The effects of chemotherapeutic drugs on human monocyte-derived dendritic cell differentiation and antigen presentation.

Recent studies indicate that chemotherapeutic agents may increase the anti-tumoral immune response. Based on the pivotal role of dendritic cells (DCs) in host tumour-specific immune responses, we investigated the effect of commonly used chemotherapeutic drugs dexamethasone, doxorubicin, cisplatin and irinotecan and glucocorticoids on monocyte-derived DCs (moDCs). Dexamethasone displayed the strongest inhibitory effect on DC differentiation. The effect of cisplatin and irinotecan was moderate, while only weak effects were noticed for doxorubicin. Surprisingly, when the functional consequence of chemotherapy-treated CD14(+) monocytes and their capacity to activate CD4(+) T responders cells were investigated, cisplatin-treated monocytes gave rise to increased T cell proliferation. However, dexamethasone, doxorubicin and irinotecan-pretreated monocytes did not stimulate any increased T cell proliferation. Further investigation of this observation revealed that cisplatin treatment during DC differentiation up-regulated significantly the interferon (IFN)-ß transcript. By contrast, no effect was evident on the expression of interleukin (IL)-1ß, tumour necrosis factor (TNF)-α, IL-6 or IFN-α transcripts. Blocking IFN-ß attenuated the cisplatin-enhanced T cell proliferation significantly. In conclusion, cisplatin treatment enhanced the immune stimulatory ability of human monocytes, a mechanism mediated mainly by the increased production of IFN-ß.

The effect of in vivo growth hormone treatment on blood gene expression in adults with growth hormone deficiency reveals potential biomarkers to monitor growth hormone therapy.

OBJECTIVE: Growth hormone (GH) replacement therapy is presently utilized in the treatment of adult GH deficiency (AGHD). Adult responses to GH treatment are highly variable and, apart from measurement of IGF-I, few tools are currently available for monitoring GH treatment progress. As GH receptors are expressed in certain blood cell types, changes in gene expression in peripheral blood can reflect perturbations induced as a result of GH therapy. DESIGN/PATIENTS: We have carried out a pilot study to identify GH-responsive genes in blood, and have assessed the utility of GH-responsive genes in monitoring GH therapy in AGHD. Blood was collected from ten women diagnosed with AGHD syndrome both before and 4 weeks after initiation of GH substitutive therapy. RNA was extracted from peripheral blood mononuclear cells (PBMCs) and changes in response to GH were detected using microarray-based gene analysis. RESULTS: All patients responded to GH replacement therapy, with serum levels of IGF-I increasing by an average of 307% (P = 0.0003) while IGFBP-3 increased by an average of 182% (P = 0.0002). Serum levels of triglycerides, LDL-C, HDL-C, APOA1 or APOB did not change after 1 month of GH treatment. By contrast, we detected an increase in Lp(a) serum levels (P = 0.0149). Using a stringent selection cutoff of P

Ezrin mediates c-Myc actions in prostate cancer cell invasion.

The forced overexpression of c-Myc in mouse prostate and in normal human prostate epithelial cells results in tumor transformation with an invasive phenotype. How c-Myc regulates cell invasion is poorly understood. In this study, we have investigated the interplay of c-Myc and androgens in the regulation of prostate cancer cell invasion. We found that c-Myc induces cell invasion and anchorage-independent growth by regulating ezrin protein expression in the presence of androgens. The activity of the ezrin promoter is controlled by androgens through c-Myc, which binds to a phylogenetically conserved E-Box located in the proximal promoter region. Besides, we also show that ezrin is an important regulator of c-Myc protein levels. These effects are achieved through androgen-induced changes in ezrin phosphorylation, which results in the regulation of downstream signals. These downstream signals involve the modulation of Akt and GSK-3beta activity resulting in increased c-Myc protein synthesis and inhibition of its degradation. In summary, we have shown a key role for ezrin as a mediator of c-Myc-induced tumorigenesis in prostate cancer cells.

Role of pituitary hormones on 17alpha-ethinylestradiol-induced cholestasis in rat.

Estrogens cause intrahepatic cholestasis in susceptible women during pregnancy, after administration of oral contraceptives, or during postmenopausal hormone replacement therapy. 17alpha-Ethinylestradiol (EE) is a synthetic estrogen widely used to cause experimental cholestasis in rodents with the aim of examining molecular mechanisms involved in this disease. EE actions on the liver are thought to be mediated by estrogen receptor alpha (ERalpha) and pituitary hormones. We tested this hypothesis by analyzing metabolic changes induced by EE in livers from hypophysectomized (HYPOX) and hypothyroid rats. Microarray studies revealed that the number of genes regulated by EE was increased almost 4-fold in HYPOX rat livers compared with intact males. Little overlap was apparent between the effects of EE in intact and HYPOX rats, demonstrating that pituitary hormones play a critical role in the hepatic effects of EE. Consistently, hypophysectomy protects the liver against induction by EE of serum bilirubin and alkaline phosphatase, two markers of cholestasis and hepatotoxicity and modulates the effects of EE on several genes involved in bile acid homeostasis (e.g., FXR, SHP, BSEP, and Cyp8b1). Finally, we demonstrate a novel mechanism of action of EE through binding and negative regulation of glucocorticoid receptor-mediated transcription. In summary, pituitary- and ERalpha-independent mechanisms contribute to development of EE-induced changes in liver transcriptome. Such mechanisms may be relevant when this model of EE-induced cholestasis is evaluated. The observation that the pharmacological effects of estrogen in liver differ in the absence or presence of the pituitary could be clinically relevant, because different drugs that block actions of pituitary hormones are now available.

Androgens induce CD-9 in human prostate tissue.

Based on microarray analyses of LNCaP and LNCaP-r prostatic cell-lines we tentatively identified CD-9 as an androgen sensitive protein. This prompted us to characterize the androgen sensitivity and the correlation to malignancy of CD-9 at the protein level. Using Western blot, RT-PCR and immunohistochemistry the expression of CD-9 was analysed in LNCaP cells stimulated during increasing time by the synthetic androgen R1881 and also in 88 specimens of human prostate cancer tissues. Expression of CD-9 was induced by R1881 in LNCaP. CD-9 was immunolocalized in human prostate tissue sections representing non-malignant tissue as well as tumour areas. In non-malignant glands CD-9 immunoreactivity was observed at the apical and lateral cell borders of luminal epithelial cells. Basal epithelial cells were largely unstained. In tumour areas CD-9 staining intensity was variable and apparently not related to primary Gleason grade. In prostate tissue from a patient under androgen ablation therapy no staining was observed in luminal epithelial cells or in the tumour areas, but some staining was observed in basal epithelial cells. CD-9 is regulated by androgens in LNCaP and present in human prostate specimens. The expression is variable and changes in expression levels. These and earlier studies using other tissues indicate that CD-9 and its cellular localization could have an important role in prostate cancer cell development.

Endoplasmic reticulum stress prolongs GH-induced Janus kinase (JAK2)/signal transducer and activator of transcription (STAT5) signaling pathway.

The desensitization of the GH-induced Janus kinase 2 (JAK2) and signal transducer and activator of transcription 5 (STAT5) signaling pathway plays a crucial role in GH regulation of hepatic genes. Previous studies have demonstrated that the inactivation of the GH-induced JAK2/STAT5 pathway is regulated by protein translation and suppressors of cytokine signaling (SOCS). In this study we sought to explore the relationships between endoplasmic reticulum stress, GH-induced JAK2/STAT5 activity and SOCS expression. 1,2-bis(o-Aminophenoxy)ethane-N,N,N,N-tetraacetic acid (acetoxymethyl)ester (BAPTA-AM), used to provoke endoplasmic reticulum stress, caused a drastic inhibition of protein translation that correlated with the phosphorylation of the eukaryotic translation initiation factor 2alpha. Both GH and BAPTA-AM caused a rapid induction of the transcription factor C/EBP homology protein (CHOP) and an additive effect was observed with combined treatment, which suggests a regulatory role of GH on endoplasmic reticulum stress. Endoplasmic reticulum stress did not interfere with the rapid GH activation of STAT5 DNA binding activity. However, BAPTA-AM prolonged the DNA binding activity of STAT5 without affecting STAT5 or JAK2 protein levels. GH-induced phosphorylation of JAK2 and STAT5 DNA binding activity were prolonged in the presence of BAPTA-AM, suggesting that endoplasmic reticulum stress prevents the inactivation of STAT5 DNA binding activity by modulating the rate of JAK2/STAT5 dephosphorylation. Like BAPTA-AM, the endoplasmic reticulum stressors dithiothreitol and A23187 also prolonged the GH-induced STAT5 DNA binding activity. We were not able to correlate BAPTA-AM effects to the GH-dependent expression of SOCS proteins or SOCS mRNA, suggesting that endoplasmic reticulum stress modulates the rate of JAK2/STAT5 dephosphorylation through mechanisms other than inhibition of SOCS expression. This study indicates that cellular stress may modulate transcription through the JAK/STAT pathway.

Microarray analysis of the in vivo effects of hypophysectomy and growth hormone treatment on gene expression in the rat.

Complementary DNA microarrays containing 3000 different rat genes were used to study the consequences of severe hormonal deficiency (hypophysectomy) on the gene expression patterns in heart, liver, and kidney. Hybridization signals were seen from a majority of the arrayed complementary DNAs; nonetheless, tissue-specific expression patterns could be delineated. Hypophysectomy affected the expression of genes involved in a variety of cellular functions. Between 16-29% of the detected transcripts from each tissue changed expression level as a reaction to this condition. Chronic treatment of hypophysectomized animals with human GH also caused significant changes in gene expression patterns. The study confirms previous knowledge concerning certain gene expression changes in the above-mentioned situations and provides new information regarding hypophysectomy and chronic human GH effects in the rat. Furthermore, we have identified several new genes that respond to GH treatment. Our results represent a first step toward a more global understanding of gene expression changes in states of hormonal deficiency.

Gene expression profile of the aging process in rat liver: normalizing effects of growth hormone replacement.

The mechanisms that control life span and age-related phenotypes are not well understood. It has been suggested that aging or at least some of its symptoms are related to a physiological decline in GH levels with age. To test this hypothesis, and to improve our understanding of the cellular and molecular mechanisms behind the aging process, we have analyzed age-induced changes in gene expression patterns through the application of DNA chip technology. In the present study, the aging process was analyzed in rat liver in the presence or absence of GH replacement. Out of 3,000 genes printed on the microarrays, approximately 1,000 were detected in the rat liver. Among these, 47 unique transcripts were affected by the aging process in male rat livers. The largest groups of age-regulated transcripts encoded proteins involved in intermediary metabolism, mitochondrial respiration, and drug metabolism. Approximately 40% of the differentially expressed gene products were normalized after GH treatment. The majority of those transcripts have previously not been shown to be under GH control. The list of gene products that showed normalized expression levels in GH-treated old rats may shed further insight on the action and mechanism behind the positive effects of GH on, for example, fuel metabolism and body composition observed in different animal and human studies.

Comparison of STAT5 mRNA levels in GH-treated male and female rats analysed by a solution hybridization assay.

In this study we describe the development of a RNA:RNA solution hybridization-RNase protection assay to quantify STAT5 mRNA in total RNA extracts from rat tissues. The assay is sensitive and reproducible. We quantified STAT5 mRNA levels in liver and thymus lymphocytes from male and female control rats and from rats treated with a single dose of recombinant human growth hormone (rhGH). No significant sex differences in the expression pattern were observed in both studied tissues, but STAT5 mRNA levels were significantly (P< 0.05) higher in liver than in thymus lymphocytes. STAT5 mRNA levels were significantly (P< 0.05) increased by a pulse of GH given to either male or female normal rats, suggesting a regulation of STAT5 gene expression in the studied tissues. In conclusion, quantitative solution hybridization-RNase protection assay of STAT5 mRNA provides a tool to further advance the study of the regulatory mechanisms involved in STAT5 gene expression.

A cDNA RDA protocol using solid-phase technology suited for analysis in small tissue samples.

cDNA representational difference analysis (cDNA RDA) is a PCR-based subtractive enrichment procedure for the cloning of differentially expressed genes. In this study, we have further developed the procedure to take advantage of solid-phase technology, and to facilitate the use of RDA when starting material is limited. Several parameters of the PCR-based generation of cDNA representations were investigated, and a solid-phase based purification step was introduced to simplify removal of digested adapter-ends and uncleaved fragments. The use of magnetic particles increased the speed of the method, and also eliminated the risk of carry-over contamination between iterative steps of subtraction and PCR amplification. The modified protocol was evaluated in monitoring differences in gene expression in (i) a rat system consisting of livers with and without growth hormone treatment, and in (ii) a human system consisting of normal colon and colon cancer.

Characterisation of novel missense mutations in the GH receptor gene causing severe growth retardation.

Two Swedish brothers, 2.5 and 4 years of age, were found to fulfil all the clinical and laboratory characteristics of Laron's syndrome. They were shown to have unique missense mutations in the GH receptor gene. Both of their parents were of normal height, but they both separately carried one of the identified gene alterations. A molecular model of the first receptor alteration suggests that a collapse in three-dimensional receptor structure most likely contributed to the GH insensitivity in these patients.

Differential cloning of growth hormone-regulated hepatic transcripts in the aged rat.

It has been suggested that aging or at least some of its symptoms are related to a physiological decline in GH levels with age. This study was performed to elucidate age-related changes in GH-dependent effects at the level of gene expression. Through the application of complementary DNA representational difference analysis (RDA) we have identified gene products that are reduced during aging in rat liver. The expression of these genes was restored upon GH treatment. Results from reverse Northern and ribonuclease protection analysis confirmed that the RDA products were truly differentially expressed. In addition to well characterized GH-regulated genes, including CYP2C12, CYP2C13, and alpha2u-globulin, we demonstrate the differential expression of at least 11 genes previously not known to be under GH control. Several hepatic transcripts encoding enzymes and receptors involved in the metabolism of protein, carbohydrates, and lipids were identified. Other RDA products consisted of transcripts encoding proteins involved in ATP synthesis, detoxification of reactive oxygen species, or immune responses. This list of GH-regulated genes in the old rat may shed further light on the action and mechanism behind the positive effects of GH on, for example, body composition and the immune system that have been observed in different animal and human studies.

Growth hormone regulation of SOCS-2, SOCS-3, and CIS messenger ribonucleic acid expression in the rat.

The SOCS (suppressors of cytokine signaling) proteins have been suggested to function as inhibitors of cytokine receptor signaling. We have analyzed SOCS-2, SOCS-3, and CIS expression in relation to GH actions in the rat. SOCS-2, SOCS-3, and CIS transcripts were detected in various GH responsive tissues, including liver, muscle, and fat. In addition to the finding that different tissues express different levels of SOCS-2, SOCS-3, and CIS messenger RNA (mRNA), the steady-state levels of these SOCS transcripts were dependent on the endocrine status of the animal. SOCS-3 expression was 5-fold higher in fat from old compared with younger rats. Hypophysectomy reduced the levels of SOCS-2 and CIS mRNA in liver, muscle, and fat, whereas SOCS-3 expression was unchanged. Using primary cultures of rat hepatocytes, GH was shown to increase SOCS-2, SOCS-3, and CIS mRNA levels with different kinetics. SOCS-3 was rapidly and transiently induced, whereas SOCS-2 and CIS were increased in a slower fashion. Glucocorticoids blocked GH-induced SOCS-3 expression in cultured hepatocytes, whereas SOCS-2 and CIS expression was potentiated. Our data fit well with a concept of SOCS proteins acting as modulators of GH signal transduction.

In vitro interaction between STAT 5 and JAK 2; dependence upon phosphorylation status of STAT 5 and JAK 2.

A working model for haematopoietic cytokine signal transduction has been hypothesised as follows. Binding of cytokines to specific receptor molecules leads to phosphorylation and activation of receptor associated members of the Janus kinase family. This is followed by tyrosine phosphorylation of the associated receptor and members of the STAT (signal transducer and activator of transcription) family of DNA-binding transcription factors. Phosphorylation is accompanied by STAT dimerisation, nuclear transport and activation of gene transcription. Activation of gene transcription is mediated by the binding of STAT dimers to palindromic STAT response elements. A number of areas of confusion remain; not least the mechanism by which multiple cytokines signal via a limited number of STATs. A role has been suggested for phosphorylated receptor tyrosine residues as STAT docking sites on activated receptor-JAK complexes. According to this model the amino acid sequence context of key tyrosine residues confers receptor specificity upon STAT activation. There is some controversy as to whether this model applies to STAT 5. The heterologous expression of STAT 5 in Sf 9 insect cells using the baculovirus expression system is described here. Protein of the correct molecular weight was expressed and found to be phosphorylated on tyrosine residues and to bind to a STAT response DNA element. This binding was dependent upon the phosphorylation status of the STAT protein. DNA binding could be abolished in vitro by treatment with a phosphotyrosine phosphatase and restored in vitro by treatment with activated recombinant JAK 2. The protein was purified to near homogeneity using a simple ion exchange/gel filtration chromatography procedure. The interaction between purified recombinant STAT 5 and JAK 2, either expressed by baculovirus or endogenously expressed in Buffalo rat liver cells, was studied. In both cases STAT 5 in its non-phosphorylated form was found to form a stable complex with activated JAK 2. Non-activated JAK 2 and phosphorylated STAT 5 were unable to participate in complex formation. The results presented provide a mechanistic basis for the activation of STAT 5 by a wide range of cytokines capable of activating JAK 2.

Desensitization of the growth hormone-induced Janus kinase 2 (Jak 2)/signal transducer and activator of transcription 5 (Stat5)-signaling pathway requires protein synthesis and phospholipase C.

Signal transducers and activators of transcription (Stat) proteins are latent cytoplasmic transcription factors that are tyrosine phosphorylated by Janus kinases (Jak) in response to GH and other cytokines. GH activates Stat5 by a mechanism that involves tyrosine phosphorylation and nuclear translocation. However, the mechanisms that turn off the GH-activated Jak2/Stat5 pathway are unknown. Continuous exposure to GH of BRL-4 cells, a rat hepatoma cell line stably transfected with rat GH receptor, induces a rapid but transient activation of Jak2 and Stat5. GH-induced Stat5 DNA-binding activity was detected after 2 min and reached a maximum at 10 min. Continued exposure to GH resulted in a desensitization characterized by 1) a rapid decrease in Stat5 DNA-binding activity. The rate of decrease of activity was rapid up to 1 h of GH treatment, and the remaining activity declined slowly thereafter. The activity of Stat5 present after 5 h is still higher than the control levels and almost 10-20% with respect to maximal activity at 10 min; and 2) the inability of further GH treatment to reinduce activation of Stat5. In contrast, with transient exposures of BRL-4 cells to GH, Stat5 DNA-binding activity could repeatedly be induced. GH-induced Jak2 and Stat5 activities were independent of ongoing protein synthesis. However, Jak2 tyrosine phosphorylation and Stat5 DNA-binding activity were prolonged for at least 4 h in the presence of cycloheximide, which suggests that the maintenance of desensitization requires ongoing protein synthesis. Furthermore, inhibition of protein synthesis potentiated GH-induced transcriptional activity in BRL-4 cells transiently transfected with SPIGLE1CAT, a reporter plasmid activated by Stat5. GH-induced Jak2 and Stat5 activation were not affected by D609 or mepacrine, both inhibitors of phospholipase C. However, in the presence of D609 and mepacrine, GH maintained prolonged Jak2 and Stat5 activation. Transactivation of SPIGLE1 by GH was potentiated by mepacrine and D609 but not by the phospholipase A2 inhibitor AACOCF3. Thus, a regulatory circuit of GH-induced transcription through the Jak2/Stat5-signaling pathway includes a prompt GH-induced activation of Jak2/Stat5 followed by a negative regulatory response; ongoing protein synthesis and intracellular signaling pathways, where phospholipase C activity is involved, play a critical role to desensitize the GH-activated Jak2/Stat5-signaling pathway.

Mitogen-activated protein kinase kinase inhibition decreases growth hormone stimulated transcription mediated by STAT5.

We have investigated the possible involvement of the MAPK pathway in the growth hormone(GH)-induced activation of one of the members of signal transducers and activators of transcription, STAT5, by using the MAPK kinase (MEK) inhibitor PD98059. PD98059 treatment of Chinese hamster ovarian cells, stably transfected with the GH receptor (CHOA cells), abolished the GH-induced MAPK activity. PD98059 decreased the amount of GH-induced STAT5 in nuclear extract with DNA-binding capacity. Furthermore, GH dependent transcription of a STAT5 regulated reporter gene was inhibited by PD98059. The MEK inhibitor did not reduce GH-stimulated nuclear translocation of STAT5. We also investigated if PD98059 differentially influences the activation of the two STAT5 homologs, STAT5a and STAT5b, which differ mainly at the C-terminal end, one of the differences being the presence of a possible MAPK phosphorylation site in STAT5a. Expression plasmids for these transcription factors were transfected into CHOA cells together with a reporter gene. GH-stimulated fold induction of transcription was reduced by PD98059 in STAT5a but not in STAT5b overexpressing cells. A MAPK phosphorylation site-mutated version of STAT5a was also transfected into CHOA cells. GH-stimulated fold induction of cotransfected reporter gene was not reduced by PD98059 in cells overexpressing mutant STAT5a. The above data show that the MAPK pathway is required for the full activation of one of the STAT5 isoforms (STAT5a).

Growth hormone-induced reorganization of the actin cytoskeleton is not required for STAT5 (signal transducer and activator of transcription-5)-mediated transcription.

We have investigated the effect of GH on the organization of the actin cytoskeleton within the cell. Human GH (hGH) treatment (50 nM) of Chinese hamster ovary (CHO) cells stably transfected with the complementary DNA for the rat GH receptor (CHO-GHR(1-638)) resulted in a reorganization of actin filaments in the cell that was not observed upon GH treatment of the untransfected parental CHO cell line. hGH initially induced depolymerization of actin stress fibers similar in magnitude to that induced by treatment of the cells with 100 nM human insulin-like growth factor I. This loss of stress fibers was observed as early as 30 sec after addition of hGH to the medium, and maximal depolymerization of stress fibers was observed between 1-4 min after addition of hGH. This was followed by a slow, but submaximal, repolymerization of the stress fibers and the formation of localized focal filamentous actin containing complexes. Similar cytoskeletal changes were observed after hGH treatment in Swiss 3T3 fibroblasts and BRL cells stably transfected with rat GH receptor complementary DNA (BRL-GHR(1-6381)). Pretreatment of CHO-GHR(1-638) cells with wortmannin (a phosphatidylinositol 3-kinase inhibitor) and verapamil (a calcium channel antagonist) both inhibited the hGH-induced actin reorganization. The integrity of the actin cytoskeleton was not required for GH-induced STAT5 (signal transducer and activator of transcription-5)-mediated transcription, as treatment of cells with cytochalasins B and D did not alter the fold stimulation of the STAT5-mediated transcriptional response to GH. We conclude that GH induces a rapid reorganization of the actin cytoskeleton by a process requiring phosphatidylinositol 3-kinase activation and calcium influx, but this cytoskeletal reorganization is not required for the STAT5-mediated transcriptional response to GH.

Specificity of transcription enhancement via the STAT responsive element in the serine protease inhibitor 2.1 promoter.

The growth hormone regulated serine protease inhibitor (SPI) 2.1 and 2.2 gene promoters have been shown to contain a response element similar to the gamma-interferon activated sequence (GAS) family of signal transducer and activator of transcription (STAT) response elements. We have investigated the STAT and cytokine specificity of the SPI 2.1 STAT responsive element using a luciferase (LUC) reporter construct and a cDNA complementation strategy in the COS 7 cell line. Growth hormone was found to stimulate SPI-LUC reporter gene expression via activation of STAT 5, but not STATs 1 or 3, which indicates that the SPI 2.1 STAT responsive element is STAT 5 specific. In addition to the growth hormone receptor, the receptors for prolactin and erythropoietin enhanced gene transcription via the SPI 2.1 STAT responsive element, which indicates that this element is, on the other hand, not cytokine specific. Activation of STAT 5 was also observed after growth hormone treatment of cells transfected with cDNA expression plasmids for several different truncated growth hormone receptor mutants, although this activation was less efficient than with the wild type receptor. Point mutation of individual tyrosines in the growth hormone receptor intracellular domain to phenylalanines had no significant effect on signal transduction via STAT 5. These data, taken together with results from experiments using the phosphatase inhibitor sodium orthovanadate, suggest that STAT 5 may not have an absolute requirement for specific phosphorylated receptor tyrosine docking sites. That receptor tyrosine residues in a variety of amino acid contexts, or phosphorylated Janus kinase (JAK) 2 alone, can facilitate STAT 5 activation could explain the observed lack of cytokine specificity in STAT 5 activation.

Insulin-like growth factor I increases bone formation in old or corticosteroid treated rats.

We studied bone induction in subcutaneous implants of demineralized bone matrix with or without insulin-like growth factor I (IGF-I) in aged or corticosteroid-treated rats. Each rat carried one pair of implants, one control and one experiment implant, containing IGF-I dissolved in a hyaluronan solution for slow release. The rats were killed after 3 weeks and the results were evaluated by measuring the calcium content of implants. Young (6-7 weeks) and old (19-27 months) rats were used. A group of young rats was treated for 1 week with subcutaneous injections of 140 micrograms/kg dexamethasone daily. Old rats produced only approximately 1% as much bone as young rats. Local delivery of IGF-I did not increase bone formation in young rats. In old rats, bone formation was increased by IGF-I, 3000 ng/implant. Corticosteroids reduced bone formation in young rats. This effect was partially reversed by local administration of IGF-I.