Antagonists of retinoic acid and BMP4 affect fetal mouse osteogenesis and odontoblast differentiation

Pathophysiology 18 (2011) 103–109

Antagonists of retinoic acid and BMP4 affect fetal mouse osteogenesis and odontoblast differentiation Despina S. Koussoulakou, Lukas H. Margaritis, Stauros L. Koussoulakos ∗ Department of Cell Biology and Biophysics, Faculty of Biology, N. & K. University of Athens, Panepistimiopolis 15784, Greece Received 16 October 2009; accepted 19 February 2010

Abstract Retinoic acid and bone morphogenetic protein (BMP4) are endogenous factors indispensable for the physiological development of vertebrates. The proximate aim of the present study was to investigate whether the natural compound citral (a retinoic acid synthesis inhibitor) and a monoclonal, anti-BMP4 antibody, administered to pregnant mice affect in the fetuses cranial osteogenesis and odontoblast differentiation. The present investigation was motivated by the fact that, retinoic acid inhibitors and BMP4 neutralizers may frequently contact human tissues (both intentional and unintentional, and/or unconsciously) inducing unanticipated effects. Our ultimate goal is the prevention of side effects and, future clinical implementation of the results. To this end, pregnant, white mice (balb-c Mus musculus) were intra-abdominally injected with either citral or anti-BMP4 antibody at the 9th gestational day. Newborns were processed within 5 h, postnatal. Results were evaluated (a) macroscopically, (b) stereoscopically, following histochemical double staining of cartilage and osseous tissues and, (c) microscopically after (c1 ) histological staining of paraffin sections, and, (c2 ) immunohistochemical detection of apoptosis. Data indicate that in vivo administration of citral (biomimicking hypovitaminosis A) caused restriction/retardation of cranial chondrogenesis and osteogenesis. Apoptosis was not detected in teeth tissues. In vivo administration of anti-BMP4 antibody resulted in a transitory interference with the normal course of odontoblast differentiation and the production of pre-dentin, whereas, delay in the ossification also included the alveoli. Animals inspected in adulthood displayed a fairly normal phenotype. It is concluded that those two substances, under their concentrations experienced, are quite safe for the public. © 2010 Elsevier Ireland Ltd. All rights reserved. Keywords: BMP4; Citral; Craniofacial variations; Odontoblasts; Teeth

1. Introduction Head and teeth formation in the mammalian embryo are extremely complex processes. Various cells of ectodermal, mesenchymal and ectomesenchymal origin interact at specific positions and times to form distinct elements. The normal development of most embryonic structures depends mainly on an intricate but highly orchestrated interplay between spatiotemporally and differentially expressed activators and their antagonistic molecules [1,2]. Among those factors significant roles play retinoic acid (RA) and bone morphogenetic protein 4 (BMP4) which are interrelated, very influential molecules in many developmental processes [3]. Several disturbances in their interplay might lead to forma∗

Corresponding author. Tel.: +30 210 7274612; fax: +30 210 7274742. E-mail address: [email protected] (S.L. Koussoulakos).

0928-4680/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pathophys.2010.02.001

tion of abnormal structures and therefore any deviation from the norm would indicate whether an exogenously administered substance is harmful for the organism. Consequently, the purpose of the present work was to further investigate the effect of RA and BMP4 signaling antagonists (citral [4,5] and anti-BMP4 antibodies [6], respectively) on fetal mouse head and teeth morphogenesis, differentiation and apoptosis, since citral is a highly consumed material and, anti-BMP4 antibodies might be induced in cases where BMP4 is used as a prospective clinical factor for the induction of reparative osteogenesis [7,8] and dentinogenesis [7,9]. RA, a derivative of vitamin A, is an important signaling molecule during embryonic development and continues to play significant roles for the adult, because it regulates a wide variety of pathways in cellular differentiation and homeostasis in various tissues and organs, head and teeth included, by inducing a series of homeobox genes [10,11]. Unnatu-

104

D.S. Koussoulakou et al. / Pathophysiology 18 (2011) 103–109

rally high quantities of RA, (hypervitaminosis A), as well as its insufficiency (hypovitaminosis A, which in fact is a worldwide problem of great magnitude) during embryonic development cause a great spectrum of congenital malformations [12,13], as well as in pathological consequences for the adult [14,15]. Therefore, the household of such a potent physiological factor should be carefully regulated. Animal cells cannot produce vitamin A de novo. The in vivo generation of vitamin A derivatives includes cellular ingestion of retinol (Vitamin A), a rate-limiting first oxidation of retinol to retinaldehyde by a retinol/alcohol dehydrogenase, and a second oxidation of retinaldehyde to RA by a retinal dehydrogenase (RALDH) [16,17]. Citral, (3,7-dimethyl-2,6-octadienal, C10 H16 O), a versatile and highly reactive aldehyde and a well known inhibitor of RALDH [18,19], is a highly consumed, particularly by housewife, product, widely used primarily as lemon flavoring in foods, beverages, candies, detergents, perfumes, cosmetics, insecticides, various toiletries and other consumer products. It is mainly found as a compound of essential oils of many plants, such as Backhousia citriodora, Melissa officinalis, Litsea cubeba, etc., which produce a large variety of edible vegetables and fruits, such as lemon, orange, lime, apricot, tomatoes, grapefruits, celery, etc. [20]. The toxicity of citral is well documented [21]. Because of its widespread use citral was nominated for toxicological study by the National Cancer Institute [22]. Curiously enough, the issue has not yet gained considerable attention, although early experiments revealed that citral actually interferes with the physiological development of a variety of organisms causing a lot of deviations from the norm, such as inhibition of odontogenesis in mice [23], pattern modification in regenerating amphibian limbs [24], p53 increased expression [25], neural tube formation abnormalities [26], prostate hyperplasia in the rats [27], etc. Apart from that, citral exerts a great variety of toxic symptoms to adults due mainly to its deleterious action on many organs [27–30]. The bone morphogenetic proteins (BMPs) are members (more than 30 known up to day in the mammalian system) of the TGF-␤ superfamily of growth factors originally identified by their bone-inducing activities [31]. Their widespread expression in many embryonic organs and tissues and in a vast spectrum of animal species (invertebrates included) suggests many roles other than in osteogenesis [32]. Their significance in embryogenesis is demonstrated by deleting specific genetic sequences (knock out) by the developing organism; with the exception of BMP5, lack of any other BMP leads to death of the organism [33]. Three examples (relevant to the present work) are sufficient to highlight the developmental significance of BMPs: (a) Their inhibition is needed for neuralization and head formation [34]; (b) BMPs stimulate the differentiation of dental pulp mesenchymal cells into odontoblast-like cells, and are investigated as agents for vital pulp therapy [35]; (c) it is well documented that a principal target of BMP4 is the gene for RALDH resulting in attenuation of its expression and, consequently, to reduced

activation of the retinoid signaling pathway [3]. In its turn, blocking RA synthesis decreases BMP4 expression [36]. The elucidation of the mode of action of morphogenetic factors is important for understanding fundamental mechanisms of development as well as for revealing the common denominator of numerous birth defects and for providing new strategies for their prevention and cure.

2. Materials and methods 2.1. Animal handling and experimentation Healthy, young, nulliparus female mice (Balb/c Mus musculus, from Hellenic Pasteur Institute) of similar age (two months old) and size (30 g) were quarantined in our laboratory for two weeks to assess their health and to allow them to acclimatize in a thermostatically maintained room (22 ± 1 ◦ C) and artificial illumination (12:12 h light/dark photoperiod, lights on at 07:00 a.m.) with standard food, tap water ad libitum and frequent air changes. All animal handling and experimentation were carried out according to the ethics recommendation of the European Communities Council Directive. After successful mating by different males, equivalent (i.e., of approximately similar age and body weight) animals were used all of them being at the 9th gestational day. Ten pregnant females were given a single injection intra-abdominally each with 3 ␮L citral (Sigma/Aldrich, Steinheim, Germany)/gram of body weight (gbw) (approx. 3 ␮M citral/gbw) and 10 others were similarly injected (after a series of tentative administrations) with 1.5 ␮g/200 ␮L of anti-BMP4/gbw mouse anti-human monoclonal antibody, (clone 3H2, Novocastra, Newcastle, U.K.). Many observations [37–39] suggest that antibody digestion or lack of absorption are not a problem in similar experiments with mice. Control animals (n = 10) were similarly injected with saline. The results were evaluated on the newborns. Within 5 h postnatal most offspring were fixed and carefully inspected by binocular stereoscopy. Two newborns from each litter (i.e., 20 newborns from the “citral” group, 20 newborns from the “anti-BMP4” group, and 20 from the control group) were anesthetized, fixed in 95% alcohol and processed for histochemical double staining (Alcian blue – Sigma/Aldrich, Steinheim, Germany, for cartilage, alizarin red S – Ferak, Berlin, Germany, for osseous tissue) [40] and clearing (in glycerine – AppliChem, Darmstadt, Germany) in order to reveal craniofacial skeletal elements. Further, 10 newborns from each group (mixed from different litters) were anesthetized and fixed in 5% aqueous formaldehyde solution and subjected to Domagk’s histochemical staining [41] of paraffin sections, and to immunohistochemical DAB staining to reveal activation of caspase-3 (Peroxidase/DAB+, Rabbit/Mouse, Dako REALTM detection Systems, Switzerland Products, 1:100 dilution of manufacturer’s stock). Many newborns (from all litters) remained with their mothers until eruption of all their teeth.

D.S. Koussoulakou et al. / Pathophysiology 18 (2011) 103–109

2.2. Statistical analysis All analyses were two-tailed for a significance level of 5%. One-way analysis of variance (ANOVA) was applied for continuous maternal and fetal variables. We consider as quantitative “different” a value far from the mean of the control group by 2 standard deviations.

3. Results The amount of the drug given to the experimental animals was chosen after a series of tentative, preliminary administrations. Quantities of citral lower than 1 ␮L/gbw did not influence mouse embryos, whereas, quantities higher than 6 ␮L/gbw resulted in retardation of embryonic development, skeletal anomalies, increased liver mass, abortions and, lethality to injected animals. 3.1. Macroscopic and stereoscopic inspection of whole newborn mice Inspection included body weight and length, limb and digit number and structure, head, eyes, ears, nostrils, whiskers. Under our chosen conditions, no significant differences were detected between experimental (anti-BMP4- or citral-treated) and control animals. Teeth erupted at the expected intervals, whereas grown animals appeared normal. On the other hand, double staining of cranial skeletal elements revealed various differences. The skull of the mouse consists of 41 bones (15 endochondral, 26 membranous). Membranous bones include the flat bones of the skull, face and jaws. Their development starts when mesenchymal cells differentiate directly into osteoblasts that synthesize the collagenous, to be mineralized osteoid. These elements are derived from the embryonic ectomesenchymal neural crest cells and the cephalic mesenchyme. The formation of each of these elements and their fusion is highly coordinated in time and space. The presence of citral and anti-BMP4 (biomimicking vitamin A deficiency) during gestation, results in a slight retardation in the process of osteogenesis. Double staining of osseous tissue (red) and mucopolysaccharides (blue) reveals that the blue colour in treated animals is restricted as compared to the control group, a fact, which suggests a disturbance in the ossification process (Fig. 1A, A1 , B, B1 ). Biomineralization as a marker of mature osteoblasts was examined using alizarin red staining. In agreement with the well established knowledge that BMPs and RA are involved in the induction of osteoblast differentiation [42,43], citral administration (i.e., lowering of the available RA) results in a decrease in the area stained by Alcian blue. 3.2. Histological staining of paraffin sections The jaws of vertebrates derive from the first pharyngeal arch which forms the nasofacial eminence and the bilateral

105

Table 1 Summary of experimental design and observed incidence of variations in newborn mice whose mothers were exposed to citral or anti-BMP4 antibody. (a)

(b)

(c)

(d)

(e)

(f)

(g)

10 (citral) 10 (anti-BMP4) 10 (saline)

[10] [10] [10]

20 [10] 20 [10] 20 [10]

10 [10] 10 [10] 10 [10]

8 [10] 10 [10] 1 [10]

7 [10] 8 [10] 0 [10]

3 [3] 3 [3] 3 [3]

(a) Number of pregnant mice exposed to the (indicated) agent. (b) [Litters]. This means that all treated pregnant mice from (a) have given viable offspring. (c) Number of newborns randomly taken from [all litters], doublestained and processed for clearing in toto. (d) Number of newborns randomly taken from [all litters] and processed for paraffin histology and apoptosis. (e) Number of newborns taken from [all litters] with cranial bone variation (particularly nasal and interparietal) as revealed by double staining in toto. (f) Number of newborns taken from [all litters] displaying by paraffin sectioning and staining disorganized alveolar bone formation and odontoblast arrangement. (g) Number of newborns taken from [3 litters] examined for caspase-3 apoptosis, and proved negative. All experimental data are significantly different from the control group using Fisher’s exact test.

maxillary and mandibular gnathic processes. These processes are fused, beginning with the mandibular ones, thus forming the mandible. The mandible belongs to viscerocranium I and is formed through cartilagenous osteogenesis from neural crest cells. Meckel’s cartilage and membranous bone formed around it also play an important role in mandibular development. Paraffin sections stained using the Domagk’s method [41] show that anti-BMP4 antibody injection in a female mouse during the 9th gestational day delays the differentiation and stratification of pre-odontoblasts and the secretion of pre-dentin (Fig. 1C1 ). Pre-dentin, on the other hand, is well formed by the already differentiated odontoblasts in normal subjects (Fig. 1C). Sections at the level of the alveolar bones display a normal pattern of osteogenesis in control specimens (Fig. 1D) with respect to anti-BMP4-treated ones (Fig. 1D1 ), where the alignment of the odontoblasts is disorganized. 3.3. Apoptosis Apoptotic cell death has been recognized for several decades to play important role in tissue morphogenesis and remodelling during embryonic development [44]. The role of apoptosis in shape formation of developing teeth is also well established [45]. Our results show that, although citral manages to retard some processes during development, similar delay in the timing of programmed cell death (which normally takes place a little earlier) cannot be detected at incisor and molar regions at this comparatively late developmental stage (Fig. 1E, E1, F, F1 , respectively) (Table 1).

4. Discussion Any variation, deviation, or abnormality in body structures might reflect a modification of the physiological, developmental program. Analysis of congenital malformations might shed light into developmental mechanisms on the one hand,

106

D.S. Koussoulakou et al. / Pathophysiology 18 (2011) 103–109

Fig. 1. Effect of citral (A1, B1, E1, F1) and anti-BMP4 (C1, D1) on cranial bone formation (A, B), odontoblast differentiation (C), alveolar bone osteogenesis (D) and apoptosis in tooth tissues (E, F), where A, B, C, D, E, F, are the corresponding controls. (A, A1, B, B1): Dorsal aspect of cleared newborn mice crania. Double staining by Alcian blue and Alizarin red S reveals mucopolysaccharides (blue colour) and osseous tissue (red colour), respectively. Intramembraneous skull roof “bones” are not yet ossified and mineralized (A1, B1), therefore they are transparent and let us see the chondrogenic already ossified basicranial bones (basioccipital = bo and basisphenoid = bs). Note lack of lambdoid suture in A1 and, absence of frontonasal (fn) eminence in B1. (C, C1, D, D1): Histological sections of newborn mouse molar buds at the crown (C) and root (D) regions, respectively. Injection of anti-BMP4 antibody in mice during their 9th gestational day (time of neural crest cell migration and formation of 1st pharyngeal arch), results in significant delay in odontoblast differentiation in embryos. Note in C1 lack in of pre-dentin layer (blue layer) between adamantinoblasts (ad) and odontoblasts (od). Similarly, delineation of odontoblasts and the osteogenic pattern of alveolar bone (ab) seem abnormal in the root (D1). (E, E1, F, F1): The immunohistochemical method applied for detection of caspase fragments in molar (m) and incisor (i) regions with (E1, F1) and without (E, F) citral intervention could not reveal apoptotic figures at this late developmental stage. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of the article.)

but on the other, it could also reveal whether various exogenous substances are potentially hazardous for a developing organism. It must be stressed, however, that the quantity of an exogeneously administered chemical substance which produces teratogenic effects differs significantly from species to species. A dramatic expression of this feature comes from

the sixties, when phocomelia was induced in many children whose mothers received during pregnancy small quantities of thalidomide [46]. Consequently, pharmacological doses considered as safe for various animal models might be highly dangerous for people. Admittedly, results of animal studies cannot be directly translated to human beings; however, the

D.S. Koussoulakou et al. / Pathophysiology 18 (2011) 103–109

inevitable extrapolation of information gathered from animal studies to humans is well justified due to the plethora of anatomical similarities and the conservative nature of developmental and physiological processes [47,48]. Numerous studies have revealed the importance of RA and BMP4 in various morphogenetic processes during embryonic development; lack of the adequate quantities needed during gestation results in a lot of birth defects, craniofacial deformities possessing a prominent role among them [49,50]. Thus, the embryonic craniofacial development may be assigned as a good model for testing effects and side effects of drugs, in vivo. In addition, hypovitaminosis A might be phenocopied by two, widely known substances, the intrinsic morphogenetic factor BMP4 and the highly consumed natural product, citral. For revealing the common denominator of countless birth defects, for assessment of potential risks encountered by endothermal mammalian embryos during gestation and for providing new strategies for their prevention and cure, some pregnant mice were administered with antiBMP4 monoclonal antibody, whereas, others received citral, a known inhibitor of retinoic acid synthesis. Our results seem to be clear and consistent with the established knowledge. 4.1. Macroscopic and stereoscopic inspection of newborns The presence of appropriate quantities of BMP4 and RA in various parts of the fetal body at the right time is considered absolutely necessary for the normal course of development [9,51]. Numerous antagonistic molecules [52] play important roles in regulating availability and concentration of RA and BMP4. In this study a BMP4 neutralizer and a RA inhibitor were administered to pregnant mice but macroscopic inspection of the newborns did not reveal any changes in the normal phenotype. This may suggest that the specific quantity and way of in vivo anti-BMP4 antibody and citral administration does not entail any risk for the newborns and their mothers and may be considered safe. The most probable explanation for this observation is that the free quantities of residual BMP4 and RA are sufficient to induce the normal phenotype [53]. In contrast to the exterior, normal appearance of the offspring, double staining of bone and cartilage reveals a delay in osteogenesis. Delay in osteogenesis observed in this study is in full agreement with the results of other studies [31,42,54,55] related to direct involvement of BMP4 and its antagonists in the developmental processes of neurulation, chondrogenesis and head formation [56]. Since a quantity of BMP4 antagonist is introduced in the pregnant animal, limitation of BMP4 allows the expression of growth and survival factors (e.g., FGF8), elongating, thus, the life of mesenchymal cells and their transition to osteoblasts [57]. In addition, the presence of citral has probably inhibited partially (transiently) retinaldehyde dehydrogenase which in turn diminishes the quantity of RA, therefore lesser activation of the osteocalcin gene [50] resulting in modified pattern of osseous tissue production.

107

4.2. Histological staining of paraffin sections This study compares 4 ␮m histological sections of tooth germs of newborn mice treated with citral and anti-BMP4 antibody during organogenesis to those of non-treated. The insufficiency in RA (caused by citral) and the presence of anti-BMP4 antibody (limiting free BMP4 molecules) allows for a limited expression of the Shh gene which affects the expression of Bmps genes. This, in turn, impedes transformation of mesenchymal cells to osteoblasts [31,32]. It is known that differentiation of pre-odontoblasts is induced by the ameloblast layer [58], which, as shown in Fig. 1C1 , is well formed. Furthermore, thickening of the basal membrane separating pre-ameloblasts from pre-odontoblasts is essential. Consequently, in order to fully understand the role of BMP4 in odontoblast differentiation and apply this knowledge clinically, further studies on the temporospatial expression of BMP4 are necessary. In this case too, lack of deviation from the normal pattern supports indications about the safety of anti-BMP4 antibody use. 4.3. Apoptosis The absence of coloured sediments in both experimental and control animals (Fig. 1E, E1 , F, F1 ) could be interpreted either as a failure to detect apoptosis (e.g., a caspaseindependent apoptosis) or as absence of apoptosis. Since (i) our protocol works very well with positive controls, (ii) our method is standardized to its maximum efficacy (see materials and methods), and (iii) the animals develop to a normal phenotype, we conclude that the physiological apoptosis has happened earlier than anticipated. 5. Conclusion Controlled, experimental administration of citral and anti-BMP4 antibody to pregnant mice affects transiently the developmental course of cranial osteogenesis and dentinogenesis in the fetuses, without noticeable effects by achievement of adulthood and therefore seems to be both safe for the organism and the offspring and, concerning anti-BMP4, promising for intended, guided odontoblast differentiation. Acknowledgements This study was funded by the Special Research Account of the National and Kapodistrian University of Athens (70/4/5709, S.L.K.) and by the State Scholarships Foundation (D.S.K.) of Greece. References [1] X.P. Wang, M. Suomalainen, C.J. Jorgez, M.M. Matzuk, S. Werner, I. Thesleff, Follistatin regulates enamel patterning in mouse incisors by

108

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11] [12]

[13]

[14] [15] [16]

[17]

[18]

[19]

[20]

[21]

D.S. Koussoulakou et al. / Pathophysiology 18 (2011) 103–109 asymmetrically inhibiting BMP signaling and ameloblast differentiation, Dev. Cell. 7 (2004) 719–730. D.B. Dean, J.T. Watson, B.R. Moed, Z. Zhang, Role of bone morphogenetic proteins and their antagonists in healing of bone fracture, Front. Biosci. 14 (2009) 2878–2888. L.M. Hoffman, K. Garcha, K. Karamboulas, M.F. Cowan, L.M. Drysdale, W.A. Horton, T.M. Underhill, BMP action in skeletogenesis involves attenuation of retinoid signalling, J. Cell Biol. 174 (1) (2006) 101–113. J.E. Kronmiller, T. Nguyen, W. Berndt, Instruction by retinoic acid of incisor morphology in the mouse embryonic mandible, Arch. Oral Biol. 40 (7) (1995) 589–595. Y. Song, J.N. Hui, K.K. Fu, J.M. Richman, Control of retinoic acid synthesis and FGF expression in the nasal pit is required to pattern the craniofacial skeleton, Dev. Biol. 276 (2) (2004) 313–329. S. Wislet-Gendebien, F. Bruyère, G. Hans, P. Leprince, G. Moonen, B. Rogister, Nestin-positive mesenchymal stem cells favour the astroglial lineage in neural progenitors and stem cells by releasing active BMP4, BMC Neurosci. 5 (2004) 33–45. L. Shum, X. Wang, A.A. Kane, G.H. Nuckolls, BMP4 promotes chondrocyte proliferation and hypertrophy in the endochondral cranial base, Int. J. Dev. Biol. 47 (6) (2003) 423–431. H. Peng, A. Usas, B. Gearhart, A. Olshanski, H.C. Shen, J. Huard, Converse relationship between in vitro osteogenic differentiation and in vivo bone healing elicited by different populations of muscle-derived cells genetically engineered to express BMP4, J. Bone Miner. Res. 19 (4) (2004) 630–641. A. Ohazama, A. Tucker, P.T. Sharpe, Organized tooth-specific cellular differentiation stimulated by BMP4, J. Dent. Res. 84 (7) (2005) 603–606. S. Koussoulakos, H.J. Anton, Vitamin A as a probe for investigating growth, differentiation and morphogenesis, Biol. Bull. 18 (4) (1992) 313–331. M. Maden, Retinoids and spinal cord development, J. Neurobiol. 66 (7) (2006) 726–738. H. Nau, G. Tzimas, L. Dencker, E.J. Lammer, W.J. Scott, Teratogenicity of vitamin A and retinoids, in: R. Blomhoff (Ed.), Vitamin A in Health and Disease, Dekker, New York, 1998, pp. 615– 663. M. Nobakht, A. Zirak, M. Mehdizadeh, P. Tabatabaeei, Teratogenic effects of retinoic acid on neurulation in mice embryos, Pathophysiology 13 (1) (2006) 57–61. F. Chytil, Retinoic acid: biochemistry, pharmacology, toxicology and therapeutic use, Pharm. Rev. 36 (Suppl.) (1984) 93–100. K.L. Penniston, S.A. Tanumihardjo, The acute and chronic toxic effects of vitamin A, Am. J. Clin. Nutr. 83 (2) (2006) 191–201. M. Clagett-Dame, H.F. DeLuca, The role of vitamin A in mammalian reproduction and embryonic development, Annu. Rev. Nutr. 22 (2002) 347–381. I. Gagnon, G. Duester, P.V. Bhat, Kinetic analysis of mouse retinal dehydrogenase type-2 (RALDH2) for retinal substrates, Biochim. Biophys. Acta 596 (1) (2002) 156–162. A. Kikonyogo, D.P. Abriola, M. Dryjanski, R. Pietruszko, Mechanism of inhibition of aldehyde dehydrogenase by citral, a retinoid antagonist, Eur, J. Biochem. 262 (3) (1999) 704–712. F. Di Renzo, M.L. Broccia, E. Giavini, E. Menegola, Citral, an inhibitor of retinoic acid synthesis, attenuates the frequency and severity of branchial arch abnormalities induced by triazole-derivative fluconazole in rat embryos cultured in vitro, Reprod. Toxicol. 24 (3–4) (2007) 326–332. D.C. Robacker, L.B. Hendry, Neral and geranial: components of the sex pheromone of the parasitic wasp, Itoplectis conquisitor, J. Chem. Ecol. 3 (1977) 563–577. A.J. Hayes, B. Markovic, Toxicity of Australian essential oil Backhousia citriodora (lemon myrtle). Part 2. Absorption and histopathology following application to human skin, Food Chem. Toxicol. 41 (10) (2003) 1409–1416.

[22] NIH: evaluation of the toxic and carcinogenic potential of citral administered orally in laboratory animals, NIH Guide 21 (3) (1992), http://grants.nih.gov/grants/guide/notice-files/not92-008.html. [23] J.E. Kronmiller, C.S. Beeman, T. Nguyen, W. Berndt, Blockade of the initiation of murine odontogenesis in vitro by citral, an inhibitor of endogenous retinoic acid synthesis, Arch. Oral Biol. 40 (1995) 465–652. [24] S.R. Scadding, Citral, an inhibitor of retinoic acid modifies pattern formation during limb regeneration in the axolotl, Ambystoma mexicanum, Rev. Canad. Zool. 77 (11) (1999) 1835–1837. [25] P.J. Duerksen-Hughes, J. Yang, O. Ozcan, p53 induction as a genotoxic test for twenty-five chemicals undergoing in vivo carcinogenicity testing, Environ. Health Perspect. 107 (10) (1999) 805– 812. [26] M. Griffith, M.H. Zile, Retinoic acid, midkine, and defects of secondary neurulation, Teratology 62 (2) (2000) 123–133. [27] M Scolnik, M. Konichezky, G. Tykochinsky, C. Servadio, A. Abramovici, Immediate vasoactive effect of citral on the adolescent rat ventral prostate, Prostate 25 (1) (1994) 1–9. [28] E. Golomb, M. Scolnik, R. Koren, C. Servadio, U. Sandbank, A. Abramovici, Effects of senescence and citral on neuronal vacuolar degeneration in rat pelvic ganglia, Neurotoxicology 22 (1) (2001) 73–77. [29] W.J. Waddell, Thresholds of carcinogenicity of flavours, Toxicol. Sci. 68 (2) (2002) 275–279. [30] N.B. Ress, J.R. Hailey, R.R. Maronpot, J.R. Bucher, G.S. Travlos, J.K. Haseman, D.P. Orzech, J.D. Johnson, M.R. Hejtmancik, Toxicology and carcinogenesis studies of microencapsulated citral in rats and mice, Toxicol. Sci. 71 (2) (2003) 198–206. [31] D. Chen, M. Zhao, G.R. Mundy, Bone morphogenetic proteins, Growth Factors 22 (4) (2004) 233–241. [32] H. Schliephake, Bone growth factors in maxillofacial skeletal reconstruction, Int. J. Oral Maxillofac. Surg. 31 (5) (2003) 469– 484. [33] W. Balemans, W. Van Hul, Extracellular regulation of BMP signaling in vertebrates: a cocktail of modulators, Dev. Biol. 250 (2) (2002) 231–250. [34] J.H. Bennett, P. Hunt, P. Thorogood, Bone morphogenetic protein-2 and -4 expression during murine orofacial development, Arch. Oral Biol. 40 (9) (1995) 847–854. [35] A.S. Tucker, K.L. Matthews, P.T. Sharpe, Transformation of tooth type induced by inhibition of BMP signalling, Science 282 (5391) (1998) 1136–1138. [36] S. Malpel, C. Mendelsohn, W.V. Cardoso, Regulation of retinoic acid signaling during lung morphogenesis, Development 127 (14) (2000) 3057–3067. [37] S. Szabó, Z. Magyar, The site of protein absorption from the peritoneal cavity, Res. Exp. Med. 161 (1973) 335–338. [38] L.C. Sait, M. Galic, J.D. Price, K.R. Simpfendorfer, D.A. Diavatopoulos, T.K. Uren, P.H. Janssen, O.L. Wijburg, R.A. Strugnell, Secretory antibodies reduce systemic antibody responses against the gastrointestinal commensal flora, Int. Immunol. 19 (3) (2007) 257–265. [39] B. Rippe, How to assess transport in animals? Perit. Dial. Int. 29 (Suppl. 2) (2009) 32–35. [40] C. Kimmel, C. Trammel, A rapid procedure for routine double staining of cartilage and bone in fetal and adult animals, Stain Technol. 56 (1981) 271–273. [41] B. Romeis, Mikroskopische Technik, Urban & Schwarzenberg, Muenchen, 1948. [42] Z. Yu, Y. Xing, All-trans retinoic acid inhibited chondrogenesis of mouse embryonic palate mesenchymal cells by down-regulation of TGF-beta/Smad signaling, Biochem. Biophys. Res. Commun. 340 (3) (2006) 929–934. [43] Y. Oki, S. Watanabe, T. Endo, K. Kano, Mature adipocyte-derived dedifferentiated fat cells can trans-differentiate into osteoblasts in vitro and in vivo only by all-trans retinoic acid, Cell Struct. Funct. 33 (2) (2008) 211–222.

D.S. Koussoulakou et al. / Pathophysiology 18 (2011) 103–109 [44] C. Penaloza, S. Orlanski, Y. Ye, T. Entezari-Zaher, M. Javdan, Z. Zakeri, Cell death in mammalian development, Curr. Pharm. Des. 14 (2) (2008) 184–196. [45] E. Matalova, A.S. Tucker, P.T. Sharpe, Death in the life of a tooth, J. Dent. Res. 83 (1) (2004) 11–16. [46] J. Knobloch, U. Rüther, Shedding light on an old mystery: thalidomide suppresses survival pathways to induce limb defects, Cell Cycle 7 (9) (2008) 1121–1127. [47] R.W. Tyl, N. Chernoff, J.M. Rogers, Altered axial skeletal development, Birth Defects Res. B Dev. Reprod. Toxicol. 80 (6) (2007) 451– 472. [48] D.S. Koussoulakou, L.H. Margaritis, S.L. Koussoulakos, A curriculum vitae of teeth: evolution, generation, regeneration, Int. J. Biol. Sci. 5 (3) (2009) 226–243. [49] C.C. Zhu, G. Yamada, M. Blum, Retinoic acid teratogenicity: the role of goosecoid and BMP-4, Cell. Mol. Biol. (Noisy-le-grand) 45 (5) (1999) 617–629. [50] A. Oliva, F. Della Ragione, M. Fratta, G. Marrone, R. Palumbo, V. Zappia, Effect of retinoic acid on osteocalcin gene expression in human osteoblasts, Biochem. Biophys. Res. Commun. 191 (3) (1993) 908–914. [51] M. Hind, A. Gilthorpe, S. Stinchcombe, M. Maden, Retinoid induction of alveolar regeneration: from mice to man? Thorax 64 (5) (2009) 451–457.

109

[52] R.W. Stottmann, R.M. Anderson, J. Klingensmith, The BMP antagonists Chordin and Noggin have essential but redundant roles in mouse mandibular outgrowth, Dev. Biol. 240 (2) (2001) 457–473. [53] W. Liu, J. Selever, D. Murali, X. Sun, S.M. Brugger, L. Ma, R.J. Schwartz, R. Maxson, Y. Furuta, J.F. Martin, Threshold-specific requirements for BMP4 in mandibular development, Dev. Biol. 283 (2) (2005) 282–293. [54] J.H. Bennett, P. Hunt, P. Thorogood, Bone morphogenetic protein-2 and -4 expression during murine orofacial development, Arch. Oral. Biol. 40 (9) (1995) 847–854. [55] C.E. Hotchkiss, J. Latendresse, S.A. Ferguson, Oral treatment with retinoic acid decreases bone mass in rats, Comp. Med. 56 (6) (2006) 502–511. [56] R. Meech, D.B. Edelman, F.S. Jones, H.P. Makarenkova, The homeobox transcription factor Barx2 regulates chondrogenesis during limb development, Development 132 (9) (2005) 2135–2146. [57] A. Suzuki, K. Tanimoto, S. Ohno, Y. Nakatani, K. Honda, N. Tanaka, T. Doi, M. Ohno-Nakahara, K. Yoneno, M. Ueki, K. Tanne, The metabolism of hyaluronan in cultured rabbit growth plate chondrocytes during differentiation, Biochim. Biophys. Acta 1743 (1–2) (2005) 57–63. [58] I. Thesleff, A. Vaahtokari, The role of growth factors in determination and differentiation of the odontoblastic cell lineage, Proc. Finn. Dent. Soc. 88 (Suppl. 1) (1992) 357–368.