An Integrated Gene Regulatory Network Controls Stem Cell Proliferation in Teeth

Epithelial stem cells reside in specific niches that regulate their self-renewal and differentiation, and are responsible for the continuous regeneration of tissues such as hair, skin, and gut. Although the regenerative potential of mammalian teeth is limited, mouse incisors grow continuously throughout life and contain stem cells at their proximal ends in the cervical loops. In the labial cervical loop, the epithelial stem cells proliferate and migrate along the labial surface, differentiating into enamel-forming ameloblasts. In contrast, the lingual cervical loop contains fewer proliferating stem cells, and the lingual incisor surface lacks ameloblasts and enamel. Here we have used a combination of mouse mutant analyses, organ culture experiments, and expression studies to identify the key signaling molecules that regulate stem cell proliferation in the rodent incisor stem cell niche, and to elucidate their role in the generation of the intrinsic asymmetry of the incisors. We show that epithelial stem cell proliferation in the cervical loops is controlled by an integrated gene regulatory network consisting of Activin, bone morphogenetic protein (BMP), fibroblast growth factor (FGF), and Follistatin within the incisor stem cell niche. Mesenchymal FGF3 stimulates epithelial stem cell proliferation, and BMP4 represses Fgf3 expression. In turn, Activin, which is strongly expressed in labial mesenchyme, inhibits the repressive effect of BMP4 and restricts Fgf3 expression to labial dental mesenchyme, resulting in increased stem cell proliferation and a large, labial stem cell niche. Follistatin limits the number of lingual stem cells, further contributing to the characteristic asymmetry of mouse incisors, and on the basis of our findings, we suggest a model in which Follistatin antagonizes the activity of Activin. These results show how the spatially restricted and balanced effects of specific components of a signaling network can regulate stem cell proliferation in the niche and account for asymmetric organogenesis. Subtle variations in this or related regulatory networks may explain the different regenerative capacities of various organs and animal species.


Introduction
Stem cells reside in specific niches that regulate their selfrenewal and differentiation. In general, these niches consist of various types of neighboring differentiated cells, which provide a milieu of extracellular matrix and signaling molecules that regulate the unique behavior of stem cells. Epithelial stem cells have been identified in several adult tissues that undergo continuous turnover, such as hair, skin, feather, and gut [1][2][3][4][5][6]. Although mammals have lost the capacity for recurrent tooth renewal, some mammals, such as rodents, have incisor teeth that grow continuously. On the basis of cell cycle kinetics, DiI tracing, and the location of cells that retain bromodeoxyuridine (BrdU) label long term, it is thought that the epithelial stem cells of rodent incisors reside in the stellate reticulum core of the cervical loops ( Figure 1A and 1B; [1,7]). An interesting feature of the mouse incisor is that the cervical loop on the labial side is much thicker than that on the lingual side. It contains abundant stellate reticulum cells in its core, whereas the lingual cervical loop is very thin and only contains a few stellate reticulum cells. However, the molecular mechanisms underlying the asymmetric growth and size of these cervical loops are unknown.
Most dental epithelial stem cells give rise to enamel-secreting ameloblasts. The enamel layer covers the dentin layer produced by mesenchymal odontoblasts. Besides the asymmetric growth and morphology of the cervical loops, enamel also exhibits asymmetric distribution in the mouse incisors. Enamel-producing ameloblasts differentiate only along the labial aspect of the mouse incisor, so enamel covers only the labial surface of the tooth, whereas the lingual surface is enamel-free and covered only by dentin ( Figure  1A). This, together with the continuous growth and wear of the mouse incisor, maintains its characteristic sharpness, which is crucial to its function for gnawing. We showed previously that ameloblast differentiation on the labial side of the incisor is induced by bone morphogenetic protein 4 (BMP4) from mesenchymal odontoblasts, and that Follistatin inhibits BMP function in lingual-side dental epithelium, preventing enamel formation there [8]. Consistent with this, ameloblasts differentiate on both sides of incisors in Follistatin À/À mice, whereas overexpression of Follistatin in the dental epithelium (Keratin 14 [K14]-Follistatin mice) inhibits ameloblast differentiation and enamel formation. Hence, the asymmetric expression of Follistatin in the dental epithelium contributes to the asymmetry of enamel formation in the mouse incisors [8]. However, although the basis for the selective differentiation of ameloblasts on the labial incisor surface is partly understood, the mechanisms that underlie the regulation of the stem cells that give rise to ameloblast progenitors is unknown. The dental mesenchyme has important functions in influencing the epithelial stem cell niche. Fibroblast growth factor (FGF) signals from the mesenchyme have been shown to regulate Notch signaling in the epithelium [1], and FGF10 has been identified as a necessary signal for epithelial stem cell maintenance, based on the hypoplastic morphology of the cervical loop in Fgf10 À/À mice [9]. Although FGF10 and presumably Notch1 are important in stem cell maintenance, they are expressed in similar patterns on both the labial and lingual sides of the mouse incisor [1,10]. So far, the earliest gene reported to be expressed asymmetrically in the developing mouse incisor is Fgf3. From embryonic day 16 (E16) onwards, Fgf3 expression is restricted to a small area of dental papilla mesenchyme on the labial side directly underneath the thick labial cervical loop [9].
The present study started from the observation that incisor growth was markedly slowed in K14-Follistatin mice, and that the cervical loops were severely hypoplastic, whereas in Follistatin À/À mutants, the lingual cervical loop was enlarged. This suggested the hypothesis that Follistatin might in some way inhibit the maintenance or proliferation of dental epithelial stem cells. We show that a complex, highly integrated, and spatially regulated signaling network underlies the effect. Fgf3 expression in the mesenchyme underlies the cervical loops and correlates with their size, and FGF3 cooperates with FGF10 as a mesenchymal signal that stimulates the proliferation of dental epithelial stem cells and transit amplifying (TA) cells. BMP4 represses Fgf3 expression, whereas Activin, which is preferentially expressed in labial mesenchyme, inhibits the repressive effect of BMP4 and restricts Fgf3 expression to the labial dental mesenchyme, resulting in the large, labial-side stem cell niche. Moreover, unlike its function in ameloblast differentiation, Follistatin may antagonize the function of Activin, rather than BMP in the cervical loop area, and thereby limit the proliferation of lingual stem cells. This results in a small, lingual stem cell niche and, collectively, accounts for the asymmetry of the incisor. Thus, the number of dental epithelial stem cells and their spatial disposition are regulated by a complex signaling network involving Activin, BMP, FGF, and Follistatin that mediates the communication between mesenchymal and epithelial cells in the stem cell niche. These results show how the spatial regulation of signals can account for asymmetric organogenesis

Results/Discussion
We observed that incisor growth was markedly slowed in K14-Follistatin transgenic mice that overexpress the trans-

Author Summary
Stem cells reside in specific niches that regulate their self-renewal and differentiation, and are responsible for the continuous regeneration of tissues. Although the regenerative potential of mammalian teeth is limited, mouse incisors grow continuously throughout life and contain stem cells at their proximal ends in the so-called cervical loops. We have used a combination of mouse mutant analyses, organ culture experiments, and gene expression studies to identify the key signaling molecules that regulate epithelial stem cell proliferation in the cervical loop stem cell niche. We show that signals from the adjacent mesenchymal tissue regulate epithelial stem cells and form a complex regulatory network with epithelial signals. Stem cell proliferation is stimulated by fibroblast growth factor 3 (FGF3), and bone morphogenetic protein 4 (BMP4) represses Fgf3 expression. In turn, Activin inhibits the repressive effect of BMP4 and Follistatin antagonizes the activity of Activin. We also show that spatial differences in the levels of Activin and Follistatin expression contribute to the characteristic asymmetry of rodent incisors, which are covered by enamel only on their labial (front) side. We suggest that subtle variations in this or related regulatory networks may explain the different regenerative capacities and asymmetric development of various organs and animal species.
forming growth factor b (TGF-b) antagonist Follistatin in the dental epithelium (Figures 2A, 2B, and S1). Histological analysis showed that in contrast to wild type, the labial cervical loop of K14-Follistatin newborns was severely hypoplastic and resembled the lingual cervical loop in size ( Figure  2C and 2D). Conversely, in Follistatin À/À embryos, which die at birth [8,11], the lingual cervical loops frequently showed marked overgrowth and resembled the labial cervical loop in size ( Figure 2E). These observations suggested that Follistatin influences epithelial stem cell proliferation in the incisor tooth germ.
To test this hypothesis, we performed BrdU incorporation experiments to assay cell proliferation. In wild-type mice, proliferative activity was most obvious in the labial inner dental epithelium, representing actively dividing TA cells that undergo further division and gradually become pre-ameloblasts and ameloblasts. BrdU-labeled cells were less apparent in the central core of the cervical loop where putative slowcycling stem cells are located, and were very sparse in lingual cervical loop epithelium ( Figure 2F). In contrast, in K14-Follistatin mice, the number of stellate reticulum cells, as well as proliferating epithelial cells, were markedly decreased in the labial cervical loop ( Figures 2G and S1). Hence, Follistatin overexpression strongly inhibits dental epithelial cell proliferation and results in a compromised stem cell niche. Conversely, Follistatin À/À incisor germs contained numerous BrdU-positive cells in both labial and lingual cervical loops ( Figures 2F, 2H, and S1). This switch from lingual to labial cervical loop morphology in Follistatin À/À embryos demonstrates that Follistatin functions in the generation of incisor asymmetry through inhibition of epithelial stem cell and TA cell proliferation.
The maintenance and proliferation of stem cells in the cervical loop epithelium depends upon FGFs that are expressed in the subjacent dental mesenchyme [1,9]. Fgf3 expression partly overlaps that of Fgf10, but has a more restricted pattern in that it underlies only labial-side TA cells. In vitro bead implantation assays showed that, similar to FGF10 [1], FGF3 stimulated epithelial cell proliferation in E15 incisors ( Figure S1). In K14-Follistatin incisors, Fgf3 expression was completely down-regulated ( Figure 2I and 2J). Moreover, in Follistatin À/À mice, Fgf3 was ectopically expressed in lingual mesenchyme adjacent to the TA cells of the enlarged cervical loop ( Figure 2K). In contrast, Fgf10 expression appeared unchanged from wild type (unpublished data). These results indicate that Follistatin prevents the expression of Fgf3 in the lingual dental mesenchyme, whereas forced expression of Follistatin leads to down-regulation of Fgf3 in the labial epithelium.
(M-O) At P1, folding of Fgf3 À/À molar epithelium is aberrant and shallower than normal; this phenotype is more severe in Fgf3 À/À ; Fgf10 þ/À molars. (P and Q) Histological sections of E13 molar region in wild-type and Fgf3 À/À ; Fgf10 À/À double mutants show that the development of both maxillary and mandibular molars is arrested prior to bud stage in the mutant. One out of 90 double-mutant embryos survived until E13. Scale bar represents 100 lm (A-C and M-Q). doi:10.1371/journal.pbio.0050159.g003 labial cervical loops were smaller than wild type ( Figure 3C). Moreover, at 5 wk, both Fgf3 À/À and Fgf3 À/À ; Fgf10 þ/À lower incisors were white and lacked characteristic yellow-brown pigment, and the compound mutant incisors were thin and frequently broken ( Figure 3D-3F). Ground sections of Fgf3 À/À incisors showed that an enamel layer was present ( Figure 3G and 3H), but a defect in enamel structure was indicated by the premature wear of molar teeth ( Figure 3K). Ground sections of Fgf3 À/À ; Fgf10 þ/À incisors showed that the enamel layer was either very thin or missing ( Figure 3I). In addition, the molars of Fgf3 À/À and Fgf3 À/À ; Fgf10 þ/À mice were smaller than wild-type teeth, and at P1, the folding of Fgf3 À/À molar epithelium was aberrant, the cusps were shallow, and this phenotype became more severe in Fgf3 À/À ; Fgf10 þ/À molars ( Figure 3K, 3L, 3N, and 3O). These data indicate that FGF3 and FGF10 function cooperatively during tooth development. This was further supported by the observation that in an Fgf3 À/À ; Fgf10 À/À double null mutant embryo, molar development was arrested prior to the bud stage ( Figure 3Q), whereas in the single knockouts, the molars, albeit smaller than wild type, underwent complete development ( Figure 3K; [9]).
It appears that although FGF3 and FGF10 are not required for the early differentiation of ameloblasts during embryonic crown morphogenesis [9], they regulate the maintenance and proliferation of epithelial stem cells in the cervical loops later in the erupted incisors. Thus, during incisor development, Fgf3 and Fgf10 genetically interact to maintain the epithelial stem cell pool that provides a continuous supply of ameloblast progenitors. Moreover, although FGF3 is not necessary for early cervical loop morphogenesis, the asymmetric expression of FGF3 appears to be sufficient to generate the asymmetric labial and lingual stem cell niches in mouse incisors. Several lines of evidence support this view. First, FGF3 protein induces intense cell proliferation in incisor epithelium ( Figure S1). Second, Fgf3 is expressed asymmetrically in incisor dental mesenchyme, in a spatial manner that correlates with differences in cervical loop sizes. Last, ectopic Fgf3 expression was detected in the lingual dental mesenchyme underlying the enlarged cervical loop in Follistatin À/À incisors. These data suggest a model whereby the lingual cervical loop receives only an FGF10 signal, which maintains the basal epithelial stem cell proliferation that would be required for the continuous incisor growth. In contrast, the labial aspect receives both FGF3 and FGF10 signals, resulting in a larger stem cell niche with an increased number of proliferating stem cells. Because the cervical loop defects in K14-Follistatin mice are more severe than those in Fgf3 À/À mice, other molecules besides FGF3 may also be affected by Follistatin overexpression.
To explore the mechanism by which Follistatin, an extracellular antagonist of Activin and BMP [12], inhibited Fgf3 expression, we first compared the expression of Fgf3, Activin bA, Bmp2, Bmp4, Bmp7, and Follistatin during incisor development. In E14 dental mesenchyme, striking co-expression of Fgf3, Activin bA, and Bmp4 was observed; Follistatin was also expressed in the mesenchyme, but at higher levels in adjacent dental epithelium ( Figure 4A). At E15, mesenchymal Fgf3 and Bmp4 expression persisted whereas Activin bA became restricted to the labial mesenchyme directly underlying the cervical loop epithelium; much weaker expression was observed around the lingual cervical loop (Figure 4A). At E16, Fgf3 expression also became asymmetric, being undetectable lingually and overlapping with Activin bA labially; Bmp4 was intensely expressed around both lingual and labial cervical loops, as well as in mesenchymal preodontoblasts on both sides. Bmp2 and Bmp7 were expressed symmetrically in dental papilla mesenchyme at all the stages examined ( Figure  S2). Follistatin was expressed widely in E15 dental epithelium. However, at E16, Follistatin was down-regulated in the labial inner dental epithelium, but persisted in the outer dental epithelium. Hence, there was an obvious gap between these Follistatin-expressing cells and the dental papilla mesenchyme in the cervical loop area. On the lingual side, the thin layers of dental epithelium continued to express Follistatin intensely ( Figure 4A). Activin receptor-like kinase 3 (Alk3), the BMP receptor type 1A, and Alk4, the Activin receptor type 1B, were expressed intensely in the incisor epithelium ( Figure  4B).
To determine the inter-relationships between the different components of this gene regulatory network, we used organ cultures of dissected E16 incisors. Because we had shown previously that Follistatin antagonizes the inductive effect of BMP in differentiating ameloblasts [8], we first inserted beads soaked in recombinant BMP4 into the cervical loop region of incisor explants and examined the expression of Fgf3 and Fgf10. Unexpectedly, after 1 d in culture, Fgf3 expression was markedly down-regulated by BMP4 in the dental mesenchyme, whereas BMP4 had no effect on Fgf10 expression ( Figure 5A-5C). We next inserted beads soaked in the BMP inhibitor Noggin into the incisor cervical loop region. Ectopic Fgf3 expression was strongly induced in the lingual dental mesenchyme directly underlying the dental epithelium ( Figure 5D, arrow), confirming that BMPs inhibit Fgf3 expression in incisor dental mesenchyme.
Consistent with this conclusion, K14-Noggin transgenic mice that overexpress Noggin in dental epithelium exhibit markedly overgrown incisors with increased epithelial proliferation in both the labial and lingual cervical loops [13]. Similar to Follistatin À/À mice and Noggin bead experiments, we detected ectopic Fgf3 expression in the mesenchyme underlying the lingual cervical loops in K14-Noggin incisors ( Figure  S3). Collectively, these results demonstrate that BMPs repress Fgf3 expression in dental mesenchyme in vivo, thereby negatively regulating epithelial stem cell proliferation in the mouse incisor.
The observation that Follistatin and BMPs have similar effects on the cervical loop, i.e., they both inhibit Fgf3 expression and growth of the epithelium, indicates that Follistatin is unlikely to act as a BMP antagonist in the cervical loop. We therefore chose to investigate the activity of Activin A, another target of Follistatin, in this system. Remarkably, similar to Noggin beads, Activin A beads also induced ectopic Fgf3 expression in lingual incisor mesenchyme directly underlying the dental epithelium ( Figure 5E, arrow). Thus, Activin A and BMP4 have opposite effects on Fgf3 expression. When Activin A and BMP4 beads were placed together in incisor explants, the inhibition of Fgf3 by BMP4 was abrogated ( Figure 5F). Next, we studied the effect of Activin on the proliferation of the cervical loop epithelium in cultured explants. When present for 1 or 2 d in incisor explant culture medium, Activin A did not significantly affect cervical loop growth (unpublished data). However, after 4 d, it had induced marked overgrowth in both labial and lingual cervical loops, with extra budding and increased cell proliferation (Figures 5G, 5H, 5J, and 5K). Consistent with the bead experiments, ectopic Fgf3 was also induced in the lingual dental mesenchyme directly underneath the enlarged cervical loop (Figure 5M and 5N). Because the Activin null mutant mice lack incisors and die shortly after birth [14], we used the selective inhibitor of ALK receptors (SB431542) to prevent endogenous Activin/TGF-b signaling in the cultured incisor explants [15]. The addition of SB431542 in culture medium had no detectable affect after 1 or 2 d of culture (unpublished data), but after 4 d, the labial cervical loop was thinner than normal, with a reduced number of stellate reticulum cells, and epithelial cell proliferation was decreased in both labial and lingual cervical loops ( Figure 5I and 5L). Also Fgf3 expression was down-regulated in the dental mesenchyme ( Figure 5M and 5O). These findings supported the role of Activin as a positive regulator of epithelial stem cell proliferation.
Several lines of evidence support the idea that the effects of Activin and BMP on Fgf3 expression are indirect and that they are likely mediated via the epithelium. First, Activin beads did not induce Fgf3 expression in dental mesenchyme when the dental epithelium was removed, indicating either an indirect effect or a co-requirement for additional factor(s) ( Figure 6G). Second, the effects of Activin and the Activin inhibitor on epithelial proliferation was delayed, being detectable after 4 d in culture ( Figure 5G, 5H, 5J, and 5K), but not after 1 d ( Figure S1, and unpublished data). Third, the ectopic Fgf3 expression in the Activin bead experiments was localized directly underneath the dental epithelium, not around the Noggin or Activin beads ( Figure 5D and 5E). Fourth, the BMP and Activin receptors, Alk3 and Alk4, respectively, were intensely expressed in the dental epithelium. These results suggest that the regulation of Fgf3 expression in the dental mesenchyme by Activin and BMP is indirect and that dental epithelium is an indispensable part In situ hybridization for the indicated genes in incisor tooth germs from E14 to E18. (A) Note the symmetric mesenchymal expression of Bmp4 in the cervical loop regions and the onset of asymmetric mesenchymal expression of Activin bA and Fgf3 at E15 and E16, respectively (arrows). Follistatin is expressed throughout the thin layer of lingual dental epithelium as well as in the labial outer dental epithelium, but is down-regulated in the inner dental epithelium labially (arrow). (B) The BMP and Activin receptors Alk3 and Alk4, respectively, are expressed in the epithelium on both labial and lingual sides between E16 and E18. Scale bar represents 200 lm. doi:10.1371/journal.pbio.0050159.g004 of this regulatory network. Thus, Activin acts as a positive regulator, whereas BMP4 is a negative regulator of Fgf3 expression, and hence of epithelial stem cell proliferation in mouse incisors, and the balance between these two mesenchymally expressed TGF-b superfamily members dictates the rate of epithelial stem cell proliferation. The finding that the preferential expression of Activin in labial mesenchyme precedes that of Fgf3 by a day ( Figure 4A) further supports the conclusion that Activin acts upstream of Fgf3.
Lastly, we addressed the regulation of Fgf3 and Activin expression in incisor mesenchyme. Beads soaked in Activin A did not induce Fgf3 expression and neither did FGF3 induce Activin in dental mesenchyme ( Figure 6G and 6H). Both dental epithelium and beads soaked in FGF9 efficiently induced Fgf3 and Activin mesenchymal expression in explants ( Figure 6C-6F). However, Fgf9 expression in cervical loop epithelium becomes labially restricted later than Activin and Fgf3 do in the adjacent labial mesenchyme (Figures 4A and S3; [16]). Therefore, FGF9 alone cannot account for the preferential expression of either gene in labial mesenchyme. Activin is so far the earliest gene showing asymmetric expression in incisor dental mesenchyme. Collectively, these studies integrate FGF, BMP, Activin, and Follistatin into a signaling network whose balance regulates the proliferation of epithelial stem cells and TA cells in the continuously growing rodent incisor ( Figure 7A and 7B). In our model, mesenchymally expressed FGF3 and FGF10 are key effectors of epithelial cell proliferation. BMP4, which is symmetrically expressed in both labial and lingual dental mesenchyme, actively represses Fgf3 expression, and therefore acts as an endogenous brake on epithelial stem cell and TA cell proliferation. As discussed above, this effect of BMP4 apparently is indirect and involves as yet unidentified signaling events from the epithelium to the mesenchyme regulating Fgf3 expression. Fgf10 expression was not affected by BMP. Thus, modulation of FGF-induced epithelial stem cell proliferation by BMP4 and Activin is presumably restricted to the FGF3-dependent component. We hypothesize that the labial cervical loop receives both FGF3 and FGF10 signals and thus becomes hyperplastic relative to the lingual cervical loop, which receives only FGF10. FGF10 may help maintain the basal level of epithelial stem cell proliferation needed to support continuous growth of the tooth and dentinogenesis on the lingual surface [17].
A key feature of the model is that beginning at E15, Activin is expressed asymmetrically. Activin is weakly expressed in lingual dental mesenchyme but strongly expressed in labial dental mesenchyme. On the basis of our findings, we suggest that Activin abrogates the repressive effect of BMP4 on labial Fgf3 expression and thereby allows FGF3 to promote epithelial stem cell and TA cell proliferation in the labial cervical loop. This would indicate that the balance between two TGF-b superfamily members controls Fgf3 expression and thereby modulates epithelial stem cell proliferation. The continued intense expression of Follistatin in lingual dental epithelium would be expected to antagonize residual Activin in lingual mesenchyme, preserving the repressive effect of BMP on lingual Fgf3; whereas on the labial side, there is a gap between Follistatin-expressing outer dental epithelium and the subjacent dental papilla mesenchyme. Therefore, Follistatin may not be able to antagonize the activity of Activin, allowing intense expression of Fgf3 labially. Thus, this regulatory network model explains the increased number of stem cells in the labial compared to the lingual cervical loop, and the asymmetry of the epithelial stem cell niche in the mouse incisor. Although the modulation of proliferation and differentiation in other epithelial stem cell niches has been associated with variations in growth and morphogenesis [1][2][3][4][5][6], this study is the first to demonstrate how fine tuning of a signaling network within a stem cell niche may directly account for asymmetric organogenesis.
The composition of this gene regulatory network has important implications for organ regeneration. The observation that Activin, Bmp2, Bmp4, Bmp7, Fgf3, and Fgf10 expression is confined to the mesenchyme underlines the importance of the dental papilla mesenchyme in regulation of the epithelial stem cell niche. Significantly, mesenchyme also has a key inductive role in maintaining the regenerative potential of other organs. For example, in hair regeneration, several of these signals were also among the dermal papilla mesenchyme signature genes identified in a recent microarray analysis [18]. The delineation of this network also has evolutionary implications. Although continuous tooth development is often regarded as intrinsic to the rodent incisor, other rodents such as the vole (Microtus sp.) possess continuously growing molars with comparable stem cell niches [10]. Moreover, the Madagascar lemur Aye-aye, Daubentonia madagascariensis, a primate, also possesses continuously growing incisors [19]. These observations suggest that the dental epithelial stem cell niche is both robust enough to be retained throughout evolution, yet flexible enough to account for differences in the sizes, shapes, growth rates, and regenerative capacities of teeth in different animals. The finetuning of stem cell proliferation by precise modulation of signaling networks may be a general mechanism that accounts for the different regenerative capacities of other organs as well.
Cell proliferation assays. E18.5 pregnant mice were injected intraperitoneally with 1.5 ml/100 g body weight of BrdU solution (Zymed Laboratories/Invitrogen, http://www.invitrogen.com/ antibodies) and killed after 2 h. In in vitro experiments, 10 lM BrdU was added to culture medium for the last 2 h of culture. Embryonic heads and cultured tissues were fixed in 4% PFA, embedded in paraffin, and immunodetected using a BrdU detection kit (Zymed Laboratories/Invitrogen). The cell proliferation indices were determined by counting the BrdU-positive and -negative cells in the cervical loop epithelium and mesenchyme in defined areas. The cells of the labial-and lingual-side mesenchyme were determined from a 200 lm 3 400 lm-wide region parallel to epithelium. The initial secretory-stage odontoblasts were chosen for the border of counting cells in the labial and lingual dental epithelium, respectively. The mean values were calculated from the two independent observers' data pools. When the numbers of proliferating cells in K14-Follistatin and Follistatin À/À were compared to the wild type (Student t-test), all the differences were highly significant.    Figure S3. Ectopic Fgf3 Expression in K14-Noggin Incisor Mesenchyme and Expression of Fgf9 in Cervical Loop Epithelium (A) In K14-Noggin mouse incisors at the newborn stage, ectopic Fgf3 is detected in lingual dental mesenchyme directly underneath lingual dental epithelium. (B) At E17, Fgf9 is expressed in both labial and lingual dental epithelium (arrows). (C) At P2, Fgf9 expression is confined only to labial-side dental epithelium (arrow; [14]). Scale bars represent 200 lm. Found at doi:10.1371/journal.pbio.0050159.sg003 (504 KB PDF).