Distinct mechanisms are responsible for osteopenia and growth retardation in OASIS-deficient mice

Bone 48 (2011) 514–523

Contents lists available at ScienceDirect

Bone j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / b o n e

Distinct mechanisms are responsible for osteopenia and growth retardation in OASIS-deficient mice Tomohiko Murakami a,⁎, Shin-ichiro Hino a, Riko Nishimura b, Toshiyuki Yoneda b, Akio Wanaka c, Kazunori Imaizumi a,d,⁎ a

Division of Molecular and Cellular Biology, Department of Anatomy, Faculty of Medicine, University of Miyazaki, 5200 Kihara, Kiyotake, Miyazaki 889-1692, Japan Department of Molecular and Cellular Biochemistry, Graduate School of Dentistry, Osaka University, 1-8 Yamadaoka, Suita, Osaka 565-0871, Japan Department of Anatomy and Neuroscience, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8521, Japan d Department of Biochemistry, Graduate School of Biomedical Sciences, Hiroshima University, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8551, Japan b c

a r t i c l e

i n f o

Article history: Received 2 July 2010 Revised 14 October 2010 Accepted 27 October 2010 Available online 1 November 2010 Edited by: Bjorn Olsen Keywords: Osteopenia Bone formation Osteoblast Endoplasmic reticulum Growth retardation

a b s t r a c t Old astrocyte specifically induced substance (OASIS), which is a new type of endoplasmic reticulum (ER) stress transducer, is a basic leucine zipper transcription factor of the CREB/ATF family that contains a transmembrane domain and is processed by regulated intramembrane proteolysis in response to ER stress. OASIS is selectively expressed in certain types of cells such as astrocytes and osteoblasts. We have previously demonstrated that OASIS activates transcription of the type I collagen gene Col1a1 and contributes to the secretion of bone matrix proteins in osteoblasts, and that OASIS−/− mice exhibit osteopenia and growth retardation. In the present study, we examined whether osteopenia in OASIS−/− mice is rescued by OASIS introduction into osteoblasts. We generated OASIS−/− mice that specifically expressed OASIS in osteoblasts using a 2.3-kb osteoblast-specific type I collagen promoter (OASIS−/−;Tg mice). Histological analysis of OASIS−/−;Tg mice revealed that osteopenia in OASIS−/− mice was rescued by osteoblast-specific expression of the OASIS transgene. The decreased expression levels of type I collagen mRNAs in the bone tissues of OASIS−/− mice were recovered by the OASIS transgene accompanied by the rescue of an abnormal expansion of the rough ER in OASIS−/− osteoblasts. In contrast, growth retardation in OASIS−/− mice did not improve in OASIS−/−;Tg mice. Interestingly, the serum levels of growth hormone (GH) and insulin-like growth factor (IGF)-1 were downregulated in OASIS−/− mice compared with those in wild-type mice. These decreased GH and IGF-1 levels in OASIS−/− mice did not change when OASIS was introduced into osteoblasts. Taken together, these results indicate that OASIS regulates skeletal development by osteoblast-dependent and -independent mechanisms. © 2010 Elsevier Inc. All rights reserved.

Introduction The endoplasmic reticulum (ER) is a critical cellular compartment in which secretory and transmembrane proteins are folded or processed [1,2]. Various cellular stress conditions lead to the accumulation of unfolded or misfolded proteins in the ER lumen. When unfolded proteins accumulate in the ER lumen, eukaryotic cells activate a system termed the unfolded protein response (UPR) to deal with the unfolded proteins and avoid cellular damage [3–5]. In mammalian cells, monitoring of the ER lumen and signaling for the UPR are mediated by three major transducers, inositol-requiring 1, ⁎ Corresponding authors. T. Murakami is to be contacted at the Division of Molecular and Cellular Biology, Department of Anatomy, Faculty of Medicine, University of Miyazaki, 5200 Kihara, Kiyotake, Miyazaki 889-1692, Japan. Fax: + 81 985 85 9851. K. Imaizumi, Present address: Department of Biochemistry, Graduate School of Biomedical Sciences, Hiroshima University, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8551, Japan. Fax: + 81 82 257 5134. E-mail addresses: [email protected] (T. Murakami), [email protected] (K. Imaizumi). 8756-3282/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2010.10.176

PKR-like endoplasmic reticulum kinase and activating transcription factor 6 (ATF6). These transducers sense unfolded proteins in the ER lumen and transduce signals to the cytoplasm and nucleus for the transcription of UPR target genes, translational attenuation of global protein synthesis and ER-associated degradation (ERAD). ER stresses caused by genetic mutations and environmental factors are associated with various diseases, such as diabetes, neurodegeneration and osteogenesis imperfecta (OI) [6–8]. Old astrocyte specifically induced substance (OASIS) is a basic leucine zipper (bZIP) transcription factor of the CREB/ATF family that contains a transmembrane domain [9–11]. Its structure is very similar to that of ATF6. OASIS is cleaved at the membrane by regulated intramembrane proteolysis in response to ER stress. Its cleaved cytoplasmic domain, which contains the bZIP domain, translocates into the nucleus where it activates the transcription of target genes. Unlike classical ER stress transducers that are expressed ubiquitously, OASIS has a unique expression pattern that is limited to certain tissues and cells, including astrocytes and osteoblasts [9–14]. CREBH, BBF2H7 and AIbZIP have similar structures to OASIS, and their expression

T. Murakami et al. / Bone 48 (2011) 514–523

levels show tissue- or cell type-specific patterns [15–18]. These proteins, including OASIS, have been reported to function as tissuespecific ER stress transducers that convert ER stress to the transcription of target genes for development, differentiation, maturation or other cell type-specific events in various tissues including bone tissues [13–18]. Osteoblasts and osteoclasts are specialized cells responsible for bone formation and resorption, respectively [19,20]. Osteoblasts synthesize type I collagen, which is the major bone matrix protein and it is essential for proper bone formation and strength [21]. Mutations in the genes that encode type I collagen or defects in quality control or posttranslational modification of type I collagen are linked to misfolding or decreased synthesis of type I collagen, and consequently cause OI [22–24]. Therefore, protein quality control in the ER for secreted bone matrix proteins is important for the formation of integrated bone. We have previously demonstrated that OASIS activates the transcription of the type I collagen gene Col1a1 and contributes to the secretion of bone matrix proteins in osteoblasts [13]. Expression and activation of OASIS in osteoblasts are promoted by bone morphogenetic protein 2, the signaling of which promotes synthesis of large amounts of bone matrix proteins in osteoblasts and then causes mild ER stress. OASIS−/− mice exhibit severe osteopenia involving a decrease in type I collagen in the bone matrix, similar to OI type I, and they also show growth retardation [13,25]. The osteopenia in OASIS−/− mice results from a dysfunction of osteoblasts, which show decreased transcription of type I collagen and abnormal expansion of the rough ER with the accumulation of bone matrix proteins. While secretion of bone matrix proteins is impaired in OASIS−/− mice, expression levels of bone matrix proteins other than type I collagen are increased and expression level of type I collagen is decreased [13]. Therefore, type I collagen is decreased in the bone matrix of OASIS−/− mice. In contrast, it remains unclear whether growth retardation in OASIS−/− mice is caused by osteoblast dysfunction. In the present study, we generated OASIS−/− mice that specifically expressed OASIS in osteoblasts (OASIS−/−;Tg mice) and investigated whether expression of OASIS in osteoblasts rescues osteopenia, osteoblast dysfunction, and growth retardation in OASIS−/− mice.

515

Material and methods Generation of OASIS transgenic mice The generation of OASIS−/− mice has been previously described [13]. OASIS transgenic (Tg) mice were generated using a pJ251 vector containing an osteoblast-specific 2.3-kb Col1a1 promoter, FLAGtagged full-length mouse OASIS cDNA and an SV40 polyadenylation signal. The linearized construct was injected into fertilized eggs derived from C57BL/6CrSlc mice by SLC, Inc., Shizuoka, Japan. Transgene integration was identified by genomic polymerase chain reaction (PCR). The primers used for the genomic PCR were 5′GCCCAGGCCAGTCGTCGGAGCAG-3′ and 5′-CACGCTCCGTGTTCCACATC-3′. The tissue specificity of the transgene expression was examined by reverse transcription (RT)-PCR. Although several lines of transgenic mice were obtained, we demonstrated the results from a strain that specifically expressed the OASIS transgene in bone tissues (Fig. 1B). The transgenic line was maintained in OASIS+/− [13] background. OASIS−/−;Tg, OASIS−/− and wild-type (WT) mice were generated by mating OASIS+/− and OASIS+/−;Tg mice. All experiments were performed with the consent of the Animal Care and Use Committees of Miyazaki University. Growth hormone treatment Three-week-old mice were subcutaneously injected with 3 μg/g per day of recombinant growth hormone (GH) (Wako Pure Chemical Industries, Osaka, Japan) for 3 weeks. This dosage of GH treatment was determined according to a previously published protocol [26,27]. Control mice were injected with saline. Cell cultures Primary cultures of osteoblasts were prepared from the calvariae of postnatal 3–4-day-old OASIS−/− and OASIS−/−;Tg mice. The calvariae were digested with 0.1% collagenase (Wako Pure Chemical Industries, Osaka, Japan) and 0.2% dispase (Gibco BRL, Rockville, MD). The isolated cells were grown in alpha-modified Eagle's medium

Fig. 1. Generation of OASIS−/− mice that specifically express OASIS in osteoblasts. (A) Construction of an OASIS transgene containing an osteoblast-specific Col1a1 promoter and an SV40 polyadenylation signal (Poly A). Arrows indicate the annealing sites of the genomic PCR primer. Tg: OASIS transgenic mice. The right panel shows the results of genomic PCR for genotyping of the OASIS transgene. The expected band (0.5 kbp) is detected. (B) RT-PCR analysis of RNA from the indicated tissues of 4-week-old OASIS Tg mice. OASIS is specifically expressed in the bone tissue. (C) Western blot analysis of lysates from primary cultures of osteoblasts using an anti-OASIS antibody. Full-length OASIS (OA-full) and the cleaved Nterminal fragment of OASIS (OA-N) are expressed in OASIS−/− osteoblasts by the OASIS transgene.

516

T. Murakami et al. / Bone 48 (2011) 514–523

(Gibco BRL) supplemented with 10% fetal calf serum. The medium was changed every 3 days. The cell culture supernatant was collected at 24 h after the last medium change. Micromass culture was modified according to previously published protocols [17]. Briefly, mesenchymal cells were prepared from the limbs of embryonic day (E) 11.5 mice and digested with 0.1% trypsin and 0.1% collagenase D. A total of 2 × 107 cells/ml were plated and maintained in α-MEM supplemented with 10% fetal calf serum, 100 ng/ml BMP-2 (Sigma-Aldrich, Tokyo, Japan), 50 μg/ml ascorbic acid and 5 mM β-glycerophosphate. Micromass cultures were stained with alcian blue or alizarin red S. Alcian blue and alizarin red S stains were solubilized with 6 M guanidine hydrochloride and 100 mM cetylpyridinium chloride, respectively. Absorbance was measured using a spectrophotometer at 630 nm for alcian blue stain and 570 nm for alizarin red S stain. RT-PCR Total RNA was isolated from cells or bone tissue using an RNeasy Mini Kit (Qiagen, Valencia, CA) according to the manufacturer's protocol. First-strand cDNA was synthesized in a 40-μl reaction volume using a random primer (Takara, Shiga, Japan) and Moloney murine leukemia virus reverse transcriptase (Invitrogen, Carlsbad, CA). PCR was performed in a total volume of 30 μl using Paq 5000 DNA polymerase (Stratagene, La Jolla, CA) and the following primer sets were used: β-actin, 5′-TCCTCCCTGGAGAAGAGCTAC-3′ and 5′-TCCTGCTTGCTGATCCACAT-3′; OASIS, 5′-CCTTGTGCCTGTCAAGATGGAG-3′ and 5′-GCAGCAGCCATGGCAGAGGAG-3′; FLAG OASIS, 5′-ACTAGTCCACCATGGACTACAAGGACGAC-3′ and 5′-GCAGCAGCCATGGCAGAGGAG-3′; Col1a1, 5′-CCCCAACCCTGGAAACAGAC-3′ and 5′-GGTCACGTTCAGTTGGTCAAAGG-3′; Col1a2, 5′-GCAATCGGGATCAGTACGAA-3′ and 5′-CTTTCACGCCTTTGAAGCCA-3′. The PCR products were resolved by electrophoresis in a 4.8% acrylamide gel. Western blotting and ELISA For western blotting analysis, proteins were extracted from cells or tissues using a cell extraction buffer comprising 0.05 M Tris–HCl (pH 8.0), 0.15 M NaCl, 5.0 mM EDTA, 1% NP-40 and a protease inhibitor cocktail (MBL, Nagoya, Japan) at 4 °C. After centrifugation, the soluble protein in the extract was quantified. Samples were loaded onto 8% SDS-polyacrylamide gels. Protein-equivalent samples were subjected to western blotting analyses. The anti-OASIS antibody used has been described previously [13]. Anti-β-actin antibody was used as a control and was purchased from Sigma (St. Louis, MO). For measurements of GH, we used a Mouse Growth Hormone ELISA Kit (Linco Research R&D, Charles, MO) according to the manufacturer's instructions. For measurements of insulin-like growth factor (IGF)-1, we used a Mouse IGF-1 ELISA Kit (R&D Systems, Minneapolis, MN), in which there is no significant cross-reactivity or interference by IGF binding proteins or IGF-II. Histomorphometric, radiographic and μCT analyses Histomorphometric, radiographic and μCT analyses were performed for WT, OASIS−/− and OASIS−/−;Tg mice. Calcein doublelabeling of the tibiae in mice was performed by calcein injections (16 mg/kg body weight) at 6-day intervals. The tibiae were fixed in 70% ethanol, and the undecalcified bones were embedded in glycolmethacrylate. Sections of 3-μm thickness were cut longitudinally in the proximal region of the tibia and stained with toluidine blue. Histomorphometry was performed with a semiautomatic image analyzing system (Measure6; System Supply, Nagano, Japan) linked to a light microscope. Histomorphometric measurements were made at a magnification of × 400 in the secondary spongiosa area from the growth plate–metaphyseal junction. The nomenclature, symbols and units used were those recommended by the Nomenclature Commit-

tee of the American Society for Bone and Mineral Research. Radiographs were obtained using a soft X-ray apparatus (TRS-1005; Sofron, Tokyo, Japan). Micro-CT analysis was performed using a micro-CT system (Scan Xmate-A090S; Comscantecno, Kanagawa, Japan). Histological staining, electron microscopy and immunostaining For histological analysis, mouse skeletons were fixed in 10% formalin and then decalcified with 10% EDTA. Van Gieson staining was performed using 6-μm paraffin sections and standard protocols. For electron microscopy, bone tissue was fixed in 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer (Agar Scientific Ltd., Essex, UK) for 48 h at 4 °C, decalcified in 10% EDTA (pH 7.4) for approximately 4 weeks at 4 °C, post-fixed in 2% osmium tetroxide in 0.1 M cacodylate buffer, dehydrated, infiltrated and embedded in EPON812. Ultra-thin sections were stained with uranyl acetate and lead citrate and examined using an election microscope (Model 7100; Hitachi, Tokyo, Japan) operated at 80 kV. Immunostaining procedures were performed as previously described [13]. An anti-osteocalcin antibody was obtained from Santa Cruz Biotechnology, Inc. Rabbit antibodies against the C-propeptide of the type I collagen chain (LF41) [28] were provided by Dr. L. W. Fisher (National Institutes of Health, Bethesda, MD). Cells were visualized with a confocal microscope (Olympus FV1000D, Tokyo, Japan). Results Generation of OASIS−/− mice that specifically express OASIS in osteoblasts We generated OASIS Tg mice in which expression of the OASIS transgene was targeted to osteoblasts using a 2.3-kb osteoblast-specific type I collagen promoter (Fig. 1A). The tissue specificity of the transgene expression was assessed by RT-PCR, and the results confirmed that the transgene was only expressed in bone tissues, and not in other tissues including the cerebrum, kidney, spleen, liver, heart and muscle (Fig. 1B). OASIS Tg mice were viable and grew normally in comparison with their WT littermates (data not shown). We then generated OASIS−/− mice that specifically expressed OASIS in osteoblasts (OASIS−/−;Tg mice) by mating OASIS+/− and OASIS+/−;Tg mice. Western blot analysis using primary cultures of osteoblasts isolated from OASIS−/− and OASIS−/−; Tg mice confirmed that OASIS protein was expressed in OASIS−/− osteoblasts by osteoblast-specific expression of the OASIS transgene (Fig. 1C). Rescue of osteopenia in OASIS−/− mice by osteoblast-specific expression of the OASIS transgene Radiographic analysis showed that bone density in OASIS−/− mice was decreased compared with that in WT mice and that this decreased bone density was rescued by expression of OASIS in osteoblasts (Fig. 2A). Spontaneous fractures were often seen in OASIS−/− mice but they were not observed in OASIS−/−;Tg mice. Micro-CT and histological analysis revealed that decreases in both the trabecular and cortical bone thicknesses of OASIS−/− mice were rescued by expression of the OASIS transgene (Figs. 2B and C). Van Gieson staining, which specifically stains collagen fibrils, showed that the amount of collagen fibrils in the bone tissues of OASIS−/− mice was decreased compared with that in WT mice and that this decrease was improved by expression of the OASIS transgene (Fig. 2C). Taken together, these results indicated that the osteopenia involving a decreased amount of type I collagen in the bone matrix was rescued by osteoblast-specific expression of the OASIS transgene. Histomorphometric analysis of bones in OASIS−/−;Tg mice To further investigate the effects of the OASIS transgene on the bone tissues of OASIS−/− mice, we performed histomorphometric

T. Murakami et al. / Bone 48 (2011) 514–523

517

Fig. 2. Rescue of osteopenia in OASIS−/− mice by OASIS expression in osteoblasts. (A) X-ray images of the lower bodies of WT, OASIS−/− and OASIS−/−;Tg male mice at 12 weeks of age. The arrows indicate spontaneous fractures in the OASIS−/− mouse. Osteopenia in the OASIS−/− mouse is rescued in the OASIS−/−;Tg mouse. (B) μCT images of metaphysis (upper panels) and diaphysis (lower panels) in the femurs of the indicated male mice at 12 weeks of age (scale bar = 1 mm). (C) Van Gieson staining of trabecular (upper panels) and cortical (lower panels) bones in the femurs of the indicated male mice at 12 weeks of age (upper panels, scale bar = 500 μm; lower panels, scale bar = 100 μm). Van Gieson staining is specific for collagen fibrils. Note that the decreased area and intensity of van Gieson staining in the OASIS−/− bone tissues are rescued by expression of the OASIS transgene.

518

T. Murakami et al. / Bone 48 (2011) 514–523

Fig. 3. Histomorphometric analysis of bone tissues in OASIS−/−;Tg mice. (A) Histomorphometric analyses of the bone volume/tissue volume (BV/TV), trabecular thickness (Tb.Th) and trabecular number (Tb.N) in tibiae from WT, OASIS−/− and OASIS−/−;Tg female mice at 12 weeks of age (n = 4). (B–D) Bone formation-related parameters in histomorphometric analysis of trabecular bone in tibiae from the indicated female mice at 12 weeks of age. The mineralizing surface/bone surface (MS/BS), mineral apposition rate (MAR) and bone formation rate/bone surface (BFR/BS) are significantly rescued in OASIS−/−;Tg mice compared with those in OASIS−/− mice (n = 4) (B). Calcein double-labeling of tibiae in the indicated mice (C). The osteoblast surface/bone surface (Ob.S/BS) is significantly rescued in OASIS−/−;Tg mice compared with that in OASIS−/− mice (n = 4) (D). (E) Bone resorption-related parameters in histomorphometric analysis of trabecular bone in tibiae from the indicated female mice at 12 weeks of age. The eroded surface/bone surface (ES/BS), osteoclast number/bone perimeter (N.Oc/B.Pm) and osteoclast surface/bone surface (Oc.S/BS) are shown (n = 4). Data are expressed as the mean ± SE. Statistical significance was analyzed by ANOVA with Ryan's method. *p b 0.05; **p b 0.01.

T. Murakami et al. / Bone 48 (2011) 514–523

519

surface (BFR/BS) in OASIS−/− mice were improved by expression of the OASIS transgene, with recovery of the osteoblast surface/bone surface (Ob.S/BS) (Figs. 3B–D). In contrast, the eroded surface/bone surface (ES/ BS), osteoclast number/bone perimeter (N.Oc/B.Pm) and osteoclast surface/bone surface (Oc.S/BS) were not significantly different between OASIS−/− and OASIS−/−;Tg mice (Fig. 3E). These results indicated that the rescue of osteopenia in OASIS−/−;Tg mice was caused by improvement of osteoblast function and activity, but it was not due to inhibition of bone resorption. Expression of type I collagen mRNAs and ultrastructure in the bone tissues of OASIS−/−;Tg mice Fig. 4. Rescue of the decreased expression of type I collagen mRNAs in OASIS−/− osteoblasts by osteoblast-specific expression of the OASIS transgene. RT-PCR analysis of RNA from bone tissues of 4-week-old WT, OASIS−/− and OASIS−/−;Tg mice is shown. Type I collagen mRNA levels in OASIS−/−;Tg mice are recovered to levels in WT mice.

analysis of the bone tissues in WT, OASIS−/− and OASIS−/−;Tg mice. Decreases in the bone volume/tissue volume (BV/TV) and trabecular thickness (Tb.Th) in OASIS−/− mice were improved by OASIS expression in osteoblasts, whereas the trabecular number (Tb.N) was not altered (Fig. 3A). Decreases in the mineralizing surface/bone surface (MS/BS), mineral apposition rate (MAR) and bone formation rate/bone

We have previously reported that OASIS promotes the transcription of the type I collagen gene Col1a1 in osteoblasts [13]. To confirm the function of OASIS for the transcription of Col1a1, we performed RT-PCR analyses of type I collagen mRNAs (Col1a1 and Col1a2) in the bone tissues of OASIS−/−;Tg mice. We found that the expression levels of type I collagen mRNAs were downregulated in OASIS−/− mice compared with those in WT mice and that this decreased expression in OASIS−/− mice recovered to levels in WT mice by expression of the OASIS transgene (Fig. 4). Osteoblasts have a well-developed ER for synthesizing large amounts of bone matrix proteins [29,30]. However, osteoblasts in

Fig. 5. Rescue of abnormal rough ER expansion in OASIS−/− osteoblasts by expression of the OASIS transgene. (A) Representative electron microscopic images of osteoblasts in WT, OASIS−/− and OASIS−/−;Tg mice at 12 weeks of age are shown. The lower panels show higher magnification images of the boxed areas in the upper panels (upper panels, scale bar = 2 μm; lower panels, scale bar = 1 μm). OASIS expression in osteoblasts rescues abnormal rough ER expansion in the osteoblasts of OASIS−/− mice. (B) Immunohistochemistry using anti-procollagen 1a1 (LF41) and anti-osteocalcin (Ocn) antibodies in WT, OASIS−/−, and OASIS−/−;Tg osteoblasts of the calvaria in 4-week-old mice. Scale bar, 5 μm. Note that both proteins are accumulated in OASIS−/− osteoblasts (arrowheads) and this accumulation of proteins is rescued in OASIS−/−;Tg osteoblasts. B indicates bone.

520

T. Murakami et al. / Bone 48 (2011) 514–523

OASIS−/− mice show abnormal expansion of the rough ER, which contains large amounts of bone matrix proteins [13]. Therefore, we examined whether abnormal expansion of the rough ER in osteoblasts of OASIS−/− mice is rescued by OASIS expression in osteoblasts. Electron microscopic examinations revealed that abnormal rough ER expansion observed in OASIS−/− osteoblasts was rescued by expression of the OASIS transgene (Fig. 5A). In our previous report, we found that osteoblasts in OASIS−/− mice exhibit an accumulation of procollagen and osteocalcin with abnormal expansion of the rough ER [13]. In the current study, we examined whether accumulation of these bone matrix proteins was improved by the OASIS transgene using immunohistochemistry. We found that the OASIS transgene rescued the accumulation of procollagen and osteocalcin in OASIS−/− mice (Fig. 5B).

A

Growth retardation in OASIS−/− mice is not rescued by osteoblast-specific expression of the OASIS transgene OASIS−/− mice show growth retardation from 3–4 weeks of age [13]. We investigated whether growth retardation in OASIS−/− mice is rescued by osteoblast-specific expression of the OASIS transgene. As shown in Figs. 6A and B, the body weight and length of OASIS−/− mice were downregulated compared with those in WT mice and this downregulation in OASIS−/− mice was not rescued by expression of the OASIS transgene. The food intakes of both OASIS−/− and OASIS−/−; Tg mice were normal (Fig. 6C). In mammals, postnatal growth is mainly regulated by GH and IGF-1 [31]. Therefore, we measured levels of serum GH and IGF-1 in OASIS−/− mice at 4 weeks of age (Figs. 6D and E). We found that IGF-1 levels were significantly downregulated in OASIS−/−

B

Body weight (g)

25

Body length (cm)

10

20 9 15 8 10 WT -/-

5

7

- / - ;Tg 0

1

2

3

4

5

6

7

8

9 10 11 12

6

WT

-/-

- / - ;T g

(week )

C

Food intake (mg / day/ g body weight)

D GH (ng / ml) 10

2 00

8

1 50

6 1 00 4 50 0

2 WT

E

-/-

- / - ;T g

IGF- 1 (ng / ml)

0

WT

F

-/-

- / - ;T g

IGF- 1 (ng / ml)

1 .5 4 00 1 .2 3 00 0 .9 2 00

0 .6

1 00 0

0 .3

WT

-/-

- / - ;T g

0 WT

-/-

- / - ;T g

Fig. 6. Growth retardation in OASIS−/− mice is not rescued by OASIS expression in osteoblasts. (A) Average body weight curves of WT, OASIS−/− and OASIS−/−;Tg female mice (n = 6). (B) Body length (distance from the tip of the nose to the base of the tail) of indicated female mice at 12 weeks of age. The growth retardation of OASIS−/− mice is not rescued by OASIS expression in osteoblasts (n = 10). (C) Food intakes of the indicated female mice at 12 weeks of age (n = 5). (D, E) Serum GH and IGF-1 levels in the indicated female mice at 4 weeks of age. Both GH and IGF-1 levels are downregulated in OASIS−/− and OASIS−/−;Tg mice compared with those in WT mice (GH, n = 11; IGF-1, n = 10). (F) Levels of IGF-I secreted into the culture supernatants of primary osteoblasts from the indicated mice (n = 4). Data are expressed as the mean ± SE. Statistical significance was analyzed by ANOVA with Ryan's method. **p b 0.01.

T. Murakami et al. / Bone 48 (2011) 514–523

521

Fig. 7. Chondrocyte differentiation in micromass cultures of mesenchymal cells. (A) Mesenchymal cells were maintained as micromass cultures for the indicated days. Cells were stained with alcian blue for cartilage matrix proteins or alizarin red S for cartilage mineralization. (B) Quantification of absorbance from cells stained with alcian blue (top) and alizarin red S (bottom) (n = 3). Data are expressed as the mean ± SE. There were no significant differences between WT, OASIS−/−, and OASIS−/−;Tg cultures.

mice compared with those in WT mice while GH levels were moderately downregulated in OASIS−/− mice. The downregulation of both GH and IGF-1 in OASIS−/− mice was not rescued by OASIS expression in osteoblasts. IGF-1 is secreted in many types of cells including osteoblasts [32]. To examine whether a decrease in IGF-1 is caused by OASIS deficiency in osteoblasts, we measured IGF-1 levels in culture supernatants of primary osteoblasts isolated from WT, OASIS−/− and OASIS−/−;Tg mice (Fig. 6F). Secreted IGF-1 levels in both OASIS−/− and OASIS−/−;Tg cells were comparable with those in WT cells. These results indicated that the decrease in GH and IGF-1 levels in OASIS−/− mice was not caused by OASIS deficiency in osteoblasts. Differentiation of OASIS−/− chondrocytes In our previous study, electron microscopic analysis showed that OASIS−/− osteoblasts, but not OASIS−/− chondrocytes, exhibit

abnormal expansion of rough ER [13]. In situ hybridization for OASIS mRNA showed that OASIS is highly expressed in osteoblasts but not in chondrocytes. In contrast, RT-PCR analysis using cDNA from chondrocyte cultures showed that OASIS mRNA is detected in chondrocyte cultures [17]. Moreover, BBF2H7, which is structurally very similar to OASIS, is highly expressed in chondrocytes and is essential for chondrogenesis [17]. These data raise a possibility that OASIS deficiency in chondrocytes contributes to growth retardation in OASIS−/− mice through a defect in chondrocyte differentiation in the growth plate. To investigate this possibility, we examined the effect of OASIS deficiency on chondrocyte differentiation using micromass culture of mesenchymal cells isolated from WT, OASIS−/−, and OASIS−/−;Tg mice (Fig. 7). Alcian blue and alizarin red S stainings in OASIS−/− and OASIS−/−;Tg cells were comparable with those in WT cells at each time point, indicating that chondrocyte differentiation is normal in OASIS−/− and OASIS−/−;Tg cells.

A

B

Increase in body weight (g)

Increase in body length (cm)

C IGF- 1 (ng / ml )

1 .8

10

400 9

1 .6

300

8

control

1 .4

200

1 .2

100

7

GH

6 5 -/-

- / - ;T g

1 .0

-/-

- / - ;T g

0

-/-

- / - ;T g

Fig. 8. Growth retardation in OASIS−/− mice is improved by GH treatment. (A) An increase in body weight of WT, OASIS−/− and OASIS−/−;Tg female mice treated with GH or saline for 3 weeks is shown (n = 4). (B) An increase in body length (distance from the tip of the nose to the base of the tail) of the indicated female mice treated with GH or saline for 3 weeks is shown (n = 4). (C) Serum IGF-1 levels in the indicated female mice treated with GH or saline for 3 weeks (n = 4). Data are expressed as the mean ± SE. Statistical significance was analyzed by Student's t-tests. *p b 0.05.

522

T. Murakami et al. / Bone 48 (2011) 514–523

Mice with defects in the IGF-1 pathway exhibit growth retardation from the embryonic stage or early postnatal stage [33,34]. In contrast, mice with defects in the GH/IGF-1 pathway grow normally for 2 weeks after birth and then develop postnatal growth retardation [35,36]. Since growth retardation is observed from 3–4 weeks of age in OASIS−/− mice with the decrease in serum GH and IGF-1 levels, it is possible that this is caused by a defect in the GH/IGF-1 axis. To investigate this possibility, we examined whether growth retardation in OASIS−/− and OASIS−/−;Tg mice was rescued by treatment with GH. Three-week-old OASIS−/− and OASIS−/−;Tg mice were subcutaneously injected with 3 μg/g per day of recombinant GH for 3 weeks. We found that body weight and length in OASIS−/− and OASIS−/−; Tg mice were increased by GH treatment with an increase in serum IGF-1 levels (Fig. 8). These results indicate that growth retardation in OASIS−/− mice is caused by a defect in the GH/IGF-1 axis.

mus. OASIS has been reported to be expressed not only in osteoblasts, but also in astrocytes and several other tissues [9,38]. More detailed analyses of OASIS-expressing cells other than osteoblasts could help determine the mechanism of growth retardation in OASIS−/− mice. Further investigation using OASIS conditional knockout mice is required to elucidate the in vivo functions of OASIS in osteoblasts and other cells. In summary, osteoblast-specific expression of the OASIS transgene rescued osteopenia in OASIS−/− mice through improvement of osteoblast function and activity, whereas growth retardation in OASIS−/− mice could not be rescued by the OASIS transgene. Growth retardation in OASIS−/− mice was caused by the downregulation of serum GH and IGF-1, as indicated by the finding that GH treatment improved growth retardation. The downregulation of GH and IGF-1 in OASIS−/− mice was independent of OASIS deficiency in osteoblasts. Therefore, we conclude that OASIS regulates skeletal development by osteoblast-dependent and -independent mechanisms.

Discussion

Acknowledgments

In this study, we examined whether osteopenia in OASIS−/− mice was rescued by specific expression of OASIS in osteoblasts. Radiographic, μCT and histological analyses showed that osteoblast-specific expression of the OASIS transgene rescued osteopenia in OASIS−/− mice with an improvement in the amount of collagen fibrils in the bone matrix. Histomorphometric analysis revealed that the rescue of osteopenia by the OASIS transgene was caused by improvement of osteoblast function and activity, but was not due to inhibition of bone resorption. Moreover, OASIS expression in osteoblasts recovered the decreased expression of type I collagen mRNAs in the bone tissues of OASIS−/− mice and abnormal expansion of the rough ER in OASIS−/− osteoblasts, with improvement of accumulation of bone matrix proteins such as procollagen and osteocalcin. Taken together, these results support the findings in our previous study that OASIS contributes to the transcription of type I collagen and secretion of bone matrix proteins in osteoblasts [13]. We generated several Tg mouse lines, but expression of the OASIS transgene was detected in only one line. Bone histomorphometric parameters in OASIS−/−;Tg mice of this line were not completely improved compared with those of WT mice (Fig. 3). The reason for this finding is unclear. One possibility is that OASIS needs to be expressed at an earlier stage of osteoblast differentiation than that activated by the 2.3 kbp Col1a1 promoter used in this study, because this line sufficiently expresses exogenous OASIS compared with endogenous OASIS (Fig. 4). We found that growth retardation in OASIS−/− mice was not rescued by osteoblast-specific expression of the OASIS transgene. Serum GH and IGF-1 levels were downregulated in OASIS−/− and OASIS−/−;Tg mice compared with those in WT mice. However chondrocyte differentiation was normal in micromass cultures of mesenchymal cells isolated from OASIS−/− and OASIS−/−;Tg mice. Growth retardation in OASIS−/− mice was caused by a defect in the GH/IGF-1 axis as supported by our finding that GH treatment improved growth retardation in OASIS−/− and OASIS−/−;Tg mice with improvement of serum IGF-1 levels. Circulating levels of IGF-1, which is mainly produced in the liver, are regulated by GH [35]. GH is produced in the pituitary gland and its secretion is mainly regulated by GH-releasing hormone and somatostatin in the hypothalamus [37]. We determined OASIS expression in the liver, pituitary gland and hypothalamus. OASIS expression was not detected in the liver and its expression levels in the pituitary gland and hypothalamus were extremely low compared with those in bone tissues (data not shown; 9,38). Therefore, we speculate that the growth retardation observed in OASIS−/− mice is caused through other mechanisms that regulate the pituitary gland or hypothalamus rather than a dysfunction of the liver, pituitary gland or hypothala-

We thank Ms. Ikuyo Tuchimoti, Tomoko Kawanami and Yukiko Motoura for technical support and Dr. L. W. Fisher (National Institutes of Health, USA) for kindly providing the LF41 antibody. This work was supported by grants from the Japan Society for the Promotion of Science KAKENHI (#22020030 and #21790323), the Naito Foundation, the Sumitomo Foundation, the Mochida Memorial Foundation for Medical and Pharmaceutical Research, the Astellas Foundation for Research on Metabolic Disorders, the Uehara Memorial Foundation, the Ube Foundation and the Toray Science Foundation.

Growth retardation in OASIS−/− mice is rescued by treatment with GH

References [1] Gething MJ, Sambrook J. Protein folding in the cell. Nature 1992;355:33–45. [2] Ellgaard L, Molinari M, Helenius A. Setting the standards: quality control in the secretory pathway. Science 1999;286:1882–8. [3] Mori K. Tripartite management of unfolded proteins in the endoplasmic reticulum. Cell 2000;10:451–4. [4] Ron D. Translational control in the endoplasmic reticulum stress response. J Clin Invest 2002;110:1383–8. [5] Schroder M, Kaufman RJ. ER stress and the unfolded protein response. Mut Res 2005;569:29–63. [6] Hotamisligil GS. Endoplasmic reticulum stress and the inflammatory basis of metabolic disease. Cell 2010;140:900–17. [7] Lindholm D, Wootz H, Korhonen L. ER stress and neurodegenerative diseases. Cell Death Differ 2006;13:385–92. [8] Lisse TS, Thiele F, Fuchs H, Hans W, Przemeck GK, Abe K, et al. ER stress-mediated apoptosis in a new mouse model of osteogenesis imperfecta. PLoS Genet 2008;4: e7. [9] Honma Y, Kanazawa K, Mori T, Tanno Y, Tojo M, Kiyosawa H, et al. Identification of a novel gene, OASIS, which encodes for a putative CREB/ATF family transcription factor in the long-term cultured astrocytes and gliotic tissue. Brain Res 1999;69:93–103. [10] Kondo S, Murakami T, Tatsumi K, Ogata M, Kanemoto S, Otori K, et al. OASIS, a CREB/ATF-family member, modulates UPR signalling in astrocytes. Nat Cell Biol 2005;7:186–94. [11] Murakami T, Kondo S, Ogata M, Kanemoto S, Saito A, Wanaka A, et al. Cleavage of the membrane-bound transcription factor OASIS in response to endoplasmic reticulum stress. J Neurochem 2006;96:1090–100. [12] Nikaido T, Yokoya S, Mori T, Hagino S, Iseki K, Zhang Y, et al. Expression of the novel transcription factor OASIS, which belongs to the CREB/ATF family, in mouse embryo with special reference to bone development. Histochem Cell Biol 2001;116:141–8. [13] Murakami T, Saito A, Hino SI, Kondo S, Kanemoto S, et al. Signalling mediated by the endoplasmic reticulum stress transducer OASIS is involved in bone formation. Nat Cell Biol 2009;11:1205–11. [14] Chihara K, Saito A, Murakami T, Hino SI, Aoki Y, Sekiya H, et al. Increased vulnerability of hippocampal pyramidal neurons to the toxicity of kainic acid in OASIS-deficient mice. J Neurochem 2009;110:956–65. [15] Zhang K, Shen X, Wu J, Sakaki K, Saunders T, Rutkowski DT, et al. Endoplasmic reticulum stress activates cleavage of CREBH to induce a systemic inflammatory response. Cell 2006;124:587–99. [16] Kondo S, Saito A, Hino SI, Murakami T, Ogata M, Kanemoto S, et al. BBF2H7, a novel transmembrane bZIP transcription factor, is a new type of endoplasmic reticulum stress transducer. Mol Cell Biol 2007;27:1716–29. [17] Saito A, Hino SI, Murakami T, Kanemoto S, Kondo S, Saitoh M, et al. Regulation of endoplasmic reticulum stress response by a BBF2H7-mediated Sec23a pathway is essential for chondrogenesis. Nat Cell Biol 2009;11:1197–204.

T. Murakami et al. / Bone 48 (2011) 514–523 [18] Nagamor I, Yomogida K, Ikawa M, Okabe M, Yabuta N, Nojima H. The testesspecific bZip type transcription factor Tisp40 plays a role in ER stress responses and chromatin packaging during spermiogenesis. Genes Cells 2006;11:1161–71. [19] Martin TJ, Sims NA. Osteoclast-derived activity in the coupling of bone formation to resorption. Trends Mol Med 2005;11:76–81. [20] Matsuo K, Irie N. Osteoclast–osteoblast communication. Arch Biochem Biophys 2008;473:201–9. [21] Viguet-Carrin S, Garnero P, Delmas PD. The role of collagen in bone strength. Osteoporos Int 2006;17:319–36. [22] Dalgleish R. The human type I collagen mutation database. Nucleic Acids Res 1997;25:181–7. [23] Marini JC, Forlino A, Cabral WA, Barnes AM, San Antonio JD, et al. Consortium for osteogenesis imperfecta mutations in the helical domain of type I collagen: regions rich in lethal mutations align with collagen binding sites for integrins and proteoglycans. Hum Mutat 2007;28:209–21. [24] Morello R, Bertin TK, Chen Y, Hicks J, Tonachini L, et al. CRTAP is required for prolyl 3-hydroxylation and mutations cause recessive osteogenesis imperfecta. Cell 2006;127:291–304. [25] Sekiya H, Murakami T, Saito A, Hino SI, Tsumagari K, Ochiai K, et al. Effects of the bisphosphonate risedronate on osteopenia in OASIS-deficient mice. J Bone Miner Metab 2010;28:384–94. [26] Kashimada K, Yamashita T, Tsuji K, Nifuji A, Mizutani S, Nabeshima Y, et al. Defects in growth and bone metabolism in klotho mutant mice are resistant to GH treatment. J Endocrinol 2002;174:403–10. [27] Buul V, Brande V. The Snell dwarfmouse. I. General growth pattern, before and during growth hormone and thyroxine therapy. Acta Endocrinol 1978;89:632–45.

523

[28] Fisher LW, Stubbs JT, Young MF. Antisera and cDNA probes to human and certain animal model bone matrix noncollagenous proteins. Acta Orthop Scand Suppl 1995;266:61–5. [29] Karsenty G, Wagner EF. Reaching a genetic and molecular understanding of skeletal development. Dev Cell 2002;2:389–406. [30] Franz-Odendaal TA, Hall BK, Witten PE. Buried alive: how osteoblasts become osteocytes. Dev Dyn 2006;235:176–90. [31] Efstratiadis A. Genetics of mouse growth. Int J Dev Biol 1998;42:955–76. [32] Ohlsson C, Bengtsson BA, Isaksson OG, Andreassen TT, Slootweg MC. Growth hormone and bone. Endocr Rev 1998;19:55–79. [33] Baker J, Liu JP, Robertson EJ, Efstratiadis A. Role of insulin-like growth factors in embryonic and postnatal growth. Cell 1993;75:73–82. [34] Li Y, Iida K, O'Neil J, Zhang P, Li S, Frank A, et al. PERK eIF2alpha kinase regulates neonatal growth by controlling the expression of circulating insulin-like growth factor-I derived from the liver. Endocrinology 2003;144:3505–13. [35] Lupu F, Terwilliger JD, Lee K, Segre GV, Efstratiadis A. Roles of growth hormone and insulin-like growth factor 1 in mouse postnatal growth. Dev Biol 2001;229: 141–62. [36] Sjogren K, Bohlooly YM, Olsson B, Coschigano K, Tornell J, Mohan S, et al. Disproportional skeletal growth and markedly decreased bone mineral content in growth hormone receptor−/− mice. Biochem Biophys Res Commun 2000;267:603–8. [37] Giustina A, Veldhuis JD. Pathophysiology of the neuroregulation of growth hormone secretion in experimental animals and the human. Endocr Rev 1998;19:717–97. [38] Omori Y, Imai J, Suzuki Y, Watanabe S, Tanigami A, Sugano S. OASIS is a transcriptional activator of CREB/ATF family with a transmembrane domain. Biochem Biophys Res Commun 2002;293:470–7.