The local inactivation of glucocorticoid as regulator for the development of glucocorticoid-induced osteoporosis

Martina Blaschke; Ina Dressel; Regine Köpp; Gabriele Armbrecht; Stephan Sehmisch; Mladen V Tzvetkov; Frank Streit; Holger M Reichardt; Claus-C Glüer; Heide Siggelkow

doi:10.1093/pnasnexus/pgaf315

. 2025 Oct 17;4(11):pgaf315. doi: 10.1093/pnasnexus/pgaf315

The local inactivation of glucocorticoid as regulator for the development of glucocorticoid-induced osteoporosis

Martina Blaschke ^1,^2,^2,^✉, Ina Dressel ³, Regine Köpp ⁴, Gabriele Armbrecht ⁵, Stephan Sehmisch ⁶, Mladen V Tzvetkov ⁷, Frank Streit ⁸, Holger M Reichardt ⁹, Claus-C Glüer ¹⁰, Heide Siggelkow ^11,^12,¹³

Editors: Anirudha Karvande, David Brenner

PMCID: PMC12596113 PMID: 41210664

Abstract

Exogenous glucocorticoids (GCs) like prednisone are used to treat inflammatory diseases in nearly 10% of older patients. This increases osteoporosis and the risk of fractures. Until now, the negative effect on bone is thought to be a direct effect mediated exclusively by the GC receptor. However, GC effects are also locally regulated at a prereceptor level by 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1). Here, we investigated the role of 11β-HSD1 in the metabolism of exogenous GC in both human mesenchymal-progenitor-cell models and in patients undergoing GC treatment. We performed experiments focusing on regulation, activity, and effects of 11β-HSD1 on the conversion of prednisone to prednisolone and back. Subsequently, GC metabolites were analyzed in combination with adipogenic and osteogenic differentiation. We also analyzed 216 patients treated with prednisolone or methylprednisolone for different inflammatory diseases. Bone mineral density, fractures, and history of falls were investigated in combination with genotyping for single nucleotide polymorphisms of HSD11B1 as parameter of 11β-HSD1 activity. Our in vitro experiments prove that not only the activation of prednisone to prednisolone but also the reverse step of inactivation is catalyzed by 11β-HSD1 with corresponding influence on cell differentiation. In fact, in patients the inactivation of prednisolone seems to be the dominant effect influencing bone mineral density. Our results change the understanding of GC responsiveness in patient treatment and further highlight the significance of prereceptor GC regulation by 11β-HSD1.

Keywords: prednisolone, 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1), oxidation, reduction, inflammatory disease

Significance Statement.

The recent discovery that corticosterone is activated and inactivated peripherally at the site of inflammation in rodents has transformed our understanding of glucocorticoid (GC) metabolism in terms of both physiology and treatment. Using different human bone-derived cell models, we can now also demonstrate in humans that peripheral GC metabolism is important for both prednisone and prednisolone treatment. Our results are further supported by bone mineral density data in patients treated with GCs for various inflammatory diseases, as well as by the association with single nucleotide polymorphisms of the GC-activating and -inactivating enzyme, 11β-hydroxysteroid dehydrogenase 1.

Introduction

Glucocorticoids (GCs) continue to be a mainstay of treatment during the acute phase of many inflammatory diseases (1, 2). Nearly 3% of patients are treated with GC, the proportion increasing to nearly 10% as patient age increases (3, 4). However, GC therapy is also a serious risk factor in the development of fractures in men and women, depending on the dosage and duration of the treatment (5, 6). Systemically applied GCs induce a negative calcium balance, deteriorate gonadal function, and induce muscle atrophy (7, 8). In addition, there are also direct GC effects on bone turnover, including an immediate increase in bone resorption and a decrease in bone formation in the long term (9). Recently, the conception of how GCs act therapeutically was profoundly altered by the novel finding that these hormones, which are inactivated in the kidneys by 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD2), are reactivated at sites of inflammation by 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1), thereby restoring their anti-inflammatory effects (10).

The enzyme 11β-HSD1 is predominantly active in liver and adipose tissue and is able to convert inactive endogenous and exogenous GCs (such as cortisone and prednisone) into their active counterparts (such as cortisol and prednisolone) (11, 12). Dysregulation of 11β-HSD1, especially in adipose tissue, has been associated with metabolic syndrome and type 2 diabetes mellitus (13, 14).

However, 11β-HSD1 is also expressed at lower levels in a number of other tissues such as bone, suggesting that it also plays a role in GC metabolism at these sites (15, 16). In particular, it has been observed that HSD11B1 gene expression, 11β-HSD1 activity, and cortisol serum concentrations were elevated in elder subjects (17, 18), possibly explaining the increase in cortisol with aging (18). We and others further demonstrated that single nucleotide polymorphism (SNP) variants of the HSD11B1 gene affect physiological cortisol levels and the severity of age-related osteoporosis (18, 19). Namely, it was found that SNPs of the HSD11B1 gene that correlated with higher cortisol levels after dexamethasone suppression were also associated with lower bone mineral density (BMD) (18). Moreover, GCs caused an increase in bone marrow fat content concomitant to decreasing BMD. This is supported by several in vivo and in vitro studies demonstrating that GCs alter the balance between osteogenesis and adipogenesis (20–25). Recent data obtained in patients with Cushing's syndrome with hypercortisolism-induced skeletal fragility further unveiled an increase in bone marrow fat content of the vertebrae, as measured by vertebral magnetic resonance spectroscopy (26). The notion that 11β-HSD1 might be involved in the production of bone marrow fat is also supported by in vitro studies demonstrating that increased 11β-HSD1 expression in mesenchymal cells shifts the balance from an osteogenic toward an adipogenic phenotype via the production of cortisol from cortisone (25).

The role of 11β-HSD1 is less evident concerning the effects of GC therapy, e.g. with prednisone or prednisolone on bone. The metabolism of prednisone and prednisolone differs significantly from the regulation of endogenous cortisol. Endogenous cortisol is tightly regulated by the hypothalamic–pituitary–adrenal axis in response to physiological needs, such as stress and the circadian rhythm. However, prednisone and prednisolone are synthetic GCs that are metabolized in the liver and exert systemic effects independent of the body's feedback mechanisms. Prednisolone has a higher affinity for GC receptors than cortisol does, resulting in more potent anti-inflammatory effects (12). Compared with cortisol, prednisolone has a longer half-life, greater oral bioavailability, and altered plasma protein binding, resulting in prolonged systemic exposure. The transformation of prednisone into its active form, prednisolone, depends primarily on hepatic 11β-HSD1, though local activation has also been shown (12, 27). The induction of 11β-HSD1 and suppression of 11β-HSD2 in inflammatory conditions have been demonstrated before (28). Recent studies in rodents associated the predominant anti-inflammatory effect of exogenously applied GCs, which were systemically inactivated by 11β-HSD2, with the local reactivation by 11β-HSD1 at the site of inflammation (10). 11β-HSD2 gene expression was detectable in fetal osteoblasts, however, very low or undetectable in adult bone (28). Hence, this enzyme is unavailable to regulate GC homeostasis in bone cells locally (25, 29). In contrast, 11β-HSD1 expressed in bone could in principle adopt both functions, namely to reduce inactive forms of GCs to active metabolites as well as oxidize active forms for instance to inactive cortisone and prednisone, thereby regulating local GC effects on bone. 11β-HSD1 is described as bidirectional, meaning that it can both activate (reduce) GCs by converting cortisone to cortisol, and inactivate (oxidize) cortisol by converting it to cortisone. The direction of this catalysis depends primarily on substrate concentrations and the presence of the cofactor nicotinamide adenine dinucleotide phosphate (NADPH = reduced form/NADP = oxidized form) (30–32). In cell-free enzyme preparations or liver microsomes, it has been reported that the conversion of substrates no longer functions because the cofactors, especially NADPH via regeneration by the glucose-6-phosphate dehydrogenase system, are no longer available (11, 30). In liver cells, as well as in osteoblasts, 11β-HSD1 primarily functions as a reductase, activating cortisone by reducing it to cortisol using NADPH as a reductive cofactor. 11β-HSD1 is located inside the inner leaflet of the ER in close association with hectose-6-phosphate dehydrogenase (H6PDH), a powerful generator of NADP(H) (31). Unexpectedly, we have been unable to demonstrate any oxidation of the active form in bone-derived mesenchymal progenitor cells or immortalized osteoblast-like cells when investigating local cortisol metabolism (25). However, the metabolism of therapeutically applied GCs differs from physiological cortisol regulation (27), suggesting that the control of GC metabolism might be different under such conditions. In addition, large differences in the effect of GC on individuals appear to exist, possibly as a consequence of differences in 11β-HSD1 expression (17 Chap. 8). It is against this background that we hypothesized that 11β-HSD1 activity is involved in the development of GC-induced osteoporosis by activating or inactivating externally applied GCs. Thus, we investigated the effects of 11β-HSD1 expression and activity on the conversion of prednisone to prednisolone and vice versa in vitro using a human mesenchymal cell model. In addition, we set out to address this issue in an in vivo model. Owing to species differences in the regulation of GC metabolism, we preferred a human model to experiments in rodents. To this end, we investigated a large cohort of patients with different inflammatory diseases treated with prednisolone or methylprednisolone with respect to BMD as well as the number of falls and fractures. To detect the influence of 11β-HSD1 in these patients, we identified the three SNPs rs11811440, rs1000283, and rs932335 located in HSD11B1 and focused our analysis on rs11811440 with respect to bone and muscle parameters.

Results

Regulation of 11β-HSD1 expression by pro-inflammatory cytokines

To examine the regulation of 11β-HSD1 under inflammatory conditions, experiments in the human mesenchymal stem-cell-line SCP-1 were performed during osteogenic differentiation. A cytokine cocktail composed of interleukin-1β (IL-1β; 1 pg/mL), tumor necrosis factor alpha (TNFα; 5 pg/mL), and IL-6 (10 pg/mL) was applied to SCP-1 cells. Changes in 11β-HSD1 gene expression compared with control conditions were determined by quantitative RT-PCR. It is worth noting that the employed cytokine concentrations correspond to approximately one-tenth of those found in sera of patients with Crohn's disease (CD) during the acute phase of the disease (33–35). Addition of the cytokine cocktail during the last 48 h of the 14-day culture period resulted in a 1.4-fold increase in mRNA levels of 11β-HSD1 compared with untreated cells (Fig. 1A). Application of a high concentration of TNFα (1 ng/mL) increased 11β-HSD1 gene expression by as much as a factor of 100, which could be reversed by adding ascending concentrations of a neutralizing anti-TNFα antibody (Fig. 1B).

Fig. 1. — Analysis of 11β-HSD1 normalized gene expression in SCP-1 cells using quantitative RT-PCR after 14 days in culture under osteogenic differentiation conditions (10 mM glycerol-phosphate, 173 µm ascorbic acid-phosphate). A) A cytokine cocktail (1 pg/mL IL-1β, 5 pg/mL TNFα, 10 pg/mL IL-6; n = 3 individual experiments; n = 3–4 biological replicates) was applied during the last 48 h of the culture period; P = 0.0046; t = 3.18; df = 21. B) 1 ng/mL TNFα (plus an unspecific IgG or 0.01, 0.1, 1.0, and 8.0 µg/mL anti-TNFα antibody; n = 1 individual experiment; n = 6 biological replicates) was applied during the last 48 h; P < 0.0001; t = 16.23; df = 10. Statistical analysis was performed using two-tailed Student's t test. (mean ± SEM).

Taken together, our findings indicate that a combination of pro-inflammatory cytokines as detected in patients with CD leads to an increase in 11β-HSD1 gene expression, mainly caused by TNFα.

Conversion of GCs during osteogenic differentiation of SCP-1 cells

The most common medication to treat autoimmune inflammation is the application of synthetic GCs such as prednisone or prednisolone. Here, we studied the conversion of these two compounds in SCP-1 cells cultured for 21 days under osteogenic conditions. First, we measured the reduction of inactive prednisone to active prednisolone as well as the oxidation of active prednisolone to inactive prednisone, both catalyzed by 11β-HSD1. To this end, the release of prednisolone (or prednisone) into the medium was monitored after treatment with prednisone (or prednisolone) for 48 h at different stages of osteogenic differentiation. The concentration of each GC in the medium was increased after addition of the respective other GC form, and the effect was stronger after 21 days compared with the 14-day time point (Fig. 2A and B). This finding confirms that a conversion of prednisone to prednisolone and vice versa takes place under osteogenic conditions in SCP-1 cells. Notably, the elevated GC secretion was mirrored by 11β-HSD1 mRNA levels (Fig. 2C and D). Gene expression of 11β-HSD1 increased 10.4-fold during the 21 days of osteogenic differentiation and was further enhanced by GC treatment at the end of the culture period, 4-fold by prednisone, and 8-fold by prednisolone. Notably, conversion of prednisone to prednisolone (11.2%) was more effective than the conversion of cortisone to cortisol (3.3%), as demonstrated previously (25) and documented in Fig. S2.

Fig. 2. — 11β-HSD1 gene expression (A, B) and conversion of GCs (C, D) in SCP-1 cells during 21 days in culture under osteogenic conditions. A, B) Gene expression of 11β-HSD1 (n = 4 biological replicates) in the absence (open circle) or (A) presence (closed circle) of 5 µM prednisone (Student's t test two-tailed⁺, *P < 0.0001; t = 10.10; df = 6) or 5 (B) µM prednisolone (Student's t test two-tailed, *P = 0.01; t = 3.96; df = 5 and ⁺P = 0.0001; t = 8.91; df = 6) during the last 48 h. C) Release of prednisolone after incubation with 5 µM prednisone during the last 48 h (n = 3 biological replicates). D) Release of prednisone after incubation with 5 µM prednisolone during the last 48 h (n = 3 biological replicates).The spontaneous oxidation of prednisolone was subtracted from prednisone values (see Fig. S1), occasionally resulting in negative values. *Control vs. GC on day 21; ⁺day 0 vs. day 21 under control conditions.

Influence of prednisone activation in SCP-1 cells and human mesenchymal progenitor cells on osteogenic differentiation

Effects of prednisone on typical GC target genes as well as on adipogenic and osteogenic markers were analyzed in SCP-1 cells to monitor the influence of GCs on differentiation (Fig. 3). Primary cells (human mesenchymal progenitor cells, hMSCs) are well known to have a higher differentiation potential compared with cell lines. Hence, we set out to confirm our findings in SCP-1 cells by performing the same experiments in hMSC cultures. The selective inhibitor 10j was used to confirm the specificity of the 11β-HSD1-catalyzed GC conversion.

Treatment of SCP-1 cells or hMSCs with prednisone resulted in its conversion to prednisolone, which was completely inhibited in the presence of the specific 11β-HSD1 inhibitor 10j (Fig. 3A and B). In parallel, 11β-HSD1 mRNA levels (Fig. 3C) were increased by 9-fold in SCP-1 and 21-fold in hMSCs compared with control conditions (Fig. 3D). The newly synthesized prednisolone resulted in activation of the GC target genes DUSP1 (2.6-fold) and GILZ (6.3-fold) in SCP-1 (Fig. 3C). Comparable results were obtained in hMSCs (Fig. 3D). Furthermore, we could demonstrate a strong increase in gene expression of the adipogenic markers aP₂ (5.6-fold) and lipoprotein lipase (LPL) (9.6-fold) in SCP-1 cells. Interestingly, a 12-fold increase was also detected for DKK1 (Fig. 3E), which is a major inhibitor of the bone anabolic Wnt-signaling pathway and known to be involved in bone formation (36). The induction of the adipogenic differentiation program by osteogenic stimulation in hMSCs was stronger (aP2: 7.5-fold; LPL: 11.7-fold), as shown in Fig. 3F. This was accompanied by a 26-fold increase in DKK1 expression, which demonstrates a significant impact on Wnt signaling and negatively affects osteogenic differentiation.

Effects on the expression of genes characteristic of osteogenic differentiation, however, were only marginal in SCP-1 cells (Fig. 3G). ALP expression was unchanged by prednisolone, and OCN expression was increased 1.9-fold. Contrary to our results in SCP-1 cells, expression of the osteogenic marker genes OCN and ALP even tended to decrease after prednisone treatment in hMSCs (Fig. 3H). Collectively, the results obtained in primary hMSCs largely confirm the findings made in SCP-1 cells and indicate that prednisone induces a shift of the differentiation program from osteogenesis to adipogenesis after its reduction to prednisolone.

All changes in gene expression depended on 11β-HSD1, as they were essentially abolished in the presence of its inhibitor, 10j. Although the inhibitor was effective, as demonstrated by prednisolone measurements, activation was not completely inhibited in hMSCs, resulting in unchanged levels of 11β-HSD1 mRNA (Fig. 3D). Our data suggest that the residual activity of 11β-HSD1 resulted in a small amount of prednisolone, which was sufficient to maintain elevated 11β-HSD1 and LPL gene expression while inhibiting DKK and aP2 expression in these cells. Therefore, our data demonstrate that a very small amount of prednisolone can maintain active adipogenic differentiation. The different effects on DKK1 and adipogenic markers may suggest distinct effects at various stages of differentiation. Human primary cells undergo various stages of differentiation, including the development of other cell types, such as adipocytes. This has been demonstrated in previous publications (25, Chap. 561; 24 Chap. 105). The variability in differentiation of hMSCs depends on the site of the bone specimen, the patient's age, the number of cells used for seeding, and especially the time and confluence in culture (37, Chap. 106 Chap. 199; 38, Chap. 281; 39, Chap. 3; 40, Chap. 587). Responsiveness to different stimuli also varies depending on the stage of differentiation (37, Chap. 199). This has also been demonstrated following stimulation with GCs (41, Chap. 180). These experiments, which used different seeding densities for cells with different rates of cell division, did not produce completely identical results at the various time points.

Taken together, our findings suggest that prednisolone inactivation depends on 11β-HSD1 activity and is presumably caused by transcriptional regulation. Furthermore, the simultaneous production of prednisolone promotes adipogenic differentiation while having minimal impact on osteogenic markers.

Influence of prednisolone inactivation in SCP-1 cells and hMSCs on osteogenic differentiation

Given the known capacity of 11β-HSD1 to oxidize GCs in the absence of 11β-HSD2, we wondered whether inactivation of prednisolone was relevant in the differentiation of SCP-1 cells or hMSCs (Fig. 4).

Conversion of prednisolone to prednisone was almost completely inhibited by addition of 10j to SCP-1 cells (Fig. 4A), while the increase of 11β-HSD1 gene expression by prednisolone was unaffected by addition of 10j in SCP-1 cells (Fig. 4C). In SCP-1 cells, application of prednisolone significantly increased gene expression of the GC target genes DUSP1 (1.9-fold) and GILZ (6.2-fold; Fig. 4C), and promoted adipogenesis based on the 4.8-fold induction of aP₂, the 9.9-fold induction of LPL, and the 14.8-fold induction of DKK1 (Fig. 4E). Effects on osteogenic differentiation were low, with a slight increase in OCN expression (Fig. 4G). Addition of 10j did not affect the expression of any of the analyzed genes. This finding indicates that all the observed effects of GCs on the differentiation program of SCP-1 cells are mostly mediated by their reduced form prednisolone.

In hMSCs, treatment with prednisolone resulted in the generation of prednisone (Fig. 4B), and increased expression of 11β-HSD1 (Fig. 4D), a strong induction of the GC target genes DUSP1 (6.2-fold) and GILZ (11.2-fold; Fig. 4D), and tremendously promoted adipogenic differentiation (aP₂: 20.5-fold, LPL: 13.4-fold, DKK1: 15.9-fold; Fig. 4F). In contrast, effects on osteogenic differentiation were only moderate with only a slight increase in ALP expression and a 37% decrease in OCN expression (Fig. 4H).

Conversion of prednisolone to prednisone in hMSCs was blocked by the 11β-HSD1 inhibitor 10j as expected (Fig. 4B), resulting in a 65% lower release of the inactive GC. However, 11β-HSD1 expression was further increased, whereas mRNA levels of DUSP1 and GILZ remained unaltered (Fig. 4D). Interestingly, also the adipogenic markers aP₂ and LPL as well as DKK1 further increased in the presence of 10j (Fig. 4F). In contrast, although significance was not reached, the expression of OCN and ALP was reversed (Fig. 4H). These findings suggest that the stimulation of adipogenic differentiation by prednisolone in hMSCs was further promoted in the presence of 10j, presumably due to the inhibition of prednisolone oxidation to prednisone, thereby shifting the equilibrium of both forms of GC toward their active form. The results from Figs. 3B and 4 support the hypothesis that an increase in oxidative activity (Fig. 4B, D, F, and H) is accompanied by a decrease in the activity of the enzyme responsible for converting prednisone to prednisolone (Fig. 3B, D, F, and H).

Prednisolone or methylprednisolone therapy in patients

From these in vitro results, we conclude that the 11β-HSD1 enzyme activity in hMSC-derived osteoblasts regulates the local activation of GC treatment and thereby cell differentiation. If this is of relevance for bone health in vivo, SNPs of HSD11B1 should influence the effect of GC treatment in vivo.

To examine the outcome of GC therapy in relation to 11β-HSD1 enzyme activity in vivo, we used data from 216 patients enrolled in the PSIOD study (Prevalence of Steroid-Induced Osteoporosis in Germany) who received orally applied prednisolone or methylprednisolone. The patient characteristics are summarized in Table 1.

Table 1.

Patient characteristics.^a

Parameter	All patients n = 216 (100%)	Male n = 80 (38%)	Female n = 136 (62%)
Age mean ± SD (range)	68 ± 6 (55–80)	68 ± 7 (55–79)	67 ± 6 (56–79)
BMI mean ± SD (range)	27.1 ± 4.1 (18–41)	26.9 ± 3.4 (19–36)	27.3 ± 4.4 (18–41)
Indication for GC therapy (n)
Rheumatism	81 (38%)	23 (29%)	58 (43%)
Pulmonary disease^b	67 (31%)	33 (41%)	34 (25%)
Polymyalgia	21 (10%)	9 (11%)	12 (9%)
Other rheumatic disease	16 (7%)	5 (6%)	11 (8%)
Other	31 (14%)	10 (13%)	21 (15%)
Prednisolone (n)	140 (65%)	50 (63%)	90 (66%)
Cumulative (g/m²)^cdose	15.5 ± 17.9	22.7 ± 23.4	11.6 ± 12.5
Daily (mg/m²) dose^c	3.8 ± 2.6	3.9 ± 2.1	3.7 ± 2.8
Duration (years)	9.3 ± 9	12.2 ± 10	8.2 ± 8
Methylpredisolone (n)	76 (35%)	30 (38%)	46 (34%)
Cumulative (g/m²)^c dose	14.2 ± 17.8	16.6 ± 21.8	12.6 ± 14.5
Daily dose (mg/m²)^c	4.1 ± 3.4	4.0 ± 3.3	4.1 ± 3.5
Duration (years)	9.6 ± 8	9.4 ± 10	9.1 ± 9
Vertebral osteoporotic fractures (fx)^d
Number of fx	n = 153	n = 75	n = 78
Number of patients with fx	n = 51 (24%) (1–12/pat)	n = 23 (28%) (1–8/pat)	n = 28 (21%) (1–12/pat)
Fx grading by Genant^d in n patients
Grade 1	n = 6	n = 3	n = 3
Grade 2	n = 20	n = 10	n = 10
Grade 3	n = 25	n = 10	n = 15
Peripheral fractures^e (fx)
Number of fx	n = 193	n = 72	n = 121
Number of patients with fx	n = 122 (56%)	n = 42 (53%)	n = 80 (59%)
Forearm	n = 25 (12%) (1–4/pat)	n = 6 (8%) (1–3/pat)	n = 19 (14%) (1–4/pat)
Rib	n = 18 (8%) (1–7/pat)	n = 9 (11%) (1–7/pat)	n = 9 (7%) (1–3/pat)
Hip	n = 5 (2%) (1/pat)	n = 2 (3%) (1/pat)	n = 3 (2%) (1/pat)
Other locations	n = 74 (34%) (1–6/pat)	n = 25 (31%) (1–6/pat)	n = 49 (36%) (1–6/pat)
Falls^e
Number of falls Number of patients with falls	n = 90 n = 47 (22%)	n = 32 n = 20 (24%)	n = 58 n = 27 (20%)
	(1–10/pat)	(1–6/pat)	(1–10/pat)
DXA measurement
BMD femur (g/cm²)	0.856 ± 0.141	0.900 ± 0.141	0.832 ± 0.135
T-score femur	−0.907 ± 1.047	−0.91 ± 0.94	−0.90 ± 1.11
Number of patients	n = 215	n = 80	n = 135
BMD spine (g/cm²)	0.912 ± 0.175	0.947 ± 0.180	0.891 ± 0.169
T-score spine	−1.38 ± 1.56	−1.31 ± 1.63	−1.43 ± 1.52
Number of patients	n = 206	n = 78	n = 128

Open in a new tab

^aNot all data were available for all patients.

^bAsthma, COPD, or other pulmonary disease.

^cDoses are given as equivalence doses calculated per body surface.

^dData collected by X-ray analysis and evaluated applying grading according to Genant and Jergas (42).

^eData collected through patient history and questionnaire.

BMD at the spine and at the femoral neck correlated to the occurrence of vertebral osteoporotic fractures (spine r = −3.22; P = 0.003; femoral neck, r = −0.50; P < 0.001; univariate logistic regression).

Duration of GC therapy (P = 0.01) and cumulative dosage (P = 0.005) of prednisolone were significantly different between males and females. The cumulative prednisolone or methylprednisolone equivalent dosage per body surface correlated to vertebral fractures (r = 0.02; P = 0.015). In contrast, there was no correlation between BMD values and parameters of GC therapy.

The SNPs rs11811440, rs1000283, and rs932335 in the HSD11B1gene have been previously reported as playing a role in the BMD of osteoporotic patients and were associated with serum cortisol levels after dexamethasone treatment (18). Patients analyzed for the rare rs11811440 A/A genotype were also analyzed for the rare genotype in rs1000283 and rs932335, with three exceptions; hence, we focus in the following on the results of SNP rs11811440. Analysis of all 216 patients revealed that the genotype call rate was 100% and that this SNP was in Hardy–Weinberg equilibrium (P > 0.2, Pearson’s χ² test). The most frequent genotype was C/C (67%; n = 145), followed by C/A (29%; n = 63). The homozygous allele combination A/A was detected in eight patients only (4%). HSD11B1 genotypes did not significantly correlate with daily prednisolone dosages used. The prednisolone equivalence dosage was similar in all three SNP rs11811440 groups (Fig. 5). The lowest cumulative (13.9 ± 22.0 g/m²) as well as daily dosage (2.9 ± 0.9 mg/m²) was applied in patients with the A/A allele combination (Fig. 5A).

Considering the importance to GC conversion in patients receiving (methyl)-prednisolone, we compared the relevance of the HSD11B1 SNP rs11811440 for the T score at the spine and femoral neck in all patients as well as in patients older than 65 years (Fig. 5B). Interestingly, an SNP dependence was observed in patients treated with (methyl)-prednisolone, revealing an inverse correlation between the number of A alleles and the T score measured in these patients. More specifically, a significant linear relationship could be demonstrated (all patients: spine: R² = 0.051; P = 0.001; femur: R² = 0.023; P = 0.025) between the T score at the spine or femoral neck and the occurrence of the minor A allele of the SNP (one-way ANOVA: spine: P = 0.003; femoral neck: P = 0.063). In patients older than 65 years, the relationship was even more significant (patients ≥65: spine: R² = 0.07; P = 0.002; femoral neck: R² = 0.06; P = 0.003), and T scores for the spine were the lowest in patients with A/A genotype as well (patients ≥65 years, one-way ANOVA: spine: P = 0.001; femur: P = 0.008) (Fig. 5B, Table 1).

Finally, we analyzed the contribution of gender, age (Fig. 5C) and number of vertebral osteoporotic fractures as determined by X-ray analysis, peripheral fractures and falls as acquired by the patient questionnaire regarding the HSD11B1 SNP rs11811440 (Fig. 4D and E). The patients in all three SNP groups had a similar contribution of gender and age. No differences regarding number of vertebral osteoporotic fractures, or peripheral fractures or falls (Fisher’s exact testing; P = 0.54; P = 0.62 and P = 0.20) were detected. Collectively, our data unveil that SNP rs11811440 in the HSD11B1 gene impacts features of osteoporosis in patients treated with GC.

Discussion

In a number of inflammatory diseases, therapeutic GCs such as prednisolone are still the mainstay of treatment in the early phase of treatment. However, their use is limited by the systemic and local side effects. Effects on bone are initially characterized by high bone resorption. As the duration of therapy increases, a decrease in bone formation occurs through reduced osteoblast and osteocyte function, leading to reduced bone strength and fractures (43). In addition, systemically applied GCs induce a negative calcium balance, deteriorate gonadal function, and induce muscle atrophy (7, Chap. 135; 8 Chap. 578) further increasing the risk of falls and fractures. We and others previously demonstrated that SNPs in HSD11B1 affect physiological cortisol levels, the severity of age-related osteoporosis (18) and also fractures (19). Moreover, it has already been suggested that 11β-HSD1 may also influence the efficacy of GC therapy on bone (12, 17).

In this study, we investigated how 11β-HSD1 influences bone metabolism using both an in vitro model and an in vivo study population. Our results suggest that 11β-HSD1 acts by both activating prednisone and inactivating prednisolone in bone.

In hMSCs, we demonstrated that a combination of pro-inflammatory cytokines results in an increase of 11β-HSD1 gene expression, mainly caused by TNFα. These data obtained in human cells are in line with recently published findings in a murine model of polyarthritis. In their work, Fenton and colleagues presented a novel hypothesis of GC activation in chronic inflammatory diseases, namely that GCs are systematically inactivated and only locally activated by 11β-HSD1, with increasing activity depending on the inflammatory response and TNFα. Similarly, our model was mainly influenced by a dose-dependent effect of TNFα. Hence, GC effects on bone during inflammatory diseases also depend on 11β-HSD1. Our data are confirmed by the targeted deletion of 11β-HSD1 in osteoblasts and osteoclasts and in bone biopsies from patients with rheumatoid arthritis compared with osteoarthritis (44).

Under osteogenic conditions, we induced the differentiation of immortalized human progenitor cells SCP-1 toward the osteogenic lineage. Using electron microscopy, demonstrated the co-expression of osteogenic and adipogenic markers in the same cells under osteogenic culture conditions, which typically include dexamethasone and ascorbic acid (24 Chap. 105). In this study, however, we did not present data obtained using adipogenic differentiation media. Instead, we only examined the adipogenic and osteogenic differentiation of cells via osteogenic differentiation. Our data, along with data from other groups, show that even low levels of GCs in the osteogenic medium are sufficient to induce dedifferentiation from an osteoblastic to an adipogenic phenotype. This is a concomitant event (45, Chap. 588).

We detected a differentiation-dependent increase in 11β-HSD1 gene expression and a correspondent conversion of prednisone to prednisolone and vice versa in this human cell model. Conversion of prednisone to prednisolone was more effective than the conversion of cortisone to cortisol as demonstrated previously (25). The bioavailability of prednisolone is influenced by several factors, like binding affinity, substrate availability, localization in cytosol or the endoplasmatic reticulum and the redox state dependent on NADPH and hectose-6-phosphate dehydrogenase, a powerful generator of NADP(H) (31). Adipogenesis increased within 48 h, whereas the activity of the Wnt pathway decreased as a result of the higher DKK1 expression without any effect on osteogenic markers. The specificity of the 11β-HSD1 activity was proven by the use of a specific inhibitor and by the absence of 11β-HSD2 gene expression. The main effect on the differentiation of SCP-1 was mostly mediated by the reduced form without any evidence of the inactivation of prednisolone. However, the differentiation potential of primary human cells differs from that of immortalized cell systems. Human primary cells undergo various stages of differentiation, including the development of other cell types, such as adipocytes. This has been demonstrated in previous publications (24, 25). Therefore, primary cells represent mixed cell cultures that more closely resemble the in vivo situation in bone, but they also have the disadvantage of higher heterogeneity and variation in results. Our results with hMSCs and the 11β-HSD1 enzyme inhibitor 10j suggest that increased oxidative activity is associated with reduced activity of the enzyme responsible for converting prednisone to prednisolone.

In our previous study in which we analyzed osteoporosis patients, we found that five polymorphisms in the HSD11B1 gene significantly correlated with BMD. All five SNPs are located in intron 5 and highly genetically linked. One of them, rs11811440, consistently demonstrated the strongest correlation with BMD, which points toward a reduced rate of fractures in homozygous carriers (18). Based on our results and another study indicating that the two highly genetically linked variants rs1000283 and rs932335 were associated with BMD at the femoral neck and with fracture risk in Korean postmenopausal women with osteoporosis, we hypothesized that the 11β-HSD1 activity represented by SNP rs11811440 for HSD11B1 may also influence the effectivity of GC therapy on bone. In this study, we therefore investigated 216 patients treated with prednisolone or methylprednisolone for a number of inflammatory diseases. Prednisolone has a longer half-life and greater oral bioavailability, which is partly due to its lipophylic properties. It is not influenced by daytime or stress and binds to albumin, resulting in prolonged systemic exposure compared with cortisol, which has different pharmacokinetics. Prednisolone binds more tightly to the 11β-HSD1 enzyme and to the GC receptor 10 times faster. The cumulative prednisolone or methylprednisolone equivalent dose per body surface correlated to vertebral fractures. However, there was no correlation of BMD to cumulative or daily dose of prednisolone equivalent. The allele combination was in agreement with our results obtained in patients analyzed for osteoporosis or in other studies (18). No significant differences concerning age, cumulative, or daily prednisolone equivalent dose was evident between SNP rs11811440 alleles, which could have influenced the results for BMD, falls, and fractures. SNP rs11811440 alleles demonstrated a significant correlation with BMD T score at the spine and a tendency at the femoral neck, the difference becoming significant in patients ≥65 years of age. Patients with the AA allele demonstrated the lowest BMD values. Regarding the prednisolone equivalent, patients with the AA allele had the lowest doses of all patients and a clearly lower range; this effect was therefore clearly unrelated to the higher doses of GC. These data indicate that the effect of GC therapy on bone is dependent on the SNP rs11811440 allele combination and that the group with the C/C allele has a higher BMD.

At first glance, these data are in contrast to previous studies on osteoporosis patients with regard to the investigated physiological cortisol levels (18, 19). Here, the C/C allele was associated with lower BMD levels at the spine and femoral neck (19) and correlated with an increase in suppressed cortisol levels after dexamethasone suppression (18). Hence, the C/C allele represents an increase in 11β-HSD1 enzyme function under physiological conditions. The question arises perhaps as to how our data on patients treated with GC can be thus explained. Our experimental data prove the increased oxidative capacity of 11β-HSD1, thereby inactivating the equivalent prednisolone to prednisone with an effect on MSC differentiation. This effect is influenced by inhibiting the 11β-HSD1 enzyme. The oxidative capacity of the enzyme is considerably higher than that in the oxidation of cortisone to cortisol as demonstrated in earlier studies (25). From these results, our data at hand can be explained by an effect of the SNP rs11811440 allele of HSD11B1 on the oxidation of prednisolone to prednisone equivalent. Patients with the C/C allele inactivate more prednisolone than those with either the C/A or AA allele, followed by a lower concentration in the body as well as the bone, thus explaining higher BMD levels. Our data are supported by the notion that the patients with the AA allele were administered the lowest range of GC cumulative and daily doses, although the differences were not significant. The effect of the reduced inactivation capacity of the AA allele of HSD11B1 encoding for the 11β-HSD1 enzyme thus indicates that higher local prednisolone equivalent doses are available to attain the anti-inflammatory effect, resulting in possibly lower daily doses administered.

Our results in patients treated with GC for inflammatory disease are highly relevant in light of the recently identified important role of the 11β-HSD1 enzyme in exclusively activating GC locally at the site of inflammation. The significance of this enzyme in elder subjects was already demonstrated in a previous study of ours in patients evaluated for osteoporosis (18). Based on our results, it becomes evident that not only the activation but also the inactivation of GC by 11β-HSD1 in bone is of major importance in the treatment of patients with inflammatory disease and elder subjects.

A number of publications have clearly demonstrated interest in this field by reporting on the development of new substances that influence 11β-HSD1 enzyme activity (46, 47). Data from patients with Cushing's disease and syndrome who were protected from severe adverse effects of excess endogenous GCs due to a functional deficit in 11β-HSD1 activity (48) supported the development of specific enzyme inhibitors. The 11β-HSD1 inhibitor AZD4017 has been found to regulate skin function in humans with type 2 diabetes, improving wound healing and epidermal integrity while increasing water loss (49). AZD4017 was also tested in a randomized, double-blind, placebo-controlled trial involving 55 postmenopausal women with osteopenia over a period of 90 days. Despite >90% inhibition of 11β-HSD1 activity, there was no effect on bone formation or resorption, as determined by various bone turnover markers (49). The authors concluded that, in normal physiology, changes in local cortisol metabolism mediated by 11β-HSD1 expression in osteoblasts do not limit the rate of bone formation or resorption (49). A recent prospective, randomized, double-blind, placebo-controlled study of healthy probands showed that 11β-HSD1 inhibition with AZD4017 could mitigate adverse GC effects without compromising their anti-inflammatory actions (50). Future clinical studies will need to evaluate whether inhibiting 11β-HSD1 also in inflammatory diseases alleviates side effects without reducing the anti-inflammatory action of GCs at the local site. These data highlight the importance of local GC metabolism in GC treatment.

Our results provide new data on the metabolism of 11β-HSD1 in humans with inflammatory diseases, in which 11β-HSD1 is strongly induced and influences the activation and inactivation of prednisolone. Taken together, our results emphasize the importance of investigating the dual function of 11β-HSD1 in the activation and inactivation of local GCs.

A limitation of our study is its low power to identify the effects of the rs11811440 SNP on the frequency of fractures and falls. Since patients with the allele combination A/A represent only 4% of our study cohort, the data need to be confirmed in a larger group of patients. From the Korean data on vertebral fracture prevalence of 7.44% in postmenopausal women and associated with SNPs of HSD11B1 (19), we calculated a power analysis for the estimated sample size to be able to prove a fracture risk reduction. In the Korean study, the dominant A/A allele was associated with a 50% fracture risk reduction (19). To detect a difference of 7.44 vs. 3.72% in fracture rates between the groups with 80% power and α = 0.05, at least 4,375 postmenopausal women would be needed so that ∼175 are in the A/A group. However, GC-induced osteoporosis is complex and BMD values do not correlate well with fractures, even proving to be independent as also seen in our study. Besides the effect on bone, a number of systemic effects of GC on calcium metabolism, gonadal hormones, and IGF-1 could also possibly influence falls and fractures. Therefore, further studies on falls and fractures in patients treated with GCs should include the rs11811440 SNPs or highly genetically linked variants in their analyses, together with other parameters that influence bone health. It may well be that the main effect of 11β-HSD1 enzyme concerning bone is local. We must point out that we are only presenting correlational data. Our study is limited by the absence of functional assays that validate the direct effect of rs11811440 on 11β-HSD1 oxidative activity.

In summary, the conversion of prednisone to prednisolone and vice versa in human models of MSCs is catalyzed by 11β-HSD1. Inflammation is followed by an increase in 11β-HSD1 expression, mainly mediated by TNFα. The enzyme 11β-HSD1 also catalyzes the inactivation of prednisolone to prednisone with effects on cell differentiation. In patients treated for inflammatory disease with prednisolone, the oxidative capacity of the enzyme mainly influences BMD values, as demonstrated by the analysis of the SNP rs11811440 allele of HSD11B1. Owing to the large number of individuals treated with GC (4), our data have huge impact on the treatment of inflammatory diseases. Our data would support the use of HSD11B1 SNPs to adapt treatment regimens employing GC in patients with inflammatory diseases. Further investigations into patients and the development of substances influencing 11β-HSD1 enzyme function should therefore also focus on its oxidative capacity.

Materials and methods

Cell culture

Materials for cell culture were obtained from Nunc (Roskilde, Denmark), fetal calf serum (FCS) from Lonza (Cologne, Germany), and cell culture media and the medium supplements (antibiotics and glutamine) from GIBCO-BRL (Eggenstein, Germany). All other reagents were purchased from Sigma (Taufkirchen, Germany), unless otherwise stated.

Isolation and culture of human mesenchymal progenitor cells and SCP-1 cells

Spongiosa was isolated from patients during surgery and bone fragments were prepared as described earlier (33, 37, 51). hMSCs, which have the potential to differentiate into osteoblasts, were cultured according to our established protocol (25, 52). Analysis by fluorescence-activated cell sorting revealed a surface-marker profile consistent with a mesenchymal origin. FITC-conjugated mouse-anti-human monoclonal antibodies obtained from BD Pharmingen (San Diego, CA, USA) were employed at a dilution of 1:50. SCP-1 cells, a single-cell-derived hMSC cell line which expresses human telomerase reverse trascriptase (hTERT) after lentiviral gene transfer and displays a stem-cell-like character as demonstrated by its osteogenic, adipogenic, and chondrogenic differentiation potential, were kindly provided by M. Schieker (Munich) (53). SCP-1 cells from passage 80 were cultured in DMEM (high glucose), 10% FCS, penicillin/streptomycin, and 2 mM glutamine.

Osteogenic differentiation

Osteogenic differentiation generally includes the use of dexamethasone at a concentration of 10 nM. However, in our study, we investigated GC metabolism and omitted dexamethasone from the initial stimulation.

HMSCs were trypsinized and plated in 6-well plates at a density of 1.7 × 10⁵ cells/mL. Subsequently, osteogenic differentiation was induced by adding 10 mM beta-glycerophosphate, 10 µM ascorbic acid 2-phosphate, and 50 nM 1.25-OH vitamin D_3.

SCP-1 cells were plated at a lower density of 4 × 10⁴ cells/mL in 6-well plates. Osteogenic differentiation was induced by adding 10 mM beta-glycerophosphate, 10 µM ascorbic acid 2-phosphate, and 50 nM 1.25-OH vitamin D₃.

RNA isolation and cDNA synthesis

Total RNA was isolated using the Qiagen RNeasy Mini Kit, and purity was confirmed based on the 260/280 ratio. For each sample, 250–500 ng of RNA were reverse transcribed into cDNA using M-MLV reverse transcriptase, as described previously (52). RT-PCR analysis was performed using a Peq-Lab Primus 96 thermal cycler in a total volume of 40 µL as reported earlier (24). Quantitative RT-PCR analysis was performed using an ABI Prism StepOnePlus or an ABI 7500 Instrument (Applied Biosystems, Darmstadt, Germany) employing the SYBR Green Reaction Master Mix from the same company and the primers listed in Table S1. Primers were either synthesized by Invitrogen or Metabion (Planegg, Germany). The housekeeping gene ACTB was used for normalization, because its expression was found to be constant under the different experimental conditions; relative expression levels of target genes were calculated with the Ct method (54).

Pharmacological manipulations

SCP-1 cells or hMSCs were cultured for 19 days under osteogenic conditions as described previously (25, 34). Subsequently, the chemical compound 10j (CAS 1009373-58-3; Merck compound 385581) was applied at a concentration of 1 µM specifically to inhibit 11β-HSD1 activity. Six hours later, either 5 µM prednisone or 5 µM prednisolone dissolved in 0.1% Bovine serum albumin (BSA) were added to SCP-1 cells and incubated for another 42 h. Treatment with 10j alone served as control to identify inhibitor-specific effects.

Analysis of GC concentrations by LC–MS/MS

To determine GC concentrations in the cell-culture supernatant, the medium was changed to FCS-free medium containing only 0.1% BSA. Then, prednisone (dissolved in ethanol) or prednisolone (dissolved in methanol) was added for 48 h depending on experimental conditions. Incubation with 0.05% ethanol or 0.05% methanol in 0.1% BSA served as vehicle controls. To monitor spontaneous reduction (prednisone to prednisolone) or oxidation (prednisolone to prednisone), GCs were incubated for 48 h at 37 °C in the absence of cells. While no spontaneous reduction of prednisone to prednisolone occurred, the spontaneous oxidation of prednisolone to prednisone was detected in four independent experiments. The mean values for prednisone detected as spontaneous oxidation product (n = 2–3) in every individual prednisone experiment was subtracted from every prednisone measurement (n = 3), which occasionally resulted in negative values (Figs. 3 and S1). At the end of the experiment, cell-culture supernatants were removed, immediately frozen, and stored at −20 °C until analysis. LC–MS/MS measurement was performed by the central analytic laboratory facility of University Medical Center Göttingen (55).

Application of cytokines and cytokine-neutralizing antibodies

To induce osteogenic differentiation, SCP-1 cells were treated with 10 mM beta-glycerophosphate and 173 µM ascorbic acid 2-phosphate on day 7 and cultured for another 14 days. During the last 48 h, a cytokine cocktail containing 1 pg/mL IL-1β, 5 pg/mL TNFα and 10 pg/mL IL-6 or 1 ng/mL TNFα in combination with different concentrations (0, 0.01, 0.1, 1.0, and 8 µg/mL) of an anti-TNFα antibody was added. This selection of cytokines was based on findings made in patients with Crohns’ disease (33) and their ability to induce osteoclast-stimulating activity in osteoblasts (34). An IgG antibody served as an isotype control. The anti-TNFα antibodies were preincubated with TNFα for 2 h prior to being added to the cell culture. RNA was isolated at the end of the experiment on day 21 and stored for further analysis.

Patients

We included 216 patients treated with prednisolone or methylprednisolone recruited between 2002 and 2004 in Berlin for the PSIOD study into our analysis. The original study was performed in two phases. The first phase was intended to identify the number of patients taking GC medication in the general population between 18 and 80 years of age. Hence, a stratified random sample was selected from addresses on the register at the residents’ registration office Berlin in Germany and contacted by regular mail. For the second phase, the appointment at the clinic was used to identify the number of prevalent vertebral fractures by assessing upper and lumbar spine radiographs in patients between 55 and 80 years old currently taking GCs orally. These patients were recruited out of the patients at the institute or were referred from other physicians. Patients were included if they were taking GC orally at the time of inclusion and the 3 months prior at a dose ≥2.5 mg/day prednisolone equivalent with an interruption of fewer than 14 days. The prednisolone equivalent was calculated individually for the different medications (56).

Patients with diseases influencing measurement by dual X-ray absorptiometry (DXA) or X-ray were excluded. Hence, patients were excluded if they presented hip arthroplasty of both sides and more than two fractured vertebrae in the lumbar spine in the region of L1 to L4. In addition, patients unable to answer questions concerning their medical history or fill out questionnaires were excluded.

Dual X-ray absorptiometry

DXA procedures to measure BMD were performed with a Delphi Hologic QDR-4500 device at the proximal femur or the lumbar spine. Calibration was performed with the European Spine Phantom.

Fracture

For vertebral fracture evaluation, lateral X-rays of the upper and lower spine were performed (Th4 to L5). Osteoporotic fractures were corrected for non-osteoporotic forms and quantitatively analyzed following an algorithm modified after Genant and Jergas (42). Aspects of spine deformity were graded according to the difference between the anterior to the medium or posterior height of the individual vertebra or the posterior height of directly adjacent upper or lower vertebrae. A vertebral fracture was defined with one of these indices as being <0.8, meaning a >20% reduction in height. Peripheral fractures were documented when taking the patient history, which included a fracture questionnaire.

Detection of falls

The history of falls was determined using a standard questionnaire, which included questions on falls during the last year.

Genotyping

Genomic DNA was isolated from venous blood samples using an automated solid-phase extraction method (EZ1 DNABlood 350-ml Kit) with the Bio-robot EZ1 (both from Qiagen, Hilden, Germany) according to the manufacturer's instructions. The SNPs were genotyped using a multiplex single-base primer extension. Preamplification was performed using a Qiagen Multiplex PCR Kit (Qiagen) according to the manufacturer's instructions. The primer extension reaction was carried out using a SNaPshot multiplex kit (Life Technology, Darmstadt, Germany), also according to the manufacturer's instructions. Details on the primers and reaction conditions used are available in Table S1. The reaction products were analyzed on an ABI Prism 3130xl genetic analyser (Life Technology), and the genotype was determined with GeneMapper v3.7 (Life Technology).

Thirty percent of all samples were genotyped in duplicate to verify the results. The control for the sample switching was performed as described previously (18, 57).

Statistics

Data were analyzed using Prism GraphPad 5 software (San Diego, USA) and IBM-SPSS software version 27. All data are presented as means (±SEM). Statistical significance was determined using an unpaired t test for cell-culture data (n = 3–6 biological replicates; n = 2 technical replicates, no data excluded). Patients’ data (all patients included) were tested for normal distribution (Kolmogorov–Smirnov), and either Kruskal–Wallis, Mann–Whitney U (non-normative distribution), or one-way ANOVA (normative distribution) group testing was applied. Pearson or spearman correlation and linear regression analysis of patient data was performed as well as applying univariate logistic regression models. All tests were performed two-tailed. The χ² or Fisher's exact test was used to determine differences between expected frequencies. The alpha level was set at P < 0.05.

Study approval

All patients, whose bone material was used for cell culture, provided informed consent. The institutional review board of Göttingen University Medical Center approved the study (09/05/01), which was conducted in accordance with good clinical practice and the Declaration of Helsinki. Patients who were included into the PSIOD study (Berlin part) provided written consent to the study after reading the patient information. The institutional review board of the Benjamin Franklin University Hospital, Free University of Berlin, approved the PSIOD study (Berlin part).

Supplementary Material

pgaf315_Supplementary_Data

pgaf315_supplementary_data.pdf^{(192KB, pdf)}

Acknowledgments

The authors thank A. Entwistle for his assistance with the English version of the manuscript.

Contributor Information

Martina Blaschke, Clinic of Gastroenterology, Gastrointestinal Oncology and Endocrinology, University Medical Center, 37075 Göttingen, Germany; Medical Care Center (MVZ) Endokrinologikum Göttingen, 37075 Göttingen, Germany.

Ina Dressel, Clinic of Gastroenterology, Gastrointestinal Oncology and Endocrinology, University Medical Center, 37075 Göttingen, Germany.

Regine Köpp, Clinic of Gastroenterology, Gastrointestinal Oncology and Endocrinology, University Medical Center, 37075 Göttingen, Germany.

Gabriele Armbrecht, Center for Muscle and Bone Research, Charité, 12203 Berlin, Germany.

Stephan Sehmisch, Clinic for Trauma Surgery, Orthopedics and Reconstructive Surgery, University Medical Center, 37075 Göttingen, Germany.

Mladen V Tzvetkov, Department of General Pharmacology, University Medicine Greifswald, 17487 Greifswald, Germany.

Frank Streit, Department of Clinical Chemistry, University Medical Center, 37075 Göttingen, Germany.

Holger M Reichardt, Institute for Cellular and Molecular Immunology, University Medical Center, 37075 Göttingen, Germany.

Claus-C Glüer, Biomedical Imaging Section, Department of Radiology and Neuroradiology, MOINCC, 24118 Kiel, Germany.

Heide Siggelkow, Clinic of Gastroenterology, Gastrointestinal Oncology and Endocrinology, University Medical Center, 37075 Göttingen, Germany; Medical Care Center (MVZ) Endokrinologikum Göttingen, 37075 Göttingen, Germany; Clinic for Trauma Surgery, Orthopedics and Reconstructive Surgery, University Medical Center, 37075 Göttingen, Germany.

Supplementary Material

Supplementary material is available at PNAS Nexus online.

Funding

This study was supported by grant no. 91 Elsbeth-Bonhoff-Foundation and in part by the German Research Foundation (DFG) with the grants TZ 74/4-1 and SI 493/7-1. The authors acknowledge support by the Open access publication funds of the Göttingen University.

Author Contributions

Martina Blaschke (Conceptualization, Formal analysis, Investigation, Writing—original draft, Writing—review & editing), Ina Dressel (Investigation), Regine Köpp (Data curation, Investigation), Gabriele Armbrecht (Data curation, Project administration, Writing—review & editing), Stephan Sehmisch (Methodology), Mladen V. Tzvetkov (Methodology, Project administration, Supervision, Writing—review & editing), Frank Streit (Methodology), Holger M. Reichardt (Investigation, Supervision, Writing—original draft), Claus-C. Glüer (Methodology, Supervision), and Heide Siggelkow (Conceptualization, Project administration, Supervision, Writing—original draft, Writing—review & editing)

Data Availability

All data associated with this study are present in the paper or the Supplementary Materials (Anonymous patients’ data are provided as excel file). Additional cell culture data will be available to all qualified academic researchers and provided under a material transfer agreement for research use only. The availability of clinical data is dependent on the patients’ informed consent.

References

1. Bruscoli S, Febo M, Riccardi C, Migliorati G. 2021. Glucocorticoid therapy in inflammatory bowel disease: mechanisms and clinical practice. Front Immunol. 12:691480. 10.3389/fimmu.2021.691480 [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Reichardt SD, et al. 2021. The role of glucocorticoids in inflammatory diseases. Cells. 10(11):2921. 10.3390/cells10112921 [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Meijer OC, Pereira AM. 2019. Three percent annually on systemic glucocorticoids: facts, worries and perspectives. Eur J Endocrinol. 181(6):C23–C28. 10.1530/EJE-19-0555 [DOI] [PubMed] [Google Scholar]
4. Laugesen K, Jorgensen JOL, Petersen I, Sorensen HT. 2019. Fifteen-year nationwide trends in systemic glucocorticoid drug use in Denmark. Eur J Endocrinol. 181(3):267–273. 10.1530/EJE-19-0305 [DOI] [PubMed] [Google Scholar]
5. den Uyl D, Bultink IE, Lems WF. 2011. Advances in glucocorticoid-induced osteoporosis. Curr Rheumatol Rep. 13(3):233–240. 10.1007/s11926-011-0173-y [DOI] [PMC free article] [PubMed] [Google Scholar]
6. van Staa TP, Leufkens HG, Cooper C. 2002. The epidemiology of corticosteroid-induced osteoporosis: a meta-analysis. Osteoporos Int. 13(10):777–787. [DOI] [PubMed] [Google Scholar]
7. Weinstein RS. 2011. Clinical practice. Glucocorticoid-induced bone disease. N Engl J Med. 365(1):62–70. [DOI] [PubMed] [Google Scholar]
8. Lane NE. 2019. Glucocorticoid-induced osteoporosis: new insights into the pathophysiology and treatments. Curr Osteoporos Rep. 17(1):1–7. 10.1007/s11914-019-00498-x [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Yao W, et al. 2008. Glucocorticoid excess in mice results in early activation of osteoclastogenesis and adipogenesis and prolonged suppression of osteogenesis: a longitudinal study of gene expression in bone tissue from glucocorticoid-treated mice. Arthritis Rheum. 58(6):1674–1686. 10.1002/art.23454 [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Fenton C, et al. 2021. Local steroid activation is a critical mediator of the anti-inflammatory actions of therapeutic glucocorticoids. Ann Rheum Dis. 80(2):250–260. 10.1136/annrheumdis-2020-218493 [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Diederich S, et al. 2002. 11beta-hydroxysteroid dehydrogenase types 1 and 2: an important pharmacokinetic determinant for the activity of synthetic mineralo- and glucocorticoids. J Clin Endocrinol Metab. 87(12):5695–5701. [DOI] [PubMed] [Google Scholar]
12. Tomlinson JW, et al. 2004. 11beta-hydroxysteroid dehydrogenase type 1: a tissue-specific regulator of glucocorticoid response. Endocr Rev. 25(5):831–866. [DOI] [PubMed] [Google Scholar]
13. Gathercole LL, Stewart PM. 2010. Targeting the pre-receptor metabolism of cortisol as a novel therapy in obesity and diabetes. J Steroid Biochem Mol Biol. 122(1–3):21–27. [DOI] [PubMed] [Google Scholar]
14. Nair S, et al. 2004. 11beta-Hydroxysteroid dehydrogenase type 1: genetic polymorphisms are associated with Type 2 diabetes in Pima Indians independently of obesity and expression in adipocyte and muscle. Diabetologia. 47(6):1088–1095. [DOI] [PubMed] [Google Scholar]
15. Stegk JP, Ebert B, Martin HJ, Maser E. 25 2009. Expression profiles of human 11beta-hydroxysteroid dehydrogenases type 1 and type 2 in inflammatory bowel diseases. Mol Cell Endocrinol. Mar. 301(1–2):104–108. 10.1016/j.mce.2008.10.030 [DOI] [Google Scholar]
16. Diederich S, et al. 2000. 11Beta-hydroxysteroid-dehydrogenase isoforms: tissue distribution and implications for clinical medicine. Eur J Clin Invest. 30(Suppl 3):21–27. [DOI] [PubMed] [Google Scholar]
17. Cooper MS, et al. 2002. Osteoblastic 11beta-hydroxysteroid dehydrogenase type 1 activity increases with age and glucocorticoid exposure. J Bone Miner Res. 17(6):979–986. [DOI] [PubMed] [Google Scholar]
18. Siggelkow H, et al. 2014. Genetic polymorphisms in 11beta-hydroxysteroid dehydrogenase type 1 correlate with the postdexamethasone cortisol levels and bone mineral density in patients evaluated for osteoporosis. J Clin Endocrinol Metab. 99(2):E293–E302. [DOI] [PubMed] [Google Scholar]
19. Hwang JY, et al. 2009. HSD11B1 polymorphisms predicted bone mineral density and fracture risk in postmenopausal women without a clinically apparent hypercortisolemia. Bone. 45(6):1098–1103. [DOI] [PubMed] [Google Scholar]
20. Meunier P, Aaron J, Edouard C, Vignon G. 1971. Osteoporosis and the replacement of cell populations of the marrow by adipose tissue. A quantitative study of 84 iliac bone biopsies. Clin Orthop. 80(147):147–154. [DOI] [PubMed] [Google Scholar]
21. Gimble JM, Robinson CE, Wu X, Kelly KA. 1996. The function of adipocytes in the bone marrow stroma: an update. Bone. 19(5):421–428. [DOI] [PubMed] [Google Scholar]
22. Nuttall ME, Gimble JM. 2000. Is there a therapeutic opportunity to either prevent or treat osteopenic disorders by inhibiting marrow adipogenesis? Bone. 27(2):177–184. [DOI] [PubMed] [Google Scholar]
23. Justesen J, et al. 2001. Adipocyte tissue volume in bone marrow is increased with aging and in patients with osteoporosis. Biogerontology. 2(3):165–171. 10.1023/a:1011513223894 [DOI] [PubMed] [Google Scholar]
24. Ponce ML, et al. 2008. Coexpression of osteogenic and adipogenic differentiation markers in selected subpopulations of primary human mesenchymal progenitor cells. J Cell Biochem. 104(4):1342–1355. [DOI] [PubMed] [Google Scholar]
25. Blaschke M, et al. 2021. The rise in expression and activity of 11beta-HSD1 in human mesenchymal progenitor cells induces adipogenesis through increased local cortisol synthesis. J Steroid Biochem Mol Biol. 210:105850. 10.1016/j.jsbmb.2021.105850 [DOI] [PubMed] [Google Scholar]
26. Ferrau F, et al. 2020. High bone marrow fat in patients with Cushing's syndrome and vertebral fractures. Endocrine. 67(1):172–179. 10.1007/s12020-019-02034-4 [DOI] [PubMed] [Google Scholar]
27. Diederich S, et al. 2004. Pharmacodynamics and pharmacokinetics of synthetic mineralocorticoids and glucocorticoids: receptor transactivation and prereceptor metabolism by 11beta-hydroxysteroid-dehydrogenases. Horm Metab Res. 36(6):423–429. [DOI] [PubMed] [Google Scholar]
28. Cooper MS, et al. 2001. Modulation of 11beta-hydroxysteroid dehydrogenase isozymes by proinflammatory cytokines in osteoblasts: an autocrine switch from glucocorticoid inactivation to activation. J Bone Miner Res. 16(6):1037–1044. [DOI] [PubMed] [Google Scholar]
29. Cooper MS, et al. 2000. Expression and functional consequences of 11beta-hydroxysteroid dehydrogenase activity in human bone. Bone. 27(3):375–381. [DOI] [PubMed] [Google Scholar]
30. Bujalska IJ, et al. 2005. Hexose-6-phosphate dehydrogenase confers oxo-reductase activity upon 11 beta-hydroxysteroid dehydrogenase type 1. J Mol Endocrinol. 34(3):675–684. 10.1677/jme.1.01718 [DOI] [PubMed] [Google Scholar]
31. Chapman K, Holmes M, Seckl J. 2013. 11beta-hydroxysteroid dehydrogenases: intracellular gate-keepers of tissue glucocorticoid action. Physiol Rev. 93(3):1139–1206. 10.1152/physrev.00020.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
32. Lakshmi V, Monder C. 1988. Purification and characterization of the corticosteroid 11 beta-dehydrogenase component of the rat liver 11 beta-hydroxysteroid dehydrogenase complex. Endocrinology. 123(5):2390–2398. 10.1210/endo-123-5-2390 [DOI] [PubMed] [Google Scholar]
33. Siggelkow H, et al. 2009. Erythrocyte sedimentation rate as an osteoporosis risk factor in patients with active Crohn's disease. Osteology. 18:209–216. [Google Scholar]
34. Blaschke M, et al. 2018. IL-6, IL-1beta, and TNF-alpha only in combination influence the osteoporotic phenotype in Crohn's patients via bone formation and bone resorption. Adv Clin Exp Med. 27(1):45–56. 10.17219/acem/67561 [DOI] [PubMed] [Google Scholar]
35. Blaschke M, et al. 2018. Crohn's disease patient serum changes protein expression in a human mesenchymal stem cell model in a linear relationship to patients’ disease stage and to bone mineral density. J Clin Transl Endocrinol. 13:26–38. 10.1016/j.jcte.2018.06.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Li J, et al. 2006. Dkk1-mediated inhibition of Wnt signaling in bone results in osteopenia. Bone. 39(4):754–766. 10.1016/j.bone.2006.03.017 [DOI] [PubMed] [Google Scholar]
37. Siggelkow H, et al. 1999. Development of the osteoblast phenotype in primary human osteoblasts in culture: comparison with rat calvarial cells in osteoblast differentiation. J Cell Biochem. 75(1):22–35. [PubMed] [Google Scholar]
38. Siggelkow H, et al. 1998. In vitro differentiation potential of a new human osteosarcoma cell line (HOS 58). Differentiation. 63(2):81–91. [DOI] [PubMed] [Google Scholar]
39. Siggelkow H, Justesen J, Clausen C, Kassem M. 2003. Aging is associated with decreased maximal life span and accelerated senescence of bone marrow stromal cells. Bone. 33(6):919–926. [DOI] [PubMed] [Google Scholar]
40. Siggelkow H, et al. 1989. Relationships between histomorphometric features of bone formation and bone cell characteristics in vitro in renal osteodystrophy. J Clin Endocrinol Metab. 69(6):1166–1173. 10.1210/jcem-69-6-1166 [DOI] [PubMed] [Google Scholar]
41. Pierotti S, Gandini L, Lenzi A, Isidori AM. 2008. Pre-receptorial regulation of steroid hormones in bone cells: insights on glucocorticoid-induced osteoporosis. J Steroid Biochem Mol Biol. 108(3-5):292–299. [DOI] [PubMed] [Google Scholar]
42. Genant HK, Jergas M. 2003. Assessment of prevalent and incident vertebral fractures in osteoporosis research. Osteoporos Int. 14(Suppl 3):S43–S55. 10.1007/s00198-002-1348-1 [DOI] [Google Scholar]
43. Weinstein RS, Jilka RL, Parfitt AM, Manolagas SC. 1998. Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. J Clin Invest. 102(2):274–282. [DOI] [PMC free article] [PubMed] [Google Scholar]
44. Fenton CG, et al. 2022. 11beta-Hydroxysteroid dehydrogenase type 1 within osteoclasts mediates the bone protective properties of therapeutic corticosteroids in chronic inflammation. Int J Mol Sci. 23(13):7334. 10.3390/ijms23137334 [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Kollmer M, Buhrman JS, Zhang Y, Gemeinhart RA. 2013. Markers are shared between adipogenic and osteogenic differentiated mesenchymal stem cells. J Dev Biol Tissue Eng. 5(2):18–25. 10.5897/JDBTE2013.0065 [DOI] [PMC free article] [PubMed] [Google Scholar]
46. Burns DM, et al. 2022. Discovery of a novel 2-spiroproline steroid mimetic scaffold for the potent inhibition of 11β-HSD1. Bioorg Med Chem Lett. 73:128884. 10.1016/j.bmcl.2022.128884 [DOI] [PubMed] [Google Scholar]
47. Zhang C, et al. 2022. Discovery of 1'-(1-phenylcyclopropane-carbonyl)-3H-spiro[isobenzofuran-1,3'-pyrrolidin]-3-one as a novel steroid mimetic scaffold for the potent and tissue-specific inhibition of 11β-HSD1 using a scaffold-hopping approach. Bioorg Med Chem Lett. 69:128782. 10.1016/j.bmcl.2022.128782 [DOI] [PubMed] [Google Scholar]
48. Tomlinson JW, et al. 2002. Absence of cushingoid phenotype in a patient with Cushing's disease due to defective cortisone to cortisol conversion. J Clin Endocrinol Metab. 87(1):57–62. 10.1210/jcem.87.1.8189 [DOI] [PubMed] [Google Scholar]
49. Abbas A, et al. 2022. Effect of AZD4017, a selective 11β-HSD1 inhibitor, on bone turnover markers in postmenopausal osteopenia. J Clin Endocrinol Metab. 107(7):2026–2035. 10.1210/clinem/dgac100 [DOI] [PMC free article] [PubMed] [Google Scholar]
50. Othonos N, et al. 2023. 11β-HSD1 inhibition in men mitigates prednisolone-induced adverse effects in a proof-of-concept randomised double-blind placebo-controlled trial. Nat Commun. 14(1):1025. 10.1038/s41467-023-36541-w [DOI] [PMC free article] [PubMed] [Google Scholar]
51. Siggelkow H, et al. 2003. Cytokines, osteoprotegerin, and RANKL in vitro and histomorphometric indices of bone turnover in patients with different bone diseases. J Bone Miner Res. 18(3):529–538. [DOI] [PubMed] [Google Scholar]
52. Siggelkow H, Schmidt E, Hennies B, Hufner M. 2004. Evidence of downregulation of matrix extracellular phosphoglycoprotein during terminal differentiation in human osteoblasts. Bone. 35(2):570–576. [DOI] [PubMed] [Google Scholar]
53. Bocker W, et al. 2008. Introducing a single-cell-derived human mesenchymal stem cell line expressing hTERT after lentiviral gene transfer. J Cell Mol Med. 12(4):1347–1359. [DOI] [PMC free article] [PubMed] [Google Scholar]
54. Livak KJ, Schmittgen TD. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)). Method. Methods. 25(4):402–408. 10.1006/meth.2001.1262 [DOI] [PubMed] [Google Scholar]
55. Streit F, Armstrong VW, Oellerich M. 2002. Rapid liquid chromatography-tandem mass spectrometry routine method for simultaneous determination of sirolimus, everolimus, tacrolimus, and cyclosporin A in whole blood. Clin Chem. Jun. 48(6 Pt 1):955–958. [Google Scholar]
56. Forth W, Henschler D, Rummel D. Allgemeine und spezielle Pharmakologie und Toxikologie. 27th ed. Brockhaus BI, Brockhaus FA, Mannheim, 1987. p. 591–630. [Google Scholar]
57. Tzvetkov MV, Meineke I, Sehrt D, Vormfelde SV, Brockmoller J. 2010. Amelogenin-based sex identification as a strategy to control the identity of DNA samples in genetic association studies. Pharmacogenomics. 11(3):449–457. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

pgaf315_Supplementary_Data

pgaf315_supplementary_data.pdf^{(192KB, pdf)}

Data Availability Statement

[pgaf315-B1] 1. Bruscoli S, Febo M, Riccardi C, Migliorati G. 2021. Glucocorticoid therapy in inflammatory bowel disease: mechanisms and clinical practice. Front Immunol. 12:691480. 10.3389/fimmu.2021.691480 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B2] 2. Reichardt SD, et al. 2021. The role of glucocorticoids in inflammatory diseases. Cells. 10(11):2921. 10.3390/cells10112921 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B3] 3. Meijer OC, Pereira AM. 2019. Three percent annually on systemic glucocorticoids: facts, worries and perspectives. Eur J Endocrinol. 181(6):C23–C28. 10.1530/EJE-19-0555 [DOI] [PubMed] [Google Scholar]

[pgaf315-B4] 4. Laugesen K, Jorgensen JOL, Petersen I, Sorensen HT. 2019. Fifteen-year nationwide trends in systemic glucocorticoid drug use in Denmark. Eur J Endocrinol. 181(3):267–273. 10.1530/EJE-19-0305 [DOI] [PubMed] [Google Scholar]

[pgaf315-B5] 5. den Uyl D, Bultink IE, Lems WF. 2011. Advances in glucocorticoid-induced osteoporosis. Curr Rheumatol Rep. 13(3):233–240. 10.1007/s11926-011-0173-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B6] 6. van Staa TP, Leufkens HG, Cooper C. 2002. The epidemiology of corticosteroid-induced osteoporosis: a meta-analysis. Osteoporos Int. 13(10):777–787. [DOI] [PubMed] [Google Scholar]

[pgaf315-B7] 7. Weinstein RS. 2011. Clinical practice. Glucocorticoid-induced bone disease. N Engl J Med. 365(1):62–70. [DOI] [PubMed] [Google Scholar]

[pgaf315-B8] 8. Lane NE. 2019. Glucocorticoid-induced osteoporosis: new insights into the pathophysiology and treatments. Curr Osteoporos Rep. 17(1):1–7. 10.1007/s11914-019-00498-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B9] 9. Yao W, et al. 2008. Glucocorticoid excess in mice results in early activation of osteoclastogenesis and adipogenesis and prolonged suppression of osteogenesis: a longitudinal study of gene expression in bone tissue from glucocorticoid-treated mice. Arthritis Rheum. 58(6):1674–1686. 10.1002/art.23454 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B10] 10. Fenton C, et al. 2021. Local steroid activation is a critical mediator of the anti-inflammatory actions of therapeutic glucocorticoids. Ann Rheum Dis. 80(2):250–260. 10.1136/annrheumdis-2020-218493 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B11] 11. Diederich S, et al. 2002. 11beta-hydroxysteroid dehydrogenase types 1 and 2: an important pharmacokinetic determinant for the activity of synthetic mineralo- and glucocorticoids. J Clin Endocrinol Metab. 87(12):5695–5701. [DOI] [PubMed] [Google Scholar]

[pgaf315-B12] 12. Tomlinson JW, et al. 2004. 11beta-hydroxysteroid dehydrogenase type 1: a tissue-specific regulator of glucocorticoid response. Endocr Rev. 25(5):831–866. [DOI] [PubMed] [Google Scholar]

[pgaf315-B13] 13. Gathercole LL, Stewart PM. 2010. Targeting the pre-receptor metabolism of cortisol as a novel therapy in obesity and diabetes. J Steroid Biochem Mol Biol. 122(1–3):21–27. [DOI] [PubMed] [Google Scholar]

[pgaf315-B14] 14. Nair S, et al. 2004. 11beta-Hydroxysteroid dehydrogenase type 1: genetic polymorphisms are associated with Type 2 diabetes in Pima Indians independently of obesity and expression in adipocyte and muscle. Diabetologia. 47(6):1088–1095. [DOI] [PubMed] [Google Scholar]

[pgaf315-B15] 15. Stegk JP, Ebert B, Martin HJ, Maser E. 25 2009. Expression profiles of human 11beta-hydroxysteroid dehydrogenases type 1 and type 2 in inflammatory bowel diseases. Mol Cell Endocrinol. Mar. 301(1–2):104–108. 10.1016/j.mce.2008.10.030 [DOI] [Google Scholar]

[pgaf315-B16] 16. Diederich S, et al. 2000. 11Beta-hydroxysteroid-dehydrogenase isoforms: tissue distribution and implications for clinical medicine. Eur J Clin Invest. 30(Suppl 3):21–27. [DOI] [PubMed] [Google Scholar]

[pgaf315-B17] 17. Cooper MS, et al. 2002. Osteoblastic 11beta-hydroxysteroid dehydrogenase type 1 activity increases with age and glucocorticoid exposure. J Bone Miner Res. 17(6):979–986. [DOI] [PubMed] [Google Scholar]

[pgaf315-B18] 18. Siggelkow H, et al. 2014. Genetic polymorphisms in 11beta-hydroxysteroid dehydrogenase type 1 correlate with the postdexamethasone cortisol levels and bone mineral density in patients evaluated for osteoporosis. J Clin Endocrinol Metab. 99(2):E293–E302. [DOI] [PubMed] [Google Scholar]

[pgaf315-B19] 19. Hwang JY, et al. 2009. HSD11B1 polymorphisms predicted bone mineral density and fracture risk in postmenopausal women without a clinically apparent hypercortisolemia. Bone. 45(6):1098–1103. [DOI] [PubMed] [Google Scholar]

[pgaf315-B20] 20. Meunier P, Aaron J, Edouard C, Vignon G. 1971. Osteoporosis and the replacement of cell populations of the marrow by adipose tissue. A quantitative study of 84 iliac bone biopsies. Clin Orthop. 80(147):147–154. [DOI] [PubMed] [Google Scholar]

[pgaf315-B21] 21. Gimble JM, Robinson CE, Wu X, Kelly KA. 1996. The function of adipocytes in the bone marrow stroma: an update. Bone. 19(5):421–428. [DOI] [PubMed] [Google Scholar]

[pgaf315-B22] 22. Nuttall ME, Gimble JM. 2000. Is there a therapeutic opportunity to either prevent or treat osteopenic disorders by inhibiting marrow adipogenesis? Bone. 27(2):177–184. [DOI] [PubMed] [Google Scholar]

[pgaf315-B23] 23. Justesen J, et al. 2001. Adipocyte tissue volume in bone marrow is increased with aging and in patients with osteoporosis. Biogerontology. 2(3):165–171. 10.1023/a:1011513223894 [DOI] [PubMed] [Google Scholar]

[pgaf315-B24] 24. Ponce ML, et al. 2008. Coexpression of osteogenic and adipogenic differentiation markers in selected subpopulations of primary human mesenchymal progenitor cells. J Cell Biochem. 104(4):1342–1355. [DOI] [PubMed] [Google Scholar]

[pgaf315-B25] 25. Blaschke M, et al. 2021. The rise in expression and activity of 11beta-HSD1 in human mesenchymal progenitor cells induces adipogenesis through increased local cortisol synthesis. J Steroid Biochem Mol Biol. 210:105850. 10.1016/j.jsbmb.2021.105850 [DOI] [PubMed] [Google Scholar]

[pgaf315-B26] 26. Ferrau F, et al. 2020. High bone marrow fat in patients with Cushing's syndrome and vertebral fractures. Endocrine. 67(1):172–179. 10.1007/s12020-019-02034-4 [DOI] [PubMed] [Google Scholar]

[pgaf315-B27] 27. Diederich S, et al. 2004. Pharmacodynamics and pharmacokinetics of synthetic mineralocorticoids and glucocorticoids: receptor transactivation and prereceptor metabolism by 11beta-hydroxysteroid-dehydrogenases. Horm Metab Res. 36(6):423–429. [DOI] [PubMed] [Google Scholar]

[pgaf315-B28] 28. Cooper MS, et al. 2001. Modulation of 11beta-hydroxysteroid dehydrogenase isozymes by proinflammatory cytokines in osteoblasts: an autocrine switch from glucocorticoid inactivation to activation. J Bone Miner Res. 16(6):1037–1044. [DOI] [PubMed] [Google Scholar]

[pgaf315-B29] 29. Cooper MS, et al. 2000. Expression and functional consequences of 11beta-hydroxysteroid dehydrogenase activity in human bone. Bone. 27(3):375–381. [DOI] [PubMed] [Google Scholar]

[pgaf315-B30] 30. Bujalska IJ, et al. 2005. Hexose-6-phosphate dehydrogenase confers oxo-reductase activity upon 11 beta-hydroxysteroid dehydrogenase type 1. J Mol Endocrinol. 34(3):675–684. 10.1677/jme.1.01718 [DOI] [PubMed] [Google Scholar]

[pgaf315-B31] 31. Chapman K, Holmes M, Seckl J. 2013. 11beta-hydroxysteroid dehydrogenases: intracellular gate-keepers of tissue glucocorticoid action. Physiol Rev. 93(3):1139–1206. 10.1152/physrev.00020.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B32] 32. Lakshmi V, Monder C. 1988. Purification and characterization of the corticosteroid 11 beta-dehydrogenase component of the rat liver 11 beta-hydroxysteroid dehydrogenase complex. Endocrinology. 123(5):2390–2398. 10.1210/endo-123-5-2390 [DOI] [PubMed] [Google Scholar]

[pgaf315-B33] 33. Siggelkow H, et al. 2009. Erythrocyte sedimentation rate as an osteoporosis risk factor in patients with active Crohn's disease. Osteology. 18:209–216. [Google Scholar]

[pgaf315-B34] 34. Blaschke M, et al. 2018. IL-6, IL-1beta, and TNF-alpha only in combination influence the osteoporotic phenotype in Crohn's patients via bone formation and bone resorption. Adv Clin Exp Med. 27(1):45–56. 10.17219/acem/67561 [DOI] [PubMed] [Google Scholar]

[pgaf315-B35] 35. Blaschke M, et al. 2018. Crohn's disease patient serum changes protein expression in a human mesenchymal stem cell model in a linear relationship to patients’ disease stage and to bone mineral density. J Clin Transl Endocrinol. 13:26–38. 10.1016/j.jcte.2018.06.002 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B36] 36. Li J, et al. 2006. Dkk1-mediated inhibition of Wnt signaling in bone results in osteopenia. Bone. 39(4):754–766. 10.1016/j.bone.2006.03.017 [DOI] [PubMed] [Google Scholar]

[pgaf315-B37] 37. Siggelkow H, et al. 1999. Development of the osteoblast phenotype in primary human osteoblasts in culture: comparison with rat calvarial cells in osteoblast differentiation. J Cell Biochem. 75(1):22–35. [PubMed] [Google Scholar]

[pgaf315-B38] 38. Siggelkow H, et al. 1998. In vitro differentiation potential of a new human osteosarcoma cell line (HOS 58). Differentiation. 63(2):81–91. [DOI] [PubMed] [Google Scholar]

[pgaf315-B39] 39. Siggelkow H, Justesen J, Clausen C, Kassem M. 2003. Aging is associated with decreased maximal life span and accelerated senescence of bone marrow stromal cells. Bone. 33(6):919–926. [DOI] [PubMed] [Google Scholar]

[pgaf315-B40] 40. Siggelkow H, et al. 1989. Relationships between histomorphometric features of bone formation and bone cell characteristics in vitro in renal osteodystrophy. J Clin Endocrinol Metab. 69(6):1166–1173. 10.1210/jcem-69-6-1166 [DOI] [PubMed] [Google Scholar]

[pgaf315-B41] 41. Pierotti S, Gandini L, Lenzi A, Isidori AM. 2008. Pre-receptorial regulation of steroid hormones in bone cells: insights on glucocorticoid-induced osteoporosis. J Steroid Biochem Mol Biol. 108(3-5):292–299. [DOI] [PubMed] [Google Scholar]

[pgaf315-B42] 42. Genant HK, Jergas M. 2003. Assessment of prevalent and incident vertebral fractures in osteoporosis research. Osteoporos Int. 14(Suppl 3):S43–S55. 10.1007/s00198-002-1348-1 [DOI] [Google Scholar]

[pgaf315-B43] 43. Weinstein RS, Jilka RL, Parfitt AM, Manolagas SC. 1998. Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. J Clin Invest. 102(2):274–282. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B44] 44. Fenton CG, et al. 2022. 11beta-Hydroxysteroid dehydrogenase type 1 within osteoclasts mediates the bone protective properties of therapeutic corticosteroids in chronic inflammation. Int J Mol Sci. 23(13):7334. 10.3390/ijms23137334 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B45] 45. Kollmer M, Buhrman JS, Zhang Y, Gemeinhart RA. 2013. Markers are shared between adipogenic and osteogenic differentiated mesenchymal stem cells. J Dev Biol Tissue Eng. 5(2):18–25. 10.5897/JDBTE2013.0065 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B46] 46. Burns DM, et al. 2022. Discovery of a novel 2-spiroproline steroid mimetic scaffold for the potent inhibition of 11β-HSD1. Bioorg Med Chem Lett. 73:128884. 10.1016/j.bmcl.2022.128884 [DOI] [PubMed] [Google Scholar]

[pgaf315-B47] 47. Zhang C, et al. 2022. Discovery of 1'-(1-phenylcyclopropane-carbonyl)-3H-spiro[isobenzofuran-1,3'-pyrrolidin]-3-one as a novel steroid mimetic scaffold for the potent and tissue-specific inhibition of 11β-HSD1 using a scaffold-hopping approach. Bioorg Med Chem Lett. 69:128782. 10.1016/j.bmcl.2022.128782 [DOI] [PubMed] [Google Scholar]

[pgaf315-B48] 48. Tomlinson JW, et al. 2002. Absence of cushingoid phenotype in a patient with Cushing's disease due to defective cortisone to cortisol conversion. J Clin Endocrinol Metab. 87(1):57–62. 10.1210/jcem.87.1.8189 [DOI] [PubMed] [Google Scholar]

[pgaf315-B49] 49. Abbas A, et al. 2022. Effect of AZD4017, a selective 11β-HSD1 inhibitor, on bone turnover markers in postmenopausal osteopenia. J Clin Endocrinol Metab. 107(7):2026–2035. 10.1210/clinem/dgac100 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B50] 50. Othonos N, et al. 2023. 11β-HSD1 inhibition in men mitigates prednisolone-induced adverse effects in a proof-of-concept randomised double-blind placebo-controlled trial. Nat Commun. 14(1):1025. 10.1038/s41467-023-36541-w [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B51] 51. Siggelkow H, et al. 2003. Cytokines, osteoprotegerin, and RANKL in vitro and histomorphometric indices of bone turnover in patients with different bone diseases. J Bone Miner Res. 18(3):529–538. [DOI] [PubMed] [Google Scholar]

[pgaf315-B52] 52. Siggelkow H, Schmidt E, Hennies B, Hufner M. 2004. Evidence of downregulation of matrix extracellular phosphoglycoprotein during terminal differentiation in human osteoblasts. Bone. 35(2):570–576. [DOI] [PubMed] [Google Scholar]

[pgaf315-B53] 53. Bocker W, et al. 2008. Introducing a single-cell-derived human mesenchymal stem cell line expressing hTERT after lentiviral gene transfer. J Cell Mol Med. 12(4):1347–1359. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgaf315-B54] 54. Livak KJ, Schmittgen TD. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)). Method. Methods. 25(4):402–408. 10.1006/meth.2001.1262 [DOI] [PubMed] [Google Scholar]

[pgaf315-B55] 55. Streit F, Armstrong VW, Oellerich M. 2002. Rapid liquid chromatography-tandem mass spectrometry routine method for simultaneous determination of sirolimus, everolimus, tacrolimus, and cyclosporin A in whole blood. Clin Chem. Jun. 48(6 Pt 1):955–958. [Google Scholar]

[pgaf315-B56] 56. Forth W, Henschler D, Rummel D. Allgemeine und spezielle Pharmakologie und Toxikologie. 27th ed. Brockhaus BI, Brockhaus FA, Mannheim, 1987. p. 591–630. [Google Scholar]

[pgaf315-B57] 57. Tzvetkov MV, Meineke I, Sehrt D, Vormfelde SV, Brockmoller J. 2010. Amelogenin-based sex identification as a strategy to control the identity of DNA samples in genetic association studies. Pharmacogenomics. 11(3):449–457. [DOI] [PubMed] [Google Scholar]

PERMALINK

The local inactivation of glucocorticoid as regulator for the development of glucocorticoid-induced osteoporosis

Martina Blaschke

Ina Dressel

Regine Köpp

Gabriele Armbrecht

Stephan Sehmisch

Mladen V Tzvetkov

Frank Streit

Holger M Reichardt

Claus-C Glüer

Heide Siggelkow

Roles

Abstract

Significance Statement.

Introduction

Results

Regulation of 11β-HSD1 expression by pro-inflammatory cytokines

Fig. 1.

Conversion of GCs during osteogenic differentiation of SCP-1 cells

Fig. 2.

Influence of prednisone activation in SCP-1 cells and human mesenchymal progenitor cells on osteogenic differentiation

Fig. 3.

Influence of prednisolone inactivation in SCP-1 cells and hMSCs on osteogenic differentiation

Fig. 4.

Prednisolone or methylprednisolone therapy in patients

Table 1.

Fig. 5.

Discussion

Materials and methods

Cell culture

Isolation and culture of human mesenchymal progenitor cells and SCP-1 cells

Osteogenic differentiation

RNA isolation and cDNA synthesis

Pharmacological manipulations

Analysis of GC concentrations by LC–MS/MS

Application of cytokines and cytokine-neutralizing antibodies

Patients

Dual X-ray absorptiometry

Fracture

Detection of falls

Genotyping

Statistics

Study approval

Supplementary Material

Acknowledgments

Contributor Information

Supplementary Material

Funding

Author Contributions

Data Availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases