Selective up-regulation of functional CXCR4 expression in erythroid cells by HIV-1 Tat protein

Abstract

CXCR4 is the high affinity receptor for the SDF-1α chemokine and represents the main coreceptor for HIV-1 T-tropic strains. The surface expression of CXCR4 was analysed in CD34⁺ haematopoietic progenitors, induced to differentiate along the erythroid or granulocytic lineages, in liquid cultures supplemented or not with HIV-1 Tat protein. At concentrations as low as 1–10 ng/ml, synthetic Tat protein significantly increased the surface expression of CXCR4 in erythroid but not in granulocytic cells. The Tat-mediated up-regulation of surface CXCR4 was accompanied by a concomitant increase of CXCR4 mRNA and total CXCR4 protein content in cells developing along the erythroid lineage after 6–10 days of culture. Moreover, addition of SDF-1α (200 ng/ml) induced a significant higher rate of apoptosis in Tat-treated erythroid cells in comparison with control cells. These results demonstrated for the first time a direct positive role in haematopoietic gene regulation of Tat protein, and suggest the possible involvement of Tat in HIV-1-induced anaemia.

Introduction

Beside showing an invariable loss of CD4⁺ T lymphocytes, patients infected by human immunodeficiency virus type 1 (HIV-1) frequently exhibit bone marrow abnormalities associated with peripheral blood cytopenias, such as anaemia, neutropenia and thrombocytopenia, occurring alone or in various combinations [1,2].

The discovery that some chemokine receptors act as coreceptors for HIV-1 env protein [3], represented a great advancement in our understanding of HIV-1 pathogenesis. CXCR4 and CCR5 are the most important HIV-1 envelope coreceptors [4]. In fact, all HIV-1 strains use either CXR4 (T cell strain or X4) or CCR5 (R5 strains) or both coreceptors (RX54). In this respect, it has been clearly established that both CD4 and CXCR4 are expressed on the surface of a subset on CD34⁺ cells [5–8], while the expression of CCR5 was extremely low or undetectable in these cells [6]. In spite of the expression of both CD4 and CXCR4, CD34⁺ haematopoietic progenitor cells are not prone to infection with either X4 or R5 strains of HIV-1 [2,9]. Consistently, a productive infection of haematopoietic progenitors does not appear to be a key pathogenetic element in AIDS-related cytopenias [2, 8, 10].

The pathogenesis of HIV-1-related cytopenias and CD34⁺ progenitor cell impairment is likely multifactorial, but several studies demonstrated that the induction of CD34⁺ haematopoietic cell death, triggered by the interaction between HIV-1 gp120 and the surface of haematopoietic cells represents a major mechanism of peripheral blood cytopenias [11]. It has also been shown that the interaction between HIV-1 virions and/or gp120 and the surface of haematopoietic progenitors elicits the induction of different signal transduction pathways leading to haematopoietic derangement with minimal or no productive infection of haematopoietic progenitors. The role of CXCR4 in the pathogenesis of peripheral blood cytopenias is not fully elucidated yet, even though it may to be related to envelope-mediated cell inhibition. Moreover, we and other authors have recently shown that the interaction between SDF-1α and CXCR4 affects the survival/maturation of haematopoietic cells belonging to the erythroid [12] or megakaryocytic [13] lineages.

In this context, it was demonstrated that HIV-1 Tat protein up-regulates CXCR4 expression on the surface of CD4 T lymphocytes [14]. Of note, Tat protein, which is essential for an efficient replication of HIV-1, can be actively released by HIV-1 infected cells affecting the proliferation and survival of infected and/or uninfected cells by autocrine/paracrine mechanisms [15]. It has been demonstrated by our and other groups that Tat protein interacts with a variety of surface receptors eliciting the activation of different signal transduction pathways that regulate a large array of cellular genes [16–19]. The contribution of extracellular Tat to the progression of viral infection is underlined by the ability of neutralizing anti-Tat antibodies to decrease the viral load both in vitro and in vivo [20,21]. Although several studies were performed on the interaction of HIV-1 and CD34⁺ cells, little is known on the interaction between haematopoietic cells and regulatory HIV-1 proteins which play a fundamental role in the AIDS pathogenesis.

The aim of this study was to investigate the interaction(s) between extracellular Tat and haematopoietic progenitor cells. In particular, the surface expression of CXCR4 coreceptor was monitored in cells induced to differentiate along the granulocytic and erythroid lineages in the presence or absence of Tat protein.

Materials and methods

Isolation of CD34⁺ progenitor cells from cord blood

For the isolation of primary CD34⁺ cells from 20 cord blood specimens, maternal informed consent was obtained according to the Helsinki declaration of 1975 and the St. Orsola General Hospital Ethical Committee guidelines. Mononuclear cells were isolated from cord blood samples from normal full-term newborn infants by Ficoll-Paque (d = 1·077 g/ml, Pharmacia, Uppsala, Sweden) and rinsed in RPMI 1640 (Gibco, Grand Island, NY, USA). The CD34⁺ progenitor cells were purified from mononuclear cells by means of Mini-MACS system (Miltenyi Biotech, Germany) able to separate CD34⁺ cells following manufacturers’ instructions. The purity rate of CD34⁺ cells was determined by flow cytometry using a monoclonal antibody (mAb) that recognizes a separate epitope of the CD34 molecule (HPCA-2; Becton-Dickinson, Lincoln Park, NJ, USA) directly labelled with fluorescein (FITC-mAb).

CD34⁺ cultures, differentiation and Tat treatment

Purified CD34⁺ cells were seeded in Xvivo-20 medium (BioWhittaker, Walkersville, MD, USA) containing nucleosides (10 mg/ml each), 0·5% bovine serum albumin (BSA), 100 nmol/l BSA-adsorbed cholesterol, 10 µg/ml insulin, 200 µg/ml iron-saturated transferrin and 10 nmol/l 2-β-mercaptoethanol (all purchased from Sigma, St Louis, MO, USA). Granulocytic cultures were obtained by seeding the cells (5 × 10⁴/ml) in the presence of stem cell factor (SCF; 50 ng/ml, Roche, Mannheim, Germany), interleukin-3 (rIL-3, 10 ng/ml, Genzyme, Cambridge, MA, USA) and granulocyte colony stimulating factor (G-CSF, 10 ng/ml Genzyme). Erythroid cultures were obtained by seeding the cells (5 × 10⁴/ml) in the presence of SCF (50 ng/ml), rIL-3 (1 ng/ml) and erythropoietin (EPO, 4 U/ml Roche). At specific time points (0, 2, 6, 10, 15 days), cells were counted and an aliquot was stained and analysed by flow cytometry or employed for slot-blot or RT-PCR technique as described below. Cell density was readjusted to 5 × 10⁴/ml by adding fresh medium and appropriate cytokine cocktail. Tat protein was added at different concentration (1–500 ng/ml; Tecnogen, Caserta, Italy) for 24 h, afterwards cultures were washed and replaced with fresh medium containing cytokines without Tat. As negative control, HIV-1 recombinant p24 (Intracell, Cambridge, MA, USA) was used. In some experiments, E and G cultures were treated with 200 ng/ml of SDF-1α protein (Pharmingen, La Jolla, CA, USA), which was added in culture at least 24 h after Tat withdrawal.

Phenotipic analysis of cell surface molecules

At specific days of liquid culture, the expression of glycophorin A or CD15, surface markers was evaluated by staining with anti-Glycophorin A (Pharmingen), anti-CD15 (Becton-Dickinson) monoclonal antibody (mAbs) labelled with isotiocyanate of fluorescein (FITC) whereas CXCR4 was evaluated through 12G5 anti-CXCR4 mAb (Pharmingen) revealed by phycoerythrin (PE)-conjugated goat antimouse IgG (DAKO, Carpinteria, NJ, USA). Aliquots of 2 × 10⁵ cells were stained with 5 µl of each mAb in 200 µl of PBS containing 1% of fetal calf serum (FCS; BioWhittaker) at 4°C for 30 min Irrelevant isotype-matched control mAbs were employed to assess the unspecific fluorescence. After staining procedures, samples were analysed by FACScan cytometer (Becton-Dickinson) using LYSIS II program (Becton-Dickinson). At least 10 000 events are collected for each sample.

Slot-blot protein analysis

The technique was performed as previously described [22]. Briefly, 5 × 10⁵ cells were washed in PBS and resuspended in SDS-buffer (2·3%SDS, 5% mercaptoethanol, 0·015 m TrisHCl, pH 6·8, 10% glycerol) and filtered on Nitrocellulose membrane (Amersham, Arlington, UK) by using the Slot-blot apparatus (Fermentas, Vilnius, Lithuania). The membrane was saturated by 3% BSA in 0·015 m of PBS (pH 7·4) for 30 min at room temperature and then with 3% gelatin-PBS for 30 min at room temperature. The latter blocking solution was removed and the membrane was treated with unlabelled anti-CXCR4 mAb (Pharmingen) diluted 1 : 50 with PBS for 4 h at room temperature. After several PBS washings, a second incubation with a peroxidase-labelled rabbit antimouse IgG labelled with peroxidase (1 : 100 in PBS; Dako) was performed for 2 h at room temperature. After several washes, chemiluminescent horseradish peroxidase substrate (ECL, Amersham) was added and the membrane was developed on autoradiography film (Kodak XOMAT, Eastman, New York, NY, USA). The loading of the samples in each slot-blot well was checked by parallel slot-blot stained with antitubulin mAb (Sigma) stained or by stripping and staining with antitubulin mAb (Sigma) of the same slot-blot membrane.

RT-PCR analysis of CXCR4 mRNA

Total RNA was extracted from samples by High pure total RNA kit (Roche, Germany) following the manufacturers’ instructions. A first semiquantitative approach was performed by classical RT-PCR. RT-PCR was carried out for 30 cycles (94°C for 1 minute, 60°C for 1 minute and 72°C for 1 minute) by using Stratagene RT-PCR kit (Stratagene, La Jolla, CA, USA) analysing total RNA serially diluted (5 fold for each dilution). CXCR4 specific employed oligonucleotides were 5′AGCTGTTGGCTGAAAA GGTGGTCTATG3′ (forward) and 5′GCGCTTCTGGTGG CCCTTGGAGTGTG3′ (reverse) for a 251-bp amplicon and the gliceraldeyde−3-phosphatase deydrogenase (GAPDH) specific oligonucleotides were 5′AGCAATGCCTCCTGCACCACC AAC3′ (forward) and 5′CCGGAGGGGCCATCCACAGCTC3′ (reverse). The amplified products were run in 2·5% agarose gel electrophoresis.

Apoptosis detection

Apoptotic cell death was evaluated by FACScan through FITC-conjugated Annexin V (FITC-Annexin V) staining (Roche). Erythroid cells were costained with FITC-Annexin V and antiglycophorin A mAb conjugated with PE (Pharmingen) whereas granulocytic cells were costained with FITC-Annexin V and CD15 mAb conjugated with PE (Pharmingen). The samples were analysed by flow cytometry as previously described [12].

Statistical analysis

Results were expressed as means ± standard deviations (SD) of three or more experiments performed in duplicate or triplicate. Statistical analysis was performed using two-tailed Student's t-test.

Results

Tat selectively up-regulates CXCR4 expression in cells of the erythroid lineage without affecting cell viability and maturation

In order to achieve highly purified haematopoietic progenitor cells, umbelical cordon blood CD34⁺ cells were enriched by means of Mini-MACS separation procedure. After each immunomagnetic separation, the purity of CD34⁺ cells as assessed by flow cytometric procedure was constantly higher than 95% (data not shown). Accordingly to previous studies underlying the central role of CXCR4 in the homing of haemopoietic progenitors (12, 23, 24), the average percentage of freshly isolated cord blood CD34⁺ cells coexpressing CXCR4 protein on their cell surface was 45% (range 20–72% in 20 samples analysed) (Fig. 1).

Fig. 1.

CXCR4 membrane surface expression on freshly purified CD34⁺ cells evaluated by flow cytometry. The samples were stained by irrelevant isotype matched mAb plus goat PE-conjugated polyclonal antimouse IgG antibody (a) or by anti-CXCR4 mAb followed by goat PE-conjugated polyclonal antimouse IgG Ab (b). A representative experiment is shown.

Highly purified CD34⁺ cells were grown in serum-free liquid suspension cultures supplemented with IL-3, SCF and lineage-specific growth factors, such as EPO or G-CSF, in order to allow differentiation towards the erythroid (E) or granulocytic (G) lineages, respectively. As expected (12,23) under culture conditions, CXCR4 surface expression rapidly dropped. The degree of maturation was analysed by flow cytometry analysing the expression of glycophorin A and CD15, which represent specific differentiation membrane markers of the E and G lineages, respectively (Fig. 2a,b). Addition in culture of Tat protein did not significantly affected the viability (data not shown) and the surface expression of glycophorin A or CD15 (Fig. 2a,b protein expression during all assayed time-points in E, and G cultures. Interestingly, however, in E cultures, Tat protein increased the number of CXCR4 expressing cells (P < 0·01) at both day 6 (21% versus 9%) and 10 (18% versus 8%) of culture (Figs 3a, 4). On the other hand, HIV-1 p24 protein, used as negative control, did not elicit any CXCR4 protein increase. Moreover, Tat did not induce any significant increase of CXCR4 protein in G cultures (Fig. 3b). Furthermore, the mean fluorescence intensity analysis of flow cytometry assay demonstrated that Tat treatment (10 ng/ml) determines higher mean fluorescence intensity in CXCR4 positive E culture cells (Fig. 5).

Fig. 2.

Phenotypic analysis of glycophorin A (a) or CD15 (b) differentiation marker expressing cells in Tat-treated (10 ng/ml) (▪) or untreated E (□) or untreated G () cultures, by means of flow cytometry procedure, at day 2, 6, 10, 15. Data are expressed as percentage of glycophorin A (a) or CD15 (b) positive cells and represent the means ± SD of four independent experiments performed in duplicate.

Fig. 3.

Effect of Tat on CXCR4 expression in E and G cultures. Phenotypic analysis of surface CXCR4 was performed through flow cytometry procedure in Tat-treated (10 ng/ml)(□), p24 treated (10 ng/ml) (▪) or (a)untreated E () and (b)untreated G () cultures at day 2, 6, 10, 15. Data are expressed as percentage of CXCR4 positive cells and represent the means ± SD of four independent experiments performed in duplicate.

Fig. 4.

Flow cytometry analysis of CXCR4 membrane surface expression in E cultures. (a) Negative control represented by E cultures stained by irrelevant isotype matched mAb plus goat PE-conjugated polyclonal antimouse IgG antibody, (b) Tat untreated E cultures stained by anti-CXCR4 mAb plus goat PE-conjugated polyclonal antimouse antibody, (c) Tat-treated E cultures stained by anti-CXCR4 mAb plus goat PE-conjugated polyclonal antimouse antibody. A representative experiment is shown at day 6.

Fig. 5.

CXCR4 membrane surface expression mean calculated as mean fluorescence intensity (MFI) in CXCR4 expressing cells. For each time point CXCR4 expression in Tat-treated (10 ng/ml, ▪) or Tat-untreated (□) samples is expressed as percentage of the control represented by HIV-1 p24 (10 ng/ml). Data are expressed as means ± SD of four independent experiments performed in duplicate.

Since previous studies showed that low (picomolar/nanomolar) and high (nanomolar/micromolar) concentrations of Tat elicit distinct biological effects on different cell types [19,25–28], in the next group of experiments we employed increasing Tat concentrations (1–500 ng/ml). As shown in Fig. 6, CXCR4 up-regulation was detectable at day 6 and 10 (P < 0·01) following the treatment with low concentration (1–10 ng/ml) of Tat while a higher concentration (100–500 ng/ml) was ineffective in order to achieve a significant increase of CXCR4 protein. Although we cannot formally exclude that the inability of high concentration of Tat to up-regulate CXCR4 surface expression in E cells was due to a masking effect (29), this possibility is unlikely because Tat was washed out at least 1 day before performing phenotypic analysis of surface CXCR4.

Fig. 6.

Dose–response for Tat-induced up-regulation of CXCR4 expression in E culture at day 6 and 10. Data are expressed as percentage of CXCR4 positive cells and represent the means ± SD of three independent experiments performed in duplicate. □ Epo, Epo +1ng/ml Tat, Epo + 10ng/ml Tat, ▪ Epo + 100ng/ml Tat, () Epo + 500ng/ml Tat, Epo + 10ng/ml p24.

Tat induces an increase of both CXCR4 mRNA and protein in E cultures

In the following experiments, we sought to investigate whether CXCR4 surface up-regulation was related to increase of CXCR4 mRNA and protein or to recruitment of cytoplasmatic CXCR4 on the cell surface. For this purpose, the synthesis of CXCR4 mRNA was analysed by RT-PCR technique in Tat-treated or untreated E cultures using serial dilutions of total RNA. Tat-treated E cultures showed a clear increase of CXCR4 mRNA at day 6 and at day 10 (Fig. 7a) in comparison with controls. These data were confirmed by quantitative Real-time PCR performed by LightCycler instrument (data not shown) suggesting an involvement of Tat protein in the transcriptional regulation of CXCR4 gene. The amount of CXCR4 protein was next examined by slot-blot in whole cell extracts obtained from E cultures at different time points. At both day 6 and day 10, an increase of CXCR4 protein level was noticed in Tat-treated E cultures with respect to control cultures (Fig. 7b). HIV-1 p24 protein did not increase the CXCR4 protein amount in all time points tested (data not shown).

Fig. 7.

Evaluation of CXCR4 mRNA and protein expression. (a)serially diluted (5 folds) cellular total RNA aliquots, achieved from Tat-treated or untreated E cultures, were amplified with CXCR4 or GAPDH primers and analysed on 2·5% agarose gel electrophoresis. A typical experiment, performed at day 6, is shown. (b) total CXCR4 protein content was evaluated of Tat-treated E cultures (Lane 1) and Tat-untreated E cultures (Lane 2) by slot-blot technique. As control, tubulin protein was assayed (Lane 3 and 4).

SDF-1α protein treatment elicits a significant increase of apoptotic cells in Tat-treated E cultures

Several reports demonstrated that the interaction between SDF-1α and CXCR4 receptor could affect the proliferation and survival of CD34⁺ progenitor cells. This regulation is mainly related to SDF-1α concentration [12, 30, 31]. Thus, in an effort to elucidate the biological significance of Tat-mediated CXCR4 up-regulation in erythroid cells, we next investigated the effects of SDF-1α treatment (200 ng/ml) on E cultures treated with Tat or p24 protein. The results, shown in Fig. 8a, demonstrate that SDF-1α treatment induces a significant increase, at day 6 (P < 0·01) and at day 10 (P < 0·05), of apoptotic cells in E cultures treated with Tat (10 ng/ml) in comparison with Tat-untreated E cultures as evaluated by the analysis of glycophorin A⁺/Annexin V⁺ cells through flow cytometric procedure. In contrast, SDF-1α did not significantly increase the percentage of apoptotic cell in Tat-treated G cultures in comparison with control G cultures (Fig. 8b). The treatment of E or G culture with p24 protein (10 ng/ml) did not affect significantly the cell survival (Fig. 8a,b).

Fig. 8.

Effect of SDF-1 α (200 ng/ml) on the percentage of glycophorin A⁺/Annexin V⁺ cells (a) or on the percentage of CD15⁺/Annexin V⁺ cells (b) at various time-points of p24-treated, Tat-treated or Tat-untreated EPO (a) or G-CSF (b) -differentiated liquid culture supplemented with or without SDF-1α. Data are expressed as means ± SD of three independent experiments performed in duplicate. □ Epo or G-CSF, () Epo or G-CSF + Tat, Epo or G-CSF + SDF-1α, ▪ Epo or G-CSF + Tat + SDF-1α, Epo or G-CSF + p24, Epo or G-CSF + p24 + SDF-1α.

Discussion

Tat protein regulates viral replication and viral transcript elongation during HIV-infection [32,33]. This protein is secreted from infected cells and can be taken up by infected or uninfected neighbouring cells by paracrine/autocrine mechanism [21,34–36] Tat is able either to activate or inhibit several cellular genes, thus influencing cell proliferation and survival and therefore affecting cell biology [25,26,27,28,31,37]. Recent observations demonstrated that Tat mediates CXCR4 positive regulation in CD4⁺ T primary cells, suggesting a more complex role of Tat in the regulation of HIV-1 infection [14,38]. In spite of large studies on Tat activity in CD4⁺ T cell, macrophages and neuronal cells, little is known on the interaction between Tat and haematopoietic progenitor cells. In our study, we investigated the CD34⁺ haematopoietic progenitors during their differentiation towards erythrocytic and granulocytic lineage in the presence of Tat, in order to elucidate the ability of Tat to regulate CXCR4 expression. Our data indicate that Tat treatment increases CXCR4 expression on E culture cell membrane whereas it is completely ineffective on G cultures. Moreover, we observed that the Tat-related CXCR4 increase is detectable in E cultures at day 6–10 of liquid culture, whereas, in early and in late differentiation time points, it was not noticed any significant variation of CXCR4 content. This Tat-related induction of CXCR4 protein, documented at 6–10 days, is Tat concentration dependent (1–10 ng/ml). In fact, as described in a large array of studies performed on different cellular models, as for this experimental context, the positive biological activity of Tat is strictly connected with picomolar/nanomolar concentrations while nanomolar/micromolar concentrations did not support CXCR4 activation under these experimental conditions. Several studies demonstrated that low concentrations of SDF-1α (<1 ng/ml) promote the proliferation of adult peripheral blood CD34⁺ [31] interacting with CXCR4 whereas higher concentrations of SDF-1 α inhibit the growth of haematopoietic cells and, in particular, of erythroid progenitors [12, 30, 31]. In this context, we noticed that the Tat-treated erythroid cells are sensitive to apoptosis induced by high concentration (200 ng/ml) of SDF-1α protein adding complexity to the biological activities of Tat and SDF-1α/CXCR4 interaction.

In this study, we demonstrated that Tat protein up-regulates the expression of CXCR4 protein in haematopoietic progenitors differentiating along the erythroid lineage, but not in cells differentiating along the granulocytic lineage. CXCR4 is a member of G-protein coupled seven transmembrane domain receptor family [39], which is expressed in many cell tissues such as brain, spleen, lung [40], and blood cells including T cells, B cells and monocytes. CXCR4 receptor was also been detected on a large subset of CD34⁺ haematopoietic cells, and its high-affinity ligand, SDF-1α, displays a potent chemoattractant activity on CD34⁺ cells [24]. Remarkably, CXCR4 plays a central role as a coreceptor for T tropic HIV-1 strain infection and its presence, associated to CD4, is pivotal for cell infection and therefore for HIV-1 spreading [4,41]. In this context, it is important to point out that HIV-1 infection is often associated to anaemia and/or other peripheral blood cytopenias in a consistent number of patients (40–80%) [1,2]. Strikingly, in spite of the simultaneous presence of CD4 and CXCR4 receptors on the surface of a significant subset of CD34⁺ cells, haematopoietic progenitors rarely show a productive infection either in vivo or in vitro [11,42–44]. On the other hand, a major mechanism of haematopoietic progenitor cell loss is represented by the induction of apoptosis following the interaction between envelope gp120 protein and the CD4/CXCR4 receptor complex [2, 10, 45, 46]. Interestingly, the significant apoptotic cell number increase, induced by SDF-1α in Tat-treated E cultures, indicates a possible biological role of the enhanced CXCR4 expression in E cultures. Therefore, the ability of Tat to up-regulate the surface expression of CXCR4 on the surface of E cells likely represents a pathogenetic mechanism able to explain the loss of erythroid progenitors in a subset of HIV-1 infected patients.

In conclusion, this study represents, at our knowledge, the first evidence of Tat protein ability to positively regulate gene expression in haematopoietic cells belonging to the erythroid lineage. They also suggests that Tat protein may be involved in the pathogenesis of anaemia and represents a guide towards future investigations into the biological consequences of chemokine receptor expression and regulatory viral protein activity on haematopoietic cells.