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 - BIOCHIMICA E BIOFISICA "FRANCESCO CEDRANGOLO" - ()

Research Unit Leader

Fulvio Della Ragione

Description

The oxygen-depending processes described in the Background section, mainly involving VHL, HIF1-alpha and PHD proteins, represent the mechanisms by which mammals respond to variation of the oxygen pressure in the blood. These pathways, which are not only modulated by hypoxia, control both important metabolic processes (such as glucose metabolism) and phenotypical events including angiogenesis, cell division cycle, differentiation and apoptosis.
Recently, we discovered some alterations of VHL which hampered HIF1-alpha degradation causing the accumulation of the transcription factor. The increase of HIF1-alpha, in turn, results in the build-up of Epo and VEGF gene transcription and the stimulation of erythropoiesis and polycythemia. This form of congenital polycythemia has been observed before, with a high frequency, only in Chuvashia, a Russian region. Recently, we demonstrated that the disease is particularly frequent in South Italy (in particular in Ischia, an island of the Naples harbour, ref. 46).
Therefore, the aim of this unit of the project will be to investigate the molecular and biochemical mechanisms which link the oxygen-sensing processes and hereditary polycythemias. The results will help to shed new light on the oxygen-regulated molecular processes and on its correlated diseases.

The submitted program has the following objectives:
Task 1) Development of biochemical approaches for investigating oxygen-response pathways.

Task 2) Analysis of hypoxia-responsive pathway in BFU-E and CFU-E cells from normal and polycythemic patients.

Task 3) Characterization of the mechanisms by which subjects with VHL mutations in an heterozigosity status show a polycythemic phenotype.
A critical point of the hereditary polycythemia due to VHL mutation is the occurrence of some heterozygotes affected by the mutation present the disease while normally the heterozygotes do not show the signs of the disease. In cellular models established from these subjects, we will investigate the activity of the major steps of the oxygen-sensing pathways.


Task 4) Identification of the mechanisms by which erythroid progenitors from subjects with congenital polycythemia are extremely responsive to normal Epo level.
A major question of hereditary polycythemia, exluding the subtype due to Epo receptor alteration, is the observation that the patient erythroid precursors respond to Epo with a proliferation rate remarkably higher than the normal counterpart. Such an observation is important since it might explain the polycythemia observed in patients with normal Epo. It is our goal to evaluate the occurrence of additional endogenous mechanism, including the production of cytokines and the constitutive activation of specific pathways, which might explain the phenomenon.



Experimental plan.
The experimental plan reported below is organized into tasks that roughly follow the goals of the project reported in the respective section.

Task 1) Development of biochemical approaches for investigating oxygen-response pathways.
Our preliminary investigations demonstrated that lymphoblastoid cell lines established from B-cells of subjects affected by polycythemia due to alterations of VHL protein, show an increased expression of VEGF. Conversely, the transcription of other well-known HFI-1? target genes, i.e. Epo, SDF-1, triose-phosphate isomerase, aldolase and GAPDH, does not appear modified. Intriguingly, this observation was confirmed by the analysis of expression of pheripheral reticulocytes (46).
These findings demonstrate that one of the methods most used in the study of O2-correlated pathways (i.e. the analysis of the expression of HIF1-alpha -target genes) is clearly insufficient for identifying alterations of the oxygen-dependent mechanisms. Therefore, one of the goals of this study will be to establish novel biochemical methodologies for evidencing constitutive modifications in the oxygen-sensing pathway.

Due to the high complexity of the pathway under investigation, we will approach the problem at two different levels including. the evaluation of HIF1-alpha activity and the identification of phenotypic-dependent HIF1-alpha target genes. It is important to state that the dependence from a phenotype represent a critical point in the identification of the pattern of genes to be investigated.

It is high probable that the investigation of the activity of HIF1-alpha represents the functional point on which converge the O2-responsive mechanisms. Thus, it will be important to standardize methodologies for evaluating HIF1-alpha level and function. Some of these methodologies have been preliminarly developed in our laboratory, namely a sensitive immunoblotting for characterizing the content of the protein and an ELISA approach for investigating the activity of the protein. In addition , other techniques, including an EMSA, will be useful to study the activity in the presence of other nuclear cofactors occurring in a specific phenotype (i.e. BFU-E and CFU-E).
Indeed, as described in details in the introductory section, HIF1-alpha ?works in vivo as a multiprotein complex which includes particularly CREB, histone acetylases and histone deaceylases. An additional mechanism of HIF1-alpha regulation might be its localization. Indeed, since the protein works by forming complexes at nuclear level and recognizing a specific consensus sequence, its localization appears extremely important. Thus, a otential mechanism of regulation might be its relocalization or its compartimentalization. We have developed consolidated biochemical methodologies for obtaining extremely purified preparations of cellular compartments ( ) whose purity is evaluated by specific markers. Additionaly, the use of immunochemical techniques and confocal microscope will be essential in unravel the problem of cellular localization.

Thus, in summary the estimation of HIF1-alpha??function, as the convergence point of all the oxygen-sensing pathway, is the pivotal target for an initial study.

The identification of a panel of genes/proteins, whose increase (or decrease) are either diagnostic for the modulation of the oxygen-depending pathway(s) or important in the development of pathologies might be a second point of this task. In other words, it is possible that small variations of HIF1-alpha??are not detectable by the methodologies described above while the characterization of several genes (an array of genes) responding to HIF1-alpha??might be more sensitive. Particularly, it will be important to identify the genes O2-modulated in the erythroid precursors.

In order to construct this array, we will incubate cells with different phenotypes, (including particularly BFU-E and CFU-E) with compounds inhibiting HIF1-alpha proline hydroxylation, i.e. colbalt chloride or desferoxamine. Indeed, both the compounds interfere PHD activities, the first displacing the iron ion, the second acting as iron chelating molecule. Then, we will analyze the cellular transcriptosome from treated or untreated cells by commercial expression array or, most intriguingly, by means new methodologies based on the use of real-time PCR and libraries of fluorescent short oligos. On the basis of these data, we will select the genes mostly regulated by HIF1-alpha in order to prepare a small array of genes (not more than 100) to investigate. Obviously, it will be important, in the context of polycythemias, to evluate the expression of these genes in the erythroid phenotype. The analysis of these genes will be performed by methods of real-time PCR recently developed which allow the contemporaneous study of several genes. The use of a quantitative PCR represents a significant advancement in the study of expression pattern since it allows to get consolidated findings and not (as in the expression array) data to be verified.



Task 2) Analysis of hypoxia-responsive pathway in BFU-E and CFU-E cells from normal and polycythemic patients.

In this part of the project, we will investigate the interplay between alterations of the hypoxia-responsive pathway and the development of polycythemia. A pivotal aspects of this study is the selection of the experimental model. Indeed, in several instances the use of unsuitable cellular systems might lead to confusing conclusions.

We have selected to employ in planned experiments: 1) EBV-immortalized B lymphoblasts from normal and patients affected by congenital polycythemia and 2) BFU-E and CFU-E cells from normal and patients with congenital polycythemia. The immortalization of B-lymphocytes and BFU-E and CFU-E purification and growth will be performed from the other two units of the project. In the case of lymphoblastoid cells, the lines will be growth also in our laboratory.

The studies will be mainly dedicated to the analysis of the function of the pathways regulated by O2 and not to the description of genetic alterations. Thus, we will investigate the HIF1-alpha activity by means the methodologies described at the point 1.

Whether we will not identify functional alterations, we will investigate the transcription of the genes modulated by HIF-1?. These investigations will be performed at mRNA and protein level (by immunoblotting) with the aim of quantitatively evaluating the findings obtained by real time PCR. When we will identify HIF1-alpha quantitative or qualitative alterations, a different unit of the project (Dr. Perrotta) will sequence VHL, HIF1-alpha, cullins, PHDs and other genes involved in the O2-dependent pathways.
Whether no alterations will be evidenced, we will characterize step by step the degradation process of HIF1-alpha.
As a first approach, we will add to the cells, inhibitors of proteasome activity (LLnL and lactacystin). This should result in an increase of HIF1-alpha content. On these samples, we will analyze, the amount of ubiquitinated HIF1-alpha by using a strategy which involves HIF1-alpha immunoprecipitation and analysis of IP with antibodies against ubiquitin.

These experiments should give to us data regarding the efficiency of the ubiquitination process and thus of the activity of VHL-E3 ubiquitin ligase. Additionally, we will study the ubiquitination in vivo of HIF1-alpha by transfecting the cells with vectors encoding for Ub-Myc. After immunoprecipitation by HIF1-alpha antibodies, the immnoprecipitate material will be analyzed by anti-Myc antibodies.
Moreover, by means of recombinant HIF1-alpha, we will evaluate the rate of the transcription factor degradation in vitro by employing extracts from control and polycythemic erythroid precursors.
The two approaches reported above (i.e. the use of proteasome inhibitors and the in vitro degradation) have been successfully employed in the laboratory of the proponent to investigate the degradation in vitro of p27Kip1. Indeed, the removal of p27Kip1 (a cyclin-dependent kinase inhibitor) occurs, as in the case of HIF1-alpha, by means of ubiquitination and proteasomal removal (46).
In the case we will identify a difference in the ubiquitination and/or in the degradation rate, we will characterize by immunoblotting the various components of the VHL-E3 complex. Whether we will evidence variations, we will study the molecular mechanisms of the alteration.
We will also investigate PHD and FIH activities by following the methods reported in literature.
The obtained data will clarify the alterations of the HIF1-alpha pathways which might cause congenital erythrocytosis. Their knowledge will furnish data on the regulation of this pivotal cellular pathway.


Task 3) Characterization of the mechanisms by which subjects with VHL mutations in an heterozigosity status show a polycythemic phenotype.

An intriguing aspect of the interplay between an altered hypoxia-sensing pathway and polycythemia is the occurrence of patients with a Chuvash VHL mutation in heterozygosity showing polycythemic phenotype. Conversely, the majority of heterozygotes with this VHL mutation are completely normal. Patients with this genetic condition have been previously identified by us.


In order to face this question, we will analyze BFU-E and CFU-E cultures and lymphoblastoid cells from the heterozygote subjects for the HIF1-alpha status both normal and polycythemic.
Firstly, we will evaluate whether in the polycythemic patients will occur an increase of the HIF1-alpha activity by means the methodologies before described. Preliminar data appear to confirm the hypothesys. Subsequently, we will investigate the HIF1-alpha ubiquitination, the capability of the extracts to degrade the protein, its localization and the activity of PHD and FIH.

The data obtained should be able to identify the cellular defect which should change a begnign condition, as the heterozigosity is, in a polycythemic condition which is conversely characterized by a negative prognosis.
It is however possible, that these investigations might give negative results. In other words, the alteration of VHL gene in heterozygosity might give identical activaty of
HIF1-alpha, both in the polycythemic and not polycythemic subjects.
Finally, in the absence of all these changes we will hypothesize the occurrence of mutator genes not already identified whose localization will be investigated by specific genetic approach like single nucleotide polymorphisms (SNPs).


Task 4) Identification of the mechanisms by which erythroid progenitors from subjects with congenital polycythemia are extremely responsive to normal Epo level.

An important and frequent feature of congenital polycythemia is that the erythroid precursors (including those obtained from the peripheral blood) form more and larger colonies than the normal counterpart. Moreover, these colonies respond remarkably well to the addition of exogenous Epo when compared to the control.
Several hypotheses might explain the observation.
1. A number of experiments correlated an altered regulation of cell cycle to the erythrocytosis. Thus, we will evaluate the major components of cell division cycle in BFU-E and CFU-E, from patients and control subjects, growth in a liquid medium. Our group has a well demonstrate expertise in cell division cycle studies, including the level of cyclin, cyclin-dependent kinases (cdks) and cdk inhibitors (cdki), phosphorylation of the retinoblastoma proteins and the evaluation of the cdk activity (47, 48).
2. An additional hypothesis is that the erythroid progenitors produce cytokines that by an autocrine or by an intracellular mechanism stimulate or facilitate their growth. This mechanism is probable (but not yet proved) in the case of VHL mutations where the BFU-E and CFU-E might produce VEGF and/or Epo. It is important to note that recently it has been demonstrated that erythroid precursors synthesize VEGF and that this cytokine acting, by an intracellular mechanism, induce the proliferation of these cells..
3. Finally, we will evaluate in BFU-E and CFU-E from controls and erythrocytotic patients the status of some biochemical pathways connected to the proliferation events, and particularly the status of the Ras->Raf->MAPK pathway. Also in this field, our group of research has a well demonstrated expertise (47, 48).