In vivo Activity of the Cleaved Form of Soluble Urokinase Receptor: A New Hematopoietic Stem/Progenitor Cell Mobilizer

The urokinase-type plasminogen activator (uPA) is a serine protease that activates plasminogen to plasmin and binds to a specific high-affinity cell-surface receptor, uPAR (CD87). uPAR is formed by three homologous domains of ∼90 amino acids (D1, D2, and D3, as numbered from the NH₂ terminus) and is anchored to the cell membrane through a glycosyl-phosphatidylinositol (GPI) tail, attached to the COOH-terminal D3 domain (1). uPAR capability to transduce cell signals is likely due to its association with other cell-surface molecules, such as integrins and receptors for fMet-Leu-Phe (fMLP), a potent leukocyte chemoattractant (2). uPAR can be shed by the cell surface; soluble uPAR (suPAR) has been detected in plasma and urine from patients with various diseases (3). Both cell surface and soluble uPAR can be proteolytically cleaved in the D1-D2 linker region, thus generating truncated forms of GPI-uPAR (c-uPAR) and suPAR (c-suPAR), lacking D1. Cleaved uPAR forms can vary at the NH₂ terminus, according to the protease that cleaves the receptor and that may or may not disrupt a specific sequence, corresponding to amino acids 88-92 (SRSRY), which is involved in cell migration. In fact, cleaved suPAR or a uPAR-derived peptide containing the SRSRY sequence (uPAR_84-95) is able to activate fMLP receptors, thus inducing migration of monocytes, basophils, and hematopoietic stem cells (HSC; refs. 4–6). The human uPAR_84-95 peptide is not species specific because it exerts its chemotactic activity also on murine uPAR^−/− fibroblasts, 3T3, and LB6 cells and on rat aortic smooth muscle cells (7).

Cleaved forms of suPAR have been detected in biological fluids of individuals affected by breast and ovarian cancer, acute myeloid leukemia, and neurologic disorders (3); recently, a significant increase in SRSRY-containing c-suPAR has been observed in urine and serum from patients affected by colorectal and prostatic carcinomas as compared with healthy volunteers (8). However, a direct evidence of c-suPAR role in vivo has never been provided.

We have recently shown the involvement of c-suPAR in the mobilization of human CD34⁺ HSCs into peripheral blood (6). Mobilized peripheral blood HSCs have largely replaced bone marrow HSCs in transplantation procedures (9). Several agents are able to mobilize bone marrow HSCs; due to its potency and safety, the most common mobilizer is the granulocyte colony-stimulating factor (G-CSF; ref. 10). HSC mobilization involves β₁ and β₂ integrins, the stromal derived factor 1 (SDF1) chemokine, its receptor, CXCR4, and specific proteases; their relative role and importance have been only partially elucidated and are often controversial (11–14). Agents that down-regulate integrin activity or disrupt the SDF1-CXCR4 axis induce HSC release from bone marrow into the circulation. AMD3100, a pharmacologic antagonist of CXCR4, is an efficient HSC mobilizer both in humans and mice (15, 16).

We have recently reported that G-CSF administration to HSC healthy donors up-regulated uPAR expression in circulating myeloid precursors and monocytic cells and increased s-uPAR and c-suPAR levels in sera (6). c-suPAR and its derived peptide (uPAR_84-95) induced in vitro migration of bone marrow HSCs by activating the high-affinity fMLP receptor (FPR). The uPAR peptide also abolished in vitro migration of HSCs towards SDF1.

These in vitro findings prompted us to examine the role of c-suPAR in HSC mobilization in vivo. To this end, we first investigated whether the uPAR_84-95 chemotactic peptide was able to regulate in vitro migration of mouse M1 leukemia cells, which express the CD34 antigen and can be induced to differentiate by physiologic myelopoietic factors (17–19). The CD34 antigen is expressed also on murine hematopoietic stem and progenitor cells (HSC/HPC) capable of rapidly reconstituting and rescuing myeloablated mice (19–21). We then examined the in vivo effects of uPAR_84-95 administration on mouse HSC/HPC mobilization.

Reagents. Ninety-six-well microtiter plates were purchased from Costar (Cambridge, MA). Ficoll-Hypaque (specific gravity, 1,077), bovine serum albumin, and hydrocortisone sodium hemisuccinate were from Sigma Chemical Co. (St. Louis, MO). RPMI 1640 and heat-inactivated FCS were from Life Technologies (Gaithersburg, MD). Chemotaxis PVPF filters were purchased from Corning (Corning, NY) and the SDF1 from Peprotech (London, United Kingdom). Horseradish peroxidase–conjugated streptavidin, the biotinylation kit, and the enhanced chemiluminescence (ECL) detection kit were from Amersham International (Amersham, United Kingdom) and polyvinylidene fluoride (PVDF) filters were from Millipore (Windsor, MA). Red cell lysis buffer was from Becton Dickinson (Mountain View, CA). Phycoerythrin-conjugated antimouse CD34 monoclonal antibody was purchased from BD Biosciences PharMingen (Milan, Italy). Methylcellulose and the recombinant mouse growth factor cocktail were from Stem Cell Technologies (Vancouver, Canada). Recombinant human G-CSF (lenograstim) was purchased from Italfarmaco (Milan, Italy). The peptide uPAR_84-95 (AVTYSRSRYLEC) and its scrambled version (TLVEYYSRASCR) were prepared by PRIMM (Milan, Italy).

Cell cultures. Mouse M1 leukemia cells (17) were grown in RPMI 1640 supplemented with 10% heat-inactivated FCS.

Cell migration assay. Cell migration assays were done in Boyden chambers with 5-μm pore size PVDF polycarbonate uncoated filters. Cells (2 × 10⁵) were plated in the upper chamber in serum-free medium; 10⁻⁸ mol/L fMLP, various concentrations of uPAR-derived peptide (uPAR_84-95), 100 ng/mL SDF1, or serum-free medium was added in the lower chamber. Cells were allowed to migrate for 90 minutes at 37°C, 5% CO₂. Cells on the lower surface of the filter were then fixed in ethanol, stained with hematoxylin, and counted at 200× magnification (10 random fields per filter).

In a separate set of experiments, cells were preincubated for 30 minutes at 37°C with 10⁻⁴ mol/L fMLP, 10⁻⁷ mol/L uPAR_84-95, or buffer.

Animals and experimental design. BALB/c mice (Charles River Laboratory, Lecco, Italy), 8 weeks of age, were fed with commercial rodent chow and acidified water for some days before use. All the experiments were approved by the animal care committee of our institution. Mice were injected i.p. with the peptide uPAR_84-95 or its scrambled version (ScP) and with recombinant human G-CSF or buffer, in 200 μL PBS containing 0.1% albumin.

uPAR_84-95 or ScP was injected daily for 2 days at the indicated doses; G-CSF (0.25 mg/kg) or buffer was administered daily for 5 days. Four hours after the last injection, mice were killed by CO₂ asphyxiation and peripheral blood was obtained by cardiac puncture using a heparinized syringe. Each experimental group was composed of at least three mice.

Protein biotinylation. uPAR_84-95 (2 mg/mL) was biotinylated with a biotinylation kit (Amersham) according to the instructions of the manufacturer. Biotinylated peptide was quantitated with Micro BCA Reagent (Pierce, Rockford, IL).

Detection of circulating biotinylated uPAR_84-95 in vivo. To determine the levels of circulating biotinylated uPAR_84-95 in vivo, serum was obtained from four mice at various intervals after a single i.p. injection of 20 μg biotinylated uPAR_84-95. Triplicates of 1:10 mouse serum dilutions were analyzed by dot-blot onto a PVDF membrane. Biotinylated peptide was detected by hybridization with horseradish peroxidase–conjugated streptavidin and ECL. Dot-blots were analyzed by densitometric scanning and compared with dot-blots of various concentrations of biotinylated uPAR_84-95.

Flow cytometry analysis. Whole blood containing ∼1 × 10⁶ nucleated cells was washed once with red cell lysis buffer and then with PBS. Cells were stained with 20 μL of a phycoerythrin-conjugated anti-mouse CD34 monoclonal antibody for 20 minutes at 4°C and then analyzed by FACScan flow cytometer (Becton Dickinson). At least 5 × 10⁵ to 10 × 10⁵ nucleated cells were acquired for each sample using the CellQuest software (Becton Dickinson) and the total number of CD34⁺ cells was calculated on the basis of their percentage. Equivalent gating on isotype-matched negative controls was used for background subtraction in all assays.

Colony-forming unit assay. The total number of WBC in the peripheral blood was determined. Peripheral blood mononuclear cells (PBMC) were isolated by density gradient centrifugation and plated in methylcellulose in the presence of a recombinant mouse growth factor cocktail, as previously described, to enumerate peripheral blood progenitor cells (16). Colony forming units (CFU) granulocyte-macrophage (CFU-GM), CFU granulocyte erythroid macrophage, megakaryocyte (CFU-GEMM), and blast-forming unit-erythroid (BFU-E) were counted after 8-day incubation at 5% CO₂. Each experiment was done in triplicate.

Statistical analysis. Differences between groups were evaluated with the Student's t test. P ≤ 0.05 was considered statistically significant.

uPAR_84-95 stimulates migration of mouse CD34⁺ M1 cells and inactivates murine CXCR4. We have recently reported that the uPAR-derived peptide uPAR_84-95 stimulates migration of human CD34⁺ KG1 cells and CD34⁺ HSCs, by activating fMLP receptors, and inactivates CXCR4, the chemokine receptor primarily responsible for HSC retention in bone marrow (6, 22–25). We then investigated whether the same peptide is able to exert the same effects on mouse CD34⁺ M1 leukemia cells.

M1 cells were able to respond to the fMLP chemotactic stimulus, thus indicating that these cells express functional fMLP receptors (not shown). Therefore, fMLP receptors could potentially mediate uPAR_84-95-induced migration of CD34⁺ M1 cells, as previously shown in human CD34⁺ HSCs. In fact, uPAR_84-95 induced dose-dependent M1 cell migration (Fig. 1A); homologous desensitization with fMLP confirmed the involvement of fMLP receptors, as previously reported for human CD34⁺ HSCs (not shown).

Moreover, we showed that CD34⁺ M1 cells were able to migrate toward SDF1; preincubation with uPAR_84-95 almost abolished the SDF1-dependent chemotaxis, thus indicating that uPAR_84-95 regulates CXCR4 activity also in mouse cells (Fig. 1B).

uPAR_84-95 administration increases circulating WBCs and CD34⁺ HSCs/HPCs in mouse. In vitro migration experiments indicated that uPAR_84-95 induces migration of hematopoietic cells and regulates CXCR4 activity in mouse, suggesting that this molecule might induce release of mouse HSCs/HPCs from bone marrow and promote their migration into the circulation in vivo. Therefore, we analyzed the effect of various doses of uPAR_84-95 on mobilization of WBCs and HSCs/HPCs in mice.

Murine HSCs/HPCs express the CD34 antigen (19–21), which is largely expressed in all human HSCs; in mice, CD34 is present in HSCs that are capable of rapidly reconstituting and rescuing myeloablated mice (19–21). Thus, we analyzed CD34 expression on mobilized WBCs to identify circulating HSCs/HPCs.

uPAR_84-95 or its scrambled peptide was administered daily for 2 days at concentrations of 0.3, 1.5, 3, and 6 mg/kg. The peptide concentration range was compatible with the doses of a CXCR4 agonist peptide (26) and of a CXCR2 agonist previously used to induce HSC mobilization in mice (22). uPAR_84-95 caused rapid leukocytosis and induced a measurable increase in circulating CD34⁺ HSCs/HPCs, with a peak value at the 3 mg/kg dose (Fig. 2A and B, respectively). The increase in circulating CD34⁺ cells following uPAR_84-95 administration was similar to that obtained by administration of other agents (9). These results indicate that the uPAR-derived chemotactic peptide is able to induce dose-dependent mobilization of WBCs and CD34⁺ cells in mouse, the optimal dose being 3 mg/kg.

To determine the levels of circulating uPAR_84-95 after i.p. injection, mice were injected with biotinylated peptide; serum samples were collected at various intervals and analyzed by dot-blot. Dot-blots were analyzed by densitometric scanning and compared with dot-blots of biotinylated peptide at various concentrations. The peak of circulating biotinylated uPAR_84-95 was observed at 1 to 2 hours after peptide injection (Fig. 3) and corresponded to 1.9 and 2.1 μg/mL, respectively. These results were compatible with the observed effects of the uPAR-derived peptide on WBC and CD34⁺ HSC/HPC mobilization. However, a single stimulation with uPAR_84-95 induced only a very weak effect (not shown), thus suggesting that a “priming” is required to obtain an efficient mobilization.

uPAR_84-95 and G-CSF induce similar increases in circulating polymorphonuclear cell–enriched WBCs and HSCs/HPCs in mice. We then compared the HSC/HPC mobilizing capability of uPAR_84-95 with that of G-CSF. G-CSF was administered daily for 5 days, as previously reported (27). G-CSF-administration induced a 2.7-fold increase of circulating WBCs, as compared with control mice treated with buffer. Mobilized leukocytes were strongly enriched in polymorphonuclear cells (PMN; Table 1), in accordance with previous reports (22, 27). uPAR_84-95 was given daily for 2 days at a concentration of 3 mg/kg, the dose exerting the maximum effect in dose-response experiments (Fig. 2). Administration of the uPAR-derived peptide caused a 2-fold increase in circulating WBCs, as compared with control mice treated with the scrambled peptide. uPAR_84-95-mobilized leukocytes were enriched in PMNs, as observed in G-CSF-induced mobilization (Table 1).

We then analyzed circulating CD34⁺ HSCs/HPCs following uPAR_84-95 or G-CSF stimulation. uPAR_84-95 induced a 15-fold increase in circulating CD34⁺ HSCs/HPCs as compared with the scrambled peptide (27.4 ± 4.2/μL versus 1.8 ± 0.3/μL; P = 0.0004). Its potency was similar to that of G-CSF (25.5 ± 1.9/μL versus 1.3 ± 0.2/μL; P = 0.004; Fig. 4). In vitro colony-forming unit assays confirmed uPAR_84-95 capability to induce HSC/HPC mobilization. Indeed, significant increases in circulating levels of all types of colonies assayed were observed following uPAR_84-95 administration, as compared with the scrambled peptide injection (Fig. 5). The greatest relative increase occurred in the number of CFU-GMs (611.8 ± 266.2/mL versus 96.3 ± 31.0/mL; P = 0.04); this effect was somewhat lower than that induced by G-CSF (902 ± 189.2/mL versus 125.8 ± 32.4/mL; P = 0.026), but the difference was not statistically significant (611.8 ± 266.2/mL versus 902 ± 189.2/mL; P = 0.198).

Finally, uPAR_84-95 administration during G-CSF stimulations did not induce a further increase in circulating CFU-GM, BFU-E, and CFU-GEMM, thus indicating that uPAR_84-95 does not act sinergistically or additively with G-CSF (Fig. 5).

These results indicate that the uPAR_84-95 capability to induce mobilization of WBCs and HSCs/HPCs was similar to that of G-CSF; however, synergistic or additive effects were not observed.

uPAR is involved in cell migration, adhesion, and proliferation (2), functions that are mediated by integrins and members of the fMLP receptor family and that require the activation of G-proteins, Src-family tyrosine kinases, and extracellular signal–regulated kinases (28). A peptide epitope of uPAR, corresponding to amino acids 84-95 (AVTYSRSRYLEC) and located in the D1-D2 linking region, is endowed with potent chemotactic activity. Cleaved forms of suPAR, lacking D1 and containing this specific sequence, are unable to bind extracellular and cell-membrane uPAR ligands, such as uPA, vitronectin, and integrins (29–31); nevertheless, they acquire the capability to stimulate migration of fMLP receptor–expressing cells, even in the absence of GPI-linked uPAR (4, 7). Such suPAR fragments or the uPAR-derived peptide uPAR_84-95 can also regulate in vitro activity of chemokine receptors, such as RANTES, MCP1, and SDF1 receptors, in a fMLP receptor–dependent manner (6, 32).

Cleaved forms of suPAR lacking D1 have been detected in biological fluids of individuals affected by various diseases (3). Several proteases, such as metalloproteases, elastase, and uPA, can cleave uPAR in the D1-D2 linker region without affecting the chemotactic sequence. Thus, in vivo detectable c-suPAR could contain the chemotactic epitope and could represent an active molecule with precise and distinct functional roles rather than an intermediary product of uPAR metabolism. In fact, SRSRY-containing c-suPAR has been detected in human urine and serum; chemotactic c-suPAR levels from patients affected by colorectal and prostatic carcinomas are significantly higher than those from healthy volunteers (8).

We have recently found increased c-suPAR levels in sera of HSC donors, following G-CSF treatment (6). In vitro, the c-suPAR-derived peptide uPAR_84-95 induced migration of bone marrow HSCs and abolished SDF1-dependent human HSC migration. Both these effects induced by uPAR_84-95 could contribute to HSC release from bone marrow. To elucidate the role that such peptide might exert on HSC trafficking in vivo, we first investigated whether uPAR_84-95, which is not species specific, may induce the same effects in mice. Our results showed that uPAR_84-95 can regulate in vitro migration of mouse CD34⁺ leukemia cells; thus, it could potentially influence HSC retention in bone marrow, favoring their mobilization into the circulation. Indeed, uPAR_84-95 administration induced migration of mouse CD34⁺ HSCs/HPCs into the circulation to an extent similar to that observed by administering the widely used mobilizer agent G-CSF. uPAR_84-95-mobilized leukocytes were strongly enriched in PMNs as observed with other previously reported mobilizer agents (22, 27). The PMN increase has been proposed as an important step in HSC mobilization because PMNs produce proteases able to degrade SDF1 and/or matrix and adhesion molecules (9, 22). In vitro colony-forming cell assays, which are generally used to identify mobilized hematopoietic stem and progenitor cells (16), showed uPAR_84-95 capability to induce HSC/HPC mobilization into the circulation.

uPAR_84-95-dependent effects can be obtained after only two stimulations, instead of five stimulations required by G-CSF, probably because the specific peptide acts directly on HSCs/HPCs and not by activating secondary pathways, like G-CSF. Our findings show, for the first time, that chemotactic suPAR fragments, detected in human biological fluids, are able to regulate cell adhesion and migration also in vivo, thus suggesting that their increase in some pathologies, as in acute leukemia, might not be merely a side effect of increased uPAR expression.

Moreover, the finding that uPAR_84-95 exerts a HSC/HPC mobilizing activity similar to that of GCS-F suggests a potential use of the cleaved form of suPAR, or its derived chemotactic peptide, in the strategies to optimize HSC mobilization, for instance in G-CSF poor mobilizers. The possibility of using only the active peptide instead of the whole molecule could avoid side effects due to uPAR activities other than the chemotactic capability. Further in vivo studies are required to exclude toxic effects to verify the therapeutic potential and to evaluate possible clinical applications of this molecule.

Grant support: European Union Framework Program 6 (LSHC-CT-2003-503297, CANCERDEGRADOME), Ministero dell'Università e della Ricerca Scientifica e Tecnologica, and AIL-Salerno.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank S. Sequino for his technical assistance in handling mice.