Role of Fibroblast Growth Factor 21 (FGF21) in Undernutrition-Related ...

GROWTH

FACTORS-CYTOKINES

Role of Fibroblast Growth Factor 21 (FGF21) in Undernutrition-Related Attenuation of Growth in Mice Rita Ann Kubicky,* Shufang Wu,* Alexei Kharitonenkov, and Francesco De Luca Section of Endocrinology and Diabetes (R.A.K., S.W., F.D.L.), St. Christopher’s Hospital for Children, Drexel University College of Medicine, Philadelphia, Pennsylvania 19134; and Lilly Research Laboratories (A.K.), Indianapolis, Indiana 46285

Reduced caloric intake in mammals causes reduced skeletal growth and GH insensitivity. However, the underlying molecular mechanisms are not fully elucidated. The aim of this study was to determine whether the increased activity of fibroblast growth factor 21 (FGF21) during chronic undernutrition in mice causes GH insensitivity and growth failure. After 4 wk of food restriction, fgf21 knockout (KO) mice exhibited greater body and tibial growth than their wild-type (WT) littermates. Daily injections of recombinant human FGF21 in a subgroup of food-restricted fgf21 KO mice prevented these differences in body and tibial growth. At the end of the 4-wk food restriction, GH binding and GH receptor expression were reduced in the liver and in the growth plate of food-restricted WT mice (compared to WT mice fed ad libitum), whereas they were similar between food-restricted and ad libitum KO mice. In addition, a single injection of GH induced greater liver signal transducer and activator of transcription 5 phosphorylation and IGF-I mRNA in food-restricted KO mice than in WT mice. Lastly, in the tibial growth plate of food-restricted WT mice, FGF21 mRNA and protein expression was greater than that of WT mice fed ad libitum. In contrast, the IGF-I mRNA and protein expression was smaller. Our findings support a causative role for FGF21 in the inhibition of skeletal growth during prolonged undernutrition. Such role may be mediated by the antagonistic effect of FGF21 on GH action in the liver and, possibly, in the growth plate. (Endocrinology 153: 2287–2295, 2012)

t is well known that insufficient caloric intake inhibits skeletal growth. However, the underlying mechanisms are not fully elucidated. In rodents, short-term fasting leads to reduced GH secretion (1) as well as hepatic GH sensitivity, with the impaired GH action being primarily due to decreased GH receptor (GHR) mRNA expression in the liver (2– 4). In humans, malnutrition also induces liver GH insensitivity but, unlike rodents, is associated with increased circulating GH levels (5–7). With respect to the causative mechanisms of GH insensitivity, recent evidence suggests an important role for fibroblast growth factor 21 (FGF21).

I

FGF21 is a member of a subfamily of FGFs (which includes FGF15/19 and FGF23) (8, 9), which lack the FGF heparin-binding domain. Thus, these FGFs can diffuse away from their tissue of synthesis and function as endocrine factors. During fasting, increased expression of FGF21 induces gluconeogenesis, fatty acid oxidation, and ketogenesis. As a result, FGF21 is considered a key regulator of the metabolic adaptation to fasting (10 –12). In addition, it has been shown that transgenic mice overexpressing FGF21 exhibit reduced bone growth and hepatic GH insensitivity (13). Based on this evidence, we hypothesized that the increased activity of FGF21 during chronic undernutrition

ISSN Print 0013-7227 ISSN Online 1945-7170 Printed in U.S.A. Copyright © 2012 by The Endocrine Society doi: 10.1210/en.2011-1909 Received October 25, 2011. Accepted February 2, 2012. First Published Online February 28, 2012

* R.A.K. and S.W. contributed equally to this work. Abbreviations: C.V., Coefficient of variation; FFA, free fatty acid; FGF21, fibroblast growth factor 21; FGFR, fibroblast growth factor receptor; GHR, GH receptor; KO, knockout; KO-AL, Fgf21 KO mice fed ad libitum; KO-FR, Fgf21 KO mice food restricted; rh, recombinant human; STAT, signal transducer and activator of transcription; WT, wild type; wt, weight; WT-AL, WT mice fed ad libitum; WT-FR, WT mice food restricted.

Endocrinology, May 2012, 153(5):2287–2295

endo.endojournals.org

2287

2288

Kubicky et al.

FGF21 Causes Growth Attenuation

may cause GH insensitivity and growth failure. To test our hypothesis, we used mice with a targeted disruption of the Fgf21 locus and exposed them to restricted food intake for 4 wk. During the study period, we monitored the longitudinal bone growth, growth plate chondrogenesis, and GH action in the knockout (KO) mice and compared them with those of wild-type (WT) mice. Our findings indicate that the lack of FGF21 expression in KO mice attenuates their food restriction-associated diminished bone growth and growth plate chondrogenesis, and GH insensitivity.

Materials and Methods Generation of mice with targeted disruption of Fgf21 To generate mice with targeted disruption of the Fgf21 locus, a 6.5-kb NheI genomic fragment containing all three exons of the mouse Fgf21 gene was subcloned and used as a gene targeting vector by replacing part of exon 1 (30 bp downstream of the ATG), all of exon 2, and the 5⬘ region of exon 3 with a neomycin resistance gene (pGTN29; New England Biolabs, Ipswich, MA), thus deleting approximately 1200 bp of the genomic Fgf21 sequence and deleting the 3⬘ part of exon 1, all of exon 2, and the 5⬘ region of exon 3. Founder mice were subsequently backcrossed onto the C57/BL6 line at least 10 times before investigation.

Animal care Animal care was in accordance with the Guide for the Care and Use of Laboratory Animals [DHEW Publication No. (NIH) 85-23, revised 1988]. All procedures were approved by the Institutional Animal Care and Use Committee of Drexel University College of Medicine. Homozygous Fgf21 KO mice were crossbred with WT C57/BL6 mice; heterozygous Fgf21 KO mice (F1) were then crossbred to obtain homozygous WT and homozygous Fgf21 KO mice for subsequent investigation. Mice were maintained in a temperature (22 C)-, humidity-, and light (12-h light, 12-h dark cycle)-controlled environment. Three-week-old KO and WT mice were housed separately and allowed ad libitum access to chow and water for 1 wk (“prestudy”). During this week, the mean daily food consumption for the KO and WT mice was measured. At the beginning of the study period (4 wk), mice were assigned to four groups: two groups [WT mice fed ad libitum (WT-AL) and Fgf21 KO fed ad libitum (KO-AL)] fed ad libitum with standard chow and two groups (WT mice food restricted and Fgf21 mice KO food restricted) fed with a special diet (5001; containing two times the amount of vitamins and minerals compared with standard chow; Animal Specialties And Provisions, LLC., Quakertown, PA) provided at 50% of the amount of food consumed during the prestudy by WT and KO mice, respectively. At the end of each week, the mean daily amount of chow consumed by the WT-AL and KO-ad lib mice was calculated separately; at the beginning of the next week, WT-restricted and KO-restricted mice were given 50% of the mean daily amount of food consumed in the previous week by the WT-AL and KO-AL, respectively. A subgroup of KO-restricted mice received recombinant human


FGF21 [5 ␮g/g weight (wt)] or vehicle administered ip daily for the whole duration of the study. On the last day of the study, subgroups of WT and KO mice received an ip injection of recombinant mouse GH (5 ␮g/g wt) or vehicle; 15 min after the injection, all mice (injected and noninjected) were euthanized and their livers removed and stored at ⫺80 C for subsequent analysis. At the end of the 4-wk study period, blood was collected by ventricular puncture from each animal before euthanasia. Serum samples were obtained and stored at ⫺20 C for FGF21, GH, IGF-I, ␤-hydroxybutyrate, and free fatty acids (FFA) analysis. Weight and whole body length (from nose to anus) were measured weekly during the 4-wk study period after the animals were sedated with an injection of ketamine hydrochloride (0.02 mg/ 100 g body weight) (Phoenix Pharmaceutical, Inc., Burlingame, CA) and acepromazine (0.1 mg/100 g body weight) (Roche Applied Science, Indianapolis, IN). Serial radiographs of the tibiae were obtained weekly, with each animal placed prone on an x-ray cassette. The lengths of both tibiae were measured on the radiograph, and the average value was calculated.

Measurement of serum hormones and metabolites After collection, serum was stored at ⫺80 C before analysis. GH levels were determined in duplicate by ELISA (Millipore Corp., Billerica, MA) with intra- and interassay coefficient of variation (C.V.) of 1.7– 4.3 and 3.2– 4.9%, respectively. FGF21 levels were determined in duplicate by ELISA (Millipore Corp.) with intra- and interassay C.V. of 2.7–9.1 and 3.3– 8.4%, respectively. IGF-I levels were determined in duplicate by a mousespecific Quantikine ELISA, including a 500-fold dilution of all samples (R&D Systems, Inc., Minneapolis, MN), with intra- and interassay C.V. of 4.1–5.6 and 4.3–9.0%, respectively. FFA and ␤-hydroxybutyrate levels were determined in duplicate using an enzyme colorimetric assay (BioVision, Mountain View, CA).

Quantitative histology At the end of the in vivo animal study, tibiae were fixed with 4% paraformaldehyde overnight, decalcified with Decalcifier II (Surgipath, Richmond, IL) for 1 h, and paraffin embedded. Three 5- to 7-␮m-thick longitudinal sections from each bone were obtained and stained with hematoxylin/phloxine/saffron. Tibial growth plate, growth plate epiphyseal, proliferative, and hypertrophic zone heights (in micrometers) were measured. All measurements were performed by a single observer blinded to the treatment regimen.

Western blot analysis Frozen livers were extracted with the NE-PER system (nuclear and cytoplasmic extraction kit from Pierce, Rockford, IL) (14). Cytoplasmic and nuclear extracts were resolved by SDSPAGE and transferred onto a nitrocellulose membrane and were probed with the following primary antibodies: goat polyclonal antibody against signal transducer and activator of transcription (STAT) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), rabbit polyclonal antibody against p-Stat5 (Cell Signaling Technology, Inc., Danvers, MA), goat polyclonal antibody against IGF-I (Santa Cruz Biotechnology, Inc.), goat polyclonal antibody against FGF21 (Santa Cruz Biotechnology, Inc.), rabbit polyclonal antibody recognizing the epitope corresponding to amino acids 339 – 638 of GHR (Santa Cruz Biotechnology, Inc.), and


rabbit polyclonal antibody against ␤-actin (Sigma-Aldrich, St. Louis, MO). The blots were developed using a horseradish peroxidase-conjugated polyclonal donkey-antirabbit IgG antibody and enhanced chemiluminescence system (GE Healthcare, Princeton, NJ). Quantitation of exposed x-ray films was performed with ImageJ software (National Institutes of Health, Bethesda, MD).

RNA extraction and real-time PCR At the end of the study period, total RNA was extracted from liver and the tibial growth plate using the QIAGEN RNeasy Mini kit (QIAGEN, Inc., Valencia CA). The recovered RNA was further processed using 1st Strand cDNA Synthesis kit for RT-PCR (avian myeloblastosis virus) (Roche Diagnostics Corp., Indianapolis, IN) to produce cDNA. 1 ␮g of total RNA and 1.6 ␮g of oligo-p(dT)15 primer were incubated for 10 min at 25 C, followed by incubation for 60 min at 42 C in the presence of 20 U of avian myeloblastosis virus reverse transcriptase and 50 U of ribonuclease inhibitor in a total 20-␮l reaction. The cDNA products were directly used for PCR or stored at ⫺80 C for later analysis. Real-time quantitative PCR was carried out using StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA) in a final volume of 25 ␮l containing 1 ␮l of cDNA, 12.5 ␮l of 2⫻ SYBR Green Master Mix (Applied Biosystems), 0.1 ␮M primers (Applied Biosystems) in deoxyribonuclease-free water. Primers used were: Fgf21 genotyping primer (3⬘wt1 GTGGTGAGAACCCAGGAGCC, 3⬘com1 ATTTAGGAGTGC


2289

AGACCTGCCC, 3⬘neo1 GGTGGGCTCTATGGCTTCTG); mouse ␤-actin (TGTGATGGTGGGAATGGGTCAGAA, TGT GGTGCCAGATCTTCTCCATGT); IGF-I (GGGCATTGTGGATGAGTGTTGCTT, TGGAACGAGCTGACTTTGTAGGCT); fibroblast growth factor receptor (FGFR)-1 (CCAGTG CAT CCATGAACTCTGGGGTTCTCC, GGTCACACGGTT GGGTTTGTCCTTATCCAG); FGFR-3(AGTGCTAAATGCC TCCCACGAAGA, AGGATGGAGCATCTGTTACACGCA); ␤-klotho (CTGGCTAAGGTTCAAGTACGGAGACCTCCC, GGAGCTGAGCGATCACTAAGTGAATACGCA); and GHR (AGCCTCGATTCACCAAGTGTCGTT, CAGCTTGTCGTTGGCTTTCCCTTT). The PCR conditions were: 50 C for 2 min followed by 40 cycles at 95 C for 15 sec and 60 C for 1 min. Values were quantified using the comparative cycle threshold method (15), and samples were normalized to ␤-actin.

GH-binding activity At the end of study period, mouse hepatocytes and chondrocytes were isolated and cultured routinely in monolayer culture at 37 C in 5% CO2/95% air in DMEM supplemented with 10% fetal bovine serum, 25 mM HEPES, 2 mM L-glutamine, 50 IU/ml penicillin, and 50 ␮g/ml streptomycin (16, 17). Monolayer cultures set up in 100-mm culture dishes were detached by scraping in 4 ml of PBS containing a mixture of protease inhibitors (10 mg/ml aprotinin, 10 mM benzamidine, and 0.2 mM phenylmethylsulfonylfluoride). After centrifugation, the packed cells were resuspended in 2.5-ml ice-cold

FIG. 1. Effects of food restriction on weight change, body growth, and tibial growth of WT and Fgf21 KO mice. Body weight, whole body length (from nose to anus), and serial radiographs of both tibiae were obtained weekly. The lengths of both tibiae were measured on the radiograph, and the average value was calculated. WT-AL (n ⫽ 6) and KO-AL (n ⫽ 5); WT-FR (n ⫽ 12) and KO-FR (n ⫽ 23); KO-FR ⫹ FGF21 (n ⫽ 5), food-restricted Fgf21 KO mice injected daily with recombinant human FGF21 (5 ␮g/g wt) for the 4-wk study period. Results are expressed as mean ⫾ SE.

2290

Kubicky et al.



FIG. 2. Effects of food restriction on tibial growth plate morphology. At the end of the experimental period (4 wk), tibiae were isolated, fixed, and paraffin embedded. Three 5- to 7-␮m-thick longitudinal sections were obtained and stained with hematoxylin/phloxine/saffron. The tibial growth plate zone heights were measured by one observer blinded to the treatment groups. A, WT and Fgf21 KO mice fed ad libitum (AL). B, Foodrestricted (FR) WT and Fgf21 KO mice. WGP, Whole growth plate; EZ, epiphyseal zone; PZ, proliferative zone; HZ, hypertrophic zone. Results are expressed as mean ⫾ SE.

inhibitor mixture and disrupted by sonication. Total cellular membranes were obtained by centrifugation of the cell sonicates at 150,000 ⫻ g for 15 min at 4 C. To measure GH binding, the membrane pellets were resuspended at protein concentrations of 1 mg/ml in 300 ␮l of binding assay buffer [25 mM Tris-Cl (pH 7.4), 0.1% BSA, and 10 mM MgCl2] and incubated in triplicate with 105 cpm human 125I-GH in the presence or absence of 10 ␮g of unlabeled GH. After incubation at 4 C for 18 h, the membrane suspensions were centrifuged, washed twice with 1-ml ice-cold PBS/BSA, and radioactivity counted using a scintillation counter.

Results Serum levels and liver mRNA expression of FGF21 in WT and Fgf21 KO mice fed ad libitum or food restricted In food-restricted WT (WT-FR) mice, the serum levels of FGF21 (Supplemental Table 1, published on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org) and its mRNA expression in the liver (Supplemental Fig. 1) were significantly greater than

FIG. 3. Effects of food restriction on [125I]GH-binding and GHR expression in the liver. [125I]GH-binding activity (A), GHR mRNA (B), and protein (C) expression were measured in hepatocytes isolated from WT and Fgf21 KO mice fed ad libitum (AL) or food restricted (FR) at the end of the 4wk study period (as described in Materials and Methods).



2291

FIG. 4. Effects of food restriction on STAT5 phosphorylation and IGF-I mRNA expression in the liver. On the last day of the 4-wk study, WT-FR and KO-FR mice received an ip injection of recombinant mouse GH (5 ␮g/g wt; FR ⫹ GH) or vehicle (FR ⫺ GH); 15 min after the injection, all mice (GHinjected, vehicle-injected, and noninjected mice fed ad libitum) were euthanized and their livers removed. A, Cytoplasmic and nuclear extracts were resolved by SDS-PAGE and transferred onto a nitrocellulose membrane and probed with p-STATS and total STAT5 antibodies. B, Total RNA was extracted from the liver, and IGF-I mRNA expression was detected by real-time PCR.

those measured in WT mice fed ad libitum (WT-AL). In contrast, Fgf21 KO mice fed ad libitum (KO-AL) or food restricted (KO-FR) exhibited virtually undetectable serum levels (Supplemental Table 1) and FGF21 mRNA in the liver (Supplemental Fig. 1). Effect of food restriction on body growth, tibial growth, and growth plate thickness of WT and Fgf21 KO mice At the beginning of the 4-wk study, the mean body weight, body length, and tibial length of the four study groups (WT-AL, KO-AL, WT-FR, and KO-FR) were equivalent. By the end of the study, WT-AL and KO-AL mice experienced equivalent food intake (Supplemental Fig. 2), weight gain (Fig. 1A), body growth (Fig. 1B), tibial growth (Fig. 1C), and growth plate thickness (Fig. 2A and Supplemental Fig. 3). WT-FR and KO-FR mice had similar food intake and weight loss (Supplemental Fig. 2 and Fig. 1A, respectively); in contrast, KO-FR exhibited spared body length (KO-FR vs. WT-FR; P ⬍ 0.01) (Fig. 1B) and greater tibial growth (KO-FR vs. WT-FR; P ⬍ 0.01) (Fig. 1C) and tibial growth plate thickness (KO-FR

vs. WT-FR; P ⬍ 0.05) (Fig. 2B and Supplemental Fig. 3) compared with the WT-FR mice. The greater growth plate thickness was due to the greater height of the KO-FR hypertrophic zone. Within each of the four subgroups of mice (WT-AL, WT-FR, KO-AL, and KO-FR), there was no difference between male and female animals with respect to weight gain/loss, body growth, or tibial growth (Supplemental Table 2). In the subgroup of KO-FR mice treated with recombinant human (rh) FGF21 (5 ␮g/g wt), the mean weight loss was similar to that of the untreated KO-FR subgroup. In contrast, the body growth (KOFR⫹FGF21 vs. KO-FR; P ⬍ 0.05) (Fig. 1B) and tibial growth (KO-FR⫹FGF21 vs. KO-FR; P ⬍ 0.01) (Fig. 1C) were smaller than those of the untreated KO-FR but similar to those of untreated WT-FR mice. With respect to the serum levels of ␤-hydroxybutyrate and FFA, they were both higher in the WT-FR and KO-FR when compared with the WT-AL and KO-AL, respectively (Supplemental Table 1). However, there was no difference for neither level between WT-FR and KO-FR. These findings suggest that the lack of FGF21 expression in mice prevents the food restriction-associated body

2292

Kubicky et al.


FIG. 5. FGFR1, FGFR3, and ␤-klotho mRNA expression in the growth plate and in the liver. FGFR1, FGFR3, and ␤-klotho mRNA expression in chondrocytes and liver was measured by real-time PCR. Total RNA was extracted from chondrocytes isolated from fetal (20 days post-coitus) mouse metatarsal growth plates and from the liver of 4-wk-old mice. It was then processed as described in Materials and Methods. The mRNA expression of FGFR1, FGFR3, and ␤-klotho mRNA was normalized by ␤-actin in the same samples.

and longitudinal bone growth reduction, without affecting food intake or weight gain or loss. Effects of food restriction on systemic GH sensitivity in WT and Fgf21 KO mice At the end of the study, serum GH and IGF-I levels of WT-FR mice were two to three times higher and approximately six times lower than those measured in WT-AL, respectively (Supplemental Fig. 4, A and B). We found no difference between KO-FR and KO-AL with respect to their serum GH levels, whereas the serum IGF-I level of the former group was approximately two times lower than that of the latter group but significantly higher than the serum IGF-I of the WT-FR group (Supplemental Fig. 4, A and B). In the subgroup of KO-FR mice treated with rhFGF21, serum GH levels were higher than those of un-


treated KO-FR and WT-FR, whereas IGF-I levels were lower than those of untreated KO-FR but similar to those of untreated WT-FR (Supplemental Fig. 4, C and D). Hepatocytes isolated from mice at the end of the study revealed a significantly reduced [125I]GH binding and GHR mRNA and protein expression in WT-FR mice compared with WT-AL mice (Fig. 3, A–C, and Supplemental Table 3). Although we found a trend consistent with lower GH binding, GHR mRNA, and protein expression in KO-FR compared with KO-AL mice, the differences between the two groups did not reach statistical significance (Fig. 3, A–C, and Supplemental Table 3). Of note, GH binding, GHR mRNA, and GHR protein expression were all significantly greater in KO-FR than in WT-FR (Fig. 3, A–C, and Supplemental Table 3). In WT-FR mice injected with one dose of recombinant mouse GH 15 min before the end of the study, liver Stat5 phosphorylation (Fig. 4A and Supplemental Table 4) and IGF-I mRNA expression (Fig. 4B) were similar to those of untreated WT-FR mice. In contrast, KO-FR mice injected with recombinant mouse GH exhibited greater liver Stat5 phosphorylation (Fig. 4A and Supplemental Table 4) and IGF-I mRNA expression (Fig. 4B) than those of untreated KO-FR and GH-injected WT-FR mice. These findings indicate that food restriction in WT mice induces systemic GH insensitivity due to decreased GHR mRNA and protein expression in the liver; the lack of FGF21 expression in KO mice prevents a significant reduction in GHR expression and, in turn, GH insensitivity. Effects of food restriction on FGF21 and IGF-I expression and GH sensitivity in the growth plate of WT and Fgf21 KO mice To determine whether FGF21 modulates growth plate chondrogenesis directly at the growth plate, we first evaluated the expression of FGFR1, FGFR3 (both known to be

FIG. 6. FGF21 mRNA and protein expression in the growth plate of WT and Fgf21 KO fed ad libitum or food restricted. At the end of the 4-wk study, FGF21 mRNA (A) and protein (B) expression were detected by real-time PCR and Western blotting in the growth plate of WT and Fgf21 KO mice fed ad libitum (AL) or food restricted (FR).


preferentially bound and activated by FGF21), and ␤-klotho in the mouse tibial growth plate. mRNA expression of both receptors and ␤-klotho were detected in the growth plate (Fig. 5). At the end of the 4-wk study, we evaluated the mRNA and protein expression of FGF21 in the mouse growth plate. In WT-FR mice, FGF21 expression was significantly greater than in WT-AL, whereas it was virtually undetectable in KO-FR and KO-AL (Fig. 6, A and B, and Supplemental Table 5). In tibial growth plate chondrocytes isolated from WT-FR mice at the end of the study, [125I]GH-binding, GHR mRNA and protein expression, and IGF-I mRNA and protein expression were all reduced compared with WT-AL mice (Fig. 7 and Supplemental Table 5). No difference was found between KO-AL and KO-FR mice relative to GH binding and GHR mRNA and protein expression (Fig. 7), whereas IGF-I mRNA and protein expression was decreased in KO-FR compared with KO-AL mice. Of note, GH binding, GHR expression, and IGF-I expression in KO-FR were significantly greater than those in WT-FR (Fig. 7 and Supplemental Table 5). Thus, food restriction in WT mice induces decreased GHR receptor expression and binding and, in turn, GH


2293

sensitivity in the growth plate; the lack of FGF21 expression in KO mice prevents these effects.

Discussion Food restriction is associated with reduced skeletal growth, with such association primarily resulting from an impaired GH-IGF-I axis. Previous evidence indicates that food restriction may affect the GH-IGF-I axis both systemically and locally within the growth plate. In rodents, short-term food deprivation causes reduced GH secretion and (more consistently) action, which in turn lead to diminished IGF-I synthesis in the liver and decreased IGF-I circulating levels (1– 4). In other mammals (rabbit, sheep, cows, pigs, and humans), the fasting-associated reduced IGF-I synthesis is due only to GH insensitivity (5–7, 18 – 21). With respect to the effects of food restriction on GH action and IGF-I synthesis in the growth plate, it has been shown that 48 h of fasting in rabbits induce an up-regulation of GHR and IGF-I mRNA expression, suggesting that these changes in the growth plate may not contribute to the fasting-associated growth inhibition (21). However, GH action in the growth plate has not been studied in

FIG. 7. Effects of food restriction on [125I]GH binding, GHR expression, and IGF-I expression in the tibial growth plate. [125I]GH-binding activity (A), GHR mRNA and protein expression (B and C, respectively), and IGF-I mRNA and protein expression (B and C, respectively) were measured in chondrocytes isolated from the tibial growth plate of WT and Fgf21 KO-fed ad libitum (AL) or food restricted (FR) at the end of the 4-wk study period (as described in Materials and Methods).

2294

Kubicky et al.


other species or after prolonged food restriction. Although this evidence supports the importance of GH insensitivity as a cause of malnutrition-related growth failure, little is known on its underlying molecular mechanisms. In our study, we have demonstrated that the lack of expression of the Fgf21 gene in mice attenuates malnutrition-associated growth failure. Compared with WT mice, Fgf21 KO-AL for 4 wk experienced similar mean food consumption, weight gain, body linear growth, and tibial growth. After 4 wk of food restriction, Fgf21 KO exhibited an attenuated reduction of their body linear growth, tibial growth, and tibial growth plate thickness when compared with WT-FR mice. Although we cannot rule out that Fgf21 KO mice may have experienced a reduced metabolic rate, the fact that food intake and weight loss in WT-FR and KO-FR mice were comparable suggests only a marginal contribution of any improved energy efficiency as a causative mechanism in the attenuation of growth retardation in Fgf21 KO mice. To further support the specific effect of FGF21 on growth, we have shown that daily injections of rhFGF21 in KO mice during 4 wk of food restriction prevents the attenuation of growth failure observed in untreated Fgf21 KO mice. Previously, it has been shown that transgenic mice overexpressing FGF21 exhibit reduced tibial length compared with WT mice; thus, these findings and ours support a causative role for FGF21 in malnutrition-associated growth failure. With respect to the mechanisms underlying the FGF21mediated growth inhibition, the increased serum GH and reduced IGF-I levels in WT-FR mice (compared with WTAL) is consistent with GH insensitivity. The fact that foodrestricted Fgf21 KO mice exhibited similar GH levels compared with ad libitum Fgf21 KO mice (and serum IGF-I significantly higher than in WT-FR mice) suggests that the lack of FGF21 in mice prevents GH insensitivity. Such assumption is further supported by the significant increase in liver Stat5 phosphorylation and IGF-I mRNA expression in food-restricted Fgf21 KO mice after the injection of GH, with no GH effect being detected in WT-FR mice. Regarding the underlying molecular mechanisms of FGF21-mediated GH insensitivity, the reduced GH binding and GHR expression in WT-FR mice (and the lack of any reduction in food-restricted Fgf21 KO mice) when compared with mice fed ad libitum implicates a FGF21dependent decreased cell-surface GHR abundance as a cause of GH insensitivity. In Fgf21 KO mice, the food restriction-associated attenuated reduction of the tibial growth plate thickness (compared with WT-FR mice) indicates that the lack of FGF21 expression results in a preserved growth plate chondrogenesis. Conversely, the increased expression of


systemic FGF21 in WT-FR mice likely contributes to suppressed growth plate chondrogenesis and, in turn, longitudinal bone growth. Such effect may result from the reduced systemic IGF-I synthesis secondary to systemic GH insensitivity; on the other hand, it could also depend on the antagonistic effect of FGF21 on GH action directly at the long bones’ growth plate. Our finding of decreased GHR expression and IGF-I expression in the growth plate of WT-FR mice (along with their increased circulating GH levels) suggests a decreased GH action on growth plate chondrocytes. Circulating FGF21 (which is increased in WT-FR mice) could antagonize GH action by binding FGFR1 and/or FGFR3 (whose expression has been demonstrated in the growth plate, along with that of ␤-klotho) on growth plate chondrocytes. In addition, or alternatively, the food restriction-associated increased expression of FGF21 in the mouse growth plate may implicate a paracrine role for this growth factor in inhibiting growth plate chondrogenesis. In conclusion, our findings support a causative role for FGF21 in the inhibition of growth plate chondrogenesis and longitudinal bone growth during undernutrition. Such role may be mediated by the antagonistic effect of FGF21 on GH action in the liver and in the growth plate. Furthermore, our findings suggest that FGF21 may modulate growth plate chondrogenesis by acting both as an endocrine and a paracrine factor. Further studies are needed to elucidate the specific effects of FGF21 on growth plate chondrocyte function.

Acknowledgments Address all correspondence and requests for reprints to: Francesco De Luca, M.D., St. Christopher’s Hospital for Children, 3601 A Street, Philadelphia, Pennsylvania 19134. E-mail: [email protected]. Disclosure Summary: The authors have nothing to disclose.

References 1. Tannenbaum GS, Rorstad O, Brazeau P 1979 Effects of prolonged food deprivation on the ultradian growth hormone rhythm and immunoreactive somatostatin tissue levels in the rats. Endocrinology 104:1733–1738 2. Maes M, Underwood LE, Ketelslegers JM 1983 Plasma somatomedin-C in fasted and refed rats: close relationship with changes in somatogenic but not lactogenic binding sites. J Endocrinol 97:243– 252 3. Straus DS, Takemoto CD 1990 Effects of fasting on insulin-like growth factor-1 and growth hormone receptor mRNA levels and IGF-1 gene transcription in rat liver. Mol Endocrinol 4:91–100 4. Bornfeldt KE, Arnqvist HJ, Enberg B, Mathews LS, Norstedt G 1989 Regulation of insulin-like growth factor-1 and growth hor-


5.

6.

7.

8.

9.

10.

11.

mone receptor gene expression by diabetes and nutritional states in rat tissues. J Endocrinol 122:651– 656 Counts DR, Gwirtsman H, Carlsson LM, Lesem M, Cutler Jr GB 1992 The effect of anorexia nervosa and refeeding on growth hormone-binding protein, the insulin-like growth factors (IGFs), and the IGF-binding proteins. J Clin Endocrinol Metab 75:762–767 Ho KY, Veldhuis JD, Johnson ML, Furlanetto R, Evans WS, Alberti KG, Thorner MO 1988 Fasting enhances growth hormone secretion and amplifies the complex rhythms of growth hormone secretion in man. J Clin Invest 81:968 –975 Bang P, Brismar K, Rosenfeld RG, Hall K 1994 Fasting affects serum insulin-like growth factors (IGFs) and IGF-binding proteins differently in patients with non-insulin-dependent diabetes mellitus vs. healthy nonobese and obese subjects. J Clin Endocrinol Metab 78: 960 –967 Nishimura T, Nakatake Y, Konishi M, Itoh N 2000 Identification of a novel FGF, FGF21, preferentially expressed in the liver. Biochim Biophys Acta 1492:203–206 Kharitonenkov A, Shiyanova TL, Koester A, Ford AM, Micanovic R, Galbreath EJ, Sandusky GE, Hammond LJ, Moyers JS, Owens RA, Gromada J, Brozinick JT, Hawkins ED, Wroblewski VJ, Li DS, Mehrbod F, Jaskunas SR, Shanafelt AB 2005 FGF-21 as a novel metabolic regulator. J Clin Invest 115:1627–1635 Inagaki T, Dutchak P, Zhao G, Ding X, Gautron L, Parameswara V, Li Y, Goetz R, Mohammadi M, Esser V, Elmquist JK, Gerard RD, Burgess SC, Hammer RE, Mangelsdorf DJ, Kliewer SA 2007 Endocrine regulation of the fasting response by PPAR␣-mediated induction of fibroblast growth factor 21. Cell Metab 5:415– 425 Gälman C, Lundåsen T, Kharitonenkov A, Bina HA, Eriksson M, Hafström I, Dahlin M, Amark P, Angelin B, Rudling M 2008 The circulating metabolic regulator FGF21 is induced by prolonged fasting and PPAR␣ activation in man. Cell Metab 8:169 –174


2295

12. Reitman ML 2007 FGF21: a missing link in the biology of fasting. Cell Metab 5:405– 407 13. Inagaki T, Lin VY, Goetz R, Mohammadi M, Mangelsdorf DJ, Kliewer SA 2008 Inhibition of growth hormone signaling by the fasting-induced hormone FGF21. Cell Metab 8:77– 83 14. Wang X, Jiang J, Warram J, Baumann G, Gan Y, Menon RK, Denson LA, Zinn KR, Frank SJ 2008. Endotoxin-induced proteolytic reduction in hepatic growth hormone (GH) receptor: a novel mechanism for GH insensitivity. Mol Endocrinol 22:1427–1437 15. Schmittgen TD, Livak KJ 2008 Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc 3:1101–1108 16. Leung KC, Doyle N, Ballesteros M, Waters MJ, Ho KK 2000 Insulin regulation of human hepatic growth hormone receptors: divergent effects on biosynthesis and surface translocation. J Clin Endocrinol Metab 85:4712– 4720 17. Touvier T, Conte-Auriol F, Briand O, Cudejko C, Paumelle R, Caron S, Baugé E, Rouillé Y, Salles JP, Staels B, Bailleul B 2009 LEPROT and LEPROTL1 cooperatively decrease hepatic growth hormone action in mice. J Clin Invest 119:3830 –3838 18. Thomas GB, Mercer JE, Karalis T, Rao A, Cummins JT, Clarke IJ 1990 Effects of restricted feeding on the concentrations of growth hormone (GH), gonadotropins, and prolactin (PRL) in plasma, and on the amounts of messanger ribonucleic acid for GH, gonadotropin subunits, and PRL in the pituitary glands of adult ovariectomized ewes. Endocrinology 126:1361–1367 19. Breier BH, Bass JJ, Butler JH, Gluckman PD 1986 The somatotrophic axis in young steers: influence of nutritional status on pulsatile release of growth hormone and circulating concentrations of insulin-like growth factor-1. J Endocrinol. 11:209 –215 20. Buonomo FC, Baile CA 1991 Influence of nutrition deprivation on insulin-like growth factor-1, somatotropin, and metabolic hormones in swine. J Anim Sci. 69:755–760 21. Heinrichs C, Colli M, Yanovski JA, Laue L, Gerstl NA, Kramer AD, Uyeda JA, Baron J 1997 Effects of fasting on the growth plate: systemic and local mechanisms. Endocrinology 138:5359 –5365

You can post your CV, post an open position, or look for your next career opportunity in the targeted Career Services site. www.endo-society.org/PlacementServices