KMN-159, a novel EP4 receptor selective agonist, stimulates osteoblastic differentiation in cultured whole rat bone marrow

Abstract

KMN-159 is the lead compound from a series of novel difluorolactam prostanoid EP₄ receptor agonists aimed at inducing local bone formation while avoiding the inherent side effects of systemic EP₄ activation. KMN-159 is a potent, selective small molecule possessing pharmacokinetic properties amenable to local administration. Unfractionated rat bone marrow cells (BMCs) were treated once at plating with escalating doses of KMN-159 (1 pM to 10 μM). The resulting elevated alkaline phosphatase (ALP) levels measured 9 days post-dose are consistent with increased osteoblastic differentiation and exposure to KMN-159 at low nanomolar concentrations for as little as 30 min was sufficient to induce complete osteoblast differentiation of the BMCs from both sexes and regardless of age. ALP induction was blocked by an EP₄ receptor antagonist but not by EP₁ or EP₂ receptor antagonists and was not induced by EP₂ or EP₃ receptor agonists. Addition of BMCs to plates coated with KMN-159 24 days earlier resulted in ALP activation, highlighting the chemical stability of the compound. The expression of phenotype markers such as ALP, type I collagen, and osteocalcin was significantly elevated throughout the osteoblastic differentiation timecourse initiated by KMN-159 stimulation. An increased number of tartrate-resistant acid phosphatase-positive cells was observed KMN-159 or PGE₂ treated BMCs but only in the presence of exogenous receptor activator of nuclear factor kappa-Β ligand (RANKL). No change in the number of adipocytes was observed. KMN-159 also increased bone healing in a rat calvarial defect model with a healing rate equivalent to recombinant human bone morphogenetic protein-2. Our studies show that KMN-159 is able to stimulate osteoblastic differentiation with a very short time of exposure, supporting its potential as a therapeutic candidate for augmenting bone mass.

Keywords: Bone; Bone marrow cells; EP4 agonist; Osteoblasts; Osteogenesis.

Figure 1:. Dose responsive activation of ALP enzyme activity in young female rat BMCs by EP₄ receptor agonists.

BMCs isolated from young female rats were treated once at plating with varying doses of the indicated compounds. Cells were cultured and ALP enzyme activity determined on day 9 as described in the Methods. Compounds tested were (A) PGE₂; (B) PGE₁; (C) 11-deoxy-PGE₂; (D) 11-deoxy-PGE₁; (E) 11-deoxy-8-aza-PGE₁; (F) KMN-165; (G) KMN-80; and (H) KMN-159. EC₅₀ values are expressed as the mean ± SEM of (n) replicates. Statistical analysis of the differences was performed using ANOVA followed by Tukey test. P < 0.01 as compared to ^aPGE₂, ^bPGE₁, ^c11-deoxyPGE₂, ^d11-deoxyPGE₁, ^e11-deoxy-8-aza PGE_1, ^fKMN-165, ^gKMN-80. P < 0.05 as compared to ^a’PGE₂, ^b’PGE₁, ^c’11-deoxyPGE₂.

Figure 2:. Dose responsive activation of ALP enzyme activity in old female or young male rat BMCs by EP₄ receptor agonists.

BMCs isolated from old female rats (A-D) or young male rats (E-H) were treated once at plating with varying doses of the indicated compounds. Cells were cultured and ALP enzyme activity determined on day 9 as described in the Methods. Compounds tested were (A, E) PGE₂; (B, F) PGE₁; (C, G) KMN-80; and (D, H) KMN-159. EC₅₀ values are expressed as the mean ± SEM of (n) replicates. Statistical analysis of the differences was performed using ANOVA followed by Tukey test. P < 0.01 as compared to aPGE₂, bPGE₁, cKMN-80; P < 0.05 as compared to a’PGE₂, c’11-deoxyPGE₂ in the same animal group.

Figure 3:. Treatment of rat BMCs with PGE₂ or KMN-159 does not change total adherent cell number in culture but increases the ALP positive area.

BMCs were treated once at plating with either vehicle (●), PGE₂ (1μM; ■), or KMN-159 (1μM; ▲). On day 4 after plating, cultures were fed with 50% plating volume of fresh medium containing dexamethasone (10 nM final). (A) On days 5, 6, and 7, a 24-well plate was fixed in methanol and air dried. All plates were then stained with DAPI and the number of cells quantitated by counting nuclei using the BioTek Cytation system (n = 8 wells / treatment / day). Statistical analysis of the differences was performed using 2-way ANOVA followed by Tukey test and shows no effect of either treatment. P < 0.001 among the different time points. (B) On day 9, one 6-well plate for each treatment was fixed and stained for ALP activity. The ALP positive area was quantitated using the BioTek Cytation system (n = 6 wells / treatment). Statistical analysis of the differences was performed using ANOVA followed by Tukey test. ** P < 0.01 and * P < 0.05 as compared to vehicle control and n.s. indicates non-significance.

Figure 4:. Stimulation of Osteoblastic Differentiation by KMN-159 in Rat Bone Marrow Cells Occurs via EP₄ Activation.

(A) BMCs were treated once at plating with the indicated doses of either butaprost, an EP₂ receptor selective agonist (▲); sulprostone, an EP₃ receptor selective agonist (■); or our EP₄ selective agonist KMN-159 (●). (B) BMCs were treated with the indicated doses of KMN-159 following a 30-minute pretreatment with either vehicle (●); ONO-8711, an EP₁ selective antagonist (▲); PF-04418948, an EP₂ selective antagonist (■); or ONO-AE3–208, an EP₄ selective antagonist (○). Cells were then cultured and ALP enzyme activity determined on day 9 as described in the Materials and Methods. No induction of ALP enzyme was observed when the BMCs were treated with the antagonist compounds alone (data not shown).

Figure 5:. KMN-159 is Stable at Room Temperature and Stimulates Osteogenesis in BMCs After a Short Exposure.

(A) KMN-159 was diluted in ethanol to the final concentrations noted and added to the wells of 24-well plates (n = 8 wells / dilution). Ethanol vehicle alone was added to a corresponding set of plates. The ethanol solvent was evaporated overnight and the plates stored for 24 days at room temperature in room light. BMCs were then added to the wells, fed on day 4, harvested on day 9, and ALP enzyme activity assayed. (B, C) BMCs were treated with either 1μM KMN-159 or vehicle (0.1% ethanol) and incubated in suspension at 37°C for the indicated times. KMN-159 was removed and cells were then plated as described in Methods for determination of ALP activity (n = 8 wells / treatment / time) or mineralization (n = 6 wells / treatment / time). All cultures were fed with fresh medium containing dexamethasone (10 nM final) on day 4. ALP activity is expressed as fold increase following KMN-159 treatment over the corresponding vehicle treated wells (B). Assessment of mineralization was performed only at 30 or 60 minute treatments. Cultures were fed with osteogenic media on day 7 after plating and every 3 days thereafter, and fixed on day 21. Mineralization assessed by von Kossa staining (C). Quantification of the mineralization (D) is expressed as the total number of pixels per well that were von Kossa staining positive (■ vehicle treated; ■ KMN-159 treated). Statistical analysis of the differences was performed using 2-way ANOVA followed by Sidak test. Effect of treatment ** P < 0.01.

Figure 6:. KMN-159 Stimulates Expression of Osteoblast Phenotype Marker Genes.

BMCs isolated from female rats (<6 months) were treated once at plating with 1μM KMN-159 (■) or vehicle (0.1% ethanol; ■) and plated in 12-well plates. On day 4 after plating, all cultures were fed with fresh medium containing dexamethasone (10 nM final). Beginning on day 7 after plating and every 3 days thereafter, cultures were fed with osteogenic medium. RNA was prepared as indicated in Methods. Expression of GAPDH was used as a reference. Statistical analysis of the differences was performed using 2-way ANOVA followed by Tukey test. P < 0.001 effect of time and treatment. * P < 0.05; ** P < 0.01 and *** P < 0.001 as compared to vehicle control at the same timepoint.

Figure 7:. KMN-159 Stimulates Osteoclastogenesis but not Adipogenesis in vitro.

(A, B) For assessment of osteoclastogenesis, BMCs isolated from male rats (<6 months) were treated once at plating with the indicated doses of KMN-159, PGE₂, or vehicle (0.1% ethanol) in phenol red-free MEMα supplemented with 10% FCS and 16 ng/mL each RANKL and M-CSF and plated in 24-well plates (n = 5 – 6 wells). Cells were fixed on day 14 and stained for TRAP enzyme activity. Statistical analysis of the differences was performed using ANOVA followed by Tukey test. * P < 0.05; *** P < 0.001 as compared to vehicle control. (C) For assessment of adipogenesis, BMCs isolated from male rats (<6 months) were treated at plating and at medium replacement with 1μM KMN-159, 1μM PGE₂, or vehicle (0.1% ethanol) in MEMα supplemented with 10% FCS, 10 μg/mL insulin, and 1 μM dexamethasone (n = 8 wells). Cultures were fixed on day 12 and stained for lipids with Oil Red O. Comparison of Oil Red O staining between vehicle or either PGE₂ or KMN-159 treated cells showed no significant change.

Figure 8:. KMN-159 Stimulates Repair of a Calvarial Defect in Rats.

The ability of KMN-159 to accelerate bone formation in vivo was tested using a 2.6 mm cranial osteotomy model in 3 month old female rats. Vehicle (●), KMN-159 (■) and rhBMP-2 (▲) were prepared in CPC containing DBM as described in Methods. Following placement of the CPC-DBM mixtures into the defects, the rats were scanned weekly for 8 weeks using dental cone beam CT scanner. The area of non-mineralized defect was determined each week, compared to the area measured in the same animal immediately after surgery (week 0), and the percent defect repair calculated. Data points are expressed as mean ± SEM (n = 5). Statistical analysis of the differences was performed using 2-way ANOVA followed by Tukey test. P < 0.001 effect of time and treatment. ** P < 0.05 (KMN-159 compared to vehicle control) and *** P < 0.001 (KMN-159 and rhBMP-2 as compared to vehicle control at the same time point).