| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 is an essential molecule downstream of Foxn1, the gene responsible for the nude mutation, in normal hair development
* Laboratory of Genome and Biosignal, Tokyo University of Pharmacy and Life Sciences, Hachioji, Tokyo, Japan;
Department of Biological Resources, Integrated Center for Sciences, Ehime University, Shitsukawa, Toon City, Ehime, Japan;
Department of Biological Sciences, Graduate School of Science, Tokyo Metropolitan University, Hachioji, Tokyo, Japan; and
Department of Lipid Biochemistry, Graduate School of Medicine, Kobe University, Chuou-ku, Kobe City, Japan
1Correspondence: Laboratory of Genome and Biosignal, Tokyo University of Pharmacy and Life Sciences, 1432-1 Horinouchi, Hachioji, 192-0392 Tokyo, Japan. E-mail: kfukami{at}ls.toyaku.ac.jp
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1, a key molecule in the phosphoinositide signaling pathway, and nude mice show similar hair abnormalities, such as lack of cuticle and bending. We also found that expression of hair keratins was remarkably decreased in skin of PLC-1 knockout mice. Furthermore, expression of PLC-1 was induced in Foxn1-transfected U2OS cells. In addition, we showed that PLC-1 expression was remarkably decreased in skin of nude mice. In skin and keratinocytes of nude mice as well as PLC-1 KO mice, activation of PLC downstream effectors, such as PKC and nuclear factor of activated T cells, was impaired. These results indicate that PLC-1 is an essential molecule downstream of Foxn1 in normal hair formation, and strongly suggest that hairlessness in nude mice is caused by insufficient expression of PLC-1.—Nakamura, Y., Ichinohe, M., Hirata, M., Matsuura, H., Fujiwara, T., Igarashi, T., Nakahara, M., Yamaguchi, H., Yasugi, S., Takenawa, T., Fukami, K. Phospholipase C-1 is an essential molecule downstream of Foxn1, the gene responsible for the nude mutation, in normal hair development.
Key Words: phosphoinositide • nude mice • hair follicle • hair keratin
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Although the immunological functions of Foxn1 have been studied intensively, there have been relatively few studies of its functions in skin. Foxn1 regulates expression of hair keratins, which are essential for normal hair structure (2)
. However, how Foxn1 regulates hair keratin expression and hair formation is largely unknown.
Phospholipase C (PLC) hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP2) to generate two second messengers, diacylglycerol (DG) and inositol 1,4,5-trisphosphate (IP3). DG mediates the activation of protein kinase C (PKC) (3)
, and IP3 releases calcium from intracellular stores (4)
. PLC can be categorized into six types, β, 
, , 
, 
, and 
, on the basis of sequence homology and activation mechanism (5
, 6)
. The -type PLCs, 1, 3, and 4, are thought to be the primary forms expressed in mammals. Among these, PLC-1 is expressed abundantly in most tissues (7)
. We previously generated PLC-1 knockout (KO) mice and found that PLC-1 KO mice show marked hair loss (8)
. However, the detailed molecular mechanism that underlies this hair loss in PLC-1 KO mice is unclear.
In the present study, we found that PLC-1 is a molecule downstream of Foxn1 and is essential for Foxn1-induced hair keratin expression and normal hair shaft formation. Furthermore, we found that the nude mouse is a phenocopy of the PLC-1 KO mouse, suggesting that skin defects observed in nude mice are caused by insufficient expression of the PLC-1 gene.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 KO mice were generated as described previously (8)
. Antibodies against PLC-1 (BD Pharmingen, San Diego, CA, USA), Foxn1 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), β-actin (Chemicon International, Temecula, CA, USA), PKC
(BD Pharmingen), PKC (BD Pharmingen), and phospho-PKC (pan) (Cell Signaling Technology, Inc., Danvers, MA, USA) were purchased. Mouse cDNAs encoding full-length PLC-1 and Foxn1 were each inserted into pcDNA 3.1 (Invitrogen, Carlsbad, CA, USA). A mouse cDNA encoding a nude (nu) Foxn1 was generated from total RNA of skin from nude mice by reverse transcription (RT)-polymerase chain reaction (PCR) and inserted into the pcDNA 3.1. Various sized promoter sequences of the PLC-1 gene were obtained from genomic DNA of ICR mice by PCR and inserted into pGL3 Basic (Promega, Madison, WI, USA). Reporter plasmid containing three mutated Foxn1 binding sites was generated by replacement of three 5'-ACGC-3' sequences located at –1891 to –1888, –366 to –363, and +93 to +96 of the PLC-1 gene by 5'-ACAC-3'. pNFAT-Luc plasmid was purchased from Stratagene (La Jolla, CA, USA).
Histochemistry and in situ hybridization
Histochemistry was performed as described previously (7)
. For in situ hybridization analysis, skin samples were fixed in 4% paraformaldehyde, and 8-µm-thick cryosections were prepared. In situ hybridization with digoxigenin-labeled riboprobes for mouse PLC-1 (accession: AF133125; nucleotide: 94–2364), mouse Foxn1 (accession: X81593; nucleotide: 1041–1781), and mHa3 (accession: X75650; nucleotide: 1–1288) were performed as described (9)
.
RT-PCR
Total RNAs were isolated from skin samples and U2OS cells with an RNeasy Mini Kit (Qiagen, Valencia, CA, USA). Template cDNAs were synthesized from total RNA with SuperScript III (Invitrogen). Five-fold serial dilutions of templates were prepared, and PCR amplification was performed with primer pairs for human PLC-1 (5'-GGCACCTGAGCAGAATATACCC-3' and 5'-TGGGTTCCTCCAGCCTAGTC-3'), mouse PLC-1 (5'-CCTGACACCACCTTCAACTC-3' and 5'-TGCAGTGTTGTTCCTAACGG-3'), mouse Foxn1 (5'-GCTTCCCAGCAGAGGCTTGGTGC-3' and 5'-CTGGATGCATTGGGTGCAGAGGCTG-3'), mHa3 (5'-TGAGATCAACACGTATAGGG-3' and 5'-CTTCACAGCTCAGATCTCTC-3'), mHa5 (5'-GAGGTTGGTTTGGAGAAGGCAT-3' and 5'-CATACCAGTCTTCAGCATCCCT-3'), human β-actin (5'-TGTGATGGTGGGAATGGGTCAG-3' and 5'-TTTGATGTCACGCACGATTTCC-3'), mouse β-actin (5'-TTGTTACCAACTGGGACGACATGG-3' and 5'-GATCTTGATCTTCATGGTGCTAGG-3'), human PLC-β1 (5'-CAGAGTGTCTTAACAGAAGTGGAAG-3' and 5'-GACCGGATCATCTCTGTCTTCTCC-3'), human PLC-β2 (5'-GCAAGGGCTCTCGCAAGAAGAGGAG-3' and 5'-CCTCGTACTCTGCCAGCGCCTCCTTC-3'), human PLC-β3 (CCTGGCTCAGGCACAGGCTGAGGGC-3' and 5'-GCCGACGGATGGAGTTGACTGACTC-3'), human PLC-β4 (5-CTTTGGACTGCTTCATGGCAAGTC-3' and 5'-CCATGCAGAAGTTACACTGCACGCAAG-3'), human PLC-1 (5'-GAAGAGAAGATTGGCACAGAACGTG-3' and 5'-GTCCACCACAAACTCTGTCTTCTGC-3'), human PLC-2 (5'-GTGGAGACGAAGGCTGACAGCATC-3' and 5'-GGAGGAGTAAAGTTCCTCTTCGCTC-3'), human PLC-3 (5'-GTCAGGCACAATGCCCGCCAGCTG-3' and 5'-GATGAAGAGCGTGGCTGGTGACAGTG-3'), human PLC-4 (5'-CAGCACAATACTTGGCAGTTAAGCC-3' and 5'-CCTTGTTGCATGCAGGTCCAAGGCAG-3' and 5'-CAGCACAATACTTGGCAGTTAAGCC-3'), human PLC- (5'-GTTCCGCGTTCACTTCGAAGATCTTG-3' and 5'-TTTGCATGATCTCCTCTTCTGGACC-3'), human PLC- (5'-CCAGTTGCCTCTTACTCATTCATC-3' and 5'-GCTCAAGGCTCTCACCCATTCTGG-3') human PLC-1 (5'-GATGTAGGAGCCGGTGGAGTGGCG-3' and 5'-CCTGTACCCAAGCAGTTGGCACATC-3'), and human PLC-2 (5'-TCAGAGGCGGAGCAGGACCAGG-3' and 5'-CGTGGCTGCCAAGTCCAAGAGCC-3').
Northern blot analysis
Total RNAs were isolated from skin samples of mice with an RNeasy Mini Kit (Qiagen). Each RNA sample was the pool of RNA from three mice. Northern blot analysis was carried out by using digoxigenin-dUTP-labeled probes for gapdh (accession: M32599; nucleotide: 125–920), mHa1 (accession: M27734; nucleotide: 1331–1551), mHa2 (accession: X75649; nucleotide: 1259–1475), mHa3 (accession: X75650; nucleotide: 1007–1204), mHa4 (accession: AA589585; nucleotide: 222–422), mHb5 (accession: AI892769; nucleotide: 213–415), and mHb6 (accession: X99143; nucleotide: 1461–1660) with a DIG Northern Starter kit (Roche Diagnostics, Mannheim, Germany), according to the manufacturer’s instructions. Relative expression levels of hair keratins were determined with Image J software.
Scanning electron microscopy
Skins were fixed with 3% glutaraldehyde in 0.1 M phosphate buffer (pH 7.3). Specimens were postfixed with 1% osmium tetroxide in water for 30 min at 4°C and stained en bloc with 1% tannic acid in water for 30 min at room temperature and 1% osmium tetroxide in water for 30 min at 4°C. Samples were then dehydrated through a graded series of ethanols and immersed in t-butyl alcohol, freeze-dried, coated with platinum, and examined with a Hitachi S-800 scanning electron microscope.
Cell culture
U2OS and HaCaT cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum at 37°C under 5% CO2. U2OS cells were transfected with various plasmids with FuGENE 6 transfection reagent (Roche) according to the manufacturer’s instructions. A suspension culture of HaCaT cells with polyhydroxyethylmethacrylate-coated dishes was performed as described previously (10)
. Primary murine keratinocytes were isolated and cultured as described previously (8)
.
Reporter gene assays
Reporter gene assays were performed as described previously (11)
. Briefly, the reporter plasmids, pRL-CMV (Wako Pure Chemicals, Osaka, Japan) expressing sea pansy luciferase under the control of the cytomegalovirus promoter as an internal control, and the indicated expression plasmids were transfected into U2OS cells. At 30 h after the transfection, the luciferase activities were determined in extracts of the transfected cells using a PicaGene Dual-Luciferase Reporter assay system (Toyo Ink, Tokyo, Japan). The luciferase activities were normalized to the sea pansy luciferase activity from the pRL-CMV vector. The results are presented as means and SD of the relative luciferase activity from at least three independent experiments. A luciferase assay with the pNFAT-Luc plasmid was performed as described (8)
with a slight modification.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 KO mice display phenotypes strikingly similar to those of nude mice at the histological level1 KO mice show marked hair loss. However, the detailed molecular mechanism that underlies this hair loss in PLC-1 KO mice is unclear. In the present study, we noticed that the appearance of a nude mouse is very similar to that of a PLC-1 KO mouse (Fig. 1
B, C). Both PLC-1 KO and nude mice lack fur development, whereas control mice have well-developed coats (Fig. 1A-C
). We then investigated histological structures of skin from control, PLC-1 KO, and nude mice (Fig. 1D-I
). Hematoxylin and eosin (HE) staining of skin sections from PLC-1 KO and nude mice revealed that the hair shafts of both mice are bent and fail to penetrate the epidermis at postnatal day (P) 8 (Fig. 1E, F
). At P28, hair follicles of PLC-1 KO mice are more severely disrupted than those of nude mice (Fig. 1H, I
). Furthermore, scanning electron microscopy confirmed the presence of many hair shafts with normal structure in control mice (Fig. 1J, M
), whereas few hair shafts emerged from the skin surface of PLC-1 KO and nude mice. We also observed that these few emerging hair shafts in PLC-1 KO and nude mice were flattened, lacked cuticles, and were bent (Fig. 1K, L, N, O
). These findings revealed that skin phenotypes of PLC-1 KO and nude mice were remarkably similar.
|
PLC-1 KO mice have molecular defects similar to those of nude mice
The striking resemblance of skin phenotypes between PLC-1 KO mice and nude mice suggested to us that PLC-1 KO mice might have molecular defects similar or identical to those in nude mice. In nude mice, the gene encoding the transcription factor Foxn1 is spontaneously mutated. This mutation is a single base pair deletion in exon 3 of the gene. This deletion causes a frameshift, resulting in a premature stop codon and truncated protein that lacks DNA-binding and transcriptional activities (12)
. Thus, this mutation also affects downstream target genes of Foxn1. In the skin of nude mice, one of these target genes is mHa3, an acidic hair keratin that is essential for normal hair shaft structure (2)
. In nude mice, the Foxn1 mutation leads to insufficient mHa3 expression and abnormal hair shaft structures. Given the striking similarity between nude and PLC-1 KO mice, we hypothesized that expression of mHa3 might also be decreased in skin of PLC-1 KO mice. To test this possibility, we examined mHa3 mRNA levels in skin of control and PLC-1 KO mice at P28. Northern blot analysis revealed that mHa3 was expressed abundantly in skin of control mice, whereas mHa3 expression was decreased remarkably in skin of PLC-1 KO and nude mice (Fig. 2
A). Because many hair follicles of PLC-1 KO mice change into abnormal cysts at P28 (arrow in Fig. 1H
), we cannot exclude the possibility that reduced expression of mHa3 in PLC-1 KO mice is a secondary effect of destruction of hair follicles and formation of abnormal cysts. To address this possibility, we next examined expression of mHa3 in skin from control and PLC-1 KO mice at P0, when hair follicles of PLC-1 KO mice show no apparent abnormalities. RT-PCR analysis showed that expression of mHa3 was also reduced in skin of neonatal PLC-1 KO mice (Fig. 2B
), strongly suggesting that impairment of mHa3 expression in PLC-1 KO mice is caused directly by lack of the PLC-1 gene. In addition to mHa3, the expression of some hair keratins is reported to be decreased in skin of nude mice (13)
. Therefore, we examined the expression of hair keratins in skin of PLC-1 KO mice at P8. Expression of acidic and basic hair keratins, including mHa1, mHa2, mHa3, mHa4, mHb5, and mHb6, was decreased in skin of PLC-1 KO mice to an extent similar to that of nude mice (Fig. 2C
). In contrast with other hair keratins, mHa5 is expressed normally in skin of nude mice (13)
. RT-PCR analysis revealed that mHa5 expression was not affected in skin of PLC-1 KO mice (Fig. 2D
). These results indicate that PLC-1 KO mice possess a skin phenotype similar to that of nude mice at both the histological and molecular levels.
|
PLC-1 KO mice show no macroscopic abnormalities of the thymus and whiskers
Nude mice are known for immunodeficiency caused by athymia (14)
. Therefore, we next examined whether PLC-1 KO mice also have the athymic phenotype. We found that control and PLC-1 KO mice have apparently normal thymuses at P5 (Supplemental Fig. S1A, B), whereas a well-developed thymus was not observed in nude mice (Supplemental Fig. S1C). It is possible that there are functional defects in thymus, such as abnormal thymocyte development, in PLC-1 KO mice; therefore, we examined fluorescence-activated cell sorting (FACS) profiles of control and PLC-1 KO thymocytes (Supplemental Fig. S1D, E). FACS analysis with anti-CD4 and anti-CD8 antibodies revealed that the distribution of thymocytes was normal in PLC-1 KO mice (Supplemental Fig. S1E). This result indicates that the transitions from double-negative to double-positive and from double-positive to single-positive thymocytes occurred normally in thymuses of PLC-1 KO mice. In addition, the whiskers of control and PLC-1 KO mice are long and straight (Supplemental Fig. S1F, G), whereas those of nude mice are curly and short (Supplemental Fig. S1H). These results indicate that there are some tissue-dependent differences in phenotype between PLC-1 KO and nude mice.
Temporal and spatial expression of PLC-1 and Foxn1 are similar
In nude mice, the gene encoding the transcription factor Foxn1 is spontaneously mutated and loses its transcriptional activities (12)
, leading to insufficient hair keratin expression and abnormal hair shaft structures. Given that PLC-1 KO mice show a skin phenotype similar to that of nude mice, it is likely that both PLC-1 and Foxn1 play roles in the same signaling pathway. Therefore, we first compared the time course of expression of these two genes during differentiation of a human keratinocyte cell line, HaCaT. Western blot analysis revealed that expression of both PLC-1 and Foxn1 peaked at 12 h after suspension-induced differentiation and that expression of both genes decreased within 24 h after the induction of differentiation (Fig. 3
A). This result suggests that there are some correlations between expression of PLC-1 and Foxn1 in keratinocyte cultures. We next examined localization of PLC-1 and Foxn1 mRNAs in hair follicles at P28. In situ hybridization analysis revealed that both PLC-1 and Foxn1 were expressed in the same regions of hair follicles, where mHa3 was expressed (Fig. 3B-D
). These observations suggest that expression of PLC-1, Foxn1, and mHa3 may be correlated in hair follicles.
|
We then examined how expression of PLC-1, Foxn1, and mHa3 changes during the hair cycle. In situ hybridization analysis revealed that all three genes were expressed in the same regions of hair follicles at P8, the growth phase for hair shafts and follicles (Fig. 3E, H, K
). Expression of these genes decreased rapidly at P20, which is the resting phase of hair growth (Fig. 3F, I, L
). Subsequently, PLC-1, Foxn1, and mHa3 mRNAs were detected again at P28, a growth phase (Fig. 3G, J, M
), indicating that PLC-1, Foxn1, and mHa3 were expressed simultaneously during the stage of hair shaft generation. This tight correlation between expression of PLC-1, Foxn1, and mHa3 strongly suggests that expression of these genes is regulated in the same signaling pathway.
Expression of PLC-1 is induced by Foxn1 in U2OS cells
Given the strong correlation between the patterns of expression of PLC-1 and Foxn1, it is likely that Foxn1 regulates the expression of PLC-1. To assess this possibility, we examined the effect of Foxn1 on the expression of PLC-1. PLC-1 was upregulated during keratinocyte differentiation (Fig. 3A
). In addition, Foxn1 induces the early stages of keratinocyte differentiation (15)
. Therefore, it is difficult to rule out the possibility that a change in PLC-1 expression by Foxn1 is due to keratinocyte differentiation induced by Foxn1. Thus, keratinocytes are inappropriate for these experiments. We therefore used U2OS osteosarcoma cells, which do not undergo differentiation on Foxn1 expression. RT-PCR analysis revealed that the amount of PLC-1 mRNA was increased remarkably by exogenously expressed Foxn1 but not by the nude (nu) mutant of Foxn1 (Fig. 4A
). This result indicated that Foxn1 induces PLC-1 expression at the mRNA level. In addition, we examined whether the expression of other PLC isozymes is affected by exogenously expressed Foxn1. Expression of PLC-β1, -β2, -1, -2, -3, -4, -, and -1 was detected in U2OS cells transfected with empty vector, and expression was not increased when Foxn1 was exogenously expressed (Fig. 4B
). We did not detect expression of PLC-β3, -β4, -2, or - in U2OS cells transfected with empty vector or vector encoding Foxn1 (data not shown). Thus, among all known PLC isozymes, only PLC-1 expression is induced by Foxn1, indicating that Foxn1 specifically induces PLC-1 expression. We then analyzed the sequence of the PLC-1 promoter crucial for Foxn1-induced PLC-1 expression by generating mutants of the PLC-1 promoter driving the luciferase reporter. It has been reported that the DNA-binding domain of Foxn1 binds to an 11-bp consensus sequence containing the invariant tetranucleotide 5'-ACGC in vitro (16)
. Therefore, we analyzed the sequence of the promoter and 5'-untranslated regions of the PLC-1 gene and found three potential Foxn1-binding sites. We generated reporter plasmids containing the luciferase gene under the control of the PLC-1 promoter with wild-type and mutated Foxn1-binding sites. A luciferase assay showed similar increases in luciferase activity in both reporter plasmids on Foxn1 expression (Fig. 4C
). This result strongly suggested that Foxn1 indirectly promotes the activity of the PLC-1 promoter. We then determined the region of the PLC-1 promoter crucial for activation. For this purpose, we generated reporter plasmids containing various lengths of the PLC-1 promoter. Luciferase activities of reporter plasmids containing positions –1976 to +127, –465 to +127, –159 to +127, and –83 to +127 of the PLC-1gene were elevated by exogenously expressed Foxn1 (Fig. 4D
). However, elevation of luciferase activity was not observed with a reporter plasmid containing positions –18 to +127 of the PLC-1 gene (Fig. 4D
). This result indicated that the sequence from –83 to –18 of the PLC-1 gene is essential for Foxn1-mediated activation of the PLC-1 promoter.
|
We then examined whether Foxn1 induces PLC-1 expression at the protein level. Western blot analysis revealed that the amount of PLC-1 protein was increased remarkably by exogenously expressed Foxn1 but not by the nu mutant of Foxn1 (Fig. 4E
). Spontaneous nude mutation of the Foxn1 gene (R255X) is also found in humans (17)
and the phenotype of humans with the nude mutation is identical to that of nude mice. Therefore, we examined whether Foxn1 (R255X) could induce PLC-1 expression. Western blot analysis revealed that PLC-1 expression was not upregulated by Foxn1 (R255X). In contrast to Foxn1-induced PLC-1 expression, we found that PLC-1 did not affect expression of Foxn1 (Fig. 4F
). These results indicated that Foxn1 functions as an upstream molecule of PLC-1 and induces PLC-1 expression at both the transcriptional and translational levels.
Expression of PLC-1 is decreased in skin of nude mice
Because Foxn1 induces expression of PLC-1 in U2OS cells and because Foxn1 and PLC-1 showed similar expression patterns in hair follicles, we expected that the level of expression of PLC-1 would be altered in skin of nude mice. In situ hybridization analysis revealed that PLC-1 was abundantly expressed in hair follicles of control mice (Fig. 5
A), whereas only faint expression of PLC-1 was observed in hair follicles of nude mice (Fig. 5B
). In addition, we performed Western blot analysis and confirmed that expression of PLC-1 was remarkably decreased in skin of nude mice (Fig. 5C
). We then examined expression of PLC-1 in skin of control and nude mice at different time points. We observed a similar, weak level of expression of PLC-1 in skin of both mice at P0. However, the difference in expression of PLC-1 became evident as early as P8. At this stage, strong expression of PLC-1 was observed in skin of control mice, whereas expression of PLC-1 was still low in skin of nude mice. This difference in PLC-1 expression became more remarkable at P28 (Fig. 5D
). Finally, we examined whether Foxn1 expression was altered in skin of PLC-1 KO mice and found that expression of Foxn1 was not changed in the absence of PLC-1 (Fig. 5E
). These results indicate that Foxn1 is the upstream regulator of PLC-1 expression.
|
Activation of PLC downstream effectors is impaired in skin and keratinocytes of nude mice
We showed previously that activation of downstream effectors of PLC was impaired in skin or keratinocytes of PLC-1 KO mice (8)
. Therefore, we examined changes in activity of these downstream effectors in skin and keratinocytes of nude mice. At first, we examined expression level and activity of PKCs. Western blot analysis revealed that the amount of PKC- and - protein was remarkably decreased in skin of nude and PLC-1 KO mice at P8, compared with that of wild-type skin (Fig. 6
A). Then we examined whether down-regulation of some PKC isozymes causes reduced PKC activity. The antibody against phosphorylated PKCs was used for detection of activated PKCs. Activated PKCs are observed in epidermis and the upper part of hair follicles in control skin, whereas signals of activated PKCs are only faintly observed in skin of nude and PLC-1 KO mice (Fig. 6B
). We next examined activity of nuclear factor of activated T cells (NFAT), a transcription factor activated by intracellular calcium elevation, in keratinocytes using a luciferase assay. NFAT activity in keratinocytes from nude mice and PLC-1 KO mice was less than that in keratinocytes from wild-type mice (Fig. 6C
). These results indicate that the downstream signal of PLC is impaired in skin and keratinocytes of nude mice.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 KO mice. For this purpose, we screened for mice that displayed a phenotype similar to that of PLC-1 KO mice and found that nude mice show reduced expression of hair keratins, including mHa3 (Fig. 2A-C
), and have a hairless phenotype similar to that of PLC-1 KO mice (Fig. 1)
. Therefore, we hypothesized that PLC-1 KO and nude mice have defects in the same signaling pathway. Deletion of a single base pair nucleotide in the Foxn1 gene results in a loss-of-function mutation and the phenotype of nude mice (12)
. Therefore, we examined whether exogenously expressed Foxn1 affects the expression of PLC-1 and found that Foxn1 induces expression of PLC-1 at both the transcriptional and translational levels (Fig. 4)
. We used U2OS osteosarcoma cells in the present study. Given that Foxn1 induced expression of PLC-1 in U2OS cells, the regulation of PLC-1 expression by Foxn1 may be a general phenomenon not limited to hair follicles. We also found that the transcriptional activity of PLC-1 was upregulated by exogenously expressed Foxn1 in COS7 cells (data not shown).
Expression of hair keratins, such as mHa1, mHa2, mHa3, mHa4, mHb5, and mHb6, is lower in the skin of PLC-1 KO mice than in the skin of control mice (Fig. 2C
). In addition, we confirmed that Foxn1 was expressed normally in the skin of PLC-1 KO mice (Fig. 5E
). These results indicate that Foxn1 is insufficient for the normal expression of hair keratins and that both Foxn1 and PLC-1 are required. Given that expression of hair keratins in the skin of PLC-1 KO and nude mice was strikingly similar, PLC-1 is likely to be in close functional proximity to Foxn1 with respect to hair keratin induction.
How PLC-1 induces the expression of hair keratins remains to be determined. In our system, Foxn1 induced PLC-1 expression but not hair keratin expression. It has been reported that exogenous expression of Foxn1 does not induce expression of hair keratins even in isolated keratinocytes (18)
. The reason may be that other molecules specifically expressed in hair follicles, in addition to Foxn1 and PLC-1, are required for hair keratin expression. PLC-1 has been reported to translocate into the nucleus (19)
. In addition, PIP2, the substrate of PLC-1, regulates the expression of genes (20)
. Therefore, one possible mechanism may be that PLC-1 regulates hair keratin expression by hydrolysis of PIP2 in nuclei.
We showed that the protein levels of total and activated PKC are impaired in skin and keratinocytes of nude mice. It was reported that protein levels of total and activated PKCs are increased in nude keratinocytes compared with those of wild-type keratinocytes (21)
. In this study, we used extracts of epidermis and skin sections from P8 mice. This difference in experimental systems may cause the different results.
The skin phenotype of nude mice has not been studied as intensively as the immunological phenotype, and there are only a few reports regarding the genes that expressed differently and cause the hairless phenotype in nude mice. Therefore, our finding helps clarify the molecular mechanisms underlying hairlessness of nude mice. There are a few reports regarding the genes induced by Foxn1 in skin. In addition to PLC-1, Akt was also identified as a Foxn1-inducible gene in keratinocyte (18)
. Because Akt, like PLC-1, is a key molecule in the phosphoinositide signaling pathway, it is possible that the phosphoinositide signaling pathway is important for hair shaft formation. Recently, expression of desmocolin2 (Dsc2) was also reported to be regulated by Foxn1, and its expression decreased in the skin of nude mice (22)
. We found that expression of Dsc2 mRNA in PLC-1 KO mice was not changed compared with that in control mice (Supplemental Fig. S2A). This result suggests that expression of Dsc2 is regulated by Foxn1 independently of PLC-1 in contrast to hair keratins.
Other than Foxn1, there are some genes that regulate hair shaft formation. Among these, Lef1 (23)
and Hoxc13 (24
, 25)
play critical roles in hair shaft formation. However, we did not find any change in expression of these genes between control and PLC-1 KO mice (Supplemental Fig. S2B–E). This indicates that the hairlessness in PLC-1 KO mice is not caused by reduced expression of Lef1 and Hoxc13. In addition to Lef1 and Hoxc13, bone morphogenic protein (BMP) signaling plays important roles in hair formation (26
27
28
29)
. Mice in which BMP signaling is inhibited exhibit a hairless phenotype similar to that of PLC-1 KO and/or nude mice, reduced expression of Foxn1, and a subsequent decrease in hair keratin expression (30)
. Therefore, it may be interesting to examine whether forced expression of PLC-1 rescues the hairless phenotype of these mice.
Nude mice are known to have immunological defects caused by athymia, and, therefore, we examined the morphology and function of the thymus in PLC-1 KO mice. In contrast to the striking similarities in skin abnormalities between PLC-1 KO and nude mice, we found no obvious abnormality in the thymus of PLC-1 KO mice (Supplemental Fig. S1A–E). It is possible that Foxn1 regulates expression of many genes in thymus and that PLC-1 is one of dispensable genes for thymic development. In other words, PLC-1 is essential for Foxn1-mediated fur development but is not essential for thymic development. This finding may contribute to our understanding of the phenotypes of nude mice.
Spontaneous nude mutation of the Foxn1 gene is also found in humans (17)
, and the phenotype of humans with the nude mutation is identical to that of nude mice. We found that Foxn1 (R255X), the Foxn1 mutation observed in humans, did not induce expression of PLC-1 (Fig. 4E
). Therefore, it is likely that expression of PLC-1 is decreased in humans with this mutation. Clarification of the mechanism by which PLC-1 regulates Foxn1-induced hair keratin expression may lead to improvements in therapies for abnormalities observed in humans with the mutation.
|
ACKNOWLEDGMENTS |
|---|
Received for publication June 20, 2007. Accepted for publication September 6, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 and -3 are essential in the trophoblast for placental development. Mol. Cell. Biol. 25,10979-109881 is required for skin stem cell lineage commitment. EMBO J. 22,2981-2991[CrossRef][Medline]1 associates with importin β1 and translocates into the nucleus in a Ca2+-dependent manner. FEBS Lett. 579,4949-4954[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  J. D. Stallings, Y. X. Zeng, F. Narvaez, and M. J. Rebecchi Phospholipase C-{delta}1 Expression Is Linked to Proliferation, DNA Synthesis, and Cyclin E Levels J. Biol. Chem., May 16, 2008; 283(20): 13992 - 14001. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
