|Home | About | Journals | Submit | Contact Us | Français|
To analyze the effects of nicotine on palatal fusion inhibition in vitro and determine if nicotine modulated transforming growth factor β3 or phosphatidylinositol-3 kinase signaling. A second objective was to determine the localization and regulation of nicotinic receptors in the medial edge epithelia (MEE) during palatal fusion.
Palatal shelves from embryonic day (E) 13.5 mice were cultured in serum free media and treated with 0, 0.06, 0.6, or 6 mM nicotine, nicotinic receptor antagonist α-bungarotoxin, or the combination of nicotine and α-bungarotoxin. Tissues harvested at 72 h were analyzed for epithelial-mesenchymal transformation (EMT) and fusion. MEE samples collected at 20 h were analyzed for phosphorylated Akt-Ser473, phosphorylated Smad2, and nicotinic receptors.
Nicotine inhibited palatal fusion in vitro in a dose dependent manner. Activated Akt-Ser473 was greater in control MEE than in nicotine treated tissues; while there was no difference in activated Smad2 between groups. The α7 subunit of nicotinic receptor was expressed in MEE during palate fusion and increased in nicotine treated tissues. Alpha-bungarotoxin did not rescue the nicotine treated palates.
Nicotine treatment had no effect on Smad2, but caused a down regulation of the PI-3 kinase pathway that may have contributed to inhibiting palatal fusion in vitro.
It is generally accepted that exposure to nicotine has adverse effects on fetal development. Higher mortality rates and lower birth weights are associated with maternal smoking during pregnancy (1–3). Animal studies have shown that nicotine administration during pregnancy resulted in delayed implantation and decreased fetal weight (4), abnormal incisor and molar formation (5, 6), and delayed tongue development (7). Maternal cigarette smoking as a risk factor for cleft palate has been demonstrated by epidemiological studies (8–10). In addition, fetal cleft palates can be induced in goats by maternal gavage of nicotinana glauca (11, 12). Cleft palate was also induced in mice by daily maternal nicotine hemisulfate injections from gestational days 6–15 (13).
Normal palatal development requires multiple cell–cell interaction and procedures. Two palatal shelves initially project vertically from the maxillary processes into the oral cavity on embryonic day 12 (E12) in mice and grow on both sides of the tongue. The palatal shelves reorient to a horizontal position superior to the tongue (E13 in mice) and grow to meet in the midline (14, 15). Palatal shelves are composed of craniopharyngeal ectoderm and neural crest derived mesenchyme (16). At these early developmental stages, two epithelial layers cover the developing palatal shelves, the superficial periderm and the basal epithelium. The periderm sloughs off and exposes the basal layer prior to the adherence of opposing palatal shelves (17). Basal epithelia of the medial edges of the contacting shelves form new desmosomes on the apical surface and adhere one another to form an epithelial seam (18). The seam epithelia thin to one layer and divide into islands. They eventually disappear, resulting in the final confluence of the mesenchyme and complete fusion of the palate (14, 19). In the investigation of medial edge epithelia (MEE) disappearance and resulting mesenchymal confluence, several studies suggested that most of the MEE, if not all, go through epithelial-mesenchymal transformation (EMT) (18, 20, 21). During this phenotypic change, MEE lose cell–cell adhesion, change cell shape, degrade the basement membrane, and migrate into mesenchyme. At the same time as MEE transform into mesenchyme, the epithelia on the oral surface become masticatory mucosa, while the epithelia on the nasal surface differentiate into pseudostratified ciliated columnar cells lining the airspace. Disturbances at any stage during this process, i.e., defective palatal shelf growth, failed or delayed shelf elevation, or failure of EMT and fusion can result in cleft palate.
Many biological agents regulate EMT during palate development. There are numerous experiments that support the hypothesis that transforming growth factor (TGF) β3 contributes to palatal fusion (22–25). Tgfβ3 and type II Tgfβ receptors (TβR-II) are localized in the MEE before fusion (26). Tgfβ3 transgenic and knockout mice have cleft palate as their only craniofacial birth defect (22, 23). When palatal shelves from Tgfβ3 knockout mice were cultured, the midline epithelia failed to go through EMT (24). The addition of TGFβ3 into culture media rescued the fusion of the palates from Tgfβ3-null mice (25). Interestingly, although chickens have naturally open palates, the cultured palatal shelves fused when TGFβ3 was added into the media (27).
Some of the TGFβ3 downstream signaling molecules are included in the Smad family (28, 29). Although Smad expression was detected during palatal fusion (30), they have not been proven to be essential factors for EMT. Phosphatidylinositol-3 (PI-3) kinases an alternative downstream signal in the TGFβ pathway as it is involved in actin reorganization, matrix metalloproteinases (MMP) production, and cell mobility (31, 32). Recently, we have demonstrated that LY294002, a specific inhibitor of PI-3 kinase, blocked EMT and palate fusion in vitro (33).
There is evidence that nicotine regulates the production and release of fibroblast growth factor (FGF), TGFβ, and extracellular matrix in smooth muscle cells (34, 35) and vascular endothelial cells (36). Whereas, nicotine modulates EGF and TGFβ in cervical cancer cells (37, 38). Therefore, if nicotine inhibits EMT and palate fusion, it is important to determine the effects of nicotine on growth factors and signaling pathways involved in palate development, such as TGFβ3 and PI-3 kinase.
It is generally recognized that in the neural system, nicotine elicits its effects by interacting with nicotinic acetylcholine receptors (nAChRs). These receptors are also localized in epithelial cell populations (39, 40). Nicotine regulates the expression of nAChRs subunits and the receptors are involved in modulating cell shape, motility, proliferation and differentiation (39–42). We hypothesize that if nAChRs are present in MEE during palate development, they mediate the toxic effects of smoking.
In the current study, we investigated if nicotine inhibits EMT and palatal fusion in vitro and if the effect is regulated through TGFβ3 or PI-3 kinase signaling. We also studied the localization and regulation of nAChRs in MEE during EMT and palatal fusion.
Timed-pregnant CD 1 mice (Harlan Sprague Dawley Inc., Indianapolis, IN, USA) were used in these studies. Animals were maintained under standard conditions at a 10 h:14 h light:dark cycle. Female mice were mated overnight and the day of vaginal plug was timed as day 0. In CD 1 mice, the palate shelves elevate above the tongue between E13 and E13.5 days. Therefore, pregnant females were killed by cervical dislocation at 13.5 day and the embryos were dissected out of the amniotic sacs into a dish of ice-cold Hanks’ balanced salt solution (HBSS; Gibco, Grand Island, NY, USA). The stage of development was determined using morphological features.
The surface epithelia were labeled with (5 and 6) carboxy 2,7′ dichlorofluorescein diacetate succinimidyl ester (CCFSE; Molecular Probes) according to the methods of Griffith and Hay (43). Briefly, 1:500 dilution of 10 mM CCFSE stock solution (in dimethyl sulphoxide, DMSO) was prepared in HBSS and warmed at 37°C prior to use. After mandibles and tongues were removed from the embryos, upper heads were exposed to CCFSE for 1.5 h at 37°C. For controls, tissues were exposed to DMSO under the same conditions. The upper heads were then rinsed in HBSS to remove the unincorporated CCFSE followed by dissection and organ culture of the palatal shelves. At the completion of the experiment, the whole palates were evaluated with confocal microscopy. Complete z-series were obtained, and representative signal optical images from the EMT region were used for figures.
Palatal shelves were dissected from E13.5 mouse embryos and placed nasal side down on HATF filters (0.45 μM pore size; Millipore Corp., Bedford, MA, USA) with their medial edges in contact. The tissues were suspended at the air-media (BGJb; Gibco) interface on a triangular-shaped wire grid in an organ culture dish. Nicotine treated groups were supplemented with nicotine hemisulfate (Sigma, St Louis, MO, USA) at concentrations of 0.06, 0.6, or 6 mM. α-bungarotoxin (α-BTX) was added at 1 or 10 μM (Sigma). In the groups treated with both nicotine and antagonist, tissues were incubated with α-BTX for 1 h at 37°C before both nicotine and α-BTX were added into the media. The medium barely covered the apical surface of the palates. Tissues were incubated at 37°C in a humidified gas mixture (5% CO2 and 95% air) and media was changed every 24 h. Experiments were repeated 16 times with a total of 120 embryos from 22 pregnant mice.
Cultured palatal shelves collected at 72 h were fixed in freshly prepared 4% paraformaldehyde/phosphate-buffered saline (PBS) (pH = 7.4) for 30 min. After rinsing in PBS, the tissues were processed for paraffin embedding. Serial sections (8 μm thickness) were collected and numbered in sequence from rostral to caudal. Light microscopic analysis of hematoxylin and eosin stained sections (H and E) was completed and a Zeiss Axioplan was used to capture images. Serial sections were also evaluated for the expression of laminin. After blocking with 10% normal goat serum/PBS, the tissues were incubated in 1:30 rabbit anti-laminin (Sigma) for 2 h at room temperature except for negative controls. Following rinsing, the primary antibody was detected by 1:100 fluorescein isothiocyanate conjugated goat-anti-rabbit antiserum (Jackson Immuno-Research, PA, USA).
Samples harvested at 20 h were infiltrated in 50/50 PBS/ornithine carbamoyl transferase embedding medium (Sakura Finetec, CA, USA) for 5 min before being mounted as frozen blocks and stored in −80°C freezer. Serial 8 μm coronal sections were collected and fixed in acetone at −20°C for 20 min. After blocking with 10% goat serum, the samples were incubated overnight at 4°C in 1:100 primary antibodies (phosphorylated Akt-Ser473, Santa Cruz Biotechnology, CA, USA; phosphorylated Smad2-Ser 465/467, Upstate Biotechnology, NY, USA; α-7 nicotinic receptor subunit, Biodesign International, ME, USA) in 3% goat serum except negative controls. Following rinsing, the primary antibody was detected by 1:100 Alexa Fluor® 488 goat-anti-rabbit antiserum (Molecular Probes, OR, USA). After mounting in Slowfade™ media (Molecular Probes, Eugene, OR, USA), the images were obtained using a Leica SP2 confocal microscope. All treatment groups were viewed at the same settings for each antibody. All secondary antibody controls were negative. Images were arranged with Adobe Photoshop and PageMaker programs, with careful attention to equal image enhancement for all treatment groups.
Cryostat sections were collected and fixed as described above. Slides were then incubated with 1:75 Alexa Fluor® 488 α-bungarotoxin (α-BTX; Molecular Probes) for 1.5 h at 37°C. Negative control was incubated in Alexa Fluor® 594 α-BTX in the same conditions. After rinsing and mounting in Slowfade™ media, the images were obtained using a Leica SP2 confocal microscope with an excitation wavelength of 488 μm. All treatment groups were analyzed at the same microscope settings on the same day and arranged with Adobe Photoshop and PageMaker programs.
Immunohistochemical experiments and α-BTX labeling were analyzed with the Leica upright TCS-SPII confocal microscope, equipped with argon and helium neon lasers with excitation wavelengths of 457, 488, 514, 543 and 633 μm. The tissues were analyzed with 10× and 20× air lenses. En face images of the carboxyfluorescein labeled whole mount palates were obtained from oral to nasal epithelia at approximately 3-μm intervals. Optical images from representative depths were analyzed, enhanced and stored. Images were arranged with Adobe Photoshop and PageMaker programs.
Cultured palatal shelves were harvested at 20 h (n = 15 palates/treatment group/experiment). Tissues from different experiments (n = 4) were kept separately. To reduce the contamination by mesenchyme, the middle 1 mm (0.5 mm on either side of the MEE) of the palates were dissected and placed in lysis buffer (10 mM Tris buffer, 150 mM NaCl, 1 mM ethylenediaminetetraacetic acid, 0.1% TX100, and 0.1% NP40) with 1% protease inhibitor, phosphotase inhibitor cocktails I and II (Sigma-Aldrich, St. Louis, MO, USA). The tissues were then sonicated and centrifuged at 20,800 g for 30 min and the resultant supernatant was saved. Duplicate aliquots of each sample were used for Lowry assays to determine protein concentration. Proteins were separated on 7.5% sodium dodecyl sulphate-polyacrylamide gel electrophoresis pre-cast gels (Bio-Rad, Hercules, CA, USA) and transferred onto Immobilon-P polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA). The non-specific binding sites on the membranes were blocked by incubating in blocking buffers: 2% BSA-Tris-buffered saline with 0.1% Tween 20 (TBST) for phosphorylated antibodies for 1 h at room temperature; and 5% milk-TBST for α-7 nicotinic receptor subunit (α-7AchR) antibody overnight at 4°C. This was followed by 12-h incubation in primary antibodies at 4°C (1:100 Akt/2% BSA-TBST; 1:500 Smad2/5% milk-TBST; 1:800 nicotinic receptor/5% milk-TBST). After rinsing in TBST, membranes were incubated in secondary antibodies conjugated with horseradish peroxidase for 1–2 h at room temperature. Proteins were detected using femtoLucent™ sans (Geno Technology Inc., MO, USA) and probes were observed on Kodak Image station 440CF (Eastman Kodak Digital Science™, Rochester, NY, USA). Images were arranged with Adobe Photoshop and with PageMaker programs. Experiments were repeated twice from separate litters and the Western blots were analyzed with densitometry. The mean densitometry value was graphed.
The palates went through different degrees of EMT and fusion after 72 h in vitro. Five stages of fusion (1–5) were scored according to confocal and histomorphological observations (33). We calculated the mean fusion score (MFS) by taking the stage (1–5) times the number of embryos, and divided the sum by the total number of embryos in that group. For example, to calculate the MFS for the controls the following calculation was made: [(1)(3) + (5)(4) + (16)(5)]/22 = 4.67 (Table 1). Therefore groups with more complete palatal fusion had MFS closer to 5. MFS was compared between groups using Kruskal–Wallis test, a non-parametric test (distribution-free) used to compare three or more independent groups of sampled data. p-Value of <0.01 was considered statistically significant.
Palatal tissues were collected from 88 E13.5 mouse embryos using 14 pregnant mice (Table 1). To follow the fate of MEE, a lipophilic marker, CCFSE was used to label the palatal epithelia (43) for confocal analysis. Dissected palatal shelves were placed nasal surface down in pairs with their medial edges in contact in organ culture. Cultures were maintained 20, 48 or 72 h and media were changed daily.
Confocal microscopy was used to examine the overall tissue architecture of cultured palate samples by obtaining complete z-series through the palatal fusion zone. Typical z-series images were obtained from the oral surface to a depth of 1 mm (Levels 1–3 in Fig. 1N) at 3-μm intervals, and representative sections (Fig. 1A–K) demonstrate that CCFSE could be used to label the surface epithelia and medial edge epithelium (MME). As the MEE transformed into mesenchyme the label became weaker (Fig. 1K–M). The nasal surface was not reached due to the thickness of the tissues (Approximately 1.5 mm).
One example of the data obtained from a complete z-series from a control palate cultured for 48 h demonstrated that the oral surface epithelium (Level 1, Fig. 1N) was uniformly labeled with CCFSE (Fig. 1A–D). The palatal shelves were not apposed in the apical sections (Fig. 1A–D), similar to (Level 1, Fig. 1N). The cleft between the two palatal shelves did not contain CCFSE stained tissues and appears as a black area (arrows, Fig. 1B, C). In optical sections toward the middle of the tissue (Level 2, Fig. 1N), only midline epithelia in the fusion zone were intensely labeled (Fig. 1, F–H). These epithelia adhered and thinned into a single layer from anterior to posterior (arrows, Fig. 1G, H). The midline epithelia (Boxed area in Fig. 1I, K) were examined with higher magnification (Fig. 1L, M) to illustrate that increased space between the labeled epithelial cells was observed. These cells had lost cell–cell adhesion, changed cell shape, and were migrating lateral into the pre-existing mesenchyme (arrows, Fig. 1L, M). Confocal analysis was more informative than paraffin section analysis for overall detection of EMT throughout the palatal cultures. However, the cross sectional data obtained from H & E staining and immunohistochemistry (Fig. 2D, G) provided additional important information concerning the tissue structure (Fig. 2D–F) and basement membrane proteins (Fig. 2G–I).
Control palates (n = 21/22) completely fused (16 stage 5, and five stage 4) for a mean fusion score (MFS) of 4.67 by 72 h (Table 1). The labeling was more discretely distributed in the fusion zone (arrows, Fig. 2A) than at 48 h (arrows, Fig. 1L, M), indicating the 72-h cultured palates had achieved mesenchymal confluence and fusion. Palatal shelves cultured with 0.06 mM nicotine progressed through normal EMT and achieved stages 4 or 5 with a MFS of 4.75 after 72 h (Table 1). Whereas, palates treated with 0.6 mM nicotine had intensely CCFSE labeled midline epithelia that were continuous along the fusion zone after 48 h of culture (data not shown). Persistent midline epithelia were also observed after 72 h in the presence of 0.6 mM nicotine (Fig. 2B, mee), although the seam appeared broken at a few places along the midline (arrowheads, Fig. 2B). These palate samples were scored as partially fused (stage 3). We also observed one of nine analyzed tissues achieved stage 4 fusion (Table 1), with only small epithelial islands persisting in the midline. The overall MFS for 0.6 mM nicotine treatment group was 3.11, which was significantly lower (p < .01) than the control and 0.06 mM nicotine treatment groups.
In the palates treated with 6 mM nicotine, confocal observations of CCFSE labeling and histochemical analysis demonstrated persistent epithelia in the midline after 48 or 72 h, and mesenchymal confluence was not achieved (Fig. 2C). Using our scoring protocol, 23 of 26 palates were stage 1. In addition, three samples were scored as stage 2, providing a total MFS of 1.16, significantly (p < .01) different from control palatal fusion scores.
Palate fusion and disappearance of MEE in samples cultured for 72 h were confirmed by hematoxylin and eosin (H and E) stained cross sections. No continuous epithelia remained in the midline region (arrow, Fig. 2D) and mesenchyme formed a confluent layer of control samples. We also observed that the epithelia in the medial region of oral surface (o) disappeared in most of the fused palate samples (Asterisk, Fig. 2D). These epithelia were presumably MEE in origin.
The size and thickness of the palatal shelves of the nicotine treated tissues were similar to the control group (Fig. 2D–F); however, proliferation analysis was not completed in the current investigation. In the palate tissues treated with 0.6 or 6 mM nicotine, two distinct layers of cuboidal shaped MEE persisted in the fusion zone on the nasal and oral sides (arrow, Fig. 2E, F). The midline seam was broken but an epithelial island remained (arrowhead, Fig. 2E) in the tissues treated with 0.6 mM, but not 6 mM nicotine.
As breakdown of the basement membrane is a critical step in EMT, we examined the presence of laminin in the cultured palates. Laminin expression was observed under the oral epithelia (arrowheads, Fig. 2G). However, the fusion zones of the control-cultured samples (arrow, Fig. 2G) were negative for laminin, suggesting that the basement membrane of the MEE was completely degraded. Some areas of laminin staining were observed beneath the oral epithelia in both control and nicotine treated palates (Fig. 2G–I). At high magnification some appeared to be the developing blood vessels, while other areas were not identified. In palates treated with nicotine (0.6 or 6 mM) laminin remained in the fusion zone (mee, arrowhead, Fig. 2H and arrows, 2I) demonstrating that the basal lamina had not degraded in these palates. Laminin was detected along the fusion zone (arrows, Fig. 2I), continuous with that of the oral surface epithelia. This result further demonstrated that the MEE did not thin into one layer and the intact basal lamina persisted in the tissues. Therefore, basement membrane degradation, an essential event during EMT, did not occur in nicotine treated palate samples.
In summary, according to our scoring system, the MFS for the 0.6 and 6 mM nicotine treatment groups are 3.11 and 1.16, respectively (Table 1). These values were significantly lower (p < 0.01) than the control and 0.06 mM nicotine treatment groups (4.67 and 4.75, respectively). Therefore, in the presence of nicotine, palatal fusion was inhibited in a dose dependent manner.
As demonstrated by the dose response study, the MEE in control palates was actively transforming into mesenchyme by 48 h and completely transformed by 72 h. To study the signal transduction activation, an earlier time in the EMT process was needed. It was determined that by after 1 day in culture, the palatal shelves had adhered and started to transform, but a substantial population of the MEE remained in the midline (Fig. 3A). Therefore, the tissues were harvested at 20 h to capture the cells in the transforming MEE phenotype. In addition, as the MEE were the focus of the research, cultured palates were either frozen and sectioned for immunohistochemistry or further harvested for Western blot analysis. To enrich the protein samples for MEE, the midline seam was harvested separately from the whole tissues and pooled by treatment group to obtain enough protein.
Several signaling pathways have been implicated in the EMT process including TGFβ3 (25, 27, 44) and PI-3 kinase (33). PI-3 kinase activates 3-phosphoinostide-dependent protein kinases (PDKs) (45, 46). Integrin-linked kinase (ILK) is a PDK that is involved in EMT, present in the developing palate MEE (47) and phosphorylates Akt on its serine 473 site (48, 49). Smad2 acts as a downstream signaling molecule when TGFβ3 binds to its cell surface receptors. Therefore, the effects of nicotine on PI-3 kinase and TGFβ3 signaling during palate fusion were evaluated by determining the levels of phosphorylated Akt (Ser473) and phosphorylated Smad2 (Ser465/467) by immunohistochemistry (Fig. 3A, B, E, F) and Western blot analysis (Fig. 3C, G).
The midline epithelia were positive for phosphorylated- Akt (Ser473) as observed from a single confocal optical image of a frozen section (Fig. 3A). The fluorescence was discontinuous (arrow, Fig. 3A) and no staining was observed in the fusion zone on the nasal area MEE (asterisk, Fig. 3A), suggesting phosphorylated Akt decreased in the transformed MEE. The palates cultured with 6 mM nicotine had low levels of phosphorylated Akt in the fusion zone (arrow, Fig. 3B) and surface epithelia. Western blot analysis of the middle 1 mm (0.5 mm on either side of the MEE) of the tissues from duplicate experiments confirmed the immunohistochemical results. The phosphorylated-Akt (Ser473) from the control tissues (Fig. 3C, Con) was threefold higher than nicotine treated tissues (Fig. 3C, 6 Nic) in densitometry analysis of duplicate experiments (Fig. 3D).
In contrast, activated Smad2 was not different between controls and nicotine treated palates after 20 h in culture. The MEE stained in a similar pattern and intensity (arrows, Fig. 3E, F) in surface and midline epithelia in both tissues. The discrete fluorescence in the control tissue (asterisk, Fig 3E) suggested down-regulation of phosphorylated Smad2 in transformed cells. In contrast, activated Smad2 was observed throughout the midline epithelia in 6 mM nicotine treated tissues. Protein analysis demonstrated that nicotine treatment (6 Nic, Fig. 3G) did not alter the level of phosphorylated-Smad2 analyzed by densitometry in duplicate experiments (Fig. 3H).
α-Bungarotoxin irreversibly binds many subunits of nAChRs with high affinity and blocks their activation. Fluorescent α-bungarotoxin (α-BTX) was used to determine the localization of nAChRs in palate tissues. Cryostat cross-sections from embryonic mouse palates cultured for 20 h were evaluated. Fluorescent α-BTX was observed in the nasal (n), oral (o), and midline epithelia (mee) (arrows, Fig. 4A) of the E14.5-day in vivo mouse palate during normal fusion. Staining was not continuous along the midline (asterisks, Fig. 4A), suggesting that some MEE transformed into mesenchyme and no longer expressed the midline epithelial protein, nAChRs. The location and intensity of fluorescent α-BTX in the palates cultured under control conditions were similar to E14.5 mouse palates (arrow, Fig. 4B), with more gaps in MEE (asterisks, Fig. 4B). In the 6 mM nicotine treated palates, however, the staining was continuous in the fusion zone and the intensity was greater in the epithelia, especially in MEE, than the control cultured palates (arrow, Fig. 4C). In addition, the mesenchymal cells and other areas were intensely labeled lateral to the MEE (arrowheads, Fig. 4C). Some of these areas also stained with laminin, indicating that they may be epithelial cell populations (data not shown). This pattern of nAChR localization was not observed in control-cultured samples (Fig. 4B).
The α-7 nicotinic receptor subunit is one of the nAChRs expressed in epithelia and is selectively, irreversibly bound by α-BTX (41, 50, 51). Western blot analysis demonstrated that α-7 nAChR subunit increased 3.3-fold in tissues treated with nicotine (6 Nic Fig. 4E) compared with control tissues (Con, Fig. 4E) quantified by densitometry from duplicate experiments (Fig. 4F).
To investigate if nicotine inhibited palate fusion by directly binding to nAChRs, culture media was supplemented with 1 μM or 10 μM α-BTX, with or without 6 mM nicotine. Tissues treated with 1 or 10 μM α-BTX fused after 72 h in culture except for two stage 3 (partial) fusions (Table 1). The tissue morphology resembled those of control-cultured palates (Fig. 1D, G). Only a few discrete fluorescent islands were observed in a single optical image of CCFSE labeled whole mount palate (arrows, Fig. 5A). There were no epithelia remaining in the fusion zone in H and E stained sections (Fig. 5B). The combination of 6 mM nicotine and α-BTX did not rescue the palates from non-fusion, although two of 12 samples were partially fused (Table 1). The CCFSE labeled MEE were continuous anterior-posteriorly (Fig. 5B) and two layers of epithelia persisted in the fusion zone (arrows, Fig. 5D).
The location and intensity of fluorescent α-BTX staining of the samples treated with the combination of 6 mM nicotine and α-BTX (Fig. 4D) were similar to control palates (Fig. 4B). Western blot analysis showed a decrease in α-7 nAChR subunit compared with nicotine treated tissues, although the level was higher than control samples analyzed from duplicate experiments (BTX+6 Nic, Fig. 4E).
Using lipophilic cellular markers (DiI or CCFSE) or retroviral vectors, several research groups traced the fate of MEE (20, 21, 43, 52). These experiments concluded that the majority of MEE, if not all, remain in the tissue but go through cellular changes and obtain mesenchymal characteristics. In our study, CCFSE (43) was used to label the surface epithelia before culture. The CCFSE diffuses through the cell membrane, and intracellular esterase cleaves the acetates to release the fluorophore. The product is packaged into intracellular compartments as a lipid insoluble compound, which can pass through gap junctions into the basal layer but does not get into the underlying mesenchyme. Therefore, it exclusively labels the epithelial cell population. The direct entry of CCFSE into periderm enables this epithelial layer to be more intensely labeled than the basal MEE (53). During in vitro palate fusion, peridermal cells go through apoptosis (18). This would explain why the intensity of CCFSE labeling in our completely fused palatal cultures was low. Other studies have suggested that there may be dye spreading during sectioning of tissues labeled using DiI or CCFSE (54). In these experiments, the potential dye spreading was avoided by analyzing whole-mount tissues with confocal microscopy. The current experiments clearly demonstrated that CCFSE labeled epithelial cells and the mesenchyme formed confluence after 72 h of culture in control conditions.
The amount of phosphorylated Smad2 in the MEE of 20-h nicotine treated palates did not change compared with controls. The immunohistochemistry demonstrated that Smad2 was approximately the same in controls and nicotine treated MEE, but more MEE cells remained in the midline of the nicotine treated palates (Fig. 3).
A previous study in our lab demonstrated that the inhibition of PI-3 kinase activity by LY 294002 resulted in the failure of EMT and palatal fusion (33). The mechanisms suggested were: (1) PI-3 kinase may be involved in initiation of palatal fusion through new desmosome formation; (2) PI-3 kinase may regulate basement membrane degradation by modulating MMP production; (3) PI-3 kinase may be downstream of TGFβ3 during palate fusion. In addition, ILK may be a signaling molecule downstream of PI-3 kinase regulating EMT (47, 48). Activated ILK phosphorylates its downstream effector Akt on the serine 473 site. Therefore, we examined the level of phosphorylated Akt (Ser473) and demonstrated that it was significantly decreased in the 6 mM nicotine treated tissues after 20 h, suggesting nicotine may have inhibited EMT and palatal fusion through interference with the PI-3 kinase pathway.
The receptors in cholinergic transmission are composed of the muscarinic acetylcholine receptors, which are G-coupled receptors; and the nAChRs. The neuronal nAChRs are a family of ACh-gated ion channels consisting of eight α and three β subunits and are also present in non-neuronal cells, including keratinocytes (51, 55). Addition of ACh or nicotine to cultured human keratinocytes modifies their proliferation, differentiation, and motility (40, 41, 50), suggesting the presence of functional nAChRs in keratinocytes. An interesting aspect of nAChR regulation is that chronic exposure to nicotine can result in an increase in the number of nAChRs as observed in the human brains of tobacco smokers (56, 57). The regulation of nAChRs is in contradiction with the general rule that over-exposure to agonists decreases the number of receptors. Depending on the nature of the agonist and the length of exposure, nAChRs go through rapid reversible desensitization, permanent inactivation, and post-transcriptional up regulation to compensate for the non-functional receptors (58–60). This regulation mechanism also explains the biphasic effects of nicotine on keratinocytes. Addition of ACh or nicotine at 0.1 to 10 mM increased keratinocyte adhesion and migration within 30 min in culture (61), but long-term exposure resulted in up regulated α7 nAChR subunit and decreased keratinocyte migration (50). It was suggested that α4, β2 and α7 nAChR subunits are more sensitive to up regulation and desensitization (62, 63). We detected nAChRs in the mouse palate MEE using fluorescent α-BTX. Our results from protein analysis suggested that the α7 nAChR subunit increased in tissues treated with 6 mM nicotine for 20 h. These data lead to the speculation that normal functions of nAChRs are essential during EMT in palate development; while an overdose of nicotine resulted in inactivated although up regulated nAChRs.
Two possible reasons may explain why α-BTX did not rescue palatal non-fusion. First, the concentration of α-BTX may have been too low in relation to the dosage of nicotine. In a recent study (64) 1 μM α-BTX was used to reverse the effects of 10 μM nicotine in keratinocyte cultures. In contrast, in this study 10 μM α-BTX was applied to palates that were treated with 6 mM nicotine. Secondly, nAChRs may be required for EMT and palatal fusion as antagonist treatment resulted in inactivation of the receptors. Although α-BTX treatment alone did not inhibit palatal fusion, addition of either d-tubocurarine or mecamylamine at 1 mM concentration in culture media resulted in partial fusion or non-fusion of the palates with MFS of 3.0 and 3.71, respectively (data not shown). Supporting our hypothesis that nAChRs may be required for EMT and palatal fusion. The observation that α-BTX alone did not inhibit fusion can be explained as (i) α-BTX is a very specific nAChR antagonist and only blocks the activation of homomeric α7, α8, and α9 subunits (51), and (ii) the highest concentration used was 10 μM instead of millimolar concentrations. In addition, mice produced with a substitution of a threonine for a leucine at position 247 in the channel domain have reduced levels of α7 nAChR and die shortly after birth because of neurological problems, however they do not have cleft palates (65).
In neural cell cultures, α7-receptor subunit associates with the PI-3 kinase p85 subunit, stimulates Ca2+ influx, and activates Akt (66). If this is also true in epithelia, it is possible that the nicotine inactivation of nAChRs in the MEE inhibited EMT and fusion through the PI-3 kinase - Akt pathway.
Other in vitro studies also used millimolar nicotine concentrations (50). These nicotine concentrations were significantly higher than the level of nicotine found in the arterial blood from smokers and may not be relevant to an in vivo situation. Although these experiments did not examine the possible function of nicotine metabolites, such as cotinine, they clearly demonstrated that nicotine inhibited palatal fusion in vitro. The decreased active Akt and increased α7 nAChR subunits may suggest that the PI-3 kinase pathway may be down stream of nAChR signaling for EMT during palate fusion.
We thank Jesus Acevedo, Tamara Field, Jan Lalor, Steve Lin and Petra Moessner for their technical assistance. We are also grateful to the Baylor Oral Health Foundation, Tobacco Endowment Fund, and NIH EY08886 for supporting the study.
*This work was supported by, Grant Sponsor: Baylor Oral Health Foundation; NIH (Grant No.: EY-08886); Tobacco Endowment Fund, Texas A & M University System (Grant No.: 304-202850-4013).