Retardation in Bone Resorption after Bone Marrow Ablation in Klotho ...

0013-7227/00/$03.00/0 Endocrinology Copyright © 2000 by The Endocrine Society

Vol. 141, No. 1 Printed in U.S.A.

Retardation in Bone Resorption after Bone Marrow Ablation in Klotho Mutant Mice* TERUHITO YAMASHITA, HIROYUKI YOSHITAKE, KUNIKAZU TSUJI, NANAKO KAWAGUCHI, YO-ICHI NABESHIMA, AND MASAKI NODA Department of Molecular Pharmacology, Medical Research Institute, Tokyo Medical and Dental University (T.Y., H.Y., K.T., N.K., M.N.), Tokyo 101-0062; the Department of Pathology and Tumor Biology, Graduate School of Medicine, Kyoto University (Y.-i.N.), Kyoto 606-8501; and the Core Research for Evolutional Science and Technology, Japan Science and Technology Corporation, Saitama 332-0012, Japan ABSTRACT The klotho gene mutant mice exhibit both osteopetrotic phenotype, including elongation of trabeculae in the epiphyses of long bones and vertebral bodies, and osteopenic phenotype, such as thin cortical bones in the diaphyses of these bones. These diverse features raise the question of whether the klotho gene defect results in alteration in bone resorption in vivo. Therefore, we examined the effect of the klotho gene defect on bone resorption by using bone marrow ablation model. At 1 week after bone marrow ablation, trabecular bones were formed in the ablated marrow cavity to levels higher than those in unablated bones in both klotho mutant and wild-type mice. At 2 weeks postsurgery, newly formed trabecular bones were resorbed in wild-type

mice to resume normal bone marrow and trabecular bone volume fraction as reported previously. In contrast, the newly formed trabecular bones in the ablated marrow in klotho mutant mice remained at levels similar to those at 1 week. The defect in the bone resorption phase in klotho mutant mice is associated with site-specific reduction of the number and size of osteoclasts in klotho mutant mice. Moreover, the expression levels of osteoprotegerin messenger RNA in the ablated femora of klotho mutant mice were higher than those in wildtype mice. These results indicate that lack of klotho gene expression suppressed bone resorption that should normally take place 2 weeks after bone marrow ablation. (Endocrinology 141: 438 – 445, 2000)

K

LOTHO IS A recently identified gene that encodes a putative type I transmembrane protein consisting of 1014 amino acid residues in mice and rats (1, 2) and 1012 amino acid residues in human (3). Notably, klotho mutant mice that are hypomorphic with regard to klotho gene expression exhibit human aging-related phenotypes, including atherosclerosis, neural degeneration, skin and gonadal atrophy, calcification of soft tissues and die within 3–4 months. These manifestations are reminiscent of the precocious aging syndrome in humans. The klotho gene is expressed at high levels in kidney (1), and it is postulated that a soluble form of Klotho protein could exist and act as a humoral factor (3, 4). Although the expression of the klotho gene has not been detected in the bones of wild-type mice even by RT-PCR analysis, the presence of abnormalities in the bones of klotho mutant mice suggests that Klotho protein could act as a humoral signaling molecule. Although klotho mutant mice were first reported to

exhibit osteopenia in the diaphyseal cortical bones (1), the presence of osteopetrotic phenotype, including high levels of trabecular bones in the epiphyseal and metaphyseal regions, in klotho mutant mice (5, 6) indicates that the effect of the lack of Klotho protein may not be a simple reduction of the levels of overall bone mass. These diverse features raised the question of whether the klotho gene defect affects bone resorption in vivo. Although it is desirable to examine the direct effects of Klotho protein on osteoclasts, difficulty in obtaining the Klotho protein hampered this direction of research. Therefore, we were forced to use in vivo assays to evaluate the effect of the klotho gene defect on bone resorption. Bone marrow ablation in the long bones of rodents causes vigorous new bone formation within the first week and then subsequent rapid bone resorption in the second week to regenerate bone marrow with normal levels of trabecular bones. It is a highly reproducible in vivo assay to evaluate bone formation and resorption (7–9). We found that bone resorption that occurred in the second week postsurgery was significantly impaired in klotho mutant mice, indicating the presence of a defect in bone resorption in vivo. We also observed site-specific morphological changes in osteoclasts and high levels of osteoprotegerin (OPG)/osteoclastogenesis inhibitory factor (OCIF) messenger RNA (mRNA) expression in the ablated femora in klotho mutant mice.

Received July 9, 1999. Address all correspondence and requests for reprints to: Dr. Masaki Noda, Department of Molecular Pharmacology, Medical Research Institute, Tokyo Medical and Dental University, 3–10, Kanda-Surugadai 2-Chome, Chiyoda-ku, Tokyo, 101-0062, Japan. E-mail: [email protected]. * This work was supported by grants-in-aid from the Japanese Ministry of Education (11877357, 11152209, 11770791, 10044246, and 0930734) grants from Core Research for Evolutional Science and Technology of Japan Science and Technology Corp. (to J.S.T.), a grant from the Research for the Future Program of the Japan Society for the Promotion of Science (96100205), and grants from Traffic Medicine Foundation, Interdisciplinary Cancer Research Foundation, Inamori Foundation, Multidisciplinary Cancer Research Foundation, National Space Development Agency of Japan, and Marine and Fire Insurance Foundation.

Materials and Methods Animals Homozygous klotho mutant mice and control littermates were obtained by mating heterozygous klotho mutant mice (C3H) that were established as previously described (1).

438

BONE RESORPTION DEFECT AFTER MARROW ABLATION Marrow ablation procedure Either klotho mutant homozygotes or wild-type mice (4 – 6 weeks old) were used for experiments. At least four animals were used per group. Bone marrow was ablated in the right femora of each animal as follows. Under general anesthesia, a hole was made in the intercondylar regions of the femora by inserting a 26-gauge needle. Then the content of the bone marrow was removed by using no. 10 –30 dental files in a gradient manner and finally by inserting a 0.6-mm diameter Kirschner wire up to the proximal ends of the femora to ensure completion of marrow ablation. The left femora were untreated and were used as an internal control. Animals were killed either at the time of surgery or 1 and 2 weeks postsurgery, and soft x-ray radiographs of the bones were taken before processing the bones for histological examination. Half of the bones were subjected to histological examination, and the other half were used for RNA preparation.

Tissue preparation and RNA extraction Femora were fixed with 4% paraformaldehyde in PBS (pH 7.4) and were subsequently subjected to microcomputed tomography (microCT) examination before histological analysis. For RT-PCR analysis, femora were collected before operation and at 1 and 2 weeks postsurgery, and marrow tissues were flushed out from them to extract total RNA according to the acid guanidine isothiocyanate-phenol/chloroform method.

MicroCT analysis The bones were subjected to microCT analysis using Musashi (model NXCP-C80, NS-ELEX, Tokyo, Japan). This apparatus is equipped with a microfocus x-ray tube with a spot size of 6 3 8 mm. Analysis was conducted at 40 kVp and 100 mA to obtain the best contrast between bones and soft tissues. The digital data were reconstructed to obtain CT images in 1024 3 1024 pixel matrices. The resolution was approximately 11 mm. Trabecular bone volume in a square area of 30 3 300 pixels (0.28 3 2.8 mm) in the metaphyseal region of the femora was quantitated using the Luzex-F Image analyzing system (NIRECO, Tokyo, Japan).

Histological analysis After microCT examination, femora were decalcified in 10% EDTA (pH 7.4), dehydrated, and then embedded in glycol methacrylate. Fivemicron thick sections were prepared and stained for tartrate-resistant acid phosphatase (TRAP) followed by staining with toluidine blue. Histomorphometry was conducted to quantify the number of osteoclasts and osteoclast surface as defined by Parfitt et al. (10). The size and area of more than 60 osteoclasts were measured using the sections containing TRAP-positive MNCs. Ferret’s minimum distance between two parallel lines was used as an indicator of the height of osteoclasts. To evaluate osteoclast number and size, two square areas (230 3 230 mm/area) in each of the three separate regions were examined as described below. The first region was adjacent to the growth plate (GP), being 100 mm away from the boundary between hypertropic chondrocytes and ossified matrix. The second and third regions were 800 and 2000 mm away from this boundary respectively.

Semiquantitative RT-PCR analysis RT-PCR analysis was performed using primers specific to OPG/OCIF and RANKL/TRANCE/ODF/OPGL genes and one tube RT-PCR system (Roche Molecular Biochemicals, Mannheim, Germany). The first DNA synthesis was conducted using AMV reverse transcriptase at 55 C for 30 min followed by denaturation at 94 C for 30 sec, annealing at 55 C for 30 sec, and polymerization at 68 C for 45 sec. Subsequently, the amplification cycle was repeated 30 –38 times. The cycle number was determined so that the PCR product levels were amplified within a linear range. Ethidium bromide-stained DNA bands were quantitated using an image analyzer Bio-1D system (VILBER LOURMAT, France). As a control, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA levels were also estimated by RT-PCR. The levels of OPG/OCIF or RANKL/TRANCE/ODF/OPGL relative to GAPDH expression levels were calculated and compared between klotho mutant and wild-type mice.

439

Oligomer sets used for OPG/OCIF, RANKL/TRANCE/ODF/OPGL and GAPDH were 59-GCA CAT TTG GCC TCC TGC TAA TTC-39 (forward, OPG) and 59-ACT CTC GGC ATT CAC TTT GGT CCC-39 (reverse, OPG), 59-TCA TCT CTG TGG TAG TAG TGG CTG-39 (forward, RANKL) and 59-TTA GGA GCA GTG AAC CAG TCG AAG-39 (reverse, RANKL), and 59-ACC ACA GTC CAT GCC ATC AC-39 (forward, GAPDH) and 59-TCC ACC ACC CTG TTG CTG TA-39 (reverse, GAPDH), respectively.

Statistical evaluation The results were presented as the mean 6 sem. Comparison between the values for ablated and control bones in individual animals were evaluated using paired Wilcoxon’s signed rank test. The mean values at each time point were analyzed with Mann-Whitney’s U test. P , 0.05 was considered significantly different.

Results New trabecular bone formation is normal in klotho mutant mice, whereas subsequent bone resorption is impaired after bone marrow ablation in klotho mutant mice

One week after marrow ablation, soft x-ray pictures demonstrated an increase in the radiopacity in the ablated bone marrow cavity in the right femur in both klotho mutant and wild-type mice compared with the pictures of the corresponding regions in the bones taken at the time of ablation (data not shown). Two weeks after ablation, radiographs of the right femora in wild-type mice became indistinguishable from those in nonoperated left femora due to the reduction of the radiopacity in the ablated marrow as reported previously (7–9), whereas the ablated femora of the klotho mutant mice still remained to show high radiopacity especially at the distal ends (data not shown). The changes in radiopacity could be due to the alteration in cortical bones, trabecular bones, or both. To visualize the changes in trabecular bone structures and those in cortical bone thickness in the femora after bone marrow ablation, we conducted microCT in the epiphyseal to metaphyseal regions of femora within the midsagittal plane. MicroCT examination demonstrated that ablated bone marrow cavity of the right femora (Fig. 1, B and F) was filled with newly formed trabecular bones that were more abundant than the trabecular bones in untreated (left) femora (Fig. 1, A and E) at 1 week postsurgery in both klotho mutant (Fig. 1, B and A) and wild-type mice (Fig. 1, F and E), respectively. Cortical bone thickness was not altered by bone marrow ablation in both klotho mutant (Fig. 1, B compared with A) and wild-type mice (Fig. 1, F compared with E), indicating that the major reason for the increase in radiopacity is due to trabecular bone formation in the ablated marrow space. By 2 weeks after operation, trabecular bones in the ablated femora of wildtype mice were resorbed, and the pattern and amount of these bones were similar between control and ablated femora (Fig. 1, G vs. H) as reported previously. In contrast, ablated (right) femur of klotho mutant mice even at 2 weeks postsurgery showed abundant trabecular bones compared with nonablated (left) femora (Fig. 1, D vs. C). Cortical thickness again was not altered in either (klotho mutant and wild-type) genotype at 2 weeks after bone marrow ablation (Fig. 1, D and H) and was similar to the cortical morphology at 1 week (Fig. 1, B vs. F). Quantitation of the trabecular bones on the microCT sec-

440

BONE RESORPTION DEFECT AFTER MARROW ABLATION

Endo • 2000 Vol 141 • No 1

FIG. 1. Two-dimensional microCT analysis. Newly formed trabecular bones remained in the diaphyseal region of the ablated femur in klotho mutant mice at 2 weeks after operation (D vs. H). A–D, The klotho mutant mice; E–H, wildtype littermates. Left panels, One week after surgery; right panels, 2 weeks after surgery.

tions indicated that basal levels (before surgery) of bone volume/tissue volume (BV/TV) were about 15% and 7% in klotho mutant and wild-type mice, respectively (Fig. 2A). One week after bone marrow ablation, BV/TV levels in the right femora were increased to about 23% and 22% in klotho mutant mice and wild-type mice, respectively (Fig. 2B, closed bars). By 2 weeks after operation, BV/TV in the ablated side in wildtype (Fig. 2C, 1/1, closed bar) was reduced to about 8%, a level similar to that in the control (left) side (Fig. 2C, 1/1, open bar). In contrast, in klotho mutant mice, the trabecular bone volume in the right femora still remained about 23% even at 2 weeks postsurgery (Fig. 2C, kl/kl, closed bar). This BV/TV value of the ablated femur of klotho mutant mice was significantly higher than that in the left femora (Fig. 2C, kl/kl, open bar; P , 0.05). The klotho mutant mice revealed site-specific reduction in the number and size of TRAP-positive multinucleated cells (MNCs)

Subsequent to our initial observation that osteoclast number is reduced in intact klotho mutant mice (5), we discovered that reduction of osteoclast number is only observed at the ends of trabecular bones in the regions away from the GP and not in the trabecular bones close to GP (Yamashita, T., and M. Noda, unpublished data). We, therefore, examined the number and shape of the osteoclasts in the trabecular bones

in the ablated marrow in three discrete regions that differ in their distance by 100, 800, and 2000 mm away from the GP. For examination of osteoclast number and shape, we chose day 8 postsurgery for the morphological evaluation, because it was hard to investigate the three regions immediately after or 2 weeks after surgery in wild-type mice due to the absence by removal or disappearance of proper amount of trabecular bones, respectively. The numbers of TRAP-positive MNCs on the surface of trabecular bones (N.Oc/BS) in klotho mutant mice in the regions 100 or 800 mm away from the GP were slightly lower (20% and 35%, respectively) than those in wild-type mice, but the difference was not statistically significant. In contrast, TRAP-positive MNCs number (N.Oc/ BS) in klotho mutant mice was reduced by more than 80% in the regions 2000 mm away from the GP in the marrow of the ablated femur on day 8 compared with that in wild-type mice (Table 1A; N.Oc/BS, 1.4 6 1.4 cells/mm in klotho mutant vs. 8.8 6 1.5 cells/mm in the wild-type; P , 0.05). Osteoclast surface was also reduced in klotho mutant mice in a sitespecific manner in the region 2000 mm away from the GP (Table 1B; Oc.S/BS, 2.9 6 2.9 in klotho mutant vs. 24.6 6 5.1 in wild-type mice; P , 0.05), as it was not observed at the 100 mm region. These data indicated that osteoclast number in klotho mutant mice was reduced in a site-specific manner, at least at 8 days postsurgery. Moreover, histological examination also uncovered that


441

FIG. 2. Quantitation of trabecular bone volume fraction (BV/TV) in the metaphyseal regions of femora based on two-dimensional microCT analysis. Trabecular bone volume in the femora of klotho mutant mice was 2-fold higher than that in wild-type mice even before surgery. After 1 week of operation, new trabecular bone filled the cavity of ablated femur (right columns) to similar degrees in wild-type and klotho mutant mice (B). Two weeks after marrow ablation (C), the bone volume fraction of the ablated femora in wild-type mice was reduced to levels similar to those in the nonoperated femora. On the other hand, newly formed trabecular bones in ablated femora of klotho mutant mice remained at high levels. TABLE 1. The osteoclast number (N.Oc/BS) and osteoclast surface (Oc.S/BS) in the trabecular bones in the regions 100, 800, and 2000 mm away from the growth plate of the ablated femora of klotho mutant and wild-type mice Genotype

A) N.Oc/BS (cells/mm trabecular bone perimeter) 1/1 (n 5 5) kl/kl (n 5 4) B) Oc.S/BS [ratio of the length of osteoclast surface against the trabecular bone surface (%)] 1/1 (n 5 5) kl/kl (n 5 4)

Nonablateda 100 mm

Within ablated area 100 mm

800 mm

2000 mm

6.2 6 1.4 6.7 6 1.3

8.5 6 2.9 6.8 6 1.6

6.8 6 0.1 4.4 6 0.1

8.8 6 1.5 1.4 6 1.4b

18.0 6 4.3 20.9 6 7.8

31.3 6 2.6 25.1 6 1.9

19.2 6 4.0 11.6 6 2.9

24.6 6 5.1 2.9 6 2.9b

Values indicate the mean 6 SEM. The numbers in parentheses indicate those of bone samples. a The region 100 mm away from the growth plate in the left femur where ablation was not conducted. b The difference in the values between wild-type (1/1) and klotho mutant mice (kl/kl) is statistically significant (P , 0.05).

the osteoclast morphology in vivo was also altered in klotho mutant mice in a similar site-specific manner. Morphologies of osteoclasts adjacent to (100 mm away from) the GP were similar between klotho mutant and wild-type mice (Fig. 3, B and F, respectively). Quantitation also indicated a similar area size (the area of individual TRAP-positive MNCs was measured) and height (maximum height of the osteoclasts in the axis vertical to the bone surface) of osteoclasts in the region 100 mm away from GP in wild-type and klotho mutant mice (Fig. 4, 100 mm away from the GP). In contrast, the mean size and height of osteoclasts in the region 2000 mm away from the growth plate in klotho mutant mice were reduced by 42% (P , 0.05) and 60% (P , 0.05), respectively (Fig. 4; 2000 mm away from the GP), compared with those of the osteoclasts in the corresponding regions in wild-type mice. Intermediate levels of reduction in osteoclast size and height were seen in the region 800 mm away from GP in klotho

mutant compared with those in wild-type mice (Fig. 4; 800 mm away from the GP). The klotho mutant mice show high expression levels of OPG/OCIF mRNA in the bone marrow tissue

To examine whether klotho deficiency affects the regulation of molecules related to osteoclastogenesis and osteoclastic function, we examined the expression levels of OPG/OCIF and RANKL/TRANCE/ODF/OPGL mRNA by semiquantitative RT-PCR analysis using total RNA extracted from the bone marrow tissues of the ablated (right) femora as well as untreated (left) femora. The time courses of the expression of OPG/OCIF or RANKL mRNAs were examined by determining the relative levels of expression calculated based on the ratios of the amplified products of OPG/OCIF or RANKL against those of GAPDH (Fig. 5). In the preoperation period,

442


Endo • 2000 Vol 141 • No 1

FIG. 3. Site-specific reduction of the number and size of TRAP-positive multinucleated cells in the epiphyseal/metaphyseal regions of the ablated femora in klotho mutant mice. At 8 days after surgery, sections are prepared and stained for TRAP. TRAP-positive MNCs in klotho mutant mice (A–D) and wild-type mice (E–H) are shown. In the marrow of the ablated femora of klotho mutant mice, fewer TRAP-positive MNCs were observed in the regions 2000 mm away from GP of the distal end of the femur (D) compared with TRAP-positive MNCs in the corresponding regions of the wild-type mice (H). The numbers of osteoclasts 100 mm away from the GP in klotho mutant and wild-type mice (B vs. F, respectively) were similar. Bar: A and E, 400 mm; B–D and F–H, 20 mm. Arrows indicate osteoclasts.

OPG/OCIF mRNA levels in klotho mutant mice were more than 50% of those in the wild-type (kl/kl, 1.56 6 0.15; wild-type, 1.00 6 0.09; P , 0.05). At 1 week postsurgery, both klotho mutant mice and wild-type resumed trabecular bones and marrow tissues in the ablated (right) femora. These bone marrow tissues in klotho mutant and wild-type expressed slightly more OPG/OCIF mRNA at 1 week (kl/ kl, 2.07 6 0.64; wild-type, 1.44 6 0.06; Fig. 5B) than presurgery levels. Two weeks after ablation, OPG/OCIF gene expression in the marrow tissues of the ablated (right) femur of klotho mice stayed at high levels (2.15 6 0.47), whereas in wild-type mice OPG/OCIF expression levels were reduced significantly (0.51 6 0.22; Fig. 5B). We also examined RANKL mRNA expression levels and found that they were similar throughout these experiments (Fig. 5, C and D). Discussion Impairment in trabecular bone reduction and high OPG/ OCIF expression levels in bone marrow ablation experiments in klotho mutant mice

The previous observations that klotho mutant mice exhibit both osteopenia in the cortical bones in diaphyses (1) and osteopetrosis in the epiphyses/metaphyses (5, 6) raised the

question of whether klotho deficiency affects bone resorption. In this report we clearly demonstrated that bone resorption activity in vivo in klotho mutant mice is significantly impaired when examined by bone marrow ablation experiments. Furthermore, our data indicated that OPG/OCIF expression levels in bone marrow tissues in klotho mutant mice are consistently higher than those in wild-type mice during the 2-week time course. To our knowledge this is the first study to examine the dynamic changes in the levels of OPG/OCIF expression in bone marrow ablation experiments even in wild-type mice. In wild-type mice, vigorous bone formation and then resorption take place in the first and second weeks, respectively, after bone marrow ablation, and these distinct periods coincide with the over 40% elevation (statistically significant, P , 0.05) of the OPG/OCIF expression levels and then 50% reduction (statistically significant, P , 0.05) relative to the presurgery period, respectively. The increase in OPG expression in the wild-type mice at 1 week after ablation in the presence of no change in RANKL expression is interesting. We do not have any explanation for this mechanism at this point. It might reflect an increase in the OPG-expressing cell population or enhancement of OPG expression by injuryrelated cytokines in the healing marrow, which has relatively


FIG. 4. Quantitation of the number and size of the osteoclasts in the ablated femora of klotho mutant mice. The number and size of the osteoclasts in the femora of klotho mutant mice were quantified as described in Materials and Methods. The asterisk indicates that the difference is statistically significant (P , 0.05).

high cellularity and may be rich in injury-related cytokines at this time point. After bone marrow ablation in klotho mutant mice, OPG/ OCIF expression was increased by about 25% (from 1.56 6 0.15 to 2.07 6 0.64) in the first week. However, a striking difference compared with the wild-type mice was the absence of the reduction in the levels of OPG/OCIF expression (2.15 6 0.47) in the second week. RANKL (TRANCE/ODF/ OPGL) expression levels, on the other hand, were not altered in klotho mutant and wild-type mice during the 2-week course of the bone marrow ablation experiments. These data indicated clearly the abnormal regulation of OPG/OCIF expression in klotho mutant mice. In addition, our data indicated a correlation between OPG/OCIF levels and those in bone resorption in the bone marrow ablation model. Klotho protein is predicted to be an enzyme-like molecule; however, its direct function is not known. Our data indicated for the first time that at least in bone, klotho mutation results in an increase in the levels of an important cytokine regulator, OPG/OCIF. Site-specific reduction in the number and size of osteoclasts in klotho mutant mice

Quantitation of osteoclast number in the three distinct regions of the epiphyseal trabecular bones indicated that klotho mutant mice showed interesting site-specific reduction in osteoclast number in the most distal region relative to the GP. This could explain at least in part why klotho mutant mice show elongation of the trabecular bones. This site-specific reduction of osteoclast number was not observed in wildtype mice, as the numbers of osteoclasts in the three regions

443

are virtually similar (100 mm, 8.5 6 1.3; 800 mm, 6.8 6 0.1; 2000 mm, 8.8 6 1.5). These osteoclast numbers were similar to the number in the nonablated side in wild-type mice (6.2 6 1.4). We also examined osteoclast number in the nonablated bones in klotho mice, and our data reveal that the dependence of the reduction in osteoclast number on the relative distance from the growth plate can be observed in nonablated bones in klotho mice but not in wild-type animals (data not shown), confirming that retardation of the bone resorption associated with klotho mutant abnormality with regard to the number of osteoclasts is a specific feature of the klotho mutant mice. Osteoclastogenesis in the bone marrow after bone marrow ablation is clearly a local phenomenon, and hence, it is certainly possible that a study of osteoclastogenesis in response to some local factors might reveal a defect. An in vitro osteoclastogenesis study has shown that not only systemic factors such as vitamin D and PTH, but also local factors such as interleukin-1, interleukin-6 (including other gp130-dependent cytokines), PGE2, and tumor necrosis factor are capable of promoting osteoclastogenesis. It appears, therefore, that many, if not all, of these stimuli eventually act to stimulate expression of RANKL. RANKL functions as a local factor that directly binds to its receptor, RANK, being expressed on the surface of the progenitor cells for osteoclasts. We, therefore, examined the levels of RANKL in the ablated marrow. However, as shown in the RT-PCR analyses, RANKL expression levels were not altered significantly. These observations suggest that certain local factors independent from RANKL may be involved in the reduction of bone resorption in klotho mutant mice. High levels of OPG/OCIF expression in klotho mutant mice compared with wild-type mice could partly explain the reduction in the number and size of osteoclasts in klotho mutant mice. However, as OPG/OCIF appears to be expressed in both skeletal and nonskeletal tissues, and it could act as a humoral factor systemically (11, 12), site-specific reduction in osteoclastic number and size in the klotho mutant mice cannot be explained by the increase in OPG/OCIF expression measured using the entire marrow tissues to give averaged values (including all three locations). In fact, our preliminary experiments indicated that OPG expression levels in lung and spleen of klotho mutant mice were higher than those in wild-type animals (data not shown). Questions regarding whether OPG/OCIF could act somehow site specifically in klotho mutant mice or whether OPG/ OCIF expression is also site specific in the bone marrow tissues depending on the distance from the growth plate in these mutant, but not in wild-type, mice require future investigation. There is also a possibility that if more bone is being formed in the klotho mutant mice than in wild-type mice at 2 weeks, less of the surface may be available for resorption and so account for the reduced osteoclast number. To address this point, we measured osteoblast surface in all three sites. The results indicated that osteoblast surface values at 100, 800, and 2000 mm away from the GP in wild-type mice (n 5 5) were 44.8 6 3.5%, 69.9 6 13.3%, and 57.4 610.2%, respectively, and those in klotho mutant mice (n 5 4) were 25.9 6 7.7%, 50.9 6 16.5%, and 81.3 6 12.4% respectively. The ratios of klotho mice over wild-type mice at each of the three regions, 100, 800, and 2000 mm away from the GP, were 0.58,

444


Endo • 2000 Vol 141 • No 1

FIG. 5. OPG/OCIF mRNA expression in the marrow tissues after ablation is high in klotho mutant mice. OPG/OCIF expression levels relative to GAPDH mRNA levels in the ablated femora were determined by semiquantitative RT-PCR as described in Materials and Methods. A statistically significant difference is indicated by # (P , 0.05). Closed circles, The klotho mutant mice; open circles, wild-type mice.

0.73, and 1.42, respectively. However, statistical evaluation did not show that the difference was statistically significant. Site-specific reduction of osteoclast number and size is a unique phenomena in klotho mutant mice. This type of site specificity has not been described in other osteopetrotic mice or in drug-induced osteopetrosis. Therefore, our observation of site-specific phenomena regarding osteoclasts may give a clue to understanding of the molecular function of the Klotho protein. Trabecular bone formation after bone marrow ablation is normal in klotho mutant mice

In this paper we report that a bone marrow ablation study in klotho mutant mice revealed relatively normal bone formation in the first week before the observation of impaired bone resorption in the second week. After ablation of bone marrow, bone marrow tissues regenerate in a highly reproducible manner starting with hematoma formation just after operation. One week after marrow ablation, trabecular bone volume reaches at its maximal level, and then vigorous bone resorption takes place within the second week. Two weeks later, bone marrow in the ablated bones returns to normal morphology that is indistinguishable from that in untreatedside femur in wild-type mice (7, 8). As mentioned, at least in the marrow ablation experiments, osteoblastic function in klotho mutant mice is normal. However, in the initial examinations (1, 5, 6), the cortical bones in klotho mutant mice were osteopenic. It appears, therefore, that the loss of klotho gene expression would not affect rapid bone formation, such as that taking place in the marrow ablation study, but it could affect slow or long term bone formation and/or remodeling to maintain steady state levels of cortical bones. This possibility still does not contradict our observation that OPG/OCIF expression levels are

relatively high in klotho mutant mice, as OPG/OCIF per se does not influence osteoblastic function directly. It has been proposed that osteoblasts and osteoclasts activate the functions of each other. We propose that klotho deficiency enhances OPG/OCIF expression levels for a long term, and these sustained high levels of OPG/OCIF might suppress osteoclasts, resulting in the long term reduction of as an yet unknown osteoclast-derived signal(s) that might, in turn, activate osteoblastic function. If this hypothesis is true, certain osteoblastic functions, such as those in the normal remodeling phase under the control of the signals elicited by osteoclasts, would be affected, whereas other osteoblastic functions unrelated to the influence from osteoclastic activity, such as the formation of new trabecular bones in the first week of the recovery phase after bone marrow ablation, would not be affected by the loss of klotho gene expression. High osteoprotegerin expression levels in klotho mutant mice

OPG/OCIF has been shown to be an inhibitor of osteoclastogenesis and osteoclastic function (11–14). However, its expression has been observed in many nonskeletal tissues, such as kidney, heart, placenta, liver, and intestine, in addition to bone, cartilage, and bone marrow (11). As the klotho gene is not expressed in bone but is highly expressed in kidney and to a lesser extent in other soft tissues, it is intriguing to determine whether the link between klotho expression and OPG/OCIF expression can be observed in one of these extraskeletal tissues, such as kidney, in that Klotho protein might suppress OPG/OCIF expression in wild-type animals. We are currently trying to obtain Klotho protein to investigate its direct action on the regulation of OPG/OCIF expression. Alternatively, OPG/OCIF expression is regulated by vitamin D3 (suppression) or ionized calcium and


estrogen (activation). We cannot exclude the possibility that the klotho gene product may also act through modulation of such regulators as indirect pathways. As klotho mutant mice show defects in gonadal development and possibly are in low estrogen status, OPG/OCIF up-regulation would not occur directly through estrogen regulation. Future studies should elucidate the mechanism of the enhancement of OPG/OCIF expression by the loss of the product of the klotho gene. Acknowledgments We thank Drs. Makoto Kuro-o, Hidehiro Ozawa, and Norio Amizuka for their helpful discussion.

References 1. Kuro-o M, Matsumura Y, Aizawa H, Kawaguchi H, Suga T, Utsugi T, Ohyama Y, Kurabayashi M, Kaname T, Kume E, Iwasaki H, Iida A, Shirakiiida T, Nishikawa S, Nagai R, Nabeshima Y 1997 Mutation of the mouse klotho gene leads to a syndrome resembling ageing. Nature 390:45–51 2. Ohyama Y, Kurabayashi M, Masuda H, Nakamura T, Aihara Y, Kaname T, Suga T, Arai M, Aizawa H, Matsumura Y, Kuro-o M, Nabeshima Y, Nagail R 1998 Molecular cloning of rat klotho cDNA: markedly decreased expression of klotho by acute inflammatory stress. Biochem Biophys Res Commun 251:920 –925 3. Matsumura Y, Aizawa H, Shiraki-Iida T, Nagai R, Kuro-o M, Nabeshima Y 1998 Identification of the human klotho gene and its two transcripts encoding membrane and secreted klotho protein. Biochem Biophys Res Commun 242:626 – 630 4. Shiraki-Iida T, Aizawa H, Matsumura Y, Sekine S, Iida A, Anazawa H, Nagai R, Kuro-o M, Nabeshima Y 1998 Structure of the mouse klotho gene and its two transcripts encoding membrane and secreted protein. FEBS Lett 424:6 –10 5. Yamashita T, Nifuji A, Furuya K, Nabeshima Y, Noda M 1998 Elongation of the epiphyseal trabecular bone in transgenic mice carrying a klotho gene locus mutation that leads to a syndrome resembling aging. J Endocrinol 159:1– 8

445

6. Yamashita T, Nabeshima Y, Noda M, High-resolution uCT (micro-computed tomography) analyses of the abnormal trabecular bone structures in klotho gene mutant mice. J Endocrinol, in press 7. Suva LJ, Seedor JG, Endo N, Quartuccio HA, Thompson DD, Bab I, Rodan GA 1993 Pattern of gene expression following rat tibial marrow ablation. J Bone Miner Res 8:379 –388 8. Tanaka H, Barnes J, Liang CT 1996 Effect of age on the expression of insulinlike growth factor-I, interleukin-6, and transforming growth factor-b mRNAs in rat femurs following marrow ablation. Bone 18:473– 478 9. Shimizu T, Mehdi R, Yoshimura Y, Yoshikawa H, Nomura S, Miyazono K, Takaoka K 1998 Sequential expression of bone morphogenetic protein, tumor necrosis factor, and their receptors in bone-forming reaction after mouse femoral marrow ablation. Bone 23:127–133 10. Parfitt AM, Mathews CH, Villanueva AR, Kleerekoper M, Frame B, Rao DS 1983 Relationships between surface, volume, and thickness of iliac trabecular bone in aging and in osteoporosis. Implications for the microanatomic and cellular mechanisms of bone loss. J Clin Invest 72:1396 –1409 11. Simonet WS, Lacey DL, Dunstan CR, Kelley M, Chang MS, Luthy R, Nguyen HQ, Wooden S, Bennett L, Boone T, Shimamoto G, DeRose M, Elliott R, Colombero A, Tan HL, Trail G, Sullivan J, Davy E, Bucay N, Renshaw-Gegg L, Hughes TM, Hill D, Pattison W, Campbell P, Boyle WJ 1997 Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89:309 –319 12. Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Mochizuki S, Tomoyasu A, Yano K, Goto M, Murakami A, Tsuda E, Morinaga T, Higashio K, Udagawa N, Takahashi N, Suda T 1998 Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95:3597–3602 13. Yasuda H, Shima N, Nakagawa N, Mochizuki SI, Yano K, Fujise N, Sato Y, Goto M, Yamaguchi K, Kuriyama M, Kanno T, Murakami A, Tsuda E, Morinaga T, Higashio K 1998 Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): a mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 139:1329 –1337 14. Hakeda Y, Kobayashi Y, Yamaguchi K, Yasuda H, Tsuda E, Higashio K, Miyata T, Kumegawa M 1998 Osteoclastogenesis inhibitory factor (OCIF) directly inhibits bone- resorbing activity of isolated mature osteoclasts. Biochem Biophys Res Commun 251:796 – 801