Angiogenesis has been postulated to be critical for the pathogenesis of multiple myeloma, a neoplastic disease characterized by abnormal proliferation of malignant plasma cells in the bone marrow (BM). Cleavage of the N- and C-terminal regions of circulating chromogranin A (CgA, CHGA), classically an antiangiogenic protein, can activate latent antiangiogenic and proangiogenic sites, respectively. In this study, we investigated the distribution of CgA-derived polypeptides in multiple myeloma patients and the subsequent implications for disease progression. We show that the ratio of pro/antiangiogenic forms of CgA is altered in multiple myeloma patients compared with healthy subjects and that this ratio is higher in BM plasma compared with peripheral plasma, suggesting enhanced local cleavage of the CgA C-terminal region. Enhanced cleavage correlated with increased VEGF and FGF2 BM plasma levels and BM microvascular density. Using the Vk*MYC mouse model of multiple myeloma, we further demonstrate that exogenously administered CgA was cleaved in favor of the proangiogenic form and was associated with increased microvessel density. Mechanistic studies revealed that multiple myeloma and proliferating endothelial cells can promote CgA C-terminal cleavage by activating the plasminogen activator/plasmin system. Moreover, cleaved and full-length forms could also counter balance the pro/antiangiogenic activity of each other in in vitro angiogenesis assays. These findings suggest that the CgA-angiogenic switch is activated in the BM of multiple myeloma patients and prompt further investigation of this CgA imbalance as a prognostic or therapeutic target. Cancer Res; 76(7); 1781–91. ©2016 AACR.

Multiple myeloma is a plasma cell malignancy characterized by abnormal clonal proliferation and accumulation of plasma cells in the bone marrow (BM). This results in a variety of clinical manifestations including anemia, osteolytic bone lesions, hypercalcemia, and renal failure. Most cases of myeloma also feature the production of a paraprotein (also called "M protein"), an abnormal immunoglobulin that can cause kidney problems (1). The cross-talk between myeloma and endothelial cells in the BM and the consequent activation of the angiogenesis process (i.e., the formation of new blood vessels from preexisting vessels) is critical for the pathogenesis of multiple myeloma (2–4). Accordingly, BM microvessel density increases parallel to disease progression, is an independent prognostic factor for survival, and correlates with established parameters of disease activity in patients (3–11).

Physiological and pathological angiogenesis are tightly regulated by the coordinated action of anti- and proangiogenic factors (12–14). Among the wide range of angiogenesis regulators so far discovered, recent studies have shown that chromogranin A (CgA) may have an important role in the regulation of angiogenesis (15, 16). CgA is a glycosylated, sulfated, and phosphorylated protein, 439 residue-long, stored in the secretory vesicles of many neuroendocrine cells and neurons (17), and exocytotically released in circulation together with the costored hormones, to reach 0.5 to 1 nmol/L levels in normal conditions (18). Tissue-specific intragranular and extracellular proteolytic processing of CgA leads to production of various bioactive peptides involved in the regulation of angiogenesis, metabolism, and cardiovascular system (18). Regarding angiogenesis, it has been recently shown that CgA contains: (i) a functional antiangiogenic site in the C-terminal region 410–439; (ii) a latent antiangiogenic site in the N-terminal region 1 to 76, and (iii) a latent proangiogenic site in the region 352–372 (15, 19, 20). These sites are activated by proteolytic cleavage of Q76-K77 and R373-R374 bonds, respectively. Accordingly, full-length CgA1-439 and the N-terminal fragment CgA1-76 (called vasostatin-1) inhibit angiogenesis in various angiogenesis assays, whereas the fragment CgA1-373 can stimulate angiogenesis (15). Mechanistic studies have shown that full-length CgA and the N-terminal fragment vasostatin-1 can inhibit endothelial cell migration, motility, sprouting, invasion, and capillary-like structure formation induced by VEGF, a potent proangiogenic factor, as well as the proangiogenic activity of basic fibroblast growth factor (FGF2), another important factor involved in the regulation of angiogenesis (15, 21). This fragment can also inhibit the nuclear translocation of hypoxia inducible factor (HIF)-1α, a master regulator of angiogenesis, in endothelial cells (22). However, the fragments CgA1-373 and CgA352-372 can induce the release of FGF2 from endothelial cells (15, 19). Hematological studies have shown that biologically relevant levels of full-length CgA1-439 and fragment CgA1-76 plus lower levels of other fragments lacking the C-terminal region are present in circulation in normal subjects. Blood coagulation causes, in a thrombin-dependent manner, conversion of CgA1-439 into the proangiogenic fragment CgA1-373 (15). Thus, CgA-related circulating polypeptides form a balance of anti- and proangiogenic factors tightly regulated by N- and C-terminal proteolysis.

Although abnormal levels of immunoreactive CgA have been detected in the blood of patients with neuroendocrine tumors and in several other pathological conditions, including renal failure, heart failure, rheumatoid arthritis, atrophic gastritis, inflammatory bowel disease, sepsis, or in subjects treated with proton pump inhibitors (PPI, a class of drugs commonly used to treat acid peptic disorders; refs. 18, 23, 24), little information is available regarding the levels of pro- and antiangiogenic CgA-derived polypeptides in the blood of cancer patients and no information at all regarding hematologic malignancies, including multiple myeloma.

To fill this gap, we investigated whether the balance of CgA-derived polypeptides (which may favor angiostasis in normal condition) is altered in multiple myeloma patients. To this aim we analyzed the extent of N- and C-terminal proteolytic processing of CgA in multiple myeloma patients, using plasma samples obtained from peripheral blood and BM, and in immunocompetent Vk*MYC transgenic mice, a model that has demonstrated high biologic fidelity to the human disease (25). We show that the balance of anti-/proangiogenic CgA polypeptides is tipped toward a proangiogenic state in the BM, with potentially important pathophysiologic implications.

Patients and plasma samples

Peripheral blood plasma samples and BM aspirates were obtained from 31 patients with multiple myeloma at diagnosis [active, n = 25; smoldering, n = 5; MGUS = 1; age 64.06 ± 12.7 years (mean ± SD)], after informed written consent and with ethical approval from the institutional review board. Patient's characteristics, disease staging, treatment with PPIs, and other therapies are reported in Supplementary Table S1. BM plasma samples were obtained by centrifugation of BM aspirates and cryopreserved in the gas phase of liquid nitrogen. Normal plasma samples were obtained from 25 normal donors [males/females, 13/12; age, 57.8 ± 13.9 years (mean ± SD)] that were not taking PPIs or other drugs.

Myeloma and endothelial cells

Human myeloma cell lines (KMS12-PE, KMS28-BM, U266), genotyped and tested for mycoplasma infection, were cultured in RPMI 1640 medium (Lonza) containing 10% fetal bovine serum. Human umbilical vein endothelial cells (HUVEC) were cultured in endothelial cell growth medium-2 (Lonza).

CgA fragments and antibodies

Full-length human CgA1-439, three fragments lacking the C-terminal region (CgA1-409, CgA1-400, CgA1-373), and one fragment corresponding to the N-terminal region (CgA1-76) were prepared by recombinant DNA technology as described (see Fig. 1A for a schematic representation of CgA; ref. 15). Monoclonal antibodies (mAb) B4E11 and 5A8, and polyclonal antisera α-439, α-410-439, α-373, and α-76 were described previously (15, 26, 27). The polyclonal antiserum α-373 was prepared by immunizing a rabbit with the CgA368-373 peptide coupled to keyhole-limpet hemocyanin. The antiserum, after pre-adsorption on a chromatographic column bearing full-length CgA1-439, recognizes the fragments CgA1-373, but not full-length CgA (Supplementary Fig. S1A). Full details on antibody preparation and epitopes can be found in Supplemental Methods and Supplementary Table S2.

Figure 1.

Sandwich ELISAs and levels of different CgA-related analytes in peripheral blood and BM plasma samples obtained from normal subjects and multiple myeloma patients. A, schematic representation of CgA and of its N-terminal, central, and C-terminal regions. B, schematic representation of the sandwich-ELISAs used in hematological studies: mAb B4E11 or 5A8 (capture antibodies); α-439, α-410-439, α-373, α-76, and α-FRs (detection antibodies). C, levels of different CgA-related analytes (*439, *439+436, *373, *76, and *439+436+FRs) as measured by ELISA in peripheral blood (PB) plasma obtained from normal subjects (n = 25) or paired peripheral blood and BM plasma from multiple myeloma (MM) patients (n = 31). CgA levels in BM plasma samples from patients untreated with PPIs and without renal failure (-PPI, -RF; n = 11) are also shown. The absolute values (without normalization) are shown. RF, renal failure. D, levels of *373 in peripheral blood (PB) and BM plasma of patients with levels higher than those found in normal subjects (n = 14). Lines show paired values. E, levels of analyte *373 relative to *439 or *76 (analyte ratios). Dots represent the analyte values for each patients; bars (mean ± SEM). Statistical analysis: Mann–Whitney test, except for the comparison of paired samples (paired peripheral blood and BM plasma values of multiple myeloma patients; Wilcoxon test). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Figure 1.

Sandwich ELISAs and levels of different CgA-related analytes in peripheral blood and BM plasma samples obtained from normal subjects and multiple myeloma patients. A, schematic representation of CgA and of its N-terminal, central, and C-terminal regions. B, schematic representation of the sandwich-ELISAs used in hematological studies: mAb B4E11 or 5A8 (capture antibodies); α-439, α-410-439, α-373, α-76, and α-FRs (detection antibodies). C, levels of different CgA-related analytes (*439, *439+436, *373, *76, and *439+436+FRs) as measured by ELISA in peripheral blood (PB) plasma obtained from normal subjects (n = 25) or paired peripheral blood and BM plasma from multiple myeloma (MM) patients (n = 31). CgA levels in BM plasma samples from patients untreated with PPIs and without renal failure (-PPI, -RF; n = 11) are also shown. The absolute values (without normalization) are shown. RF, renal failure. D, levels of *373 in peripheral blood (PB) and BM plasma of patients with levels higher than those found in normal subjects (n = 14). Lines show paired values. E, levels of analyte *373 relative to *439 or *76 (analyte ratios). Dots represent the analyte values for each patients; bars (mean ± SEM). Statistical analysis: Mann–Whitney test, except for the comparison of paired samples (paired peripheral blood and BM plasma values of multiple myeloma patients; Wilcoxon test). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Close modal

Mass spectrometry analysis of CgA fragments

Samples were desalted by reverse-phase chromatography using C18 ZipTip resin (Millipore). Mass spectrometry analysis was performed using an LTQ-Orbitrap XL mass spectrometer (Thermo Scientific) equipped with a nanoelectrospray ion source (Proxeon Biosystems). Full scan mass spectra were acquired in the LTQ Orbitrap mass spectrometer in the mass range m/z 700 to 1500 Da.

Immunoassays

Human full-length CgA and CgA fragments in plasma samples were detected using various sandwich ELISAs, called *439−, *439+436−, *76−, and *439+436+FRs−ELISA, described previously (15). These assays, based on the use of antibodies against different epitopes of CgA (see Supplementary Table S2), were calibrated using CgA1-439 (*439−, *439+436−, and *439+436+FRs−ELISA) or with CgA1-76 (*76−ELISA). These ELISAs can selectively detect: (i) N-terminal fragments cleaved after residue Q76 (*76−ELISA); (ii) full-length CgA (*439−ELISA); (iii) full-length CgA with or without the C-terminal tripeptide sequence 437-439 (*439+436−ELISA), and (iv) full-length CgA plus fragments containing the N-terminal and all or part of the central and C-terminal regions (*439+436+ FRs−ELISA) (15).

The fragment CgA1-373 was analyzed by a new sandwich-ELISA (called *373−ELISA) based on the use of mAb-B4E11 in the capture step, the antibody α-373 in the detection step, and CgA1-373 as a standard. Assay validation experiments showed that this ELISA selectively detects CgA1-373, but not CgA1-439, CgA1-409, CgA1-400, and CgA1-76 (Supplementary Fig. S1B). All these assays cannot detect murine CgA and CgA fragments. A schematic representation of the five ELISAs used in this study is shown in Fig. 1B. VEGF and FGF2 were analyzed using the ELISA Quantikine Kits (R&D Systems).

Rat aortic ring angiogenesis assay

The rat aortic ring (RAR) assay, based on measurement of the number of capillary-like structures spontaneously sprouting from rat aorta rings after 6 days in cell culture, was performed as described in Supplementary Methods. Basal angiogenesis, obtained without addition of FGF2 or VEGF, was examined in the absence or presence of various doses of CgA polypeptides.

In vivo studies in the Vk*MYC model

Studies on animal models (Vk*MYC mice) were approved by the Ethical Committee of the San Raffaele Scientific Institute and done according to the prescribed guidelines. Vk*MYC transgenic mice spontaneously develop clinically relevant multiple myeloma as a result of activation of the c-myc oncogene in maturing B cells under control of the kappa light chain (Vk) promoter (25). Disease development was monitored by measuring the levels of paraprotein (M-spike) in the serum, by serum protein electrophoresis, as previously described (28). Vk*MYC mice develop a monoclonal gammopathy starting at 30 weeks of age that progress slowly over time. Mice 55 to 80 weeks old were injected with 3 μg of recombinant human CgA1-439. Blood was collected after 1 and 24 h in heparinized tubes and immediately centrifuged (2000 × g, 15 min) using a refrigerated centrifuge. Plasma samples were analyzed by *439− and *439+436+FRs−ELISAs to assess human CgA fragmentation. Serum samples were also prepared in parallel for M-spike quantification.

Immunohistochemical analysis of BM microvascular density

BM microvessel density in Vk*MYC mice was analyzed by immunofluorescence microscopy as follows: bones were fixed with formalin, decalcified with Fix Decal (Pro-Eco), embedded in paraffin, and sectioned (4 μm). Vessels were stained by incubation with anti-CD31 antibody (1:200, overnight; Neomarkers) followed by biotinylated secondary antibody and avidin-horseradish peroxidase (BioCare). Slides were then incubated for 5 minutes with the 3,3′-diaminobenzidine (DAB) and, after washing, with Mayer-Hematoxylin (BioOptica). All sections were blindly evaluated by an expert hematopathologist.

Analysis of BM microvessel density in multiple myeloma patients and, as a control, in uninvolved BM biopsies from patients with nodal diffuse large B-cell lymphoma, was performed using formalin-fixed, paraffin embedded, and decalcified sections (3 μm thick) of BM biopsies stained with an anti-CD34 antibody (Novocastra, diluted 1:200), after antigen retrieval with a Tris-EDTA buffer at pH 9.

Full-length CgA and various fragments are present in the peripheral blood of normal subjects and multiple myeloma patients (at diagnosis) in variable amounts

To assess the extent of proteolytic processing of circulating CgA in normal subjects and multiple myeloma patients, we have analyzed samples of peripheral blood plasma with five different sandwich ELISAs specific for various CgA-derived polypeptides (see Supplementary Table S2 for antibody and assay description, analyte sequences, and codes, and Fig. 1A and B for a schematic representation of CgA, antibody epitopes, and assays).

The results showed that various CgA-derived polypeptides are present in the peripheral blood of normal subjects, including: (i) full-length CgA1-439 (called *439), (ii) fragments lacking residues 437–439 (CgA1-436); (iii) extremely low amounts of CgA1-373 (called *373); (iv) CgA1-76 (called *76); and (v) fragments lacking the C-terminal region 410–439 but containing central region epitopes (FRs; Fig. 1C, first bars).

Analysis of peripheral blood plasma from multiple myeloma patients at diagnosis revealed that the proangiogenic fragment *373, but not the antiangiogenic *439 and *76 polypeptides, were increased (Fig. 1C, second bars). As a consequence, the *373/*439 and *373/*76 ratios (i.e., the ratio of pro-/antiangiogenic forms) were significantly increased in the peripheral blood of patients compared to normal subjects (Fig. 1E, first and second bars). These data point to increased cleavage of CgA C-terminal region in multiple myeloma patients, particularly at residue R373.

Full-length CgA is cleaved at R373 in the BM of multiple myeloma patients

To assess whether cleavage of CgA at R373 occurred within secretory cells, during its transport in the blood, or within the BM, we then analyzed plasma samples obtained from the BM of multiple myeloma patients by ELISA. Interestingly, we observed that BM plasma contained even lower levels of *439 and higher levels of *373, compared to peripheral blood plasma (Fig. 1C, third bars, and Fig. 1D). These data suggest that cleavage at R373 occurred in the BM. In contrast, no changes of *76 was observed. Thus, the *373/*439 and *373/*76 ratios were further increased in the BM of patients (Fig. 1E, third bars), suggesting that the balance of pro-/antiangiogenic CgA polypeptides was tipped toward a proangiogenic state in the BM.

Treatment with PPIs and renal failure are major causes of increased secretion of CgA in multiple myeloma patients

The cause and the source of the increased CgA levels, observed in certain multiple myeloma patients, were then investigated. No CgA was detected in the supernatant of various cultured myeloma cell lines (data not shown), arguing against a role of myeloma cells as a source of CgA. More likely, CgA was released in circulation by the neuroendocrine system. Considering that a large fraction of our study population were taking PPIs or had renal failure, two conditions known to enhance the circulating levels of CgA, we evaluated each analyte in the BM of patients who were not taking PPIs and had no renal failure. Although these patients had normal levels of total CgA (*439+436+FRs; Fig. 1C bottom, compare the last bar with the first bar), they had low levels of *439 and high levels of *373 (Fig. 1C) and, consequently, markedly elevated *373/*439 and *373/*76 ratios (Fig. 1D, last bars). These data strongly suggest that the CgA normally released in circulation by the neuroendocrine system was proteolytically processed at residue R373 in the BM of multiple myeloma patients.

To assess the specific effects of PPI and renal failure on CgA levels in peripheral blood and BM plasma, we then stratified the patients according to these variables. As expected, renal failure tended to increase all variables in peripheral blood and BM (Supplementary Fig. S2A). PPI treatment was associated with increased levels of *439 and other fragments, but not of *76 (Supplementary Fig. S2B). These data and the results of previous studies showing that renal failure and PPI treatment are associated with increased CgA levels (29–32) suggest that both these factors were important causes of the increased circulating levels of CgA in multiple myeloma patients. However, considering that renal failure can be associated with more advanced disease (as also the need of PPI treatment), we cannot exclude a direct role of the disease itself. Furthermore, as PPI-treated patients were also taking steroids and/or other drugs (Supplementary Table S1), we cannot exclude that also other pharmacological treatments could have contributed to the increased CgA in those patients.

The associations of CgA and fragments with disease parameters and stage were also analyzed. *76 or *439+436+FRs negatively correlated with M-spike, whereas *439 positively correlated with creatinine (Supplementary Table S3). No significant association was observed between each analyte and the International or Durie and Salmon Staging Systems (ISS and DSS) for all patients (Supplementary Fig. S3A). However, we observed a higher *373/*439 ratio in three out of six PPI-untreated patients at DSS III (Supplementary Fig. S3B, right).

The cleavage of CgA C-terminal region is associated with FGF2 and VEGF levels and increased microvessel density in the BM of multiple myeloma patients

We then investigated whether CgA cleavage in the BM of multiple myeloma patients for which BM biopsis were available was associated with established parameters of angiogenesis activation. Interestingly, the enhanced ratio of pro-/antiangiogenic forms (e.g., *373/*439) in BM plasma was associated with enhanced production of VEGF and FGF2 in the BM, two potent proangiogenic cytokines (Fig. 2A and B and Supplementary Table S4). Of note, *373/*439, VEGF, and FGF2 tended to be higher in patients with advanced disease stage (Fig. 2A and B). No significant correlation was observed between VEGF and FGF2 (r = 0.412, P = 0.057).

Figure 2.

The proteolytic processing of CgA is associated with increased levels of VEGF and FGF2 in the BM of patients with multiple myeloma. Levels of VEGF (A) and FGF2 (B) in the BM of multiple myeloma patients at diagnosis with low (<2, n = 8) or high (>2, n = 9) *373/*439 ratio (an index of CgA fragmentation). VEGF, FGF2, *373, and *439 analytes were measured by ELISA (box-plots with median, interquartile and 5–95 percentile values). ***, P < 0.001 (Mann–Whitney test).

Figure 2.

The proteolytic processing of CgA is associated with increased levels of VEGF and FGF2 in the BM of patients with multiple myeloma. Levels of VEGF (A) and FGF2 (B) in the BM of multiple myeloma patients at diagnosis with low (<2, n = 8) or high (>2, n = 9) *373/*439 ratio (an index of CgA fragmentation). VEGF, FGF2, *373, and *439 analytes were measured by ELISA (box-plots with median, interquartile and 5–95 percentile values). ***, P < 0.001 (Mann–Whitney test).

Close modal

Furthermore, the sum of standardized values of the ratios between FRs (which includes *373 and other fragments lacking the C-terminal region, like CgA1-372, potentially proangiogenic) and *439 or *76 (antiangiogenic) significantly correlated with the microvessel density (MVD) in patients (Fig. 3A and B). No statistically significant values were obtained with single ratios or with single variables, suggesting that the overall balance of pro- and antiangiogenic CgA polypeptides is more important than single factors. These data indicate that proteolytic cleavage of CgA C-terminal region and the consequent change in the relative levels of pro-/antiangiogenic forms of CgA was associated with angiogenesis in the BM of patients.

Figure 3.

The cleavage of CgA C-terminal region is associated with increased microvessel density in the BM of multiple myeloma patients. A, BM microvessel density in multiple myeloma (MM; black bars) and controls (i.e., uninvolved BM biopsies taken at staging in patients with nodal diffuse large B-cell lymphoma; white bars). BM microvessel density was analyzed by immunohistochemistry after staining with an anti-CD34 antibody (bars, mean ± SD). B, linear regression analysis and correlation (r = 0.478; P = 0.045, Pearson test) between microvessel density and the sum of the standardized values (Z) of the ratios between FRs and *76 or *439. Z-score of FRs/76 and FRs/439 was calculated as Z = (Xm)/s, where X is the value, m the mean, and s is the standard deviation of population (n = 18 multiple myeloma patients).

Figure 3.

The cleavage of CgA C-terminal region is associated with increased microvessel density in the BM of multiple myeloma patients. A, BM microvessel density in multiple myeloma (MM; black bars) and controls (i.e., uninvolved BM biopsies taken at staging in patients with nodal diffuse large B-cell lymphoma; white bars). BM microvessel density was analyzed by immunohistochemistry after staining with an anti-CD34 antibody (bars, mean ± SD). B, linear regression analysis and correlation (r = 0.478; P = 0.045, Pearson test) between microvessel density and the sum of the standardized values (Z) of the ratios between FRs and *76 or *439. Z-score of FRs/76 and FRs/439 was calculated as Z = (Xm)/s, where X is the value, m the mean, and s is the standard deviation of population (n = 18 multiple myeloma patients).

Close modal

In contrast, no significant correlation was observed between MVD and VEGF or FGF2 (r = 0.176, P = 0.484; r = 0.135, P = 0.595, respectively).

The anti- and proangiogenic forms of CgA can regulate each other in the RAR angiogenesis assay

To assess whether CgA cleavage and changes in the relative levels of pro-/antiangiogenic forms of CgA may regulate angiogenesis, we analyzed the effect of full-length CgA and CgA1-373, alone and in combination, on the spontaneous capillary sprouting from RARs cultured in collagen gels, a widely used angiogenesis assay. In a previous study we showed that 0.2 to 1 nmol/L CgA1-439 can inhibit the proangiogenic effects of FGF2 and VEGF, and basal angiogenesis, whereas CgA1-373 can promote spontaneous angiogenesis at 1 to 9 nmol/L concentrations (15). As expected, CgA1-373, alone (3 nmol/L), significantly enhanced the number of capillary-like structures spontaneously outgrowing from rat aorta rings, whereas 1 nmol/L CgA1-439 decreased the number of capillary-like structures (Fig. 4A). When we added both components to the cultures, we observed neutralization of each other. Similar experiments were performed with CgA1-76, a fragment previously shown to inhibit the angiogenic effects of FGF2 and VEGF (15). Although this fragment could not inhibit the spontaneous sprouting of capillaries from aortic rings, 1 nmol/L CgA1-76 could neutralize CgA1-373 (Fig. 4B). These data suggest that changes in the relative levels of anti-/proangiogenic CgA fragments, as observed in the BM of patients, can regulate the anti-/proangiogenic activity of each other and, consequently, can contribute to promote angiogenesis in the BM of patients. In particular, the increase of CgA1-373 and the concomitant decrease of CgA1-439 in the BM of patients may tip the balance toward a pro-angiogenis state in the BM microenvironment. However, the inhibitory activity of CgA1-76 against CgA1-373, FGF2 and VEGF suggests that this fragment, which is produced also in physiological conditions, may serve to buffer the proangiogenic effects of these agents at a systemic level.

Figure 4.

Effect of CgA1-439, CgA1-373, and CgA1-76 (alone or in combination) on angiogenesis in the RAR assay. A and B, RARs were incubated with the indicated CgA fragments. Open circles correspond to number of capillaries sprouting from each aortic ring at day 6. Box plots with median, interquartile, and 5–95 percentile values are also shown. The number of aorta rings tested is also indicated (n). *, P < 0.05; **, P < 0.01; ****, P < 0.0001 (Mann–Whitney test).

Figure 4.

Effect of CgA1-439, CgA1-373, and CgA1-76 (alone or in combination) on angiogenesis in the RAR assay. A and B, RARs were incubated with the indicated CgA fragments. Open circles correspond to number of capillaries sprouting from each aortic ring at day 6. Box plots with median, interquartile, and 5–95 percentile values are also shown. The number of aorta rings tested is also indicated (n). *, P < 0.05; **, P < 0.01; ****, P < 0.0001 (Mann–Whitney test).

Close modal

Human CgA systemically administered to transgenic Vk*MYC mice is proteolytically processed at the C-terminal region

To provide further experimental evidence that multiple myeloma is associated with enhanced cleavage of CgA and angiogenesis, we performed a study using Vk*MYC transgenic mice, a clinically relevant model of multiple myeloma. Notably, in this model, mice having an M-spike >12% of total serum proteins, that is mice with more advanced disease (25), had a BM MVD two-fold greater than those with M-spike <12% (Fig. 5A), suggesting that angiogenesis was active in these mice.

Figure 5.

BM microvessel density, M-spike, and proteolytic processing of exogenous CgA in transgenic Vk*MYC mice. A, BM microvessel density in 55- to 80-week-old Vk*MYC mice (n = 10) with M-spike < or >12% of total serum proteins. BM microvessel density was analyzed by immunofluorescence microscopy after staining with anti-CD31 antibodies. B, proteolytic cleavage of exogenous CgA in Vk*MYC mice (n = 39) with M-spike < or >12% of total serum proteins, as evaluated by measuring *373 and *439 by ELISA 1 h after administration (i.v.) of 3 μg of human CgA1-439. The *373/*439 ratio is shown. C, proteolytic cleavage of exogenous CgA in normal mice (n = 5) and Vk*MYC mice (n = 29) with different levels of M-spike. Plasma samples were collected from each mouse 24 h after administration of 3 μg of human CgA1-439 and analyzed by *439- and *439+436+FRs-ELISA. The correlation between *439/*439+436+FRs ratio (an index of C-terminal fragmentation) and M-spike is shown. Bars, mean ± SEM; *, P < 0.05; **, P < 0.01; ***P < 0.001 (t-test, two tails).

Figure 5.

BM microvessel density, M-spike, and proteolytic processing of exogenous CgA in transgenic Vk*MYC mice. A, BM microvessel density in 55- to 80-week-old Vk*MYC mice (n = 10) with M-spike < or >12% of total serum proteins. BM microvessel density was analyzed by immunofluorescence microscopy after staining with anti-CD31 antibodies. B, proteolytic cleavage of exogenous CgA in Vk*MYC mice (n = 39) with M-spike < or >12% of total serum proteins, as evaluated by measuring *373 and *439 by ELISA 1 h after administration (i.v.) of 3 μg of human CgA1-439. The *373/*439 ratio is shown. C, proteolytic cleavage of exogenous CgA in normal mice (n = 5) and Vk*MYC mice (n = 29) with different levels of M-spike. Plasma samples were collected from each mouse 24 h after administration of 3 μg of human CgA1-439 and analyzed by *439- and *439+436+FRs-ELISA. The correlation between *439/*439+436+FRs ratio (an index of C-terminal fragmentation) and M-spike is shown. Bars, mean ± SEM; *, P < 0.05; **, P < 0.01; ***P < 0.001 (t-test, two tails).

Close modal

When we injected recombinant human CgA1-439 (3 μg, i.v.) to 55- to 80-week-old mice with M-spike >12%, we observed that the *373/*439 ratio in plasma samples taken 1 h later was increased, compared to mice with M-spike <12% (Fig. 5B), pointing to proteolytic cleavage of CgA in the C-terminal region. Accordingly, a similar experiment performed with normal and Vk*MYC mice showed that the *439/*439+436+FRs ratio (which is equivalent to 1 when CgA is intact and to 0 when it is totally cleaved in the C-terminal region) significantly and inversely correlated with the M-spike (Fig. 5C). These results support the hypothesis that disease progression is associated with cleavage of CgA. Considering that CgA was injected into mice intravenously, these data also support the hypothesis that the cleavage of CgA observed in patients occurred in the extracellular environment, likely the BM microenvironment, and not in neuroendocrine secretory cells.

Myeloma and endothelial cells can promote CgA C-terminal cleavage by activating plasminogen to plasmin

We then tested the hypothesis that multiple myeloma cells and/or proliferating endothelial cells, both present in the BM of patients with enhanced angiogenesis, could cleave CgA. To test these hypotheses we incubated full-length CgA with three human myeloma cell lines (KMS12, KMS28, U266) or with endothelial cells (HUVECs) and monitored CgA fragmentation by ELISA (6 h later). No cleavage was observed, arguing against a direct role of these cells in CgA fragmentation. However, when we added plasminogen to the cultures, we observed that two myeloma cell lines (KMS12, U266) and endothelial cells caused significant loss of immunoreactivity in the *439-ELISA, but not in the *439+436+FRs-ELISA, pointing to cleavage of CgA C-terminal region (Fig. 6A). It appears, therefore, that both myeloma and endothelial cells can promote CgA cleavage, although to a different extent. No significant degradation of CgA was observed with cultured HS-5 BM stromal cells or with human peripheral blood mononuclear cells (data not shown). Interestingly, the fragment *373 was detected in the supernatant of endothelial cells, but not of myeloma cells. Furthermore, cleavage was completely prevented by the addition of plasminogen-activator inhibitor-1 (PAI-1) to the cultures. Based on these data we hypothesized that myeloma and endothelial cells produced a plasminogen activator and that plasmin, thereof, was the proteolytic enzyme responsible for CgA cleavage.

Figure 6.

Myeloma and endothelial cells induce CgA C-terminal cleavage by activating plasminogen to plasmin. A, effect of myeloma cells, HUVECs, plasminogen, and PAI-1 on CgA fragmentation. HUVECs and human myeloma cell lines KMS12, KMS28, U266 (5 × 105 cells/well in 100 μL of PBS containing 0.5% of bovine serum albumin) were cultured for 24 h at 37°C in the presence or absence of plasminogen (PG, 8.8 μg/mL) and PAI-1 (1 μg/mL). Then, CgA1-439 (300 ng/mL) was added to the cell supernatants and further incubated for 6 h. The supernatants were collected and analyzed by *439- and *439+436+FRs-ELISAs. The levels of each analyte are shown as percentage of control. Bars, mean ± SEM; *, P < 0.05; **, P < 0.01; ***P < 0.001; ****P < 0.0001 (t-test, two tails). B, degradation of exogenous CgA1-439 upon incubation in peripheral blood (PB) and BM plasma samples. Full-length CgA1-439 (300 ng/mL) was added to three different pools of peripheral blood and BM plasma samples from six multiple myeloma patients (two patients/pool) in the presence or absence of PAI-1, left to incubate for 2 h at room temperature, and analyzed by *439- and *439+436+FRs-ELISAs. The analyte *439/*439+436+FRs ratio (an index of C-terminal degradation) is shown (mean ± SEM). BM, but not peripheral blood plasma, caused CgA degradation. Degradation was significantly inhibited by PAI-1. Bars, mean ± SEM; ***P < 0.001 (Wilcoxon test, paired ± PAI-1 values). C, measurement of plasmin activity in peripheral blood and BM plasma of multiple myeloma patients (pool 1, 2, and 3, as indicated). Plasmin assay was performed using the SensolyteRh110 fluorimetric assay (AnaSpec), based on cleavage of a synthetic fluorogenic substrate in the absence (solid line) and the presence (dashed line) of the selective plasmin inhibitor D-Val-Phe-Lys chloromethyl ketone. Product formation was monitored using a spectro-fluorimeter. RFU, relative fluorescence units; MM, multiple myeloma.

Figure 6.

Myeloma and endothelial cells induce CgA C-terminal cleavage by activating plasminogen to plasmin. A, effect of myeloma cells, HUVECs, plasminogen, and PAI-1 on CgA fragmentation. HUVECs and human myeloma cell lines KMS12, KMS28, U266 (5 × 105 cells/well in 100 μL of PBS containing 0.5% of bovine serum albumin) were cultured for 24 h at 37°C in the presence or absence of plasminogen (PG, 8.8 μg/mL) and PAI-1 (1 μg/mL). Then, CgA1-439 (300 ng/mL) was added to the cell supernatants and further incubated for 6 h. The supernatants were collected and analyzed by *439- and *439+436+FRs-ELISAs. The levels of each analyte are shown as percentage of control. Bars, mean ± SEM; *, P < 0.05; **, P < 0.01; ***P < 0.001; ****P < 0.0001 (t-test, two tails). B, degradation of exogenous CgA1-439 upon incubation in peripheral blood (PB) and BM plasma samples. Full-length CgA1-439 (300 ng/mL) was added to three different pools of peripheral blood and BM plasma samples from six multiple myeloma patients (two patients/pool) in the presence or absence of PAI-1, left to incubate for 2 h at room temperature, and analyzed by *439- and *439+436+FRs-ELISAs. The analyte *439/*439+436+FRs ratio (an index of C-terminal degradation) is shown (mean ± SEM). BM, but not peripheral blood plasma, caused CgA degradation. Degradation was significantly inhibited by PAI-1. Bars, mean ± SEM; ***P < 0.001 (Wilcoxon test, paired ± PAI-1 values). C, measurement of plasmin activity in peripheral blood and BM plasma of multiple myeloma patients (pool 1, 2, and 3, as indicated). Plasmin assay was performed using the SensolyteRh110 fluorimetric assay (AnaSpec), based on cleavage of a synthetic fluorogenic substrate in the absence (solid line) and the presence (dashed line) of the selective plasmin inhibitor D-Val-Phe-Lys chloromethyl ketone. Product formation was monitored using a spectro-fluorimeter. RFU, relative fluorescence units; MM, multiple myeloma.

Close modal

In agreement with this hypothesis, Western blot analysis of myeloma and endothelial cells showed that both cell types could produce the urinary-type plasminogen activator, an enzyme that can efficiently convert plasminogen to plasmin (Supplementary Fig. S4).

To assess whether the plasminogen activator/plasmin system was activated in the BM of multiple myeloma patients to an extent sufficient to cleave CgA, we spiked peripheral blood or BM plasma samples with human full-length CgA and analyzed each sample by *439-ELISA and *439+436+FRs-ELISA after incubation for 2 h in the absence or presence of PAI-1. Little or no C-terminal degradation occurred in peripheral blood plasma samples, as suggested by the observation that the *439/*439+436+FRs ratio remained close to 1 in these conditions (Fig. 6B). In contrast, this ratio dropped to 0.4 upon incubation in BM plasma, pointing to C-terminal degradation. Notably, degradation was significantly inhibited by PAI-1 (Fig. 6B), supporting the hypothesis that plasminogen activators were present in the BM in an amount sufficient to cause plasmin formation and CgA cleavage. Accordingly, when we measured the plasmin activity with the SensolyteRh110 fluorimetric assay, a very sensitive commercial kit, we observed that (i) BM plasma, but not peripheral blood plasma, could cleave the plasmin fluorogenic substrate and (ii) cleavage was completely inhibited by the selective plasmin inhibitor D-Val-Phe-Lys chloromethyl ketone (Fig. 6C). These findings further support the hypothesis that the plasminogen activator/plasmin system was activated in the BM of patients.

The R373R374 dibasic site of CgA C-terminal region is efficiently cleaved by plasmin

To verify that plasmin can indeed cleave the C-terminal region of CgA and to identify cleavage sites, we incubated full-length CgA with low amounts of plasmin-Sepharose and monitored CgA fragmentation by SDS-PAGE, Western blotting, ELISAs, and mass spectrometry. SDS-PAGE and Western blot analysis with various antibodies against the N-terminal (mAb B4E11), central (α-FRs, α-373), and C-terminal region (α-410-439, α-439) showed that plasmin can indeed cleave CgA to generate large fragments lacking the C-terminal region, including, albeit not limited to, CgA1-373 (Fig. 7A). This view was supported by the results of ELISA and mass spectrometry analysis (Fig. 7B and C). Interestingly, although various cleavage sites were identified in the C-terminal region of CgA by mass spectrometry (Fig. 7C and D), cleavage at R373 occurred faster than other sites (Fig. 7B and C). Of note, prolonged incubation with plasmin led to loss of immunoreactivity in the *373-ELISA, likely because of excessive cleavage at multiple sites. This may explain the lack of *373 in the supernatant of multiple myeloma cells that efficiently cleaved CgA.

Figure 7.

Cleavage of CgA C-terminal region by plasmin. A–C, SDS-PAGE, Western blotting, ELISA, and mass spectrometry analysis of CgA before and after incubation with plasmin-sepharose. CgA (150 μg/mL, in 680 μL of PBS) was incubated, for the indicated times, with 300 μL of a plasmin-sepharose suspension (1:3 in PBS), centrifuged, and analyzed by SDS-PAGE and Western blotting (A) with the following antibodies: mAb B4E11, α-FRs, α-373, α-410-439, and α-439 (see Materials and Methods and Fig. 1B for epitope location). Samples were also analyzed by ELISA as indicated (B) and, in the case of 2 and 24 h digestions, by electrospray mass spectrometry (C). The sequence of CgA and the identified plasmin cleavage sites at 24 h (arrows) are also shown (D).

Figure 7.

Cleavage of CgA C-terminal region by plasmin. A–C, SDS-PAGE, Western blotting, ELISA, and mass spectrometry analysis of CgA before and after incubation with plasmin-sepharose. CgA (150 μg/mL, in 680 μL of PBS) was incubated, for the indicated times, with 300 μL of a plasmin-sepharose suspension (1:3 in PBS), centrifuged, and analyzed by SDS-PAGE and Western blotting (A) with the following antibodies: mAb B4E11, α-FRs, α-373, α-410-439, and α-439 (see Materials and Methods and Fig. 1B for epitope location). Samples were also analyzed by ELISA as indicated (B) and, in the case of 2 and 24 h digestions, by electrospray mass spectrometry (C). The sequence of CgA and the identified plasmin cleavage sites at 24 h (arrows) are also shown (D).

Close modal

These findings, overall, suggest that both multiple myeloma and endothelial cells can contribute to CgA cleavage and that the plasminogen activator/plasmin system is an important mechanism for CgA fragmentation.

We have previously shown that circulating CgA1-439 (an antiangiogenic protein) can work as an angiogenic switch positively activated by cleavage at R373 and negatively regulated by cleavage at Q76, with CgA1-373 and CgA1-76 being capable of exerting pro- and antiangiogenic activities, respectively (15). This work provides experimental evidence to suggest that the CgA-angiogenic switch is positively activated in multiple myeloma.

In particular, the results of hematological studies performed with a series of ELISAs capable of discriminating between intact and cleaved molecules show that various CgA-derived polypeptides are present in the peripheral blood of multiple myeloma patients at diagnosis, including CgA1-439 (called *439), CgA1-373 (*373), CgA1-76 (*76), and other fragments. Remarkably, the relative levels of circulating pro- and antiangiogenic molecules (e.g. *373/*439 and *373/*76) were higher in patients than in normal subjects, suggesting that a proteolytic mechanism capable of “turning-on” the CgA-angiogenic switch was active in multiple myeloma patients. Furthermore, the results of hematological studies show that the BM plasma of patients contained higher levels of *373 and lower levels of *439 than peripheral-blood plasma, suggesting that cleavage at R373 occurred in the BM. At variance, cleavage at Q76 likely occurred in secretory neuroendocrine cells (and/or in circulation) and only in minor part in the BM, as suggested by the modest increase of *76 observed in the BM plasma compared to peripheral-blood plasma.

The results also show that certain multiple myeloma patients had higher circulating levels of total CgA compared to normal subjects. Which are the sources and the mechanisms underlying such increase? As cultured myeloma cells do not release CgA in the supernatant (not shown), it is likely that abnormal CgA was produced by the (neuro)endocrine system. Notably, in our study population, patients having abnormal CgA were taking PPIs and/or had renal failure, that is two conditions known to be associated with increased circulating CgA. Indeed, PPIs, which are drugs commonly used in the treatment of acid peptic disorders, can enhance two to three times, and in certain subjects even up to 10 times, the circulating levels of CgA (depending on administration schedule), by inducing CgA-positive enterochromaffin-like cell hyperplasia (30–32). Thus, enterochromaffin-like cells were likely the major source of increased CgA levels in PPI-treated patients. The mechanism of CgA increase consequent to renal failure is different, being likely related to a decline of glomerular filtration rate and accumulation of the CgA pool released by the diffuse neuroendocrine system, as previously reported (29). Thus, the cellular sources of aberrant CgA in PPI-treated patients and in patients with renal failure were likely different. We cannot exclude, however, that in some patients also the disease, by itself, may cause CgA elevation, considering that disease progression may be associated with occurrence of renal failure and increased need for PPI treatment.

Remarkably, patients who were not taking PPIs and had no renal failure had normal levels of total CgA, still had abnormally low levels of *439 and high levels of *373 in the BM (hence high *373/*439 ratio). This observation lends support to the concept that even the normal pool of circulating CgA was cleaved in the BM of patients, changing the local balance of pro-/antiangiogenic forms.

These findings raise the question as to whether changes in the balance of pro-/antiangiogenic CgA molecules in the BM might contribute to promote local angiogenesis. The following considerations suggest that CgA and its fragments may indeed have a role in the regulation of angiogenesis in patients. First, we have previously shown that CgA1-439 and CgA1-76 can inhibit angiogenesis in various in vivo and in vitro angiogenesis assays, whereas the fragment CgA1-373 promotes angiogenesis, at pathophysiologically relevant concentrations (15). Second, the results of the angiogenesis assays of this study show that CgA1-439 and CgA1-373 can counterbalance the activity of each other. Thus, the increased cleavage of CgA at residue R373 and other sites in the C-terminal region may tip the balance toward proangiogenic effects. Third, the present results of hematological studies in multiple myeloma patients show that the *373/*439 ratio correlates with the BM plasma levels of VEGF and FGF2, two potent proangiogenic factors known to play key roles in the cross-talk between myeloma cells, stromal cells, and endothelial cells (4, 33). Notably, in previous studies we have shown that CgA1-439 can inhibit the antiangiogenic activity of FGF2 and VEGF, whereas CgA1-373 can promote the release of FGF2 from cultured endothelial cells (15). Finally, the results of MVD analysis in BM tissue sections obtained from patients show positive correlation between MVD and the extent of proteolytic cleavage of CgA C-terminal region. These findings suggest that CgA and its fragments may indeed represent new players in the regulation of angiogenesis in the BM microenvironment of multiple myeloma patients.

Regarding the proteolytic mechanism responsible for the cleavage of the CgA C-terminal region in multiple myeloma patients, the results of in vitro experiments performed with cell cultures suggest that multiple myeloma cells and proliferating endothelial cells can induce CgA cleavage in a plasminogen-dependent manner and that this effect can be inhibited by PAI-1, a potent inhibitor of plasminogen activators. These findings point to the plasminogen activator/plasmin system as an important component of the CgA cleavage mechanism. Accordingly, the results of biochemical and immunological studies of CgA after treatment with low doses of plasmin show that indeed this enzyme can cleave CgA and generate large fragments lacking the C-terminal region, including CgA1-373. Interestingly, previous studies showed that myeloma cells express the urinary-type plasminogen activator (uPA) and the uPA receptor, a cell surface protein (34). This system is present also in the angiogenic endothelium and its activation is known to be important for angiogenesis (35, 36). It is therefore possible that plasmin activation, which may likely occur in the BM microenvironment of multiple myeloma patients (because of the presence of both myeloma cells and angiogenic endothelial cells), can start a vicious cycle involving local cleavage of circulating CgA, activation of angiogenesis, plasma cell proliferation, CgA cleavage, and so on. According to this view, we observed that Vk*MYC transgenic mice with advanced disease (having levels of M-spike >12%) and high MVD could cleave exogenous CgA more efficiently than mice whose M-spike was <12%. Furthermore, we detected plasmin activity in the BM plasma of patients, but not in their peripheral blood plasma. We cannot exclude, however, that other proteases in addition to plasmin (e.g., thrombin, known to cleave the R373–R374 bond (15) or other proteases produced by osteoclasts or other cells of the immune system present in the BM) are brought into play in the regulation of CgA activity. Notably, the high conservation of R373–R374 dibasic cleavage site, flanking residues, and adjacent CgA352-372 proangiogenic site, previously identified (15, 19), in human, mouse, rat, bovine, and horse CgA (Supplementary Fig. S5) further suggests that this proteolytic mechanism is biologically relevant.

We observed previously that the antiangiogenic activity of CgA occurs with U-shaped dose–response curve, the effect being lost at high CgA concentrations (>5 nmol/L; ref. 15). However, a dose-dependent response is likely to occur in multiple myeloma, as lower concentrations were measured (range 0.01–1.3 nmol/L). Furthermore, no U-shaped curves were observed with CgA1-373 up to 9 nmol/L, or with CgA1-76 up to 5 nmol/L (15), that is, with concentrations within the range of circulating CgA in multiple myeloma patients. The U-shaped curve and loss of activity could be relevant for neuroendocrine tumors in which CgA is released locally by neuroendocrine cancer cells in very high amounts, which is not the case of multiple myeloma.

In conclusion, the results of this study indicate that CgA, after being released in circulation by the neuroendocrine system, is proteolytically cleaved in the BM of multiple myeloma patients, tipping the local balance of anti-/proangiogenic CgA polypeptides toward a proangiogenic state. This finding may have important pathological implications and may stimulate clinical studies aimed at assessing whether detection of CgA levels and its fragmentation have prognostic values in terms of response to therapy and patient survival, and/or to assess whether pharmacological alteration of the CgA balance, for example by administration of antiangiogenic fragments, might have therapeutic effects.

P.L. Bergsagel reports receiving a commercial research grant from Novartis and Constellation. P.L. Bergsagel is also a consultant/advisory board member for Mundipharma, Sanofi, Janssen, Novartis, Incyte, Pfizer, and Onyx. No potential conflicts of interest were disclosed by the other authors.

Conception and design: M. Bianco, A. Corti

Development of methodology: M. Bianco, A.M. Gasparri, B. Colombo, F. Curnis, A. Sacchi, P.L. Bergsagel

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): M. Bianco, A.M. Gasparri, B. Colombo, S. Girlanda, M. Ponzoni, M.T.S. Bertilaccio, A. Calcinotto, A. Sacchi, M. Chesi, P.L. Bergsagel, F. Ciceri, M. Marcatti

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M. Bianco, A.M. Gasparri, F. Curnis, A. Calcinotto, M. Bellone, G. Tonon, F. Ciceri, A. Corti

Writing, review, and/or revision of the manuscript: M. Bianco, M. Ponzoni, M.T.S. Bertilaccio, E. Ferrero, M. Ferrarini, M. Bellone, G. Tonon, F. Caligaris-Cappio, A. Corti

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): M. Bianco, A.M. Gasparri, A. Corti

Study supervision: M. Bianco, F. Caligaris-Cappio, A. Corti

The authors thank Annapaola Andolfo for performing mass spectrometry analysis of CgA fragments.

This work was supported by Associazione Italiana per la Ricerca sul Cancro (AIRC, Special Program Molecular Clinical Oncology 5×1000–9965 and IG-14338).

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.

1.
Dimopoulos
MA
,
Kastritis
E
,
Anagnostopoulos
A
. 
Hematological malignancies: myeloma
.
Ann Oncol
2006
;
17
Suppl 10
:
137
43
.
2.
Vacca
A
,
Ribatti
D
. 
Bone marrow angiogenesis in multiple myeloma
.
Leukemia
2006
;
20
:
193
9
.
3.
Ria
R
,
Reale
A
,
De Luisi
A
,
Ferrucci
A
,
Moschetta
M
,
Vacca
A
. 
Bone marrow angiogenesis and progression in multiple myeloma
.
Am J Blood Res
2011
;
1
:
76
89
.
4.
Jakob
C
,
Sterz
J
,
Zavrski
I
,
Heider
U
,
Kleeberg
L
,
Fleissner
C
, et al
Angiogenesis in multiple myeloma
.
Eur J Cancer
2006
;
42
:
1581
90
.
5.
Vacca
A
,
Ribatti
D
,
Roncali
L
,
Ranieri
G
,
Serio
G
,
Silvestris
F
, et al
Bone marrow angiogenesis and progression in multiple myeloma
.
Br J Haematol
1994
;
87
:
503
8
.
6.
Sezer
O
,
Niemoller
K
,
Jakob
C
,
Zavrski
I
,
Heider
U
,
Eucker
J
, et al
Relationship between bone marrow angiogenesis and plasma cell infiltration and serum beta2-microglobulin levels in patients with multiple myeloma
.
Ann Hematol
2001
;
80
:
598
601
.
7.
Rajkumar
SV
,
Mesa
RA
,
Fonseca
R
,
Schroeder
G
,
Plevak
MF
,
Dispenzieri
A
, et al
Bone marrow angiogenesis in 400 patients with monoclonal gammopathy of undetermined significance, multiple myeloma, and primary amyloidosis
.
Clin Cancer Res
2002
;
8
:
2210
6
.
8.
Xu
JL
,
Lai
R
,
Kinoshita
T
,
Nakashima
N
,
Nagasaka
T
. 
Proliferation, apoptosis, and intratumoral vascularity in multiple myeloma: correlation with the clinical stage and cytological grade
.
J Clin Pathol
2002
;
55
:
530
4
.
9.
Pruneri
G
,
Ponzoni
M
,
Ferreri
AJ
,
Decarli
N
,
Tresoldi
M
,
Raggi
F
, et al
Microvessel density, a surrogate marker of angiogenesis, is significantly related to survival in multiple myeloma patients
.
Br J Haematol
2002
;
118
:
817
20
.
10.
Kumar
S
,
Witzig
TE
,
Greipp
PR
,
Rajkumar
SV
. 
Bone marrow angiogenesis and circulating plasma cells in multiple myeloma
.
Br J Haematol
2003
;
122
:
272
4
.
11.
Munshi
NC
,
Wilson
C
. 
Increased bone marrow microvessel density in newly diagnosed multiple myeloma carries a poor prognosis
.
Semin Oncol
2001
;
28
:
565
9
.
12.
Folkman
J
. 
Angiogenesis: an organizing principle for drug discovery?
Nat Rev Drug Discov
2007
;
6
:
273
86
.
13.
Italiano
JE
 Jr.
,
Richardson
JL
,
Patel-Hett
S
,
Battinelli
E
,
Zaslavsky
A
,
Short
S
, et al
Angiogenesis is regulated by a novel mechanism: pro- and antiangiogenic proteins are organized into separate platelet alpha granules and differentially released
.
Blood
2008
;
111
:
1227
33
.
14.
Ribatti
D
. 
Endogenous inhibitors of angiogenesis: a historical review
.
Leuk Res
2009
;
33
:
638
44
.
15.
Crippa
L
,
Bianco
M
,
Colombo
B
,
Gasparri
AM
,
Ferrero
E
,
Loh
YP
, et al
A new chromogranin A-dependent angiogenic switch activated by thrombin
.
Blood
2013
;
121
:
392
402
.
16.
Helle
KB
,
Corti
A
. 
Chromogranin A: a paradoxical player in angiogenesis and vascular biology
.
Cell Mol Life Sci
2014
;
72
:
339
48
.
17.
Corti
A
,
Ferrero
E
. 
Chromogranin A and the endothelial barrier function
.
Curr Med Chem
2012
;
19
:
4051
8
.
18.
Helle
KB
,
Corti
A
,
Metz-Boutigue
MH
,
Tota
B
. 
The endocrine role for chromogranin A: a prohormone for peptides with regulatory properties
.
Cell Mol Life Sci
2007
;
64
:
2863
86
.
19.
Theurl
M
,
Schgoer
W
,
Albrecht
K
,
Jeschke
J
,
Egger
M
,
Beer
AG
, et al
The neuropeptide catestatin acts as a novel angiogenic cytokine via a basic fibroblast growth factor-dependent mechanism
.
Circ Res
2010
;
107
:
1326
35
.
20.
Maestroni
S
,
Maestroni
A
,
Ceglia
S
,
Tremolada
G
,
Mancino
M
,
Sacchi
A
, et al
Effect of chromogranin A-derived vasostatin-1 on laser-induced choroidal neovascularization in the mouse
.
Acta Ophthalmol (Copenh)
2014
;
93
:
e218
22
.
21.
Belloni
D
,
Scabini
S
,
Foglieni
C
,
Veschini
L
,
Giazzon
A
,
Colombo
B
, et al
The vasostatin-I fragment of chromogranin A inhibits VEGF-induced endothelial cell proliferation and migration
.
FASEB J
2007
;
21
:
3052
62
.
22.
Veschini
L
,
Crippa
L
,
Dondossola
E
,
Doglioni
C
,
Corti
A
,
Ferrero
E
. 
The vasostatin-1 fragment of chromogranin A preserves a quiescent phenotype in hypoxia-driven endothelial cells and regulates tumor neovascularization
.
FASEB J
2011
;
25
:
3906
14
.
23.
Ceconi
C
,
Ferrari
R
,
Bachetti
T
,
Opasich
C
,
Volterrani
M
,
Colombo
B
, et al
Chromogranin A in heart failure; a novel neurohumoral factor and a predictor for mortality
.
Eur Heart J
2002
;
23
:
967
74
.
24.
Corti
A
. 
Chromogranin A and the tumor microenvironment
.
Cell Mol Neurobiol
2010
;
30
:
1163
70
.
25.
Chesi
M
,
Robbiani
DF
,
Sebag
M
,
Chng
WJ
,
Affer
M
,
Tiedemann
R
, et al
AID-dependent activation of a MYC transgene induces multiple myeloma in a conditional mouse model of post-germinal center malignancies
.
Cancer Cell
2008
;
13
:
167
80
.
26.
Corti
A
,
Longhi
R
,
Gasparri
A
,
Chen
F
,
Pelagi
M
,
Siccardi
AG
. 
Antigenic regions of human chromogranin A and their topographic relationships with structural/functional domains
.
Eur J Biochem
1996
;
235
:
275
80
.
27.
Ratti
S
,
Curnis
F
,
Longhi
R
,
Colombo
B
,
Gasparri
A
,
Magni
F
, et al
Structure-activity relationships of chromogranin A in cell adhesion. Identification of an adhesion site for fibroblasts and smooth muscle cells
.
J Biol Chem
2000
;
275
:
29257
63
.
28.
Chesi
M
,
Matthews
GM
,
Garbitt
VM
,
Palmer
SE
,
Shortt
J
,
Lefebure
M
, et al
Drug response in a genetically engineered mouse model of multiple myeloma is predictive of clinical efficacy
.
Blood
2012
;
120
:
376
85
.
29.
O'Connor
DT
,
Mahata
SK
,
Taupenot
L
,
Mahata
M
,
Livsey Taylor
CV
,
Kailasam
MT
, et al
Chromogranin A in human disease
.
Adv Exp Med Biol
2000
;
482
:
377
88
.
30.
Vlasveld
LT
,
van ‘t Wout
J
,
Castel
A
. 
False elevation of chromogranin A due to proton pump inhibitors
.
Netherl J Med
2011
;
69
:
207
.
31.
Giusti
M
,
Sidoti
M
,
Augeri
C
,
Rabitti
C
,
Minuto
F
. 
Effect of short-term treatment with low dosages of the proton-pump inhibitor omeprazole on serum chromogranin A levels in man
.
Eur J Endocrinol
2004
;
150
:
299
303
.
32.
Sanduleanu
S
,
Stridsberg
M
,
Jonkers
D
,
Hameeteman
W
,
Biemond
I
,
Lundqvist
G
, et al
Serum gastrin and chromogranin A during medium- and long-term acid suppressive therapy: a case-control study
.
Aliment Pharmacol Ther
1999
;
13
:
145
53
.
33.
Ribatti
D
,
Nico
B
,
Vacca
A
. 
Importance of the bone marrow microenvironment in inducing the angiogenic response in multiple myeloma
.
Oncogene
2006
;
25
:
4257
66
.
34.
Hjertner
O
,
Qvigstad
G
,
Hjorth-Hansen
H
,
Seidel
C
,
Woodliff
J
,
Epstein
J
, et al
Expression of urokinase plasminogen activator and the urokinase plasminogen activator receptor in myeloma cells
.
Br J Haematol
2000
;
109
:
815
22
.
35.
Pepper
MS
,
Montesano
R
,
Mandriota
SJ
,
Orci
L
,
Vassalli
JD
. 
Angiogenesis: a paradigm for balanced extracellular proteolysis during cell migration and morphogenesis
.
Enzyme Protein
1996
;
49
:
138
62
.
36.
Breuss
JM
,
Uhrin
P
. 
VEGF-initiated angiogenesis and the uPA/uPAR system
.
Cell Adh Migr
2012
;
6
:
535
615
.