There has been substantial progress in the understanding of the hereditary lymphedemas. However, the molecular mechanism is not always obvious, even when the genetic lesion has been identified. In 2009, Alders et al established the link between mutations in the CCBE1 gene and Hennekam lymphangiectasia–lymphedema syndrome, a human hereditary condition with lymphedema as a characteristic feature. Later, genetic experiments in zebrafish and mice indicated that the CCBE1 gene interacts with the lymphangiogenic vascular endothelial growth factor (VEGF)-C receptor-3 pathway, but the nature of the interaction remained elusive. In this article, Jeltsch et al show that the activity of VEGF-C is regulated by collagen- and calcium-binding epidermal growth factor domains 1 (CCBE1), which facilitates the proteolytic activation of a latent “pro” form of VEGF-C by the A disintegrin and metalloprotease with thrombospondin motifs-3 (ADAMTS3) metalloproteinase via a novel mode of growth factor activation. The in vivo data in the article show that CCBE1 is a potential therapeutic tool for the modulation of lymphangiogenesis and angiogenesis in a variety of diseases that involve the lymphatic system, such as lymphedema or lymphatic metastasis. In particular, application of VEGF-C has been scheduled for clinical trials to improve the incorporation of lymph node transplants into the lymphatic vascular system after mastectomy and axillary lymph node surgery. CCBE1 is a powerful activator of VEGF-C that can facilitate therapeutic lymphangiogenesis. CCBE1 and ADAMTS3 could also provide new targets for inhibition of tumor angiogenesis, lymphangiogenesis, and metastasis.
CCBE1 Enhances Lymphangiogenesis via A Disintegrin and Metalloprotease With Thrombospondin Motifs-3–Mediated Vascular Endothelial Growth Factor-C Activation
Abstract
Background—
Hennekam lymphangiectasia–lymphedema syndrome (Online Mendelian Inheritance in Man 235510) is a rare autosomal recessive disease, which is associated with mutations in the CCBE1 gene. Because of the striking phenotypic similarity of embryos lacking either the Ccbe1 gene or the lymphangiogenic growth factor Vegfc gene, we searched for collagen- and calcium-binding epidermal growth factor domains 1 (CCBE1) interactions with the vascular endothelial growth factor-C (VEGF-C) growth factor signaling pathway, which is critical in embryonic and adult lymphangiogenesis.
Methods and Results—
By analyzing VEGF-C produced by CCBE1-transfected cells, we found that, whereas CCBE1 itself does not process VEGF-C, it promotes proteolytic cleavage of the otherwise poorly active 29/31-kDa form of VEGF-C by the A disintegrin and metalloprotease with thrombospondin motifs-3 protease, resulting in the mature 21/23-kDa form of VEGF-C, which induces increased VEGF-C receptor signaling. Adeno-associated viral vector–mediated transduction of CCBE1 into mouse skeletal muscle enhanced lymphangiogenesis and angiogenesis induced by adeno-associated viral vector–VEGF-C.
Conclusions—
These results identify A disintegrin and metalloprotease with thrombospondin motifs-3 as a VEGF-C–activating protease and reveal a novel type of regulation of a vascular growth factor by a protein that enhances its proteolytic cleavage and activation. The results suggest that CCBE1 is a potential therapeutic tool for the modulation of lymphangiogenesis and angiogenesis in a variety of diseases that involve the lymphatic system, such as lymphedema or lymphatic metastasis.
Introduction
Vascular endothelial growth factor (VEGF) -C is the main driver of lymphangiogenesis in embryonic development and in various lymphangiogenic processes in adults.1 It acts via VEGF receptor (VEGFR) -3 and, in its proteolytically processed mature form, also via VEGFR-2. Deletion of the Vegfc gene in mice results in failure of lymphatic development because of the inability of newly differentiated lymphatic endothelial cells to migrate from the cardinal veins to sites where the first lymphatic structures form.2,3 This phenotype could be rescued by the application of recombinant VEGF-C.2 For the rescue, a mature form of VEGF-C was used, which lacked the N- and C-terminal propeptides. In cells secreting endogenous VEGF-C, these propeptides need to be proteolytically cleaved off from the central VEGF homology domain in order for VEGF-C to reach its full signaling potential.4 VEGF-C can also activate the main angiogenic receptor VEGFR-2 when both propeptides are cleaved off.4 Hence, the mature VEGF-C can stimulate also angiogenesis.
Clinical Perspective on p 1971
Mutations in VEGF-C and VEGFR-3 have been shown to result in hereditary lymphedema.5–7 Another hereditary condition with lymphedema as a cardinal symptom is Hennekam lymphangiectasia–lymphedema syndrome.8 In a subset of clinically diagnosed patients, mutations in the CCBE1 gene were found responsible for the disease,9,10 but it has been unclear how the mutant collagen- and calcium-binding epidermal growth factor domains 1 (CCBE1) causes the lymphatic phenotype. CCBE1 is a 2-domain protein with an N-terminal potential Ca-binding domain with epidermal growth factor–like repeats and a C-terminal domain with collagen-like repeats. Most of the known mutations in the CCBE1 gene are point mutations affecting its N-terminal domain; only 2 of all the identified mutations affect the collagen-like domain. All of the human mutations are expected to result in a functionally impaired CCBE1 protein but not in complete lack of CCBE1, which is likely incompatible with survival on the basis of gene deletion studies.11
In Vegfc-deficient embryos the differentiation of lymphatic endothelial cells from blood vascular endothelial cells in the cardinal veins appears unaffected, but they fail to egress from the cardinal veins.2,3 Compared with Vegfc-deficient embryos, the migration deficiency of nascent lymphatic endothelial cells in Ccbe1-deficient embryos is only partial. They form abnormal sprouts that fail to segregate from the cardinal veins, and the egressing lymphatic endothelial cells are unable to coalesce into discrete lymphatic structures.3,11 Hence, the earlier developmental block in the Vegfc-deficient embryos was attributed to the lack of a migration signal provided by VEGF-C, whereas the later block in Ccbe1-deficient embryos was attributed to a defect in endothelial cell migration, perhaps because of the lack of migratory cues from the extracellular matrix, of which CCBE1 is thought to be a component.12
Both Ccbe1+/− and Vegfc+/− heterozygous embryos show a reduction of Prox1-positive endothelial cells emigrating from the cardinal veins.3 The double heterozygous Ccbe1+/−;Vegfc+/− embryos have an aggravated version of this phenotype, suggesting that CCBE1 and VEGF-C participate synergistically in the lymphatic separation,3 which is also supported by data from zebrafish.12
In the present study, we have explored the link between CCBE1 and VEGF-C using both in vitro and in vivo assays and report that CCBE1 affects lymphangiogenesis by enhancing the cleavage of VEGF-C by the A disintegrin and metalloprotease with thrombospondin motifs-3 (ADAMTS3) metalloprotease, which removes the N-terminal propeptide from pro–VEGF-C, resulting in the mature, fully active VEGF-C.
Methods
Transfections, Metabolic Labeling, and Protein Analysis
293T and 293S GnTI− cells were (co)transfected with expression constructs coding for the indicated proteins. Twenty-four hours after the transfection, the cells were metabolically labeled with [35S]-cysteine/[35S]-methionine (PerkinElmer, Waltham, MA), and 48 hours later, conditioned cell culture medium and lysates were harvested. For the short-term labeling experiments, harvesting was performed after 24 hours. To produce unlabeled protein, the culture media were exchanged and supernatants and lysates harvested 48 hours later. After immunoprecipitation, the samples were electrophorated in 4% to 20% SDS-PAGE. For autoradiography, gels were dried and exposed to phosphoimager plates or x-ray film. For the immunodetection, the proteins were transferred to nitrocellulose. Specific signals were detected by enhanced chemiluminescence. Quantitation of the autoradiographies and Western blots was performed from the laser scanner readouts or scanned x-ray film using the ImageJ software (National Institutes of Health, Bethesda, MD).
Ba/F3-VEGFR/EpoR Assays
Stimulation of VEGFR-3 Phosphorylation
Near confluence porcine aortic endothelial cells expressing VEGFR-3 or VEGFR-3 plus neuropilin-2 were washed with PBS and starved over night in DMEM 0.2% BSA. ΔNΔC–VEGF-C, pro–VEGF-C, and CCBE1Δ175 were diluted to 0.02, 0.40, and 5.00 μg/mL in 1 mL of DMEM/0.1% BSA and incubated at 37°C for 30 minutes. The cells were stimulated for 10 or 30 minutes to detect phosphorylation of VEGFR-3 or downstream signaling proteins and then washed with ice-cold PBS. To cross-link proteins, the cells were washed twice with PBS, and purified proteins were applied in PBS (ΔNΔC–VEGF-C 100 ng/mL, pro–VEGF-C 1000 ng/mL, and CCBE1Δ175 at 25–50 μg/mL). After 3.5 minutes DTSSP (ThermoScientific, Waltham, MA) was added to a final concentration of 2 mmol/L, and cross-linking was performed for 6.5 minutes at 37°C. Cells were washed once with ice-cold TBS, lysed with 1% Triton X-100, and the immunoprecipitated fraction or the total lysate analyzed by SDS-PAGE/Western blot.
VEGFR-3 Trafficking in Human Umbilical Vein Endothelial Cells
Human umbilical vein endothelial cells stably transfected with the pMXs–VEGFR-3–green fluorescent protein vector16 were grown on glass-bottom microwells (MatTek Co, Ashland, MA) for 24 hours. The FCS concentration was reduced to 0.5%, and 12 hours thereafter, the cells were placed in an incubator (36°C and 5% CO2) on a Zeiss LSM 5 DUO Confocal microscope and treated with pro–VEGF-C or ΔNΔC–VEGF-C (100 ng/mL). The green fluorescent protein signal was recorded at the 488-nm wavelength. The human VEGFR-3 blocking antibody hF4-3C5 was used at 5 μg/mL.
Recombinant Adeno-Associated Viral Vector Production
In Vivo Experiments
Tibialis anterior muscles of FVB/N male mice were injected with 1:1 mixed solutions of AAV9s encoding mouse (m)CCBE1-V5, mVEGF-C, or HSA. The AAV9-HSA and AAV9–ΔNΔC–mVEGF-C single vectors were used as negative and positive controls, correspondingly. The total concentration of the vector particles in a single injected dose was 6×1010. Three weeks after transduction, the mice were euthanized by CO2 overdose. The tibialis anterior muscles were isolated, embedded into O.C.T. (Sakura Finetek Europe, Alphen aan den Rijn, The Netherlands), sectioned (10-μm thickness), and stained for the lymphatic (Lyve-1, Prox-1) and blood vascular markers (platelet endothelial cell adhesion molecule-1), as well as smooth muscle cell/pericyte (smooth muscle actin) and leukocyte (CD45) markers, followed by Alexa-conjugated secondary antibodies (Molecular Probes, Invitrogen, Life Technologies, Carlsbad, CA). Fluorescent images were obtained in an Axioplan microscope (Carl Zeiss AG, Oberkochen, Germany); the objectives were as follows: 10× numerical aperture = 0.3 and working distance = 5.6 mm and 20× numerical aperture = 0.5 and working distance = 2.0 mm; the camera was a Zeiss AxioCamHRm 14-bit greyscale charge-coupled device; the acquisition software was Zeiss AxioVision 4.6. Quantification of the areas stained for Lyve-1, platelet endothelial cell adhesion molecule-1, and smooth muscle actin was done as described previously.17 Shortly, we used the ImageJ software measuring the percentage of pixels that showed values above background in the appropriate color channel using the “measure” function. Prox-1 positive nuclei were counted manually. The detection of luciferase activity in EGFP/Luc Vegfr3EGFP/Luc mice was performed as described previously.19 The National Board for Animal Experiments of the Provincial State Office of Southern Finland approved all of the animal experiments carried out in this study.
Statistical Analysis
Significance of the differences was determined using 1-way ANOVA. When equal variances were assumed, the Tukey test was used as a post hoc test; when variances were assumed unequal, the Games-Howell test was used. For the Ba/F3 assays, separate ANOVAs were used for each concentration. The EC50s of the Ba/F3 assays were calculated using the 4-parameter logistic nonlinear regression model and the ReaderFit software (Hitachi Solutions America, Ltd, South San Francisco, CA). Four-parameter logistic nonlinear regression model calculations with and without weighting gave essentially similar results. Error bars in the figures indicate the SDs.
Results
CCBE1 Enhances VEGF-C Processing and Release, Resulting in Increased VEGFR-3 Activation
CCBE1 was detected as a protein of 40 to 55 kDa molecular weight in both cell lysates and conditioned media of 293T cells transfected with a CCBE1 expression vector (Figure 1A).20 However, most of the secreted CCBE1 migrated as a diffuse band of ≈100 kDa. Transfected VEGF-C was expressed as the uncleaved 58 kDa precursor, C-terminally processed 29/31 kDa the pro–VEGF-C form, and fully processed 21 kDa the mature form (Figure 1B, lane 1).4,21 However, when VEGF-C and CCBE1 were cotransfected, the amounts of the unprocessed VEGF-C and pro–VEGF-C were reduced, and the mature, fully activated VEGF-C became the major species (Figure 1B, lane 2).
Cotransfection with CCBE1 also facilitated the release of VEGF-C, because the cell-layer associated amount was reduced by 80% in lysates of the cotransfected cells in short-term labeling experiments (Figure IA in the online-only Data Supplement, compare lanes 5 through 8). A similar accelerated release of VEGF-C was achieved, when the C-terminal domain of VEGF-C was removed (Figure IB in the online-only Data Supplement).
Conditioned medium from the CCBE1/VEGF-C cotransfected cultures stimulated the growth and survival of Ba/F3–VEGFR-3/EpoR and Ba/F3–VEGFR-2/EpoR cells better than the medium of cells transfected with VEGF-C alone, whereas medium from CCBE1 transfected cells alone showed very little activity, indicating that the enhanced release and cleavage resulted in increased levels of active VEGF-C (Figure 1C and Figure IC in the online-only Data Supplement). Notably, CCBE1 also slightly promoted the survival of Ba/F3–VEGFR-3/EpoR cells, presumably because of increased processing and activation of endogenous VEGF-C made by the cells.
CCBE1 Enhances VEGF-C Processing In Trans
In the developing mouse and zebrafish embryo, CCBE1 is expressed in cells adjacent to developing lymphatic vessels.3,12 We thus determined whether CCBE1 production in trans by other cells also enhances VEGF-C processing. We transfected separate cultures of 293T cells with VEGF-C or CCBE1 and mixed the cell populations 24 hours after transfection. Alternatively, we mixed CCBE1-transfected cells with cells stably expressing VEGF-C. In these experiments, CCBE1 increased the efficiency of the extracellular processing of VEGF-C but not its release (Figure ID in the online-only Data Supplement).
CCBE1 Enhances the Lymphangiogenic Activity of VEGF-C In Vivo
Proteolytic processing of VEGF-C has been shown to increase its receptor affinity and biological activity.4,17 To investigate whether CCBE1 enhances VEGF-C–induced lymphangiogenesis in vivo, we transduced mouse tibialis anterior muscles with AAV9 expressing CCBE1 (AAV9–CCBE1) alone or together with AAV9–VEGF-C in a 1:1 ratio or AAV9–HSA as a negative control. AAV9 encoding the mature, activated form of VEGF-C (ΔNΔC–VEGF-C) was used as a positive control.
Three weeks after the AAV transduction, the muscles were analyzed by immunohistochemistry using markers for endothelial cells (platelet endothelial cell adhesion molecule-1), lymphatic endothelial cells (lymphatic vessel endothelial hyaluronan receptor (Lyve)-1, Prox1), and leukocytes (CD45). In this assay, both VEGF-C and ΔNΔC–VEGF-C stimulated lymphangiogenesis. ΔNΔC–VEGF-C gave a considerably stronger response at the same viral dose and stimulated additionally angiogenesis (Figure 2 and Figure II in the online-only Data Supplement, bottom row). This suggested that the proteolytic processing of VEGF-C was inefficient in the AAV9-transduced muscle.
However, when VEGF-C was cotransduced with CCBE1, lymphangiogenesis was significantly enhanced, as shown by the Lyve-1 and Prox-1 staining (Figure 2). Similar to the ΔNΔC–VEGF-C transduced muscle, significantly more angiogenesis and leukocyte recruitment were observed (Figures II and III in the online-only Data Supplement). These results indicated that CCBE1 enhances VEGF-C processing also in vivo.
To corroborate these findings, we used the AAV9s encoding the various VEGF-C forms and CCBE1 in mice heterozygous for a Vegfr3EGFP/Luc allele to monitor lymphangiogenesis by optical bioluminescent imaging in vivo.19 We detected strong luciferase signals in mice cotransduced with the AAVs encoding VEGF-C and CCBE1, weaker signals in mice transduced with VEGF-C or CCBE1 alone, and no bioluminescent signals in mice transduced with HSA (Figure 3).
VEGF-C and CCBE1 Are Processed by the ADAMTS3 Procollagenase
Our attempts to demonstrate a physical interaction of VEGF-C and CCBE1 were unsuccessful (Figure IVA in the online-only Data Supplement). We thus assumed that the CCBE1–VEGF-C interaction is short lived and/or indirect, perhaps mediated by the protease that removes the N-terminal propeptide of VEGF-C. We stably expressed CCBE1 in 293T cells, purified the protein, and subjected it to tryptic digestion followed by mass spectrometry. The most abundant copurified protease was ADAMTS3. Efficient N-terminal processing of pro–VEGF-C was obtained when ADAMTS3 was expressed together with VEGF-C in 293T cells (Figure 4A). To analyze whether CCBE1 enhances the ADAMTS3-mediated VEGF-C cleavage, the amounts of ADAMTS3 used for VEGF-C cleavage were titrated. When CCBE1-, VEGF-C–, and ADAMTS3-conditioned media were mixed in a ratio of 60:30:1, the ADAMTS3-mediated cleavage of VEGF-C was more efficient in the presence of CCBE1 than without (Figure 4B), and a corresponding medium had growth-promoting activity in the VEGFR-3/EpoR-expressing Ba/F3 cells (Figure 4C). When the culture media of the ADAMTS3 cotransfected samples were precipitated with ADAMTS3 antibodies or streptactin and analyzed in Western blotting with antibodies recognizing the C-terminus of CCBE1, the specific CCBE1 band migrated at 25 kDa, which corresponds to the collagen-like domain of CCBE1 (Figure 4D), indicating that ADAMTS3 may cleave CCBE1 between the epidermal growth factor and collagen homology domain. Interestingly, the DU-4475 cells produced only uncleaved CCBE1 (Figure IVB in the online-only Data Supplement), which did not promote VEGF-C activation (Figure IVC in the online-only Data Supplement).
VEGF-C Cleavage by Plasmin Is Not Influenced by CCBE1
As published previously,22 VEGF-C was efficiently cleaved by plasmin (Figure VA in the online-only Data Supplement). The fragments obtained with low amounts of plasmin activated VEGFR-3, but this activity was lost at high plasmin concentrations (Figure VB in the online-only Data Supplement). Edman degradation of the final products revealed the N-terminal sequence KTQC and a complete lack of the N-terminal helix, which is incompatible with VEGFR-3 activation.23 CCBE1 did not affect the efficiency of plasmin cleavage (Figure VC in the online-only Data Supplement).
ADAMTS3 Produced by 293T Cells Processes VEGF-C to the Mature Form
The N-terminus of the mature VEGF-C generated by incubation with recombinant, purified ADAMTS3 was identical to that reported for mature VEGF-C produced by 293 cells4 (Figure VIA in the online-only Data Supplement). VEGF-D was not cleaved by ADAMTS3 under the same conditions (Figure VIB in the online-only Data Supplement), despite featuring a similar cleavage motif (Figure VIC in the online-only Data Supplement).
Apart from ADAMTS3, 2 other proteases, ADAMTS2 and ADAMTS14, belong to the procollagenase subfamily of ADAMTS proteases.24 Interestingly, the ADAMTS1 gene deletion in mice results in deficient ovarian lymphangiogenesis.25 However, unlike ADAMTS3, ADAMTS1, 2, or 14 did not cleave VEGF-C (Figure VID in the online-only Data Supplement).
We found that the cell lines that produce active, mature VEGF-C (293T, 293T-CCBE1, and PC-3 cells) express ADAMTS3, whereas the cell lines that were unable or extremely inefficient in producing active VEGF-C (CHO and NIH-3T3), expressed very little or no ADAMTS3 (Figure VII in the online-only Data Supplement). Furthermore, when ADAMTS3 was silenced in 293T cells by using lentiviral short-hairpin RNA, the VEGF-C cleavage was inhibited (Figure VIIIA in the online-only Data Supplement).
VEGF-C/VEGF-D chimeras generated by propeptide swapping were not subject to ADAMTS3 cleavage (Figure VIIIB and VIIIC in the online-only Data Supplement). Interestingly, however, 79% of VEGF-C processing was inhibited by the purified C-terminal propeptide and 43% by the N-terminal propeptide, whereas the VEGF homology domain or HSA gave no inhibition (Figure VIIID), suggesting that the VEGF-C propeptides are necessary but not sufficient for VEGF-C recognition by ADAMTS3.
The N-Terminal Domain of CCBE1 Enhances Pro–VEGF-C Cleavage to the Mature Form
Because of the difficulty in expressing sufficient amounts of full-length CCBE1, we investigated whether the isolated N-terminal domain of CCBE1 (CCBE1Δ175) can increase VEGF-C activity. We stimulated VEGFR-3–transfected porcine aortic endothelial cells with pro–VEGF-C, which resulted in very little VEGFR-3 phosphorylation when compared with mature VEGF-C (Figure 5A, lanes 1 and 2). When the recombinant CCBE1Δ175 was added to pro–VEGF-C, VEGFR-3 phosphorylation was strongly increased (Figure 5A, lane 3). Analysis of VEGFR-3 coprecipitated proteins from the pro–VEGF-C stimulated cells indicated that both pro–VEGF-C and mature VEGF-C are bound to the receptor in the presence of CCBE1Δ175 (Figure 5B, compare lanes 2 and 3). To identify which form of VEGF-C was bound to the phosphorylated VEGFR-3 receptor, we applied purified CCBE1Δ175 and biotinylated, purified pro–VEGF-C to cultures of porcine aortic endothelial–VEGFR-3 cells in PBS for 210 seconds and cross-linked VEGFR-3–associated proteins for 390 seconds. Precipitation and analysis of tyrosyl phosphorylated proteins indicated that mature VEGF-C is bound to activated VEGFR-3 when both CCBE1Δ175 and pro–VEGF-C are used for the stimulation (Figure 5C). Pro–VEGF-C alone did not coprecipitate with VEGFR-3, unless VEGFR-3 was coexpressed with the VEGF-C coreceptor neuropilin-2 (Figure 5D). However, even then, pro–VEGF-C induced very little phosphorylation of VEGFR-3 (data not shown).
Pro–VEGF-C Can Act as a Competitive Inhibitor of Mature VEGF-C
We next analyzed the ability of pro–VEGF-C to inhibit VEGFR-3 activation by mature VEGF-C. Indeed, preincubation of lymphatic endothelial cells with high amounts of pro–VEGF-C inhibited their ability to respond to mature VEGF-C (Figure 6A). Unlike mature VEGF-C, pro–VEGF-C did not stimulate the endocytosis of VEGFR-3 or the phosphorylation of the Erk, Akt, or endothelial NO synthase downstream signaling proteins in blood vascular endothelial cells or lymphatic endothelial cells (Figure 6B and 6C).
Discussion
CCBE1 is essential for embryonic lymphangiogenesis.10–12 However, it has been unclear how it controls the lymphangiogenic response. Here we show that CCBE1 acts by regulating the cleavage of pro–VEGF-C into its active form. During its biosynthesis, unprocessed VEGF-C first undergoes a cleavage in the C-terminal part, resulting in pro–VEGF-C, and subsequently in the N-terminal part, yielding the mature form of VEGF-C.4 Proprotein convertases such as furin mediate the C-terminal cleavage of VEGF-C,26 but the protease that cleaves the N-terminal propeptide has not been clearly defined. We show that, whereas CCBE1 does not cleave VEGF-C, it greatly enhances the ADAMTS3-mediated N-terminal cleavage and activation of pro–VEGF-C. The N-terminal cleavage process seems inefficient in the majority of cultured cell lines, thus little of the pro–VEGF-C gets activated. Because of the remarkable difference in the lymphangiogenic potential between pro–VEGF-C and mature VEGF-C,17 it has been assumed that, analogous to VEGF-A,27,28 the proteolytic environment would be a critical determinant controlling VEGF-C bioavailability and activity in vivo.4,29 Our data indicate that CCBE1 expression in tissues could regulate VEGF-C activation in the lymphatic endothelial microenvironment in a spatially controlled manner.
We were unable to demonstrate a direct interaction between CCBE1 and VEGF-C, but CCBE1 interacted with the metalloproteinase ADAMTS3, as shown by mass spectrometry and a functional assay. ADAMTS3 cleavage of pro–VEGF-C was enhanced by CCBE1, whereas plasmin cleavage was not. The expression pattern of ADAMTS3 makes it a more likely candidate for VEGF-C activation during embryonic lymphangiogenesis than plasmin,30,31 but in wound healing and other invasive processes, where plasminogen becomes activated, VEGF-C activation (and deactivation) may occur via plasmin.
Although pro–VEGF-C is known to bind to VEGFR-3,4,32,33 it did not bind to or activate VEGFR-3 on its own in the porcine aortic endothelial–VEGFR-3 cells. However, when we introduced neuropilin-2, we could establish binding, yet we detected very little VEGFR-3 phosphorylation. This explains the competitive inhibition of mature VEGF-C activity by pro–VEGF-C in lymphatic endothelial cells, which express neuropilin-2.34,35
When we applied pro–VEGF-C with CCBE1 and cross-linked proteins that were bound to activated VEGFR-3, we detected mature VEGF-C. Thus, a rapid CCBE1-assisted cleavage of receptor-bound pro–VEGF-C by a cell-surface–associated protease appears responsible for the CCBE1 enhancement of pro–VEGF-C signaling activity. This is consistent with the fact that only little of the cleavage activity is released into the medium. The demonstration of CCBE1 enhancement in conditioned cell culture supernatants required carefully titrated amounts of ADAMTS3, pro–VEGF-C, and CCBE1, whereas the CCBE1 enhancement during the short VEGFR-3 phosphorylation period of 10 minutes was robust. Endothelial cells express ADAMTS3,36 most of which likely remains cell surface–associated because of its thrombospondin motif, which contains the high-affinity SVTCG binding site for CD36.37
We propose the model of VEGFR-3 activation shown in Figure 7. First, CCBE1 enables pro–VEGF-C binding to VEGFR-3. After binding, pro–VEGF-C becomes a substrate for proteases such as ADAMTS3, and the resulting in situ–generated mature VEGF-C initiates signaling. Such in situ activation of pro–VEGF-C could contribute to the lack of blood vascular effects of VEGF-C in some in vivo models,38 despite the ability of mature VEGF-C to activate VEGFR-2. The generation of mature VEGF-C also occurred in the culture medium, albeit much less efficiently. This could explain the modest angiogenesis that accompanied the prominent lymphangiogenic effect in vivo.39,40 Alternatively, in some instances, the angiogenic effect may be mediated via VEGFR-3.41
CCBE1 expression is spatially and temporally correlated with the migration routes of endothelial cells that bud from the cardinal veins.3,12 We could detect low amounts of CCBE1 in most cultured cell lines tested. Perhaps matrix association of CCBE1 via vitronectin11 could lead to high local CCBE1 concentrations, focusing ADAMTS3 activity to areas where VEGF-C activity is needed, for example, at sites where nascent lymphatic endothelial cells emigrate from the venous compartment. This resembles the concentration of plasminogen activator activity by vitronectin to cell surfaces and the extracellular matrix by binding to the urokinase-type plasminogen activator/soluble urokinase-type plasminogen activator receptor complex.42
VEGF-D, which is the closest homolog of VEGF-C,43,44 was not cleaved by ADAMTS3, and CCBE1 did not have any effect on its activation. Alignment of VEGF-C and VEGF-D orthologs reveals that both contain multiple potential plasmin cleavage sites in the linker connecting the N-terminal propeptide with the VEGF homology domain (Figure VIC in the online-only Data Supplement). Preferential cleavage at one site over the other might explain why limited exposure to plasmin activates VEGF-C, whereas longer exposure results in VEGF-C inactivation. The net charge of the polypeptide segment between the first potential plasmin cleavage site and the ADAMTS3 cleavage site in VEGF-C is very different from that in VEGF-D. This could explain the differential action of both plasmin and ADAMTS3 on these substrates despite similar cleavage motifs.
The cleavage motif of ADAMTS3 in VEGF-C is the same as the ADAMTS2 motif in procollagens (FA[AP]↓),45 which have been until now the only known substrates of ADAMTS3.46 This motif is also found in bone morphogenetic protein-2 and pleiotrophin. ADAMTS3 cleavage likely requires additional interactions, because exogenously added C-terminal VEGF-C propeptide and, to a lesser extent, also the N-terminal propeptide were able to compete with the cleavage. Surprisingly, a CCBE1 cleavage product appeared in the supernatants of ADAMTS3-transfected cells, although CCBE1 lacks the FA[AP]↓ motif. The cleavage of CCBE1 into 2 separate domains may be a prerequisite for its activity, because the inability of the conditioned medium of the DU-4475 cell line to activate pro–VEGF-C was associated with a lack of CCBE1 cleavage products.
In addition to the cleavage by ADAMTS3, CCBE1 also accelerated the release of VEGF-C. Although the C-terminal cleavage is not a prerequisite for secretion,26 the presence of the C-terminal propeptide slowed VEGF-C release compared with truncated VEGF-C forms lacking this propeptide in metabolic labeling pulse-chase experiments, indicating that facilitation of the C-terminal processing may be responsible for the enhanced release. However, the relevance of this finding is not yet clear, because it is uncertain whether cells that express both CCBE1 and VEGF-C exist in vivo.
On the basis of our findings, the lymphatic vessel defects seen in animal models lacking CCBE1 can be explained, because CCBE1 appears essential for VEGF-C activation. Furthermore, a decrease of CCBE1 expression together with other lymphangiogenic genes in the postnatal period in some tissues (Jeltsch et al, unpublished data, 2014) and potential additional substrates of ADAMTS3 suggest that CCBE1 also has lymphangiogenesis-independent roles, which may explain some of the other features of Hennekam lymphangiectasia–lymphedema syndrome.8,47 Finally, as shown by the in vivo data, CCBE1 may offer a useful target for the modulation of VEGF-C activity, which could be used for therapeutic stimulation and inhibition of lymphangiogenesis and perhaps also angiogenesis.48
Acknowledgments
We thank Marcel Lackner for assistance in the VEGFR-3 phosphorylation studies, Tanja Laakkonen for the generation of the AAVs, Tapio Tainola for DNA sequencing and laboratory management, and Markku Varjosalo and Sini Miettinen from the Proteomics Unit of the Institute of Biotechnology for mass spectrometry analysis and protein sequencing. Dr Bronek Pytowski (ImClone LLC/Eli Lilly & Co) generously provided the hF4-3C5 antibody and Drs Steven Stacker and Mark Achen the VD1 antibody.
CLINICAL PERSPECTIVE
Supplemental Material
File (002779_supplemental_material.pdf)
- Download
- 21.33 MB
References
1.
Alitalo K. The lymphatic vasculature in disease. Nat Med. 2011;17:1371–1380.
2.
Karkkainen MJ, Haiko P, Sainio K, Partanen J, Taipale J, Petrova TV, Jeltsch M, Jackson DG, Talikka M, Rauvala H, Betsholtz C, Alitalo K. Vascular endothelial growth factor C is required for sprouting of the first lymphatic vessels from embryonic veins. Nat Immunol. 2004;5:74–80.
3.
Hägerling R, Pollmann C, Andreas M, Schmidt C, Nurmi H, Adams RH, Alitalo K, Andresen V, Schulte-Merker S, Kiefer F. A novel multistep mechanism for initial lymphangiogenesis in mouse embryos based on ultramicroscopy. EMBO J. 2013;32:629–644.
4.
Joukov V, Sorsa T, Kumar V, Jeltsch M, Claesson-Welsh L, Cao Y, Saksela O, Kalkkinen N, Alitalo K. Proteolytic processing regulates receptor specificity and activity of VEGF-C. EMBO J. 1997;16:3898–3911.
5.
Gordon K, Schulte D, Brice G, Simpson MA, Roukens MG, van Impel A, Connell F, Kalidas K, Jeffery S, Mortimer PS, Mansour S, Schulte-Merker S, Ostergaard P. Mutation in vascular endothelial growth factor-C, a ligand for vascular endothelial growth factor receptor-3, is associated with autosomal dominant milroy-like primary lymphedema. Circ Res. 2013;112:956–960.
6.
Irrthum A, Karkkainen MJ, Devriendt K, Alitalo K, Vikkula M. Congenital hereditary lymphedema caused by a mutation that inactivates VEGFR3 tyrosine kinase. Am J Hum Genet. 2000;67:295–301.
7.
Karkkainen MJ, Ferrell RE, Lawrence EC, Kimak MA, Levinson KL, McTigue MA, Alitalo K, Finegold DN. Missense mutations interfere with VEGFR-3 signalling in primary lymphoedema. Nat Genet. 2000;25:153–159.
8.
Van Balkom ID, Alders M, Allanson J, Bellini C, Frank U, De Jong G, Kolbe I, Lacombe D, Rockson S, Rowe P, Wijburg F, Hennekam RC. Lymphedema-lymphangiectasia-mental retardation (Hennekam) syndrome: a review. Am J Med. Genet. 2002;112:412–421.
9.
Alders M, Mendola A, Adès L, Al Gazali L, Bellini C, Dallapiccola B, Edery P, Frank U, Hornshuh F, Huisman SA, Jagadeesh S, Kayserili H, Keng WT, Lev D, Prada CE, Sampson JR, Schmidtke J, Shashi V, van Bever Y, Van der Aa N, Verhagen JM, Verheij JB, Vikkula M, Hennekam RC. Evaluation of clinical manifestations in patients with severe lymphedema with and without CCBE1 mutations. Mol Syndromol. 2013;4:107–113.
10.
Alders M, Hogan BM, Gjini E, Salehi F, Al-Gazali L, Hennekam EA, Holmberg EE, Mannens MM, Mulder MF, Offerhaus GJ, Prescott TE, Schroor EJ, Verheij JB, Witte M, Zwijnenburg PJ, Vikkula M, Schulte-Merker S, Hennekam RC. Mutations in CCBE1 cause generalized lymph vessel dysplasia in humans. Nat Genet. 2009;41:1272–1274.
11.
Bos FL, Caunt M, Peterson-Maduro J, Planas-Paz L, Kowalski J, Karpanen T, van Impel A, Tong R, Ernst JA, Korving J, van Es JH, Lammert E, Duckers HJ, Schulte-Merker S. CCBE1 is essential for mammalian lymphatic vascular development and enhances the lymphangiogenic effect of vascular endothelial growth factor-C in vivo. Circ Res. 2011;109:486–491.
12.
Hogan BM, Bos FL, Bussmann J, Witte M, Chi NC, Duckers HJ, Schulte-Merker S. Ccbe1 is required for embryonic lymphangiogenesis and venous sprouting. Nat Genet. 2009;41:396–398.
13.
Achen MG, Roufail S, Domagala T, Catimel B, Nice EC, Geleick DM, Murphy R, Scott AM, Caesar C, Makinen T, Alitalo K, Stacker SA. Monoclonal antibodies to vascular endothelial growth factor-D block its interactions with both VEGF receptor-2 and VEGF receptor-3. Eur J Biochem. 2000;267:2505–2515.
14.
Stacker SA, Vitali A, Caesar C, Domagala T, Groenen LC, Nice E, Achen MG, Wilks AF. A mutant form of vascular endothelial growth factor (VEGF) that lacks VEGF receptor-2 activation retains the ability to induce vascular permeability. J Biol Chem. 1999;274:34884–34892.
15.
Mäkinen T, Jussila L, Veikkola T, Karpanen T, Kettunen MI, Pulkkanen KJ, Kauppinen R, Jackson DG, Kubo H, Nishikawa S, Ylä-Herttuala S, Alitalo K. Inhibition of lymphangiogenesis with resulting lymphedema in transgenic mice expressing soluble VEGF receptor-3. Nat Med. 2001;7:199–205.
16.
Ghalamkarpour A, Holnthoner W, Saharinen P, Boon LM, Mulliken JB, Alitalo K, Vikkula M. Recessive primary congenital lymphoedema caused by a VEGFR3 mutation. J Med Genet. 2009;46:399–404.
17.
Anisimov A, Alitalo A, Korpisalo P, Soronen J, Kaijalainen S, Leppänen VM, Jeltsch M, Ylä-Herttuala S, Alitalo K. Activated forms of VEGF-C and VEGF-D provide improved vascular function in skeletal muscle. Circ Res. 2009;104:1302–1312.
18.
Michelfelder S, Varadi K, Raupp C, Hunger A, Körbelin J, Pahrmann C, Schrepfer S, Müller OJ, Kleinschmidt JA, Trepel M. Peptide ligands incorporated into the threefold spike capsid domain to re-direct gene transduction of AAV8 and AAV9 in vivo. PLoS One. 2011;6:e23101.
19.
Martínez-Corral I, Olmeda D, Diéguez-Hurtado R, Tammela T, Alitalo K, Ortega S. In vivo imaging of lymphatic vessels in development, wound healing, inflammation, and tumor metastasis. Proc Natl Acad Sci U S A. 2012;109:6223–6228.
20.
Murad S, Grove D, Lindberg KA, Reynolds G, Sivarajah A, Pinnell SR. Regulation of collagen synthesis by ascorbic acid. Proc Natl Acad Sci U S A. 1981;78:2879–2882.
21.
Karpanen T, Alitalo K. Molecular biology and pathology of lymphangiogenesis. Annu Rev Pathol. 2008;3:367–397.
22.
McColl BK, Baldwin ME, Roufail S, Freeman C, Moritz RL, Simpson RJ, Alitalo K, Stacker SA, Achen MG. Plasmin activates the lymphangiogenic growth factors VEGF-C and VEGF-D. J Exp Med. 2003;198:863–868.
23.
Leppänen VM, Tvorogov D, Kisko K, Prota AE, Jeltsch M, Anisimov A, Markovic-Mueller S, Stuttfeld E, Goldie KN, Ballmer-Hofer K, Alitalo K. Structural and mechanistic insights into VEGF receptor 3 ligand binding and activation. Proc Natl Acad Sci U S A. 2013;110:12960–12965.
24.
Porter S, Clark IM, Kevorkian L, Edwards DR. The ADAMTS metalloproteinases. Biochem J. 2005;386:15–27.
25.
Brown HM, Dunning KR, Robker RL, Pritchard M, Russell DL. Requirement for ADAMTS-1 in extracellular matrix remodeling during ovarian folliculogenesis and lymphangiogenesis. Dev Biol. 2006;300:699–709.
26.
Siegfried G, Basak A, Cromlish JA, Benjannet S, Marcinkiewicz J, Chrétien M, Seidah NG, Khatib AM. The secretory proprotein convertases furin, PC5, and PC7 activate VEGF-C to induce tumorigenesis. J Clin Invest. 2003;111:1723–1732.
27.
Park JE, Keller GA, Ferrara N. The vascular endothelial growth factor (VEGF) isoforms: differential deposition into the subepithelial extracellular matrix and bioactivity of extracellular matrix-bound VEGF. Mol Biol Cell. 1993;4:1317–1326.
28.
Lee S, Jilani SM, Nikolova GV, Carpizo D, Iruela-Arispe ML. Processing of VEGF-A by matrix metalloproteinases regulates bioavailability and vascular patterning in tumors. J Cell Biol. 2005;169:681–691.
29.
Skobe M, Hamberg LM, Hawighorst T, Schirner M, Wolf GL, Alitalo K, Detmar M. Concurrent induction of lymphangiogenesis, angiogenesis, and macrophage recruitment by vascular endothelial growth factor-C in melanoma. Am J Pathol. 2001;159:893–903.
30.
Le Goff C, Somerville RP, Kesteloot F, Powell K, Birk DE, Colige AC, Apte SS. Regulation of procollagen amino-propeptide processing during mouse embryogenesis by specialization of homologous ADAMTS proteases: insights on collagen biosynthesis and dermatosparaxis. Development. 2006;133:1587–1596.
31.
Castellino FJ, Ploplis VA. Structure and function of the plasminogen/plasmin system. Thromb Haemost. 2005;93:647–654.
32.
Lee J, Gray A, Yuan J, Luoh SM, Avraham H, Wood WI. Vascular endothelial growth factor-related protein: a ligand and specific activator of the tyrosine kinase receptor Flt4. Proc Natl Acad Sci U S A. 1996;93:1988–1992.
33.
Kukk E, Lymboussaki A, Taira S, Kaipainen A, Jeltsch M, Joukov V, Alitalo K. VEGF-C receptor binding and pattern of expression with VEGFR-3 suggests a role in lymphatic vascular development. Development. 1996;122:3829–3837.
34.
Kärpänen T, Heckman CA, Keskitalo S, Jeltsch M, Ollila H, Neufeld G, Tamagnone L, Alitalo K. Functional interaction of VEGF-C and VEGF-D with neuropilin receptors. FASEB J. 2006;20:1462–1472.
35.
Xu Y, Yuan L, Mak J, Pardanaud L, Caunt M, Kasman I, Larrivée B, Del Toro R, Suchting S, Medvinsky A, Silva J, Yang J, Thomas JL, Koch AW, Alitalo K, Eichmann A, Bagri A. Neuropilin-2 mediates VEGF-C-induced lymphatic sprouting together with VEGFR3. J Cell Biol. 2010;188:115–130.
36.
Aung HH, Lame MW, Gohil K, He G, Denison MS, Rutledge JC, Wilson DW. Comparative gene responses to collected ambient particles in vitro: endothelial responses. Physiol Genomics. 2011;43:917–929.
37.
Li WX, Howard RJ, Leung LL. Identification of SVTCG in thrombospondin as the conformation-dependent, high affinity binding site for its receptor, CD36. J Biol Chem. 1993;268:16179–16184.
38.
Jeltsch M, Kaipainen A, Joukov V, Meng X, Lakso M, Rauvala H, Swartz M, Fukumura D, Jain RK, Alitalo K. Hyperplasia of lymphatic vessels in VEGF-C transgenic mice. Science. 1997;276:1423–1425.
39.
Rissanen TT, Markkanen JE, Gruchala M, Heikura T, Puranen A, Kettunen MI, Kholová I, Kauppinen RA, Achen MG, Stacker SA, Alitalo K, Ylä-Herttuala S. VEGF-D is the strongest angiogenic and lymphangiogenic effector among VEGFs delivered into skeletal muscle via adenoviruses. Circ Res. 2003;92:1098–1106.
40.
Karpanen T, Egeblad M, Karkkainen MJ, Kubo H, Ylä-Herttuala S, Jäättelä M, Alitalo K. Vascular endothelial growth factor C promotes tumor lymphangiogenesis and intralymphatic tumor growth. Cancer Res. 2001;61:1786–1790.
41.
Tammela T, Zarkada G, Wallgard E, Murtomäki A, Suchting S, Wirzenius M, Waltari M, Hellström M, Schomber T, Peltonen R, Freitas C, Duarte A, Isoniemi H, Laakkonen P, Christofori G, Ylä-Herttuala S, Shibuya M, Pytowski B, Eichmann A, Betsholtz C, Alitalo K. Blocking VEGFR-3 suppresses angiogenic sprouting and vascular network formation. Nature. 2008;454:656–660.
42.
Chavakis T, Kanse SM, Yutzy B, Lijnen HR, Preissner KT. Vitronectin concentrates proteolytic activity on the cell surface and extracellular matrix by trapping soluble urokinase receptor-urokinase complexes. Blood. 1998;91:2305–2312.
43.
Leppänen VM, Jeltsch M, Anisimov A, Tvorogov D, Aho K, Kalkkinen N, Toivanen P, Ylä-Herttuala S, Ballmer-Hofer K, Alitalo K. Structural determinants of vascular endothelial growth factor-D receptor binding and specificity. Blood. 2011;117:1507–1515.
44.
Stacker SA, Stenvers K, Caesar C, Vitali A, Domagala T, Nice E, Roufail S, Simpson RJ, Moritz R, Karpanen T, Alitalo K, Achen MG. Biosynthesis of vascular endothelial growth factor-D involves proteolytic processing which generates non-covalent homodimers. J Biol Chem. 1999;274:32127–32136.
45.
Dombrowski KE, Prockop DJ. Cleavage of type I and type II procollagens by type I/II procollagen N-proteinase: correlation of kinetic constants with the predicted conformations of procollagen substrates. J Biol Chem. 1988;263:16545–16552.
46.
Fernandes RJ, Hirohata S, Engle JM, Colige A, Cohn DH, Eyre DR, Apte SS. Procollagen II amino propeptide processing by ADAMTS-3: insights on dermatosparaxis. J Biol Chem. 2001;276:31502–31509.
47.
Al-Gazali LI, Hertecant J, Ahmed R, Khan NA, Padmanabhan R. Further delineation of Hennekam syndrome. Clin Dysmorphol. 2003;12:227–232.
48.
Norrmén C, Tammela T, Petrova TV, Alitalo K. Biological basis of therapeutic lymphangiogenesis. Circulation. 2011;123:1335–1351.
Information & Authors
Information
Published In
Copyright
© 2014 American Heart Association, Inc.
Versions
You are viewing the most recent version of this article.
History
Received: 2 February 2013
Accepted: 18 February 2014
Published online: 19 February 2014
Published in print: 13 May 2014
Keywords
Subjects
Authors
Disclosures
Dr Alitalo has been a consultant for Laurantis Pharma Oy. The other authors report no conflicts.
Sources of Funding
The Academy of Finland (grants 265982, 272683, 273612, and 273817), European Research Council (ERC-2010-AdG-268804 and Marie Curie Actions FP7/2007-2013REA 317250), Leducq Foundation (11CVD03), and Finnish Foundation for Cardiovascular Research are acknowledged for funding. Dr Holopainen was supported by the Jalmari and Rauha Ahokas Foundation and the K. Albin Johansson Foundation.
Metrics & Citations
Metrics
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
- Vascular endothelial growth factor signaling in health and disease: from molecular mechanisms to therapeutic perspectives, Signal Transduction and Targeted Therapy, 10, 1, (2025).https://doi.org/10.1038/s41392-025-02249-0
- Meningeal lymphatic drainage: novel insights into central nervous system disease, Signal Transduction and Targeted Therapy, 10, 1, (2025).https://doi.org/10.1038/s41392-025-02177-z
- Understanding the Lymphatic System: Tissue-on-Chip Modeling, Annual Review of Biomedical Engineering, 27, 1, (73-100), (2025).https://doi.org/10.1146/annurev-bioeng-110222-100246
- ADAMTS2: More than a procollagen N-proteinase, Genes & Diseases, (101686), (2025).https://doi.org/10.1016/j.gendis.2025.101686
- Trastornos que afectan a la vascularización cutánea: malformaciones vasculares, EMC - Dermatología, 59, 1, (1-13), (2025).https://doi.org/10.1016/S1761-2896(24)50025-5
- Regulation of VEGFR3 signaling in lymphatic endothelial cells, Frontiers in Cell and Developmental Biology, 13, (2025).https://doi.org/10.3389/fcell.2025.1527971
- Direct specification of lymphatic endothelium from mesenchymal progenitors, Nature Cardiovascular Research, 4, 1, (45-63), (2025).https://doi.org/10.1038/s44161-024-00570-5
- Prohibitin 2 orchestrates long noncoding RNA and gene transcription to accelerate tumorigenesis, Nature Communications, 15, 1, (2024).https://doi.org/10.1038/s41467-024-52425-z
- Mesenchymal Osr1+ cells regulate embryonic lymphatic vessel formation, Development, 151, 17, (2024).https://doi.org/10.1242/dev.202747
- Epidermal growth factor-like domain 7 drives brain lymphatic endothelial cell development through integrin αvβ3, Nature Communications, 15, 1, (2024).https://doi.org/10.1038/s41467-024-50389-8
- See more
Loading...
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginPurchase Options
Purchase this article to access the full text.
eLetters(0)
eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.
Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.