GPR30 Estrogen Receptor Expression in the Growth Plate Declines as Puberty Progresses

Abstract

Objective: Our objective was to study whether GPR30 is expressed in the epiphyseal growth plate and its potential role as a modulator of pubertal growth.

Background: Estrogens play a crucial role in the regulation of skeletal maturation and longitudinal bone growth. We have previously shown that both estrogen receptors (ERs) α and β are expressed in the human epiphyseal growth plate. Recently, a membrane-bound ER referred to as GPR30 was discovered, but the role played by this receptor in the regulation of longitudinal bone growth is not yet known.

Patients/Methods: Biopsies were collected from the tibial growth plates of 14 boys and seven girls that underwent epiphyseal surgery to arrest longitudinal bone growth. The patients were in different stages of puberty and suffered from inequality in leg length or extreme tall stature. Paraffin-embedded sections of the growth plates were used to detect expression of the GPR30 protein.

Results: The highest level of GPR30 expression was observed in hypertrophic chondrocytes, although cells with positive immunostaining were also detected in the resting zone. In contrast, no immunoreactivity was found in the proliferative zone. During pubertal progression there was a clear decline in the level of GPR30 expression in both boys and girls.

Conclusions: The novel ER GPR30 is expressed in the human growth plate, and the level of expression declines during pubertal progression. Although a relationship between GPR30 expression and age may underlie the observed pubertal decline in the GPR30 level, our observations suggest that this receptor could be involved in the modulation of longitudinal bone growth during puberty.

REGULATION OF LONGITUDINAL bone growth and skeletal maturation in humans by estrogens has been extensively characterized and applied in clinical practice for many years now (1). The phenotypes of a few rare individuals, i.e. one man carrying a mutation that inactivates the estrogen receptor (ER) α (2) and a few men and women with a deficiency in aromatase activity (3, 4), have revealed definitively that estrogens play an important role in the pubertal growth spurt and epiphyseal closure. Three ERs, i.e. ERα, ERβ, and the recently discovered membrane-bound ER GPR30 (5, 6), have been identified to date, and in connection with the development of novel selective ER modulators (SERMs) for clinical application, it is necessary to know which of these ERs is expressed in the cartilage of the epiphyseal growth plate. We have previously demonstrated that both ERα and ERβ are expressed in epiphyseal cartilage throughout puberty (7, 8), but it is not yet known whether the newly discovered ER GPR30 is also expressed in the human growth plate during pubertal development.

Patients and Methods

Patients and tissue preparation

Tissue samples were taken from 14 boys and seven girls subjected to epiphyseal surgery at different stages of puberty to arrest longitudinal leg growth in patients suffering from an inequality in leg length or extreme tall stature (Table 1). In connection with this surgery, biopsies from the tibial and femoral growth plates were obtained using a bone marrow biopsy needle (8 gauge; Gallini Medical Products and Services, Modena, Italy), and subsequently fixed in 4% formaldehyde, decalcified, and embedded in paraffin exactly as described previously (8). The pubertal staging of the patients was performed by a trained pediatric endocrinologist. This study was preapproved by the local ethics committee at Karolinska Hospital, Stockholm, Sweden, and informed consent was obtained from the patient and both parents.

TABLE 1.

Background characteristics and levels of GPR30 expression in our subjects

Patient no.	Sex	Age (yr:months)	Tanner pubertal stage	GPR30 in the resting zone (% positive cells)	GPR30 in the hypertrophic zone (% positive cells)	Diagnosis
1	F	9:6	B1	26.4	54.3	Leg-length difference
2	F	12:7	B1–2	14.9	36.0	Leg-length difference
3	F	12:8	B2	3.9	42.7	Leg-length difference
4	F	12:6	B2	10.1	27.1	Constitutional tall stature
5	F	14:2	B2–3	5.5	30.0	Leg-length difference
6	F	13:8	B3	6.8	5.0	Leg-length difference and CHL
7	F	12:6	B5	6.7	nd	Leg-length difference
8	M	12:3	G1	25.2	32.5	Marfan syndrome
9	M	12:3	G3	2.4	4.5	Leg-length difference and FF
10	M	14:3	G3	0.7	27.5	Leg-length difference
11	M	14:1	G3	9.8	51.0	Constitutional tall stature
12	M	15:4	G3–4	7.5	52.0	Leg-length difference
13	M	14:7	G3–4	17.8	51.0	Leg-length difference
14	M	16:0	G4	12	9.8	Leg-length difference and FF
15	M	13:5	G4	5.2	3.0	Leg-length difference
16	M	14:7	G4	12.4	10.2	Leg-length difference and FTFF
17	M	16:4	G4	11.1	45.4	Constitutional tall stature
18	M	14:3	G4–5	0.5	2.6	Leg-length difference and FF
19	M	14:0	G4–5	6.7	16.8	47, XYY syndrome
20	M	14:1	G4–5	2.6	15.2	Leg-length difference
21	M	12:2	G5	7.6	35.4	Klinefelter’s syndrome

CHL, Congenital hip luxation; F, female; FF, femur fracture; FTFF, femur, tibia, and fibula fractures; M, male; nd, not determined.

Immunohistochemistry

Immunohistochemical localization of GPR30 was performed in demineralized, deparaffinized sections of cartilage from human tibial and femoral growth plates. For this purpose, intrinsic peroxidase activity was inhibited by incubation with 3% H₂O₂ for 10 min and nonspecific binding blocked by incubation with 3% goat serum. Antigen retrieval was achieved in citrate buffer (pH 6.0) for 15 min at 95 C and the GPR30 protein detected by incubation with antiserum from rabbit immunized with human GPR30 [diluted 1:1500; a kind gift from Dr. E. Prossnitz, University of New Mexico, Albuquerque, NM (5)] overnight at +4 C. Thereafter, bound primary antibodies were detected using biotin-conjugated goat antirabbit antibodies (diluted 1:1000; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA), followed by incubation with avidin-conjugated peroxidase (Vector Laboratories, Burlingame, CA). The resulting peroxidase activity was detected by incubation in a solution of 3,3′-diaminobenzidine (Vector Laboratories) for 3 min.

The specificity of the anti-GPR30 antibodies used here has been well documented earlier (5). To confirm further this specificity under the present experimental conditions, human liver was used as a positive control (Fig. 1A) (9) and preimmune rabbit serum as a negative control (Fig. 1F).

Fig. 1.

Expression of GPR30 in the human growth plate. Immunohistochemical localization of GPR30 expression in human liver (A), growth plate (B), and the resting (C), proliferative (D), and hypertrophic (E and F) zones of the growth plate. B–F, The growth plate from a girl at Tanner stage 1 (patient no. 1, Table 1). F, A negative control using preimmunized serum. The arrows indicate typical positive chondrocytes and the arrowheads positive prehypertrophic chondrocytes. Bars, 100 μm.

Image analysis

Digital images of stained sections were collected using a Nikon Eclipse E800 microscope (Nikon Corp., Tokyo, Japan) equipped with an Olympus DP70 digital camera (Olympus Sverige AB, Solna, Sweden). The total number of cells and positive cells in the resting and hypertrophic zones of the growth plates was counted manually by an observer who was unaware of the pubertal stage of the patient involved. The epiphyseal and metaphyseal bones and areas where the tissue sections were detached from the slides were excluded. Chondrocytes distal to the epiphyseal bone and proximal to the proliferative columns were considered to be resting chondrocytes. Flat chondrocytes with a height less than 10 μm, arranged in columns were defined as proliferative, whereas bigger chondrocytes were designated as hypertrophic. The results are presented as the percentages of cells staining positively for GPR30.

Statistics

For evaluation of developmental trends, regression analysis was performed using the pubertal stage or chronological age as the independent variable and the percentage of GPR30-positive cells as the dependent variable. A P value less than 0.05 was considered statistically significant.

Results

Immunodetection of GPR30 within the human growth plate

Despite the fact that the specificity of the anti-GPR30 antibody used here has been well demonstrated earlier (5), we decided first to confirm this specificity under our experimental conditions by staining tissues that are known to express GPR30 (9). Thus, strong positive immunostaining was detected in sections from both human liver (Fig. 1A) and heart (data not shown). Furthermore, with the same experimental procedure, immunopositive cells were also detected in the epiphyseal cartilage of the growth plate (Fig. 1B).

More detailed analysis of the growth plate revealed relatively weak immunostaining in the resting zone (Fig. 1C), an absence of positive cells in the proliferative zone (Fig. 1D), and abundant expression of GPR30 in the hypertrophic layer (Fig. 1E). Some prehypertrophic chondrocytes stained positively (Fig. 1, D and E, arrowheads). The GPR30 immunostaining was consistently localized only in the cytoplasm (Fig. 1) (data not show). Using preimmune rabbit serum as a negative control, no staining at all was detected in the growth plate (Fig. 1F).

Down-regulation of the expression of GPR30 during pubertal progression

Clear differences in the level of GPR30 expression in the growth plate cartilage of patients at different stages of puberty were observed. The level of this expression was higher in early than late puberty. Analysis of all patients clearly revealed a linear decline in the level of GPR30 expression in the resting zone of the growth plate (percentage of positive cells = 19.7 − 3.2 × the Tanner pubertal stage; R = −0.56; P < 0.01) as well as a statistically nonsignificant tendency toward a similar decline in the hypertrophic zone (percentage of positive cells = 47.1 − 6.1 × the Tanner pubertal stage; R = −0.4; P = 0.078) during pubertal progression. Separate analysis of boys and girls revealed that this decline in the resting zone is occurring in both boys and girls (Fig. 2A). However, in the case of the hypertrophic zone, the level of GPR30 expression was significantly correlated to the pubertal stage (R = −0.89; P < 0.05) in girls, but not in boys (R = −0.26; P = 0.36; Fig. 2B). In addition, a clear correlation between the level of GPR30 expression and chronological age was observed in girls (resting zone: R = −0.88, P < 0.001; hypertrophic zone: R = −0.74, P = 0.094), but not in boys (resting zone: R = −0.006, P = 0.98; hypertrophic zone: R = +0.28, P = 0.34).

Fig. 2.

Decline in the level of GPR30 expression during pubertal progression. The relationship between the percentage of GPR30-positive cells in resting (A and C) and hypertrophic (B and D) chondrocytes, and Tanner pubertal stage (A and B) and chronological age (C and D) was analyzed by linear regression. The correlation coefficients (R), P values (p), and regression lines for girls (solid) and boys (dashed) are indicated.

Nonetheless, the total percentage of hypertrophic chondrocytes expressing GPR30 in boys decreased by 50% from early to late puberty: from 36 ± 8% (six boys in Tanner stages 1–3.5) to 17 ± 5% (eight boys in Tanner stages 4–5) (P = 0.058, as analyzed with the unpaired t test). Moreover, for both boys and girls whose only diagnosis was a difference in leg length, thereby excluding patients with constitutional tall stature, and Marfan, XYY, and Klinefelter syndromes, the decline in the percentage of GPR30-positive hypertrophic chondrocytes was linearly correlated to pubertal progression (percentage of positive cells = 59.3 − 11.1 × the Tanner pubertal stage; R = −0.6; P < 0.05).

Discussion

Here, expression of GPR30 in the cartilage of the human growth plate is demonstrated for the first time. This expression is localized to the hypertrophic and resting zones, with no immunoreactivity being observed in the proliferative zone of the epiphyseal growth plate. Moreover, the level of GPR30 expression declines as puberty progresses in both boys and girls.

Interestingly, the expression of both ERα and ERβ in human epiphyseal cartilage, demonstrated previously by ourselves and others (7, 10), occurs in all zones of the growth plate (8), in contrast to the absence of GPR30 expression in the proliferative zone. This limited localization suggests that GPR30 may be involved in specific processes related to chondrogenesis in the epiphyseal growth plate. More specifically, we believe that GPR30 is not involved in the regulation of cell proliferation but might rather be involved in the recruitment of resting zone chondrocytes into columns and hypertrophic chondrocyte differentiation. Moreover, the down-regulation of GPR30 expression during pubertal progression, in contrast to the stable levels of expression of ERα and ERβ (8), indicates that GPR30 might modulate the responsiveness of epiphyseal cartilage to estrogens at different stages of puberty. Also noteworthy is the observation that the decline in the level of GPR30 expression in the resting zone occurs in parallel in boys and girls, whereas the corresponding decline in the hypertrophic zone occurs earlier in girls, just like the pubertal growth spurt and growth plate fusion, suggesting that GPR30 might be involved in the modulation of pubertal growth. Furthermore, the sex difference in GPR30 expression could also be related to the more rapid and pronounced increase in the levels of circulating estrogen in girls than in boys during progression through puberty (11–13). Because a relationship was also observed between GPR30 expression in the growth plate and age, at least in girls, we cannot exclude that this may contribute to the observed decline in expression of this receptor during pubertal development.

Targeting ERs with SERMs is a potentially useful approach to regulate growth (14). Therefore, it is of interest to note that the activity of GPR30 is modulated by SERMs in a fashion that differs from the effects of these compounds on the classical ERs (6). Moreover, a selective activator of GPR30 has recently been developed (15). These findings indicate that selective modulation of the actions of GPR30 may be possible and might represent a useful strategy for therapeutic modulation of linear growth. Nevertheless, the physiological functions of GPR30 remain to be elucidated.

In conclusion, using immunohistochemistry we have demonstrated here that GPR30 is expressed in the human growth plate of both girls and boys. This expression is localized to the resting and hypertrophic zones, being absent from the proliferative zone, which suggests that GPR30 may be involved in chondrogenesis. Moreover, the level of GPR30 expression declines as puberty progresses, indicating that this receptor may be involved in the modulation of pubertal growth.

Acknowledgments

We thank Drs. Ola Nilsson and Henrik Wehtje for tissue collections.

This study was supported by the Swedish Research Council, Stiftelsen Frimurare Barnhuset i Stockholm, Sällskapet Barnavård, HKH Kronprinsessan Lovisas förening för Barnasjukvård, and Wera Ekströms Stiftelse.

Disclosure Statement: The authors have nothing to declare.

Abbreviations

 
• ER

  Estrogen receptor;

 
• SERM

  selective ER modulator.