfunction are a critical foundation for future advances in the therapeutic approaches to treat individuals with CLCN7 mediated osteopetrosis.

Supplementary Material

Supplementary Fig 1

Supplementary Figure 1. Rolipram and roflumilast increase the size of ADO2 OCL. BMMs were cultured with the PDE4 inhibitors, rolipram or roflumilast, for 5–6 days to form mature multinucleated OCLs. The area (size) of TRAP-positive multi-nucleated OCLs were determined (shown as μm2 × 103) determined using ImagePro Software. OCLs were generated from 4 individual mice. OCLs in 4 replicate wells per mouse were scored, with the mean ± SEM shown in the graphs. Please see Figures 4A and 5A for comparison graphs and images of TRAP-positive OCLs in vehicle and rolipram or roflumilast treated groups, respectively.

Supplementary Fig 2

Supplementary Figure 2. Rolipram and roflumilast stimulate ADO2 OCL resorption activity. Mature multinucleated OCLs from WT and ADO2 BMMs were cultured on cortical bone slices in the presence of A) rolipram or B) roflumilast, two PDE4 antagonists. Conditioned media was assayed for CTX-I and the data normalized for the number of TRAP-positive OCLs. The data from 2–5 independent experiments was consolidated and analyzed together for statistical significance, with the 0 drug WT and ADO2 data representing 5 independent experiments. Please refer to Figures 4–5 for representative data graphs with their respective OCLs images. Graphs represent mean ± SD. Asterisks (*) indicates p<0.05 for comparison to the vehicle control for each genotype, whereas hash (#) indicates statistical significance for comparison between genotypes at matching drug concentrations (p<0.05).

Acknowledgments.

We thank previous student members of our respective laboratories for technical assistance with osteoclast assays. These studies were funded in part by funds from the school of dentistry to AB.

Grant support

This work was supported by the US National Institutes of Health grants AR069583 (MJE) and AR080076 (AB)

Footnotes

Disclosure statement

The authors declare that they have no conflict of interest.

REFERENCES

  • 1.Stauber T, Weinert S, Jentsch TJ (2012) Cell biology and physiology of CLC chloride channels and transporters. Comprehensive Physiology 2:1701–1744 [DOI] [PubMed] [Google Scholar]
  • 2.Poroca DR, Pelis RM, Chappe VM (2017) ClC Channels and Transporters: Structure, Physiological Functions, and Implications in Human Chloride Channelopathies. Front Pharmacol 8:151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Jentsch TJ, Pusch M (2018) CLC Chloride Channels and Transporters: Structure, Function, Physiology, and Disease. Physiol Rev 98:1493–1590 [DOI] [PubMed] [Google Scholar]
  • 4.Jentsch TJ, Stein V, Weinreich F, Zdebik AA (2002) Molecular structure and physiological function of chloride channels. Physiol Rev 82:503–568 [DOI] [PubMed] [Google Scholar]
  • 5.Jentsch TJ, Poët M, Fuhrmann JC, Zdebik AA (2005) Physiological functions of CLC Cl- channels gleaned from human genetic disease and mouse models. Annu Rev Physiol 67:779–807 [DOI] [PubMed] [Google Scholar]
  • 6.Graves AR, Curran PK, Smith CL, Mindell JA (2008) The Cl-/H+ antiporter ClC-7 is the primary chloride permeation pathway in lysosomes. Nature 453:788–792 [DOI] [PubMed] [Google Scholar]
  • 7.Guzman RE, Grieschat M, Fahlke C, Alekov AK (2013) ClC-3 is an intracellular chloride/proton exchanger with large voltage-dependent nonlinear capacitance. ACS Chem Neurosci 4:994–1003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Matsuda JJ, Filali MS, Volk KA, Collins MM, Moreland JG, Lamb FS (2008) Overexpression of CLC-3 in HEK293T cells yields novel currents that are pH dependent. Am J Physiol Cell Physiol 294:C251–262 [DOI] [PubMed] [Google Scholar]
  • 9.Neagoe I, Stauber T, Fidzinski P, Bergsdorf EY, Jentsch TJ (2010) The late endosomal ClC-6 mediates proton/chloride countertransport in heterologous plasma membrane expression. J Biol Chem 285:21689–21697 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Picollo A, Pusch M (2005) Chloride/proton antiporter activity of mammalian CLC proteins ClC-4 and ClC-5. Nature 436:420–423 [DOI] [PubMed] [Google Scholar]
  • 11.Scheel O, Zdebik AA, Lourdel S, Jentsch TJ (2005) Voltage-dependent electrogenic chloride/proton exchange by endosomal CLC proteins. Nature 436:424–427 [DOI] [PubMed] [Google Scholar]
  • 12.Lange PF, Wartosch L, Jentsch TJ, Fuhrmann JC (2006) ClC-7 requires Ostm1 as a beta-subunit to support bone resorption and lysosomal function. Nature 440:220–223 [DOI] [PubMed] [Google Scholar]
  • 13.Meadows NA, Sharma SM, Faulkner GJ, Ostrowski MC, Hume DA, Cassady AI (2007) The expression of Clcn7 and Ostm1 in osteoclasts is coregulated by microphthalmia transcription factor. J Biol Chem 282:1891–1904 [DOI] [PubMed] [Google Scholar]
  • 14.Leisle L, Ludwig CF, Wagner FA, Jentsch TJ, Stauber T (2011) ClC-7 is a slowly voltage-gated 2Cl(−)/1H(+)-exchanger and requires Ostm1 for transport activity. Embo j 30:2140–2152 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Weinert S, Jabs S, Hohensee S, Chan WL, Kornak U, Jentsch TJ (2014) Transport activity and presence of ClC-7/Ostm1 complex account for different cellular functions. EMBO Rep 15:784–791 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Brandt S, Jentsch TJ (1995) ClC-6 and ClC-7 are two novel broadly expressed members of the CLC chloride channel family. FEBS Lett 377:15–20 [DOI] [PubMed] [Google Scholar]
  • 17.Kornak U, Kasper D, Bösl MR, Kaiser E, Schweizer M, Schulz A, Friedrich W, Delling G, Jentsch TJ (2001) Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and man. Cell 104:205–215 [DOI] [PubMed] [Google Scholar]
  • 18.Schulz P, Werner J, Stauber T, Henriksen K, Fendler K (2010) The G215R mutation in the Cl-/H+-antiporter ClC-7 found in ADO II osteopetrosis does not abolish function but causes a severe trafficking defect. PloS one 5:e12585. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Henriksen K, Gram J, Schaller S, Dahl BH, Dziegiel MH, Bollerslev J, Karsdal MA (2004) Characterization of osteoclasts from patients harboring a G215R mutation in ClC-7 causing autosomal dominant osteopetrosis type II. Am J Pathol 164:1537–1545 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Alam I, Gray AK, Chu K, Ichikawa S, Mohammad KS, Capannolo M, Capulli M, Maurizi A, Muraca M, Teti A, Econs MJ, Del Fattore A (2014) Generation of the first autosomal dominant osteopetrosis type II (ADO2) disease models. Bone 59:66–75 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Chu K, Snyder R, Econs MJ (2006) Disease status in autosomal dominant osteopetrosis type 2 is determined by osteoclastic properties. J Bone Miner Res 21:1089–1097 [DOI] [PubMed] [Google Scholar]
  • 22.Alam I, Gerard-O’Riley RL, Acton D, Hardman SL, Hong JM, Bruzzaniti A, Econs MJ (2021) Chloroquine increases osteoclast activity in vitro but does not improve the osteopetrotic bone phenotype of ADO2 mice. Bone 153:116160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Delaisse JM, Søe K, Andersen TL, Rojek AM, Marcussen N (2021) The Mechanism Switching the Osteoclast From Short to Long Duration Bone Resorption. Front Cell Dev Biol 9:644503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.von Kleist L, Haucke V (2012) At the crossroads of chemistry and cell biology: inhibiting membrane traffic by small molecules. Traffic 13:495–504 [DOI] [PubMed] [Google Scholar]
  • 25.Rahman N, Ramos-Espiritu L, Milner TA, Buck J, Levin LR (2016) Soluble adenylyl cyclase is essential for proper lysosomal acidification. J Gen Physiol 148:325–339 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Tasken K, Aandahl EM (2004) Localized effects of cAMP mediated by distinct routes of protein kinase A. Physiol Rev 84:137–167 [DOI] [PubMed] [Google Scholar]
  • 27.Taylor SS, Buechler JA, Yonemoto W (1990) cAMP-dependent protein kinase: framework for a diverse family of regulatory enzymes. Annual review of biochemistry 59:971–1005 [DOI] [PubMed] [Google Scholar]
  • 28.Biel M, Zong X, Ludwig A, Sautter A, Hofmann F (1999) Structure and function of cyclic nucleotide-gated channels. Rev Physiol Biochem Pharmacol 135:151–171 [DOI] [PubMed] [Google Scholar]
  • 29.Finn JT, Grunwald ME, Yau KW (1996) Cyclic nucleotide-gated ion channels: an extended family with diverse functions. Annu Rev Physiol 58:395–426 [DOI] [PubMed] [Google Scholar]
  • 30.Bos JL (2003) Epac: a new cAMP target and new avenues in cAMP research. Nat Rev Mol Cell Biol 4:733–738 [DOI] [PubMed] [Google Scholar]
  • 31.Houslay MD, Adams DR (2003) PDE4 cAMP phosphodiesterases: modular enzymes that orchestrate signalling cross-talk, desensitization and compartmentalization. Biochem J 370:1–18 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Soderling SH, Beavo JA (2000) Regulation of cAMP and cGMP signaling: new phosphodiesterases and new functions. Curr Opin Cell Biol 12:174–179 [DOI] [PubMed] [Google Scholar]
  • 33.Inoue D, Watanabe R, Okazaki R (2016) COPD and osteoporosis: links, risks, and treatment challenges. Int J Chron Obstruct Pulmon Dis 11:637–648 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Kawamatawong T (2017) Roles of roflumilast, a selective phosphodiesterase 4 inhibitor, in airway diseases. Journal of thoracic disease 9:1144–1154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Scuvee-Moreau J, Giesbers I, Dresse A (1987) Effect of rolipram, a phosphodiesterase inhibitor and potential antidepressant, on the firing rate of central monoaminergic neurons in the rat. Arch Int Pharmacodyn Ther 288:43–49 [PubMed] [Google Scholar]
  • 36.Huang S, Eleniste PP, Wayakanon K, Mandela P, Eipper BA, Mains RE, Allen MR, Bruzzaniti A (2014) The Rho-GEF Kalirin regulates bone mass and the function of osteoblasts and osteoclasts. Bone 60:235–245 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Kang KS, Hong JM, Horan DJ, Lim KE, Bullock WA, Bruzzaniti A, Hann S, Warman ML, Robling AG (2019) Induction of Lrp5 HBM-causing mutations in Cathepsin-K expressing cells alters bone metabolism. Bone 120:166–175 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Chen X, Zhang K, Hock J, Wang C, Yu X (2016) Enhanced but hypofunctional osteoclastogenesis in an autosomal dominant osteopetrosis type II case carrying a c.1856C>T mutation in CLCN7. Bone Res 4:16035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Alatalo SL, Ivaska KK, Waguespack SG, Econs MJ, Vaananen HK, Halleen JM (2004) Osteoclast-derived serum tartrate-resistant acid phosphatase 5b in Albers-Schonberg disease (type II autosomal dominant osteopetrosis). Clinical chemistry 50:883–890 [DOI] [PubMed] [Google Scholar]
  • 40.Alam I, McQueen AK, Acton D, Reilly AM, Gerard-O’Riley RL, Oakes DK, Kasipathi C, Huffer A, Wright WB, Econs MJ (2017) Phenotypic severity of autosomal dominant osteopetrosis type II (ADO2) mice on different genetic backgrounds recapitulates the features of human disease. Bone 94:34–41 [DOI] [PubMed] [Google Scholar]
  • 41.Murrills RJ, Dempster DW (1990) The effects of stimulators of intracellular cyclic AMP on rat and chick osteoclasts in vitro: validation of a simplified light microscope assay of bone resorption. Bone 11:333–344 [DOI] [PubMed] [Google Scholar]
  • 42.Park YG, Kim YH, Kang SK, Kim CH (2006) cAMP-PKA signaling pathway regulates bone resorption mediated by processing of cathepsin K in cultured mouse osteoclasts. Int Immunopharmacol 6:947–956 [DOI] [PubMed] [Google Scholar]
  • 43.Ramaswamy G, Kim H, Zhang D, Lounev V, Wu JY, Choi Y, Kaplan FS, Pignolo RJ, Shore EM (2017) Gsalpha Controls Cortical Bone Quality by Regulating Osteoclast Differentiation via cAMP/PKA and beta-Catenin Pathways. Sci Rep 7:45140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.McSorley T, Stefan E, Henn V, Wiesner B, Baillie GS, Houslay MD, Rosenthal W, Klussmann E (2006) Spatial organisation of AKAP18 and PDE4 isoforms in renal collecting duct principal cells. Eur J Cell Biol 85:673–678 [DOI] [PubMed] [Google Scholar]
  • 45.Stefan E, Wiesner B, Baillie GS, Mollajew R, Henn V, Lorenz D, Furkert J, Santamaria K, Nedvetsky P, Hundsrucker C, Beyermann M, Krause E, Pohl P, Gall I, MacIntyre AN, Bachmann S, Houslay MD, Rosenthal W, Klussmann E (2007) Compartmentalization of cAMP-dependent signaling by phosphodiesterase-4D is involved in the regulation of vasopressin-mediated water reabsorption in renal principal cells. J Am Soc Nephrol 18:199–212 [DOI] [PubMed] [Google Scholar]
  • 46.Takami M, Cho ES, Lee SY, Kamijo R, Yim M (2005) Phosphodiesterase inhibitors stimulate osteoclast formation via TRANCE/RANKL expression in osteoblasts: possible involvement of ERK and p38 MAPK pathways. FEBS Lett 579:832–838 [DOI] [PubMed] [Google Scholar]
  • 47.Park H, No AL, Lee JM, Chen L, Lee SY, Lee DS, Yim M (2010) PDE4 inhibitor upregulates PTH-induced osteoclast formation via CRE-mediated COX-2 expression in osteoblasts. FEBS Lett 584:173–180 [DOI] [PubMed] [Google Scholar]
  • 48.Souness JE, Griffin M, Maslen C, Ebsworth K, Scott LC, Pollock K, Palfreyman MN, Karlsson JA (1996) Evidence that cyclic AMP phosphodiesterase inhibitors suppress TNF alpha generation from human monocytes by interacting with a ‘low-affinity’ phosphodiesterase 4 conformer. Br J Pharmacol 118:649–658 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Tenor H, Hatzelmann A, Church MK, Schudt C, Shute JK (1996) Effects of theophylline and rolipram on leukotriene C4 (LTC4) synthesis and chemotaxis of human eosinophils from normal and atopic subjects. Br J Pharmacol 118:1727–1735 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Hatzelmann A, Morcillo EJ, Lungarella G, Adnot S, Sanjar S, Beume R, Schudt C, Tenor H (2010) The preclinical pharmacology of roflumilast--a selective, oral phosphodiesterase 4 inhibitor in development for chronic obstructive pulmonary disease. Pulmonary pharmacology & therapeutics 23:235–256 [DOI] [PubMed] [Google Scholar]
  • 51.Rabe KF (2011) Update on roflumilast, a phosphodiesterase 4 inhibitor for the treatment of chronic obstructive pulmonary disease. Br J Pharmacol 163:53–67 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Stauber T, Wartosch L, Vishnolia S, Schulz A, Kornak U (2023) CLCN7, a gene shared by autosomal recessive and autosomal dominant osteopetrosis. Bone 168:116639. [DOI] [PubMed] [Google Scholar]
  • 53.Waguespack SG, Koller DL, White KE, Fishburn T, Carn G, Buckwalter KA, Johnson M, Kocisko M, Evans WE, Foroud T, Econs MJ (2003) Chloride channel 7 (ClCN7) gene mutations and autosomal dominant osteopetrosis, type II. J Bone Miner Res 18:1513–1518 [DOI] [PubMed] [Google Scholar]
  • 54.Henriksen K, Sørensen MG, Jensen VK, Dziegiel MH, Nosjean O, Karsdal MA (2008) Ion transporters involved in acidification of the resorption lacuna in osteoclasts. Calcif Tissue Int 83:230–242 [DOI] [PubMed] [Google Scholar]
  • 55.Henriksen K, Sorensen MG, Nielsen RH, Gram J, Schaller S, Dziegiel MH, Everts V, Bollerslev J, Karsdal MA (2006) Degradation of the organic phase of bone by osteoclasts: a secondary role for lysosomal acidification. J.Bone Miner.Res. 21:58–66 [DOI] [PubMed] [Google Scholar]
  • 56.Al-Bari AA, Al Mamun A (2020) Current advances in regulation of bone homeostasis. FASEB Bioadv 2:668–679 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Krogstad DJ, Schlesinger PH (1987) The basis of antimalarial action: non-weak base effects of chloroquine on acid vesicle pH. Am J Trop Med Hyg 36:213–220 [DOI] [PubMed] [Google Scholar]
  • 58.Collins MP, Forgac M (2020) Regulation and function of V-ATPases in physiology and disease. Biochim Biophys Acta Biomembr 1862:183341. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Nicholson GC, Moseley JM, Yates AJ, Martin TJ (1987) Control of cyclic adenosine 3’,5’-monophosphate production in osteoclasts: calcitonin-induced persistent activation and homologous desensitization of adenylate cyclase. Endocrinology 120:1902–1908 [DOI] [PubMed] [Google Scholar]
  • 60.Doorn J, Siddappa R, van Blitterswijk CA, de Boer J (2012) Forskolin enhances in vivo bone formation by human mesenchymal stromal cells. Tissue Eng Part A 18:558–567 [DOI] [PubMed] [Google Scholar]
  • 61.Siddappa R, Mulder W, Steeghs I, van de Klundert C, Fernandes H, Liu J, Arends R, van Blitterswijk C, de Boer J (2009) cAMP/PKA signaling inhibits osteogenic differentiation and bone formation in rodent models. Tissue Eng Part A 15:2135–2143 [DOI] [PubMed] [Google Scholar]
  • 62.Siddappa R, Doorn J, Liu J, Langerwerf E, Arends R, van Blitterswijk C, de Boer J (2010) Timing, rather than the concentration of cyclic AMP, correlates to osteogenic differentiation of human mesenchymal stem cells. J Tissue Eng Regen Med 4:356–365 [DOI] [PubMed] [Google Scholar]
  • 63.Mongillo M, McSorley T, Evellin S, Sood A, Lissandron V, Terrin A, Huston E, Hannawacker A, Lohse MJ, Pozzan T, Houslay MD, Zaccolo M (2004) Fluorescence resonance energy transfer-based analysis of cAMP dynamics in live neonatal rat cardiac myocytes reveals distinct functions of compartmentalized phosphodiesterases. Circ Res 95:67–75 [DOI] [PubMed] [Google Scholar]
  • 64.Perry SJ, Baillie GS, Kohout TA, McPhee I, Magiera MM, Ang KL, Miller WE, McLean AJ, Conti M, Houslay MD, Lefkowitz RJ (2002) Targeting of cyclic AMP degradation to beta 2-adrenergic receptors by beta-arrestins. Science 298:834–836 [DOI] [PubMed] [Google Scholar]
  • 65.Ainatzoglou A, Stamoula E, Dardalas I, Siafis S, Papazisis G (2021) The Effects of PDE Inhibitors on Multiple Sclerosis: a Review of in vitro and in vivo Models. Curr Pharm Des 27:2387–2397 [DOI] [PubMed] [Google Scholar]
  • 66.Bundschuh DS, Eltze M, Barsig J, Wollin L, Hatzelmann A, Beume R (2001) In vivo efficacy in airway disease models of roflumilast, a novel orally active PDE4 inhibitor. J Pharmacol Exp Ther 297:280–290 [PubMed] [Google Scholar]
  • 67.Cho ES, Yu JH, Kim MS, Yim M (2004) Rolipram, a phosphodiesterase 4 inhibitor, stimulates osteoclast formation by inducing TRANCE expression in mouse calvarial cells. Archives of pharmacal research 27:1258–1262 [DOI] [PubMed] [Google Scholar]
  • 68.Alam I, Hardman SL, Gerard-O’Riley RL, Acton D, Parker RS, Hong JM, Bruzzaniti A, Econs MJ. Effect of Roflumilast, a Selective PDE4 Inhibitor, on Bone Phenotypes in ADO2 Mice. Calcif Tissue Int. 2024. Feb 1. doi: 10.1007/s00223-023-01180-2. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Fig 1

Supplementary Figure 1. Rolipram and roflumilast increase the size of ADO2 OCL. BMMs were cultured with the PDE4 inhibitors, rolipram or roflumilast, for 5–6 days to form mature multinucleated OCLs. The area (size) of TRAP-positive multi-nucleated OCLs were determined (shown as μm2 × 103) determined using ImagePro Software. OCLs were generated from 4 individual mice. OCLs in 4 replicate wells per mouse were scored, with the mean ± SEM shown in the graphs. Please see Figures 4A and 5A for comparison graphs and images of TRAP-positive OCLs in vehicle and rolipram or roflumilast treated groups, respectively.

NIHMS1999840-supplement-Supplementary_Fig_1.tif (542.9KB, tif)
Supplementary Fig 2

Supplementary Figure 2. Rolipram and roflumilast stimulate ADO2 OCL resorption activity. Mature multinucleated OCLs from WT and ADO2 BMMs were cultured on cortical bone slices in the presence of A) rolipram or B) roflumilast, two PDE4 antagonists. Conditioned media was assayed for CTX-I and the data normalized for the number of TRAP-positive OCLs. The data from 2–5 independent experiments was consolidated and analyzed together for statistical significance, with the 0 drug WT and ADO2 data representing 5 independent experiments. Please refer to Figures 4–5 for representative data graphs with their respective OCLs images. Graphs represent mean ± SD. Asterisks (*) indicates p<0.05 for comparison to the vehicle control for each genotype, whereas hash (#) indicates statistical significance for comparison between genotypes at matching drug concentrations (p<0.05).

NIHMS1999840-supplement-Supplementary_Fig_2.tif (535.8KB, tif)

RESOURCES