The Critical Role of Src Homology Domain 2–Containing Tyrosine Phosphatase-1 in Recombinant Human Erythropoietin Hyporesponsive Anemia in Chronic Hemodialysis Patients
Shigeru Akagi,
Haruo Ichikawa,
Tatsuo Okada,
Ai Sarai,
Taro Sugimoto,
Hisanori Morimoto,
Takashi Kihara,
Ai Yano,
Kazushi Nakao,
Yoshio Nagake,
Jun Wada and
Hirofumi Makino
Department of Medicine and Clinical Science, Okayama University Graduate School of Medicine and Dentistry, Okayama, Japan
Correspondence to Dr. Jun Wada, Department of Medicine and Clinical Science, Okayama University Graduate School of Medicine and Dentistry, 2-5-1, Shikata-cho, Okayama 700-8558, Japan. Phone: +81-86-235-7235; Fax: +81-86-222-5214; E-mail: junwada{at}md.okayama-u.ac.jp
The molecular mechanism of anemia that is hyporesponsive torecombinant human erythropoietin (rHuEPO) in hemodialysis patientswithout underlying causative factors has not been investigatedfully in hematopoietic stem cell system. Circulating CD34+ cells(1 x 104) were isolated from rHuEPO hyporesponsive hemodialysispatients (EPO-H; n = 9), patients who were responsive to rHuEPO(EPO-R; n = 9), and healthy control subjects (n = 9). The patientswith known causes of EPO hyporesponsiveness were eliminatedfrom the current study. The cells were cultured in STEM PRO34 liquid medium, supplemented with rHuEPO, IL-3, stem cellfactor, and granulocyte-macrophage colony stimulating factorfor 7 d and then transferred to a semisolid methylcelluloseculture medium for performing burst forming unit-erythroid (BFU-E)colony assay. Expression of src homology domain 2 (SH2)-containingtyrosine phosphatase-1 (SHP-1), phosphorylated Janus kinase2 (p-JAK2), and phosphorylated signal transducer and activatorof transcription 5 (p-STAT5) was assessed with Western blotanalysis. In EPO-H patients, SHP-1 antisense or scrambled S-oligoswere included in the culture medium, and its effects were evaluated.The number of circulating CD34+ cells was not statisticallydifferent among the three groups, and their proliferation rateswere similar for 7 d in culture. However, BFU-E colonies weresignificantly decreased in EPO-H patients compared with EPO-Rand control groups. The mRNA and protein expression of SHP-1and p-SHP-1 was significantly increased, whereas that of p-STAT5was reduced in EPO-H patients. The inclusion of SHP-1 antisenseS-oligo in culture suppressed SHP-1 protein expression associatedwith p-STAT5 upregulation, increase in p-STAT5–regulatedgenes, and partial recovery of BFU-E colonies. In EPO-H hemodialysispatients, the EPO signaling pathway is attenuated as a resultof dephosphorylation of STAT5 via upregulation of SHP-1 phosphataseactivity, and SHP-1 may be a novel target molecule to sensitizeEPO action in these patients.
Since 1990, administration of recombinant human erythropoietin(rHuEPO) has dramatically improved the status of anemia in chronichemodialysis (CHD) patients. Also, it has reduced administrationof frequent blood transfusion and incidence of complicationsrelated to heart disease and contributed to a better outcomein mortality and morbidity (1). However, it is still difficultto overcome rHuEPO hyporesponsiveness, in which CHD patientshave persistent severe anemia despite high-dosage rHuEPO administrationand correction of underlying factors, such as depleted ironstorage (2), inadequate dialysis (3), severe hyperparathyroidism(4), and deficiency of water-soluble vitamins (5). Previously,we isolated circulatory CD34+ hematopoietic stem cells in hHuEPOhyporesponsive CHD (EPO-H) patients, and they were subjectedto colony assay using semisolid methylcellulose culture medium.A reduced number of burst-forming units-erythroid (BFU-E) wasnoted in EPO-H patients. Using cDNA array technique, we alsoobserved downregulation of various cytokines and growth factors,including IL-6, IL-9, vascular endothelial growth factor, granulocyte-macrophagecolony-stimulating factor (GM-CSF), and leukemia inhibitoryfactor, and of receptors for thrombopoietin, IL-9, and colony-stimulatingfactor 1 (6). These results suggest that specific impairmentof autocrine/paracrine regulatory loops in BFU-E cells contributesto the development of rHuEPO hyporesponsiveness.
To gain further insights into the molecular mechanism of rHuEPOhyporesponsiveness, we used this assay system and surveyed thegene expression profile of EPO signal transduction moleculesthat belong to the Janus-associated kinase 2 (JAK2) signal transducerand activators of transcription 5 (STAT5) pathway in BFU-E–derivedcells using cDNA array technique. Although several signalingpathways have been described in the literature (7–9),JAK2/STAT5 seems to be the major downstream signaling moleculesof the EPO receptor in hematopoietic progenitor cells (HPC).Upon EPO binding to its receptor, EPO receptors dimerize, andthis allows adjacent JAK2, a tyrosine kinase associated withthe intracellular domain of EPO receptor, to approximate witheach other with resultant transphosphorylation of active sitesof JAK2 (10). After the activation of JAK2, it subsequentlyphosphorylates tyrosine residues of the intracellular domainsof EPO receptor. After the binding of STAT5 to phosphorylatedEPO receptors, STAT5 is phosphorylated by JAK2, and it thendissociates from EPO receptors. STAT5 then forms dimer and translocatesinto the nucleus to activate the transcription of specific genes,including mouse oncostatin M protein and hemoglobin (11).
The src homology domain 2 (SH2)-containing tyrosine phosphatase-1(SHP-1) is another partner in JAK2-STAT5 signaling pathway inerythroid progenitors. SHP-1, also known as hematopoietic cellphosphatase, is a protein tyrosine phosphatase located in thecytoplasm of hematopoietic cells originally discovered in humanbreast carcinoma cDNA library (12). According to previous reports,SHP-1 binds to the negative regulatory domain of EPO receptorvia its src-homology 2 (SH2) domains and causes dephosphorylationof JAK2, thus functioning as a negative regulator of intracellularsignal transduction (13). Patients with polycythemia vera havediminished expression of SHP-1 mRNA in their colony-formingunit-erythroid (CFU-E) cells, which indicate a pivotal rolefor SHP-1 in the regulation of human erythroid progenitors (14).By surveying gene expression of these signaling molecules inBFU-E cells derived from EPO-H patients, we found that SHP-1mRNA was significantly upregulated compared with rHuEPO-responsiveCHD (EPO-R) patients. We hypothesized that upregulated SHP-1suppresses the intracellular signaling events after EPO bindingto its specific receptor, and it contributes to the persistentrHuEPO hyporesponsiveness in CHD patients. In the present investigation,we isolated circulating CD34+ cells from EPO-R and EPO-H patients,and they were cultured in the liquid medium, STEM PRO 34, whichsupports the growth and expansion of hematopoietic stem cells.These culture systems enabled us to investigate the proteinexpression and tyrosine phosphorylation status of JAK2-STAT5signaling molecules, including SHP-1. After investigating theupregulated protein levels and phosphorylation of SHP-1 in EPO-Hpatients, we further elucidated the relevance of this signalingpathway in EPO-H patients using antisense oligodeoxynucleotidetechnology.
EPO-H patients (n = 9) and EPO-R patients (n = 9) were enrolledin this study. All CHD patients underwent hemodialysis therapythree times a week for at least 1 yr by using polysulfon dialysismembrane and bicarbonate dialysate (Kindaly-AF2P; Fuso, Osaka,Japan). Age- and body mass index–matched healthy controlsubjects (CON; n = 9) without renal dysfunction and anemia werealso recruited. EPO-H patients had hematocrit (Ht) levels <25%during past 24 wk despite intravenous administration of rHuEPO,9000 IU/wk, the maximum administration dosage regulated by theFederation of National Health Insurance in Japan. EPO-R patientsmaintained their Ht levels >30% with <1500 IU/wk intravenousrHuEPO administration for 24 wk. All CHD patients met the followingcriteria: (1) on HD therapy for >1 yr; (2) no evidence ofmalignancy, chronic infection, chronic inflammation, renal cysts,or diabetes; (3) no history of angiotensin-converting enzymeinhibitors or angiotensin II receptor antagonists within thepast 6 mo; (4) no hematologic diseases that cause anemia, suchas iron deficiency, pure red cell aplasia, and - and -thalassemia;(5) no gastrointestinal bleeding; (6) malnutrition; and (7)underdialysis or secondary hyperparathyroidism. Written informedconsent was obtained from all patients before the investigation.
Selection and Culture of CD34+ Cells
Blood samples were withdrawn into heparinized tubes from arteriovenousfistula at the initiation of HD. Isolation of circulating CD34+cells was performed as described previously (6). First, peripheralblood mononuclear cells were isolated by Ficoll-Paque PLUS (PharmaciaBiotech, Piscataway, NJ). Then, early hematopoietic progenitors(CD34+ cells) were isolated using Dynal CD34 progenitor cellselection system (Dynal AS, Oslo, Norway) as follows. In brief,peripheral blood mononuclear cells were incubated with DynabeadsM-450 CD34 and rosetted CD34+ cells, washed with isolation buffer(PBS without Ca2+ and Mg2+, 2% BSA, 0.6% citrate, and 100 IU/mlpenicillin-streptomycin solution) and then released from Dynabeadsusing DETACHaBEAD CD34. At this point, 1 x 104 CD34+ cells weresubjected to colony assay described in the next section. Thepurified 1 x 104 CD34+ cells were transferred into STEM-PRO34 SFM liquid medium (Invitrogen, Carlsbad, CA) supplementedwith 3 IU/ml rHuEPO, 50 ng/ml recombinant human stem cell factor(rHuSCF), 10 ng/ml recombinant human GM-CSF (rHuGM-CSF), and10 ng/ml recombinant human IL-3 (rHuIL-3) (15). Cells were thenseeded at a density of 1 x 104 cells/ml in 24-well plates andincubated at 37°C in a humidified atmosphere flushed with5% CO2/5% O2/90% N2, and culture was maintained for 7 d.
Colony Assay of HPC
For assessing differentiation potential of erythroid progenitors,HPC that were cultured in liquid medium for 7 d were transferredinto 1 ml of semisolid methylcellulose medium (Methocult H4434;Stem Cell Technologies, Vancouver, BC, Canada). Cells were seededat the density of 1 x 103/ml in 35-mm Petri dishes. The methylcellulosemedium contained 30% FCS, 1% BSA, 2 mM L-glutamine and 10–4M2-mercaptoethanol, 3 IU/ml rHuEPO, 50 ng/ml rHuSCF, 10 ng/mlrHuGM-CSF, and 10 ng/ml rHuIL-3. The dishes were incubated at37°C in a humidified atmosphere with 5% CO2/5% O2/90% N2for 14 d, and the number of BFU-E colonies was counted usingan inverted microscope.
Western Blot Analysis of SHP-1, Phospho-JAK2, and Phospho-STAT5
The rabbit polyclonal antibodies against SHP-1, phospho-JAK2(p-JAK2), and phospho-STAT5 (p-STAT5) were obtained from UpstateBiotechnology (Lake Placid, NY). Anti–glyceraldehyde-3-phosphatedehydrogenase (GAPDH) antibody (Sigma Aldrich, St. Louis, MO)was used to verify the equal loading of proteins. The culturedCD34+ cells in liquid medium (1 x 106 cells/ml) were washedtwice at 4°C in STEM-PRO 34 in 1.5-ml microfuged tubes andcentrifuged. The pellets were lysed in 400 µl of RIPAbuffer that contained 150 mM Tris-HCl, 150 mM NaCl, 1% NP-40,1% Triton-X, 0.5% SDS, 0.5% sodium deoxycholic acid (DOC), 0.5µg/ml leupeptin, 2.0 µg/ml aprotinin, and 1 mM PMSF.The extracts were centrifuged for 30 min at 14,000 rpm, andthe supernatants were collected for immunoblot analysis. Thesamples (10 µg of protein) were heated in Laemmli bufferat 100°C for 5 min, subjected to 10% SDS-PAGE, and blottedon polyvinylidene difluoride membranes. The blots were immersedin PBS-Tween (pH 7.4; 137 mM NaCl, 8.1 mM Na2HPO412H2O, 2.7mM KCl, 1.5 mM KH2PO4, and 0.1% Tween 20) that contained 5%nonfat dry milk for 1 h, then were incubated with primary antibodiesagainst SHP-1, p-JAK2, and p-STAT5 overnight at 4°C. Afterwashing three times for 5 min in PBS-Tween, the blots were incubatedwith secondary anti-rabbit antibody conjugated to horseradishperoxidase (Upstate Biotechnology) for 1 h at room temperatureand washed again three times. Autoradiograms were developedusing ECL-Plus system (Amersham Life Science, Arlington Heights,IL). The optical density ratio of each band to that of GAPDHwas determined using NIH image program version 1.61 (NationalInstitutes of Health, Bethesda, MD) (16).
Assay of SHP-1 Tyrosine Phosphorylation
In assay of SHP-1 phosphorylation, 400 µg of CD34+ celllysates was preadsorbed using 40 µl of protein A SepharoseCL4B (Amersham) in the presence of normal rabbit serum. Supernatantswere incubated with 4 µg of anti–SHP-1 antibody,and then immune complexes were adsorbed with 80 µl ofprotein A Sepharose CL4B overnight at 4°C. The samples werecentrifuged, and immunoprecipitates were washed four times inRIPA buffer. The proteins were eluted with 30 µl of Laemmlibuffer and subjected to SDS-PAGE. The blots were incubated withmouse monoclonal anti-phosphotyrosine antibody (Cell Signaling,Beverly, MA) and visualized by ECL Western blotting system.
Quantitative Real-Time Reverse Transcription–PCR
To quantify the mRNA expression of SHP-1, bcl-xL, oncostatinM, and hemoglobin 1 in the HPC, we performed real-time reversetranscription–PCR (RT-PCR). First-stranded cDNA were synthesizedfrom 1.0 µg of total RNA using ribonuclease H-MMLV (moloneymurine leukemia virus) reverse transcriptase and were addedto the Lightcycler-Mastermix (0.5 µM specific primers,4 mM MgCl2, and 2 µl of Master SYBR Green; Roche Diagnostics,Mannheim, Germany). PCR was optimized and performed in a real-timePCR cycler (Lightcycler; Roche Diagnostics) (17). For SHP-1,the denaturation at 95°C for 10 min was followed by 40 cyclesof denaturation at 95°C for 10 s, annealing at 62°Cfor 5 s, and extension at 72°C for 10 s. For bcl-xL, oncostatinM, hemoglobin 1, and GAPDH, the denaturation at 95°C for10 min was followed by 40 cycles of denaturation at 95°Cfor 10 s, annealing at 60°C for 15 s, and extension at 72°Cfor 12 s. The copy number of SHP-1, bcl-xL, oncostatin M, andhemoglobin 1 cDNA was normalized with a housekeeping gene (GAPDH),and the relative expression ratios were calculated. Gene-specificprimers for human SHP-1 (Genbank accession no. NM–002831),bcl-xL (Z23115), oncostatin M (NM_020530), hemoglobin 1 (NM_0002831),and GAPDH (M33197) were as follows: SHP-1 forward (5'-GCCTGGACTGTGACATTGAC-3'),SHP-1 reverse (5'-ATGTTCCCGTACTCCGACTC-3'), bcl-xL forward (5'-TTCCTGACGGGCATGACT-3'),bcl-xL reverse (5'-AGGATGTGGTGGAGCAGA-3'), oncostatin M forward(5'-AGACCCTCAATGCCACAC-3'), oncostatin M reverse (5'-TCTCCAA-GTCTCGATGTTCA-3'), hemoglobin 1 forward (5'-CGGTCAACTTCAAGCTCCTAA-3'),hemoglobin 1 reverse (5'-CCA- AGGGGCAAGAAGCAT-3'), GAPDH forward(5'-TGAACGGGAAGCTCACTGG-3'), and GAPDH reverse (5'-TCC- ACCACCCTGTTGCTGTA-3').
SHP-1 Antisense Experiment Using CD34+ Cells Derived from EPO-H Patients
Forty micrograms of phosphorothioated SHP-1 antisense oligodeoxynucleotide(ODN), 5'-GAGGTCTCGGTGAAACCACCTCACCATCCT-3' (18), or phosphorothioatedscrambled ODN, 5'-GGCCAGGTCCCATCGACACCGATACATCTT-3', was incorporatedinto 0.1 AU of hemagglutination virus of Japan (HVJ)-envelopevector (GenomeONE-Neo; Ishihara Sangyo, Osaka, Japan). CD34+cells were isolated from EPO-H and CON groups, and 1 x 104 cellswere suspended in 1 ml of STEM-PRO 34 liquid medium. HVJ-E vectorcarrying SHP-1 antisense ODN was added into culture media andcentrifuged at 2000 rpm for 1 h at 37°C. The cells werefurther cultured in STEM-PRO 34 SFM Complete Medium supplementedwith rHuEPO, rHuSCF, rHuGM-CSF, and rHuIL-3 for 7 d. They werethen subjected to Western blot analysis, quantitative real-timeRT-PCR, and BFU-E colony assay as described above. The transfectionefficiency into CD34+ cells was evaluated by using pEGFPLuc(BD Biosciences, Palo Alto, CA), pcDNA3.1/V5-His-TOPO/lacZ (Invitrogen),and FITC-labeled SHP-1 antisense ODN. Two days after the DNAtransfection into CD34+ cells using HVJ-E vector, luciferaseactivity was evaluated by Luciferase Reporter Assay Kit (BDBiosciences), and -galactosidase activity was visualized by-Gal staining set (Roche).
Statistical Analyses
Data are presented as mean ± SD. Statistical analysiswas performed by Wilcoxon nonparametric test. P < 0.05 wasconsidered significant.
Clinical Characteristics of CHD Patients
The EPO-H group revealed significantly lower Ht (22.3 ±1.50%), even after the intravenous administration of a highdosage of rHuEPO, whereas the EPO-R group had a higher Ht (32.9± 1.2%) with low dosage of rHuEPO (Table 1). It has beenreported that hemoglobin concentration improves in 90 to 95%of patients treated (4), and we also found that 5% of CHD patientsmet criteria of the EPO-H group in our hospitals (Table 2).Serum erythropoietin, ferritin, and percentage of transferrinsaturation revealed that the possibility of iron deficiencywas excluded in both EPO-H and EPO-R groups. The nutritionalstatus (body mass index and serum albumin levels), intact parathyroidhormone levels, the efficacy of HD (Kt/V urea values), and C-reactiveprotein did not show any significant differences between thetwo groups, and the patients in EPO-H group lacked clinicalevidence of malnutrition, inadequately controlled hyperparathyroidism,underdialysis, and inflammation. The number of isolated CD34+cells before culture was 1.16 ± 0.42 x 104/ml in theEPO-H group and 1.39 ± 0.39 x 104/ ml in the EPO-R group,and there was no statistical difference (Table 1).
Table 2. Known causes of EPO hyporesponsiveness in CHD patients (N = 256)
Colony Assays for BFU-E and GM-CFU Using Circulating CD34+ Cells
An equal number of purified CD34+ cells (1 x 104 cells) wassubjected to semisolid methylcellulose cultures and maintainedfor 14 d. The number of BFU-E was significantly less in theEPO-H group compared with the CON and EPO-R groups (Figure 1A).Furthermore, colony size was significantly smaller in the EPO-Hgroup compared with the CON and EPO-R groups (Figure 1B). Thenumber of GM-CFU was not significantly different among the threegroups (Figure 1C).
Figure 1. Number and area of burst forming unit-erythroid (BFU-E)-derived colonies in methylcellulose cultures. Circulating CD34+ cells (1 x 104 cells) were cultured in methylcellulose with 10 ng/ml IL-3, 3 U/ml erythropoietin (EPO), 50 ng/ml stem cell factor, and 10 ng/ml granulocyte-macrophage colony-stimulating factor (GM-CSF). The number of BFU-E is significantly less in the erythropoietin-hyperresponsive (EPO-H) group compared with control (CON) and erythropoietin-responsive (EPO-R) groups (A). Colony size is significantly smaller in the EPO-H group compared with the CON and EPO-R groups (B). Numbers of GM-CFU are not significantly different among the three groups (C). *P < 0.01.
Liquid Culture of HPC and Colony Assays for BFU-E
The purified CD34+ cells (1 x 104 cells) were cultured in liquidmedium for 7 d, then the cultured cells (1 x 104 cells) weretransferred into 1 ml of semisolid methylcellulose medium andmaintained for 14 d to access BFU-E colony formation. The numberof proliferated HPC at day 7 in liquid medium showed no statisticaldifference among the CON, EPO-H, and EPO-R groups (Figure 2A).However, the number of BFU-E was significantly less in the EPO-Hgroup compared with the CON and EPO-R group (P < 0.01; Figure 2, C through F).There was no statistical difference in thenumber of GM-CFU among the CON, EPO-H, and EPO-R groups (Figure 2B).The number of BFU-E showed almost equal levels by examiningCD34+ cells before liquid culture or after 7 d of liquid culturein the CON, EPO-H, and EPO-R groups (Figures 1A and 2F). Thus,we investigated CD34+ cells after 7 d of liquid culture to elucidatemolecular mechanism of reduced BFU-E in EPO-H patients.
Figure 2. BFU-E colony assay of hematopoietic progenitor cells (HPC) of STEM PRO 34 media culture. CD34+ cells (1 x 104 cells) were cultured in liquid medium for 7 d, and there is no difference in the count of HPC in liquid medium among EPO-H, EPO-R, and CON groups (A). The cultured cells (1 x 104 cells) were transferred into 1 ml of methylcellulose cultures and maintained for 14 d for BFU-E colony assay. There is no difference in the number of GM-CFU among EPO-H, EPO-R, and CON groups (B), whereas the number of BFU-E was significantly lower in the EPO-H group (E) compared with the CON (C) and EPO-R (D) groups (P < 0.01; F). Bar = 100 µm.
Western Blot Analysis of SHP-1, p-SHP-1, p-JAK2, and p-STAT5
The relative optical density of SHP-1 to GAPDH (Figure 3E) inthe EPO-H group (1.62 ± 0.21) was significantly highercompared with the EPO-R group (0.90 ± 0.15; P < 0.01;Figure 3, A and F) and the CON group (0.69 ± 0.14; P< 0.001; Figure 3, A and F). Furthermore, tyrosine phosphorylationof SHP-1 was significantly enhanced in the EPO-H group comparedwith both CON and EPO-R groups (P < 0.001; Figure 3, B and G).In contrast, the p-STAT5/GAPDH ratio was significantly higherin the EPO-R group (1.26 ± 0.20) and the CON group (1.43± 0.14) than in the EPO-H group (0.78 ± 0.13;P < 0.01; Figure 3, C and H). Because there were no significantdifferences in p-JAK2/GAPDH among the three groups (Figure 3, D and I),it suggested that the upregulated p-SHP-1 mainly dephosphorylatedp-STAT5 in cultured HPC derived from EPO-H patients.
Figure 3. Western blot analysis of src homology domain 2 (SH2)-containing tyrosine phosphatase-1 (SHP-1), phospho-SHP-1 (p-SHP-1), phosphorylated signal transducer and activator of transcription 5 (p-STAT5), and phosphorylated Janus kinase 2 (p-JAK2) in CD34+ cells that were cultured in STEM PRO 34 media for 7 d. The cultured CD34+ cells in STEM PRO 34 media (1 x 106 cells/ml) were washed twice at 4°C in liquid media in 1.5-ml microfuged tubes and centrifuged. Then, the pellets were lysed in 400 µl of RIPA buffer, and 10 µg of protein was loaded to SDS-PAGE and subjected to Western blot analysis of SHP-1 (A), p-STAT-5 (C), and p-JAK2 (D). (A) The relative optical density of SHP-1 to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in the EPO-H group (1.62 ± 0.21) is significantly higher compared with the EPO-R group (0.90 ± 0.15; P < 0.01) and the CON group (0.69 ± 0.14, P < 0.001; F). (B) In SHP-1 phosphorylation assay, 400 µg of CD34+ cell lysates was immunoprecipitated with anti–SHP-1 antibody, then immune complexes were adsorbed with 80 µl of protein A Sepharose CL4B overnight at 4°C. The proteins were eluted with 30 µl of Laemmli buffer and subjected to Western blot analysis using mouse monoclonal anti-phosphotyrosine antibody. Tyrosine phosphorylation of SHP-1 is enhanced in the EPO-H group compared with both the CON and the EPO-R groups (G). (C) The p-STAT5/GAPDH ratio is significantly higher in the EPO-R group (1.26 ± 0.20) and the CON group (1.43 ± 0.14) than in the EPO-H group (0.78 ± 0.13; H). (D) There are no significant differences in p-JAK2/GAPDH among the three groups (I). *P < 0.01; **P < 0.001.
Quantitative Real-Time RT-PCR
mRNA was isolated from the HPC that were cultured in liquidmedium for 7 d, and they were subjected to quantitative real-timeRT-PCR. SHP-1 mRNA expression, i.e., the copy number ratio (hematopoieticcell phosphatase/GAPDH), significantly upregulated in the EPO-Hgroup (0.67 ± 0.09 x 10–2) compared with the EPO-Rgroup (0.095 ± 0.04 x 10–2) and the CON group (0.10± 0.03 x 10–2; P < 0.001; Figure 4).
Figure 4. Quantitative real-time reverse transcription–PCR of SHP-1 mRNA in CD34+ cells that were cultured in STEM PRO 34 media for 7 d. First-stranded cDNA were synthesized from 1.0 µg of total RNA using ribonuclease H-MMLV reverse transcriptase and were added to the Lightcycler-Mastermix. The copy number of SHP-1 cDNA was normalized with a housekeeping gene (GAPDH), and the relative expression ratios are calculated. SHP-1 mRNA expression, i.e., the copy number ratio (SHP-1/GAPDH), significantly upregulated in EPO-H group compared with the EPO-R and CON groups. *P < 0.001 versus CON and EPO-R.
Gene Transfection Efficiency into CD34+ Cells by HVJ-E Vector
CD34+ cells (1 x 104) were suspended in 1 ml of STEM-PRO 34liquid medium, and they were centrifuged in the presence ofHVJ-E vector only or HVJ-E vector carrying pEGFPLuc. After 2d in culture, CD34+ cells were lysed and subjected to luciferaseassay. Luciferase activity was significantly increased in pEGFPLuctransfected CD34+ cells (2207 ± 105 relative light units)compared with control (138 ± 26 RLU; Figure 5A). By transfectingpcDNA3.1/V5-His-TOPO/lacZ by HVJ-E vector, blue-stained cellswere observed by -Gal staining, and transfection efficiencywas 30%. Transfection efficiency of FITC-labeled ODN was also30% of CD34+ cells.
Figure 5. SHP-1 antisense experiment. CD34+ cells (1 x 104) were suspended in 1 ml of STEM PRO 34 liquid medium, and they were centrifuged in the presence of hemagglutination virus of Japan-envelope (HVJ-E) vector only or HVJ-E vector carrying pEGFPLuc. Luciferase activity is significantly increased in pEGFPLuc transfected CD34+ cells (2207 ± 105 relative light units) compared with CON (138 ± 26 RLU; *P < 0.001; A). CD34+ cells derived from CON and EPO-H patients were treated with HVJ-E vector carrying SHP-1 antisense oligonucleotides (ODN) or HVJ-E vector carrying scrambled ODN. They were further cultured in STEM-PRO34 culture for 7 d and subjected to BFU-E colony assay and Western blot analysis. After a subsequent 14 d in culture on methylcellulose plates, BFU-E colonies significantly recovered by the treatment of SHP-1 antisense ODN compared with scrambled ODN-treated CD34+ cells in EPO patients (B through D). In contrast, SHP-1 antisense ODN did not increase the number of BFU-E compared with scrambled ODN-treated CD34+ cells in CON. By Western blot analysis, SHP-1/GAPDH ratios were significantly lower in antisense ODN-treated CD34+ cells compared with scrambled ODN (E). In turn, p-STAT5/GAPDH ratios significantly increased by SHP-1 antisense ODN compared with scrambled ODN-treated CD34+ cells (F). The mRNA expression of p-STAT5–regulated genes such as bcl-xL and oncostatin M significantly increased by SHP-1 antisense ODN (G and H). Hemoglobin 1 chain abundantly expressed, and there were no significant differences between SHP-1 antisense-treated and scrambled ODN-treated CD34+ cells (I). *P < 0.05, **P < 0.01.
SHP-1 Antisense Experiment Using CD34+ Cells Derived from EPO-H Patients
CD34+ cells that were derived from the EPO-H and CON groupswere treated with HVJ-E vector carrying SHP-1 antisense ODNor HVJ-E vector carrying scrambled ODN. They were further culturedin liquid media and subjected to BFU-E colony assay and Westernblot analysis. After 14 d of culture on methylcellulose plates,BFU-E colonies significantly recovered by the treatment of SHP-1antisense ODN compared with scrambled ODN-treated CD34+ cellsderived from EPO-H patients (P < 0.05; Figure 5, B through D).In contrast, SHP-1 antisense ODN did not alter the numberof BFU-E compared with scrambled ODN-treated CD34+ cells inCON. Thus, we further checked the status of SHP-1 and p-STAT5by Western blot analysis and quantified mRNA in p-STAT5–regulatedgenes by real-time RT-PCR. By Western blot analysis, SHP-1/GAPDHratios were significantly decreased in the antisense ODN group(0.93 ± 0.14) compared with the scrambled ODN-treatedCD34+ cells (1.72 ± 0.18; P < 0.01; Figure 5E). Inturn, p-STAT5/GAPDH ratios significantly increased by SHP-1antisense ODN (1.78 ± 0.30) compared with scrambled ODN-treatedCD34+ cells (1.26 ± 0.28; P < 0.05; Figure 5F). Toconfirm further the functional activation of p-STAT5 in theSHP-1 antisense ODN-treated cells, we evaluated mRNA expressionof p-STAT5–regulated genes, such as bcl-xL, oncostatinM, and hemoglobin 1. Copy number ratios of bcl-xL/GAPDH (0.08± 0.001) and oncostatin M/GAPDH (0.225 ± 0.054)in the antisense ODN-treated group were significantly highercompared with scrambled ODN–treated CD34+ cells, bcl-xL/GAPDH(0.05 ± 0.001), and oncostatin M/GAPDH (0.153 ±0.013; P < 0.01; Figure 5, G and H). We found that hemoglobin1 mRNA abundantly expressed, and there were no statistical differencesbetween antisense (hemoglobin 1/GAPDH ratio, 4.88 ± 6.62)and scramble ODN-treated (4.80 ± 8.33) CD34+ cells (Figure 5I).
It has not been fully investigated how CHD patients developchronic renal failure (CRF)-associated anemia or why these patientsinvariably reveal the hyporesponsiveness to exogenous administrationof rHuEPO. There are numerous causes for EPO hyporesponsiveness,such as iron deficiency, chronic inflammation, malignancies,malnutrition, underdialysis, secondary hyperparathyroidism,and gastrointestinal bleeding. Even after the elimination ofthese causes, 5 to 10% of CHD patients still developed hyporesponsivenessto rHuEPO (4). Previously, we reported that the number of circulatingCD34+ hematopoietic progenitors in such EPO-H patients is comparableto EPO-R patients and normal subjects (6). However, the numberof peripheral blood BFU-E, examined by a methylcellulose culture,is significantly reduced in EPO-H patients despite administrationof rHuEPO. We analyzed and compared the gene expression profileby extracting mRNA from BFU-E colonies and demonstrated theimpairment of the autocrine/paracrine regulatory loop of varioushematopoietic cytokines in EPO-H patients (6). Because we collectedonly a minute amount of cells from BFU-E colonies, we were notable to examine the activation of various signaling molecules.To address this issue, we isolated circulating CD34+ cells andcultured them in STEM PRO 34 liquid culture media, supplementedwith prescribed cytokines, which enabled us to examine the activationor phosphorylation of signaling molecules downstream of thecytokine receptors. Furthermore, BFU-E colony formation of CD34+cells, isolated from peripheral blood in EPO-H and EPO-R patients,was not altered by the expansion in STEM PRO 34 liquid culture.Thus, this proliferation assay system is appropriate and suitablefor evaluation of BFU-E formation and signaling events of variouscytokine receptors. By using the proliferation assay system,we demonstrated that mRNA and protein expression and tyrosinephosphorylation of SHP-1 significantly increased in CD34+ cellsderived from EPO-H patients compared with the EPO-R group. Westernblot analysis revealed that p-STAT5 was significantly lowerin the EPO-H group compared with the EPO-R group, whereas therewas no difference in p-JAK2. Therefore, we consider that upregulatedphosphorylated SHP-1 mainly dephosphorylates p-STAT5 and thusnegatively regulates EPO signaling in HPC and contributes tothe rHuEPO hyporesponsiveness in EPO-H patients (Figure 6).It is surprising that EPO-H patients without known causes ofEPO hyporesponsiveness have similar abnormalities in the EPOsignaling pathway, SHP-1 and STAT5. In human CD34+ cells anderythroblasts, erythropoietic factors activate three signalingpathways, JAK-STAT, mitogen-activated protein kinase p42/44,and phosphatidylinositol 3-kinase/Akt axes, and these processesare regulated by orchestrated activation of multiple signalingcascades. In preliminary data, we did not recognize apparentdifferences in phosphorylation of AKT and mitogen-activatedprotein kinase p42/44 in CD34+ cells in EPO-H patients; thus,the changes in phosphorylation of SHP-1 and STAT5 seems to bespecific events in EPO-H patients.
Figure 6. Schematic drawing of signal pathways in response to EPO. EPO receptor dimerization after EPO binding is schematically shown. Dimerized EPO receptors allow two JAK2 to approximate with each other and transphosphorylate active sites of JAK2. JAK2 subsequently phosphorylates tyrosine residues in the intracellular domains of EPO receptor, and then STAT5 binds to phosphorylated EPO receptors. STAT5 is phosphorylated by JAK2, dissociates from EPO receptors, and then forms dimer and translocates into the nucleus to activate the transcription of specific genes, including mouse oncostatin M protein and hemoglobin. SHP-1 binds to the negative regulatory domain of EPO receptor via its src-homology 2 (SH2) domain and causes dephosphorylation of JAK2 and STAT5; thus, it functions as a negative regulator of EPO signal transduction.
In addition to EPO signaling, SHP-1 negatively regulates theIL-3 signaling, i.e., SHP-1 associates with the -chain of IL-3receptor via the amino-terminal SH2 domain of SHP-1 after thebinding of ligand to the receptor, which then leads to dephosphorylationof the phosphorylated tyrosine residues. Paling et al. (19)demonstrated that SHP-1 negatively regulates IL-3 signalingin BaF/3 cells, potentially via regulation of tyrosine phosphorylationof STAT5. In the process of hematopoietic cell differentiation,IL-3 mainly regulates the proliferation of immature hematopoieticprogenitors (GM-CFU and early BFU-E), whereas EPO promotes bothproliferation and differentiation of later stages of erythroidprogenitors, including late BFU-E. One can speculate that SHP-1negatively regulates early BFU-E via dephosphorylation of IL-3receptor and STAT5 and plays a role in rHuEPO hyporesponsiveness.Matsuzaki et al. (20) reported that BFU-E formation in EPO-HCHD patients was suppressed under increasing concentration ofIL-3 (0.01 to 1 ng/ml), whereas normal response in GM-CFU formationsimilar to the control subjects was observed in the presenceof low-concentration IL-3 (0.01 ng/ml). Lower sensitivity toIL-3 for BFU-E formation indicated that SHP-1–mediatedinhibition of IL-3 signaling is critical in early BFU-E coloniesand not in GM-CFU formation. However, it remains unclear whySHP-1 selectively inhibits BFU-E colony formation in EPO-H patientsbecause IL-3 signaling is also critical for differentiationof myeloid progenitor series of cells, and SHP-1 as a tumorsuppressor in lymphoma or leukemia patients has been suggested(21,22).
The human SHP-1 gene consists of 17 exons that span 17 kb, includingtwo alternative 5' exon 1, which are driven by two differentpromoters (21,23). The hematopoietic form of the SHP-1 transcriptis initiated at a downstream promoter, promoter 2, separatedby 7 kb from the upstream promoter 1. The downstream promoter2 is active exclusively in cells of the hematopoietic lineage,whereas promoter 1 is active in various cells of nonhematopoieticorigin. Promoter 1 has two characteristic motifs for expressionof SHP-1: E-box, located 190 bp upstream of the transcriptionstarting site (CAP site), and NF-B binding site, located 105bp upstream of the E-box. In contrast, promoter 2 contains aninverted GATA box, a CCAAT box, and a TATA box region. Thus,the molecular genetic approaches do not provide enough datato explain promoter 2–driven hematopoietic-specific expressionof SHP-1 and its transcriptional regulation. For instance, diminishedor abolished expression of SHP-1 protein was reported in varioushematologic cell lines: B cell lymphoma, T cell lymphoma, naturalkiller cell lymphoma, chronic myelogenous leukemia, B cell acutelymphoblastic leukemia, and polycythemia vera. In these leukemiccell lines, DNA deletion or single-base mutations in SHP-1 genewere not detected, and the absence of expression of SHP-1 mRNAcould be due to the gene silencing resulting from the methylationof the cysteine residue in the CpG islands located in the regionencompassing promoter 2 and the first exon of hematopoieticisoform.
Why does SHP-1 significantly increase in CD34+ cells derivedfrom EPO-H patients? First, we can raise the possibility thatmultiple causative factors, such as uremic toxins, inflammation,and iron deficiency, may be involved in upregulated expressionof SHP-1. However, after the exclusion of CHD patients withsuch known causes for rHuEPO hyporesponsiveness, EPO-H patientsrevealed profound EPO hyporesponsiveness. As described above,there are no characteristic transcription factor binding sitesin the promoter region of SHP-1; thus, we cannot negate thepossibility that there are unknown conditions in the uremicstate that provoke rHuEPO hyporesponsiveness. Because upregulatedexpression of SHP-1 gene persisted even after the in vitro culturein STEM PRO 34 liquid media or methylcellulose supplementedwith various growth factors, the other plausible reason forupregulated expression of SHP-1 gene in EPO-H patients may beattributed to single nucleotide gene polymorphism in promoterregions or to the modification of the degree of methylationof the SHP-1 gene. As mentioned above, absence of expressionof SHP-1 mRNA in leukemic cells could be due to the gene silencingresulting from increased methylation of the cysteine residuein the CpG islands located in the region encompassing promoter2 and the first exon of hematopoietic isoform. Thus, we mayfurther investigate whether frequency of methylation of CpGislands decreases in CD34+ cells in EPO-H patients.
Because upregulated expression of SHP-1 was noted in EPO-H patients,antisense technology was used to confirm the functional roleof SHP-1 in the pathogenesis of rHuEPO hyporesponsiveness. Therehave been some reports regarding gene transfection into hematopoieticstem cells using adenovirus (24,25), electroporation (26,27),and recombinant HVJ (28). Recently, HVJ-E vector has come intouse to minimize the damage to transfected cells (29), and thisis a novel tool that can be used easily to transfer expressionplasmids, ODN, and proteins (30). In our initial studies, transfectionefficiency of FITC-labeled SHP-1 antisense ODN was 30% withoutsignificant cell toxicity. Such transfected CD34+ cells couldbe maintained in culture media and evaluated for the BFU-E colonyformation and expression of signal transduction molecules SHP-1and STAT5. After culture in methylcellulose media, the treatmentof SHP-1 antisense ODN decreased SHP-1 protein expression andupregulated p-STAT5, and it resulted in the partial recoveryof BFU-E colony formation. The current data indicate that SHP-1is a novel therapeutic target in the treatment of EPO-H patientswho undergo CHD therapy. In addition, the inhibition of SHP-1may be beneficial to improve the responsiveness to rHuEPO inCRF patients while reducing the dosage of rHuEPO administration.
In conclusion, we demonstrated that intracellular signal transductionof EPO receptor is attenuated as a result of dephosphorylationof STAT5 via SHP-1 phosphatase activity, thus contributing torHuEPO hyporesponsiveness in hematopoietic progenitors in EPO-HCHD patients. By suppression of SHP-1 protein using antisenseODN, we confirmed that SHP-1 is relevant in the pathogenesisof rHuEPO hyporesponsiveness in CHD patients. Thus, the futureefforts should be directed for screening of specific inhibitorsof SHP-1 phosphatase activity, and these agents may improvethe quality of life in CRF patients and also reduce the financialburden in CHD patients.
Acknowledgments
This work was supported by Uehara Memorial Foundation, The NaitoFoundation, ONO Medical Foundation, Grant-in-Aid for ScientificResearch (C), Ministry of Education, Science and Culture, Japan(14571025) to J.W. and Grant-in-Aid for Scientific Research(B), Ministry of Education, Science and Culture, Japan (14370319)to H.M. We gratefully thank Dr. Isao Kumagai at Teraoka MemorialHospital, Dr. Kazuhi Taniai, and Dr. Mie Kunitomi at OkayamaCentral Hospital for helpful discussion regarding the rHuEPOhyporesponsiveness observed in CHD patients.
Drueke TB, Eckardt KU: Role of secondary hyperparathyroidism in erythropoietin resistance of chronic renal failure patients. Nephrol Dial Transplant 17 [Suppl 5]: 28–31, 2002
Lin CL, Hsu PY, Yang HY, Huang CC: Low dose intravenous ascorbic acid for erythropoietin-hyporesponsive anemia in diabetic hemodialysis patients with iron overload. Ren Fail 25: 445–453, 2003[CrossRef][Medline]
Nakao K, Wada J, Ota K, Ichikawa H, Akagi S, Okamoto A, Hida K, Nagake Y, Makino H: Perturbation of autocrine/paracrine loops of burst-forming units of erythroid-derived cells in rHuEPO-hyporesponsive hemodialysis patients. Am J Kidney Dis 41: 624–636, 2003[CrossRef][Medline]
Watowich SS, Hilton DJ, Lodish HF: Activation and inhibition of erythropoietin receptor function: Role of receptor dimerization. Mol Cell Biol 14: 3535–3549, 1994[Abstract/Free Full Text]
Yoshimura A, Arai K: Physician education: The erythropoietin receptor and signal transduction. Oncologist 1: 337–339, 1996[Free Full Text]
He TC, Jiang N, Zhuang H, Wojchowski DM: Erythropoietin-induced recruitment of Shc via a receptor phosphotyrosine-independent, Jak2-associated pathway. J Biol Chem 270: 11055–11061, 1995[Abstract/Free Full Text]
Witthuhn BA, Quelle FW, Silvennoinen O, Yi T, Tang B, Miura O, Ihle JN: JAK2 associates with the erythropoietin receptor and is tyrosine phosphorylated and activated following stimulation with erythropoietin. Cell 74: 227–236, 1993[CrossRef][Medline]
Gregory RC, Jiang N, Todokoro K, Crouse J, Pacifici RE, Wojchowski DM: Erythropoietin receptor and STAT5-specific pathways promote SKT6 cell hemoglobinization. Blood 92: 1104–1118, 1998[Abstract/Free Full Text]
Shen SH, Bastien L, Posner BI, Chretien P: A protein-tyrosine phosphatase with sequence similarity to the SH2 domain of the protein-tyrosine kinases. Nature 352: 736–739, 1991[CrossRef][Medline]
Klingmuller U, Lorenz U, Cantley LC, Neel BG, Lodish HF: Specific recruitment of SH-PTP1 to the erythropoietin receptor causes inactivation of JAK2 and termination of proliferative signals. Cell 80: 729–738, 1995[CrossRef][Medline]
Wickrema A, Chen F, Namin F, Yi T, Ahmad S, Uddin S, Chen YH, Feldman L, Stock W, Hoffman R, Platanias LC: Defective expression of the SHP-1 phosphatase in polycythemia vera. Exp Hematol 27: 1124–1132, 1999[CrossRef][Medline]
Mrug M, Stopka T, Julian BA, Prchal JF, Prchal JT: Angiotensin II stimulates proliferation of normal early erythroid progenitors. J Clin Invest 100: 2310–2314, 1997[Medline]
Melrose J, Shen B, Ghosh P: Affinity and Western blotting reveal homologies between ovine intervertebral disc serine proteinase inhibitory proteins and bovine pancreatic trypsin inhibitor. Proteomics 1: 1529–1533, 2001[CrossRef][Medline]
Oselin K, Mrozikiewicz PM, Pahkla R, Roots I: Quantitative determination of the human MRP1 and MRP2 mRNA expression in FACS-sorted peripheral blood CD4+, CD8+, CD19+, and CD56+ cells. Eur J Haematol 71: 119–123, 2003[CrossRef][Medline]
Sharlow ER, Pacifici R, Crouse J, Batac J, Todokoro K, Wojchowski DM: Hematopoietic cell phosphatase negatively regulates erythropoietin-induced hemoglobinization in erythroleukemic SKT6 cells. Blood 90: 2175–2187, 1997[Abstract/Free Full Text]
Paling NR, Welham MJ: Role of the protein tyrosine phosphatase SHP-1 (Src homology phosphatase-1) in the regulation of interleukin-3-induced survival, proliferation and signalling. Biochem J 368: 885–894, 2002[CrossRef][Medline]
Matsuzaki Y, Mizuguchi T, Kosaka M, Saito S: Analysis of circulating hematopoietic progenitors in patients with chronic renal failure under hemodialysis. Int J Hematol 63: 33–40, 1996[CrossRef][Medline]
Wu C, Sun M, Liu L, Zhou GW: The function of the protein tyrosine phosphatase SHP-1 in cancer. Gene 306: 1–12, 2003[CrossRef][Medline]
Oka T, Ouchida M, Koyama M, Ogama Y, Takada S, Nakatani Y, Tanaka T, Yoshino T, Hayashi K, Ohara N, Kondo E, Takahashi K, Tsuchiyama J, Tanimoto M, Shimizu K, Akagi T: Gene silencing of the tyrosine phosphatase SHP1 gene by aberrant methylation in leukemias/lymphomas. Cancer Res 62: 6390–6394, 2002[Abstract/Free Full Text]
Banville D, Stocco R, Shen SH: Human protein tyrosine phosphatase 1C (PTPN6) gene structure: Alternate promoter usage and exon skipping generate multiple transcripts. Genomics 27: 165–173, 1995[CrossRef][Medline]
Sakurai F, Mizuguchi H, Hayakawa T: Efficient gene transfer into human CD34+ cells by an adenovirus type 35 vector. Gene Ther 10: 1041–1048, 2003[CrossRef][Medline]
Shayakhmetov DM, Carlson CA, Stecher H, Li Q, Stamatoyannopoulos G, Lieber A: A high-capacity, capsid-modified hybrid adenovirus/adeno-associated virus vector for stable transduction of human hematopoietic cells. J Virol 76: 1135–1143, 2002[Abstract/Free Full Text]
Wu MH, Smith SL, Dolan ME: High efficiency electroporation of human umbilical cord blood CD34+ hematopoietic precursor cells. Stem Cells 19: 492–499, 2001[Abstract/Free Full Text]
Weissinger F, Reimer P, Waessa T, Buchhofer S, Schertlin T, Kunzmann V, Wilhelm M: Gene transfer in purified human hematopoietic peripheral-blood stem cells by means of electroporation without prestimulation. J Lab Clin Med 141: 138–149, 2003[CrossRef][Medline]
Jin CH, Kusuhara K, Yonemitsu Y, Nomura A, Okano S, Takeshita H, Hasegawa M, Sueishi K, Hara T: Recombinant Sendai virus provides a highly efficient gene transfer into human cord blood-derived hematopoietic stem cells. Gene Ther 10: 272–277, 2003[CrossRef][Medline]
Shimamura M, Morishita R, Endoh M, Oshima K, Aoki M, Waguri S, Uchiyama Y, Kaneda Y: HVJ-envelope vector for gene transfer into central nervous system. Biochem Biophys Res Commun 300: 464–471, 2003[CrossRef][Medline]
Kaneda Y, Nakajima T, Nishikawa T, Yamamoto S, Ikegami H, Suzuki N, Nakamura H, Morishita R, Kotani H: Hemagglutinating virus of Japan (HVJ) envelope vector as a versatile gene delivery system. Mol Ther 6: 219–226, 2002[CrossRef][Medline]
Received for publication January 16, 2004.
Accepted for publication August 31, 2004.
Related Article
This Month’s Highlights
J. Am. Soc. Nephrol. 2004 15: 2955-2958.
[Full Text][PDF]
This article has been cited by other articles:
I. C. Macdougall Novel Erythropoiesis-Stimulating Agents: A New Era in Anemia Management
Clin. J. Am. Soc. Nephrol.,
January 1, 2008;
3(1):
200 - 207.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
- and 
-thalassemia; (5) no gastrointestinal bleeding; (6) malnutrition; and (7) underdialysis or secondary hyperparathyroidism. Written informed consent was obtained from all patients before the investigation. 
5% of CHD patients met criteria of the EPO-H group in our hospitals (Table 2). Serum erythropoietin, ferritin, and percentage of transferrin saturation revealed that the possibility of iron deficiency was excluded in both EPO-H and EPO-R groups. The nutritional status (body mass index and serum albumin levels), intact parathyroid hormone levels, the efficacy of HD (Kt/V urea values), and C-reactive protein did not show any significant differences between the two groups, and the patients in EPO-H group lacked clinical evidence of malnutrition, inadequately controlled hyperparathyroidism, underdialysis, and inflammation. The number of isolated CD34+ cells before culture was 1.16 ± 0.42 x 104/ml in the EPO-H group and 1.39 ± 0.39 x 104/ ml in the EPO-R group, and there was no statistical difference (Table 1). 
). Thus, we investigated CD34+ cells after 7 d of liquid culture to elucidate molecular mechanism of reduced BFU-E in EPO-H patients. 
B binding site, located 105 bp upstream of the E-box. In contrast, promoter 2 contains an inverted GATA box, a CCAAT box, and a TATA box region. Thus, the molecular genetic approaches do not provide enough data to explain promoter 2–driven hematopoietic-specific expression of SHP-1 and its transcriptional regulation. For instance, diminished or abolished expression of SHP-1 protein was reported in various hematologic cell lines: B cell lymphoma, T cell lymphoma, natural killer cell lymphoma, chronic myelogenous leukemia, B cell acute lymphoblastic leukemia, and polycythemia vera. In these leukemic cell lines, DNA deletion or single-base mutations in SHP-1 gene were not detected, and the absence of expression of SHP-1 mRNA could be due to the gene silencing resulting from the methylation of the cysteine residue in the CpG islands located in the region encompassing promoter 2 and the first exon of hematopoietic isoform. 
