- © 2002 by American Society of Clinical Oncology
Overexpression of the Polycythemia Rubra Vera-1 Gene in Essential Thrombocythemia
- Luciana Teofili,
- Maurizio Martini,
- Myriam Luongo,
- Antonella Di Mario,
- Giuseppe Leone,
- Valerio De Stefano and
- Luigi Maria Larocca
- From the Istituto di Ematologia and Istituto di Anatomia Patologica, Università Cattolica del Sacro Cuore, Rome, Italy.
- Address reprint requests to Luigi Maria Larocca, MD, Istituto di Anatomia Patologica, Università Cattolica del Sacro Cuore, Largo F. Vito, 1, 00168 Rome, Italy; email: llarocca{at}rm.unicatt.it
Abstract
PURPOSE: To examine the utility of polycythemia rubra vera-1 (PRV-1)–specific reverse transcriptase polymerase chain reaction (RT-PCR) to discriminate essential thrombocythemia (ET) and polycythemia vera (PV) from secondary thrombocytosis (ST) or secondary erythrocytosis (SE).
PATIENTS AND METHODS: We analyzed the expression of PRV-1 in granulocytes isolated from 37 patients with ET, 37 patients with PV, 25 patients with ST, 10 patients with SE, 25 patients with secondary leukocytosis (SL), five patients with chronic myelogenous leukemia (CML), five patients with chronic idiopathic myelofibrosis (IM), five patients with myelodysplastic syndrome (MDS), and 20 normal individuals by PRV-1–specific RT-PCR. In female patients, PRV-1 expression was correlated with clonality analysis as assessed by the human androgen receptor polymorphism assay.
RESULTS: PRV-1 was not expressed in granulocytes isolated from normal individuals or from patients with ST, SE, CML, IM, MDS, and inflammatory/infectious SL. On the contrary, all ET patients, 35 of 37 PV patients, and five patients with acute postsurgery and posttraumatic SL overexpressed PRV-1. All the cases with monoclonal hematopoiesis (17 of 21 with ET and 12 of 12 with PV) expressed PRV-1, yet PRV-1 overexpression extended also over the cases of ET showing polyclonal hematopoiesis (four of 20).
CONCLUSION: The overexpression of PRV-1 seems to be a useful tool for discriminating ET and PV from ST and SE, thus offering an innovative diagnostic approach on the basis of the detection of positive diagnostic criteria instead of exclusion criteria.
THE WORLD HEALTH Organization classification of hematologic malignancies includes chronic myeloproliferative diseases (MPDs) Ph1-positive chronic myelogenous leukemia (CML), chronic idiopathic myelofibrosis (IM), essential thrombocythemia (ET), and polycythemia vera (PV).1 Nevertheless, although the presence of the Ph1 chromosome or its molecular equivalent indubitably indicates the diagnosis of CML, in the case of IM, PV, and ET there as yet is no reliable molecular indicator of disease. In particular, in PV and ET, by far the most frequent MPDs, the diagnosis is reached only when other possible causes of secondary erythrocytosis or thrombocytosis have been ruled out.2,3
Recently, Temerinac et al4 cloned a novel urokinase-type plasminogen activator receptor superfamily member gene, named polycythemia rubra vera-1 (PRV-1), which was not detectable in granulocytes from patients with secondary erythrocytosis (SE), whereas it was overexpressed in patients with PV and in some patients with ET. The natural ligand of PRV-1 has not yet been identified. Nevertheless, the physiologic expression of PRV-1 in bone marrow precursors and its detection in granulocytes from normal individuals receiving granulocyte colony-stimulating factor suggest a link between cell proliferation and PRV-1 expression.4
In this study, we investigated the expression of the PRV-1 gene by using the PRV-1–specific reverse transcriptase polymerase chain reaction (RT-PCR) in 37 patients with ET and in 25 patients with secondary thrombocytosis (ST). Normal individuals, patients affected by secondary leukocytosis (SL), PV, SE, CML, IM, and myelodysplastic syndrome (MDS), were studied as the control group. Moreover, considering that both clonal and polyclonal hematopoiesis has been reported in ET,5-7 in females patients, we matched the PRV-1 findings with the results of clonality analysis obtained by the human androgen receptor (HUMARA) polymorphism assay.
PATIENTS AND METHODS
Patients
Thirty-seven patients with ET, 37 patients with PV, 25 patients with ST, 10 patients with SE, 25 patients with SL, five patients with CML, five patients with IM, and five patients with MDS were investigated. Fourteen patients with ET and seven patients with PV were studied at the time of diagnosis. The median follow-up of patients with a pre-existent diagnosis was 24 months (range, 4 to 96 months) for ET and 19.5 months (range, 2 to 120 months) for PV. In all cases, diagnosis of PV and ET was performed according to the criteria of the PV and ET study groups.2,3 SE included nine cases of chronic obstructive pulmonary disease and one case of polycystic renal disease. ST included six splenectomized patients, 13 patients with documented infections, and six patients with sideropenic anemia. SL included 12 patients with infectious disease, eight patients with chronic inflammatory disease, four cases of acute leukocytosis after coronary artery bypass graft surgery, and one case of polytrauma. In addition, 20 healthy blood donors were investigated. Clinical and laboratory features of PV and ET patients and of individuals enrolled as control groups (normal, SE, ST, and SL) are listed in Table 1.
Table 1. Characteristics of the Patients and Normal Individuals*
Sample Collection and Granulocyte and T-Lymphocyte Isolation
All samples were collected after informed consent. Fifteen milliliters of peripheral blood was collected in sodium citrate and diluted 1:2 with Ca2+- and Mg2+-free phosphate-buffered saline (PBS). In order to remove red cells, cell suspension was mixed to 6% hydroxyethyl starch solution (Plasmasteril; Fresenius AG, Friedberg, Germany) in a ratio of 1 mL of hydroxyethyl starch to 8 mL of cell suspension and then sedimented for 90 minutes at room temperature. All nucleated cells were then recovered, washed twice, suspended in PBS, and centrifuged over Ficoll-Hypaque (Pharmacia Biotech, Uppsala, Sweden) density gradient for 30 minutes. Cells at the bottom of the vial, consisting of over 95% granulocytes (as evaluated by morphologic examination after May-Grünwald staining), were recovered, washed twice in PBS, and then stored with and without TRIzol (Gibco, Invitrogen Corporation, Carlsbad, CA) at −80°C until PRV-1 and HUMARA assays. Mononuclear cell fraction was recovered as well, washed twice, and suspended in RPMI (Hyclone, Cramlingtone, United Kingdom) supplemented with 20% heat-inactivated fetal calf serum (Stem Cell Technologies, Vancouver, British Columbia, Canada) at 2 × 106 cells/mL. After an overnight incubation at 37°C in a 5% CO2-humidified atmosphere, nonadherent cells were recovered and centrifuged, and the dried pellet was stored at −80°C until HUMARA assay. These cells consisted of over 85% T lymphocytes (as evaluated by flow cytometry analysis after staining with anti-CD3 monoclonal antibody).
RT-PCR
RNA was extracted using TRIzol reagent (Gibco) following the manufacturer’s instructions. Total RNA was eluted in 50 μL of RNAse-free water, and treated with 1 μL of RQ1 RNase-free DNase (1 U/μL) (Promega, Madison, WI) at 37°C for 15 minutes, to eliminate DNA contamination. RNA was then extracted with phenol/chloroform/isoamyl alcohol, precipitated with ethanol, and then resuspended in diethylpyrocarbonate-treated H2O. First-strand cDNA was synthesized by incubating 0.5 to 1 μg of RNA at 42°C for 50 minutes in a final volume of 20 μL of RT buffer 1× (50 mmol/L Tris HCl [pH 8.3], 75 mmol/L KCl, 3 mmol/L MgCl2) containing 100 units of Moloney murine leukemia virus reverse transcriptase (Gibco), 2.5 μmol/L random hexamers (Gibco), 1 mmol/L dNTP mix (Promega), and 10 mmol/L dithiothreitol (Gibco). Three microliters of cDNA was amplified with specific primers for PRV-1 in 25 μL of final volume, containing 1 units of Taq DNA polymerase (Gibco), 1 mmol/L of each primer, 200 mmol/L of both dNTP and 1× reaction buffer (10 mmol/L Tris HCl [pH 8.3], 50 mmol/L KCl, 1.5 mmol/L MgCl2). The PCR conditions were as follows: one cycle of 3 minutes at 95°C, followed by 30 cycles at 95°C for 40 seconds, 54°C for 40 seconds, 72°C for 40 seconds, and a final cycle of 3 minutes at 72°C. Specific primers (sense 5′-CAG TTT GGG ACA GTT CAG C-3′; antisense 5′-AAA GCG GGA GGG AGT TAA C-3′) for PRV-1 amplified a 286-bp fragment, which was sequenced in order to confirm the specificity of the PCR product. The mixture was separated on a 2% agarose gel, and after staining with ethidium bromide, the PCR product was visualized under ultraviolet light. To assess the presence of cDNA, each sample was amplified with specific primers for beta-actin, according to the method previously described.8 Bone marrow mononuclear cells were used as positive control. In all cases resulting negative for PRV-1 expression, PCR products were transferred onto nylon membrane (Amersham, Arlington Heights, IL) and hybridized with a PRV-1 phosphorus-32 (32P)-labeled specific cDNA probe. After hybridization, the membranes were exposed to autoradiography overnight.
HUMARA Assay
The HUMARA polymorphism assay was carried out in granulocytes isolated from 41 female patients. T lymphocytes were used as control.5 One microgram of DNA, isolated using DNAzol and TRIzol reagents (Gibco), was incubated with and without 20 units of HpaII (Pharmacia Biotech) at 37°C for 12 hours in a final volume of 20 μL. After 10 minutes of incubation at 95°C, 3 μL of all samples were amplified for 35 cycles (30 seconds at 95°C, 30 seconds at 60°C, and 30 seconds at 72°C) in a total volume of 30 μL. The mixtures contained 1 units of Taq platinum DNA polymerase (Gibco), 1 mmol/L of each primer (sense 5′-GTC GTG AAG GTT GCT GTT CCT CAT-3′, antisense 5′-TCC AGA ATC TGT TCC AGA GCG TGC-3′), 200 mmol/L of each dNTP, 2 mmol/L MgCl2, 2% dimethyl sulfoxide, and 1× reaction buffer (10 mmol/L Tris HCl [pH 8.3], 50 mmol/L KCl). The products were electrophoresed on the NuSieve 3:1 agarose gel (BioWhittaker Molecular Applications, Rockland, ME), stained with ethidium bromide, and visualized under ultraviolet light. When the undigested PCR products were not clearly detectable as two well-separated bands, analysis was again carried out in the same PCR mixture containing 0.134 nCi α-32P-dCTP/μL (> 3.000 Ci/mmol; ICN Biomedicals, Inc, Aurora, OH) for 28 cycles (30 seconds at 95°C, 30 seconds at 60°C, and 30 seconds at 72°C), as described by Allen et al.9 Two microliters of the PCR products was mixed with 5 μL of sequencing gel-loading buffer and then electrophoresed on a denaturing 6 mol/L urea acrylamide gel. The gel was dried and exposed to x-ray film at −80°C for 12 hours with an intensifying screen. Autoradiographs were then blindly scored. In patients older than 60 years, only results indicating a polyclonal hematopoiesis were considered, in order to exclude the possibility of age-related skewing of X-chromosome inactivation.10 All the PRV-1 and HUMARA assays were independently carried out by two different investigators (M.M. and M.L.), blinded to diagnosis and to clinical and laboratory findings of the enrolled patients.
RESULTS
The expression of PRV-1, as evaluated by RT-PCR, in representative selected cases is shown in Figs 1 and 2⇓. The results obtained in normal individuals and in patients are listed in Table 2. The PRV-1 was not expressed in granulocytes isolated from normal individuals. Moreover, no positive sample was found among patients with ST or SE, or among the 20 patients with infectious/inflammatory SL. In all these cases, the lack of PRV-1 expression was confirmed by the hybridization of the PCR products with the PRV-1 32P-labeled specific cDNA probe. On the contrary, PRV-1 was expressed in granulocytes isolated from all patients with ET (Table 2). The PRV-1 expression was not related to the platelet count, the presence of splenomegaly, or the administered treatment (Tables 1 and 2⇓). All but one of the investigated patients had bone marrow biopsy showing the histologic picture typical of ET, with hypercellularity, megakaryocytic hyperplasia, and frequent megakaryocytic aggregates. Interestingly, one patient with normal bone marrow cellularity was also PRV-1–positive. Furthermore, PRV-1 was expressed on granulocytes isolated from 35 of 37 patients with PV. As for ET patients, this finding was not related to the hematocrit values, the presence of splenomegaly, or the administered treatment (Tables 1 and 2⇓). Two patients with PV were PRV-1–negative at repeated determinations. In these patients, the PV Study Group (PVSG) diagnostic criteria were fulfilled; nevertheless, the bone marrow histology was in contrast with the diagnosis of PV. In fact, bone marrow biopsy showed the coexistence of a chronic lymphoproliferative disease together with erythroid hyperplasia in one case, and suggested a myelodysplastic disorder in the other one. Interestingly, we found PRV-1 expression in all four patients with SL after coronary artery bypass grafting surgery and in the polytraumatized patient (Fig 2 and Table 2). In these patients, WBC and platelet counts and the hemoglobin and hematocrit values did not significantly differ from those observed in patients with SL caused by infectious and inflammatory disease. Nevertheless, in all cases of postsurgery and posttraumatic leukocytosis, the onset of neutrophilia was acute, whereas the other SL patients presented high WBC counts at repeated determinations before PRV-1 evaluation, suggesting a chronic behavior of leukocytosis. Finally, no PRV-1 expression was found in samples obtained from patients with CML, IM, or MDS.
Fig 1. PRV-1 mRNA in patients with ET (lanes 1-3), PV (lanes 4-6), ST (lanes 7-9), and SE (lanes 10-12), and in controls (lanes 13-15). Lanes 16 and 17, negative and positive controls. Beta-actin expression indicates cDNA integrity. (A) Ethidium bromide staining. (B) Hybridization with a PRV-1 32P-labeled probe.
Fig 2. (A) PRV-1 mRNA in patients with acute postsurgery SL (lanes 1-3), inflammatory/infectious SL (lanes 4-6), CML (lanes 7-9), IM (lanes 10-12), and MDS (lanes 13-15). Lanes 16 and 17 are negative and positive controls, respectively. MW, molecular weight. (B) Beta-actin expression indicates cDNA integrity.
Table 2. PRV-1 Expression in Patients and Control Subjects
Although to date no doubt exists about the clonality of PV, recent studies show that in some cases ET is a polyclonal disorder.5-7 We have matched the PRV-1 findings obtained in female patients with the clonality analysis of granulocytes; T lymphocytes were used as a control cell population. Forty-one female patients were investigated: eight with ST, 21 with ET, and 12 with PV. Two patients with ST and two patients with ET aged 70 and 63 years and 70 and 61 years, respectively, had monoclonality of both granulocytes and T lymphocytes. We considered this finding to probably be because of an age-related X-chromosome skewing,10 and these patients were considered not assessable. The results of clonality analysis obtained in the 37 assessable patients are listed in Table 3. We found that in all the PRV-1–negative patients with ST, no clonal cell population was detectable in either granulocyte or T-lymphocyte fractions. As expected, all 12 PRV-1–positive patients with PV had monoclonal granulocytes, and in 11 of them this finding was corroborated by the detection of polyclonal T lymphocytes. Clonality findings in patients with ET seemed more intriguing. In fact, in agreement with other studies,5-7 we found that of the 19 assessable patients, four young individuals (21%) aged 24, 24, 43, and 49 years, respectively, had monoclonality of both granulocytes and T lymphocytes; four patients (21%) had polyclonality of both granulocytes and T lymphocytes; and 11 patients (58%) had monoclonality of granulocytes and polyclonality of T lymphocytes. If we consider not assessable the six patients with monoclonal granulocytes and T lymphocytes, our data indicate that clonal hematopoiesis is unequivocally present in 11 (73%) of 15 patients with ET. However, despite these heterogeneous findings, PRV-1 was always expressed in patients with ET, thus appearing quite independent of the presence of monoclonal hematopoiesis.
Table 3. Clonality Analysis and PRV-1 Expression in Female Patients
DISCUSSION
In this study, we showed that granulocytes isolated from patients with ET and PV share the abnormal expression of a surface receptor, named PRV-1, and that this molecular marker can be used to promptly differentiate these diseases from SE and ST. In fact, other chronic MPD, MDS, and chronic leukocytosis related to inflammatory and infectious disease were found to be consistently PRV-1–negative. We found that the PRV-1 gene was overexpressed in granulocytes mobilized after an acute distress such as surgery or polytrauma, suggesting a functional role of this gene in such situations. PRV-1 belongs to the same family of urokinase-type plasminogen activator receptor, a molecule that is clearly involved in the regulation of adhesion and migration of polymorphonucleates.11 Interestingly, it has been shown that neutrophils mobilized after an acute stress differ in CD11b/CD18 antigen expression with respect to neutrophils isolated from patients with inflammatory/infectious leukocytosis,12 suggesting that acute and chronic leukocytoses are induced by different mechanisms. The finding of PRV-1 expression in conditions of acute distress does not affect the diagnostic utility of PRV-1 overexpression analysis, considering that these conditions are usually not associated with thrombocytosis. Inflammatory/infectious leukocytoses, on the contrary, are frequently associated with high platelet counts (ie, in our series, three patients with inflammatory/infectious SL had platelet counts > 600 × 109/L and 15 patients with ST had WBC counts > 10 × 109/L) in the absence of PRV-1 expression.
The PRV-1 gene has been recently cloned by Temerinac et al4 from granulocytes of patients with PV. They used the Northern blot analysis to investigate PRV-1 expression in 30 patients with MPD and found that all 19 patients with PV and two of the six patients with ET strongly expressed PRV-1, whereas four CML patients were PRV-1–negative. In our study, we investigated PRV-1 expression in chronic MPD by using RT-PCR, which represents a more sensitive method for detecting specific RNA than Northern blot. In our series, granulocytes isolated from all 37 patients with ET and from 35 of the 37 patients with PV were PRV-1–positive. In contrast, PRV-1 RNA was not detectable in granulocytes from normal individuals or from patients with CML, IM, MDS, ST, or SE. Importantly, in all these cases, the lack of PRV-1 expression observed with ethidium bromide staining was further confirmed by the hybridization with a PRV-1 32P-labeled specific cDNA probe, suggesting that a simple RT-PCR assay, not requiring the radioactive detection step, can be sufficient to determine PRV-1 expression. Two patients who fulfilled diagnostic criteria for PVSG for PV2 were PRV-1 negative at repeated determinations. Nevertheless, in both of them, the bone marrow histology was not consistent with the diagnosis of PV. Thus, these cases suggest the following for consideration: (1) PRV-1 expression can identify only typical forms of PV, as well as the bcr/abl rearrangement is not detectable in atypical CML;13 and (2) the stringent diagnostic criteria proposed by the PVSG can include cases of SE. In contrast, the detection of the overexpression of PRV-1 may allow identification of latent forms of PV and ET, not fulfilling conventional diagnostic criteria, as already reported for the endogenous erythroid colony assay.14
Since Fialkow15 published the study on the glucose-6-phospate dehydrogenase isoenzymes in various hematologic malignancies, MPDs have been considered clonal disorders. Although to date no doubt exists about the clonality of PV, recent studies show that in some cases ET is a polyclonal disorder.5-7 Accordingly, we found that four of 15 assessable patients with ET showed polyclonal hematopoiesis. PRV-1 was positive in all patients with PV and in all patients with ET, independent of the clonality of hematopoiesis, whereas it was not expressed in patients with ST, all of whom showed polyclonal hematopoiesis. Thus, PRV-1 overexpression, as opposed to the clonality analysis, efficiently identifies all cases of ET, both monoclonal and polyclonal.
Although the exact function of PRV-1 in normal hematopoiesis remains to be established, its overexpression in primitive but not in SE and ST could reflect the dysregulation of mechanisms controlling cell proliferation in the neoplastic cells. Recently, two groups have quantified PRV-1 expression using a real-time PCR assay with TaqMan technology.16,17 They confirmed that PV and ET patients express significantly higher amounts of PRV-1 than controls and that interferon alfa–treated PV patients display a PRV-1 expression comparable to normal controls,17 suggesting that PRV-1 expression could be a marker for evaluating the response to therapy. Whereas the method that we set-up to evaluate PRV-1 expression is not a quantitative assay, undoubtedly it seems to be a rapid, inexpensive, feasible test that does not require substantial expertise in interpreting the results. Therefore, this method could be widely applied as a first tool for screening, reserving the quantitative approach for the follow-up of positive patients. In conclusion, the overexpression of PRV-1 seems to be sensitive in discriminating ET and PV from ST and SE, thus offering an innovative tool for selecting patients requiring further careful diagnostic investigations specific for ET and PV.
Acknowledgments
Supported by grants from Ministero dell’Università e della Ricerca Scientifica e Technologica and by grant no. 1998-2000 from the Associazione Italiana per la Ricerca sul Cancro.
- Received November 21, 2001.
- Accepted June 20, 2002.
