Regulatory mechanism of CCN2 production by serotonin (5-HT) via 5-HT2A and 5-HT2B receptors in chondrocytes

Abstract

Serotonin (5-hydroxytryptamine: 5-HT) is recognized as a neurotransmitter in the central nerve system and as a regulator of systemic blood pressure in the peripheral tissues. Recently, it was reported that 5-HT2 receptors (5-HT2Rs) were expressed in cartilage tissues lacking both vessels and neurons, suggesting possible novel functions of 5-HT during cartilage development and regeneration. Our previous data indicated that CCN family protein 2/connective tissue growth factor (CCN2/CTGF) plays a central role in cartilage development and regeneration. Therefore, the aim of this study was to investigate the effect of 5-HT on the production of CCN2 in chondrocytes. Firstly, we showed that the mRNAs of 5-HT2R subtypes 5-HT2AR and 5-HT2BR, were expressed in a human chondrocytic cell line, HCS-2/8; however, 5-HT2CR mRNA was not detected. In addition, exogenously added 5-HT did not affect the 5-HT2AR and 5-HT2BR expressions. Next, we demonstrated that CCN2 production was increased by treatment with a 5-HT2AR agonist and the combination of 5-HT and 5-HT2BR antagonist. In contrast, treatment with a 5-HT2BR agonist and the combination of 5-HT and 5-HT2AR antagonist decreased CCN2 production. Furthermore, we showed that phosphorylation of Akt and p38 MAPK were increased by treatment with 5-HT2AR agonist, and that phosphorylation of PKCε, PKCζ, ERK1/2 and JNK were increased by treatment with 5-HT2BR agonist. Finally, we found that 5-HT2AR was localized in the growth plate, whereas 5-HT2BR was localized in the articular cartilage. These findings suggest that 5-HT promotes CCN2 production through the 5-HT2AR in growth plates, and that it represses CCN2 production through the 5-HT2BR in articular cartilage for harmonized development of long bones.

Fig 1. Effect of 5-HT on the gene expression and protein production of 5-HT₂ receptor in HCS-2/8 cells.

(A) After HCS-2/8 cells had reached confluence, they were treated with 5-HT at a concentration of 10 μM for 24 h. Then, total RNA was isolated and real-time RT-PCR analysis was performed by using specific primers for GAPDH, TpH-1, 5-HT_2AR, 5-HT_2BR and 5-HT_2CR. The ordinate indicates the relative ratio with respect to GAPDH expression, and data represents mean and standard deviation of culture with (n = 4) or without 5-HT (n = 4). The gene expressions levels of 5-HT_2AR, 5-HT_2BR, and TpH-1 were confirmed in HCS-2/8 cells, whereas 5-HT_2CR was not detected (ND). In addition, these gene expressions were not affected by the treatment with 5-HT. (B) After HCS-2/8 cells had reached confluence, they were treated with 5-HT for 24 h. Then, Western blot analysis was performed by using antibodies recognizing the indicated proteins. The apparent molecular weights of 5-HT_2AR were 75 kDa (arrow) and 50 kDa (arrowhead). The major band of 5-HT_2BR indicated a molecular weight 55 kDa (arrow), and minor bands were found at 30 kDa (arrowheads). The major band of 5-HTT was at 53 kDa (arrow), and a minor band at 75 kDa (arrowhead). Exogenous 5-HT added had no effect on the production of these proteins. Histone H3 was used as a loading control.

Fig 2. Real-time RT-PCR analysis of COL2A1, ACAN, and CCN2 mRNAs in HCS-2/8 cells treated with BW723C86, an agonist of 5-HT_2BR.

(A) After HCS-2/8 cells had reached confluence, they were treated with BW723C86 at a concentration of 10 μM for 12 h. Then, total RNA was collected; and real-time RT-PCR analysis was performed. The amounts of these mRNAs were normalized to that amount of GAPDH mRNA. The graphs show the expression levels of (a) COL2a1, (b) ACAN, (c) CCN2 after incubation with (n = 6) or without BW723C86 (n = 6). (B) HCS-2/8 cells were treated with BW723C86 at a concentration of 10 μM for 24 h. The graphs show the expression levels of (a) COL2a1, (b) ACAN, (c) CCN2 after incubation with (n = 3) or without BW723C86 (n = 3). In all graphs, the ordinate indicates the relative ratio with respect to untreated sample (ratio = 1.0), and bars represent mean and standard deviation. The data were analyzed by Student t-test, and p < 0.05 (*) was considered significant.

Fig 3. Real-time RT-PCR analysis of COL2A1, ACAN, and CCN2 mRNAs in HCS-2/8 cells treated with the combination of 5-HT and SB204741, an antagonist of 5-HT_2BR or retanserin, an antagonist of 5-HT_2AR.

(A) HCS-2/8 cells were grown until they had reached confluence. Then, the cells were treated with 5-HT (10 μM), ritanserin, an antagonist of 5-HT_2AR (100 μM), SB204741, an antagonist of 5-HT_2BR (30 μM) or the combination of 5-HT and SB206553 or ritanserin. After 12 h, total RNA was collected; and real-time RT-PCR analysis was then performed. The amounts of these mRNAs were normalized to that amount of GAPDH mRNA. The graphs show the expression levels of (a) COL2a1, (b) ACAN, (c) CCN2 mRNAs in HCS-2/8 cells treated with vehicle control (PBS and/or DMSO, n = 4), 5-HT (n = 4), ritanserin alone (n = 3), the combination of 5-HT and ritanserin (n = 3), SB204741 alone (n = 4), or the combination of 5-HT and SB204741 (n = 4). (B) HCS-2/8 cells were treated with 5-HT, ritanserin, SB204741, or the combination of 5-HT and SB206553 or ritanserin for 24 h. The graphs show the expression levels of (a) COL2a1, (b) ACAN, (c) CCN2 in HCS-2/8 cells treated with vehicle control (PBS and/or DMSO, n = 4), 5-HT (n = 4), ritanserin alone (n = 4), the combination of 5-HT and ritanserin (n = 4), SB204741 alone (n = 4), or the combination of 5-HT and SB204741 (n = 4). In all graphs, the ordinate indicates fold induction with respect to control sample (ratio = 1.0), and bars represent mean and standard deviation. The data were analyzed by Tukey’s test for multiple comparisons, and p < 0.05 (*), p < 0.01 (**) was considered significant.

Fig 4. Protein production of CCN2 regulated by 5-HT signaling via 5-HT_2AR and 5-HT_2BR.

HCS-2/8 cells were grown until they had reached confluence. (A) The cells were treated with vehicle control (PBS and/or DMSO), 5-HT, ritanserin alone, the combination of 5-HT and ritanserin, SB204741 alone, or the combination of 5-HT and SB204741 at the indicated dose. Cell lysates were prepared 24 h later, and Western blot analysis was performed with anti-CCN2 and β-actin antibodies. The graph indicates relative densitometry to untreated controls (ratio = 1.0; dotted line) from 5 independent cultures and analyzed by Bonferroni’s test, and p < 0.05 (*) was considered significant. (B) HCS-2/8 cells were treated with 5-HT alone (10 μM), NBOH-2C-CN (10 μM or 100 μM) or BW723C86 (1 μM or 10 μM) for 24 h. Then cell lysates were prepared, and Western blot analysis was performed. Relative densitometry (untreated control = 1.0; dotted line) from 5 independent cultures are presented and analyzed by Bonferroni’s test, and p < 0.05 (*) was considered significant. (C) HCS-2/8 cells were treated with 5-HT at the concentration of 10 μM for 24 h, and the cell culture supernatant and cell layer fraction were harvested. Quantification of 5-HT was performed by using an ELISA system. The concentration of 5-HT produced by HCS-2/8 cells was determined by subtracting the 5-HT concentration in fresh media from that in conditioned media. Results are presented as the mean and standard deviations of 3 independent cultures and analyzed by Bonferroni’s test, and p < 0.05 (*) was considered significant. (D) HCS-2/8 cells were treated with 5-HT at the concentration of 10 and 50 μM for 24 h, and Western blot analysis was performed. The graph indicates relative densitometry to untreated controls (ratio = 1.0; dotted line) from 3 measurements and analyzed by Dunnett’s test, and there was no significant difference.

Fig 5. Activation of Ca²⁺ influx, and phosphorylation of Akt and PKCs in HCS-2/8 cells stimulated with 5-HT, and agonists of 5-HT_2AR and 5-HT_2BR.

(A) After HCS-2/8 cells had reached sub-confluence, the cells were pre-treated with Fluo-4AM (final concentration; 3 mmol/l) in a recording medium at 37°C. After 20 min, the culture medium was replaced with recording medium without Fluo-4AM; and these cells were then treated with 5-HT (10 μM), NBOH-2C-CN (100 μM) or BW723C86 (10 μM) for 1 min. Photographs of the cells were taken under a fluorescence microscope (upper panels). The same field was visualized by phase-contrast microscopy (lower panels). The bar represents 50 μm. (B; a) Time course of fluorescence intensity measured by using a Fluo-4 AM in HCS-2/8 cells treated with 5-HT, NBOH-2C-CN or BW723C86. The ordinate indicates the ratio of fluorescence intensity with respect to untreated sample (ratio = 1.0). Arrow indicates stimulation point (time = 0). (b) Bar graph shows the ratio of fluorescent intensity of each group at 60 seconds after treatment (dotted line in panel “a”). Results are presented as the mean and standard deviations of 8 independent cultures and analyzed by Bonferroni’s test, and p < 0.05 (*) was considered significant. (C-F) After HCS-2/8 cells had reached confluence, the cells were treated with 5-HT or the indicated agonist at the final concentrations shown. After 5 min, cell lysates were prepared; and Western blot analysis was then performed with antibodies recognizing the indicated proteins. (C) The level of phosphorylated Akt (n = 3) was increased by the treatment with NBOH-2C-CN (10 μM and 100 μM) and (D) PKCε (n = 4) and (E) PKCζ (n = 3) phosphorylation levels were increased by the treatment with BW723C86 (10 μM). (F) The level of phosphorylated PKCα (n = 3) showed no change. The total amounts of conventional PKCα (PKCα), novel PKCε (PKCε), and atypical PKCζ (PKCζ) remained unchanged by any treatment. The graph indicates relative densitometry (untreated control = 1.0; dotted line) from 3 measurements and analyzed by Dunnett’s test, and p < 0.05 (*) was considered significant.

Fig 6. Activation of MAPKs in HCS-2/8 cells treated with 5-HT or agonist of each 5-HT₂ receptor and rescue effect of inhibitors of ERK1/2 and JNK on the BW723C86-decreased CCN2 production.

(A) HCS-2/8 cells were grown until they had reached confluence. Then, the cells were treated with 5-HT, NBOH-2C-CN or BW723C86 at the indicated concentrations. After 15 min, cell lysates were prepared; and Western blot analysis was performed with the antibodies against the indicated proteins. The levels of phosphorylated ERK1/2 and JNK were increased by the treatment with BW723C86 at a concentration of 10 μM. In contrast, the levels of phosphorylated p38 MAPK were increased by the treatment with NBOH-2C-CN at the concentrations of 10 μM and 100 μM. (B) HCS-2/8 cells were grown until they had reached confluence. Then, when the cells were treated with 5-HT (10 μM) or BW723C86 (10 μM), PD98059 (MEK1 inhibitor; 50 μM) or SP600125 (JNK inhibitor; 50 μM) was applied to the cultures simultaneously. After 24 h, the cell lysates were prepared; and Western blot analysis was performed with anti-human CCN2 rabbit serum and β-actin antibody. When HCS-2/8 cells were treated with 5-HT, PD98059 or SP600125 had no effects. In contrast, when the cells were treated with BW723C86, CCN2 production was rescued by either PD98059 or SP600125. The graphs give the results of Western blotting using anti-CCN2 antibody and quantified by densitometric analysis, with normalization by the levels of β-actin. The ordinate indicates the fold change relative to untreated controls (ratio = 1.0; dotted line). The graph indicates relative densitometry (untreated control = 1.0; dotted line) from 6 independent cultures and analyzed by Bonferroni's test, and p < 0.05 (*) was considered significant.

Fig 7. Immunohistochemical analysis of whole knee joints from 60 day-old male mice by use of anti-5-HT_2AR and 5-HT_2BR antibodies.

(A-C) Sections of the frontal knee joints were stained with toluidine blue, and cartilage tissues showed metachromatic staining. The areas surrounded by the boxes are enlarged in “B” (articular cartilage tissues) and “C” (growth plate). (D-H) In the low-power magnification view of the knee joint stained with anti-5-HT_2AR (D) the areas surrounded by the boxes are enlarged (E, H). Images of “E” and “F” represent articular cartilage tissues (E) and the growth plate (F). A serial section was stained with a non-immune antibody as a negative control, and images of the same areas as seen in “E” and “F” are shown in “G” and “M”, respectively. The immunoreactivity for 5-HT_2AR was detected in cells from the proliferating to prehypertrophic regions of the growth plate. (I-M) The knee joint stained with anti-5-HT_2BR. In the low-power-magnification view (I), the areas indicated by the boxes are enlarged in “J” and “K”. Images in “J” and “K” represent articular cartilage tissues (J) and the growth plate (K). Images in “L” and “M” represent the same areas as seen in “J” and “K,” respectively, in a serial section stained with a non-immune antibody as a negative control. The immunoreactivity for 5-HT_2BR was detected in the surface layer of articular cartilage tissues. The sizes of scale bars are indicated.

Fig 8. Schematic representation of possible intracellular 5-HT signaling pathways in chondrocytes.

Results newly obtained in the present study are summarized. Since 5-HT_2AR is localized in the growth plate, 5-HT signaling via 5-HT_2AR induces Ca²⁺ influx. Then, p38 MAPK is activated by phosphorylated Akt; as a result, CCN2 production is increased. On the other hand, since 5-HT_2BR is localized in articular cartilage tissues, 5-HT signaling via 5-HT_2BR induces Ca²⁺ influx similar to 5-HT signaling via 5-HT_2AR. Then, the phosphorylated ERK1/2 level is increased through both activation of PKCε and PKCζ; as a result, CCN2 production is decreased. Furthermore, independent of PKC, 5-HT may transmit a certain unknown signal to JNK, which also inhibits the CCN2 production in chondrocytes.