Business & Economics

Comparison of ZAP-70/Syk mRNA levels with immunoglobulin heavy-chain gene mutation status and disease progression in chronic lymphocytic leukemia

The protein tyrosine kinase ZAP-70 has recently emerged as a major prognostic indicator in chronic lymphocytic leukemia (CLL). ZAP-70 is structurally and functionally homologous to Syk, a key mediator of B-cell receptor signaling. We therefore
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  haematologica/the hematology journal | 2005; 90(11) | 1533| Luca Laurenti Aleksandar PetlickovskiCarlo RumiStefania GobessiPaola PiccioniMichela TarnaniPierluigi PuggioniSara MariettiSimona SicaGiuseppe LeoneDimitar G.Efremov Comparison of ZAP-70/Syk mRNA levels withimmunoglobulin heavy-chain gene mutation status anddisease progression in chronic lymphocytic leukemia T he clinical course and outcome of chronic lymphocytic leukemia (CLL)is highly variable. Some patientshave rapidly progressive disease and dieearly in spite of therapy, whereas othersexhibit a stable disease and have a normallife span. 1 Prognosis in CLL is classicallyassessed using the Rai and Binet staging sys-tems, which are based on clinical and hema-tologic parameters. 2,3 Although effective inclassifying CLL patients into broad prognos-tic subgroups, these staging systems cannotpredict the course of the disease in individ-ual patients, in particular those with earlystage disease. Therefore, novel genetic andbiological parameters, such as immunoglob-ulin V  H gene mutation status, cytogeneticabnormalities and expression of CD38 andZAP-70, are increasingly being used to pre-dict the risk of disease progression. 4-13 Among these, the absence of somatic muta-tions in the V  H genes appears to be a partic-ularly strong prognostic indicator of aggres-sive disease. However, sequencing of rearranged V  H genes is rather cumbersomeand expensive, and therefore has not beenadopted as a routine laboratory investiga-tion. ZAP-70 is a protein tyrosine kinase whichplays a key role in proximal T-cell receptorsignaling. Until recently, it was believed thatZAP-70 is expressed only in T-cells and nat-ural killer (NK) cells. However, gene expres-sion profiling with cDNA microarraysrevealed that ZAP-70 is expressed also inCLL B cells, particularly in those withunmutated V  H genes. 14 Subsequently,expression of ZAP-70 was evaluated by sev-eral groups using different flow-cytometryprocedures. 11-13,15 These studies confirmedthe association between ZAP-70 expressionand V  H gene mutation status and showedthat CLL patients with a high percentage of ZAP-70 positive leukemic cells have signifi- From the Department of Hematology,Catholic University Hospital "A.Gemelli",Rome,Italy (LL,CR,PP,MT,PP,SS,GL);ICGEBHematology Group,Monterotondo-Outstation,CNR Campus"A.Buzzati-Traverso",Rome,Italy (AP,SG,SM,DGE).Correspondence: Dimitar Efremov,M.D.,Ph.D.,ICGEB Outstation,MonterotondoCNR Campus "Adriano Buzzati-Traverso",Via E.Ramarini 32I-00016 Monterotondo Scalo,Rome,Italy.E-mail: Luca Laurenti,M.D.Department of Hematology Catholic University Medical School Largo A.Gemelli 8,I-00168 Rome,  Background and Objectives. The protein tyrosine kinase ZAP-70 has recently emergedas a major prognostic indicator in chronic lymphocytic leukemia (CLL). ZAP-70 is struc-turally and functionally homologous to Syk,a key mediator of B-cell receptor signaling.We therefore evaluated ZAP-70 expression in CLL B cells using Syk as an intracellularstandard.Design and Methods. The relative amounts of ZAP-70 and Syk were determined in puri-fied B cells from 92 CLL patients using a novel reverse transcriptase/polymerasechain reaction (RT-PCR) procedure that co-amplifies both transcripts with equal efficien-cy. The ZAP-70/Syk mRNA ratio was correlated with V H gene mutation status,mediantreatment-free survival and FACS analysis of ZAP-70 expression. Results. ZAP-70 was expressed in the majority of cases with unmutated V H genes(88%),but also at lower levels in a substantial fraction of cases with mutated V H genes(44%). High levels of ZAP-70,defined as ZAP-70/Syk mRNA ratios above 0.25,wereobserved mainly in cases with unmutated V H genes and correlated with short treat-ment-free survival. In contrast,no difference was observed in the median treatment-free survival between patients with low ZAP-70/Syk ratios (0.05-0.25) and patientswith no or negligible ZAP-70 expression (ZAP-70/Syk<0.05). In 73 cases ZAP-70expression was investigated by RT/PCR and FACS analysis; concordance with V H genemutation status was 86% and 71%,respectively.Interpretation and Conclusions.ZAP-70 is frequently expressed in CLL B cells,but only high levels correlate with unmutated V H gene status and progressive disease.Expression of ZAP-70 can be accurately assessed by analysis of the ZAP-70/Syk mRNAratio,thus providing an alternative to FACS analysis. Key words:chronic lymphocytic leukemia,immunoglobulin V H gene mutation status,ZAP-70,Syk,prognosis.Haematologica 2005; 90:1533-1540©2005 Ferrata Storti Foundation Chronic Lymphocytic Leukemia • Research Paper   cantly shorter treatment-free and overall survival.However, the flow-cytometry analysis also revealedthat expression of ZAP-70 is a continuous variable,thus requiring an arbitrary cut-off value to define posi-tive and negative cases. 11-13 Moreover, in the two largestseries the percentage of ZAP-70-positive CLL cells wasclose to the cut-off value in a significant fraction of cases. 13,15 Considering that the flow-cytometry proce-dures are inherently prone to small imprecisions, it islikely that precise assignment of ZAP-70 positivity maynot be possible in all cases. Therefore, alternativeapproaches based on different methodological andstandardization principles would be useful in validatingthe flow-cytometry data, especially in cases withequivocal results.The possible role for ZAP-70 in the pathogenesis of CLL is also a matter of substantial interest. The signifi-cant homology with Syk indicates that ZAP-70 mayalso play a role in CLL B-cell receptor (BCR) signaling.Furthermore, ZAP-70 can reconstitute some aspects of BCR function in Syk-deficient lymphoma B cells, indi-cating that the two protein tyrosine kinases are func-tionally homologous. 16 More recently, ZAP-70 wasshown to associate directly with the BCR complex inCLL B cells and to undergo tyrosine phosphorylationfollowing stimulation by antigen. 17 In these cellsstronger activation of other BCR signaling moleculeswas also observed, indicating that ZAP-70 can enhanceBCR signal transduction. 18 These findings are intrigu-ing, considering that ZAP-70 was considered inferior toSyk as an immunoreceptor signaling molecule, due toits lower intrinsic enzymatic activity, its lower capaci-ty to associate with certain tyrosine phosphorylatedimmunoreceptor tyrosine-based activation notifs(ITAM) and its requirement for activation by Src fami-ly kinases. 19-21 Therefore, in order to understand betterhow ZAP-70 can affect BCR signaling, it would be use-ful to evaluate the relative quantities of these two relat-ed protein tyrosine kinases in CLL B cells. In the present study, we took advantage of the con-siderable homology between the coding sequences of ZAP-70 and Syk to develop an RT-PCR procedure thatallows accurate measurement of the relative levels of the two transcripts. We used this procedure to deter-mine the relative expression of ZAP-70 and Syk in puri-fied CLL B cells from 92 patients. The findings werecorrelated with the V  H gene mutation status, percent-age of ZAP-70 positive cells and clinical course of thepatients. Design and Methods Patients and cell samples Blood samples were collected from 92 patientswhose morphologic and immunophenotypic featuressatisfied standard criteria for B-cell CLL. 22 Informedconsent was obtained from all patients according tothe Declaration of Helsinki and approval for the studywas obtained from the institutional human researchcommittee at the Catholic University Medical Schoolin Rome. Sixty-two patients (67%) were male and themedian age of all the patients was 63 years (range 33to 84). At diagnosis 72 patients (78%) were in Binetstage A, 15 patients (16%) in stage B and 5 patients(6%) in stage C. The median duration of follow-upfrom diagnosis was 48 months. Fifty-seven (62%)patients had never been treated and 35 (38%) hadreceived treatment at some stage during the course of the disease. Treatment was initiated for symptomaticor progressive disease, according to National CancerInstitute Working Group criteria. 22 The patients whohad been treated had not received chemotherapy orsteroids for at least 3 months prior to the sampling. Peripheral blood mononuclear cells were separatedby Ficoll gradient centrifugation (AmershamBiosciences, Uppsala, Sweden). B cells were isolatedwith the use of magnetic beads coupled to monoclon-al antibodies specific for CD19 (Dynal Biotech, Oslo,Norway). CD19 selection typically resulted in morethan 98%, purity as assessed by flow cytometry. Insome experiments purification was also performedwith the B-cell negative isolation kit (Dynal Biotech) orusing Dynabeads M-450 (Dynal Biotech) coated withgoat anti-human IgM (Southern BiotechnologyAssociates, Inc., USA). 23 Quantification of the ZAP-70/Syk mRNA levels Measurement of ZAP-70 and Syk mRNA wasbased upon a similar procedure used for the quantifi-cation of the relative amounts of A g  and G g  globinmRNA. 24 Briefly, total cellular RNA was isolated fromthe CD19-selected B cells using Trizol reagent(Invitrogen), according to the manufacturer's instruc-tions. RNA (1 m g) was reverse transcribed using ran-dom hexamers and then PCR amplified with primersZSf 5' TCATGCAC(G)CAGCTGGACAACC 3' andZSr 5' CCAG(A)GTACTTCATC(G)CCCATGGA 3'.The ZSf primer was 5' end labeled with the fluores-cent dye TET. The PCR reactions were performed in100 m L for 30 cycles. A 2 m L aliquot of the PCR wasdigested for 2 hours with 6 U of the restrictionenzyme BstNI (New England BioLabs, Beverly, MA,USA). One tenth of the reaction volume was dena-tured and the PCR products were size-separated andquantified on an ABI PRISM 310 capillary elec-trophoresis system using the GeneScan Analysis3.1.2 software (Applied Biosystems, Forster City, CA,USA). ZAP-70 and Syk plasmid DNA standards were pre-pared by RT-PCR of mRNA from the Jurkat T-celllymphoma and BJAB B-cell lymphoma cell lines,respectively. The PCR fragments were cloned in the | 1534| haematologica/the hematology journal | 2005; 90(11)L. Laurenti et al.  pDrive cloning vector using the PCR cloning plus kit (Qia-gen, Valencia, CA, USA). Plasmids containing the ZAP-70 and Syk fragments were linearized by restrictionenzyme digestion and mixed in equimolar amounts. Ig V  H   gene sequence analysis The PCR amplification, cloning, and sequencing of V  H region genes have been described in detail elsewhere. 25 Briefly, RNA was reverse transcribed using randomhexamers and then PCR amplified with a degenerate V  H FWR1 primer in combination with C m , C g  and C a reverse primers. Samples that failed to amplify withthese combinations were amplified with a mixture of forward primers complementary to leader sequences of  V  H families 1 to 6. PCR products were purified with theQIAquick PCR purification kit (Qiagen) and eithersequenced directly or cloned with the PCR cloning plus kit(Qiagen). Sequencing was done with the BigDyeTerminator v3.1 Cycle Sequencing kit and ABI 3100genetic analyzer (Applied Biosystems). Candidategermline genes were assigned by searching the VBASEdirectory. Percentage homology was calculated bycounting the number of mutations between the 5' endof FR1 and the 3' end of FR3. Sequences with less than2% differences from germline V  H sequences were con-sidered unmutated. Immunoblot analysis Cell pellets were lysed in ice-cold 1% Nonidet P-40(NP-40) lysis buffer (10 mM Tris-HCl, pH 7.4, 5 mMEDTA, 150 mM NaCl, 0.1% sodium dodecyl sulfate(SDS), 0.1% sodium deoxycholate) containing 1:40dilution of a protease inhibitor cocktail for mammaliancells (Sigma-Aldrich). The protein concentration of eachcell lysate was determined with the RC DC ProteinAssay (Bio-Rad Laboratories, Hercules, CA, USA). Theprotein samples (20 m g/lane) were separated by 10%sodium dodecyl sulfate-polyacrylamide gel elec-trophoresis (SDS-PAGE) and transferred on Immobilon-P polyvinilidene difluoride membranes (Millipore,Bedford, MA, USA). Membranes were blotted at 4°Cwith antibodies against ZAP-70, Syk (both from CellSignaling Technology) and b -actin (Sigma, Saint Louis,MO, USA). Immunodetection was done with anti-rab-bit IgG horse-radish peroxidase-linked or anti-mouseIgG horse-radish peroxidase-linked antibodies (CellSignaling Technology) and the ECL Plus enhanced-chemiluminiscence detection system (AmershamBiosciences, Buckinghamshire, UK) with BioMax MR films (Eastman Kodak, Rochester, NY, USA). Flow cytometry analysis of ZAP-70 Fifty microliters of whole blood in EDTA containingapproximately 5 ¥ 10 5 cells were incubated at room tem-perature in the dark for 20 minutes with CD3 PE-Cy5,CD56 PE-Cy5, CD5 FITC and CD19 PE-Cy7 to identi-fy membrane antigens. After incubation, samples werecentrifuged at 1500 rpm for 5 minutes. Cell pellets wereresuspended in 100 m L of fixation medium A (CaltagLaboratories, Burlingame, CA, USA) and incubated atroom temperature in the dark for 15 minutes. Then,cells were pelleted again at 1500 rpm for 5 minutes,resuspended in 100 m L of permeabilization medium B(Caltag Laboratories) and incubated with the ZAP-70monoclonal antibody at room temperature in the darkfor one hour. The ZAP-70 antibody was R-phycoery-thrin conjugated (clone 1E 7.2, Caltag Laboratories).Samples were washed twice by centrifugation at 1500rpm for 5 minutes and resuspended in 500 m L of phos-phate-buffered saline. Finally, 1 ¥ 10 4 fixed cells wereanalyzed by flow cytometry (BD FACSCanto, BectonDickinson) using the gating strategy described byCrespo et al  . 11 Statistical analysis Correlations between ZAP-70 expression and V  H gene mutational status were analyzed using Wilcoxon’srank-sum test. The cut-off ZAP-70/Syk value that bestdiscriminated mutated from unmutated cases wasdetermined using a receiver operating characteristicplot. The median treatment-free survival, calculatedfrom diagnosis to initial therapy, was estimated by themethod of Kaplan and Meier and assessed by the log-rank test. Data for patients who had not received treat-ment were regarded as censored. Statistical analyseswere performed using the SigmaStat 3.1 program(Systat Software Inc., Richmond, CA, USA). Results Human ZAP-70 and Syk have approximately 60%homology at the mRNA level. We therefore searchedfor short stretches of sequence identity to design PCR primers expected to co-amplify both transcripts withequal efficiency. Two such sequences were identified,separated by a 150 nt region that contained BstNIrestriction enzyme sites at different locations in ZAP-70and Syk (Figure 1A). These restriction enzyme siteswere subsequently used to distinguish between theZAP-70 and Syk PCR fragments. To verify that ZAP-70 and Syk can be co-amplifiedwith equal efficiency we first tested cloned ZAP-70 andSyk cDNA. Serial dilutions of an equimolar mixture of the two fragments were subjected to PCR amplificationfor various numbers of cycles. The PCR products weredigested with BstNI and the obtained fragments of 51and 62 bp, corresponding to ZAP-70 and Syk, respec-tively, were separated and quantified by capillary elec-trophoresis on a genetic analyzer. The ZAP-70/Sykratio obtained in all instances was identical to the inputplasmid DNA (Figure 1B). haematologica/the hematology journal | 2005; 90(11) | 1535|   ZAP-70/Syk mRNA ratios in CLL  L. Laurenti et al.| 1536| haematologica/the hematology journal | 2005; 90(11)  ZAP-70/Syk mRNA levels and V  H   gene mutation status We next tested the procedure in leukemic B-cellspurified by CD19 positive selection. The purity of the leukemic cell population was confirmed by flowcytometry using antibodies specific for CD5 andCD19. Two typical results are shown in Figure 2A. Ininitial experiments the CLL B cells were also purifiedby positive selection with anti-IgM coated Dyna-beads or by negative selection with a mixture of anti-bodies specific for CD2, CD3, CD7, CD14, CD16,and CD56. No differences were observed in the ZAP-70/Syk mRNA ratios with the use of the differentpurification procedures ( data not shown ). The ZAP-70/Syk mRNA ratio was then determinedin CD19-purified leukemic B cells from 92 CLLpatients. In 37 cases (40%) ZAP-70 was either not Figure 1. Analysis of the ZAP-70/Syk mRNA ratio by quantitative RT-PCR A. Alignment of the homologous ZAP-70 and Syk nucleotide sequences. The sequences used to design the PCR primers are shown in bold upper case letters. The BstN I site in each fragment isunderlined. Following restriction enzyme digestion,51 bp and 62 bp TET-labeled fragments are obtained for ZAP-70 and Syk,respective-ly. B. An equimolar mixture of plasmids with cloned ZAP-70 and Syk cDNA fragments was diluted 10 and 100 fold and the three sam-ples were subjected to PCR, BstN I restriction enzyme digestion and capillary electrophoresis on a genetic analyzer. The ZAP-70/Sykratios obtained and the position of the ZAP-70 and Syk fragments are indicated in each diagram. The analysis shown is after 30 cyclesof PCR. Identical results were obtained when the same mixtures were subjected to 40 PCR cycles ( data not shown ). AB Figure 2.Analysis of the ZAP-70/Syk mRNAratio in CLL B cells. A. CLL B cells were pos-itively selected with CD19 Dynabeads andanalyzed by flow cytometry to confirm thepurity of the leukemic cell population,priorto investigation of the ZAP-70/Syk mRNAratio. Two representative analyses areshown (sample CLL-G71 without detectableZAP-70 mRNA and sample CLL-G112 withhigh levels of ZAP-70 mRNA). The ZAP-70/Syk ratios and the position of the ZAP-70 and Syk fragments are indicated in bothcases. B. ZAP-70/Syk mRNA ratios in 59 V H mutated and 33 V H unmutated cases withCLL. ZAP 70SykZAP 70SykZAP-70/Syk standard ZAP-70/Syk standard, 10 ¥ dilution ZAP-70/Syk standard, 100 ¥ dilution ZAP-70/Syk=1.03 ZAP-70/Syk= 0.98 ZAP-70/Syk= 1.0660 80 60 10040 80 60 10040 80ZAP-70 Syk ZAP-70 Syk ZAP-70Syk 1.601.401.201.000.800.600.400.200     Z    A    P   7    0    /    S   y    k   m    R    N    A   r   a   t    i   o Mutated CLL(n=59)Unmutated CLL(n=33) AB CD19 pos, CLL-G71 CD19 pos, CLL-G112CD19 FITC ZAP-70/SykZ/S=0.00 Z/S=0.810.33     C    D   5    P    E 10 0 10 1 10 2 10 3 10 4 CD19 FITC10 0 10 1 10 2 10 3 10 4    1    0     0    1    0    1    1    0     2    1    0     3    1    0    4     C    D   5    P    E   1    0     0    1    0    1    1    0     2    1    0     3    1    0    4 ZAP-70/Syk   90 80 90 80  ZAP-70/Syk mRNA ratios in CLLhaematologica/the hematology journal | 2005; 90(11) | 1537| detected or the amount was considered negligible(ZAP-70/Syk<0.05). ZAP-70 mRNA was present inthe remaining 55 samples (60%), with ZAP-70/SykmRNA values ranging from 0.07 to 1.58. The levelsof ZAP-70 rarely exceeded the levels of Syk (onlythree cases had a ZAP-70/Syk ratio >1.0), indicatingthat Syk is predominantly expressed in CLL B cells.To investigate the association with V  H gene muta-tion status, we compared the ZAP-70/Syk mRNAratios in CLL cells with mutated and unmutated V  H genes (Figure 2B). ZAP-70 mRNA was detected in 26 of the 59 (44%) V  H mutated and 29 of the 33 (88%)  V  H unmutated samples. However, the levels of ZAP-70 were on average significantly lower among theZAP-70 positive cases with mutated V  H genes (medi-an ZAP-70/Syk mRNA ratio 0.15, range 0.07 to 0.90)than among those with unmutated V  H genes (median0.55, range 0.13 to 1.58,  p <0.001, Mann-Whitneyrank sum test). Therefore, we investigated whether aparticular ZAP-70/Syk mRNA ratio can be used todiscriminate between the V  H mutated and V  H unmu-tated cases. Using receiver operating characteristicplot analysis we observed that a ZAP-70/Syk ratio of 0.25 could correctly assign the V  H gene mutation sta-tus in 86% of the cases. Based on this threshold 82%of the V  H unmutated and only 12% of the V  H mutat-ed samples were classified as cases with high ZAP-70expression. Nineteen V  H mutated and two V  H unmutated caseshad detectable ZAP-70 mRNA that was below the0.25 threshold. To verify that these cases expressZAP-70 protein, we performed western blottinganalysis with CLL cells purified by CD19-positiveselection from 20 representative cases. The cell lines Jurkat and BJAB were used as a positive and negativecontrol, respectively. The samples were probed withantibodies against ZAP-70, Syk and b -actin. CLL Bcells with ZAP-70/Syk values between 0.05 and 0.25had readily detectable ZAP-70 protein, although con-siderably less than the amount detected in the sam-ples with ZAP-70/Syk ratios above 0.25 (Figure 3).Only a very faint band was observed in samples withZAP-70/Syk ratios below 0.05, which again validatedthe data of the RT-PCR assay. Correlation between ZAP-70/Syk mRNA levels and time to initial therapy  We next compared the clinical course in patients withdifferent levels of ZAP-70 expression (Figure 4A). Thepatients were stratified into three groups according tothe ZAP-70/Syk mRNA ratio: (i) patients with negligi-ble/undetectable ZAP-70 (ZAP-70/Syk<0.05), (ii)patients with low ZAP-70 expression (ZAP-70/Syk=0.05-0.25), and (iii) patients with high ZAP-70expression (ZAP-70/Syk>0.25). The median treatment-free survival was significantly shorter in the group of patients with high ZAP-70 expression (30 months)compared to the groups with low or negligible/unde-tectable ZAP-70 (not reached in both groups,  p =0.0005and  p =0.00001, respectively) (Figure 4A). In contrast,the median treatment-free survival of the group withlow ZAP-70 expression was not significantly differentfrom that of the group with negligible/undetectableZAP-70 mRNA (  p =0.92), indicating that only high lev-els of ZAP-70 are associated with an unfavorable clini-cal course. The survival curves of cases with low or neg-ligible/undetectable ZAP-70 levels were similar to thesurvival curve of patients with mutated V  H genes(Figure 4B). Likewise, the median treatment-free sur-vival in patients with high ZAP-70 expression was sim-ilar to that in patients with unmutated V  H genes (18months). We also examined in detail the discordant cases forZAP-70 expression and V  H gene mutation status. Theseven V  H mutated cases with high ZAP-70 expression(>0.25) had 89 to 96% homology with a known germ-line V  H gene. Two of these patients had received firsttreatment at 51 and 60 months after diagnosis, where- Figure 3. Expression of ZAP-70 and Syk protein in CLL B cells and lymphoma cell lines. Immunoblotting was performed with antibodiesspecific for ZAP-70,Syk and b actin on lysates from purified CD19-positive CLL cells. An analysis of 14 representative cases is shown.The T-cell lymphoma cell line Jurkat and the B-cell lymphoma cell line BJAB were included as a positive and negative control for ZAP-70,respectively. B104 is a B-cell lymphoma cell line that co-expresses ZAP-70 and Syk. The V H gene mutational status and the ZAP-70/Syk mRNA ratio for each sample is shown at the top of the figure. G83 G77 G41 G22G98 G106G71U M M M M U M0.18 0.16 0.010.560.0 0.550.10G19 G13 G88 G108G45 G2 G18 Jur.B104 BJABM M U U U U U  --- 0.00 0.020.27 0.700.130.00 0.94  - 0.20 0.00 ZAP-70 SampleMutational statusZAP-70/Syk mRNA Syk b -actin
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks