SEARCH

SEARCH BY CITATION

Keywords:

  • 5-HT2B receptor;
  • RARβ;
  • limb bud;
  • micromass culture;
  • [35S]sulfate

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Retinoic acid (RA) synthesizing and metabolizing enzymes are coordinately expressed with serotonin 2B (5-HT2B) receptors at sites of epithelial–mesenchymal (E-M) interaction in the mouse embryo (Bhasin et al., 1999). The promoter of the 5-HT2B receptor contains potential RA response element (RAREs) as well as an AP-2 site. Because both retinoid and serotonergic signaling have been implicated in the regulation of chondrogenic differentiation, the present study investigated whether these signals may work together to regulate this morphogenetic process in hindlimb bud micromass cultures. Results indicate that 5-HT promotes [35S]sulfate incorporation (chondrogenic differentiation) by activation of 5-HT2B receptors, which use the mitogen activated protein kinase (p42 MAPK) signal transduction pathway, whereas RA dose-dependently inhibits sulfate incorporation and promotes expression of RARβ, which could lead to inhibition of p38 MAPK. No evidence was found to support the possibility that RA negatively regulates expression of 5-HT2B receptors. Taken together, these results suggest that 5-HT and RA may act as opposing signals to regulate chondrogenic differentiation in the developing hindlimb, possibly mediated by different MAPK signal transduction pathways. Developmental Dynamics 230:201–209, 2004. © 2004 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Serotonin (5-HT) has been shown to act as a developmental signal early in mouse embryogenesis (Lauder, 1988, 1993). At this time, 5-HT (derived from the maternal–embryonic circulation) regulates proliferation, migration, gene expression, and morphogenesis of neural crest-derived cell populations (Shuey et al., 1992, 1993; Yavarone et al., 1993a, b; Moiseiwitsch and Lauder, 1995; Choi et al., 1997, 1998; Moiseiwitsch et al., 1998; Moiseiwitsch, 2000; Nebigil et al., 2000b, 2001). In the craniofacial region, epithelial 5-HT uptake sites coincide with areas of chondrogenic differentiation in mesenchyme (Lauder et al., 1988). In mandibular micromass cultures (MMCs) grown in serum-containing medium, 5-HT receptor antagonists have been shown to differentially down-regulate cartilage proteoglycan core protein, a marker of cartilage matrix (Moiseiwitsch and Lauder, 1997). In serum-free MMCs, 5-HT exerts dose-dependent stimulatory effects on levels of insulin-like growth factor-1 (IGF-1), and inhibitory effects on cell proliferation, mediated by 5-HT receptors that activate the cyclic adenosine monophosphate/protein kinase A (cAMP/PKA) pathway (Lambert and Lauder, 1999; Lambert et al., 2001). Activation of 5-HT receptors that either negatively regulate cAMP or activate the Gq-coupled 5-HT2B receptor, promotes cell proliferation in these cultures (Buznikov et al., 2001). Stimulatory effects of 5-HT on proliferation of mandibular mesenchyme cells are consistent with mitogenic effects of 5-HT in other tissues and cell lines (Lee et al., 1991, 1999; Loric et al., 1995; Fanburg and Lee, 1997; Launay et al., 1998), including periosteal fibroblasts (Westbroek et al., 2001).

Because many molecules involved in craniofacial development have also been shown to play a role in limb development (Helms et al., 1997; Schneider et al., 1999), and 5-HT receptors are expressed in osteoblasts, osteoclasts, and osteoblast precursor cells (Bliziotes et al., 2001; Westbroek et al., 2001), effects of 5-HT on chondrogenic differentiation in the hindlimb bud were investigated.

The limb bud has been extensively used to study chondrogenic differentiation and skeletal development. Early work demonstrated that RA affects limb development in a stage- and dose-dependent manner (Shenefelt, 1972; Kochhar, 1973). Studies using either limb bud explants or dissociated cell cultures have shown that chondrogenic differentiation is dose-dependently inhibited by RA (Jiang et al., 1995; Underhill and Weston, 1998; Underhill et al., 2001). Defects caused by excess RA in vivo are similar to those caused by exposing cultured mouse embryos to ritanserin (Choi et al., 1997; Fiorica-Howells et al., 2000), a pan–5-HT2 receptor antagonist that blocks 5-HT2A, 5-HT2B, and 5-HT2C receptors (Bonhaus et al., 1995; Fiorica-Howells et al., 2000). Of interest, the promoter of the 5-HT2B receptor contains several potential retinoid response elements (Choi et al., 1997). RA-related molecules, including RA receptors, cellular RA binding protein 1 (CRABP1), and retinaldehyde dehydrogenase 2 (RALDH2, a RA synthesizing enzyme), are coordinately expressed with 5-HT2B receptors in the craniofacial region, heart, and limb buds of the mouse embryo (Bhasin et al., 1999). These observations have led to speculation about a possible functional relationship between 5-HT2B receptor and RA signaling during mouse embryogenesis (Choi et al., 1997; Lauder et al., 2000). To address this question, the present study investigates the effects of 5-HT2B receptor activation and RA signaling on chondrogenic differentiation in MMCs derived from embryonic day (E) 11.5 mouse hindlimb bud.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Serotonin and Chondrogenic Differentiation in Hindlimb Micromass Cultures

Serum-free hindlimb MMCs were treated with different doses of 5-HT (10−6 M, 10−8 M), together with nialamide (a monoamine oxidase inhibitor) and L-cysteine (an antioxidant), to prevent degradation of 5-HT. Both doses of 5-HT significantly increased [35S]sulfate incorporation, indicative of maturation of cartilage matrix (chondrogenic differentiation; Bagi and Burger, 1989), compared with vehicle controls and cultures treated with nialamide and L-cysteine (NC, Fig. 1). IGF-1 was used as a positive control, because this growth factor has been shown to promote chondrogenesis in the developing limb (Geduspan and Solursh, 1993). As expected, IGF-1 significantly promoted [35S]sulfate incorporation (Fig. 1).

thumbnail image

Figure 1. Serotonin (5-HT) promotes chondrogenic differentiation in hindlimb micromass cultures. Cultures were plated in serum-containing medium, then switched to serum-free medium at 1 day in vitro (1 DIV). Cultures were incubated with serum-free medium (control), nialamide 10−5 M plus L-cysteine 10−5 M (NC), 5-HT 10−8 M + NC, 5-HT 10−6 M + NC, or insulin-like growth factor-1 (IGF-1, 20 ng/ml; positive control) from 2 to 4 DIV. Fifteen minutes after treatment on 4 DIV, cultures were incubated with [35S]sulfate for 6 hr. [35S]Sulfate incorporation was determined by liquid scintillation counting. Asterisks indicate P < 0.05 analysis of variance, Dunnet's post hoc test). The graph represents data from six independent experiments; n = number of cultures analyzed per treatment group.

Download figure to PowerPoint

Serotonin Receptors and Chondrogenic Differentiation

The 5-HT2B receptor agonist BW723C86 (BW, Kennett et al., 1998) promoted [35S]sulfate incorporation (Fig. 2) to a similar extent as IGF-1 and 5-HT (Fig. 1). The 5-HT2B/2C antagonist SB206553 (SB, Forbes et al., 1996; Kennett et al., 1996) completely blocked the effects of both 5-HT and BW (Fig. 2), although it had no significant effects by itself, indicating that the stimulatory effects of 5-HT were mediated by the 5-HT2B receptor. Total protein was measured in MMCs treated with BW. No significant increase in protein was found (data not shown), indicating that BW did not promote cell proliferation.

thumbnail image

Figure 2. Activation of serotonin 2B (5-HT2B) receptor promotes chondrogenic differentiation in hindlimb micromass cultures. Cultures were plated in serum-containing medium then switched to serum-free medium at 1 day in vitro (DIV). Cultures were incubated with serum-free medium (control), 5-HT 10−8 M + nialamide 10−5 M plus L-cysteine 10−5 M (5-HT NC), BW 723C86 10−8 M (BW, a 5-HT2B R agonist), 5-HT 10−8 M + NC + SB206553 10−8 M (5-HT NC SB, a 5-HT2B/2C R antagonist), BW 10−8 M + SB 10−8 M (BW SB), or SB 10−8 M alone from 2 to 4 DIV. Fifteen minutes after treatment on 4 DIV, cultures were incubated with [35S]sulfate for 6 hr. [35S]Sulfate incorporation was determined by liquid scintillation counting. Asterisks indicate P < 0.05 analysis of variance (Dunnet's post hoc test). The graph represents data from five independent experiments; n = number of cultures analyzed per treatment group.

Download figure to PowerPoint

Expression of 5-HT2B Receptors in Micromass Cultures

Expression of 5-HT2B receptor protein in hindlimb MMCs was determined by using standard immunocytochemical methods, as described by Lambert and Lauder (1999). 5-HT2B receptor immunoreactivity was present in hindlimb MMCs as early as 1 day in vitro (DIV; data not shown). At 2 DIV, 5-HT2B–immunoreactive cells were located within prechondrogenic foci, as previously reported for mandibular MMCs (Buznikov et al., 2001), as well as in cells still dispersed in the monolayer (Fig. 3A,B). Changes in levels of 5-HT2B receptor protein during the culture period (1–4 DIV) were determined by using 125I-labeled protein A immunobinding (IB), as described (Lambert and Lauder, 1999; Lambert et al., 2001). Levels of 5-HT2B receptor protein increased significantly between 1 and 3 DIV, with no further increases evident at 4 DIV (Fig. 3C).

thumbnail image

Figure 3. Serotonin 2B (5-HT2B) receptor immunocytochemistry and 125I-labeled protein A immunobinding in hindlimb micromass cultures. 5-HT2B receptor-immunoreactive cells are concentrated in chondrogenic foci but are also scattered throughout the monolayer (arrows). A: At 2 days in vitro (original magnification, 180×). B: Enlargement of the boxed area in A (original magnification, 450×). C:125I-Labeled protein A immunobinding reveals 5-HT2B receptor immunoreactivity increases significantly from 1 to 3 DIV. Scale bars = 100 × in A, 40 μm in B.

Download figure to PowerPoint

5-HT2B Receptors, Chondrogenic Differentiation, and the p42 MAPK Pathway

The Gq-coupled 5-HT2B receptor activates the Ras/Raf pathway, leading to phosphorylation of MEK-1/2 and activation of p42/44 MAPK (Launay et al., 1996). To determine whether this pathway was involved in promoting chondrogenic differentiation ([35S]sulfate incorporation), after activation of the 5-HT2B receptor, hindlimb cultures were treated with either 5-HT + NC, or BW, together with PD98059 (PD), a specific MEK inhibitor, because phosphorylated MEK (pMEK) is the direct upstream activator of p42/44 MAPK (Zhang et al., 1997). PD blocked the stimulatory effects of 5-HT (10-8 M) as well as BW on [35S]sulfate incorporation but had no significant effect by itself (Fig. 4). Western blotting demonstrated that BW promotes MEK phosphorylation, whereas PD blocks this effect (Fig. 5). These results indicate that activation of the p42/44 MAPK pathway is involved in the positive regulation of chondrogenic differentiation by 5-HT2B receptors in hindlimb cultures.

thumbnail image

Figure 4. MEK inhibitor blocks effects of serotonin 2B (5-HT2B) receptor activation on chondrogenic differentiation. Cultures were plated in serum-containing medium, then switched to serum-free medium at 1 day in vitro (DIV). Cultures were incubated with serum-free medium (control), 5-HT 10180−8 M + nialamide 10−5 M plus L-cysteine 10−5 M (5-HT NC), BW 723C86 10−8 M (BW, a 5-HT2B R agonist), 5-HT 10−8 M + NC + PD98059 10−8 M (5-HT NC PD, an MEK inhibitor), BW 10−8 M + PD 10−8 M (BW PD), or PD 10−8 M alone from 2 to 4 DIV. Fifteen minutes after treatment on 4 DIV, cultures were incubated with [35S]sulfate for 6 hr. [35S]Sulfate incorporation was determined by liquid scintillation counting. Asterisks indicate P < 0.05 analysis of variance (Dunnet's post hoc test). The graph represents data from 10 independent experiments; n = number of cultures analyzed per treatment group.

Download figure to PowerPoint

thumbnail image

Figure 5. PD98059 (PD) blocks promotion of MEK phosphorylation by the serotonin 2B (5-HT2B) agonist BW 723C86 (BW). Cultures were plated in serum-containing medium, then switched to serum-free medium at 1 days in vitro (DIV). Cultures were incubated with serum-free medium (control), 5-HT 10−8 M + NC, BW 10−8 M, 5-HT 10−8 M + NC + PD98059 10−8 M (PD, a MEK inhibitor), BW 10−8 M + PD 10−8 M, or PD 10−8 M alone for 2, 5, or 10 min followed by harvesting cultures for Western blots. A: Effects of treatments (5 min) on ratio of pMEK/MEK expressed as percentage of control. The asterisk indicates BW significantly different from control (P < 0.05 analysis of variance, Dunnet's post hoc test); the plus sign indicates BW significantly different from BW + PD (P < 0.05 analysis, Dunnet's post hoc test). Graph represents data from four independent experiments, where protein was pooled from two to five cultures/treatment. B: Representative Western blots showing time course for treatment effects on phosphorylated MEK and MEK.

Download figure to PowerPoint

Retinoic Acid and Chondrogenic Differentiation in Hindlimb Cultures

As shown in Figure 6, hindlimb MMCs exposed to 10−6 M all-trans RA exhibited a decrease in [35S]sulfate incorporation, indicating reduced maturation of cartilage matrix, whereas lower doses of RA (10−7 M, 10−8 M) had no significant effect. The negative effect of the highest dose of RA is consistent with previous reports showing dose-dependent inhibitory effects of RA on skeletal development (Jiang et al., 1995; Underhill and Weston, 1998; Weston et al., 2000; Underhill et al., 2001). In addition, because RA is known to affect many different cellular processes, including proliferation, differentiation, morphogenesis, and cell death (Sulik et al., 1988; Kochhar et al., 1993; Rodriguez-Leon et al., 1999; Hoffman et al., 2003), we evaluated cell viability after RA treatments by generating single cell suspensions and counting live cells using the trypan blue exclusion method. Results showed no difference in cell viability between control and the three different concentrations of RA tested. As well, total protein was measured in MMCs treated with RA (10−6 M); there was no significant effect compared with control. These results indicate that negative effects of RA on [35S]sulfate incorporation did not result from increased cell death or inhibition of cell proliferation (data not shown).

thumbnail image

Figure 6. Retinoic acid (RA) inhibits chondrogenic differentiation in hindlimb micromass cultures. Cultures were plated in serum-containing medium, then switched to serum-free medium at 1 day in vitro (DIV). Cultures were incubated with serum-free medium (control), insulin-like growth factor-1 (IGF-1, 20 ng/ml), RA 10−8 M, RA 10−7 M or RA 10−6 M from 2 to 4 DIV. Fifteen minutes after treatment on 4 DIV, cultures were incubated with [35S]sulfate for 6 hr. [35S]Sulfate incorporation was determined by liquid scintillation counting. Asterisks indicate P < 0.05 analysis of variance (Dunnet's post hoc test). The graph represents data from four independent experiments, where n = number of cultures analyzed per treatment group.

Download figure to PowerPoint

Effects of Retinoic Acid on 5-HT2B Receptor Expression

One possible explanation for the opposite effects of RA and 5-HT2B receptor activation on [35S]sulfate incorporation could be negative regulation of 5-HT2B receptor expression by RA, as suggested by the presence of potential RA response elements (RAREs) in the promoter (Choi et al., 1997; Bhasin et al., 1999; Lauder et al., 2001). To test this possibility, real-time PCR was used to determine whether RA could decrease 5-HT2B receptor transcript expression. Hindlimb MMCs were treated with different concentrations of all-trans RA (10−8 M, 10−7 M, or 10−6 M) or untreated. Six cultures from each treatment group were pooled for RNA isolation. In addition, effects of RA on a known RA-inducible receptor, RARβ (positive control), were assessed. All samples were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a gene constitutively expressed in all mammalian cells. Table 1 shows the average fold change for each treatment group compared with control. RA promoted expression of RARβ receptor transcripts in a dose-dependent manner, with higher doses of RA leading to higher average fold change compared with control. Expression of 5-HT2B receptor transcripts was also slightly increased by RA, but the magnitude of change was not nearly as high as for RARβ. These results indicate that, although RA may promote expression of the 5-HT2B receptor in hindlimb cultures to some extent, it clearly does not negatively regulate expression of this receptor.

Table 1. Effects of RA on 5-HT2B Receptor mRNA Expression in Hindlimb Micromass Culturesa
GeneDrug treatment
RA 10−8 MRA 10−7 MRA 10−6 M
  • a

    Average fold changes ± standard deviation in RARβ or 5-HT2B receptor mRNA are shown as determined by real-time PCR. RNA was isolated from MMCs either untreated or treated with different doses of all-trans RA (10−8 M, 10−7 M, and 10−6 M) and then assessed for relative change of mRNA levels in treated versus untreated control cultures. All levels were normalized to GAPDH prior to calculation of fold change. Note that there was a dose-dependent increase in RARβ mRNA transcripts (a retinoid-inducible receptor) in response to RA, whereas there was a minimal change in 5-HT2B receptor expression. RA, retinoic acid; 5-HT2B, serotonin 2B; RARβ, RA-inducible receptor; PCR, polymerase chain reaction; MMCs, hindlimb micromass cultures; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

5-HT2B receptor1.1 ± .013.95 ± .043.74 ± 2.86
RARβ receptor15.93 ± 1.9527.9 ± 2.039.09 ± 4.72

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Accumulating evidence suggests that 5-HT regulates developmental events during craniofacial development, including tooth germ morphogenesis, neural crest migration, and chondrogenic differentiation (Lauder, 1988, 1993; Shuey et al., 1992, 1993; Moiseiwitsch and Lauder, 1995, 1997; Choi et al., 1997, 1998; Moiseiwitsch et al., 1998; Moiseiwitsch, 2000). Although the end result is quite different in terms of morphology, it is clear that signals regulating patterning and morphogenesis of both the face and limbs are highly conserved (Schneider et al., 1999; LaMantia et al., 2000). In both cases, primordia begin as buds of mesenchyme surrounded by epithelium. As development proceeds, interactions between the two tissue compartments regulate morphogenesis of composite structures (Wedden et al., 1988; Richman and Tickle, 1992; Tickle and Eichele, 1994). Many malformations, such as those seen in Smith-Lemli-Opitz and velo-cardio-facial syndromes, occur simultaneously in the face and limbs (Smith et al., 1964; Friedman et al., 1977; Munke et al., 1990; Shprintzen, 2000; Maynard et al., 2002). As well, many cognitive disorders that involve the forebrain also tend to affect other sites of E-M interactions, including the face, heart, and limbs (Goodman, 1998; LaMantia, 1999).

Results of the present study provide evidence that activation of 5-HT2B receptors promotes chondrogenic differentiation in hindlimb MMCs. The 5-HT2B agonist BW increased [35S]sulphate incorporation, but did not significantly alter protein content (data not shown), indicating that this effect was due to enhanced chondrogenic differentiation, rather than to increased proliferation of chondrogenic progenitors. This finding is consistent with previous evidence that 5-HT promotes bone development, mediated by 5-HT2B receptors (Bliziotes et al., 2001; Westbroek et al., 2001).

Results also indicate that RA negatively regulates chondrogenic differentiation, without promoting cell death or decreasing cell proliferation. Taken together, these data raise the possibility that 5-HT2B receptors and RA may exert opposing influences on differentiation in the limb bud and at other sites of E-M interaction. This is supported by our recent study showing opposing regulation of cell proliferation in the frontonasal mass by 5-HT2B and RA (Bhasin et al., 2004).

Previous studies have reported that various 5-HT receptor agonists, including 8-OH-DPAT (5-HT1A) and SC53116 (5-HT4), increase levels of IGF-1, which itself can promote chondrogenesis (Lambert and Lauder, 1999; Lambert et al., 2001). However, in preliminary studies, agonists for these receptors did not stimulate [35S]sulfate incorporation in hindlimb MMCs (data not shown), leading us to test agonists for other 5-HT receptors, including those in the Gq-coupled 5-HT2 family.

The 5-HT2B receptor activates phospholipase C beta (PLC-β), causing hydrolysis of phosphatidylinositol bisphosphate (PIP2) to generate diacylglycerol (DAG) and inositol trisphosphate (IP3). DAG then activates protein kinase C (PKC). In tumor cells and cell lines expressing the 5-HT2B receptor, activation of this receptor causes PKC to activate the p42/p44 MAPK pathway by means of p21 ras (Loric et al., 1995; Launay et al., 1996, 1998; Fiorica-Howells et al., 2000). The 5-HT2B receptor has also been implicated in cross-talk with several other signal transduction pathways, including the platelet-derived growth factor receptor tyrosine kinase, cSrc and nitric oxide synthase (Manivet et al., 2000; Nebigil et al., 2000b). Based on these studies, a specific MEK-1/2 inhibitor, PD98059 (Zhang et al., 1997), was used to determine whether promotion of [35S]sulfate incorporation by activation of 5-HT2B receptors involved the p42/p44 MAPK cascade. PD effectively blocked the stimulatory effects of 5-HT, as well as those of the 5-HT2B receptor agonist BW (Fig. 4A) and prevented phosphorylation of MEK-1/2 by BW (Fig. 4B). These results provide evidence that 5-HT2B receptors use the p42/p44 MAPK pathway to promote chondrogenic differentiation. This result is not consistent with a previous report that inhibition of MEK-1/2 by PD promotes chondrogenesis in chick limb bud micromass cultures (Oh et al., 2000). However, differences in culture conditions could help explain this discrepancy, because chick MMCs were maintained in 10% fetal calf serum for 4 DIV in the presence of PD, whereas our cultures were grown in serum-free medium and exposed to PD to block the effects of BW.

The effect of different doses of RA on [35S]sulfate incorporation was also determined (Fig. 5). The highest dose of RA (10−6 M) inhibited sulfate incorporation, consistent with previous studies showing inhibitory effects of high doses of RA on skeletal development (Jiang et al., 1995; Underhill and Weston, 1998; Underhill et al., 2001; Hoffman et al., 2003). To ensure that the inhibitory effect of RA was not due to increased cell death, cell viability was evaluated using the trypan blue exclusion method. No significant difference in numbers of viable cells was found between treatment groups (data not shown), indicating that the negative effects of RA on chondrogenic differentiation were not due to increased cell death.

Previous studies have reported similar sites of malformations after exposure of mouse embryos to excess RA or the pan–5-HT2 receptor antagonist, ritanserin (Choi et al., 1997, 1998; Lauder et al., 2000; Nebigil et al., 2000a). Malformations caused by ritanserin also resemble those observed in AP-2 knockout mice (Schorle et al., 1996; Zhang et al., 1996), consistent with evidence that the 5-HT2B receptor upstream regulatory sequence contains both potential RAREs and an AP-2 site (Choi et al., 1997; Bhasin et al., 1999; Lauder et al., 2001).

Similarities between the phenotypes of ritanserin- and RA-treated embryos, as well as AP-2 knockout mice, have led to speculation about a possible reciprocal relationship between RA and 5-HT2B receptor signaling during embryogenesis (Choi et al., 1997; Lauder et al., 2000). To determine whether RA might negatively regulate expression of the 5-HT2B receptor, real-time PCR was used to measure changes in levels of 5-HT2B receptor transcripts in hindlimb cultures after exposure to different doses of RA. RARβ was used as a positive control. Results indicate that, while expression of the 5-HT2B receptor was slightly stimulated by retinoid treatment, these effects were minimal compared with the robust stimulatory effects on RARβ, a known retinoid-induced RAR (Table 1). Clearly, there was no negative regulation of 5-HT2B by RA, as previously considered. On the other hand, the AP-2 site in the 5-HT2B promoter is still a candidate for possible indirect retinoid regulation of this receptor, because AP-2 has been shown to be induced by RA (Luscher et al., 1989). AP-2 can act as a positive or negative transcriptional regulator (Hilger-Eversheim et al., 2000). Therefore, it is not clear whether RA would promote or repress expression of 5-HT2B if mediated by AP-2. The RARβ receptor is one of the retinoid receptors thought to be involved in modulating the effects of RA on chondrogenesis in the limb, but its role appears to be primarily a negative one, because it is localized to regions of nonchondrogenesis (Kochhar et al., 1993; Jiang et al., 1995; Hoffman et al., 2003). That RA increased expression of RARβ nearly 40-fold (Table 1) could help explain the significant inhibitory effects of 10−6 M RA on chondrogenic differentiation (Fig. 5). Whether RA also promotes expression of other RARs in these cultures is not known. However, if expression of RARα or RARγ were also promoted, this could contribute significantly to the inhibitory effects of RA, because repression of these RARs is a known requirement for chondrogenesis in the hindlimb, associated with activation of p38 MAPK and PKA and with expression of the transcription factor Sox9 (Weston et al., 2000, 2002). Therefore, it would be predicted that exposure to RA at the concentration that reduced [35S]sulfate incorporation could lead to a decrease in p38 MAPK phosphorylation. Future studies will determine whether this is the case and if inhibitory effects of RA on chondrogenesis are associated with decreased activation of PKA and reduced expression of Sox 9.

Taken together, results of the present study indicate that inhibitory effects of RA on chondrogenic differentiation in hindlimb cultures are not mediated by negative transcriptional regulation of the 5-HT2B receptor but are accompanied by increased expression of RAR β, which could result in decreased activation of p38 MAPK. This possibility must be verified by further signal transduction studies. On the contrary, stimulatory effects of 5-HT2B receptor activation on chondrogenic differentiation appear to be mediated by activation of the p42/44 MAPK pathway. Therefore, RA and 5-HT may exert opposing effects on chondrogenesis in the developing hindlimb by using different MAPK pathways. Whether serotonergic and retinoid signaling exert opposing influences on morphogenesis at other sites of E-M interaction remains to be determined.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Mice

Wild-type (ICR, Harlan) mouse embryos were obtained from timed pregnancies generated in a breeding colony maintained by the Department of Laboratory Animal Medicine at the University of North Carolina at Chapel Hill (plug day = E0.5). For preparation of cultures, pregnant mice were killed by rapid cervical dislocation and embryos were harvested immediately.

Hindlimb Micromass Cultures

High-density micromass mesenchymal cell cultures (MMCs), were prepared from hindlimb buds of embryonic day (E) 11.5 mouse embryos (crown–rump length, 7.0–7.5 mm), according to a modification of methods described previously (Hassell and Horigan, 1982). Hindlimb buds were removed in phosphate buffered saline (PBS). Mesenchymal cells were separated from the epithelium by dissociation in 0.1% trypsin and 0.1% EDTA in Dulbecco's minimal essential medium (DMEM, Gibco), followed by trituration in serum-containing medium (DMEM/Ham's F-12 [Gibco] + 10% fetal calf serum [Gibco] + penicillin/streptomycin) and filtration through a nylon mesh. Viable cells were counted by using a hemocytometer (trypan blue exclusion method) and diluted to a final concentration of 15 × 106 cells/ml. A 10-μl aliquot of this suspension was spotted onto plastic coverslips (Thermanox, Fisher) in a 24-well plate and allowed to settle for 15 min. A second 10-μl aliquot was carefully added on top of the first. After 45 min, wells were flooded with 0.5 ml of serum-containing medium. After 24 hr (1 day in vitro; 1 DIV), the serum-containing medium was removed and replaced with serum-free medium (DMEM/Ham's F-12 + 0.1% bovine serum albumin [Sigma] + penicillin/streptomycin). Drug treatments were administered from 2 to 4 DIV. Final concentrations of drugs tested were nialamide (10−5 M, Sigma) + L-cysteine (10−5 M, Sigma, NC), 5-HT (10−6 M or 10−8 M, Sigma) + NC, IGF-1 (20 ng/ml, DSL, Inc.), 5-HT2B receptor agonist BW723C86 (BW; 10−8 M, Sigma), 5-HT2B/2C receptor antagonist SB206553 (SB; 10−8 M, RBI), MEK inhibitor PD98059 (PD; 10−8 M, RBI), or all-trans retinoic acid (RA; 10−6 M, 10−7 M or 10−8 M, Sigma). Cultures were incubated at 37°C with 5% CO2.

Chondrogenic Differentiation Assays ([35S]Sulfate Incorporation)

Incorporation of [35S]sulfate into the extracellular matrix can be used as a measure of chondrogenic differentiation (Bagi and Burger, 1989). Fifteen minutes after drug treatment on the 4th DIV, hindlimb MMCs were treated with 5 μCi [35S]sulfate for 6 hr. Cultures were subsequently washed with 0.5 ml of Hank's balanced salt solution (Gibco) and precipitated with 0.5 ml of 5% trichloroacetic acid (TCA) for 15 min at room temperature. Cultures were rinsed with 5% TCA and then dissolved in 0.5 ml of 1% sodium dodecyl sulfate (SDS) at 37°C for 15 min. The SDS solution was added to 5 ml of ScintoSafe Econo 2 scintillation fluid (Fisher). [35S]Sulfate incorporation was determined by liquid scintillation counting.

Western Analysis of MEK Phosphorylation

Mouse hindlimb micromass cell cultures were lysed by incubating in 100 μl of buffer containing 0.0625 M Tris-HCl, pH 6.8, 2% w/v SDS, and 10% glycerol for 10 min at room temperature followed by trituration. Protein concentrations were determined with the Bio-Rad DC assay kit (Bio-Rad, Hercules, CA) using bovine serum albumin as a standard. Lysates were frozen at −80°C until use. For Western blot analysis, 3–5 μg of micromass cell lysate per lane were separated by SDS–polyacrylamide gel electrophoresis on 10% polyacrylamide gels (Laemmli, 1970) and transferred to 0.45-μm nitrocellulose as described (Ausubel et al., 1992). The nitrocellulose membranes were incubated with antibodies recognizing MEK and phosphorylated MEK (pMEK; MEK1/2 Phospho Plus kit; Cell Signaling Technology, Beverly, MA) according to the manufacturer's protocols. Immunoreactivity was detected by using a Lumi-Glo chemiluminescence substrate (Cell Signaling Technology) and exposure to Kodak X-AR film. Quantitative densitometry was performed using the computer program NIH Image.

Characterization of Hindlimb Micromass Cultures by Immunocytochemistry

Expression of 5-HT2B receptors in hindlimb MMCs was established by using a specific 5-HT2B receptor monoclonal antibody (PharminGen) and standard immunocytochemistry protocols, as previously described (Lambert and Lauder, 1999; Lambert et al., 2001).

[125I] Protein A Immunobinding Assays

Semiquantitative analysis of relative 5-HT2B receptor protein levels in treated and control MMCs was performed by using [125I]-labeled protein A immunobinding assays, as previously described (Lambert et al., 2001).

Real-Time PCR

Hindlimb MMCs were grown as described above. Six wells per treatment group were pooled for RNA isolation. RNA was isolated from the cultures with TRIzol reagent (Invitrogen) using standard methods. Isolated RNA underwent treatment with DNase I (Ambion) to remove any DNA from the sample, followed by reverse transcription using AmpliTaq Gold DNA Polymerase (Applied Biosystems, ABI). Real-time PCR using Sybr Green was performed according to the protocols established by ABI (P/N 4304965). All samples were run in triplicate and standardized to GAPDH. All reagents were obtained from ABI. Samples were run and analyzed by using an ABI Prism 7700 Sequence Detection System. Primers were designed using Primer Express (ABI) software: 5-HT2B receptor forward primer, 5′ GCA GAT TTG CTG GTT GGA TTG 3′; 5-HT2B receptor reverse primer, 5′ GGG CCA TAT AGC CTC AAA CAT G 3′; RARβ forward primer, 5′ CGG GCA GAT CCT GGA TTT C 3′; RARβ reverse primer, 5′ GAG AGC CTT TTC CTG CAG CAT 3′; GADPH forward primer, 5′ TGT GTC CGT CGT GGA TCT GA 3′; GADPH reverse primer, 5′ CCT GCT TCA CCA CCT TCT TGA 3′.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Krystle Strand for technical assistance, and Dr. Thomas Sadler for critical reading of the manuscript and helpful advice. Drs. Vicki Bautch and Terry Magnuson also provided useful discussions during the course of this study. J.M.L. was funded by a NIDCR grant.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES