RICERCA ITALIANA     

Research program

Control mechanisms of erythropoiesis and congenital and familial polycythemias: role of oxygen-sensing pathways

University Co-ordinator

Seconda Università degli Studi di NAPOLI - PEDIATRIA - ()

Research Unit Leader

Silverio Perrotta

Description

Preliminary results
In order to enroll the patients, we used a diagnostic algorithm for polycythemia, based on serum Epo concentration in non-phlebotomized patients (1). A two-page anonymous questionnaire on diagnosis and treatment of congenital erythrocytosis (packed cell volume, PCV>0.50 in men and PCV> 0.48 in women) was mailed to AIEOP (Italian Association of Pediatric Haematology and Oncology) members (pediatric hematologists/oncologists) and to a number of SIE (Italian Society of Haematology) members (adult hematologists). This report represented the first and only nationwide survey on primary familial and congenital polycythemia. At present, we have identified 112 unselected, treated and untreated adult patients (75 men; 37 women). The patients were divided according to diagnostic algorithm previously reported. Forty-two patients with suspected Chuvash-like congenital polycythemia (high serum Epo and absence of secondary erythrocytosis) were then unrolled in the study. The median age of patients at diagnosis was 25 years (range: 1-39 years). Some patients had undergone sporadic or regular phlebotomy treatment.
Initially, twenty-two patients from 13 families with suspected Chuvash-like congenital polycythemia were analyzed (37). Twelve patients lived on the island of Ischia (Bay of Naples). Written informed consent for molecular genetic analysis, data analysis and publication was obtained from all participants.
The analysis of all the twelve patients (3 independent families) from Ischia surprisingly showed that all patients were homozygous for the VHL C598T mutation (37). We analyzed the mutation prevalence in the island, and observed a frequency higher than that observed in Chuvascia (9 heterozygotes over 64 healthy subjects, frequency of 0.0703). Importantly, when we investigated 60 healthy subjects from other Italian regions, we were totally unable to detect any VHL mutation, including the Chuvash one (37).
Subsequently, we investigated the haplotype of our patients and demonstrated that they had a pattern identical to the haplotype observed in all the previous cases with VHL C598T allele (30).
The data obtained in the novel VHL-dependent polycythemic cluster identified in the Ischia suggests that heterozygotes might have some selective advantages that favor the spread and maintenance of the mutated allele.
To address this issue, we employed a sensitive ELISA assay for detecting HIF-1alpha activity. By means of this method, we demonstrated that HIF-1alpha activity was 3-fold higher in homozygotes compared to controls (37). Surprisingly, activity was also statistically increased in heterozygous subjects. This finding represents the first demonstration that the heterozygotes present a clear biochemical difference with the respect of normal subjet.
Using real time PCR analysis, we also measured the expression of the EPO, VEGF, SDF-1 and TP1 genes, which are targets of activated HIF-1alpha. SDF-1 was chosen in view of the fact that it has recently been demonstrated to be modulated by HIF-1alpha. VEGF expression was increased in homozygotes (by more than 3-fold as assessed by gel scanning). In contrast, EPO and SDF-1 and TP1 gene transcription was identical in all samples.
The data on Epo expression are in disagreement with those already reported (20) which demonstrated a remarkable up-regulation of Epo gene transcription in lymphoblastoid cells from Chuvash patients. We cannot explain the difference but we can only hypothesize that lymphoblastoid cells do not normally express Epo but efficaciously transcribe VEGF. Thus, we believe that investigation on HIF-1alpha-dependent gene expression should take into careful consideration the cell phenotype investigated. In this view, the assay of HIF-1alpha activity might represent a more reliable methodology.
We also found three unrelated patients with the classical VHL598C>T mutation on one allele. The other allele did not show any mutation. In both the cases, the C598T mutation was inherited from the father. However, the fathers, as almost all the subjects heterozygous for the C598T mutation, did not show any clinical or laboratory signs of erythrocytosis. Importantly, up to now, only two independent cases of polycythemic patients heterozygous for the C598T mutation have been reported (6).
Subsequently, we carried out a series of analyses to look for other genetic aberrations(e.g., mutations, deletions and silencings) that could affect the apparently wild-type VHL allele. First, we sequenced the VHL gene promoter and did not find any mutations. Therefore, we analyzed the VHL transcripts (37). We retrotranscribed total RNA from EBV-transformed B-lymphoblasts of the two patients, and amplified the cDNA using primers localized in the 5’- and 3’-UTR of the VHL mRNA (37). This experiment revealed two full-length VHL transcripts with a similar efficiency.
Because the two heterozygote polycythemic patients have high Epo serum levels, it is conceivable that genetic alterations of components of the oxygen-sensing pathway other than VHL may contribute to erythrocytosis. Consequently, we looked for mutations in the HIF-1alpha; gene and in two genes (i.e. elongin B and elongin C) that encode other components of the E3 complex that ubiquitinates HIF-1alpha. No mutations were found. We concluded that the two polycythemic heterozygotes have unidentified alterations at other steps of the HIF-1alpha-related pathway.
The occurrence of the three heterozygotes affected by polycythemia suggested to us the possibility that in these cases the normal allele was silenced (at least in part) by methylation, and thus the altered allele was mostly expressed. In this case, it is obvious that the heterozygotes phenotypically behave as the homozygotes. However, in all these cases real time PCR evidenced that both the alleles (i.e. the normal and the mutated) were almost identically expressed, thus ruling out the silencing of the normal allele.

TASK 1. Characterization of patients with VHL-dependent congenital polycythemia in Ischia island.
As reported in the preliminary results section, although congenital polycythemias are relatively rare diseases, we recently evidenced a large cluster of Chuvash polycythemia in the Ischia Island. This represents the second identified cluster in the world and shows a frequency of heterozygotes extremely high (>0.07) similar to that observed in Chuvashia. The disease is due to an alteration of the von-Hippel Lindau gene that results in the production of a pVHL protein with a lower affinity for the HIF-1alpha transcription factor.
However, our patients in Ischia seem to present a clinical picture significatively different fron the Chuvash cluster, including the absence (or low incidence) of cerebral vascular events. Thus, an important aim of the project will be to accumulate additional cases and further clinical information on the Ischia patients. In addition, we plan to prepare a bank of EBV-immortalized lymphoblastoid cell bank from polycytemic patients. In this bank, we will include not only familial and congenital polycythemias of Chuvash type but all the italian cases which respond to the features reported in the flow chart previously described. A further goal of this part of the project will be the identification of genes which might be responsible for the different phenotype observed in Ischia cluster versus the Chuvashia one (putative mutator genes).
The task will be performed as follows:
1. Clinical informations will be obtained from the pediatricians, hematologists and family physicians of Ischia and/or from medical records, death certificates, relatives,and neighbours. We will organize a Meeting in Ischia with family physicians, physicians of Ospedale Anna Rizzoli (Lacco Ameno d'Ischia, Naples, Italy) and the Biologists of Hematology Laboratories. In order to avoid potentially counfounding factors in comparing the clinical findings of the two sites (i.e., Ischia and Chuvashia)we will utilize the same questionnaire employed for Chuvash population. We will record demographic and clinical characteristics and causes of death. In the case of living and collaborative patients, we will record their medical history, perform a physical examination, and collect venous blood samples.
2. We will extract DNA from peripheral blood leukocytes and identify the C->T mutation in the exon 3 of VHL (the mutation of Chuvash polycythemia) since it abolishes a Fnu4HI restriction endonuclease recognition site.
3. In order to get a more precise characterization of the polycythemic phenotype, we will determine the increased sensitivity of the erythroid progenitors (erythroid blast-forming units, BFU-Es) to Epo using an in vitro clonogenic assay. The assay will be performed in all identified VHL C598T homozygous patients.
4. We will prepare Epstein Barr Virus (EBV)-transformed lymphoblastoid cell lines of homozygotes and heterozygotes for VHLC598T mutation. In this way, we plan to prepare a large bank of italian polycythemic cell lines (including also familial non Chuvash erythrocytosis).
5. Total RNA from lymphoblastoid cell lines and cultured erythroid progenitors will be prepared and employed for investigating the expression of selected genes (by real time PCR) target of HIF1 as well as for analyzying the cell expression profile. In other words, we will use total RNA from control subjects, normal heterozygotes, polycythemic heterozygotes and polycythemic homozygotes, to identify genes that might represent the effectors responsible for the different pathological status.
6. To determine whether the eventually novel pVHL target genes are actually regulated by HIF1, we will induce hypoxia or add cobalt ion (an inhibitor of HIF1 hydroxylation) to lymphoblastoid cell line cultures and BFU-E and CFU-GM colonies. The expression of the selected genes will be evaluated by real time PCR.
7. We will compare the clinical data and expression of identified target genes in polycythemic Ischia population with the available data in Chuvash polycythemia.

TASK 2. Identification of patients with Chuvash-like congenital polycythemia from other region of Italy.
A second goal of the project will be the identification of Chuvash-like congenital erythrocytosis from the other Italian regions. Indeed, while we have observed that the polycythemia occurring in Ischia is due exclusively to classical VHL C598T mutation, the other cases we have preliminary observed in Italy (although the frequency is remarkably lower than in Ischia), appear due to new VHL mutations. Therefore, we are interested in analyzing the other VHL-related polycythemia cases in Italy, and also in preparing established EBV-immortalized cell line from these patients.
The task will be performed as follows:
1. We will sequence the full lenght VHL gene to search for mutations as routinely performed in the coordinator laboratory.
2. The identified VHLC598T homozygous patients will be processed according to the steps of Task 1 (including the characterization of expression profiles). In particular, we will evaluate if the clinical pattern and target gene expression are similar to patients of Ischia and/or of Chuvash population.
3. The new VHL identified mutations in polycythemic patients in double heterozygous state or single heterozygous state will be studied in more details. It is important to underline that preliminary results clearly show that the double heterozygous or single heterozygous patients suffer of a more severe eryhrocytosis than the VHLC598T homozygous patients. For this reason, in order to investigate the importance of the new identified mutations and the relationship between the alteration of the oxygen-sensing pathway and the gravity of clinical phenotype, we will send to the partner group pellets of immortalized peripheral lymphocytes (and/or erythroid progenitor colonies) from VHLC598T homozygous patients, double heterozygous subjects, and single heterozygous patients to evaluate the level and activity of HIF-1alpha protein. Theresults will help in the characterization of the relationship between HIF-1alpha status and clinical conditions.
The subsequent step will be the evaluation of the expression of activated HIF-1 target genes and the transcription profile in the usually employed cell models. Also in this case, various genetic conditions, including the Chuvash mutation and new VHL mutations, will be compared.
4. We will compare the clinical data and the expression of the identified target genes in these patients in order to putatively identify the molecular basis of their more severe disease. We will also try to evaluate the involvement of genes responsible for tumor predisposition in VHL syndrome in this particular subgroup of patients.

TASK 3. Polycythemic patients with elevated Epo without VHL mutations.
In addition to the congenital polycythemia due to VHL mutations, there are a significant percentage (in our statistics > 50%) of congenital polycythemias, with high (or a relatively high Epo), without any alteration to VHL gene. On the other hand,since these hereditary erythrocytosis have high erythropoietin, they are probably due to alterations to oxygen-sensing pathways. We have already available a significant number of these cases, which have been up to now characterized for clinical and biochemical parameters. Intriguingly, some of these erythrocytic patients show a serum Epo more than 50-fold higher than the normal content. On these patients, and other than we plan to enroll, we will investigate the possible occurrence of mutations in other genes involved in the oxygen-sensing pathways (including elongins B and C, CUL2, RBX1, PHD1-3, and FIH). Moreover, we will establish immortalized lymphoblastoid cells from these patients in order to compare their expression profile with those of VHL-dependent polycythemia.
Therefore, we will proceed as follows:
1. We will evaluate the clinical phenotype of the polycythemic subgroup without VHL mutations in the coding region, in order to establish if the clinical findings are similar(or not) to those of VHL mutated patients.
2. Initially, we will utilize the immortalized peripheral lymphocytes and erythroid progenitor colonies of these patients and will assay the HIF-1alpha transcription factor activity by ELISA method and immunoblotting (this study will be performed by the group of partner).
3. If HIF-1 activity is higher in these polycythemic patients than controls, we will investigate the occurrence of abnormalities that may affect the expression of the normal VHL allele or result in altered transcripts. For this reason, we will perform the sequence of the VHL gene promoter and analyze the VHL transcripts. We will retrotranscribe total RNA from EBV-transformed B lymphoblasts and amplified the cDNA using primers localized in the 5'- and 3'-UTR of the VHL mRNA.
4. Because these polycythemic patients have high Epo serum levels and increased HIF-1 activity, it is conceivable that genetic alterations of components of the oxygen-sensing pathway other than VHL could be the cause of erythrocytosis. Consequently, the coding regions of the elongins B (16p12.3, 3 exons) and C (8q21.11, 3exons), CUL2 (10p11.2-p11.1, 21 exons), RBX1 (22q13.2,5exons), PHD1(chr19q13.2,5exons), PHD2 (chr1q24.1, 5 exons) and PHD3 (chr14q13.1, 5 exons) will be screened for DNA sequence variation using a DHPLC-based protocol.
5. If we did not find any mutations, the subsequent step will be the analysis of the gene expression profiling and, then, quantitative real time PCR analysis, in order to identify other candidate genes.
The delineation of other components of the HIF pathway will provide a new list of candidate genes for mutations that would be expected to alter the poise of the entire oxygen sensing system, the consequences of such changes in oxygen sensing might include not only aberrations of hematocrit but also altered responses to altitude.