VEGF-C and VEGF-C156S in the pro-lymphangiogenic growth factor therapy of lymphedema: a large animal study

Abstract

Introduction

VEGF-C156S, a lymphangiogenesis-specific form of vascular endothelial growth factor C (VEGF-C), has been considered as a promising candidate for the experimental pro-lymphangiogenic treatment, as it lacks potential angiogenic effects. As a precursor to future clinical trials, the therapeutic efficacy and blood vascular side effects of VEGF-C and VEGF-C156S were compared in a large animal model of secondary lymphedema. Combination of lymphatic growth factor treatment and autologous lymph node transfer was used to normalize the lymphatic anatomy after surgical excision of lymphatic tissue.

Methods

Lymph vessels around the inguinal lymph node of female domestic pigs were destroyed in order to impair the normal lymphatic drainage from the hind limb. Local injections of adenoviruses (Ad) encoding VEGF-C or VEGF-C156S were used to enhance the regrowth of the lymphatic vasculature. AdLacZ (β-galactosidase) and saline injections served as controls.

Results

Both VEGF-C and VEGF-C156S induced growth of new lymphatic vessels in the area of excision, although lymphangiogenesis was notably stronger after VEGF-C treatment. Also the transferred lymph nodes were best-preserved in the VEGF-C-treated pigs. Despite the enlargement of blood vessels following the VEGF-C therapy, no signs of sprouting angiogenesis or increased blood vascular permeability in the form of increased wound exudate volumes were observed.

Conclusions

Our results show that VEGF-C provides the preferred alternative for growth factor therapy of lymphedema when compared to VEGF-C156S, due to the superior lymphangiogenic response and minor blood vessel effects. Furthermore, these observations suggest that activation of both VEGFR-2 and VEGFR-3 might be needed for efficient lymphangiogenesis.

Introduction

The lymphatic vasculature plays an essential role in the regulation of normal tissue fluid balance and participates in immune responses of the body by transporting various antigens and leukocytes into lymph nodes [13]. Impairment of normal lymphatic drainage promotes the formation of protein-rich edema in the affected limb, further leading to fibrosis, increased lipid deposition, trophic skin changes, inflammation and attenuated immune responses [4]. This progressive pathological condition, lymphedema, can vary from mild and local to prolonged, progressive, disfiguring, physically and psychosocially disabling and occasionally even life-threatening condition [1, 4].

Currently, no curative treatment for lymphedema exists. Most of the conservative treatment approaches, usually cumbersome and beneficial only to a limited extent, have been aimed at alleviating the symptoms [4, 5], while liposuction has been utilized in the management of adipose tissue hypertrophy in late-stage lymphedema [6]. Other surgical treatment options, such as lymphatic-venous shunts and lymphatic grafts, have met with limited success [5, 79]. Perhaps the most promising novel approach, autologous microvascular lymph node transfer, aims at reconstructing the normal lymphatic vascular anatomy and function after cancer treatment [10, 11]. However, in this surgical method the regrowth of lymph vessels is expected to occur spontaneously, and therefore, the incorporation of the transferred lymph node into the existing lymphatic vasculature may fail [1214]. The most recent experimental endeavors have focused on the combination of autologous lymph node transfer and targeted delivery of pro-lymphangiogenic growth factors [13, 14]. This approach can be used to reconstruct the entire lymphatic anatomy, i.e., both the lymphatic vessel network and the lymph nodes.

VEGF-C and VEGF-D (VEGF-C/D), members of the vascular endothelial growth factor (VEGF) family, have been shown to induce local and controlled lymphatic regeneration after lymphatic injury in animal models [1318]. Signaling of the VEGF growth factors occurs via tyrosine kinase receptors (VEGFRs) that are expressed in endothelial cells of blood and lymphatic vessels [1922]. The main target receptor of VEGF-C/D, VEGFR-3 [20, 23], has been demonstrated to be a key regulator of lymphangiogenesis—the formation of new lymphatic vessels from preexisting ones [2, 2426]. However, the proteolytically processed forms of VEGF-C and VEGF-D bind to and activate also VEGFR-2 [19, 20], which is considered as the main mediator of angiogenesis—the development of new blood vasculature from existing vessels [19, 22].

Although VEGF-C and VEGF-D preferentially promote the growth of lymphatic vessels through activation of VEGFR-3, high levels of these growth factors have been demonstrated to induce undesirable blood vascular changes, such as enlargement, tortuosity and increased vessel permeability, ultimately resulting in tissue edema [13, 15, 2730]. In order to minimize such adverse side effects, which are thought to result from VEGF-C/D to VEGFR-2 interaction, utilization of a point mutant form of VEGF-C, VEGF-C156S, has been studied [15, 25, 28, 31]. Unlike VEGF-C/D, VEGF-C156S is a selective agonist of only VEGFR-3 [31], and considered to be lacking effects on blood vasculature in the adult, yet capable of inducing regrowth of lymphatic vessels (Fig. 1) [15, 25, 28].

Fig. 1
figure1

Receptor signaling properties of VEGF-C, VEGF-D and VEGF-C156S. The proteolytically processed forms of VEGF-C and VEGF-D bind to both VEGFR-2 and VEGFR-3, resulting in the formation and activation of VEGFR-2 homodimers, VEGFR-2/3 heterodimers and VEGFR-3 homodimers [20, 23, 3941]. P processed, UP unprocessed

In the present study, we have compared the efficacy and potential adverse effects of VEGF-C and VEGF-C156S in a clinically relevant experimental lymph node transfer model [13, 14], mimicking the situation of lymphedema patients undergoing a lymph node transfer operation [10, 11]. Since previous reports have already demonstrated that VEGF-D has more blood vascular effects than VEGF-C [13, 28, 29], we concentrated on the differences between VEGF-C and VEGF-C156S, aiming at the selection of the best pro-lymphangiogenic molecule to be used in future clinical trials of lymphedema treatment.

Methods

In vitro comparison of adenovirus vectors

In order to preoperatively compare the expression levels of the adenoviruses, they were tested in commercially obtained primary human umbilical vein endothelial cells (HUVECs; Gibco, Life Technologies Corporation, Carlsbad, CA, USA), cultured according to the manufacturer’s guidelines and plated into 6-well plates at 150,000 cells per well. The cells were transduced using a multiplicity of infection (MOI) of 50 in replicates of six wells with adenoviruses encoding VEGF-C, VEGF-C156S or β-galactosidase (LacZ). A set of six non-transduced wells was used as a control. At 12 h after transduction, the cells were washed with Hank’s balanced salt solution (Gibco, Life Technologies Corporation, Carlsbad, CA, USA) and fed with fresh growth medium. Cell culture media were harvested at 72 h after the transduction. VEGF-C and VEGF-C156S concentrations were measured from the media by a commercial enzyme-linked immunosorbent assay (human VEGF-C ELISA; Biovendor, Brno, Czech Republic) according to the manufacturer’s instruction. Optical densities were read at 450 nm (using reference wavelength of 620 nm) with a microplate reader (Infinite 200; Tecan Group Ltd., Männedorf, Switzerland). A standard curve for interpolating VEGF-C and VEGF-C156S concentrations was formed with MasterPlex ReaderFit 2010 2.0.0.69 (MiraiBio Group of Hitachi Solutions America, Ltd., San Francisco, CA, USA), using a five-parameter logistic model, and the optical densities were converted to nanograms per milliliter.

Animal model

All experiments outlined in this article were approved by the Finnish National Animal Experiment Board (reference number ESAVI-2010-005896) and performed with female domestic pigs (Lab Animal Center, University of Eastern Finland, Kuopio, Finland), utilizing the previously developed porcine inguinal lymph node transfer model [13, 14]. Randomly chosen pigs were assigned to different treatment groups as follows: VEGF-C, n = 8; VEGF-C156S, n = 8; LacZ, n = 6, NaCl, n = 6. Preoperatively, azaperone (Stresnil 40 mg/ml; Janssen-Cilag, Vienna, Austria) and atropine (Atropine 1 mg/ml; Leiras, Helsinki, Finland) were administered with intramuscular injections as a preanesthetic medication, using weight-dependent doses. The animals were operated on under general propofol (Propofol-Lipuro 20 mg/ml; B. Braun, Melsungen, Germany) anesthesia at 15 mg/kg/h in combination with analgetic phentanyl (Fentanyl; Janssen-Cilag, Vienna, Austria) at 10 mg/kg/h, both delivered intravenously. Throughout the procedure, an adequate level of anesthesia was ensured by the lack of corneal reflex.

In contrast to the chains of lymph nodes found in humans, the pig has only one superficial lymph node located in the inguinal ring area of both hind limbs, to which the collecting lymphatic vessels in the thigh are connected. By excising lymphatic vessels surrounding this node, it is possible to locally destroy lymph flow from the hind limb and to mimic the clinical condition in secondary lymphedema patients [13, 14]. To visualize lymphatic vessels and the superficial lymph node, 0.3 ml of Patent Blue tracer dye (Guerbet, Villepinte, France) was injected intradermally into the right distal hind limb. A 10-cm skin incision was then made into the right inguinal ring area, and the superficial lymph node was located. After identifying the vascular pedicle (artery and vein) coming from the femoral vessels to the node, the inguinal lymphatic vascular network surrounding the node was destroyed by excising all afferent lymphatic vessels from a 5-cm radius and all efferent lymphatic vessels from a 3-cm radius. In addition, all fat tissue was also meticulously removed within this area. At this point, the superficial lymph node was connected to the surrounding tissues only by the 4-cm long vascular pedicle, which was left intact to simulate microsurgical anastomosis of blood vessels to the node [10]. The contralateral (left) lymph node was left untouched.

After destroying the lymphatic vessels, 1.4 × 1011 adenoviral vector (Ad) particles encoding VEGF-C156S were injected in 2.0 ml of 0.9 % NaCl into the perinodal fat surrounding the exposed lymph node. The corresponding dose for AdVEGF-C and AdLacZ was 1.0 × 1011, thus compensating for the lower expression level of AdVEGF-C156S observed preoperatively in the analysis of HUVEC culture media. Injection of physiological saline diluent solution (NaCl) was used for the control pigs. The pedicular lymph node was then rotated 90° laterally from its original position by the vascular pedicle and attached to its new location with 3–0 Vicryl sutures (Ethicon, Somerville, JN, USA), thus mimicking the lymph node transfer performed in human lymphedema patients, in which the autologous lymph node graft is vascularized, but lacks all lymphatic vessel connections [10, 11]. Finally, wounds were closed with 4–0 Polysorb sutures (Covidien, Dublin, Ireland). Postoperative analgesia was provided with carprofen (Rimadyl; Pfizer, New York, NY, USA) and buprenorphine (Temgesic; Merck & Co., Inc., Whitehouse Station, NJ, USA), and prophylactic antibiotic treatment with cefuroxime (Zinacef; GlaxoSmithKline, London, UK), using weight-dependent doses.

Postoperative wound exudate formation

Postoperatively, the pigs were monitored daily for wound exudate accumulation in the operated area by an observer without knowledge of the experimental groups. The amount of wound exudate was estimated in deciliters for at least 18 postoperative days or until the edema had fully dissipated. If the well-being or motility of a pig was evidently affected by the degree of fluid formation, the wound exudate was drained with a sterile needle and the volumes were recorded. To compensate for the skewing effect of varying fluid drainages, a sum function was established by adding the drained amounts to all subsequent volume estimates during the rest of the follow-up period.

Imaging of collecting lymphatic vessels and lymph node size

After a 2-month follow-up period, the pigs were sedated and anesthetized as described previously and injected with Patent Blue to visualize lymphatic vessels. A small collecting lymph vessel in the calf region was then cannulated using x3.5 magnification surgical loops, and Lipiodol radiography tracer (Guerbet, Villepinte, France) was injected into the vessel. Following the Lipiodol injection, X-ray lymphangiographies (still images and dynamic X-ray sequences taken at 30 fps) were recorded with a GE Innova 3100-IQ Excellence X-ray machine (GE Healthcare, Little Chalfont, UK). The pigs were then euthanized by a bolus intravenous injection of magnesium sulfate, while still under anesthesia. After euthanizing, the inguinal lymph node from the operated right side and the intact node from the left side were measured in three dimensions, and the volumes of the lymph nodes were estimated as ellipsoids using the equation \(V = \frac{4}{3}\pi abc\). The total number of lymphatic vessels and the number of lymphatic vessels connected to and draining past the superficial lymph node were later quantified from the lymphangiograms by an observer blinded to the treatment groups, using Photoshop CS5 (Adobe Systems, San Jose, California, USA).

Histology and immunohistochemistry

Tissue samples were collected from the transferred lymph nodes and their surrounding tissues, fixed in 4 % paraformaldehyde, embedded in paraffin and cut into 7-µm-thick sections. To evaluate lymph node atrophy, the sections were stained with hematoxylin–eosin. The stage of atrophy was graded separately by two persons blinded to the treatment groups, according to the following criteria: 0 = intact lymph node architecture, 1 = local formations of fibro-fatty tissue, 2 = accumulation of fibro-fatty formations, 3 = confluent areas of fibroblast proliferation or fat necrosis and 4 = widespread atrophy, only residuals of follicular structure remaining.

For immunohistochemical analysis of lymphatic vessels, tissue sections were incubated with primary antibodies against the lymphatic endothelial marker vascular endothelial growth factor receptor-3 (VEGFR-3) [32], and processed using standard avidin–biotin-HRP method, with 3′–5′-diaminobenzidine (Invitrogen Corporation, CA, USA) color substrate and methyl green counterstain. To analyze potential effects on blood vessels, another set of tissue sections was incubated with primary antibodies against platelet endothelial cell adhesion molecule-1 (PECAM-1/CD31; R&D Systems, MN, USA) and α-smooth muscle actin (α-SMA; Sigma-Aldrich, St. Louis, MO, USA), and further processed with avidin–biotin-HRP and alkaline-phosphatase systems (Vector Laboratories, CA, USA), using 3′–5′-diaminobenzidine (Invitrogen Corporation, CA, USA) and Vector Blue (Vector laboratories, CA, USA) color substrates, respectively, without counterstain.

In order to assess the mean cross-sectional areas (µm2) and densities (vessels per mm2) of lymphatic vessels proximal to the transferred lymph node, multiple tissue sections were evaluated for each animal and the best-preserved tissue section selected for further analysis. Next, several microscopic images (Olympus DP70 and Olympus AX-70; Olympus Corporation, Tokyo, Japan) were taken from the best-preserved areas of the section. The images were then organized according to the cross-sectional area and density of the lymphatic vessels and four images selected by ranking them visually: two from the areas of maximum vessel cross-sectional size (i.e., minimum density) and two from the areas of maximum lymphatic vessel density (i.e., minimum cross-sectional size). The images were analyzed with Adobe Photoshop CS5 (Adobe Systems, San Jose, California, USA), which was used to define the combined cross-sectional area (in µm2) of all lymphatic vessels (including both the lumen and the wall of the vessel) and the combined tissue area (in µm2, i.e., excluding areas without any stained tissue) from the four images. In addition, the total number of lymphatic vessels was also counted. The mean cross-sectional area was then calculated by dividing the total cross-sectional area of the lymph vessels by the total number of vessels. Similarly, the mean density was calculated by dividing the total number of lymph vessels by the total tissue area.

This procedure was also used to estimate the mean cross-sectional areas and densities of blood vessels from the PECAM-1/α-SMA double-stained tissue sections. To account for the pan-endothelial properties of PECAM-1 [33], and to distinguish blood vasculature from collecting lymphatic vessels with α-SMA positive pericytes/smooth muscle cells in the outer layer [3], only vessels with strongly immunostained endothelial cells were classified as blood vessels. This was made possible by the lower staining intensity of PECAM-1 in lymph vessel endothelium [3], demonstrated in our study by the presence of vessels with α-SMA positive outer layer, but with only borderline PECAM-1 positive endothelial cells, consequently categorized as collecting lymphatic vessels.

All analyses were done by an observer blinded to the treatment groups.

Statistical analyses

All data were analyzed using IBM SPSS Statistics 21.0.0.0 (IBM Corporation, Armonk, NC, USA), JMP Pro 10.0.0 (SAS Institute, Cary, NC, USA) or SAS 9.1.3 (SAS Institute, Cary, NC, USA). Statistical tests were performed two-tailed, and p values <0.05 were considered statistically significant. In case of non-normality or heterogeneity of variances, data were logarithmic-transformed before analysis and back-transformed to the original scale for presentation. For data not following normal distribution even after transformation, nonparametric tests were used.

The data in Figs. 4e, f, 5e, 6e, f and 7e, f were tested with one-way ANOVA, and in case of statistically significant group-wise differences, this was followed by Gabriel’s (Figs. 4e, f, 6e, f, and 7e) or Dunnett’s (Fig. 5e; using the intact group as a control) pairwise comparisons test. The data in Figs. 2, 4g and 5f were tested with Kruskal–Wallis one-way analysis of variance. For the data in Fig. 2, pairwise comparisons were performed using the Steel–Dwass method, and for the data in Fig. 5f, using the Steel method with the NaCl group as a control.

Fig. 2
figure2

VEGF-C/C156S expression levels. The amplitude of lymphatic growth factor production by the adenoviral vectors is tested preoperatively in human umbilical vein endothelial cell (HUVEC) cultures. The observed differences are later compensated by adjusting the adenoviral vector injection volumes accordingly. Horizontal lines in the graph indicate medians and boxes interquartile ranges, whereas whiskers extend from minimum to maximum values. Ns: p > 0.05; *: p < 0.05

The data in Fig. 3 were analyzed using a linear mixed-effects model. The time of the measurement (postoperative day 1 to day 18) was specified as a repeated effect, with heterogeneous first-order autoregressive as the covariance structure. The treatment group was specified as a fixed effect and wound exudate volume as a quantitative dependent variable. An interaction term between the treatment group and the time of the measurement was used to examine group-related changes during the follow-up period. Animal identifier was specified as a grouping variable to include a random intercept in the model, with variance components as the covariance structure.

Fig. 3
figure3

Postoperative development of wound exudate volume in the operated area. When the time of the measurement is ignored, the overall wound exudate volumes did not differ significantly between the groups, regardless of the treatment (p = 0.863). Volumes are expressed as geometric means

For normally distributed data, effect sizes for pairwise comparisons are reported as Hedges’ g statistic with 95 % confidence interval (CI), whereas Cliff’s delta statistic with 95 % CI is used for non-normally distributed data. Results are interpreted as a combination of both p value and effect size CI: With a significant p value and an effect size CI excluding the zero, the result is interpreted as significant and substantial. In case of a nonsignificant p value and an effect size CI excluding the zero, the result is interpreted as statistically nonsignificant, but still substantial. With a nonsignificant p value and an effect size CI including the zero, the result is interpreted as both nonsignificant and non-substantial.

Results are expressed as arithmetic means (±standard deviations) for normally distributed data (Figs. 4e, f, 5e and 6f) and geometric means (95 % CIs) for log-transformed data (Figs. 3, 6e and 7e), with error bars in the respective graphs representing 95 % CIs. For the non-normally distributed and non-transformed data (Figs. 2, 4g and 5f), results are reported as medians (interquartile ranges), with horizontal lines in graphs indicating medians and, when present, boxes interquartile ranges and whiskers extending from minimum to maximum values.

Fig. 4
figure4

VEGF-C treatment enhances the regrowth of collecting lymph vessels. ad Lymphangiogenic effects of VEGF-C and VEGF-C156S in transduced versus control pigs, illustrated by native lymphangiograms taken 2 months after the initial operation. Note also the differences in the perfusion of the transferred nodes (marked by asterisks). Afferent and efferent lymph vessels are indicated by arrows and arrowheads, respectively. Scale bars = 5 cm. e–g: Differences in the total number of collecting lymphatic vessels, the number of collecting vessels connecting to the transferred lymph node and the number collecting vessels bypassing the node. Results are reported as arithmetic means with error bars representing 95 % confidence intervals for figures e and f. For figure g, horizontal lines in the graph indicate medians and boxes interquartile ranges, whereas whiskers extend from minimum to maximum values. Ns: p > 0.05; *: p < 0.05; **: p < 0.01; ***: p < 0.001

Fig. 5
figure5

Lymph nodes in the VEGF-C group retain their size and structure better than the nodes in the other groups. ad Hematoxylin–eosin staining of the transferred lymph nodes at the end of the 2-month follow-up period. Scale bars = 1000 µm. e Size of the transferred lymph nodes, compared with intact nodes of the contralateral side. Results are expressed as arithmetic means and 95 % confidence intervals. Ns: p > 0.05; *: p < 0.05; ***: p < 0.001. f Level of lymph node atrophy as evaluated from the hematoxylin–eosin staining. Results are reported as individual scores, with horizontal lines indicating medians. Ns: p > 0.05; *: p < 0.05

Fig. 6
figure6

VEGF-C induces a robust growth of lymphatic vessels. ad VEGFR-3-stained tissue sections taken from the proximity of the transferred lymph node, 2 months after the initial operation. Lymphatic vessel endothelial cells positive for VEGFR-3 are stained in brown. Examples of VEGFR-3-stained lymph vessels are indicated by arrows. Scale bars = 250 µm. ef: Mean lymphatic vessel cross-sectional areas and densities. Results are expressed as geometric means for figure e and arithmetic means for figure f, with error bars representing 95 % confidence intervals in both figures. Ns: p > 0.05; *: p < 0.05; ***: p < 0.001

Fig. 7
figure7

VEGF-C induces enlargement of blood vessels, but does not alter vessel density. ad PECAM-1/α-SMA double-stained tissue sections taken from the proximity of the transferred lymph node at the end of the 2-month follow-up period. Blood vessel endothelial cells positive for PECAM-1 are stained in brown, while smooth muscle cells positive for α-SMA are visualized in blue. Examples of double-stained blood vessels are indicated by arrows. Scale bars = 250 µm. ef: Mean blood vessel cross-sectional areas and densities. Results are reported as geometric means for figure e and arithmetic means for figure f, with error bars representing 95 % confidence intervals in both figures. Ns: p > 0.05; *: p < 0.05

Results

AdVEGF-C and AdVEGF-C156S expression levels

Preoperative analysis of adenoviral VEGF-C and VEGF-C156S expression levels in HUVEC cultures indicated the amount of growth factor protein (Fig. 2; Table 1, Online Resource 1) to be highest in the cultures transduced with AdVEGF-C, significantly higher than in the AdVEGF-C156S (p = 0.037), AdLacZ (p = 0.023) and non-transduced (p = 0.024) cultures. Second-highest protein levels were found in the media of AdVEGF-C156S-transduced cultures, significantly higher when compared to the AdLacZ (p = 0.024) and non-transduced cultures (p = 0.025). The difference between the latter two was nonsignificant (p = 1.000) and non-substantial. In order to ensure comparable production of the growth factor protein, the observed 1.4-fold difference between AdVEGF-C and AdVEGF-C156S was compensated in the porcine model by the use of a correspondingly increased amount of AdVEGF-C156S viral particles.

Wound healing after surgical operation

No group-related differences were observed in the healing process, measured as the number and severity of postoperative surgical wound infections and duration of prophylactic antibiotic treatment; all inguinal wounds healed completely within 1 month after the operation.

Postoperative wound exudate dissipation

The surgical excision of lymphatic vasculature resulted in accumulation of wound exudate in the inguinal ring area. All visible signs of wound exudate were absent at the end of the third postoperative week, except for individual pigs in the VEGF-C, VEGF-C156S and NaCl groups, in which the wound exudate accumulates dissipated by the end of the fourth week. The length of the wound exudate formation period did not differ significantly (p = 0.521) or substantially between the treatment groups. The numbers of postoperative days needed before the final dissipation of visible wound exudate (expressed as geometric means and 95 % confidence intervals) were as follows: 14.3, 95 % CI (9.9, 20.6) for VEGF-C; 16.7, 95 % CI (13.3, 21.1) for VEGF-C156S; 12.3, 95 % CI (7.7, 19.8) for LacZ; 15.2, 95 % CI (10.7, 21.7) for NaCl.

In cases of excessive subcutaneous wound exudate volumes, the fluid was drained with a sterile needle. The number of drainages during the first 18 postoperative days was similar in all groups and the differences statistically nonsignificant (p = 0.104) and non-substantial. The numbers of drainages per pig (expressed as medians and interquartile ranges) were as follows: 1.00, 1.00–3.00 for VEGF-C; 2.00, 2.00–3.00 for VEGF-C156S; 1.00, 1.00–2.00 for LacZ; and 1.00, 1.00–1.50 for NaCl. The accumulation of wound exudate is visualized in Fig. 3.

Statistical analysis of the volumes revealed a significant main effect of the time of the measurement on the wound exudate volumes (p < 0.001). Furthermore, a significant interaction effect between the treatment group and the time of the measurement was also observed (p < 0.001). However, the main effect of the treatment group on the wound exudate volumes was nonsignificant (p = 0.863). These findings indicate that when the treatment group is ignored, the combined wound exudate volumes varied significantly between the postoperative days. Furthermore, the significant interaction effect demonstrates that the differences in the wound exudate volumes between the groups varied from postoperative day to another. Nevertheless, the nonsignificant main effect of the treatment group indicates that when the time of the measurement is ignored, the overall wound exudate volumes did not differ between the groups in a statistically significant way, regardless of the given treatment.

Lymphatic vessel regrowth and lymphatic flow function

Regrowth of new functional lymphatic vessels and, particularly, the formation of collecting lymph vessels were assessed by lymphangiography 2 months after the lymphatic vessel excision and gene transfer. As seen from the X-ray lymphangiograms (Fig. 4a–d), the recovery of lymphatic vasculature was most evident in pigs treated with either VEGF-C or VEGF-C156S (Fig. 4a, b). In contrast, the number of functional vessels spanning the operated area was markedly lower in the control-treated LacZ and NaCl groups (Fig. 4c, d). The lymphangiogenic response was quantified from the lymphangiograms as the total number of new collecting lymphatic vessels, including both the vessels connecting to the transferred lymph node and those bypassing it.

The total number of vessels (Fig. 4e; Table 2, Online Resource 1) was highest in the VEGF-C group, significantly higher than in the LacZ (p = 0.004) and NaCl (p = 0.002) groups. Similarly, also treatment with VEGF-C156S resulted in more pronounced lymphangiogenesis than control therapy, but only when compared to the NaCl group, and in a nonsignificant but substantial way: p = 0.097, g = 1.33, 95 % CI (0.10, 2.56). The rest of the group-wise differences were nonsignificant and non-substantial.

The number of vessels connecting to the transferred lymph node (Fig. 4f; Table 2, Online Resource 1) was highest in the VEGF-C group, significantly higher than in the LacZ (p = 0.001) and NaCl (p < 0.001) groups. In contrast to the total number of lymphatic vessels, treatment with VEGF-C156S did result in a significant increase in the number of vessels connected to the transferred node, but only when compared with the NaCl group (p = 0.045). The difference between the VEGF-C156S and the viral control LacZ group was nonsignificant (p = 0.056), but nonetheless, still substantial: g = 1.26, 95 % CI (0.04, 2.48). No other significant or substantial differences were observed.

Some collecting lymphatic vessels bypassing the transferred node (Fig. 4g; Table 2, Online Resource 1) were detected in most pigs, but these showed only nonsignificant and non-substantial group-wise differences.

Size and structure of the transferred lymph nodes

In order to estimate the preservation of the transferred lymph nodes (Fig. 5a–d), they were measured in three dimensions at the time of killing. When compared with the intact contralateral nodes, the lymph node volumes (Fig. 5e; Table 3, Online Resource 1) decreased nonsignificantly but substantially in the VEGF-C156S [p = 0.089, g = −1.20, 95 % CI (−2.13, −0.28)] and LacZ (p = 0.085, g = −1.22, 95 % CI (−2.27, −0.18)] groups and significantly in the NaCl group (p = 0.005). In contrast, size of the lymph nodes treated with VEGF-C did not differ significantly or substantially from their intact counterparts.

For further histological analysis, the transferred lymph nodes were cut into sections and stained with hematoxylin–eosin. Degradation of normal follicular structure and local fibro-fatty formations were detected to some extent in all of the nodes (Fig. 5a–d). However, the VEGF-C and VEGF-C156-s treated nodes had preserved their overall structure somewhat better than the LacZ and NaCl controls, in which confluent areas of fibro-fatty tissue were detected. Regular lymphatic structure had almost completely been replaced by adipose tissue in 33 % of the NaCl control nodes and by fibrotic tissue in 60 % of the LacZ nodes. A similar change, transition to fibro-fatty tissue, was noted in 25 % of the VEGF-C156S lymph nodes, whereas 100 % of the VEGF-C nodes retained their overall structure.

To quantify the changes, the sections were graded using a scale from 0 to 4, with 0 indicating intact lymph node architecture and 4, in turn, widespread atrophy. When compared to the NaCl control group, the transferred nodes had been preserved significantly better only in the VEGF-C group (p = 0.029), while the rest of the group-wise comparisons turned out nonsignificant and non-substantial (Fig. 5f; Table 4, Online Resource 1).

Lymphangiogenesis in tissue sections

To visualize and analyze not only the collecting lymphatic vessels, but also the lymphatic capillaries, tissue sections taken from the proximity of the transferred lymph node were stained for the lymphatic endothelial marker VEGFR-3. Mean cross-sectional areas (including both lumen and vessel wall) and densities of VEGFR-3 positive vessels were then determined from the photomicrographs (Fig. 6a–d).

The mean vessel area (Fig. 6e; Table 5, Online Resource 1) was largest in the VEGF-C group, nonsignificantly but substantially larger than in the LacZ group [p = 0.108, g = 1.79, 95 % CI (0.39, 3.19)] and significantly larger than in the NaCl group (p = 0.031). No other significant or substantial group-wise differences were observed.

The mean vessel density (Fig. 6f; Table 5, Online Resource 1) was highest in the VEGF-C group, significantly higher than in the VEGF-C156S (p < 0.001), LacZ (p < 0.001) and NaCl (p < 0.001) groups. The rest of the group-wise comparisons were nonsignificant and non-substantial.

Angiogenesis in tissue sections

Angiogenic effects of VEGF-C and VEGF-C156S therapy were examined using tissue sections taken from the proximity of the transferred lymph node, double stained for PECAM-1/α-SMA. Blood vessel cross-sectional areas and densities were quantified from the photomicrographs (Fig. 7a–d).

The mean vessel area (Fig. 7e; Table 6, Online Resource 1) was largest in the VEGF-C group, nonsignificantly but substantially larger than in the VEGF-C156S group [p = 0.135, g = 1.35, 95 % CI (0.11, 2.58)] and significantly larger than in the LacZ (p = 0.020) and NaCl (p = 0.020) groups. No other significant or substantial group-differences were observed. Furthermore, the mean blood vessel densities were similar and the differences between the groups were nonsignificant and non-substantial (Fig. 7f; Table 6, Online Resource 1).

Discussion

Lymphedema remains a common clinical challenge after surgical or radiation therapy of metastatic cancer. A recent systematic review and meta-analysis demonstrated that roughly one in five women with breast cancer will develop arm lymphedema [34]. We envision the lymph node transfer combined with the lymphatic growth factor therapy to be optimal for patients suffering from postmastectomy lymphedema.

Here we have compared the therapeutic potential and adverse effects of VEGF-C and its engineered variant VEGF-C156S in a lymphedema large animal model, combining surgical lymph node transfer with growth factor treatment [13, 14]. Our results show that while both VEGF-C and VEGF-C156S therapy induced regrowth of lymphatic vessels, lymphangiogenesis resulting from VEGF-C treatment was more robust. Consequently, the lymph nodes treated with VEGF-C seemed to be preserved better than those treated with VEGF-C156S. Furthermore, although VEGF-C therapy resulted in blood vessel dilation, the effect on blood vessel density was nonsignificant. Importantly, the observed wound exudate volumes did not indicate that VEGF-C would cause significant changes in vascular permeability. A summary of the study protocol and the results is shown in Fig. 8.

Fig. 8
figure8

Schematic view of the porcine lymphedema model. a Afferent and efferent lymphatic vessels surrounding the inguinal lymph node are excised. The tissue flap containing the lymph node is translocated, leaving the main artery and vein supplying the node intact. Adenoviral vectors encoding VEGF-C, VEGF-C156S or VEGF-D are administered perinodally into the flap. b Regrowth of lymphatic network is analyzed after the 2-month follow-up period. ce Summary of results from the current and relevant previous studies [15, 25, 28]. Regrowth of lymphatic vasculature and integration of the transferred lymph node can be most efficiently enhanced with perinodal injections of adenovirus vector encoding VEGF-C, while simultaneously minimizing possible adverse blood vascular effects. Proteolytic processing regulates in vivo receptor specificity and affinity of VEGF-C and VEGF-D

When originally engineered, the in vitro VEGFR-3 activating properties of VEGF-C156S were demonstrated to be equivalent to those of VEGF-C [31]. In vivo studies later indicated the selective activation of the VEGFR-3 signal transduction pathway by VEGF-C156S to result in a robust lymphangiogenesis [15, 25, 28]. Hence, VEGF-C156S was considered a suitable alternative for VEGF-C, especially due to the lack of potential adverse blood vascular effects originating from VEGFR-2 signaling [15, 25, 28]. However, these reports on the lymphangiogenic potential of VEGF-C156S focused on its impact on peripheral lymphatic capillary network, whereas in most lymphedema patients, the underlying clinical problem concerns primarily damage of the collecting lymph vessels [2, 4]. Furthermore, until now, the lymphangiogenic effect of VEGF-C156S has been estimated using small animal models and, in most cases, intact normal tissue [15, 25, 28].

In order to overcome these limitations, we compared the potential of VEGF-C156S therapy with that of VEGF-C in a clinically relevant large animal model. While treatment with VEGF-C proved potent in restoring the collecting lymphatic vessels at the site of surgical damage, the effect of VEGF-C156S therapy was more subdued. The weaker lymphangiogenic potential of VEGF-C156S may be explained by the differences in VEGF-C/D receptor properties and specificity (Figs. 1, 8c–e). Although the significance of VEGFR-3 signaling in adult tissue lymphangiogenesis has been well established [2, 24], the role of VEGFR-2 remains somewhat controversial. In the adult tissue, VEGFR-2 is mainly expressed in blood vascular endothelial cells, but weakly also in lymphatic endothelia—mostly in collecting lymphatic vessels and, in lesser extent, also in lymphatic capillaries [27, 35, 36]. VEGFR-3, on the other hand, localizes predominantly to lymphatic endothelial cells, especially in lymphatic capillaries [27, 36, 37]. This difference in expression could potentially provide an explanation for the notably stronger lymphangiogenesis resulting from the VEGF-C treatment.

While the selective activation of VEGFR-2 is not sufficient for the lymphangiogenic sprouting of new vessels [36], both VEGFR-2 and VEGFR-3 have been demonstrated to be essential and to cooperate in the early stages of lymphangiogenesis and serve redundant functions in the later stages [38]. The proteolytically processed forms of both VEGF-C and VEGF-D are known to induce formation of VEGFR-2/3 heterodimers in addition to VEGFR-2 and VEGFR-3 homodimers [20, 23, 3941], whereas binding of VEGF-C156S to its receptor results in VEGFR-3 homodimerization only (Fig. 1) [31]. VEGFR-2/3 heterodimers have already been suggested to regulate maturation of lymphatic capillaries into collecting lymph vessels [3]. Furthermore, these heterodimers are important in the angiogenic sprouting of blood vessels [40]. As angiogenesis and lymphangiogenesis have been demonstrated to be inherently related processes [42], VEGFR-2/3 heterodimers could also possess a role in lymphangiogenic sprouting.

Previous research has demonstrated that the success of autologous lymph node transfer depends on the lymphatic vascular regrowth, as the maintenance of normal lymph node structure and function requires incoming lymph flow [1214]. The effective lymphangiogenesis induced by the VEGF-C gene transfer therapy retained best both the size and histology of the transferred lymph nodes. In comparison, the results obtained with VEGF-C156S therapy or viral control vector were quite similar to each other, and only marginally better than those seen with non-viral control. Because lymph nodes perform crucial functions in several human diseases [43], their preservation and integration are of essential interest.

Besides lymphangiogenic efficacy, possible angiogenic effects present another important consideration. Previous studies have suggested that VEGF-C156S does not affect blood vessels [15, 28]. In comparison, VEGF-D treatment has been reported to increase blood vessel size and permeability [13, 2830], and according to some reports, to result also in sprouting angiogenesis [30]. VEGF-C, on the other hand, induces blood vessel enlargement in small animal models [15, 27, 28], and based on some of these studies, can also increase vascular permeability [15, 27], but does not seem to have an impact on the number of blood vessels [27, 28]. Until now, none of these effects of VEGF-C have previously been replicated in large animal models [13, 14]. The current data confirm that VEGF-C156S has no angiogenic potential, whereas VEGF-C resulted in blood vessel enlargement, but essentially no increased permeability or sprouting angiogenesis. Thus, the blood vascular effects of VEGF-C seem to be minor in comparison with VEGF-D [13].

Despite the evolution of medicine, lymphedema has remained a challenging problem with no curative treatment [1, 4, 5]. However, the results of experimental lymphatic growth factor therapy have raised new hope [43]. An attractive alternative to VEGF-C was provided by VEGF-C156S, a mutant form that induced specifically lymphangiogenesis, without the potential adverse blood vascular effects of VEGF-C [15, 28]. The present clinically relevant model of growth factor therapy demonstrated that when compared to VEGF-C156S, VEGF-C treatment provides superior lymphangiogenic efficacy, and only minor treatment-related blood vascular effects. VEGF-C does not seem to increase wound exudate formation, sprouting angiogenesis, blood vessel leakiness or lead to delayed wound healing. Results of the current study suggest that activation of both VEGFR-2 and VEGFR-3 is needed for efficient lymphatic vessel regrowth and collecting vessel formation. In conclusion, VEGF-C serves as the primary candidate for clinical trials of lymphedema.

Conclusions

Lymphedema, resulting from cancer surgery or radiation therapy, remains an unsolved problem with no curative treatment. Previous studies have demonstrated the combination of vascular endothelial growth factor (VEGF) therapy and autologous lymph node transfer to be efficient in reconstructing the damaged lymphatic network. VEGF-C156S, an engineered variant of VEGF-C, inducing specifically lymphangiogenesis via VEGF receptor-3 (VEGFR-3) activation and thus lacking potential angiogenic effects, has been suggested as an attractive alternative to its parental form VEGF-C. Our lymphedema large animal model shows that due to its superior lymphangiogenic efficacy, yet only minimal vascular side effects, VEGF-C serves as the primary candidate for future clinical trials. Furthermore, these results suggest that activation of both VEGFR-2 and VEGFR-3 might be needed for efficient lymphangiogenesis.

References

  1. 1.

    Witte MH, Bernas MJ, Martin CP, Witte CL (2001) Lymphangiogenesis and lymphangiodysplasia: from molecular to clinical lymphology. Microsc Res Tech 55:122–145. doi:10.1002/jemt.1163

    CAS  PubMed  Article  Google Scholar 

  2. 2.

    Alitalo K, Tammela T, Petrova TV (2005) Lymphangiogenesis in development and human disease. Nature 438:946–953. doi:10.1038/nature04480

    CAS  PubMed  Article  Google Scholar 

  3. 3.

    Norrmén C, Tammela T, Petrova TV, Alitalo K (2011) Biological basis of therapeutic lymphangiogenesis. Circulation 123:1335–1351. doi:10.1161/CIRCULATIONAHA.107.704098

    PubMed  Article  Google Scholar 

  4. 4.

    Rockson SG (2001) Lymphedema. Am J Med 110:288–295

    CAS  PubMed  Article  Google Scholar 

  5. 5.

    Suami H, Chang DW (2010) Overview of surgical treatments for breast cancer-related lymphedema. Plast Reconstr Surg 126:1853–1863. doi:10.1097/PRS.0b013e3181f44658

    CAS  PubMed  Article  Google Scholar 

  6. 6.

    Damstra RJ, Voesten HGJM, Klinkert P, Brorson H (2009) Circumferential suction-assisted lipectomy for lymphoedema after surgery for breast cancer. Br J Surg 96:859–864. doi:10.1002/bjs.6658

    CAS  PubMed  Article  Google Scholar 

  7. 7.

    Baumeister RG, Seifert J, Hahn D (1981) Autotransplantation of lymphatic vessels. Lancet 1:147

    CAS  PubMed  Article  Google Scholar 

  8. 8.

    Nagase T, Gonda K, Inoue K, Higashino T, Fukuda N, Gorai K, Mihara M, Nakanishi M, Koshima I (2005) Treatment of lymphedema with lymphaticovenular anastomoses. Int J Clin Oncol 10:304–310

    PubMed  Article  Google Scholar 

  9. 9.

    Felmerer G, Sattler T, Lohrmann C, Tobbia D (2012) Treatment of various secondary lymphedemas by microsurgical lymph vessel transplantation. Microsurgery 32:171–177. doi:10.1002/micr.20968

    PubMed  Article  Google Scholar 

  10. 10.

    Becker C, Assouad J, Riquet M, Hidden G (2006) Postmastectomy lymphedema: long-term results following microsurgical lymph node transplantation. Ann Surg 243:313–315. doi:10.1097/01.sla.0000201258.10304.16

    PubMed Central  PubMed  Article  Google Scholar 

  11. 11.

    Saaristo AM, Niemi TS, Viitanen TP, Tervala TV, Hartiala P, Suominen EA (2012) Microvascular breast reconstruction and lymph node transfer for postmastectomy lymphedema patients. Ann Surg 255:468–473. doi:10.1097/SLA.0b013e3182426757

    PubMed  Article  Google Scholar 

  12. 12.

    Mebius RE, Streeter PR, Brevé J, Duijvestijn AM, Kraal G (1991) The influence of afferent lymphatic vessel interruption on vascular addressin expression. J Cell Biol 115:85–95

    CAS  PubMed  Article  Google Scholar 

  13. 13.

    Lähteenvuo M, Honkonen K, Tervala T, Tammela T, Suominen E, Lähteenvuo J, Kholová I, Alitalo K, Ylä-Herttuala S, Saaristo A (2011) Growth factor therapy and autologous lymph node transfer in lymphedema. Circulation 123:613–620. doi:10.1161/CIRCULATIONAHA.110.965384

    PubMed  Article  Google Scholar 

  14. 14.

    Honkonen KM, Visuri MT, Tervala TV, Halonen PJ, Koivisto M, Lähteenvuo MT, Alitalo KK, Ylä-Herttuala S, Saaristo AM (2013) Lymph node transfer and perinodal lymphatic growth factor treatment for lymphedema. Ann Surg 257:961–967. doi:10.1097/SLA.0b013e31826ed043

    PubMed  Article  Google Scholar 

  15. 15.

    Saaristo A, Veikkola T, Tammela T, Enholm B, Karkkainen MJ, Pajusola K, Bueler H, Ylä-Herttuala S, Alitalo K (2002) Lymphangiogenic gene therapy with minimal blood vascular side effects. J Exp Med 196:719–730

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  16. 16.

    Szuba A, Skobe M, Karkkainen MJ, Shin WS, Beynet DP, Rockson NB, Dakhil N, Spilman S, Goris ML, Strauss HW, Quertermous T, Alitalo K, Rockson SG (2002) Therapeutic lymphangiogenesis with human recombinant VEGF-C. FASEB J 16:1985–1987. doi:10.1096/fj.02-0401fje

    CAS  PubMed  Google Scholar 

  17. 17.

    Tammela T, Saaristo A, Holopainen T, Lyytikkä J, Kotronen A, Pitkonen M, Abo-Ramadan U, Ylä-Herttuala S, Petrova TV, Alitalo K (2007) Therapeutic differentiation and maturation of lymphatic vessels after lymph node dissection and transplantation. Nat Med 13:1458–1466. doi:10.1038/nm1689

    CAS  PubMed  Article  Google Scholar 

  18. 18.

    Baker A, Kim H, Semple JL, Dumont D, Shoichet M, Tobbia D, Johnston M (2010) Experimental assessment of pro-lymphangiogenic growth factors in the treatment of post-surgical lymphedema following lymphadenectomy. Breast Cancer Res 12:R70–R70. doi:10.1186/bcr2638

    PubMed Central  PubMed  Article  Google Scholar 

  19. 19.

    Joukov V, Sorsa T, Kumar V, Jeltsch M, Claesson-Welsh L, Cao Y, Saksela O, Kalkkinen N, Alitalo K (1997) Proteolytic processing regulates receptor specificity and activity of VEGF-C. EMBO J 16:3898–3911. doi:10.1093/emboj/16.13.3898

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  20. 20.

    Achen MG, Jeltsch M, Kukk E, Mäkinen T, Vitali A, Wilks AF, Alitalo K, Stacker SA (1998) Vascular endothelial growth factor D (VEGF-D) is a ligand for the tyrosine kinases VEGF receptor 2 (Flk1) and VEGF receptor 3 (Flt4). Proc Natl Acad Sci USA 95:548–553

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  21. 21.

    Lohela M, Bry M, Tammela T, Alitalo K (2009) VEGFs and receptors involved in angiogenesis versus lymphangiogenesis. Curr Opin Cell Biol 21:154–165. doi:10.1016/j.ceb.2008.12.012

    CAS  PubMed  Article  Google Scholar 

  22. 22.

    Carmeliet P, Jain RK (2011) Molecular mechanisms and clinical applications of angiogenesis. Nature 473:298–307. doi:10.1038/nature10144

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  23. 23.

    Joukov V, Pajusola K, Kaipainen A, Chilov D, Lahtinen I, Kukk E, Saksela O, Kalkkinen N, Alitalo K (1996) A novel vascular endothelial growth factor, VEGF-C, is a ligand for the Flt4 (VEGFR-3) and KDR (VEGFR-2) receptor tyrosine kinases. EMBO J 15:290–298

    CAS  PubMed Central  PubMed  Google Scholar 

  24. 24.

    Karkkainen MJ, Ferrell RE, Lawrence EC, Kimak MA, Levinson KL, McTigue MA, Alitalo K, Finegold DN (2000) Missense mutations interfere with VEGFR-3 signalling in primary lymphoedema. Nat Genet 25:153–159. doi:10.1038/75997

    CAS  PubMed  Article  Google Scholar 

  25. 25.

    Veikkola T, Jussila L, Makinen T, Karpanen T, Jeltsch M, Petrova TV, Kubo H, Thurston G, McDonald DM, Achen MG, Stacker SA, Alitalo K (2001) Signalling via vascular endothelial growth factor receptor-3 is sufficient for lymphangiogenesis in transgenic mice. EMBO J 20:1223–1231. doi:10.1093/emboj/20.6.1223

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  26. 26.

    Karpanen T, Alitalo K (2008) Molecular biology and pathology of lymphangiogenesis. Annu Rev Pathol Mech Dis 3:367–397. doi:10.1146/annurev.pathmechdis.3.121806.151515

    CAS  Article  Google Scholar 

  27. 27.

    Saaristo A, Veikkola T, Enholm B, Hytönen M, Arola J, Pajusola K, Turunen P, Jeltsch M, Karkkainen MJ, Kerjaschki D, Bueler H, Ylä-Herttuala S, Alitalo K (2002) Adenoviral VEGF-C overexpression induces blood vessel enlargement, tortuosity, and leakiness but no sprouting angiogenesis in the skin or mucous membranes. FASEB J 16:1041–1049. doi:10.1096/fj.01-1042com

    CAS  PubMed  Article  Google Scholar 

  28. 28.

    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 (2003) VEGF-D is the strongest angiogenic and lymphangiogenic effector among VEGFs delivered into skeletal muscle via adenoviruses. Circ Res 92:1098–1106. doi:10.1161/01.RES.0000073584.46059.E3

    CAS  PubMed  Article  Google Scholar 

  29. 29.

    Rutanen J, Rissanen TT, Markkanen JE, Gruchala M, Silvennoinen P, Kivelä A, Hedman A, Hedman M, Heikura T, Ordén M, Stacker SA, Achen MG, Hartikainen J, Ylä-Herttuala S (2004) Adenoviral catheter-mediated intramyocardial gene transfer using the mature form of vascular endothelial growth factor-D induces transmural angiogenesis in porcine heart. Circulation 109:1029–1035. doi:10.1161/01.CIR.0000115519.03688.A2

    CAS  PubMed  Article  Google Scholar 

  30. 30.

    Kholová I, Koota S, Kaskenpää N, Leppänen P, Närväinen J, Kavec M, Rissanen TT, Hazes T, Korpisalo P, Gröhn O, Ylä-Herttuala S (2007) Adenovirus-mediated gene transfer of human vascular endothelial growth factor-D induces transient angiogenic effects in mouse hind limb muscle. Hum Gene Ther 18:232–244. doi:10.1089/hum.2006.100

    PubMed  Article  Google Scholar 

  31. 31.

    Joukov V, Kumar V, Sorsa T, Arighi E, Weich H, Saksela O, Alitalo K (1998) A recombinant mutant vascular endothelial growth factor-C that has lost vascular endothelial growth factor receptor-2 binding, activation, and vascular permeability activities. J Biol Chem 273:6599–6602

    CAS  PubMed  Article  Google Scholar 

  32. 32.

    Jussila L, Valtola R, Partanen TA, Salven P, Heikkilä P, Matikainen M, Renkonen R, Kaipainen A, Detmar M, Tschachler E, Alitalo R, Alitalo K (1998) Lymphatic endothelium and Kaposi’s sarcoma spindle cells detected by antibodies against the vascular endothelial growth factor receptor-3. Cancer Res 58:1599–1604

    CAS  PubMed  Google Scholar 

  33. 33.

    Pusztaszeri MP, Seelentag W, Bosman FT (2006) Immunohistochemical expression of endothelial markers CD31, CD34, von Willebrand factor, and Fli-1 in normal human tissues. J Histochem Cytochem 54:385–395. doi:10.1369/jhc.4A6514.2005

    CAS  PubMed  Article  Google Scholar 

  34. 34.

    DiSipio T, Rye S, Newman B, Hayes S (2013) Incidence of unilateral arm lymphoedema after breast cancer: a systematic review and meta-analysis. Lancet Oncol 14:500–515. doi:10.1016/S1470-2045(13)70076-7

    PubMed  Article  Google Scholar 

  35. 35.

    Millauer B, Wizigmann-Voos S, Schnürch H, Martinez R, Møller NPH, Risau W, Ullrich A (1993) High affinity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis. Cell 72:835–846

    CAS  PubMed  Article  Google Scholar 

  36. 36.

    Wirzenius M, Tammela T, Uutela M, He Y, Odorisio T, Zambruno G, Nagy JA, Dvorak HF, Ylä-Herttuala S, Shibuya M, Alitalo K (2007) Distinct vascular endothelial growth factor signals for lymphatic vessel enlargement and sprouting. J Exp Med 204:1431–1440. doi:10.1084/jem.20062642

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  37. 37.

    Kaipainen A, Korhonen J, Mustonen T, van Hinsbergh Victor W M, Fang G, Dumont D, Breitman M, Alitalo K (1995) Expression of the fms-like tyrosine kinase 4 gene becomes restricted to lymphatic endothelium during development. Proc Natl Acad Sci U S A 92:3566–3570

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  38. 38.

    Goldman J, Rutkowski JM, Shields JD, Pasquier MC, Cui Y, Schmökel HG, Willey S, Hicklin DJ, Pytowski B, Swartz MA (2007) Cooperative and redundant roles of VEGFR-2 and VEGFR-3 signaling in adult lymphangiogenesis. FASEB J 21:1003–1012. doi:10.1096/fj.06-6656com

    CAS  PubMed  Article  Google Scholar 

  39. 39.

    Dixelius J, Mäkinen T, Wirzenius M, Karkkainen MJ, Wernstedt C, Alitalo K, Claesson-Welsh L (2003) Ligand-induced vascular endothelial growth factor receptor-3 (VEGFR-3) heterodimerization with VEGFR-2 in primary lymphatic endothelial cells regulates tyrosine phosphorylation sites. J Biol Chem 278:40973–40979. doi:10.1074/jbc.M304499200

    CAS  PubMed  Article  Google Scholar 

  40. 40.

    Nilsson I, Bahram F, Li X, Gualandi L, Koch S, Jarvius M, Söderberg O, Anisimov A, Kholová I, Pytowski B, Baldwin M, Ylä Herttuala S, Alitalo K, Kreuger J, Claesson-Welsh L (2010) VEGF receptor 2/-3 heterodimers detected in situ by proximity ligation on angiogenic sprouts. EMBO J 29:1377–1388. doi:10.1038/emboj.2010.30

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  41. 41.

    Harris NC, Davydova N, Roufail S, Paquet-Fifield S, Paavonen K, Karnezis T, Zhang Y, Sato T, Rothacker J, Nice EC, Stacker SA, Achen MG (2013) The propeptides of VEGF-D determine heparin binding, receptor heterodimerization, and effects on tumor biology. J Biol Chem 288:8176–8186. doi:10.1074/jbc.M112.439299

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  42. 42.

    Nakao S, Zandi S, Hata Y, Kawahara S, Arita R, Schering A, Sun D, Melhorn MI, Ito Y, Lara-Castillo N, Ishibashi T, Hafezi-Moghadam A (2011) Blood vessel endothelial VEGFR-2 delays lymphangiogenesis: an endogenous trapping mechanism links lymph- and angiogenesis. Blood 117:1081–1090. doi:10.1182/blood-2010-02-267427

    CAS  PubMed Central  PubMed  Article  Google Scholar 

  43. 43.

    Hartiala P, Saaristo AM (2010) Growth factor therapy and autologous lymph node transfer in lymphedema. Trends Cardiovasc Med 20:249–253. doi:10.1016/j.tcm.2011.11.008

    CAS  PubMed  Article  Google Scholar 

Download references

Acknowledgments

We wish to acknowledge Heikki Karhunen, Minna Törrönen, Heikki Pekonen and Seija Sahrio for their excellent technical assistance. This study was funded by the Academy of Finland, the Turku University Foundation and Special Governmental Funding (EVO) allocated to Turku University Central Hospital.

Conflict of interest

Drs. Saarikko and Alitalo have consultant agreements with Herantis Pharma Plc.

Ethical standard

All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. All procedures performed in studies involving animals were in accordance with the ethical standards of the institution or practice at which the studies were conducted.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Mikko T. Visuri.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (PDF 921 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Visuri, M.T., Honkonen, K.M., Hartiala, P. et al. VEGF-C and VEGF-C156S in the pro-lymphangiogenic growth factor therapy of lymphedema: a large animal study. Angiogenesis 18, 313–326 (2015). https://doi.org/10.1007/s10456-015-9469-2

Download citation

Keywords

  • Lymphedema
  • Lymphangiogenesis
  • Angiogenesis
  • VEGF-C
  • VEGF-C156S