Members of the Dapper (Dpr)/Dact protein family are involved in the regulation of distinct signaling pathways, including TGFβ/Nodal, canonical and noncanonical Wnt pathways. Three Dpr genes, Dpr1, Dpr2 and Dpr3, are expressed in mouse embryos and in many adult tissues; however, their in vivo functions have not been reported. In this study, we generated Dpr2-deficient mice using a gene-knockout approach. Homozygous Dpr2 knockout (Dpr2–/–) embryos developed normally and postnatal Dpr2–/– mice grew to adulthood without obvious morphological or behavioral defects. We found that Dpr2 was expressed highly in epidermal keratinocytes and in hair follicles of adult mice, and that Dpr2 deficiency resulted in accelerated re-epithelialization during cutaneous wound healing. Furthermore, we demonstrated that loss of Dpr2 function enhanced the responses of keratinocytes to TGFβ stimulation, and that TGFβ signals promoted adhesion to fibronectin and migration of keratinocytes, by regulating the expression of specific integrin genes. Thus, Dpr2 plays an inhibitory role in the re-epithelialization of adult skin wounds by attenuating TGFβ signaling.
Wound healing is a dynamic and coordinated process that involves inflammation, cell proliferation, cell migration, extracellular matrix deposition and tissue remodeling. The formation of new tissue starts with the migration of keratinocytes in the wound epidermis and hair follicles (Schafer and Werner, 2007). Key players in the process of wound healing have been identified, including a diverse set of growth factors and cytokines, including transforming growth factor-β (TGFβ) (Grose and Werner, 2004; Singer and Clark, 1999; Werner and Grose, 2003).
TGFβ signaling regulates a variety of cellular events and plays an important role in embryonic development (Chen and Meng, 2004; Tian and Meng, 2006; Whitman, 1998). TGFβ ligands transduce signals through heteromeric complexes of type II and type I serine/threonine kinase receptors. Ligand binding leads to the phosphorylation of the type I receptors, which in turn phosphorylate downstream effectors, including Smad2 and/or Smad3. Activated Smad2 and/or Smad3 form complexes with Smad4 and translocate into the nucleus where they regulate the transcription of target genes (Massague and Chen, 2000).
Upregulation of TGFβ1 from platelets in response to wounding suggests that TGFβ1 regulates wound repair (Amendt et al., 2002; Schafer and Werner, 2007). However, experiments to address the role of TGFβ1 during wound healing are inconclusive (Gailit et al., 1994; Grose and Werner, 2004; Werner and Grose, 2003; Yang et al., 2001). Genetic ablation of TGFβ1 or its receptors has been shown to both reduce and enhance the rate of wound repair (Amendt et al., 2002; Brown et al., 1995; Crowe et al., 2000; Shah et al., 1999). Some researchers report that exogenous TGFβ1 inhibits keratinocyte proliferation in vitro and in vivo (Amendt et al., 2002; Hebda, 1988; Sellheyer et al., 1993). However, depending on the repair model and the intensity of TGFβ1 signaling, TGFβ1 could either inhibit (Amendt et al., 2002; Hosokawa et al., 2005) or promote re-epithelialization (Gailit et al., 1994; Reynolds et al., 2005; Reynolds et al., 2008). Therefore, the precise role of TGFβ1 and its fine regulation in re-epithelialization remains unclear.
Xenopus Dapper (Dpr) was first identified as a Dishevelled (Dsh)-interacting protein in a yeast two-hybrid screen (Cheyette et al., 2002). Through its interaction with Dsh, Dpr may help recruit or stabilize GSK3 and axin in the β-catenin degradation complex. Loss-of-function studies demonstrate that maternal Dpr is required for notochord development of Xenopus embryos (Cheyette et al., 2002). Orthologs of Xenopus Dpr and other members of Dpr family have been identified in other vertebrate species (Katoh and Katoh, 2003; Yau et al., 2005; Zhang et al., 2006). In zebrafish, Dpr2 has been shown to suppress mesoderm induction by Nodal signals by promoting the lysosomal degradation of TGFβ receptors ALK4 and ALK5 (Zhang et al., 2004b). Interestingly, zebrafish Dpr2 is also required for convergent extension movements through its role as an enhancer of the Wnt/Ca2+-PCP components stbm/tri and wnt-11/slb, whereas Dpr1 enhances Wnt8/β-catenin activity (Waxman et al., 2004). The mouse genome expresses three Dpr genes, Dpr1, Dpr2 and Dpr3, which have official gene symbols Dact1, Dact2 and Dact3 (Fisher et al., 2006; Su et al., 2007). Like Xenopus Dpr, mammalian Dpr1 inhibits Dishevelled-mediated Wnt signaling; this effect may be mediated by Dpr1 binding to Dsh, followed by lysosomal degradation of Dsh (Yau et al., 2005; Zhang et al., 2006). Unlike Dpr1, mammalian Dpr2 has little effect on canonical Wnt signaling, but it inhibits TGFβ signaling in a manner similar to that of zebrafish Dpr2 (Su et al., 2007). However, the in vivo functions of these mouse Dpr genes have not been reported.
In an attempt to investigate the in vivo function of mouse Dpr2, we generated Dpr2-deficient mice using homologous recombination. Surprisingly, we found that Dpr2–/– homozygous embryos developed normally and that Dpr2–/– mice were alive at birth and underwent normal postnatal development. However, mice lacking Dpr2 showed accelerated re-epithelialization and healing of skin wounds. We further demonstrated that accelerated migration of Dpr2–/– keratinocytes was associated with enhanced TGFβ signaling.
Generation of Dpr2-knockout mice
The mouse Dpr2 gene consists of four exons, with the translation start codon located in the first exon (Fig. 1A). In an attempt to determine Dpr2 function during embryonic development and in adult tissues, we generated Dpr2-deficient mice by deleting a 2.7 kb genomic region containing the first exon and the 5′ portion of the first intron of the Dpr2 gene. The targeting vector contained 7032 bp of the Dpr2 gene, which includes a 5′ region of 3114 bp and a 3′ region of 3918 bp (Fig. 1A). To obtain ES cells with the Dpr2 deletion, the targeting vector was electroporated into mouse TC-1 ES cells. Subsequently, we identified four independent homologously recombinant ES cell clones by PCR and Southern blotting (Fig. 1B). We injected each of the four ES cell clones into host blastocysts and obtained three chimeric mice derived from two of the ES clones, which exhibited germline transmission of the mutated Dpr2 allele. We then crossed the Dpr2+/Neo mice with EIIa-Cre transgenic mice (Lakso et al., 1996; Xu et al., 2002) to delete the Neo cassette, which was present in the targeting vector and was floxed with a pair of loxP sites (Fig. 1A and C). Homozygous Dpr2-deficient mice (Dpr2–/–) were generated by mating heterozygous Dpr2 deletion mice (Dpr2+/–) (Fig. 1D and E). No phenotypic differences were found between the animals derived from the two original ES cell lines.
To confirm disruption of the Dpr2 gene and loss of Dpr2 expression, RT-PCR and northern blot analyses were performed using total RNA from brain and kidney. RT-PCR failed to detect any Dpr2 mRNA in Dpr2–/– mice (Fig. 1F). Northern blotting showed that a 4.4 kb band appeared as expected in wild-type mouse tissues, whereas no band was detected in Dpr2–/– mice (Fig. 1G), which suggests that these mice are indeed Dpr2 deficient.
Dpr2-knockout mice develop normally
Genotyping revealed that out of 458 F2 mice, 134 were Dpr2+/+, 208 were Dpr2+/– and 116 were Dpr2–/–. Homozygous mutant mice accounted for approximately 25% of the F2 mice, which agreed with predicted ratios. Dpr2-deficient mice developed normally without any visible phenotypic changes, and their growth rate was also comparable with that of wild-type littermates. Dpr2–/– adults did not exhibit gross morphological or behavioral abnormalities. Even though Dpr2 was expressed highly in ES cells (see below), outgrowth of Dpr2–/– blastocysts was normal, the inner cell mass expanded into a multilayer mass, the trophectoderm expanded into a trophoblast monolayer, and loss of Dpr2 had no effect on cell proliferation in early mouse embryos (supplementary material Fig. S1). These observations indicate that Dpr2 may be dispensable for normal embryonic development and adult homeostasis.
Dpr2 is expressed in keratinocytes
We showed previously that mouse Dpr2 mRNA transcripts are present in embryos as early as 3 days post coitus (Su et al., 2007). In this study, we examined mouse Dpr2 expression using RT-PCR analysis of embryos at earlier stages, including unfertilized eggs and blastocysts (data not shown). We found that Dpr2 mRNA was present in embryos at all stages tested, with the exception of unfertilized eggs. We also examined Dpr2 expression in adult tissues and organs using northern blot analysis. As shown in Fig. 2A, high levels of Dpr2 expression were detected in the cerebrum, with lower levels of Dpr2 expression in the heart, kidney and testes. Moreover, Dpr2 transcripts were present at high levels in mouse ES cells and at a moderate level in skin (Fig. 2A).
In skin, Dpr2 mRNA was detected primarily in the keratinocytes of epidermis and hair follicles (Fig. 2B). To determine whether Dpr2-positive cells in skin are keratinocytes, we performed combinative experiments with in situ hybridization of Dpr2 and immunofluorescence of keratin 14 (K14), a marker for the basal layer of keratinocytes and the outer root sheath of hair follicles (Gat et al., 1998; Vassar et al., 1989). Expression patterns of Dpr2 and K14 overlapped (Fig. 2B-E), suggesting that Dpr2 may function in keratinocytes and in cutaneous healing.
Accelerated wound closure in Dpr2-deficient mice is associated with enhanced keratinocyte migration
Considering that Dpr2 has high-level expression in the keratinocytes, we attempted to investigate its potential involvement in skin wound healing. We first observed the healing rate of full-thickness cutaneous wounds in adult skin of the Dpr2–/– and wild-type littermates. The time-dependent healing of the skin wounds in both types of mice was shown in Fig. 3A. Three days after wounding, the wound area in Dpr2–/– mice was smaller than in their wild-type littermates, suggesting accelerated wound healing in Dpr2–/– mice. Measurement of the wound area using digital images of five mice per time point for each group showed that the differences between Dpr2–/– mice and wild-type mice were statistically significant on days 3, 5, and 7 after wounding (P<0.05) (Fig. 3B).
Since keratinocyte migration is a key step in wound healing and Dpr2 is primarily expressed in skin keratinocytes, we further tested the difference in the migration rate of keratinocytes in Dpr2–/– and wild-type mice based on K14 immunostaining. Dpr2–/– and wild-type littermates were sacrificed on days 3, 5, 7 and 11 after injury (five mice per time point for each group). Re-epithelialization occurred faster in Dpr2–/– mice than in wild-type mice (Fig. 4A). On the day 7 after wounding, the wounds in Dpr2–/– mice had already lost their eschar and appeared completely epithelialized. By contrast, wounds in the wild-type littermates still carried an eschar and showed partial epithelialization (Fig. 4A). As shown in Fig. 4B, the distance between the migration edge of Dpr2–/– keratinocytes was significantly shorter than that of wild-type controls. Accordingly, the migration distance of Dpr2–/– keratinocytes was longer (Fig. 4C). To determine whether this effect correlated with increased cell proliferation, we counted the number of 5′-bromo-2′-deoxyuridine (BrdU)-positive keratinocytes in day 3 wounds of Dpr2–/– and wild-type mice (Fig. 4D). The percentage of proliferating keratinocytes was comparable in Dpr2–/– and wild-type cells (Fig. 4E). These data suggest that accelerated wound healing in the absence of Dpr2 is due to enhanced re-epithelialization, which is consistent with previous reports (Werner et al., 2007).
Dpr2 deficiency enhances cell response to TGFβ signaling
It has been shown that upregulation of TGFβ1 is involved in re-epithelialization during cutaneous wound healing (Werner and Grose, 2003). We next asked whether TGFβ1 level was increased in Dpr2 mutants. Western blot analysis showed that enhanced TGFβ1 expression level within the wounds of wild-type and Dpr2–/– mice was comparable at days 3 and 5 after injury (supplementary material Fig. S2). Dpr2 has been found to modulate the duration of TGFβ signaling by controlling turnover of endocytosed type I receptors (Su et al., 2007; Zhang et al., 2004b). We hypothesized that the function of Dpr2 in wound healing might be related to its involvement in TGFβ signaling. To test this hypothesis, we examined responses to TGFβ signals and total protein levels of TβRI (ALK5) and the downstream effector Smad2 in wild-type and Dpr2–/– keratinocytes. The results showed that the expression of the (CAGA)12-luciferase reporter was induced by TGFβ1 (5 ng/ml) to a higher extent in Dpr2–/– keratinocytes relative to wild-type keratinocytes, and this inductive effect was more obvious when cells were transfected with a constitutively active form of ALK5 (caALK5) (Fig. 5A). Transfection of exogenous Dpr2 led to the efficient inhibition of TGFβ1-induced reporter expression in both wild-type and Dpr2–/– cells (Fig. 5A). These data suggest that loss of Dpr2 enhances cell responses to TGFβ signaling. Although total levels of ALK5 were comparable in wild-type and Dpr2–/– cells, TGFβ1 stimulation resulted in significantly higher levels of ALK5 in Dpr2–/– cells than in wild-type cells (Fig. 5B). Furthermore, the phosphorylated Smad2 (Smad2-P) levels were increased significantly in Dpr2–/– keratinocytes following TGFβ1 treatment (Fig. 5C). Considering that Dpr2 preferentially promotes the degradation of endocytosed (activated) TGFβ receptors (Zhang et al., 2004b), we believe that loss of Dpr2 in mutant keratinocytes causes slower degradation of activated ALK5, which in turn leads to increased amount of Smad2-P.
Accelerated migration of Dpr2–/– keratinocytes is caused by enhanced TGFβ signaling
To confirm the association between accelerated re-epithelialization in Dpr2–/– keratinocytes and enhanced TGFβ signaling, we analyzed keratinocyte migration in in vitro wound models. Wound-healing assays in cultured cells showed that TGFβ1 treatment led to accelerated cell migration of both wild-type and Dpr2–/– keratinocytes (P<0.01) (Fig. 6A and B). However, a significant difference was found in the migration distance between wild-type and Dpr2–/– keratinocytes (P<0.05). At 72 hours after treatment with TGFβ1 (5 ng/ml), the wounds were healed almost completely in Dpr2–/– cells, whereas the wounds of the wild-type cells remained creviced. Treatment with SB431542 (SB, 10 μM), an inhibitor of ALK5/ALK4 kinase activity (Laping et al., 2002), effectively inhibited TGFβ1-induced migration of both wild-type and Dpr2–/– keratinocytes compared with vehicle-treated (DMSO) controls (Fig. 6A,B; P<0.05 in wild-type cells and P<0.01 in Dpr2–/– cells), although DMSO treatment alone affected keratinocyte migration. These data suggest that the accelerated migration of keratinocytes in Dpr2-deficient mice is related to enhanced ALK5-mediated TGFβ signaling.
We then performed Transwell culture assays to confirm the association of accelerated migration of Dpr2-deficient keratinocytes with upregulation of TGFβ1 signaling. Wild-type and Dpr2–/– keratinocytes were seeded into the upper wells of Transwell chambers with medium containing TGFβ1 (5 ng/ml), and normal medium was added to the lower wells of the chambers. The number of cells migrating into the lower surface of the membrane was counted. As shown in Fig. 6C, the addition of TGFβ1 promoted the migration of both Dpr2–/– and wild-type cells (P<0.01), but the effect was significantly stronger in Dpr2–/– cells (P<0.05). These results again indicate that accelerated migration of Dpr2-deficient keratinocytes is mediated by enhanced TGFβ signaling.
Upregulated expression of integrin β1 and αv was associated with the accelerated re-epithelialization of in Dpr2–/– keratinocytes
Integrins are transmembrane glycoproteins that form αβ heterodimers and act as receptors by binding to extracellular matrix proteins (Steffensen et al., 2001). Integrins have been shown to play fundamental roles in keratinocyte adhesion and migration (Grose et al., 2002; Juhasz et al., 1993; Martin, 1997; Steffensen et al., 2001; Zambruno et al., 1995). To determine whether the difference in migration ability between wild-type and Dpr2–/– keratinocytes is associated with altered expression of integrin genes, we examined integrin mRNA expression using real-time RT-PCR. As shown in Fig. 7, TGFβ1 (5 ng/ml) stimulation led to an apparent increase of integrin β6 (Fig. 7A) and integrin αv (Fig. 7C) expression and slight upregulation of integrin β1 expression (Fig. 7B), whereas integrin α5 expression was unchanged (Fig. 7D). Furthermore, the TGFβ1-induced increase in integrin gene expression was greater in Dpr2–/– cells than in wild-type cells.
To further confirm the involvement of specific integrins in cell migration, cell adhesion assays were carried out on fibronectin-coated substrates in the presence or the absence of specific integrin antibodies. The adhesive strength of both wild-type and Dpr2–/– keratinocytes to fibronectin was enhanced after TGFβ1 treatment with stronger adhesion for Dpr2–/– keratinocytes (Fig. 7E). Importantly, the adhesive strength of both types of cells was partially inhibited by the addition of either integrin β1 (2 μg/ml, Santa Cruz) or integrin αv (2 μg/ml, Santa Cruz) antibodies, whereas the control IgG had no effect (Fig. 7E). These data suggest that surface integrins β1 and αv, are associated with the adhesive ability of keratinocytes to fibronectin. Thus, loss of Dpr2 function leads to increased expression of specific integrins stimulated by TGFβ signals in Dpr2–/– keratinocytes in a provisional wound matrix, and ultimately results in accelerated migration of these cells. Our finding is consistent with the view that upregulation of specific integrins can promote adhesion and migration of keratinocytes on the fibronectin-rich wound matrix (Juhasz et al., 1993; Martin, 1997; Steffensen et al., 2001; Weinacker et al., 1994; Zambruno et al., 1995).
Zebrafish Dpr2 modulates Nodal/TGFβ signals by promoting lysosomal degradation of their receptors (Zhang et al., 2004b). Mouse Dpr2 has also been shown to antagonize TGFβ signaling in vitro and its overexpression in zebrafish embryos leads to reduced mesodermal gene expression (Su et al., 2007). In this study, we generated Dpr2-deficient mice by using homologous recombination. We found that Dpr2 was highly expressed in keratinocytes of epidermis in adult skin, and that Dpr2 deficiency caused acceleration of keratinocyte migration in skin wounds. The accelerated migration of Dpr2–/– keratinocytes was associated with increased adhesive strength due to enhanced specific integrin expression by elevated TGFβ signaling.
Here, we have uncovered a crucial role for Dpr2 in the regulation of cutaneous wound healing through modulation of the cellular response to TGFβ1 signaling. TGFβ1 is a key regulator of wound healing that affects diverse processes during wound repair (Werner and Grose, 2003). However, the precise role for TGFβ1 in keratinocyte migration remains largely unknown (Schafer and Werner, 2007). TGFβ1 has been shown to upregulate integrin gene expression and to enhance keratinocyte migration (Gailit et al., 1994; Hebda, 1988; Zambruno et al., 1995). However, the application of TGFβ1 to wounds can either accelerate or inhibit re-epithelialization depending on the amount of TGFβ1 treatment or the wounding model used (Garlick and Taichman, 1994). Moreover, the targeted disruption of TGFβ1, its receptors or Smad proteins leads to either reduced or accelerated wound healing (Ashcroft et al., 1999; Crowe et al., 2000; Larjava et al., 1993; Shah et al., 1999). In this study, we showed that Dpr2 is expressed in epidermal and hair follicle keratinocytes, and that targeted disruption of Dpr2 led to accelerated wound healing, largely because of accelerated re-epithelialization in Dpr2–/– mice. Our data demonstrated that accelerated re-epithelialization in Dpr2-deficient mice was associated with enhanced keratinocyte migration, but not with keratinocyte proliferation. The Dpr2-deficient keratinocytes migrated faster than wild-type keratinocytes in the presence or absence of TGFβ1. Dpr2-deficient keratinocytes expressed higher levels of ALK5 and Smad2-P, indicating that the cellular responses to TGFβ1 were elevated in the absence of Dpr2. Reporter assays confirmed that exogenous Dpr2 expression inhibited ALK5-induced cellular responses to TGFβ1 in keratinocytes, and attenuation of ALK5 activity by SB431542 also led to reduced TGFβ1–mediated keratinocyte migration. Collectively, these data suggest that elevated ALK5 and activated Smad2 levels are likely to be the mechanisms underlying the accelerated re-epithelialization in skin wounds of Dpr2-deficient mice. These results are in agreement with previous studies that Dpr2 is a negative regulator of TGFβ signaling (Su et al., 2007; Zhang et al., 2004b).
During cutaneous wound healing, integrins play crucial roles in cell adhesion, spreading, and migration (Hynes, 1992), and TGFβ1 induces the expression of integrin genes in epidermal keratinocytes (Gailit et al., 1994; Zambruno et al., 1995). Keratinocytes upregulate the expression of fibronectin-binding integrins, including α5β1, αvβ5 and αvβ6, which stimulate the migration of cells on the fibronectin-rich provisional wound matrix (Juhasz et al., 1993; Martin, 1997; Steffensen et al., 2001; Weinacker et al., 1994; Zambruno et al., 1995). Integrin-β1-null keratinocytes showed impaired re-epithelialization, suggesting an essential role for integrin β1 in keratinocyte migration (Grose et al., 2002). Integrin αvβ6 gene expression is increased significantly in wound keratinocytes, and has been reported to promote keratinocyte migration (Huang et al., 1998; Thomas et al., 2001). Consistently with previous studies (Gailit et al., 1994; Stepp et al., 2007; Zambruno et al., 1995), the results of our study show that TGFβ1 upregulated the gene expression of integrins αv, β1 and β6 in both wild-type and Dpr2-deficient keratinocytes. However, integrin expression levels were significantly increased in Dpr2–/– cells compared with wild-type cells in response to TGFβ1 treatment. Furthermore, cell adhesion assays showed that Dpr2–/– keratinocytes were more adherent to fibronectin than wild-type cells, and surface integrin β1 and αv were associated with the adhesive ability of the keratinocytes. We propose that enhanced cellular responses to TGFβ1 may lead to the upregulation of integrin β1 and integrin αv gene expression in Dpr2–/– keratinocytes, resulting in accelerated re-epithelialization in Dpr2–/– mice.
Previous studies on the role of TGFβ1 during re-epithelialization show contradictory results (Grose and Werner, 2004; Werner and Grose, 2003). Our conclusion is consistent with some previous reports (Crowe et al., 2000; Gailit et al., 1994; Hebda, 1988), but is inconsistent with others (Hosokawa et al., 2005; Yang et al., 2001). This inconsistency might be due to the use of different animal models. In K14-Smad2 (Hosokawa et al., 2005) and K14-TGFβ1 (Yang et al., 2001) transgenic mice or in Sdc1-null mice (Stepp et al., 2002; Stepp et al., 2007), skin keratinocytes during the wound-healing process might have TGFβ1-signaling levels above a threshold at which cell migration is inhibited, probably because of alterations in the extracellular matrix.
Although Dpr2 is expressed in ES cells and throughout embryonic development when TGFβ signals play a crucial role (James et al., 2005; Massague and Chen, 2000; Millan et al., 1991; Paria et al., 1992; Shen, 2007), Dpr2–/– mice were viable at birth, grew to a normal size and were fertile. Furthermore, Dpr2-deficicent adults behaved normally. Although high levels of Dpr2 expression were detected in the keratinocytes of hair follicles (Fig. 2B), we failed to find any difference in hair follicle morphology (supplementary material Fig. S3A,B), hair growth rate (data not shown), and zigzag (supplementary material Fig. S3C,D) and awl (supplementary material Fig. S3E,F) hair structures between wild-type and Dpr2–/– mice. Lack of apparent abnormalities in Dpr2–/– embryos and adults may be ascribed to the existence of multiple antagonists of TGFβ signaling pathway, such as Smurf1, FKBP12, Smad7 and STRAP (Massague and Chen, 2000; Yamaguchi et al., 2006). In Dpr2–/– mice, other TGFβ antagonists may play a compensatory role in embryonic development and adult tissue homeostasis. The biological significance of Dpr2 may be revealed by crossing Dpr2-deficient mice with other mutant mice or by challenging the Dpr2–/– mice with different environmental factors.
Materials and Methods
A 3.1 kb sequence containing the 5′ sequence of Dpr2 gene was amplified by PCR from a 129/Sv mouse genomic library using two primers (5′-CCGCTCGAGCATCTTGAAGATAGAG-3′ and 5′-CCGCTCGAGTGATAAGTTAGGAAGA-3′), which contained a NotI or XhoI restriction site, respectively. The amplified fragment was inserted into the NotI-XhoI sites of the vector pLoxP (Yang et al., 1998) to generate plasmid pLoxP-Dpr2-1. A 3.9-kb 3′ homologous arm was obtained by PCR with two primers (5′-AATGAAATCCCATGCCCTCT-3′ and 5′-CCGGAACATGCCCGTCAG-3′), and inserted into the EcoRI site of the recombinant plasmid pLoxP-Dpr2-1 to generate the targeting vector pDpr2neo.
Electroporation of ES cells and generation of germline chimeras
pDpr2neo was digested with NotI and electroporated into TC-1 embryonic stem (ES) cells (Deng et al., 1994). About 15 μg genomic DNA extracted from ES cells or animal tails was digested with KpnI, resolved on a 0.8% agarose gel and transferred to nitrocellulose membrane for Southern blot with 32P-labeled probes. The probes included a 1.1-kb 3′ flanking genomic fragment, which was prepared by digesting pBKS-Dpr2 (full-length cDNA) with PstI, and a 0.75 kb Neo sequence, which was amplified using primers (5′-GCTTGGGTGGAGAGGCTATTCGGCT-3′ and 5′-GAACTCGTCAAGAAGGCGATAGAAGGCGAT-3′) from pLoxP. The 3′ flanking probe detected a 15 kb fragment from the wild-type allele and a 6.2 kb fragment from the targeted allele. The Neo probe recognized an 8 kb fragment from the mutant allele, but no fragment was detected from the wild-type allele. Targeted ES cells were injected into C57BL/6J blastocysts to generate chimeras, which were bred with C57BL/6J female mice to generate Dpr2 heterozygous mice.
Mice were genotyped by Southern blot as described above or by PCR. For PCR analysis, primer 1 (5′-GGCTGGAGTTCCTTGGAGACC-3′) and primer 2 (5′-TCTGGATCCCAGAAGCAAGCC-3′) from Dpr2 genomic sequence amplified a 714 bp fragment from the wild-type allele. The combinatory use of primer 1 and primer 3 from Neo (5′-CCAGACTGCCTTGGGAAAAGC-3′) amplified a 610 bp fragment from the mutant allele. Moreover, amplification with primer 1 (5′-CCCGAGGAATCACTGCAAGCT-3′) and primer 4 (5′-GTGGGGTCAAGGGTCAGAGGG-3′) generated a 283 bp fragment for the mutant allele without the Neo cassette.
In situ hybridization and immunofluorescence
Mice were anesthetized with 1% sodium pentobarbital by abdominal cavity injection. They were fixed from the left ventricle with 4% paraformaldehyde (PFA) for 10-15 minutes and then flushed with normal saline. The skin was removed and fixed overnight in cold 4% PFA. After transferring into 30% sucrose, the skin was sectioned using a cryostat microtome. The sections were used for in situ hybridization following the routine protocol (Zhang et al., 2004a) using antisense and sense probes. For the combination of in situ hybridization and immunofluorescence, anti-K14 (1:1000, BAbCo) and anti-digoxigenin antibodies were added simultaneously and incubated for 2 hours at 37°C. After development of color for in situ hybridization, slides were incubated for 1 hour in the secondary antibody conjugated with TRITC-conjugated anti-rabbit IgG (1:100, Zhongshan, Beijing), and then stained with 4′,6′-diamidino-2-phenylindole (DAPI) to visualize the nucleus. Slides were observed using a Nikon E600 microscope (Tokyo, Japan). These experiments were repeated at least three times to obtain reliable results.
RT-PCR and northern blot analysis
For RT-PCR, total RNA was extracted from diverse tissues of adult mice and whole embryos at various stages using TRIZOL reagent (Invitrogen), and used for reverse transcription with SuperScript III reverse transcriptase (Invitrogen). The primers used to amplify a fragment of Dpr2 coding sequence were 5′-TCACGGCTAAGGAGACAGGA-3′ and 5′-TAGACGGTCGCTGCAAACAG-3′. The glyceraldehyde phosphate dehydrogenase gene (GAPDH), which was used as the internal control, was amplified using primers 5′-CATCACTGCCACCCAGAAGA-3′ and 5′-GCTGTAGCCAAATTCGTTGT-3′.
For northern blotting, 15 μg total RNA extracted from mouse ES cells or adult tissues was electrophoresed on a 1% agarose gel in the presence of 0.6 M formaldehyde, and probed with 32P-labeled full-length Dpr2 using standard procedures (Fisher et al., 2006).
In vivo wound-healing experiments
Littermates at 6-8 weeks of the same sex were used. Four 4-mm full-thickness cutaneous biopsy punch wounds were made on either side of the mouse. The entry wounds were photographed on day 0, 1, 3, 5, 7, 9 and 11. The wound areas were determined using NIH ImageJ software. The wounded tissue was collected on various days after injury. Tissue was either fixed in 4% PFA for paraffin embedding or snap-frozen in OCT (Thermo Lifesciences).
Histological analysis and immunohistochemistry
Mouse tissues were fixed in 4% PFA at 4°C overnight, embedded in paraffin, sectioned at 6 μm, and stained with hematoxylin and eosin (H&E). For BrdU labeling, mice were injected intraperitoneally with 100 μg/g body weight of BrdU (Sigma) 2 hours before sacrifice. For immunohistochemistry and immunofluorescence, the slides were incubated with K14 antibody (1:1000, BAbCo) and BrdU antibody (1:300, Invitrogen), followed by incubation with FITC-conjugated anti-rabbit IgG and TRITC-conjugated anti-mouse IgG, or biotin-conjugated goat anti-rabbit IgG (all 1:100, Zhongshan, Beijing). Signals were detected by DAB (Zhongshan, Beijing) staining or directly by fluorescence microscopy. Slides were observed under a Nikon E600 microscope (Tokyo, Japan) with a digital camera.
Preparation and in vitro culture of keratinocytes
Primary mouse keratinocytes were isolated from skin of newborn Dpr2–/– or control mice, as described (Stepp et al., 2007). Full-thickness skin was treated with 0.25% dispase overnight at 4°C. The epidermis was separated from the dermis and dispersed by stirring into single cells, followed by suspending in Keratinocyte-SFM medium with supplements (Invitrogen). Cells were incubated in dishes coated with coating solution (10 μg/ml fibronectin, 1% v/v Vitrogen 100 Collagen, 100 μg/ml BSA, 10 mM HEPES) at 34°C in 5% CO2 for 12 hours to allow cells to attach to the bottom. Attached cells were further cultured in fresh medium, which was replaced once every 2 days.
Reporter gene assay and western blot
For all reporter gene assays, keratinocytes were plated in 24-well plates. When cells were grown to 70-80% confluency, the luciferase reporter plasmid, (CAGA)12-luciferase (0.5 μg), was cotransfected using lipofectamine2000 (Invitrogen) with Renilla construct (10 ng) and other plasmids into keratinocytes. One day after transfection, cells were treated with or without TGFβ1 (5 ng/ml) in Keratinocyte-SFM medium for 20 hours. 48 hours after transfection, cells were harvested and the luciferase activity of cell lysates was determined using a luciferase assay system (Promega). Each experiment was performed in triplicate, and the data were averaged and expressed as the mean ± s.d. of three independent experiments after normalization to Renilla activity.
Western blot analysis was performed as described (Teng et al., 2006). Keratinocytes were treated with or without TGFβ1 (5 ng/ml) for 48 hours. Cell lysates were separated by SDS-PAGE, and proteins of interest analyzed by immunoblotting and ECL. Antibodies used included ALK5 and TGFβ1 (1:1000, Santa Cruz), Smad2-P and total Smad2/3 (1:1000, Cell Signaling) and β-actin (1:4000, Sigma).
In vitro wound-healing assay
The 80-90% confluent keratinocytes in a 12-well plate were treated with 10 μg/ml mitomycin C (Sigma) for 2 hours in order to remove the influence of cell proliferation. The cells were then wounded by manually scraping the cells with a 1-ml-pipette tip. The cells were treated with or without TGFβ1 (5 ng/ml). Cell migration into the wound surface was monitored by microscopy at various time points. Quantification of migration rates was done by measuring the distance of the wound edge of the migrating cells from the start point from three independent experiments. Ten points of wound edges throughout a vision field of the microscope were measured, and the mean value was calculated as one experiment.
The assay was performed in Transwell plates with polycarbonate membrane filters containing 8-μm pores (Millipore). The membrane was coated with coating solution for 30 minutes before use. Growth medium was added to the lower well of the chamber and aliquots of 5×104 keratinocytes in 300 μl medium with or without TGFβ1 (5 ng/ml) were seeded into the upper well. After 24 hours, cells in the upper surface of the membrane were removed with a wet cotton swab. The cells on the lower surface of the membrane were fixed with ice-cold methanol and then stained with crystal violet. Each experiment was performed in triplicate, and counting was done in five randomly selected microscopic fields (magnification, ×100) within each well.
Real-time PCR was performed using Total RNA from cultured keratinocytes, and repeated at least four times (four different cell preparations) for each gene with Roche LightCycler 2.0 system using a SYBR Green assay. Expression values were normalized to HPRT expression. The primer sequences were as follows: integrin β6: 5′-TTTGTTTGAAAGAGGAGGAA-3′, and 5′-TTCTGGGTGGTACATACGG-3′; integrin β1: 5′-CCAGGGACTGACAGAAGAC-3′ and 5′-CAGCAGGCTAAACAAAGAA-3′; integrin αv: 5′-CCTCCACTGTCAGAAAGAA-3′ and 5′-GGAATTAGCCTAAGATACG-3′; integrin α5: 5′-CAGGACAAATCCCAAACAA-3′ and 5′-GCAGAACCCAGGTCATACA-3′, HPRT: 5′-ATGCCGAGGATTTGGAAAAAGTGTTT-3′ and 5′-TGTCCCCCGTTGACTGATCATTACAG-3′.
For the cell adhesion assay, wild-type and Dpr2–/– keratinocytes were treated with TGFβ1 (5 ng/ml) for 48 hours and were then released by incubation in trypsin for 8 minutes. The cells were washed with PBS and resuspended in growth medium. In order to perform cell adhesion assay, a 24-well plate was previously coated with fibronectin (50 μg/ml) at 37°C for 1 hour, leaving some wells uncoated as a negative control. The wells were washed with washing buffer (0.1% BSA in PBS) for 2 times, and then blocked with blocking buffer (0.5% BSA in PBS) in CO2 incubator for 1 hour. Followed by washing with washing buffer, 1×105 cells of each type were added to individual wells and the plate was incubated at 34°C for 2 hours. The plate was shaken and washed with washing buffer 2-3 times, followed by fixation in 4% PFA at room temperature for 15 minutes. The cells were stained with Crystal Violet for 20 minutes and wells were washed with water and dried. After addition of 2% SDS, the plate was incubated at room temperature for 30 minutes. The optical density of the retained cells was measured using a Multiscan Machine (Thermo) at 550 nm.
Statistical analysis of the data was performed using a two-tailed Student's t-test using Microsoft Excel (Microsoft, Redwood, CA). *P<0.05, **P<0.01.
We thank Chuxia Deng, Philip Leder and Fen Zhou for TC1 ES cells and pLoxP vector, Ye-Guang Chen for (CAGA)12-luciferase reporter. This work was supported by National Basic Research Program of China (2005CB522500, 2006BAI23B01-3 and 2006CB943401), National Science Foundation of China (90208002, 30221003 and 30430350), National High-Tech Research and Development Program (2006AA02Z168) and grant Z0006303041231.
Supplementary material available online at http://jcs.biologists.org/cgi/content/full/121/17/2904/DC1
- Accepted June 12, 2008.
- © The Company of Biologists Limited 2008