Introduction

　Double－stranded RNA－dependent protein kinase（PKR）is interferon（IFN）－inducible, double－stranded RNA（dsRNA） activated serine／threonine protein kinase widely expressed in mammalian cells, includ-ing osteoblasts1―4）. PKR can inhibit viral protein syn-thesis through phosphorylation of eukaryotic initia-tion factor－2α（eIF－2α）, leading to its participation in cellular antiviral responses5）. PKR has also been reported to function as a signal transducer for mediat-

ing transcriptional activation and have many roles in response to various stimuli. For example, PKR induces the production of cytokines and growth fac-tors, and, in turn, stimulates cell growth and differen-tiation6―8）. Several studies have demonstrated that PKR is a product of a tumour suppressor gene9,10）, and suggested that it is involved in the induction of apoptosis by dsRNA or TNF－α11,12）.　The signal transducers and activators of transcrip-tion（STAT）family is a group of transcription factors recognized as critical integrators of cytokine and growth factor receptor signalling13）. After cell－surface receptor binding by ligands, STAT1 is activated by the Janus kinases（JAKs）, other cell－surface recep-tor－associated kinases. In turn, activated STAT1

J. Oral Biosci. 49（3）：155－162, 2007

Review（JAOB／Rising Members Award）

The Role of Double－stranded RNA－dependent Protein Kinase in Osteoblasts

Kaya Yoshida§

Department of Histology and Oral Histology, Institute of Health Biosciences,

The University of Tokushima Graduate School

3－18－5, Kuramoto, Tokushima 770－8504, Japan

〔Received on February 25, 2007；Accepted on April 3, 2007〕

Key words：osteoblasts／PKR／Runx2／STAT1α

Abstract：Double－stranded RNA－dependent protein kinase（PKR）is interferon（IFN）－inducible, double－stranded RNA（dsRNA）－activated serine／threonine protein kinase. PKR is known to have an anti－viral function and mediate transcriptional activation. However, there are few reports concerning the role of PKR in bone metabolism. We established cells that stably expressed the PKR catalytic mutant gene（PKR－K／R）to investigate the role of PKR in osteoblast differentiation. The PKR－K／R mutant cells exhibited up－regu-lated cell proliferation and low alkaline phosphatase（ALP）activity. The PKR－K／R mutant cells were not able to form bone nodules in vitro. In the PKR－K／R mutant cells, runt－related gene 2（Runx2）－mediated transcription decreased compared to the level in control cells. The expression of STAT1α protein increased, and the protein was translocated to the nucleus in PKR－K／R mutant cells. When the expression of STAT1α protein in PKR mutant cells was suppressed using RNAi, the activity of Runx2－mediated transcription recovered to the control level. These results indicate that PKR is a stimulator of Runx2 transcription and a negative modulator of STAT1α expression. Our findings also suggest that PKR plays an important role in the differentiation of osteoblasts and calcification of bone matrix by modulating STAT1α and／or Runx2 expression.

§ Corresponding author　 E－mail：kaya@dent.tokushima－u.ac.jp.

dimerizes and translocates to the nucleus, where it then binds to DNA elements and regulates cytokine－specific gene transcription14,15）. Many cytokines use this pathway to transduce their biological response under various stresses. In bone metabolism, STAT1 promotes the differentiation of human osteoblasts and prevents apoptosis of these cells in response to IL－6－type cytokines16）. An increase in bone mass, bone mineral content, and other bone growth parameters was observed in STAT1－/－ mice, suggesting that STAT1 has a negative effect on bone formation17,18）.　PKR is one of the best characterized genes regu-lated by the JAKs／STAT signalling pathway in response to IFN19）. On the other hand, PKR interacts with STAT1, and PKR－STAT1 complexes have nega-tive effects on STAT1 DNA－binding and transactiva-tion capacities20,21）. In addition, recent reports sug-gested that STAT1 acts as an attenuator of runt－related gene 2（Runx2）, an essential factor for the dif-ferentiation of osteoblasts22）. These findings suggest that PKR is involved in bone formation by modulating STAT1 and／or Runx2 function. However, there are no

reports concerning the roles of PKR in bone metabolism.　This review focuses on the regulatory roles of PKR in the differentiation of osteoblasts. I will discuss the transcriptional factors implicated in PKR function.

A Catalytic Dominant－negative Mutant of PKR Prevents the Differentiation of Osteoblasts and

Calcification of Bone Matrix

　It was found that the expression of a dominant－negative mutant PKR（PKR－K／R）, in which the amino acid Lys at 296 is replaced with Arg, inhibits the catalytic activity of PKR23）. To investigate the effect of PKR on osteoblasts, we transfected mouse MC3T3－E1 cells with PKR－K／R cDNA24）. Figure 1A shows that the wild－type cells reached confluence at day 6, when the cell number was 20 times greater than that at plating. The PKR－K／R mutant cells grew faster than the wild－type cells, reaching confluence at day 5. The number of cells at that time was approxi-

156 K. Yoshida：PKR and Osteoblasts

Fig. 1　Acceleration of cell proliferation in PKR－K／R mutant cellsA：The wild－type（MC；open squares） and PKR－K／R mutant（K／R；closed squares）cells were plated at a density of 6,000 per 35－mm dish, cultured for various time periods, and cell numbers were counted as indicated. Each plot shows the aver-age of 4 different cultures；the bars represent the S.E.M. Experiments were per-formed in duplicate, and representative data are presented. Data were analyzed by ANOVA, and a Bonferroni／Dunu’s test was utilized to estimate the significance between means. ＊p＜0.01 compared with the wild－type cells. B：After wild－type（MC）and PKR－K／R mutant（K／R）cells had been cultured for 5 days, the morphol-ogy of the 2 cell types was examined under a microscope. Bar represents 10 μm.

A B

mately 30 times greater than that at plating. The num-ber of mutant cells was approximately 160％ of that of wild－type cells at the confluent stage. PKR－K／R mutant cells were irregular in shape and smaller in size compared with wild－type cells（Fig.　1B）. These phenotypes were earlier shown to be correlated with decreased ALP activity and also decreased accumula-tion of collagen in MC3T3－E1 cells25）, indicating that osteoblastic differentiation was inhibited in the PKR－K／R mutant cells.　In agreement with changes in cell morphology, ALP activity, which is one of the markers of osteoblast dif-ferentiation, was inhibited in the PKR－K／R mutant cells. As shown in Fig.　2, ALP activity in the wild－type cells increased as the culture time proceeded, whereas no increase in activity was observed in the PKR－K／R mutant cells during the entire culture period. At day 10, ALP activity in the wild－type cells was 6.5－fold higher than that of the PKR－K／R mutant cells.　Since MC3T3－E1 cells have the capacity to undergo osteoblastic differentiation and mineralization in vitro26）, we next examined whether PKR could influence mineralization by MC3T3－E1 cells. In 25－day wild－type cultures, mineralized bone nodules appeared as dark dots when stained by the von Kossa technique（Fig.　3）. The strongly stained black nodules were not observed in the PKR－K／R mutant cells, while only slightly stained regions were detected in the mutant cells. These results indicated that PKR mutation leads to the inhibition of osteoblast differ-entiation.

PKR Regulates Runx2 Transcriptional Activity

　The runt－related gene 2（Runx2） is a transcrip-tional factor which belongs to the runt－domain gene family. It was reported that targeted disruption of Runx2 results in the complete inhibition of bone for-mation by osteoblasts27）. In addition, the over－expres-sion of Runx2 induces nonosteogenic cells to express osteoblast－related genes in vitro28）. These reports suggest that Runx2 is essential for osteoblast differen-tiation and bone formation. To determine if the tran-scriptional activity of Runx2 is altered by the catalytic

mutation of PKR, we constructed a fused luciferase（Luc）reporter gene consisting of 3 tandem Runx2 binding elements（3×OSE）. Wild－type cells or PKR－K／R mutant cells were transfected with this con-struct, and the stably expressing cells were isolated. As shown in Fig.　4, luciferase activity in the PKR－K／R mutant cells was significantly suppressed；the activity was only approximately 10％ of that of the wild－type cells. Levels of luciferase activity in cells transfected with pcDNA 3.1 were similar to those in wild－type cells. These results indicate that PKR regu-lates Runx2－induced transactivation.

PKR Regulates Runx2 Activity through STAT1－dependent Pathway

　We next examined the expression of STAT1α pro-tein in wild－type and PKR－K／R mutant cells. Figure

157

Fig. 2　ALP activity in PKR－K／R mutant cellsALP activity was measured in the wild－type（open bars）and PKR－K／R mutant（solid bars）cells on the days indicated. Each plot shows the average of 4 different cultures, and the bars repre-sent the S.E.M. The results are representative of 3 independent experiments. Data were analyzed by ANOVA, and a Bonferroni／Dunu’s test was utilized to estimate the significance between means. ＊p＜0.01 compared with the wild－type cells.

158 K. Yoshida：PKR and Osteoblasts

Fig. 3　Bone nodule formation was prevented in PKR－K／R mutant cellsWild－type MC3T3－E1（MC）, PKR－K／R mutant（K／R）, and MC3T3－E1 cells transfected with pcDNA 3.1（NC）were cultured for 25 days. The mineralization of these cells was analyzed by the von Kossa staining method. Numerous bone nod-ules were detected in the wild－type cells and those transfected with pcDNA 3.1. However, bone nodules were not found in the PKR－K／R mutant cells.

Fig. 5　Enhanced STAT1α expression in PKR－K／R mutant cells

Wild－type（MC）and PKR－K／R mutant（K／R）cells were cultured for the indicated periods and cell lysates were prepared. Twelve micrograms of protein of each sample was separated using 12.5％ SDS－PAGE, transferred to a PVDF membrane, and incubated with anti－STAT1α spe-cific antibody. The antibody was stripped off the membra-ne, which was then re－incubated with anti－β－actin anti-body as a loading control.

Fig. 4　Runx2－mediated transcription in PKR－K／R mutant cells

Wild－type（MC）and PKR－K／R mutant（K／R）cells, and wild－type cells transfected with pcDNA 3.1（NC）were transiently transfected with the 3×OSE－Luc reporter. After a 48－h incubation, the luciferase activity was measured. The activity was normalized using the values obtained in wild－type cells as 100％. Each plot shows the average of 3 different experiments；the bars represent the S.E.M. Data were analyzed by ANOVA, and a Bonferroni／Dunu’s test was utilized to estimate the significance between means. ＊p＜0.05 com-pared with the wild－type cells. The results are representative of 3 independent experiments.

5 shows that the expression of STAT1α protein decreased in a time－dependent manner in the wild－type cells. However, the level of STAT1α protein increased in PKR－K／R mutant cells cultured for 5 days. At 10 or 15 days in culture, the expression of STAT1α protein was lower than that in cells cultured for 5 days.　Because the activated STAT1α translocates to the nucleus, we examined the effects of PKR on the cellu-lar localization of the STAT1α protein. We prepared nuclear and cytosolic fractions from cells cultured for 5 days and then examined the expression of STAT1α by Western blotting. As shown in Fig.　6, STAT1α underwent nuclear translocation in the PKR－K／R mutant cells；however, the majority of STAT1α remained in the cytoplasm in the wild－type cells or those transfected with pcDNA 3.1. The purity of the nuclear fraction was confirmed using an anti－B23 anti-body, which recognizes proteins present in the nuclear but not in the cytosolic fraction（Fig.　6）. These findings point out the possibility that PKR regulates the functions of STAT1α as well as its expression.　Next, we investigated the relationship between Runx2 transcriptional activity and STAT1α expression using RNA interference. Treatment of PKR－K／R

mutant cells for 96 h with STAT1α siRNA efficiently reduced the level of STAT1α protein in these cells, whereas expression of the STAT1α protein was not suppressed in cells treated with control siRNA（Fig.　7A）. The level of β－actin did not change in cells treated with STAT1α siRNA or control siRNA. These results indicate that the STAT1α RNA interference worked well in our system. Thus, we analyzed Runx2 transcriptional activity with STAT1α siRNA. The Runx2－mediated transcriptional activity in the PKR－K／R cells was 30％ that of the wild－type cells. How-ever, activity increased in PKR－K／R cells treated with STAT1α siRNA. The activity was almost the same as that in untreated wild－type cells. The control siRNA treatment did not increase the Runx2 tran-scriptional activity in PKR－K／R mutant cells（Fig.　7B）. These data suggest that the increase in STAT1α would result in the suppression of Runx2 activity in PKR－catalytic mutated cells. These findings are con-sistent with recent reports that STAT1 interacts with Runx2 in the cytoplasm and inhibits the nuclear local-ization and transcriptional activity of Runx2 in osteo-blasts, demonstrating the inhibitory effect of STAT1 on bone formation18,22）. PKR also interacts with STAT1, and the PKR－STAT1 complex has negative

159

Fig. 6　Nuclear translocation of STAT1α in PKR－K／R mutant cells

Wild－type（MC）and PKR－K／R mutant（K／R）cells, as well as MC3T3－E1 cells transfected with pcDNA 3.1（NC）, were cultured for 5 days. Nuclear（N）and cytosolic（C）fractions from each cell type were prepared, and the levels of STAT1α protein were analyzed by Western blotting. The antibody was stripped off the membrane, which was then re－incubated with anti－B－23 antibody as a nuclear marker.

effects on STAT1 DNA－binding and transactivation20,21）. Therefore, it is necessary to test whether STAT1 and Runx2 can form a complex in the cytoplasm in both wild－type and PKR－K／R mutant cells to understand how PKR regulates bone formation through the STAT1 and／or Runx2 pathways.

Conclusion

　Our present results clearly demonstrate that bone nodule formation does not occur in PKR－K／R mutant osteoblasts in vitro；however, no abnormality in bone formation was reported in PKR－/－ mice29）. Differ-ences in bone formation between the knockout mice and PKR－K／R mutant cells raise the possibilities that functionally redundant pathways regulating bone formation exist in PKR－/－ mice. In this case, PKR gene removal can be compensated for by another

gene during development30）, indicating that PKR can regulate bone formation while closely cooperating with other pathways. PKR is activated during infec-tion and various cellular stresses, and is identified as a component of the pro－inflammatory response31,32）. Thus, it is also possible that PKR may modulate bone metabolism through the STAT1α signalling pathway activated by stimulators that are involved in the immune response and inflammation.　In conclusion, the data presented in this review indicate that the kinase activity of PKR is required for the differentiation of osteoblasts and bone formation in vitro, although the molecular mechanism is not completely understood. It is important to elucidate how PKR regulates bone formation through the STAT1 and／or Runx2 pathways. This knowledge will contribute to a better understanding of normal bone formation and various diseases of the skeletal system.

160 K. Yoshida：PKR and Osteoblasts

Fig. 7　Effect of STAT1α siRNA on Runx2－mediated transcriptionA：Duplexes of 25－nucleotide STAT1α siRNA were used to transfect PKR－K／R mutant cells. The cell lysates were prepared 96 h after transfection, and the level of STAT1α protein was examined by Western blot analysis. The antibody was stripped off the membrane, which was then re－incubated with anti－β－actin antibody as the loading control. B：To assay the activity of the OSE2 promoter, PKR－K／R mutant（K／R）cells were transfected with STAT1α siRNA or control siRNA. At 24 h after transfection, the cells were transfected with OSE2－Luc expression vector and pRL vector. After a 48－h incubation, the luciferase activity was measured. The activity was normalized using the values obtained in wild－type cells as 100％. The experiment was performed in tripli-cate, and the bars represent the S.E.M. The results are representative of 3 independ-ent experiments. Data were analyzed by ANOVA, and a Bonferroni／Dunu’s test was utilized to estimate the significance between means.

BA

Acknowledgements

　I am grateful to Professor Tatsuji Haneji and col-laborators of the Department of Histology and Oral Histology, Institute of Health Biosciences, The Uni-versity of Tokushima Graduate School, for their sup-port and encouragement. I also thank Dr. Hiroyuki Morimoto and Dr. Hirohiko Okamura for their kind support and suggestions. This study was supported in part by a Grant－in－Aid for Scientific Research from the Ministry of Education, Science, Sports, and Cul-ture of Japan（17791304）.

References

1）Hovanessian, A.　G.：The double stranded RNA－acti-vated protein kinase induced by interferon：dsRNA－PK.　J. Interferon Res. 9：641―647, 1989.

2）Williams, B.　R.　G.：Role of the double－stranded RNA－activated protein kinase（PKR） in cell regu-lation. Biochem. Soc. Trans. 25：509―513, 1997.

3）Samuel, C.　E.：Antiviral actions of interferon. Inter-feron－regulated cellular proteins and their surpris-ingly selective antiviral activities. Virology 183：1―11, 1991.

4）Morimoto, H., Okamura, H., Yoshida, K., Kitamura S. and Haneji, T.：Okadaic acid induces apoptosis through double－stranded RNA－dependent protein kinase／eukaryotic initiation factor－2α pathway in human osteoblastic MG63 cells. J. Biochem. 136：433―438, 2004.

5）deHaro, C., Méndez, R. and Santoyo, J.：The eIF－2α kinases and the control of protein synthesis. FASEB J, 10：1378―1387, 1996.

6）Ito, T., Jagus, R. and May, W.　S.：Interleukin 3 stimu-lates protein synthesis by regulating double－stranded RNA－dependent protein kinase. Proc. Natl. Acad. Sci. USA 91：7455―7459, 1994.

7）Mundschau, L.　J. and Faller, D.　V.：Platelet－derived growth factor signal transduction through the inter-feron－inducible kinase PKR, Immediate early gene induction. J. Biol. Chem. 270：3100―3106, 1995.

8）Salzberg, S., Mandelbaum, M., Zalcberg, M. and Shainberg, A.：Interruption of myogenesis by trans-forming growth factorβ1 or EGTA inhibits expression and activity of the myogenic－associated（2’－5’）oli-goadenylate synthetase and PKR. Exp. Cell Res.

　219：223―232, 1995.

9）Clemen, M.：Tumorigenesis. Suppression with a difference. Nature 360：210―211, 1992.

10）Lengyel, P.：Tumor－suppressor genes：news about the interferon connection. Proc. Natl. Acad. Sci. USA 90：5893―5895, 1993.

11）Yeung, M.　C., Liu, J. and Lau, A.　S.：An essential role for the interferon－inducible, double－stranded RNA－activated protein kinase PKR in the tumor necrosis factor－induced apoptosis in U937 cells.：Proc. Natl. Acad. Sci. USA 93：12451―12455, 1996.

12）Der, S.　D., Yang, Y.　L., Weissmann, C. and Williams, B.　R.　G.：A double－stranded RNA－activated protein kinase－dependent pathway mediating stress－induced apoptosis. Proc. Natl. Acad. Sci. USA 94：3279―3283, 1997.

13）Schindler, C.：Cytokines and JAK－STAT signalling. Exp. Cell Res. 253：7―14, 1999.

14）Chen, W., Daines, M.　O. and Hershey, G.　K.：Turning off signal transducer and activator of transcription（STAT）：the negative regulation of STAT signalling. J. Allergy Clin. Immunol. 114：476―489, 2004.

15）Clevenger, C.　V.：Roles and regulation of stat family transcription factors in human breast cancer. Am. J. Pathol. 165：1449―1460, 2004.

16）Bellido, T., Borba, V.　Z., Roberson, P. and Manolagas, S.　C.：Activation of the Janus kinase／STAT（signal transducer and activator of transcription）signal trans-duction pathway by interleukin－6－type cytokines pro-motes osteoblast differentiation. Endocrinology 138：3666―3676, 1997.

17）Sahni, M., Raz, R., Coffin, J.　D., Levy, D. and Basili-co, C.：STAT1 mediates the increased apoptosis and reduced chondrocyte proliferation in mice overex-pressing FGF2. Development 128：2119―2129, 2001.

18）Xiao, L., Naganawa, T., Obugund, eE., Gronowicz, G., Ornitz, D.　M., Coffin, J.　D. and Hurley, M.　M.：Stat1 controls postnatal bone formation by regulating fibro-blast growth factor signaling in osteoblasts. J. Biol. Chem. 279：27743―27752, 2004.

19）Clemens, J.　E. and Elia, A.：The double－stranded RNA－dependent protein PKR：structure and func-tion. J. Interfer. Cytokine. Res. 17：503―524, 1997.

20）Wong, A.　H., Tam, N.　W., Yang, Y.　L., Cuddihy, A.　R., Li, S., Kirchhoff, S., Hauser, H., Decker, T. and Koro-milas, A.　E.：Physical association between STAT1 and the interferon－inducible protein kinase PKR and implications for interferon and double－stranded RNA

161

signaling pathways. EMBO J. 16：1291―1304, 1997.21）Tam, N.　W.　N., Ishii, T., Li, S., Wong, A.　H.　T., Cud-

dihy, A.　R. and Koromilas, A.　E.：Upregelation of STAT1 protein in cells lacking or expressing mutants of the double－stranded RNA－dependent protein kinase PKR. Eur. J. Biochem. 262：149―154, 1999.

22）Kim, S., Koga, T., Isobe, M., Kern, B.　E. Yokochi, T., Chin, Y.　E., Karsenty, G., Taniguchi, T. and Takaya-nagi, H.：Stat1 functions as a cytoplasmic attenuator of Runx2 in the transcriptional program of osteoblast differentiation. Gene Dev. 17：1979―1991, 2003.

23）Takizawa, T., Ohashi, K. and Nakanishi, Y.：Possible involvement of double－stranded RNA－activated pro-tein kinase in cell death by influenza virus infection. J. Virol. 70：8128―8132, 1996.

24）Yoshida, K., Okamura, H., Amorim, B.　R., Ozaki, A., Tanaka, H., Morimoto, H. and Haneji, T.：Double－stranded RNA－dependent protein kinase is required for bone calcification in MC3T3－E1 cells in vitro. Exp. Cell Res. 311：117―125, 2005.

25）Horiguchi, Y., Nakai, T. and Kume, K.：Effects of Bor-detella bronchiseptica dermonecrotic toxin on the structure and function of osteoblastic clone MC3T3－E1 cells. Infect. Immun. 59：1112―1116, 1991.

26）Sudo, H., Kodama, H., Amagai, Y., Yamamoto, S. and Kasai, S.：In vitro differentiation and calcification in a new clonal osteogenic cell line derived from new-born mouse calvaria. J. Cell Biol. 96：191―198, 1983.

27）Komori, T., Yagi, H., Nomura, S., Yamaguchi, A., Sasaki, K., Deguchi, K., Shimizu, Y., Bronson, R.　T., Gao, Y.　H., Inada, M., Sato, M., Okamoto, R., Kitamu-ra, Y., Yoshiki, S. and Kishimoto, T.：Targeted disrup-tion of Cbfa1 results in a complete lack of bone forma-tion owing to maturational arrest of osteoblasts. Cell 89：755―764, 1997.

28）Ducy, P., Zhang, R., Geoffroy, V., Ridall, A.　L. and Karsenty, G.：Osf2／Cbfa1：a transcriptional activa-tor of osteoblast differentiation. Cell 89：747―754, 1997.

29）Yang, Y.　L., Reis, L.　F., Pavlovic, J., Aguzzi, A., Scha-fer, R., Kumar, A., Williams, B.　R, Aguet, M. and Weissmann, C.：Deficient signaling in mice devoid of double－stranded RNA－dependent protein kinase. EMBO J. 14：6095―6106, 1995.

30）Elkins, T., Zinn, K., McAllister, L., Hoffmann, F.　M. and Goodman, C.　S.：Genetic analysis of drosophila neural cell adhesion molecule：Interaction of fasci-clin I and abelson tyrosine kinase mutations. Cell 60：565―575, 1990.

31）Williams, B.　R.：PKR；a sentinel kinase for cellular stress. Oncogene 18：6112―6120, 1999.

32）Gusella, G.　L., Musso, T., Rottschafer, S.　E., Pulkki, K. and Varesio, L.：Potential requirement of a func-tional double－stranded RNA－dependent protein kinase（PKR）for the tumoricidal activation of macro-phages by lipopolysaccharide or IFN－α β, but not IFN－γ. J. Immunol. 154：345―354, 1995.

162 K. Yoshida：PKR and Osteoblasts