Growth Hormone–Mediated Janus Associated Kinase–Signal Transducers and Activators of Transcription Signaling in the Growth Hormone–Resistant Potassium-Deficient Rat
Franz Schaefer*,
Sun-Ae Yoon,
Pouneh Nouri,
Tanny Tsao,
Padmaja Tummala,
Ellen Deng and
Ralph Rabkin
*Division of Pediatric Nephrology, University Children’s Hospital, Heidelberg, Germany; and Veterans Affairs Palo Alto Health Care System and Department of Medicine, Stanford University, Palo Alto, California
Correspondence to Dr. Ralph Rabkin, Veterans Affairs Palo Alto Health Care System (111R), 3801 Miranda Avenue, Palo Alto, CA 94304. Phone: 650-858-3985; Fax: 650-849-0213; E-mail: rabkin{at}stanford.edu
ABSTRACT. Potassium deficiency (KD) is associated with severegrowth failure, in part caused by growth hormone (GH) resistance.This study set out to determine whether the resistance couldbe caused by a defect in GH-mediated janus associated kinase–signaltransducers and activators of transcription (STAT) signalingas occurs in uremia. To this end, rats were fed a K-deficientdiet for 8 d and pair-fed controls received a K-replete diet.Animals from each group received GH or vehicle, and during thisperiod, KD rats were GH resistant; GH induced body and liverweight gain and linear body growth were severely attenuatedin these rats. In addition, signal transduction was studiedin the liver of rats that were killed 10 or 15 min after anintravenous GH bolus or vehicle. When the rats were killed,GH receptor mRNA and protein levels were similar in the twogroups. The abundance of STAT5, STAT3, and STAT1, proteins thatmediate GH signaling, was significantly increased by 40 to 130%in KD. Furthermore, GH induced a far greater increase in STAT5and STAT3 phosphorylation in this group. STAT5 phosphorylationwas enhanced fourfold even when normalized for total STAT5 content.Phosphorylated STAT5 and STAT3 proteins were also increasedin nuclear extracts, suggesting normal nuclear translocationof the activated signaling proteins. DNA binding of nuclearSTAT5 was unaltered. Thus, in KD, there is resistance to thegrowth-promoting action of GH despite hyperactivation of thejanus associated kinase–STAT signaling pathway. This suggeststhe presence of a defect distal to the nuclear binding of STATor, alternatively, a defect in a STAT-independent GH-activatedsignaling pathway.
In children with congenital tubulopathies such as Bartter syndrome,potassium (K) depletion leads to severe growth failure (1).In elderly and hospitalized patients, K deficiency (KD) is associatedwith wasting and an increased risk of death (2). The causativerole of KD for growth failure and tissue metabolism has beendemonstrated impressively in animal studies. Young rodents thatreceive a K-deficient diet stop growing within a few days (3–12),and their ability to accrete tissue protein is severely impaired(3). Involvement of the growth hormone (GH)–IGF-I systemin KD-induced growth failure has been suggested by reduced pituitaryGH release (6), diminished hepatic GH receptor binding (9),low tissue IGF-1 mRNA levels (8,10,12), and low serum IGF-1levels (4,6,10–12). Whereas these findings are compatiblewith central downregulation of the GH–IGF-1 system inKD, exogenous administration of GH fails to restore growth inKD animals (6).
Recently, the intracellular signaling pathways activated byGH have been elucidated (13). Binding of GH to its membranereceptor induces dimerization of the receptor, resulting intyrosine phosphorylation of janus associated kinase-2 (JAK2),a tyrosine kinase associated with the intracellular receptordomain. Phosphorylation induces the kinase activity of JAK2,which in turn phosphorylates the cytoplasmic domain of the GHreceptor (GHR). The GHR–JAK2 complex phosphorylates threemembers of a group of molecules known as signal transducersand activators of transcription (STAT), especially STAT5 and,to a lesser degree, STAT3 and STAT1. Upon phosphorylation, theSTAT form dimers that are directly translocated into the nucleusand bind to specific promoter sequences of GH-dependent genesto transactivate or repress their transcription. Although GHalso activates other signaling pathways through JAK2, includingthe mitogen-activated protein kinase and PI-3 kinase pathways,activation of the JAK2-STAT5 pathway seems to be essential fornormal growth.
Given the above background, we hypothesized that the markedGH resistance observed in KD might be caused by a defect inthe GHR-JAK-STAT signaling pathway. For testing this hypothesis,hepatic GHR expression and GH binding, the GH-induced tyrosinephosphorylation of JAK2 and the STAT, and the nuclear translocationand DNA binding of STAT5 were assessed in GH-resistant KD ratsand pair-fed controls.
Experimental Animals and Protocols
Pituitary-intact young male Sprague-Dawley rats that weighed75 g were conditioned for 5 d to a diet that contained 1% K(Purina, Richmond, IN). At the end of the run-in period, animalswere randomly allocated to experimental groups. The animalsthat were allocated to the low-K group were fed ad libitum witha diet that contained 0.01% KCl. The control groups continuedto receive the 1% K diet, which was given either ad libitumor as the same amount that corresponding animals in the low-Kgroup had consumed spontaneously in the previous 24 h (individualpair feeding). All animals were allowed free access to water.Five to eight animals were allocated to each subgroup.
Study A.
The low-K and the pair-fed control groups were subdivided atthe time of diet allocation into two subgroups that receivedeither twice-daily subcutaneous injections of 5 mg/kg bovineGH (bGH; Monsanto Corp., St Louis, MO) or vehicle for 8 d. Onthe eighth day, the vehicle-treated animals were fasted overnightand anesthetized with ketamine/xylazine. A small piece of liverwas ligated and removed, followed by an injection of 5 mg/kgbGH administered via the femoral vein. The remaining liver wascollected 15 min after GH injection, and the animals were killedby aortic puncture. The liver tissue was quickly excised, minced,snap-frozen in liquid nitrogen, and stored at –80°Cuntil further processing.
Study B.
In another series of animals that received the low-K diet orwere pair fed with 1% K control food for 8 d, samples were collected10 min after injection of either vehicle or 5 mg/kg bGH in separateanimals.
Study C.
In a third set of experiments designed to measure GH-stimulatedSTAT5 DNA binding activity, rats were placed on a low-K or K-repletediet for 8 d. At this time, six rats from each group received5 mg/kg GH or vehicle by intraperitoneal injection and werethen killed 1 h later under ketamine/xylazine anesthesia. Theliver was removed and frozen for electromobility shift assay(EMSA). All experiments were conducted in accordance with theNational Institutes of Health Guide for the Care and Use ofLaboratory animals.
Biochemical Measurements
Serum K was measured by flame photometry. The protein contentsof the tissue lysates were measured by the Bradford method (Bio-RadProtein Assay; Bio-Rad Laboratories, Hercules, CA).
GHR Binding Assay
Liver membranes were prepared as described previously (14).Briefly, 0.5 g of liver in 0.25 M ice-cold sucrose was homogenizedin an Ultra-Turrax homogenizer (IKA-Labortechnik, Staufen, Germany)at 20,000 rpm. After centrifugation at 1000 x g for 10 min,the supernatant was centrifuged at 100,000 x g for 60 min at4°C. The pellet, which contained crude liver membranes,was resuspended with 0.25 M sucrose, and GHR binding was measuredas described previously (15). The bGH was radiolabeled by theIodo-Gen method (Pierce Chemical, Rockford, IL). Briefly, membranealiquots (600 µg of protein) were preincubated with 3M MgCl2 for 5 min to remove endogenous prebound GH and thenincubated with 10–1 M [125I]GH at 4°C overnight. Freeand liver membrane receptor–bound [125I]GH were separatedby centrifugation at 15,000 x g for 10 min and quantified ina gamma counter. Specific [125I]GH binding was determined bydisplacement with 10–6 M unlabeled GH. Results were expressedas percentage of specific GH binding per milligram of membraneprotein.
Ribonuclease Protection Assay
The GHR probe was prepared from a 2.2-kb full-length mouse cDNAprovided by Dr. F. Talamantes (16). This cDNA encodes a sequencethat encompasses the extra- and intracellular domain of thereceptor. A glyceraldehyde-3-phosphate dehydrogenase cDNA probewas used to normalize for hybridization efficacy. The cDNA probeswere labeled with 50 µCi [32P]dCTP by a random primermethod (Multiprime DNA labeling system; Amersham Biotech, Piscataway,NJ). RNA was isolated from liver as described previously (17).Twenty micrograms of total RNA was hybridized with the radiolabeledantisense riboprobes at 42°C overnight. The mixture wasthen incubated with an RNase digestion buffer followed by theaddition of proteinase K. Ethanol-precipitated protected hybridswere separated on a polyacrylamide/urea denaturing gel. Autoradiographywas performed overnight at –20°C using a cassettewith intensifying screens. Protected bands were quantified densitometrically.
Immunoprecipitation and Western Immunoblotting
Antibodies against STAT1, STAT3, phospho-STAT1, and phospho-STAT3were obtained from New England Biolabs (Beverly, MA). The anti-STAT5antibody, which detects STAT5a and STAT5b; the anti-phosphotyrosine(-PY20) antibody; and protein A agarose were from Santa CruzBiotechnology (Santa Cruz, CA). For the detection of phospho-STAT5,an antibody from Upstate Biotechnology (Lake Placid, NY) wasused. This polyclonal antibody detects the tyrosine-phosphorylatedforms of both STAT5A and STAT5B. The GHR antibody mAb263, raisedagainst purified rat GHR (18), was purchased from Biogenesis(Kingston, NH).
A total of 10 mg of frozen liver tissue was homogenized on icewith a Polytron (Brinkmann Instruments, Westbury, NY) at maximumspeed for 30 s in 7 vol of lysis buffer (100 mM Tris [pH 7.4],1% Triton X-100) that contained 100 mM sodium pyrophosphate,100 mM NaF, 10 mM EDTA, 10 mM Na3VO4, 2 mM PMSF, and 0.1 mg/mlaprotinin. After 30 min of incubation on a shaker at 4°C,the extracts were centrifuged for 30 min at 45,000 x g to removeinsoluble material, and the supernatants were saved for furtheranalysis. For the detection of liver GHR, 50 mg of tissue washomogenized in a hot lysis buffer that contained 1% SDS and10 mM Tris (pH 7.5) with added 0.2 mM PMSF and 0.2 mM Na3VO4.Homogenates were boiled for 10 min and then centrifuged as describedabove. Nuclear extracts were obtained from 100-mg frozen tissuealiquots using the NE-PER extraction kit (Pierce) accordingto the manufacturer’s instructions. For JAK2 immunoprecipitation(IP), 5 mg of protein was diluted to 1 ml in IP buffer (containing10 mM Tris [pH 7.4], 1% Triton X, 150 mM NaCl, 1 mM EDTA, 1mM EGTA, 0.2 mM Na3VO4, 0.2 mM PMSF, and 0.5% NP-40) and incubatedwith 2 µg of JAK2 antibody overnight with constant mixingat 4°C. The immune complexes were then adsorbed to proteinA agarose beads for 2 h at 4°C and washed four times bycentrifugation in a microcentrifuge for 10 s and resuspendedin IP buffer.
The JAK2 immunoprecipitates or 50 µg of liver lysate (fordirect assay of all other proteins) were heated in Laemmli bufferat 100°C for 5 min, electrophoresed on a 7.5% SDS polyacrylamidegel, and electroblotted onto nitrocellulose membranes (Schleicher& Schuell, Keene, NH). Blots were blocked for 1 h in TBSTbuffer (10 mM Tris [pH 7.4], 138 mM NaCl, 0.05% Tween-20; SigmaChemical Co.) that contained 3 to 5% nonfat dehydrated milkor 1% BSA. Subsequently, the blots of the JAK2 IP were incubatedovernight at 4°C with antibodies against phosphotyrosineresidues (-PY20). The blots of the liver lysates were incubatedwith antibodies directed against GHR, JAK2, STAT1, STAT3, STAT5,phospho-STAT1, phospho-STAT3, or phospho-STAT5. After washingthree times for 15 min in TBST, the blots were incubated withsecondary anti-rabbit (JAK2, STAT1, STAT3, phospho-STAT1, phospho-STAT3)or anti-mouse antibody (GHR, -PY20, STAT5) conjugated to horseradishperoxidase (Santa Cruz Biotechnologies) at dilutions of 1:2000to 1:20,000 for 1 h at room temperature and then washed againthree times as described. The signal on the filter was detectedby enhanced chemiluminescence (ECL; Amersham, Arlington Heights,IL) and exposure to Kodak XAR film (Eastman Kodak, Rochester,NY). Protein expression was quantified densitometrically witha Fluor-S digital image analyzer and Multianalyst software (Bio-Rad).Relative density units refer to mean pixel density with localbackground subtraction.
EMSA
EMSA was performed as described previously by Ram et al. (19)with minor modifications. Liver nuclear extracts were obtainedfrom 100-mg frozen tissue aliquots using the NE-PER extractionkit as above. Eight micrograms of nuclear protein dissolvedin buffer (10 mM HEPES [pH 7.5], 50 mM NaCl, 0.5 mM PMSF, 0.1mM EDTA, 1 mM dithiothreitol, 0.5 mM sodium fluoride, 1 mM sodiumorthovanadate, 1 µg/ml pepstatin A, and 10% glycerol)was preincubated for 10 min at room temperature with 9 µlof gel mobility shift buffer (10 mM Tris-HCl [pH 7.5], 2 µgof poly(dI-dC), 4% glycerol, 1 mM MgCl2, 0.5 mM EDTA, 0.5 mMdithiothreitol, and 50 mM NaCl). Anti-STAT5 antibody was thenadded and incubated for 45 min at room temperature. Thereafter,the double-stranded oligonucleotide probe, 32P-end-labeled onone strand using T4 polynucleotide kinase, was added and incubatedfor 30 min. Samples were electrophoresed through nondenaturingpolyacrylamide gels (4% acrylamide, 0.05% bisacrylamide). TheDNA probe used for gel mobility shift analysis was a rat -caseinprobe (Stat 5/mammary gland factor response element, nucleotides–101 to –80), 5' GGA CTT CTT GGA ATT AAG GGA-3'(sense strand, oligonucleotide ON-257) and 5'-GTC CCT TAA TTCCAA GAA GTCC-3' (antisense strand, ON-258).
Statistical Analyses
Autoradiographic readings of GHR mRNA were normalized for glyceraldehyde-3-phosphatedehydrogenase mRNA signal strength. Similarly as indicated later,phosphorylated protein readings were normalized for the respectiveprotein levels. The control group mean was assigned a valueof 100%, and individual values are expressed relative to thisvalue. Data are given as mean ± SEM. Data were checkedfor Gaussian distribution using the Kolmogorov-Smirnov test(Sigmastat, Jandel, San Rafael, CA). Two-tailed unpaired t testswere applied for comparison of two normally distributed groups;comparisons between more than two normally distributed groupswere made by one-way ANOVA followed by pairwise multiple comparison(Student-Newman-Keuls method). For non-Gaussian distributions,the Kruskal-Wallis test followed by all-pairwise comparisons(Dunn test) was used.
Effect of KD Diet and bGH on Body Weight Gain, Liver Weight, Linear Growth, and Food Consumption
Serum K levels were significantly reduced in the low-K animals(Table 1). The low-K diet caused a marked impairment of absolutebody weight gain and longitudinal growth. Food utilization (i.e.,weight gain per unit weight of food consumed) was reduced by75% compared with the pair-fed controls. GH stimulated bodyand hepatic weight gain and longitudinal growth in the controls,whereas the KD rats were resistant to this treatment (Table 1,Figure 1).
Figure 1. Growth retardation and growth hormone (GH) resistance in potassium-deficient (KD) young rats. Young rats (initial weight 70 to 80 g) were subjected to diets that contained either 0.01% (low K) or 1% K (control). Control animals were individually pair fed (PF) with low-K animals. Starting on the first day of receiving the experimental diet, animals received either 5 mg/kg bovine GH (bGH) twice daily or vehicle (V) subcutaneously. Weight gain was suppressed in GH- and vehicle-treated low-K groups from the second day of diet onward. Error bars indicate SEM. Study A and study B were pooled for this analysis (n = 11–14 per group). Symbols indicate differences (P < 0.05) between groups. $PF-GH versus PF-V; *PF-GH versus low K–GH and low K–V; #PF-V versus low K–V and low K–GH.
GHR Gene Expression, Protein Abundance, and Binding Capacity
Hepatic GHR mRNA levels were unchanged in the low-K animalscompared with pair-fed controls (Figure 2A). Western immunoblottingshowed an increase of GHR protein abundance in the low-K groupby 25%, but this difference did not achieve statistical significance(Figure 2B). In addition to the GHR protein measurements, specificGH binding to liver membranes was performed to rule out thepresence of dysfunctional receptors. Specific GH binding tomembranes did not differ between low-K animals and pair-fedcontrols (Figure 2C).
Figure 2. Hepatic GH receptor (GHR) mRNA, protein abundance, and steady-state GH membrane binding are unaffected in K deficiency. GHR and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA was measured by Northern blot; GHR protein was measured by Western immunoblotting. GHR mRNA, protein, and GH membrane binding were analyzed in the same animals (n = 5–7 per group). GHR mRNA levels were normalized for the GAPDH signal. Bars indicate mean ± SEM.
JAK-STAT Protein Abundance Figure 3 depicts results from study B, and Figure 4 shows Westernimmunoblast from study B. The total abundance of JAK2 proteinin the liver was unchanged by the low-K diet in both experimentalseries performed. The mean (± SEM) signal strength inthe low-K animals was 103 ± 13% (NS) and 66 ±15% (NS) of the pair-fed control group mean in studies A andB (Figure 3A), respectively. Total STAT5 abundance was increasedin the low-K animals to 116 ± 6% of the control groupmean in study A (P = 0.06; Figure 4A) and to 137 ± 6%in study B (P < 0.005; Figure 3A). STAT3 and STAT1 were assessedin study B only (Figure 3A). Total STAT3 protein abundance inthe low-K animals was 235 ± 26% (P < 0.001), and STAT1was 151 ± 8% of controls (P < 0.05)
Figure 3. Analysis of janus associated kinase (JAK)–signal transducers and activators of transcription (STAT) signaling pathway. (A) Abundance of JAK2, STAT5, STAT3, and STAT1 protein in liver whole-cell lysates of low K and pair-fed control animals. Values are expressed relative to control group mean and are from study B (n = 6 per group). Bars indicate mean ± SEM. *P < 0.05; ***P < 0.0005. (B) Phosphorylation of JAK2 and STAT proteins 10 min after bGH administration. (C) GH-induced phosphorylation of JAK2 and STAT proteins relative to total abundance of respective protein.
Figure 4. Western immunoblots showing increased hepatic STAT protein abundance and GH-induced tyrosine phosphorylation of JAK2, STAT5, and STAT3 in KD rats. (A) Western immunoblots of STAT5 and STAT3 total protein and tyrosine-phosphorylated fraction in KD rats and pair-fed controls 15 min after injection of 5 mg/kg bGH (study A). (B) Western immunoblots of phosphorylated STAT5 and STAT3 in liver nuclear protein extracts in low K animals and pair-fed controls (same animals as in 4A above).
JAK-STAT Protein Phosphorylation
Basal phosphorylation of STAT5 was undetectable or very weak(0 to 0.1 arbitrary units) in both the low-K and the pair-fedcontrol animals before injection of bGH. In study A, shown inFigure 4A, 15 min after bGH injection, STAT5 phosphorylationwas moderately increased in the controls (0.1 to 0.4 arbitraryunits) but was much more prominent in the low-K animals (0.3to 0.9 arbitrary units). The mean GH-stimulated phospho-STAT5signal in the low-K group was 450 ± 93% of the controlgroup mean (P < 0.005). The relative GH-induced phosphorylationof STAT5, normalized to total STAT5 abundance, was also increasedmore than fourfold in the low-K compared with the control animals.These results were confirmed in study B. Both the absolute (400± 20% of control mean; P < 0.0001; Figure 3B) andthe relative phosphorylation of STAT5 (280 ± 9%; P <0.001; Figure 3C) were increased significantly in the bGH-injectedlow-K group. STAT3 phosphorylation was also increased in thelow-K group (P < 0.005; Figure 3B). However, this increasein phosphorylation was commensurate with the increase in totalSTAT3 protein, and there was no difference in phosphorylationrelative to total STAT3 protein level (Figure 3C). Post-growthhormone STAT1 phosphorylation was similar in low-K and controlanimals. The relative phosphorylation of JAK2 tended to be increasedin low-K animals (225 ± 58% of control mean; P = 0.06;Figure 3C), as a result of a slight increase in phosphorylatedand a slight decrease in total JAK2 abundance.
Nuclear Translocation of STAT and STAT5 DNA Binding Activity
In study A, phosphorylated STAT5 and STAT3 were also measuredin hepatic nuclear extracts. Fifteen minutes after GH injection,both phospho-STAT5 and phospho-STAT3 were detectable in thenucleus, with a markedly greater abundance in the low-K animals(Figure 4B). The binding of phosphorylated STAT5 to its specificDNA binding sequence was evaluated in nuclear extracts of studyC animals (six per group) by EMSA using a rat -casein probe(Figure 5). GH treatment induced a characteristic gel mobilityshift complex, and addition of STAT5 antibody further retardedthe mobility of the band. The signal strength of the STAT5-DNAcomplex was essentially identical in the low-K and in the pair-fedcontrols (104 ± 8 versus 100 ± 4 arbitrary units;n = 6 per group).
Figure 5. Gel mobility shift analysis of liver nuclear STAT5 DNA binding activity using a 32P-labeled -casein promoter probe. Distinct STAT5-DNA complexes are evident in the GH-treated animals but not in the vehicle (V)-treated rats, and a supershift was induced when samples were preincubated with STAT5 antibody (rats 3 and 8). There is no apparent difference in STAT5-DNA binding activity between KD and control pair-fed rats. The nuclear extracts were prepared from livers that were harvested 60 min after the intravenous injection of GH or vehicle.
In this study, we set out to determine whether a defect in GH-mediatedJAK-STAT signal transduction could be a cause of the GH resistancethat develops in the KD state and in this way contribute tothe striking growth retardation that develops in this condition.The rationale for this hypothesis was that intact JAK2-STAT5–mediatedsignaling is regarded as a prerequisite for GH-mediated bodygrowth, because mice that lack the genes for STAT5a and 5b aregrowth retarded (20), as are humans with a STAT5b mutation (21).In rats with chronic renal failure, a condition of acquiredgrowth failure associated with GH resistance, we recently demonstratedthat hepatic GH-dependent JAK-STAT signaling is impaired (22).In this condition, phosphorylation of JAK2 and STAT5, STAT3,and STAT1 is depressed as is nuclear translocation of the phosphorylatedSTAT. Unexpected in KD, we found that in contrast to renal failure,neither the GH-stimulated JAK-STAT phosphorylation nor the nucleartranslocation of phosphorylated STAT is depressed. Paradoxic,STAT3 and STAT5 tyrosine phosphorylation was significantly increasedin KD animals, and this occurred in the presence of unchangedGHR mRNA and protein expression, receptor ligand binding, andJAK2 protein levels. Moreover, binding of nuclear STAT5 to aspecific STAT5 response element probe -casein was unaffectedby KD. Thus, in the KD state, GH resistance develops despitehyperactivation of STAT5 and STAT3 phosphorylation and nucleartranslocation and intact STAT5 DNA binding. Accordingly, inthe discussion below, we consider alternative mechanisms thatmight account for the GH resistance.
It is interesting to note that the increased abundance of phosphorylatedSTAT5 and STAT3 in the KD animals after GH exposure seems toarise by different mechanisms. Increased total protein abundancewas observed for all STAT examined: STAT5, STAT3, and STAT1.Whereas the elevation of phosphorylated STAT3 in response toGH was proportionate to the increase in total STAT3 proteinconcentration, phosphorylated STAT5 was increased not only inabsolute terms but also when corrected for the increase in protein(relative phosphorylation). In other words, it turns out thatmost of the augmented STAT5 phosphorylation is accounted forby the increase in relative phosphorylation, whereas the augmentedSTAT3 phosphorylation is accounted for by the increase in STATprotein. Recently, somewhat comparable changes have been reportedin skeletal muscle of septic and endotoxin-treated rats thatalso develop GH resistance. In these animals, GH-mediated phosphorylationof STAT5 was exaggerated in gastrocnemius muscle, whereas GHfailed to stimulate IGF-1 gene expression (23). Furthermore,GH-induced STAT3 phosphorylation was increased, and this resultedin part from an increase in STAT3 protein level in the septicstate. It is tempting to speculate that the upregulation ofSTAT phosphorylation in muscle of septic rats and liver of KDrats may be an attempt to compensate for the GH-resistant state.
The hyperphosporylation of STAT5 in KD cannot be accounted forby a change in GH receptor levels or steady-state liver plasmamembrane binding, which were unaltered in the KD animals. However,it is conceivable that the increased tyrosine phosphorylationmight be caused by in vivo alterations in the kinetics of GHreceptor binding, internalization, and endosomal processing.Intracellular K depletion inhibits coated pit formation andreceptor-ligand internalization in cultured cells and in theisolated perfused rat liver (24,25). In cultured IM-9 lymphocytes,K depletion suppresses the slowly dissociating component ofGH binding and downregulation of the GHR (26). Accordingly,if these same changes are induced by KD in vivo, then they mightserve to prolong the action of GH on the JAK-STAT signal transductionpathway and thus cause the increase in STAT phosphorylationnoted.
An alternative mechanism of JAK-STAT hyperphosphorylation couldpotentially involve members of a protein family of cytokine-induciblesuppressors (SOCS), namely SOCS-1, SOCS-2, SOCS-3, and CIS (27,28).These proteins, which are inducible by GH and several othercytokines, inhibit GH-dependent JAK2 activation (29,30). Decreasedexpression of these proteins in KD could conceivably explainthe increased STAT5 phosphorylation in response to GH observed.Against this possibility is the lack of increase in basal STAT5phosphorylation in the KD rats and the short time period (10min) between the injection of GH and the measurement of STAT5phosphorylation in rats that were not previously treated withGH; changes in SOCS protein expression take considerably longerthan this.
Another potential explanation for STAT hyperphosphorylationmight be impaired deactivation of the signaling protein cascade(31–35). Thus, for example, if there were diminished tyrosinephosphatase activity in KD, then this could conceivably leadto sustained STAT5 phosphorylation (33,34). It is also possiblethat a defect in another signaling pathway activated by GH,such as the PI3-kinase or the mitogen-activated protein kinasepathway, may cause the GH-resistant phenotype in KD. Finally,it is conceivable that although signal transduction throughthe JAK-STAT pathway is intact in KD, a more distal defect thatcauses resistance to GH action might be present.
Changes in other hormonal systems such as the renin-angiotensinsystem may play an important role in the growth retardationof KD. The renin-angiotensin system has been shown to be upregulatedin the K-depleted state, and angiotensin II (AII) may be oneof the factors contributing to growth failure in KD by inducinganorexia and by suppressing IGF-1 production (36,37). WhereasAII can activate JAK/STAT signaling under certain conditions(38), it is unlikely that elevated AII levels are a cause ofthe exaggerated GH-induced STAT5 phosphorylation observed becausebasal STAT5 phosphorylation before GH administration was absentin the low-K animals. However, AII might conceivably contributeto a putative distal block in GH signal transduction. In additionto these possibilities, KD is associated with a state of IGF-1resistance (7), and this will contribute to the impaired bodygrowth but not the impaired liver growth observed in responseto GH (39,40), because hepatocytes are essentially devoid ofIGF-1 receptors (41).
In this study in which we set out to determine whether KD-inducedGH resistance could be caused by a defect in the JAK-STAT signalingpathway, we found not only that JAK2-STAT phosphorylation isunimpaired in the liver of KD rats but also, unexpected, thatGH-induced STAT5 and STAT3 phosphorylation is actually augmented.The increased phosphorylation could be accounted for by elevatedSTAT protein levels and in the case of STAT5 also by a largeincrease in relative phosphorylation. It is interesting thatSTAT5 DNA binding activity seemed to be unaffected even thoughnuclear phosphorylated STAT5 levels were increased. Taken together,this leads us to speculate that the increase in STAT phosphorylationmay be a compensatory mechanism brought into play, for example,in response to deficient upstream activation as a result oflow circulating GH concentrations (4) and/or of defective downstreamevents. In any event, it seems that resistance to GH in KD cannotbe accounted for by alterations in the JAK-STAT signal transductionpathway. Thus, resistance presumably arises because of a defectdistal to the binding of STAT to DNA or, alternatively, becauseof a defect in a STAT-independent GH-activated signaling pathway(13).
Acknowledgments
This study was supported by a grant from the Max Kade Foundation(F.S.), by a Merit Review Grant from the Research service ofthe Department of Veterans Affairs (R.R.), and funding fromthe American Heart Association (R.R.).
Simopoulos AP: Growth characteristics in patients with Bartter’s syndrome. Nephron 23: 130–135, 1979[Medline]
Rastegar R, Soleimani M, Rastergar A: Hypokalemia and hyperkalemia. Postgrad Med J 77: 759–764, 2001[Abstract/Free Full Text]
Dorup I, Clausen T: Effects of potassium deficiency on growth and protein synthesis in skeletal muscle and the heart of rats. J Nutr 62: 269–284, 1989
Flyvbjerg A, Dorup I, Everts ME, Orskov H: Evidence that potassium deficiency induces growth retardation through reduced circulating levels of growth hormone and insulin-like growth factor-I. Metabolism 40: 769–775, 1991[CrossRef][Medline]
Flyvbjerg A, Marshall SM, Frystyk J, Rasch R, Bornfeldt KE, Arnqvist HJ, Jensen PK, Pallesen G, Orskov H: Insulin-like growth factor I in initial renal hypertrophy in potassium-depleted rats. Am J Physiol 262: F1023–F1031, 1992
Dorup I, Flyvbjerg A, Everts ME, Orskov H: Effects of growth hormone on growth and muscle Na+-K+ pump concentration in K+ deficient rats. Am J Physiol 262: E511–E517, 1992
Dorup I, Flyvbjerg A: Effects of IGF-1 infusion on growth and muscle pump concentration in K+-deficient rats. Am J Physiol 264: E810–E815, 1993
Hsu FW, Tsao T, Rabkin R: The IGF-I axis in kidney and skeletal muscle of potassium deficient rats. Kidney Int 52: 363–370, 1997[Medline]
Hochberg Z, Amit T, Flyvbjerg A, Dorup I: Growth hormone (GH) receptor and GH-binding protein deficiency in the growth failure of potassium-depleted rats. J Endocrinol 147: 253–258, 1995[Abstract/Free Full Text]
Van Neck JW, Flyvbjerg A, Schuller AGP, Rosato RR, Groffen C, Van Kleffens M, Lindenbergh-Kortleve D, Dorup I, Drop SLS: IGF, type I IGF receptor and IGF-binding protein mRNA expression in kidney and liver of potassium-depleted and normal rats infused with IGF-I. J Mol Endocrinol 19: 59–66, 1997[Abstract/Free Full Text]
Fervenza F, Tsao T, Rabkin R: Paradoxical body and kidney growth in potassium deficiency. Ren Fail 23: 339–346, 2001[CrossRef][Medline]
Tsao T, Fawcett J, Fervenza FC, Hsu FW, Huie P, Sibley RK, Rabkin R: Expression of insulin-like growth factor-I and transforming growth factor-beta in hypokalemic nephropathy in the rat. Kidney Int 59: 96–105, 2001[CrossRef][Medline]
Carter-Su C, Rui L, Herington J: Role of the tyrosine kinase JAK2 in signal transduction by growth hormone. Pediatr Nephrol 14: 550–557, 2000[CrossRef][Medline]
Postel-Vinay MC: Binding of human growth hormone to rat liver membrane lactogenic and somatotropic sites. FEBS Lett 69: 137–140, 1976[CrossRef][Medline]
Fervenza FC, Tsao T, Rabkin R: Response of the intrarenal insulin-like growth factor-I axis to acute ischemic injury and treatment with growth hormone and epidermal growth factor. Kidney Int 49: 344–354, 1996[Medline]
Smith WC, Kuniyoshi J, Talamantes F: Mouse serum growth hormone (GH) binding protein has GH receptor extracellular and substituted transmembrane domains. Mol Endocrinol 3: 984–990, 1989[CrossRef][Medline]
Tsao T, Wang J, Fervenza FC, Vu TH, Jin IH, Hoffman AR, Rabkin R: Renal growth hormone-insulin-like growth factor-I system in acute renal failure. Kidney Int 47: 1658–1668, 1995[Medline]
Barnard R, Bundesen PG, Rylatt DB, Waters MJ: Monoclonal antibodies to the rabbit liver growth hormone receptor: production and characterization. Endocrinology 115: 1805–1813, 1984[Abstract]
Ram PA, Park SH, Choi HK, Waxman DJ: Growth hormone activation of Stat1, Stat3, and Stat5 in rat liver. J Biol Chem 271: 5929–5940, 1996[Abstract/Free Full Text]
Udy GB, Towers RP, Snell RG, Wilkins RJ, Park SH, Ram PR, Waxman DJ, Davey HW: Requirement of STAT5b for sexual dimorphism of body growth rates and liver gene expression. Proc Natl Acad Sci U S A 94: 7239–7244, 1997[Abstract/Free Full Text]
Kofoed EM, Hwa V, Little B, Woods KA, Buckway CK, Tsubaki J, Pratt KL, Bezzrodnik L, Jasper H, Tepper A, Heinrich JJ, Rosenfeld RG: Growth hormone insensitivity associated with a STAT5b mutation. N Engl J Med 349: 1139–1147, 2003[Free Full Text]
Schaefer F, Chen Y, Tsao T, Nouri P, Rabkin R: Impaired JAK-STAT signal transduction contributes to growth hormone resistance in chronic uremia. J Clin Invest 108: 467–475, 2001[CrossRef][Medline]
Hong-Brown LQ, Brown CR, Cooney RN, Frost RA, Lang CH: Sepsis-induced muscle growth hormone resistance occurs independent of STAT5 phosphorylation. Am J Physiol Endocrinol Metab 285: E63–E72, 2003[Abstract/Free Full Text]
Larkin JM, Brown MS, Goldstein JL, Anderson GW: Depletion of intracellular potassium arrests coated-pit formation and receptor-mediated endocytosis in fibroblasts. Cell 33: 273–285, 1983[CrossRef][Medline]
Marks GJ, Hart TK, Rush GF, Hoffstein ST, Fong KL, Budelski PJ: Internalization of recombinant tissue-type plasminogen activator by isolated rat hepatocytes is via coated pits. Thromb Haemost 63: 251–258, 1990[Medline]
Ilondo MM, Courtoy PJ, Geiger D, Carpentier JL, Rousseau GG, De Meyts P: Intracellular potassium depletion in IM-9 lymphocytes suppresses the slowly dissociating component of human growth hormone binding and the down-regulation of its receptors but does not affect insulin receptors. Proc Natl Acad Sci U S A 83: 6460–6464, 1986[Abstract/Free Full Text]
Starr R, Wilson TA, Viney EM, Murray LJL, Rayner JR, Jenkins BJ, Gonda TJ, Alexander WS, Metcalf D, Nicola NA, Hilton DJ: A family of cytokine-inducible inhibitors of signalling. Nature 387: 917–921, 1997[CrossRef][Medline]
Cooney RN: Suppressors of cytokine signaling (SOCS): Inhibitors of the JAK/STAT pathway. Shock 17: 83–90, 2002[CrossRef][Medline]
Emanuelli B, Peraldi P, Filloux C, Sawka-Verhelle D, Hilton D, van Obberghen E: SOSC-3 is an insulin-induced negative regulator of insulin signaling. J Biol Chem 275: 15985–15991, 2000[Abstract/Free Full Text]
Hansen JA, Lindberg K, Hilton DJ, Nielsen JH, Billestrup N: Mechanism of inhibition of growth hormone receptor signaling by suppressor of cytokine signaling proteins. Mol Endocrinol 13: 1832–1843, 1999[Abstract/Free Full Text]
Levy DE, Darnell JE Jr: STATs: Transcriptional control and biological impact. Nat Rev Mol Cell Biol 3: 651–662, 2002[CrossRef][Medline]
Kisseleva T, Bhattacharya S, Braunstein J, Schindler CW: Signaling through the JAK/STAT pathway, recent advances and future challenges. Gene 285: 1–24, 2002[CrossRef][Medline]
Ram PA, Waxman DJ: Interaction of growth hormone-activated STATs with SH2-containing phosphotyrosine phosphatase SHP-1 and nuclear JAK2 tyrosine kinase. J Biol Chem 272: 17694–17702, 1997[Abstract/Free Full Text]
Aoki N, Matsuda T: A cytosolic protein-tyrosine phosphatase PTP1B specifically dephosphorylates and deactivates prolactin-activated STAT5a and STAT5b. J Biol Chem 275: 39718–39726, 2000[Abstract/Free Full Text]
Shuai K: Modulation of STAT signaling by STAT-interacting proteins. Oncogene 19: 2638–2644, 2003
Brink M, Wellen J, Delafontaine P: Angiotensin II causes weight loss and decreases circulating insulin-like growth factor I in rats through a pressor-independent mechanism. J Clin Invest 97: 2509–2516, 1996[Medline]
Fervenza FC, Rabkin R: The role of growth factors and ammonia in the genesis of hypokalemic nephropathy. J Ren Nutr 12: 151–159, 2002[CrossRef][Medline]
Booz GW, Day JN, Baker KM: Interplay between the cardiac renin angiotensin system and JAK-STAT signaling: Role in cardiac hypertrophy, ischemia/reperfusion dysfunction, and heart failure. J Mol Cell Cardiol 34: 1443–1453, 2002[CrossRef][Medline]
Gautsch TA, Kandl SM, Donovan SM, Layman DK: Growth hormone promotes somatic and skeletal muscle growth recovery in rats following chronic protein-energy malnutrition. J Nutr 129: 828–837, 1999[Abstract/Free Full Text]
Rooman RP, Kuijpers G, Gresnigt R, Bloemen R, Koster JG, Van Buul-Offers SC: Dexamethasone differentially inhibits thyroxine- or growth hormone-induced body and organ growth of Snell dwarf mice. Endocrinology 144: 2553–2558, 2003[Abstract/Free Full Text]
Caro JF, Poulos J, Ittoop O, Pories WJ, Flickinger EG, Sinha MK. Insulin-like growth factor I binding in hepatocytes from human liver, human hepatoma, and normal, regenerating, and fetal rat liver. J Clin Invest 81: 976–981, 1988
Received for publication October 11, 2003.
Accepted for publication June 11, 2004.
This article has been cited by other articles:
Y. Chen, D. Sun, V. M. R. Krishnamurthy, and R. Rabkin Endotoxin attenuates growth hormone-induced hepatic insulin-like growth factor I expression by inhibiting JAK2/STAT5 signal transduction and STAT5b DNA binding
Am J Physiol Endocrinol Metab,
June 1, 2007;
292(6):
E1856 - E1862.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
75 g were conditioned for 5 d to a diet that contained 1% K (Purina, Richmond, IN). At the end of the run-in period, animals were randomly allocated to experimental groups. The animals that were allocated to the low-K group were fed ad libitum with a diet that contained 0.01% KCl. The control groups continued to receive the 1% K diet, which was given either ad libitum or as the same amount that corresponding animals in the low-K group had consumed spontaneously in the previous 24 h (individual pair feeding). All animals were allowed free access to water. Five to eight animals were allocated to each subgroup. 
-PY20) antibody; and protein A agarose were from Santa Cruz Biotechnology (Santa Cruz, CA). For the detection of phospho-STAT5, an antibody from Upstate Biotechnology (Lake Placid, NY) was used. This polyclonal antibody detects the tyrosine-phosphorylated forms of both STAT5A and STAT5B. The GHR antibody mAb263, raised against purified rat GHR (18), was purchased from Biogenesis (Kingston, NH). 
-casein probe (Stat 5/mammary gland factor response element, nucleotides –101 to –80), 5' GGA CTT CTT GGA ATT AAG GGA-3' (sense strand, oligonucleotide ON-257) and 5'-GTC CCT TAA TTC CAA GAA GTCC-3' (antisense strand, ON-258). 
