ToxSci Advance Access originally published online on December 1, 2005
Toxicological Sciences 2006 90(1):208-220; doi:10.1093/toxsci/kfj060
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Retinoid Receptor Antagonists Alter the Pattern of Apoptosis in Organogenesis Stage Mouse Limbs
Department of Pharmacology and Therapeutics, McGill University, Montréal, Québec, Canada H3G 1Y6
1 To whom correspondence should be addressed at Department of Pharmacology and Therapeutics, McGill University, 3655 Promenade Sir-William-Osler, Montréal, Québec, Canada H3G 1Y6. Fax: (514) 398–7120. E-mail: barbara.hales{at}mcgill.ca.
Received September 2, 2005; accepted November 21, 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Exposure of murine limbs in vitro to vitamin A (retinol) induces limb reduction defects and apoptosis. To assess the relative roles of the retinoic acid receptors (RARs) and the retinoid X receptors (RXRs), embryonic-day-12 murine limbs were cultured with selective RAR or RXR antagonists in the presence or absence of teratogenic concentrations of retinol. Both antagonists alone impaired limb development; in the presence of teratogenic concentrations of retinol, both attenuated limb malformations. Abnormal limb morphology, whether caused by excessive or attenuated retinoid signaling by retinol or either antagonist, respectively, was correlated with increased apoptosis after 24 h of drug exposure. We conclude that, in the developing limb, antagonists selective for either member of the RAR/RXR heterodimer attenuate retinoid signaling and block the teratogenic signaling of excess retinol. Improvements in limb morphology in the presence of either the RAR or the RXR antagonist coincided with restoration of the extent and localization of limb bud apoptosis to control patterns.
Key Words: retinoic acid; limb development; limb malformations; teratogen.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Vitamin A or retinol and its derivatives, the retinoids, are essential for proper embryonic development. Active retinoids are important signaling molecules in the regulation of cell differentiation, proliferation, and morphogenesis; the developing limb is sensitive to both excess and paucity of these compounds.
Genetic and biochemical manipulations indicate that retinoids are crucial for vertebrate limb development at two stages, in early organogenesis for forelimb bud initiation, and later in mid-organogenesis, for proper limb outgrowth (Mic et al., 2004
; Niederreither et al., 2002) In the mouse and other model species, administration of excess bioactive retinoids during mid-organogenesis leads to reductive limb defects resembling those seen in retinoid deficiency: truncations and deletions of the long bones (ulna, radius, and humerus in the forelimb) and digital deletions and fusions (Kochhar, 1973, 1985). During this susceptible period, key morphogenetic processes occurring in the developing limb bud include limb outgrowth, differentiation of the limb skeleton cartilaginous anlagen, and the selective deletion of specific cells by programmed cell death or apoptosis. Retinoids influence all three of these processes (Dupe et al., 1999; Francis-West and Tickle, 1996; Kochhar, 1985; Zakeri and Ahuja, 1994), but due to the importance of apoptosis in shaping limbs during development, we have examined the effects of retinoids on this event in the organogenesis-stage murine limb.
Limb bud apoptosis in the mouse begins about embryonic day (ED) 10.5, peaking at ED14.5, with the majority occurring in the interdigital (INZ), anterior and posterior marginal zones (AMZ and PMZ) and the apical ectodermal ridge (AER) (Zakeri and Ahuja, 1997). Precise spatiotemporal control of both the extent and the localization of apoptosis during development is essential for normal morphogenesis; exposure of the embryo to excessive retinol during mid-organogenesis, coincident with the induction of limb malformations, also causes aberrant apoptosis.
Retinoids bind to two subclasses of nuclear receptors, the retinoic acid receptors (RARs) and the retinoid X receptors (RXRs), each consisting of three members: 
, ß, and 
(Leid et al., 1992). Several isoforms of each exist, with distinct spatial and temporal expression patterns. The diverse effects of retinoids during normal development are mediated by the RAR/RXR heterodimer made up of various combinations of these isoforms (Chambon, 1994; Kastner et al., 1997). All-trans and 9-cis retinoic acid bind and activate transcription from the RARs, while just 9-cis retinoic acid binds RXRs with high affinity. On ligand binding, the heterodimers transactivate through DNA response element recognition.
In vivo studies following administration of selective retinoid receptor agonists show that RAR-selective agonists are the most potent limb teratogens, followed by ß- and -specific agonists (Arafa et al., 2000). Conversely, RXR-selective agonists are not fetotoxic or teratogenic when administered in vivo (Kochhar et al., 1996), but administration of these compounds can potentiate the teratogenic effects of an RAR-selective agonist (Elmazar et al., 2001). In sum, the presence of both the RAR and the RXR subtypes is important in the limb, and their relationships with one another in the heterodimer are complex. Yet little is known about how these receptors influence apoptosis in normal development or during the evolution of abnormal morphology.
We had two goals in the present work: (1) to investigate the relationship between retinoid-induced limb malformations and apoptosis and (2) to assess the importance of the RARs and the RXRs in this process. To do this, we targeted retinoid signaling during mid-organogenesis with retinoid receptor antagonists, first in the absence of exogenous ligand (endogenous situation), and second in the presence of teratogenic concentrations of retinol. Enzymes metabolizing retinol to both all-trans and 9-cis retinoic acid are present in the limb (Horton and Maden, 1995), thus generating ligands for both the RAR and RXR receptor subfamilies. ED12 limb buds were treated in an in vitro culture system using selective pan-RAR (BMS 453) and pan-RXR (HX603) antagonists on their own, as well as in combination with exogenous all-trans retinol acetate (Vitamin A).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Limb bud cultures and drug treatments.
Limb buds were cultured as previously described (Ali-Khan and Hales, 2003
; Huang and Hales, 2002). Briefly, ED12 embryos were dissected from timed-pregnant CD1 mice (Charles River Canada Inc., St. Constant, QC). At this stage embryos were 5–6 mm in length and exhibited 42–48 somite pairs. Forelimbs were excised, pooled, and cultured in roller bottles, as previously described (Kwasigroch et al., 1984). All animal studies were done in full compliance with the guidelines of The Canadian Council on Animal Care.
All-trans retinol acetate (Vitamin A) (Sigma, St. Louis, MO) dissolved in 100% ethanol, was added to the respective cultures with a final percentage of ethanol of no more than 0.5%. The concentrations used in this work were selected based on previous studies showing that the dose-response relationship for retinol is relatively shallow (Ali-Khan and Hales, 2003). An intermediate concentration produced relatively mild effects; therefore, for investigation of the full range of retinoid agonist effects, 1.25 µM (1 IU) and 62.5 µM (50 IU) were chosen. Experimental drugs were added to culture bottles just prior to the start of the culture period, and combinations of drugs were added immediately after one another.
BMS 189453 (BMS 453), a competitive pan-RAR antagonist, was a gift from Bristol-Myers Squibb (Wallingford, CT). Its selectivity at the , ß, and RAR-subtypes has been assessed by transactivation and direct binding assay (Chen et al., 1995; Yang et al., 1999). HX603, a pan-RXR antagonist, was a gift from Dr. H. Kagechika, University of Tokyo. It is a highly specific competitive antagonist at the , ß, and RXRs; its selectivity was assessed by transactivation assay and the HL-60 differentiation assay (Ebisawa et al., 1999). These compounds were dissolved in DMSO and added to the designated cultures with a final percentage of DMSO of 0.05% or less. Stock solutions of BMS 453 and HX603 were stored, protected from light, at –20°C between usages. The concentrations used were selected based on a pilot study in which limbs were cultured for 6 days, titrating BMS 453 and HX603 against retinol to attain a range that produced the desired result. A separate control study examining the effects of DMSO and ethanol on limb development showed that ethanol at the concentrations used in these studies had no significant effect on any of the parameters assessed. DMSO produced small but significant reductions in limb score and limb area of 3% (p = 0.04) and 2% (p = 0.02), respectively; there was no statistically significant interaction between the two vehicles. The addition of BMS 453 concentrations higher than 5 µM and HX603 concentrations greater than 20 µM to limb cultures resulted in a total arrest of limb development, as evidenced by a complete lack of both chondrogenesis and limb outgrowth (data not shown).
ßRARE-lacZ staining.
ßRARE-lacZ homozygote males were mated with CD-1 females to generate heterozygous mutant embryos in house. ED12 embryonic limbs were excised and cultured as described above in the presence or absence of retinol and/or BMS 453 for 24 h. Limbs were processed and stained with X-gal overnight as previously described (Rossant et al., 1991). Limbs were viewed under a dissecting microscope (Wild Heerbrugg 99067, Heerbrugg, Switzerland), and digital images were acquired (JVC GCQX3HD Digital Camera, Japan).
Gene expression analysis.
Limbs were removed after 3 h of culture, rinsed with PBS, and stored at –20°C in RNAlater RNA Stabilization Reagent (Qiagen, Mississauga, ON, Canada). Total RNA from limbs was extracted using the RNeasy Micro Kit (Qiagen, Mississauga, ON, Canada) and quantified by spectrophotometric UV analysis. RNA was then diluted to a working concentration of 10 ng/µl, and Quantitect One-Step SYBR Green RT-PCR (Qiagen, Mississauga, ON, Canada) was completed using the Roche LightCycler according to the manufacturer's instructions (Roche Diagnostics, Laval, QC, Canada). PCR thermal cycling parameters were: 95°C for 15 min (1 cycle), then 94°C for 15 sec, 55°C for 30 sec, and 72°C for 20 sec (for 45 cycles). Each treatment consisted of RNA from five separate culture experiments; each sample was measured in duplicate. Embryonic hindlimb tissues were used to make 1, 25, 50, and 100 ng/µl RNA stocks for standard curves for quantification. Primers were generated with Primer3 software (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) and synthesized at the Sheldon Biotechnology Centre (McGill University, Montreal, QC, Canada); expression was normalized against 18SRNA, and melting curve analyses were done following each PCR to determine the output and detection quality (i.e., formation of primer-dimers). Murine 18 S RNA; accession # X00686; Left primer; aaacggctaccacatccagg. Right primer; cctccaatggatcctcgtta. Murine Cyp26a1; accession # NM_007811; Left primer; ttcgggttgctctgaagact. Right primer; tcctccaaatggaatgaagc.
Limb morphology.
After 3 days of culture the media were replaced without the addition of experimental drugs, and the cultures were continued for 3 more days. After the 6-day culture period, limbs were fixed overnight in Bouin's fixation, stained with 0.1% toluidine blue (Fisher Scientific, Montréal, QC) in 70% ethanol overnight, dehydrated, and cleared with cedar wood oil. Limbs were viewed under a dissecting microscope (Wild Heerbrugg 99067, Heerbrugg, Switzerland), and digital images were acquired (JVC GCQX3HD Digital Camera, Japan). Using computer-assisted image analysis (MCID-7, Image Research Inc., St. Catherine's, ON, Canada), limb and relative cartilage (cartilage area/total limb area) areas were measured as endpoints for overall growth and chondrogenesis, respectively. The extent and quality of differentiation was assessed using the Limb Morphogenetic Differentiation Scoring System developed by Neubert and Barrach in 1977. Five to seven independent experiments were completed per treatment group, with a total of 15–25 limbs assessed per treatment group (n = 5–7).
TUNEL staining.
After a 24-h culture period, limbs were fixed with Bouin's fixative at 4°C for 3 h, dehydrated, and embedded in paraffin. Sections of 5 µm were cut and mounted on slides. Limb cells undergoing apoptosis were localized using the ApopTag Peroxidase in situ Apoptosis Detection Kit (Intergen Company, Purchase, NY) as previously described (Ali-Khan and Hales, 2003). The only changes to this protocol were that proteinase K (20 µg/ml, Sigma, St. Louis, MO) and hydrogen peroxide treatments were reduced to 7 min each at room temperature. Positively labeled apoptotic cells were visualized with DAB (3,3-diaminobenzidine) solution (DAB Substrate Kit for peroxidase, Vector Laboratories Inc., Burlington, ON). Limbs were counterstained with methylene blue, observed on a Leitz Laborlux D microscope, and images were acquired at 40x and 400x magnification (CoolSNAP RS photometrics digital camera and software). Negative controls for TUNEL staining were done concurrent to experimental samples. Examples of these from our lab have been published previously (Huang and Hales, 2002). Using computer-assisted image analysis (MCID-7, Image Research Inc., St. Catherine's, ON, Canada), the extent of interdigital apoptosis was estimated by quantifying the TUNEL-positive area in two 8.5-µm2 target zones, between the second and third, and third and fourth digits in each limb that was assessed. The two 8.5-µm2 target rectangles were positioned on each section so that the top margin was aligned with the apical edge of the limb squarely within the interdigital region, and TUNEL-positive cells were picked out by setting color and intensity detection thresholds for the brown DAB-stained cells. Only limb sections cut at a similar depth and angle were chosen for assessment. Five to eight independent experiments were completed per treatment group; the mean TUNEL-positive area per limb was recorded for a total of 15–20 limbs per treatment group (n = 5–8).
Statistical analysis.
Six-day cultured limb morphological data and 24-h cultured limb TUNEL data were analyzed using two-way ANOVA (group 1 = retinol, group 2 = BMS 453 or HX603) and the post hoc Holm-Sidak Multiple Comparison Test (Sigmastat Statistical Software, SPSS Inc., Chicago, Il). Data are expressed as means ± SEM. The minimum level of significance was p < 0.05 for all tests.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Cyp26a1 Expression
First, we confirmed that retinol was functioning in our system as a retinoid agonist and BMS 453 and HX603 were functioning as retinoid antagonists. ED12 CD-1 limb buds were cultured for 3 h in the presence or absence of retinol with or without the addition of BMS 453 or HX603. Expression levels of the retinoid-responsive gene Cyp26a1 were assessed by real-time PCR as a measure of retinoid activity (White et al., 1997
). Retinol treatment induced dose-dependent increases in expression at 1.25 and 62.5 µM (Figs. 1A and 1B). In the presence of either concentration of retinol, BMS 453 at 0.5 or 2.5 µM, or HX603 at 5 or 12 µM drastically decreased Cyp26a1 expression to baseline. This decrease was significant for both antagonists in the presence of 62.5 µM retinol. In the absence of retinol, Cyp26a1 levels were very low; thus, neither antagonist decreased expression appreciably below baseline.
|
RARE-lacZ Activity
Recent studies have indicated that BMS 453 may have mixed agonist/antagonist activity at the ßRARs (Matt et al., 2003
). To clarify the activity of BMS 453 in our system, we cultured ED12 ßRARE-lacZ limbs (Rossant et al., 1991) with the same paradigm used throughout this study. ED 12 limbs were cultured for 24 h in the presence or absence of retinol alone, with or without BMS 453 to assess the effect of the agonist and antagonist on retinoid signaling (Fig. 1C). Blue lacZ staining indicates the presence of bioactive retinoids capable of transactivating through retinoid heterodimer and RARE-binding. Control limbs showed positive staining in the interdigital regions and the proximal mesenchyme (Fig. 1C:a). The addition of 1.25 µM exogenous retinol greatly intensified lacZ staining in all these areas, most strikingly in the zone of polarizing activity (ZPA) (Fig. 1C:d); 0.5 µM BMS 453 reduced both endogenous retinoid activity and that induced by exogenous retinol to almost undetectable levels (Figs. 1C:b and 1C:e). LacZ staining at 2.5 µM BMS 453 was still inhibited below that seen in the corresponding control and retinol-treated limbs, but there was a slight increase in staining compared to limbs exposed to 0.5 µM BMS 453. This finding may reflect ßRAR agonist activity by BMS 453 at 2.5 µM. Overall, however, these Cyp26a1 and ßRARE-lacZ reporter results indicate that retinol and BMS453/HX603 are acting as bone fide retinoid agonist and antagonists, respectively, in the mid-organogenesis limb bud.
Both a Paucity and an Excess of Retinoid Signaling through the RARs or RXRs Result in Limb Malformations
Effects of exogenous retinol.
To assess effects on overall limb outgrowth and the patterning and differentiation of the limb cartilage, anlagen ED12 CD-1 limb buds were cultured for 6 days in the presence or absence of retinol. Chondrogenesis in the cultured limb mesenchyme was assessed by staining with toluidine blue for glycosaminoglycan expression, a marker of chondrocyte differentiation. The control limb (Fig. 2:a) showed proper outgrowth and development. The radius and ulna showed normal lengthening, and chondrogenesis had progressed well in both the handplate and the long bones. Figures 2:a and 3:a illustrate variation typical in controls; compare development of the phalanges, metacarpals, and carpalia. Culturing with retinol dramatically affected limb morphology (Figs. 2d, 2g, 3d and 3g). At 1.25 µM retinol, limb outgrowth was reduced and chondrogenesis altered. The radius and ulna were shortened, while the metacarpalia were lengthened and spatulate; some cartilage elements had fused or showed ill-defined margins. Exposure to 62.5 µM retinol caused almost complete inhibition of chondrogenesis and limb outgrowth, confirming that excessive signaling through the retinoid pathway with a retinoid agonist induces dose-dependent reductive limb defects.
|
|
Effects of antagonists.
To determine how attenuating signaling through these receptor families would affect limb development, limbs were cultured with BMS 453, a pan-RAR antagonist, or HX603, a pan-RXR antagonist (Figs. 2b, 2c, 3b, and 3c). Culture with either antagonist induced dose-dependent limb malformations. At 0.5 µM BMS 453, overall limb size was reduced, including shortened long bones, fused carpalia, and decreased chondrogenesis. At 2.5 µM BMS 453, differentiation of the limb cartilagenous anlagen was further retarded, and in some cases, skeletal elements were missing altogether. In contrast, although both 5 and 12 µM HX603 reduced overall limb size and differentiation of the handplate, the long bones remained relatively well lengthened, thick, and heavily chondrified. Thus, abrogating retinoid signaling through either the RARs or the RXRs during mid-organogenesis caused characteristic limb malformations; however, the resultant abnormal morphology differed between BMS 453 and HX603, as well as from the phenotype induced by exposure to excess retinol.
Effects of retinol in the presence of antagonists.
To investigate the roles of the RARs and the RXRs in mediating retinoid-induced teratogenesis, BMS 453 or HX603 was added to limb bud cultures with teratogenic concentrations of retinol (Figs. 2e, 2f, 2h, 2i, 3e, 3f, 3h, and 3i). Limbs cultured with 0.5 µM BMS 453, in the presence of 1.25 µM retinol (Fig. 2:e), showed marked improvements in long bone outgrowth, and also in the definition and development of the carpalia and metacarpalia. Limbs in this treatment group showed a phenotype similar to that of the control limbs (compare limbs in Figs. 2:e and 2:a). Exposure to 0.5 µM BMS 453 with 62.5 µM retinol also improved long bone, carpalia, and handplate differentiation, and some long bone outgrowth was restored. Exposure to 2.5 µM BMS 453 did not improve morphology further; it reduced the success of limb development in the presence of both 1.25 and 62.5 µM retinol, although not to the extent seen in limbs cultured with retinol alone (Figs. 2:f and 2:i). This result may reflect the rebound in retinoid signaling at 2.5 µM BMS 453 seen in ßRARE-lacZ reporter limbs; residual agonism at ßRARs could account for the lack of dose-response improvement in morphology seen with BMS 453 at the higher concentration.
HX603 improved limb malformations caused by 1.25 µM retinol in a dose-dependent manner (Fig. 3: compare d, e, f). Limbs cultured with 12 µM HX603 showed a phenotype close to control (compare Figs. 3:a, 3:d, and 3:f); long bones were better elongated and metacarpalia had a more normal size and shape. Limbs cultured with 5 µM HX603 (Fig. 3:e) showed intermediate morphology. Similarly, the addition of HX603 to limbs cultured with 62.5 µM retinol improved limb development compared to those cultured with retinol alone (Figs. 3:g–3:i). Exposure to 5 µM HX603 improved outgrowth and allowed some differentiation of the cartilage elements. Limbs cultured with 12 µM HX603 manifested poorly chondrified elements and little to no outgrowth. Hence, attenuating signaling through both the RARs with BMS 453 and the RXRs with HX603 inhibited the teratogenic effects of excessive retinol and considerably improved limb morphology.
Quantification.
Further analysis of the interaction between retinol and BMS 453, and retinol and HX603 is presented in Figures 4 and 5, respectively. First, limb development was scored using the Limb Morphogenetic Differentiation Scoring System developed by Neubert and Barrach (1977). This takes into account the size, shape, and number of cartilage elements and extent of their differentiation. Culture with either retinol, BMS 453, or HX603 alone significantly decreased limb score compared to controls in a dose-dependent manner (Figs. 4A and 5A). The addition of 0.5 µM BMS 453 or either concentration of HX603 in the presence of 1.25 or 62.5 µM retinol significantly improved limb score compared to retinol alone. Furthermore, compared with 0.5 µM BMS 453, culture with 2.5 µM BMS 453 with either concentration of retinol lowered scores (Fig. 4A). At 1.25 µM retinol, 0.5 µM BMS 453 and 12 µM HX603 restored limb scores to close to control values.
|
|
As a measure of chondrogenesis, we quantified the percentage cartilage of each limb (Figs. 4B and 5B). Exposure to 1.25 µM retinol alone did not significantly affect percent cartilage; exposure to 62.5 µM retinol caused a significant decrease. Either BMS 453 or HX603 alone significantly decreased percent cartilage compared to control and, in combination with 1.25 µM retinol, increased percent cartilage to close to control values, at 2.5 µM BMS 453 and 12 µM HX603, respectively.
To evaluate the effect of the agonist and either antagonist on overall limb growth, and on outgrowth, limb area was measured (Figs. 4C and 5C). Retinol alone at 62.5 µM significantly reduced limb area. HX603 alone at both concentrations and BMS 453 alone at 2.5 µM also reduced limb area significantly. In the presence of 62.5 µM retinol, BMS 453 at 0.5 µM and HX603 at 12 µM increased limb area compared to limbs cultured with retinol alone.
Thus, a general decrease in limb score, percent cartilage, and limb size was observed when retinoid signaling was either augmented or attenuated with the addition of the agonist, retinol, or antagonists, BMS 453 and HX603. BMS 453, especially at 0.5 µM, and HX603 at 12 µM, were most successful in restoring the limb score, percentage cartilage, and limb size of retinol-exposed limbs to close to control values.
Altering the Balance of Retinoids in the Mid-Organogenesis-Stage Limb Bud Increases Apoptosis in Susceptible Regions of the Limb
A key morphogenetic event in limb development during mid-organogenesis is tissue remodeling by programmed cell death. To probe the origins of the limb malformations described above, we assessed apoptosis in the limbs 24 h after exposure to retinol, BMS 453, or HX603. This time point was selected based on the well-developed and specific apoptotic response triggered in retinol-exposed limbs at this time (Ali-Khan and Hales, 2003). Limb buds that were cultured for 24 h with retinol, BMS 453, or HX603 in the same paradigms as the previous experiment were fixed, sectioned, and stained by the TUNEL method (Figs. 6 and 9).
|
|
Effects of retinol.
Culturing limb buds with retinol alone led to a dose-dependent increase in apoptosis that preceded and was correlated with the severity of malformations observed in 6-day cultured limbs (Figs. 6a, 6d, 6g, 9a, 9d, and 9g) (Ali-Khan and Hales, 2003
). This retinol-induced apoptosis occurred preferentially in zones that normally undergo programmed cell death in the mid-organogenesis-stage limb, albeit to an exaggerated and inappropriate extent. Control limbs showed few apoptotic cells, mainly localized to the apical ectodermal ridge (AER), the anterior and posterior marginal zones (AMZ and PMZ), and the interdigital zones (INZ) (interdigital regions of the same limbs are shown in Figs. 7a, 7d, 7g, 10a, 10d, and 10g). Culture with 1.25 µM retinol showed a similar pattern of apoptosis to control, but with increased numbers of dying cells (Figs. 6d, 7d, 9d, and 10d). There were fewer TUNEL-positive cells in the AER, but more in the INZ, AMZ, and PMZ, as well as around the developing long bones (Figs. 6d and 9d). Limbs cultured in 62.5 µM retinol showed greatly upregulated apoptosis in the AMZ and PMZ, the AER, between and around the developing radius and ulna, and in expanded INZs (Figs. 6, 7, 9, and 10:g). Some limbs in this group showed what appeared to be cell fragments in the INZ and along the AER that did not stain TUNEL-positive, suggesting that, in response to high concentrations of retinol, some cells may undergo alternative cell-death pathways rather than apoptosis.
|
|
Effects of antagonists.
Both BMS 453 and HX603 showed dose-dependent upregulation of apoptosis; interestingly however, they showed differential effects in different limb regions (Figs. 6a-c, 9a-c; high magnification images of the same limbs in Figs. 7a-c, 10a-c). BMS 453 at 0.5 µM increased TUNEL-staining in the AER and to a small extent in the INZs, and in the AMZ and PMZ (Figs. 6b and 7b). At 2.5 µM, it greatly extended regions of TUNEL-positive cells throughout the zeugopod mesoderm, the AMZ and PMZ, the AER, and in the INZs. Exposure to 5 µM HX603 also produced apoptosis in the INZ, the AMZ, and PMZ, but interestingly, in many limbs a band of TUNEL-positive cells was apparent below and parallel to the AER (Figs. 9b and 10b). Such a band was rarely seen in limbs in any of the retinol or BMS 453–treated groups. 12 µM HX603 induced a similar pattern, but with more TUNEL-positive cells; extensive apoptosis above the AER-parallel band of apoptosis led to disintegration of the tissue. TUNEL-positive cells also increased in the zeugopod mesenchyme soft tissue regions (Figs. 9c and 10c). Thus, decreasing retinoid signaling through both the RARs and the RXRs increased apoptosis in the developing limbs in a characteristic, specific, and dose-dependent fashion.
Effects of retinol in the presence of antagonists.
Next, we asked if the addition of the antagonists modified retinol-induced apoptosis (Figs. 6e, 6f, 6h, 6i, 7e, 7f, 7h, 7i, 9e, 9f, 9h, 9i, 10e, 10f, 10h, and 10i). In the presence of 1.25 µM retinol, 0.5 µM BMS 453 and 12 µM HX603 were successful in reducing aberrant retinol-induced apoptosis to close to control levels and patterns (Figs. 6e, 6f, 9e, and 9f, compare with Figs. 6a, 6d, 9a, 9d, 7a, 7d, 10a, and 10d). In particular, for both antagonists there were far fewer TUNEL-positive cells apparent in the INZs, but there was also reduced staining in the zeugopod mesenchyme, the AMZ, and PMZ. With 1.25 µM retinol, HX603 showed a dose-dependent reduction of retinol-induced apoptosis; however 2.5 µM BMS 453 upregulated apoptosis in susceptible regions (Figs. 6f and 7f). At 62.5 µM retinol, both BMS 453 and HX603 were less successful in reducing the numbers of dying cells, although zones of cells death were less extensive and the limbs were better elongated and outgrown than those treated with retinol alone (Figs. 6g-i, 7g-i, 9g-i, and 10g-i). The higher concentrations of BMS 453 and HX603 combined with 62.5 µM retinol induced degeneration of the limb tissue and widespread cell death; cells stained TUNEL positive in the AER and in the INZs (Figs. 6i, 7i, 9i, and 10i). Again, some of the cell fragments and cells with pycnotic nuclei in treated limbs did not stain TUNEL-positive, indicating that at high concentrations these compounds may induce nonspecific effects.
Thus, culturing limbs with BMS 453 and HX603 in the presence of teratogenic concentrations of retinol modified the amount of cell death occurring in limbs. BMS 453 at 0.5 µM and HX603 at 12 µM had the greatest success in reverting the pattern and amount of apoptosis occurring in the retinol-treated limbs to control-like levels. At the lower concentration of retinol, both antagonists restored the incidence of apoptosis in the limbs to close to control. After exposure to 62.5 µM retinol and the lower concentration of either antagonist, more cells died by apoptosis rather than by necrosis or an alternative cell death pathway. In contrast, exposure to the higher concentrations of either BMS 453 or HX603 in the presence of retinol induced extensive cell death in the limbs. For all treatments, limb areas destined to differentiate into cartilage were the most resistant to the apoptosis-inducing effects of retinol, BMS 453, or HX603. The extent of apoptosis seen in limb sections after 24-h culture corresponded to the severity of malformations seen after allowing limbs to develop for 6 days. Decreasing signaling through the RARs produced a pattern of apoptosis distinct from that seen when signaling through the RXRs was blocked. These differences in apoptosis translated to correspondingly different abnormal limb morphology after culture for 6 days.
Quantification.
The effects of retinol, BMS 453, or HX603 on apoptosis were quantified (Figs. 8 and 11). These results reflected the histological changes reported above; exposure to 1.25 or 62.5 µM retinol induced a significant dose-dependent increase in TUNEL-staining compared to control (p < 0.001). Exposure to the higher concentration of BMS 453, 2.5 µM, significantly increased the TUNEL-positive area compared to control limbs (p < 0.05) (Fig. 8). At 1.25 µM retinol, 0.5 µM BMS 453 significantly decreased, and 2.5 µM significantly increased the TUNEL-positive area (p < 0.05). At 62.5 µM retinol, both 0.5 and 2.5 µM BMS 453 significantly decreased the TUNEL-positive areas (p < 0.05). Increasing concentrations of HX603 produced a dose-dependent increase in the area of TUNEL-positive staining in the representative zones (p < 0.005). At 1.25 µM retinol, HX603 induced a dose-dependent, although not significant, decrease in TUNEL-positive staining. At 62.5 µM retinol, 5 µM HX603 produced a significant decrease in TUNEL-positive area (p < 0.001).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Specific antagonists targeted to either member of the RAR/RXR heterodimer attenuate retinoid signaling in the mid-organogenesis mouse limb bud, increasing apoptosis and causing limb malformations. Given that exposure to a retinoid agonist upregulates apoptosis, it seemed intuitive that retinoid antagonists would downregulate apoptosis. Instead, a U-shaped dose-response curve, with respect to abnormal morphology and to the induction of apoptosis, was observed. These findings reiterate that the proper balance of bioactive retinoids is essential for normal limb development and extend this axiom to the induction of apoptosis.
Qualitative and quantitative differences in the extent of drug-induced apoptosis translate to corresponding differences in the severity and phenotype of malformations, underlining the crucial role cell death plays in development. Many teratogens with disparate mechanisms of action upregulate apoptosis in the limb (Ali-Khan and Hales, 2003; Huang and Hales, 2002; Zakeri and Ahuja, 1994, 1997); this effect is cell specific and occurs preferentially in areas of the embryo predestined for normal developmental programmed cell death by apoptosis (Ahuja et al., 1997; Alles and Sulik, 1989; Zakeri and Ahuja, 1997). We have shown that the patterns of ectopic apoptosis induced by the three chemicals used in this study, retinol, BMS 453, and HX603, occur preferentially in the same broad regions, the INZ, PMZ, AMZ, and the AER, while areas destined to differentiate into cartilage are relatively resistant. The intracellular environment in susceptible cells is likely to be closer to the apoptotic threshold and therefore primed for the reception of pro-apoptotic signaling, containing, for example, lower levels of pro-survival proteins such as Bcl-2 (Novack and Korsmeyer, 1994) or survivin (Mirkes and Little, 2000). Also, to be retinoid responsive, cells require the presence of nuclear receptors, coactivators, and accessory proteins; retinoid machineries, including various subtypes of the RARs and RXRs, are expressed throughout organogenesis in the limb in defined spatiotemporal patterns (For review see, Dolle et al., 1990; Mollard et al., 2000; Niederreither et al., 2002).
While both agonist and antagonist treatments enhance apoptosis and induce dysmorphogenesis, the mechanisms underlying these endpoints may be very different. The localization of active retinoid is tightly constrained in the developing limb (Rossant et al., 1991); thus the addition of retinoid receptor ligands, whether they are retinoic acid, BMS 453 or HX603, is likely to induce chaotic messaging. Altered retinoid signaling would lead to an altered phenotype in different populations of limb cells depending on their retinoid requirements and on their susceptibility to apoptosis. Either decreased or increased signaling would disturb normal gene expression profiles, which may be manifested as death in highly susceptible cells, or as altered patterning in more resistant areas, such as the limb bud core. In addition, a further level of complexity is indicated by recent work showing that retinoid receptors function to repress gene expression in the absence of ligand (for review see, Weston et al., 2003).
The RARs and the RXRs play key roles in generating limb malformation. Exposure to retinol, BMS 453, and HX603 lead to qualitative differences in abnormal phenotype. These differing patterns may be partially a function of the different affinities of each ligand for various RAR or RXR receptor subtypes and isoforms, leading to qualitative and quantitative differences in gene expression and subsequent apoptosis; to wit, retinoid receptor subtypes mediating apoptosis in the AMZ and PMZ are distinct from those in the INZs (Dupe et al., 1999). The apoptotic patterns induced by BMS 453 were also different from those induced by retinol, favoring the explanation of varying isoform affinity.
The discovery that a selective RXR-antagonist, HX603, induced apoptosis and limb malformations on its own and, additionally, countered the effects of excess retinol in the limb was interesting. RXR agonists are not teratogenic, RXR homodimers have not to this point been detected in the limb, and transcriptional effects due to RXR-binding with other receptors, such as LXR or NGFI-B, have not been highlighted as important in normal or abnormal limb development (Jiang et al., 1995; Sucov et al., 1995).
While several studies have failed to detect 9-cis retinoic acid in the embryo under physiological conditions (Mic et al., 2003; Ulven et al., 2000), after maternal administration of teratogenic doses of retinol or retinoic acid, concentrations of both 9-cis and all-trans retinoic acid increase rapidly (Horton and Maden, 1995; Mic et al., 2003; Tzimas et al., 1996). Hence, transcriptional activity through the RXR portion of the heterodimer may only be activated under teratogenic conditions, explaining the success of HX603 in improving retinol-induced defects. The RXR mutant is the only retinoid receptor knock-out to be completely resistant to the teratogenic effects of excess Vitamin A on limb development, emphasizing the importance of the RXRs in transducing retinoid signaling within the RAR/RXR heterodimer (Sucov et al., 1995).
Diverse studies show that the ligand-bound status of the RAR partner in the RAR/RXR heterodimer affects the ability of the RXR partner to bind ligand and transactivate (Chambon, 1996; Leid et al., 1992; Nagpal et al., 1993; Vivat et al., 1997). Likewise, the status of the RXR partner affects RAR activity; while suboptimal concentrations of an RAR-agonist generated a very weak footprinting pattern at an RARE, the addition of an RXR-selective ligand greatly enhanced promoter occupancy and increased the stability of the heterodimer (Minucci et al., 1997). Transactivation studies have shown that HX603 inhibits the activity of the RAR/RXR heterodimer, unlike previously synthesized RXR antagonists, which act as RAR/RXR heterodimer agonists (Ebisawa et al., 1999).
Thus, both BMS 453 and HX603 downregulate aberrant retinol-induced apoptosis and ameliorate retinol-induced limb defects. These data indicate that both retinoid receptor subfamilies, the RARs and the RXRs, play key roles in generating limb malformation.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Supplementary data are available online at http://toxsci.oxfordjournals.org/.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by the Canadian Institutes for Health Research. We should like to thank Dr. Janet Rossant for the gift of the ßRARE-lacZ reporter mice, and Eugene Galdones for generating and maintaining the colony. Also Dr. Hiroyuki Kagechika and Bristol Myers Squibb for their gifts of the retinoid antagonists, HX603 and BMS 189453, respectively, and to Dr. Chris Zusi for his technical advice regarding BMS 189453. Thank you also to Dr. Bernard Robaire for his critical reading of the manuscript.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Ahuja, H. S., James, W., and Zakeri, Z. (1997). Rescue of the limb deformity in hammertoe mutant mice by retinoic acid-induced cell death. Dev. Dyn. 208, 466–481.[CrossRef][Web of Science][Medline]
Ali-Khan, S. E., and Hales, B. F. (2003). Caspase-3 mediates retinoid-induced apoptosis in the organogenesis-stage mouse limb. Birth Defects Res. A Clin. Mol. Teratol. 67, 848–860.[CrossRef][Web of Science][Medline]
Alles, A. J., and Sulik, K. K. (1989). Retinoic-acid-induced limb-reduction defects: Perturbation of zones of programmed cell death as a pathogenetic mechanism. Teratology 40, 163–171.[CrossRef][Web of Science][Medline]
Arafa, H. M., Elmazar, M. M., Hamada, F. M., Reichert, U., Shroot, B., and Nau, H. (2000). Selective agonists of retinoic acid receptors: Comparative toxicokinetics and embryonic exposure. Arch. Toxicol. 73, 547–556.[CrossRef][Web of Science][Medline]
Chambon, P. (1994). The retinoid signaling pathway: Molecular and genetic analyses. Semin. Cell Biol. 5, 115–125.[CrossRef][Medline]
Chambon, P. (1996). A decade of molecular biology of retinoic acid receptors. Faseb J. 10, 940–954.[Abstract]
Chen, J. Y., Penco, S., Ostrowski, J., Balaguer, P., Pons, M., Starrett, J. E., Reczek, P., Chambon, P., and Gronemeyer, H. (1995). RAR-specific agonist/antagonists which dissociate transactivation and AP1 transrepression inhibit anchorage-independent cell proliferation. EMBO J. 14, 1187–1197.[Web of Science][Medline]
Dolle, P., Ruberte, E., Leroy, P., Morriss-Kay, G., and Chambon, P. (1990). Retinoic acid receptors and cellular retinoid binding proteins. I. A systematic study of their differential pattern of transcription during mouse organogenesis. Development 110, 1133–1151.
Dupe, V., Ghyselinck, N. B., Thomazy, V., Nagy, L., Davies, P. J., Chambon, P., and Mark, M. (1999). Essential roles of retinoic acid signaling in interdigital apoptosis and control of BMP-7 expression in mouse autopods. Dev. Biol. 208, 30–43.[CrossRef][Web of Science][Medline]
Ebisawa, M., Umemiya, H., Ohta, K., Fukasawa, H., Kawachi, E., Christoffel, G., Gronemeyer, H., Tsuji, M., Hashimoto, Y., Shudo, K., et al. (1999). Retinoid X receptor-antagonistic diazepinylbenzoic acids. Chem. Pharm. Bull. (Tokyo) 47, 1778–1786.[Medline]
Elmazar, M. M., Ruhl, R., and Nau, H. (2001). Synergistic teratogenic effects induced by retinoids in mice by coadministration of a RARalpha- or RARgamma-selective agonist with a RXR-selective agonist. Toxicol. Appl. Pharmacol. 170, 2–9.[CrossRef][Web of Science][Medline]
Francis-West, P., and Tickle, C. (1996). Limb development. Curr. Top. Microbiol. Immunol. 212, 239–259.[Web of Science][Medline]
Horton, C., and Maden, M. (1995). Endogenous distribution of retinoids during normal development and teratogenesis in the mouse embryo. Dev. Dyn. 202, 312–323.[Web of Science][Medline]
Huang, C., and Hales, B. F. (2002). Role of caspases in murine limb bud cell death induced by 4-hydroperoxycyclophosphamide, an activated analog of cyclophosphamide. Teratology 66, 288–299.[CrossRef][Web of Science][Medline]
Jiang, H., Penner, J. D., Beard, R. L., Chandraratna, R. A., and Kochhar, D. M. (1995). Diminished teratogenicity of retinoid X receptor-selective synthetic retinoids. Biochem. Pharmacol. 50, 669–676.[CrossRef][Web of Science][Medline]
Kastner, P., Mark, M., Ghyselinck, N., Krezel, W., Dupe, V., Grondona, J. M., and Chambon, P. (1997). Genetic evidence that the retinoid signal is transduced by heterodimeric RXR/RAR functional units during mouse development. Development 124, 313–326.[Abstract]
Kochhar, D. M. (1973). Limb development in mouse embryos. I. Analysis of teratogenic effects of retinoic acid. Teratology 7, 289–298.[CrossRef][Web of Science][Medline]
Kochhar, D. M. (1985). Skeletal morphogenesis: Comparative effects of a mutant gene and a teratogen. Prog. Clin. Biol. Res. 171, 267–281.[Medline]
Kochhar, D. M., Jiang, H., Penner, J. D., Beard, R. L., and ChandraratnaAs (1996). Differential teratogenic response of mouse embryos to receptor selective analogs of retinoic acid. Chem. Biol. Interact. 100, 1–12.[CrossRef][Web of Science][Medline]
Kwasigroch, T. E., Skalko, R. G., and Church, J. K. (1984). Mouse limb bud development in submerged culture: Quantitative assessment of the effects of in vivo exposure to retinoic acid. Teratog. Carcinog. Mutagen. 4, 311–326.[CrossRef][Web of Science][Medline]
Leid, M., Kastner, P., and Chambon, P. (1992). Multiplicity generates diversity in the retinoic acid signalling pathways. Trends Biochem. Sci. 17, 427–433.[CrossRef][Web of Science][Medline]
Matt, N., Ghyselinck, N. B., Wendling, O., Chambon, P., and Mark, M. (2003). Retinoic acid-induced developmental defects are mediated by RARbeta/RXR heterodimers in the pharyngeal endoderm. Development 130, 2083–2093.
Mic, F. A., Molotkov, A., Benbrook, D. M., and Duester, G. (2003). Retinoid activation of retinoic acid receptor but not retinoid X receptor is sufficient to rescue lethal defect in retinoic acid synthesis. Proc. Natl. Acad. Sci. U.S.A. 100, 7135–7140.
Mic, F. A., Sirbu, I. O., and Duester, G. (2004). Retinoic acid synthesis controlled by Raldh2 is required early for limb bud initiation and then later as a proximodistal signal during apical ectodermal ridge formation. J. Biol. Chem. 279, 26698–26706.
Minucci, S., Leid, M., Toyama, R., Saint-Jeannet, J. P., Peterson, V. J., Horn, V., Ishmael, J. E., Bhattacharyya, N., Dey, A., Dawid, I. B., et al. (1997). Retinoid X receptor (RXR) within the RXR-retinoic acid receptor heterodimer binds its ligand and enhances retinoid-dependent gene expression. Mol. Cell. Biol. 17, 644–655.[Abstract]
Mirkes, P. E., and Little, S. A. (2000). Cytochrome c release from mitochondria of early postimplantation murine embryos exposed to 4-hydroperoxycyclophosphamide, heat shock, and staurosporine. Toxicol. Appl. Pharmacol. 162, 197–206.[CrossRef][Web of Science][Medline]
Mollard, R., Viville, S., Ward, S. J., Decimo, D., Chambon, P., and Dolle, P. (2000). Tissue-specific expression of retinoic acid receptor isoform transcripts in the mouse embryo. Mech. Dev. 94, 223–232.[CrossRef][Web of Science][Medline]
Nagpal, S., Friant, S., Nakshatri, H., and Chambon, P. (1993). RARs and RXRs: Evidence for two autonomous transactivation functions (AF-1 and AF-2) and heterodimerization in vivo. EMBO J. 12, 2349–2360.[Web of Science][Medline]
Niederreither, K., Vermot, J., Schuhbaur, B., Chambon, P., and Dolle, P. (2002). Embryonic retinoic acid synthesis is required for forelimb growth and anteroposterior patterning in the mouse. Development 129, 3563–3574.
Novack, D. V., and Korsmeyer, S. J. (1994). Bcl-2 protein expression during murine development. Am. J. Pathol. 145, 61–73.[Abstract]
Neubert, D. and Barrach, H.-J. (1977). Techniques applicable to study morphogenetic differentiation of limb buds in organ culture In Methods in pre-natal toxicology; evolution of embryotoxic effects in experimental animals. (D. Neubert, H.-J. Merker, T. Knaisgroch, Eds.), pp. 241–251. Georg Theme Pubs., Stuttgart, Germany.
Rossant, J., Zirngibl, R., Cado, D., Shago, M., and Giguere, V. (1991). Expression of a retinoic acid response element-hsplacZ transgene defines specific domains of transcriptional activity during mouse embryogenesis. Genes Dev. 5, 1333–1344.
Sucov, H. M., Izpisua-Belmonte, J. C., Ganan, Y., and Evans, R. M. (1995). Mouse embryos lacking RXR alpha are resistant to retinoic-acid-induced limb defects. Development 121, 3997–4003.[Abstract]
Tzimas, G., Collins, M. D., Burgin, H., Hummler, H., and Nau, H. (1996). Embryotoxic doses of vitamin A to rabbits result in low plasma but high embryonic concentrations of all-trans-retinoic acid: Risk of vitamin A exposure in humans. J. Nutr. 126, 2159–2171.
Ulven, S. M., Gundersen, T. E., Weedon, M. S., Landaas, V. O., Sakhi, A. K., Fromm, S. H., Geronimo, B. A., Moskaug, J. O., and Blomhoff, R. (2000). Identification of endogenous retinoids, enzymes, binding proteins, and receptors during early postimplantation development in mouse: Important role of retinal dehydrogenase type 2 in synthesis of all-trans-retinoic acid. Dev. Biol. 220, 379–391.[CrossRef][Web of Science][Medline]
Vivat, V., Zechel, C., Wurtz, J. M., Bourguet, W., Kagechika, H., Umemiya, H., Shudo, K., Moras, D., Gronemeyer, H., and Chambon, P. (1997). A mutation mimicking ligand-induced conformational change yields a constitutive RXR that senses allosteric effects in heterodimers. EMBO J. 16, 5697–5709.[CrossRef][Web of Science][Medline]
Weston, A. D., Blumberg, B., and Underhill, T. M. (2003). Active repression by unliganded retinoid receptors in development: Less is sometimes more. J. Cell Biol. 161, 223–228.
White, J. A., Beckett-Jones, B., Guo, Y. D., Dilworth, F. J., Bonasoro, J., Jones, G., and Petkovich, M. (1997). cDNA cloning of human retinoic acid-metabolizing enzyme (hP450RAI) identifies a novel family of cytochromes P450. J. Biol. Chem. 272, 18538–18541.
Yang, L., Munoz-Medellin, D., Kim, H. T., Ostrowski, J., Reczek, P., and Brown, P. H. (1999). Retinoic acid receptor antagonist BMS453 inhibits the growth of normal and malignant breast cells without activating RAR-dependent gene expression. Breast Cancer Res. Treat. 56, 277–291.[CrossRef][Web of Science][Medline]
Zakeri, Z. F., and Ahuja, H. S. (1994). Apoptotic cell death in the limb and its relationship to pattern formation. Biochem. Cell Biol. 72, 603–613.[Web of Science][Medline]
Zakeri, Z. F., and Ahuja, H. S. (1997). Cell death/apoptosis: Normal, chemically induced, and teratogenic effect. Mutat. Res. 396, 149–161.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. M. Rogers Casting a Broad Network: Fishing for Mechanisms of Retinoid Teratogenicity Toxicol. Sci., November 1, 2006; 94(1): 1 - 2. [Full Text] [PDF]
|
|
|
|
|
|
S. E. Ali-Khan and B. F. Hales Novel Retinoid Targets in the Mouse Limb during Organogenesis Toxicol. Sci., November 1, 2006; 94(1): 139 - 152. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||