Microvascular Research 73 (2007) 214 – 223 www.elsevier.com/locate/ymvre
Lymphangiogenesis following obstruction of large postnodal lymphatics in sheep Azadeh Jila a , Harold Kim b , Vicky P.K.H Nguyen b , Daniel J. Dumont b , John Semple c,d , Dianna Armstrong a , Eva Seto a , Miles Johnston a,⁎ a
Neuroscience Research Program, Department of Laboratory Medicine and Pathobiology, Sunnybrook Health Sciences Centre, University of Toronto, 2075 Bayview Avenue, Toronto, Ontario, Canada M4N 3M5 b Department of Molecular and Cell Biology, Sunnybrook Health Sciences Centre, University of Toronto, 2075 Bayview Avenue, Toronto, Ontario, Canada M4N 3M5 c Advanced Regenerative Tissue Engineering Centre, Sunnybrook Health Sciences Centre, University of Toronto, 2075 Bayview Avenue, Toronto, Ontario, Canada M4N 3M5 d Department of Surgery, WomenTs College Hospital, University of Toronto, 76 Grenville Street, Toronto, Ontario, Canada M5S 1B2 Received 15 August 2006; accepted 9 November 2006 Available online 26 December 2006
Abstract We examined the impact of lymph flow obstruction in large post-nodal lymphatic vessels in sheep. A silk ligature was placed 2 cm downstream from the prescapular or popliteal lymph node and tightened to interrupt flow. At 6, 12 and 16 weeks after lymph flow blockage, a network of small interconnecting lymphatics (∼ 10–40 μm in diameter) could be observed in the vicinity of the ligature. These were identified using antibodies to the lymphatic endothelial markers LYVE-1 or VEGFR-3 or unequivocally, with the upstream intraluminal injection of the non-specific cell dye CFDA-SE. The observed lymphangiogenesis coincided with increased levels of Prox1, Tie2 (Y992) phosphorylation, MAPK activation, and decreased Akt activition. In the popliteal preparations, saline was infused into the prenodal ducts upstream of the regeneration site. The slopes of the inflow pressure versus flow relationships were 17.3 ± 3.6, immediately after vessel obstruction, 36.2 ± 9.6 at 6 weeks and 15.0 ± 5.3 at 12– 16 weeks. For comparison, the average slope in a completely intact popliteal system was 3.1 ± 0.3 (from a previous publication). The resistance to flow remained high up to 12–16 weeks after flow obstruction suggesting that normal flow parameters had not been achieved over this time. The lymph node appeared to have some role in limiting the impact of post-nodal lymph obstruction, a function that appeared to be compromised by lymph stasis. © 2006 Elsevier Inc. All rights reserved. Keywords: Lymphangiogenesis; Lymphatic vessels; Lymph nodes; Lymph flow obstruction
Introduction Lymphangiogenesis is generally a vigorous process but there are many situations in which lymph flow is compromised after lymphatic injury. The lymphedema associated with cancerrelated lymph node dissection is a good example (Armer et al., 2001; Rockson, 2001) and if left untreated, the edema can lead to recurring infections, impaired limb function, psychosocial problems and in extreme cases, malignant complications and
⁎ Corresponding author. Fax: +416 480 5737. E-mail address:
[email protected] (M. Johnston). 0026-2862/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.mvr.2006.11.003
life-threatening infections. There is considerable interest in reversing lymphatic dysfunction in lymphedema by applying appropriate molecular agents directly to affected tissues (Szuba et al., 2002) or by introducing the molecules through gene therapy approaches (Saaristo et al., 2002). While molecular therapies show theoretical promise, the enhancement of tissue drainage following pharmacologically induced lymphangiogenesis will provide numerous challenges. Many questions remain; in particular, we know very little about the physiological properties of the newly formed vessels and how the transport capabilities of regenerating ducts are integrated functionally into a lymphatic network that includes lymph nodes as well as pre- and post-nodal collectors.
A. Jila et al. / Microvascular Research 73 (2007) 214–223
While the general tendency in lymphangiogenesis research is to remove a block of tissue and therefore, destroy many lymphatic vessels, there is some merit in simplifying the model such that only one vessel is injured or obstructed. When individual lymphatic trunks are severed, ‘sprouts’ emerge from the severed ends and with contributions from the surrounding areas, eventually re-establish flow (Gray, 1939). In this regard, lymphatic vessels in sheep may provide unique opportunities to study lymphangiogenesis since the collecting ducts are relatively large and can be manipulated individually. The objective of this study was to determine if the obstruction of large postnodal lymphatic vessels in sheep would induce a consistent lymphangiogenic response and if so, to investigate the impact of new vessel formation on lymph transport over time. Materials and methods A total of 40 sheep were used in this investigation. All experiments outlined in this paper have been approved by the ethics committee at Sunnybrook Health Sciences Centre and conform to the guidelines set by the Canadian Council on Animal Care and the Animals for Research Act of Ontario. Sheep were anesthetized initially by I.V. injection of sodium pentothal. Subsequently, 2.0–3.5% isofluorane was delivered through an endotracheal tube via a Narkomed 2 respirator for surgical maintenance. The surgical area was shaved and prepped with alcohol and betadine. Temgesic was given post surgically and as required thereafter to treat post-surgical pain. Antibiotic (Duplocillin) was administered I.M. 1 day prior to, and 2 days after surgery.
Surgical interruption of lymph flow The lymphatics investigated in this report were large postnodal ducts and while the diameters of the vessels were quite variable, they approximated 1 mm for popliteals, 0.5–1.0 mm for prescapular and 1–2 mm for mesenteric vessels. To access the popliteal and prescapular vessels, an incision was made through the skin and the muscle was separated, exposing the node of interest. For access to the mesenteric lymphatic vessels, a laparotomy was performed and the mesenteric lymph node chain was observed to lie close to the ileal–cecal junction. To aid in the visualization of the post-nodal lymphatics, a 1% solution of Evans blue dye (in saline) was injected either into the subcapsular sinus of the upstream node or introduced in multiple sites into the drainage basin of the node. Within seconds the dye entered the postnodal ducts. To obstruct lymph flow, a 2-0 silk ligature was placed around the post-nodal vessel at least 2 cm downstream from the node and the ligature tightened to interrupt lymph flow. In all cases, the surgical sites were closed and the animals returned to their holding pens. The sheep were sacrificed 6, 12 or 16 weeks after surgery.
Evans blue studies In six sheep, the animals were anesthetized as outlined earlier and the site of obstruction was opened to visual inspection. Evans blue dye was injected into the subcapsular sinus of the upstream lymph node in an attempt to visualize newly formed vessels. The movement of dye past the ligature into the downstream postnodal segment was taken to indicate that some degree of fluid continuity had been restored.
Fluoroscopy In 3 sheep (6 limbs) between 8 and 9 weeks post ligation, a mobile fluoroscopy system was used (BV Pulsera, Philips) to visualize the lymphatic vessels at various times post-ligation. An X-ray contrast medium (10–30 ml,
215
Lipiodol, EZ-EM Canada, Therepex) was injected either into an upstream popliteal prenodal vessel or directed into the popliteal lymph node.
Immunohistochemistry A total of 15 sheep were used to assess lymphangiogenesis using two immunofluorescence staining methods (intact preparations—3 animals, 6 weeks post ligation—6 animals, 12 weeks—4 animals, 16 weeks—2 animals). In one technique, the vessel endothelium was stained with the non-specific cell dye CFDA-SE (Molecular Probes # C-1157). This provided unequivocal identification of the lymphatics since the dye was introduced directly into the lumens of the vessels through injection into the subcapsular sinus of the upstream lymph node. 5(6)-CFDA SE [5 (and 6)-carboxyfluorescein diacetate, succinimidyl ester] (molecular weight 557.47) diffuses passively into cells and remains nonfluorescent until its acetate groups are cleaved by intracellular esterases. It fluoresces green. Additionally, we stained the sections with antibodies to LYVE1 or VEGFR-3 (tagged with Cy3-red), molecular markers relatively specific for lymphatic endothelial cells. Tissue blocks were removed surgically. The samples were embedded in Tissue-Tek O.C.T compound, placed in a base mold and frozen immediately at −80 °C. Sections (7 μm thick) were cut using a Leica cryostat model Leica CM 3050S-3-1-1 and were placed on glass microscope slides. The slides were washed 5 times in phosphate buffered saline (PBS—0.1 M) and blocked for 1 h at room temperature with 10% goat serum in phosphate buffered saline (PBS). After washing with PBS, the sections were incubated overnight at 4 °C with 1:50 or 1:100 dilutions of rabbit, anti-human LYVE-1 primary antibody (Research Diagnostics Inc) or similar dilutions of rabbit anti-human VEGFR-3 (Research Diagnostics Inc). The next day, the sections were washed with PBS and incubated with 1:100 dilutions of goat, anti-rabbit IgG antibody tagged with Cy3 (Jackson ImmunoResearch). In controls the primary antibody to the molecular markers was omitted. Finally, the slides were mounted in aquapolymount and coverslipped. Immunofluorescence microscopy was performed with a Zeiss Axiovert 100 M laser scanning confocal microscope. The argon and helium/neon lasers were set to wavelengths of 488 and 543 nm for excitation of CFDA-SE and Cy3 respectively.
Molecular analysis Tissues were harvested from 3, 6-week post ligation prescapular preparations for molecular analysis. Comparisons were made with tissues extracted in similar areas in non-ligated preparations. The samples were frozen immediately in liquid Nitrogen and stored at − 80 °C before use. Samples were homogenized in RIPA lysis buffer for 30 min on ice (10 mM NaH2PO4 pH7.5, 150 mM NaCl, 1% NP-40, 0.1% SDS, 1% Sodium Deoxycholate, 10 mM NaF, 2 mM EDTA, protease inhibitor cocktail; Complete-EDTA free, Roche USA), cleared by centrifugation and the supernatants collected for further analysis. Alternatively, tissues were placed in Trizol (GibcoBRL) and processed following the manufacturers protocol. In brief, the homogenized tissues and 200 μl of chloroform were added to 1 ml Trizol. Following centrifugation at 10,000×g for 15 min at 4 °C, the upper phase was removed and 300 μl of 100% ethanol was added per 1 ml of Trizol. After 5 min incubation at room temperature, DNA was isolated by centrifugation at 2000×g for 5 min at 4 °C. Proteins were then precipitated from the phenol–ethanol supernatant by 1.5 ml isopropyl alcohol per 1 ml Trizol. After 10 min incubation at room temperature, the protein precipitate was isolated at 12,000×g for 10 min at 4 °C. This precipitate was washed 3 times in 95% ethanol and resuspended in 1% SDS. Equal amounts of protein from the ligated and control tissues were separated by SDS-PAGE. Proteins were transferred to polyvinylidene fluoride (PVDF, Perkin Ekmer) transfer membrane for immunoblotting. With the exception of phospho-Tie2 which was blocked in 7% BSA, all membranes were blocked in 3% milk/TBS for 60 min. The primary antibodies used for detection were phospho-Tie2 Y992 (Cell Signaling Technology USA); phospho-MAPK (Cell Signaling Technology USA, 9E10); phospho-Akt S473 (Cell Signaling Technology USA); Tie2 (Santa Cruz USA, C20); MAPK (Cell Signaling Technology USA); Akt (Cell Signaling Technology USA); Prox-1 (Upstate Biotech USA); and β-actin (Sigma USA). Proteins were visualized by enhanced chemiluminescence (Pierce USA).
216
A. Jila et al. / Microvascular Research 73 (2007) 214–223
Fig. 1. Schematic illustrating details of experimental protocol.
Infusion studies In order to obtain information on fluid conductance/resistance across the site of regeneration, we perfused the site and monitored inflow pressures in 11 animals. The popliteal lymphatic network provided several advantages for this approach. Unlike most others, the popliteal lymph node is one of the nodes in which a drainage basin exclusive to the node can be defined (lower limb). Another advantage of this system is that the pre- and post-nodal popliteal ducts are easy to identify within the drainage basin of the node and some of these are of sufficient size for cannulation. In sheep, usually only one vessel drains the node (in rare occasions there are 2). Multiple pre-nodal ducts (6–12) (Heath and Brandon, 1983) enter the popliteal lymph node at various locations along the convex portion of the node. Since there is no known collateral lymphatic circulation in this area, we expected that ligation of the post-nodal duct would obstruct all lymph flow from the lower hindlimb. Evans blue dye has long been the gold standard for the identification of lymphatic vessels in vivo. When injected subcutaneously, the dye binds to protein and readily enters absorbing lymphatics. It is distributed rapidly through the lymphatic network and outlines the collecting vessels clearly. A small amount of Evans blue dye (1.0% in saline) was injected into multiple sites of the hind limb just above the hoof to permit visualization of several of the popliteal prenodal ducts. One of these was dissected free of connective tissue and cannulated in the direction of flow using a 26-gauge angiocatheter. The angiocatheter was connected to a stopcock, which in turn was attached to a
syringe pump (Kd Scientific, model #260). The free arm of the stopcock was attached to a pressure transducer (Cobe CDX disposable) and pressure was recorded on a data acquisition system (A Tech Industries). The infusate was 0.9% saline or an ‘artificial lymph’. Lymph on average contains about 40% of the concentration of plasma proteins (Yoffey and Courtice, 1970). Therefore, heparinized autologous plasma was diluted with saline to achieve a similar protein concentration. Based on past experience (Kim et al., 2003), infusion rates were varied incrementally from 0.2 to 1.0 ml/h and the inflow pressures recorded continuously. The pressure data were captured at a rate of 1/s on a computer based data acquisition system. Pressures were monitored for a minimum of 3 min at each inflow rate. As perfusion rates increased, the inflow pressures also rose. A schematic illustrating the features of the experimental design is provided in Fig. 1. In most preparations, an equilibrium pressure had been attained after a few minutes of infusion. For analysis, the equilibrium pressures were averaged over 60 s and plotted against the flow rate. The slopes of the relationships (an estimate of resistance to flow) were determined from regression analysis. All data was expressed as the mean ± S.E.M. The data were analyzed with a 2 Factor ANOVA with Greenhouse–Geisser correction. We interpreted P < 0.05 as significant.
Results In 6 preliminary experiments Evans blue dye was injected into the subcapsular sinus of the upstream node in an attempt to find evidence of new vessel growth around the site of lymph flow blockages. In some popliteal and prescapular preparations we could observe small lymphatics in the area of the ligature that had taken up the dye. In several but not all cases, dye was observed downstream of the site of obstruction suggesting that some fluid was able to bypass the area of the ligature as early as 6 weeks after flow obstruction. We did not observe any evidence of regeneration in the obstructed mesenteric preparations. This may have been due to the complexity of the lymphatic circulation at this location. While the vessels are very
Fig. 2. Identification of lymphatic vessels. For unequivocal identification of these vessels, a non-specific cell dye CFDA-SE was infused into the subcapsular sinus of the upstream node. With confocal microscopy, the vessels were clearly visible having stained green (A—longitudinal, B—cross section). A small blood vessel (arrow) can be seen on the surface of the lymphatic in (A). When this approach was combined with anti-human LYVE-1 antibodies and goat, anti-rabbit IgG tagged with Cy3 (red) the stained sections appeared yellow-orange indicating co-existence of the specific lymphatic endothelial molecular marker and by virtue of its method of introduction, the non-endothelial specific cell dye CFDA-SE (C). In these examples, postnodal mesenteric lymphatics are illustrated.
A. Jila et al. / Microvascular Research 73 (2007) 214–223
accessible, the mesenteric nodes are packed together in a chain or in some cases, are fused to form several long nodes. It is possible that collateral vessels allow the lymph to bypass the impediment in a given vessel thereby reducing the stimulus to form new ducts at the site of obstruction. This phenomenon would be less likely to occur in popliteal and prescapular vessels since lymph from the appropriate catch basin funnels into single lymph nodes with one (usually) postnodal duct.
217
Immunohistochemistry At the commencement of this study, it was unclear whether the currently available antibodies for the established lymphatic endothelial molecular markers would be effective at elucidating lymphatics in sheep tissues. Furthermore, considering that lymphatic molecular markers in other species exhibit some degree of promiscuity we wanted to combine molecular marker
Fig. 3. Formation of new lymphatics after lymph flow obstruction. Confocal microscopy images of regenerated postnodal lymphatics. At 6 and 12 weeks after lymph flow obstruction, an irregular network of small interconnecting lymphatics could be observed in the vicinity of the ligature in the prescapular and popliteal preparations (A–D). The newly formed lymphatics were variable in size but generally ranged from 10 to 20 μm in diameter. Based on immunohistochemistry studies, the vessel networks did not appear to differ significantly between 6 and 12 weeks. However, at 16 weeks, the vessel networks appeared less dense and the few vessels observed were generally larger (E). In non-ligated ‘control’ preparations we either could not observe any lymphatic vessels or portions of the original single postnodal duct were visible (F). (A) Prescapular lymphatic vessel 6 weeks after ligation. CFDA-SE (green) and goat, anti-rabbit IgG (Cy3-red). Primary Ab to LYVE-1 omitted. (B) Prescapular lymphatic vessel 6 weeks after ligation. CFDA-SE, rabbit, anti-human LYVE-1 and goat, anti-rabbit IgG (Cy3). (C) Popliteal lymphatic vessel 12 weeks after ligation. CFDA-SE, rabbit, anti-human LYVE-1 and goat, anti-rabbit IgG (Cy3). (D) Prescapular lymphatic vessel 12 weeks after ligation. CFDA-SE, rabbit, anti-human LYVE-1 and goat, anti-rabbit IgG (Cy3). (E) Prescapular vessels 16 weeks after ligation. CFDA-SE, rabbit, anti-human LYVE-1 and goat, antirabbit IgG (Cy3). (F) Prescapular vessel (non-ligated popliteal vessel). CFDA-SE, rabbit, anti-human LYVE-1 and goat, anti-rabbit IgG (Cy3).
218
A. Jila et al. / Microvascular Research 73 (2007) 214–223
approaches with a method that provides unequivocal identification of these vessels. The latter was achieved using the nonspecific cell dye CFDA-SE. Due to the large size of sheep lymphatics and the ease by which their attendant lymph nodes could be identified, we were able to inject CFDA-SE into the subcapsular sinus of the upstream node. With confocal microscopy, the vessels were clearly visible having stained green (Figs. 2A, B). As well, we observed that sheep lymphatics stained for LYVE-1 and VEGFR-3, with the most consistent staining occurring with the antibodies to the lymphatic receptor for hyaluronan (LYVE-1). When both approaches were combined in the same tissue sample, the sections appeared yellow-orange. This indicated co-existence of the specific lymphatic endothelial molecular marker (secondary antibody tagged with Cy3-red) and by virtue of its method of introduction, the non-endothelial specific cell dye CFDA-SE (green). An example of staining with CFDA-SE and LYVE-1 is provided in Fig. 2C. Images with antibodies to VEGFR-3 are not shown. At 6 weeks (Figs. 3A, B) and 12 weeks (Figs. 3C, D) after lymph flow blockade, a labyrinth of small vessels was observed in the area surrounding the ligature in the prescapular and popliteal preparations. These appeared to be organized into a ‘chicken-wire’ network and in places one could see what appeared to be budding of endothelial cells from the main branches (arrow in Fig. 3B). The newly formed lymphatics were quite variable in size but generally ranged between 10 and 20 μm in diameter. A few animals were assessed at 16 weeks and at this time fewer vessels were observed. These ducts appeared to be on average slightly larger in size (20–40 μm) (Fig. 3E). When tissues were extracted in similar areas in nonligated preparations, small lymphatics were usually not visible although in a few tissue sections, a limited amount of staining was observed (Fig. 3F). This was likely due to the section encompassing parts of the original postnodal duct. Molecular assessment Consistent with the observation that active lymphangiogenesis was occurring at the site of ligation, an increase in the levels of Prox-1 was observed 6 weeks after lymph flow blockage compared to controls (Fig. 4A). Furthermore, an increase in the activation of the receptor Tie2 correlated with an increase in the activation of downstream pathways involved in endothelial cell proliferation such as MAPK (Figs. 4B, C). Interestingly, Akt activation appeared to decrease with Tie2 activation in regenerating samples (Fig. 4D). These results suggest that the observed increase in lymphangiogenesis coincided with the presence of a number of known molecular markers involved in lymphatic vessel development. Perfusion studies In an effort to assess the ease by which fluid could move across the site of regeneration, we perfused the popliteal system at variable flow rates and monitored the inflow pressures. To limit the number of animals used, in some cases, both limbs
Fig. 4. Molecular markers associated with regeneration were expressed in the area proximal to the ligature. A) An increase in the transcription factor Prox-1, B) as well as an increase in Tie2 activation, was observed in ligated samples (L) relative to non-ligated tissues (C). Consistent with an increase in mitogenesis associated with vessel regeneration, C) an increase in MAPK activity was found in post-ligated samples relative to non-ligated tissues. D) Akt activity was found to decrease in post-ligation preparations relative to non-ligated samples.
were perfused and considered to be separate experiments (see legend to Fig. 6). For a successful perfusion experiment, several criteria had to be met. First, pressures had to increase as flows were elevated. In several preparations, leakage of infusate prevented the pressure rise. Second, equilibrium pressures had to be attained for each rate of infusion. In a few studies with the diluted plasma, the infusate appeared to interact with the lymphatic vessels in unpredictable ways causing rapid increases and decreases in pressure and we were unable to establish equilibrium conditions. Since the pre-nodal lymphatics are contractile, we attributed this observation to plasma effects on lymphatic smooth muscle. Consequently, in most preparations we reverted to saline infusions as in our earlier study (Kim et al., 2003). In several cases, the pre-nodal lymphatics were unusually small and we had difficulty cannulating the duct. Perfusions were successful in 17 limbs (11 animals). Incremental saline or ‘artificial lymph’ infusions into the popliteal pre-nodal ducts at flow rates between 0.2 and 1.0 ml/h were accompanied by increasing inflow pressures. When the infusions were carried out immediately after ligation of the popliteal post-nodal duct, equilibrium pressures at a flow rate of 1.0 ml/h peaked at an average of 17.9 ± 3.5 cm H2O and the average slope of the pressure–flow relationships was 17.3 ± 3.6. An example is provided in Fig. 5A. It was of interest to note however, that equilibrium pressures were attained at each flow rate in the obstructed series indicating that the ‘blocked system’ was able to accommodate the inflow at least when challenged at relatively low rates of infusion. Otherwise, the pressures would
A. Jila et al. / Microvascular Research 73 (2007) 214–223
219
for flows between 0.2 and 1.0 ml/h and these have been included in Fig. 6B along with the averaged perfusion data obtained in this study. In completely intact preparations the average slopes were 3.1 ± 0.3 and the peak pressure at 1.0 ml/h infusion rate was 4.5 ± 0.5 cm H2O. Therefore, as expected,
Fig. 5. Examples of pressure versus flow relationships during perfusions of the popliteal lymphatic network. The arrows indicate incremental changes in the infusion rate. (A) Recording taken immediately after ligation of the post-nodal duct. (B) Recording taken 6 weeks after ligation. (C) Recording taken 12 weeks after ligation.
have continued to rise and no equilibrium would have been reached. At 6 weeks post ligation, we had evidence that new vessels were being formed and the Evans blue test indicated that a portion of the dye was able to pass the site of obstruction. Therefore, we were surprised to observe that the slopes of the pressure–flow relationships 6 weeks after ligation were higher (average of 36.2 ± 9.6) than immediately after obstruction and that the peak pressures at 1.0 ml/h were greater as well (37.4 ± 9.3). An example is provided in Fig. 5B. By 12–16 weeks the average slopes (15.0 ± 5.3) and average peak pressures (16.8 ± 4.9 cm H2O) had declined and reached levels similar to those measured immediately after flow obstruction. An example at 12 weeks is illustrated in Fig. 5C. That some fluid continuity had been restored in the obstructed vessels was confirmed with fluoroscopy. In the example illustrated in Fig. 6A taken at 9 weeks after ligation, the contrast agent was injected into the upstream popliteal lymph node and could be observed in the downstream postnodal duct. However, the vessels proximal to the node in the vicinity of the ligation site were very faint probably due to their small size and the limited resolution of the fluoroscopic procedure. In a previous study (Kim et al., 2003), we infused intact popliteal preparations (i.e. no ligation to obstruct flow) using identical methods to those employed here. We extracted the data
Fig. 6. (A) Visualization of lymphatic regeneration after postnodal obstruction using fluoroscopy (9 weeks post ligation). In this example, the contrast agent was able to pass through the lymph node and enter the downstream post-nodal duct. However, the connections between the node and the downstream vessel (white circle) are faint probably reflecting the small size of the regenerating ducts. (B) Averaged inflow pressure–flow data. Open circles—immediately after ligation (n = 6 limbs, 3 animals); closed triangles—6 weeks after ligation (n = 6 limbs, 4 animals); open triangles—12–16 weeks after ligation (n = 5 limbs, 4 animals); closed circles—for comparison, data obtained in non-ligated intact preparations has been included from an earlier publication (n = 14, 11 animals; Kim et al., 2003). The pressure flow relationships immediately after lymph flow obstruction and 12–16 weeks later were very similar. At 6 weeks, the slopes were higher. A repeated measures ANOVA with Greenhouse–Geisser correction indicated that the data from 6 weeks was not significant different from that collected immediately after flow obstruction or at 12–16 weeks. However, in one animal from the 6 week group the slope of the regression line was very low (4.8) compared to the others in this group (39.5, 45.4, 35.9, 18.4, 72.9). We cannot say whether this was due to more effective lymphangiogenesis and restoration of fluid continuity in this animal or if there was some unobservable leak in the system. In any event, with the one low value removed, we observed a significant interaction effect between the 6 week and immediately after ligation group (p = 0.0078) and the 12–16 week group (p = 0.011). * Data taken from Kim et al. (2003).
220
A. Jila et al. / Microvascular Research 73 (2007) 214–223
interruption of post-nodal lymph flow increased the resistance to flow and even over 12–16 weeks, on average normal flow parameters had not been restored. This meant, that over this period, higher pressures were required to establish a given flow rate. It should be noted however, that in two preparations at 16 weeks, the slopes were 6.5 and 4.2, which were reasonably close to values we obtained for infusions into completely intact preparations. This suggested that the integrity of the lymphatic system in some animals may have been re-established more quickly than in others. Discussion Lymphangiogenesis has been investigated for many years (Witte et al., 2001; Yoffey and Courtice, 1970). Most of the early experimental approaches have focused on the creation of lymphedema models. Commonly, a large cylinder of tissue is removed from a limb or ear retaining only the major blood vessels and nerves (Piller and Clodius, 1985). In a more modern variant of this method, a cylinder of tissue was removed from the mouse-tail and a collagen dermal equivalent inserted to bridge the gap between the proximal and distal wound margins (Swartz and Boardman, 2002; Boardman and Swartz, 2003). Histological assessment of molecular markers has been used to assess the presence or absence of lymphatic vessels with attempts to correlate these findings with the clinical status of the animal (Cunnick et al., 2001). A limitation of the commonly used methods to assess lymphangiogenesis is that the transport properties of the newly formed vessels are difficult to determine. The analysis of regeneration in larger vessels has the potential to provide additional opportunities to examine key functional parameters. In this regard, the pre- and post-nodal lymphatic vessels in sheep are very large and accessible. With this in mind, our objective was to develop a large vessel model of lymphangiogenesis that would in time permit quantitative assessment of several lymph transport parameters and help to elucidate how the various anatomical elements are integrated to facilitate lymph transport when vessel injury occurs. We observed that the interruption of lymph flow through the popliteal or prescapular sheep post-nodal lymphatic vessels resulted in the generation of a network of small lymphatics. We were able to identify these vessels unequivocally by introducing a non-specific cell dye (CFDA-SE) upstream into the lumens of the vessels, which stained the lymphatic endothelial cells. Additionally, the regenerating lymphatic network could be stained with antibodies to the lymphatic endothelial markers LYVE-1 or VEGFR-3. Furthermore, lymphatic markers such as the transcription factor Prox-1 and the tyrosine kinase Tie2 have also been found within the regenerating tissue. The transcription factor prox-1 is the earliest genetic marker involved in lymphatic development. Deletion of this gene in mice results in embryonic lethality due to multiple defects, however most significantly these embryos develop no lymphatics (Wigle and Oliver, 1999; Hong et al., 2002; Wigle et al., 2002). This is due to the inability of early lymphatic endothelial cells to commit to the lymphatic cell fate. To this end, it was proposed that prox-1 is involved in cell fate determination and
differentiation of lymphatic endothelial cells from blood endothelium. The fact that Prox-1 levels increased post-ligation and that vessels within the regeneration area stained with antibodies to LYVE-1 suggested an increase in lymphatic endothelial cells in the area of vessel obstruction (Figs. 3 and 4). Recently, knockout and knock-in mouse studies have suggested that the Tie2-Angiopoietin pathway plays a role in lymphatic development (Gale et al., 2002). Ablation of the Angiopoietin-2 gene results in a lymphatic defect characterized by the buildup of chylous ascites and the inability of the maturing lymphatic vasculature to coalesce into a functioning system. Genetically replacing Angiopoietin-2 with Angiopoietin-1 results in the rescue of the lymphatic defect suggesting that both ligands act as redundant agonists (Gale et al., 2002). Complementary studies in which Angiopoietin-1 was overexpressed both in vivo and in vitro were consistent with the notion that Angiopoietin-1 initiates lymphangiogenesis (Morisada et al., 2005; Tammela et al., 2005). Indeed, the regeneration observed in post-ligation samples in our study correlated with an increase in Tie2 receptor activity, increased MAPK activity (Harfouche et al., 2003; Kanda et al., 2005) and decreased Akt function. This inverse relationship between Akt and MAPK has been described previously. It has been suggested that Akt along with its role in cell survival might have a function in regulating mitogenesis. The activation of Akt has been found to negatively regulate the Ras pathway via Raf phosphorylation on serine residue 259, which upon phosphorylation associates with 14-33 protein resulting in Raf inactivation (Rommel et al., 1999; Zimmermann and Moelling, 1999). Conversely, a decrease in Akt activation may augment Raf activation thereby promoting mitogenesis. Therefore, crosstalk between proliferation and survival signals may modulate an optimal level of biological function. Indeed, our data suggests that vessel regeneration coincides with an increase in MAPK activity that is not negatively regulated by a concomitant elevation in Akt. Notably, Akt activity is not completely abolished in ligated samples suggesting that a low level of activity may be needed to sustain survival (Kontos et al., 1998; Kim et al., 2000a,b). While the specifics of how Akt levels are modulated by angiogenic growth factors during lymphangiogenesis are unclear, we have demonstrated recently that Dok2 (an adapter molecule that associates with a number of receptors including Tie2) has the ability to negatively influence Akt activity (Van Slyke et al., 2005). This correlation appears to be a general phenomenon of the Dok family of proteins (Wick et al., 2001) suggestive of a potential mechanism for Akt regulation mediated by Dok2. Preliminary data suggests that an increase in Dok2/DokR activation correlates with regeneration (data not shown). Taken together, these molecular events indicate that an active proliferative process is occurring after obstruction to lymph flow in postnodal lymphatic vessels. Presumably, the regeneration of new vessels represented an attempt by the host to re-establish fluid continuity across the site of obstruction. However, the inflow pressure versus flow relationships extracted from the perfusion studies indicated that a considerable impediment to flow existed at all times tested
A. Jila et al. / Microvascular Research 73 (2007) 214–223
after ligation. The resistance to flow afforded by the lymphatic collecting ducts is normally very low (Aukland and Reed, 1993). While lymph nodes are known to provide resistance to lymph transport and this has been estimated to be 50 to 200 times greater than that provided by the lymph trunks (Papp et al., 1971; Aukland and Reed, 1993), we expected that the nodal resistance would remain constant and that any change in flow patterns would be due to the functional status of the lymphatic vessels that were being formed to bypass the downstream site of obstruction. However, the data imply that the impact of vessel ligation on the ‘lymphatic system’ is complex and that some of our assumptions might have been incorrect. In some preparations we perfused the popliteal network immediately after blocking flow in the post-nodal duct leaving the node. We chose infusion rates that were similar to normal flows in these vessels based on our own experience (0.2–1 ml/h). At each rate of infusion, an equilibrium inflow pressure was attained. It is of course possible that pressures would have increased rapidly with no equilibrium reached if we had infused at a faster rate since the compensatory mechanisms will no doubt have some finite limit. However, under the flow levels tested, the system appeared to be capable of accommodating the inflow. Since the perfused pre-nodal ducts emptied directly into the node and all flow entering the node passed to the single post-nodal vessel, it became apparent that fluid must have been lost from the system otherwise pressures would have risen to much higher levels. The most likely site for fluid loss was in the lymph nodes. Lymph nodes have an extensive network of blood capillaries that are in close contact with lymph in transit through the nodes (Belisle and Sainte-Marie, 1990; Salvador et al., 1992). Since the colloid osmotic pressure of prenodal lymph is lower than that of blood (i.e. the protein content of lymph is considerably less than that of blood; Yoffey and Courtice, 1970), protein-free fluid is absorbed from the lymph into the blood capillaries in the nodes to establish an equilibrium of Starling forces across the lymph–blood barrier (Adair et al., 1982, Adair and Guyton, 1983, 1985). The loss of water into the nodal capillaries would likely be enhanced in our studies due to the nature of the infusate (in most experiments saline with no added protein). Studies in which artificial lymph (containing appropriate levels of protein) has been infused through nodes have indicated that the protein concentration of lymph can be increased from 40% to 400% after passage through the node. The loss of water into the nodal vasculature depends on the flow rate with slower flows allowing greater exchange. In our studies, the fact that 1) equilibrium pressures were observed for each induced flow rate and that 2) we did not observe any obvious edema in the limbs of lymph obstructed animals suggests that fluid loss from the nodes may have been quite effective in dampening the impact of downstream obstruction to lymph flow. In addition, while there was considerable variability in the infusion studies and further investigation is required, the data suggested that there was a greater impediment to flow while post-nodal lymphangiogenesis was occurring (6 weeks) than that observed immediately after the post-nodal duct was completely blocked. It is difficult to conceptualize how
221
lymphangiogenesis downstream of a lymph node would increase the impediment to flow under these conditions unless possibly, ligation of the post-nodal vessel altered the ability of the node to remove water. One might hypothesize that downstream lymph obstruction would cause lymph stasis and possibly lymph coagulation within the node. Without a constant movement of lymph through the sinuses of the node, the potential for water loss into the capillaries would be reduced. It is also possible that the outflow blockage would increase the lymph pressure within the node and secondarily reduce the blood flow through the tissue. By limiting vascular perfusion, there would be less opportunity for water loss into the blood. In order to assess the conductance/resistance of the regenerating site, one would have to bypass the node and perfuse the site directly. Due to the relatively large size of the post-nodal vessel it is possible to perform this experiment although great care would have to be taken during the cannulation of the upstream portion of the vessel to avoid damaging the newly formed and presumably fragile ducts. Over 12–16 weeks, the average slopes of the pressure–flow relationships declined compared to the levels observed at 6 weeks. This may have been due to the generation of new downstream lymphatics or the re-opening of the sinuses in the node. Additionally, when lymph flow is obstructed chronically, there is also evidence to suggest that interstitial protein is diverted directly into the local vasculature through endogenously formed lymphaticovenous communications (Aboul-Enein et al., 1984; Pain et al., 2004). Whatever the nature of the compensatory changes occurring after lymph flow obstruction, average values for the slopes of the pressure–flow relationships at 12–16 weeks never fell below those observed immediately after vessel ligation. Therefore, even though lymphangiogenesis was clearly occurring and fluoroscopy studies indicated some degree of pre- and postligation fluid continuity, the transport properties on average were not improving at least with regards to the induced flow method of analysis. We did observe that the slopes of the pressure–flow relationships had declined in two preparations to levels that approximated those in completely intact non-ligated systems (6.5 and 4.3). This suggested in these animals at least, that changes in the lymphatic network as noted above helped to re-establish relatively normal flow. In this regard, it will be important to assess the accommodation of the lymphatic network using other methods that may provide a more sensitive measure of lymphatic performance. In one such technique, radioactive albumin could be infused into the popliteal pre-nodal ducts. From the appearance of the tracer in plasma, the mass transport rate of the protein could be quantified at various times after lymph flow obstruction. In summary, the data suggest the following: 1) In complex lymphatic systems (mesenteric for example), post-nodal lymph flow obstruction did not result in the formation of new ducts perhaps as proximal collateral vessels were capable of maintaining adequate tissue drainage. However, post-nodal lymph flow obstruction in collecting lymphatic vessels in which collaterals are few or
222
A. Jila et al. / Microvascular Research 73 (2007) 214–223
absent (popliteal and prescapular) gave rise to the formation of a network of new ducts presumably in an effort to reestablish flow across the area of blockage. 2) These vessels could be identified as lymphatics unequivocally by infusing a non-specific cell dye (CFDA-SE) into the lumen of the upstream duct. The newly regenerating vessels also express the lymphatic markers LYVE-1 and VEGFR-3. Regeneration also correlated with changes in Prox-1 levels, Tie2, MAPK and Akt activity. 3) Standard methods used to assess the effectiveness of lymphangiogenesis such as fluoroscopy can be misleading since they may show an intact lymphatic network whereas more sensitive methods of analysis illustrate the persistence of a lymph transport deficit. Perfusion studies indicated that an increased resistance to induced flow persisted from the time of flow obstruction to 12–16 weeks. Therefore, normal flow parameters had not been restored over this period. 4) The lymph nodes appeared to dampen the impact of postnodal lymph flow obstruction probably by the loss of water into the nodal vasculature. This function appeared to be compromised by lymph stasis. Studies of lymphangiogenesis following lymph flow obstruction in sheep could provide new opportunities to investigate the physiological properties of regenerating lymphatic vessels and help to determine how the newly formed vessels are assimilated into the lymph transport network over time. In particular, the interaction and integration of lymphatics and lymph nodes in response to flow obstruction would seem to warrant further investigation. Acknowledgments This work was funded by grants from the Canadian Institutes of Health Research/Canadian Breast Cancer Research Alliance operating grant and from the Advanced Regenerative Tissue Engineering Centre (ARTEC), Sunnybrook Health Sciences Centre. We also wish to thank M. Katic (Department of Research Design and Biostatistics, Sunnybrook Health Sciences Centre) for assistance in the computational analyses of the data. References Aboul-Enein, A., Eshmawy, I., Arafa, S., Abboud, A., 1984. The role of lymphovenous communication in the development of postmastectomy lymphedema. Surgery 95, 562–566. Adair, T.H., Guyton, A.C., 1983. Modification of lymph by lymph nodes. II. Effect of increased lymph node venous blood pressure. Am. J. Physiol. 245, H616–H622. Adair, T.H., Guyton, A.C., 1985. Modification of lymph by lymph nodes. III. Effect of increased lymph hydrostatic pressure. Am. J. Physiol. 249, H777–H782. Adair, T.H., Moffatt, D.S., Paulsen, A.W., Guyton, A.C., 1982. Quantitation of changes in lymph protein concentration during lymph node transit. Am. J. Physiol. 243, H351–H359. Armer, J.M., Heppner, P.P., Mallinckrodt, B., 2001. The Secret epidemic: lymphedema. J. Women's Cancer 3, 35–41. Aukland, K., Reed, R.K., 1993. Interstitial–lymphatic mechanisms in the control of extracellular fluid volume. Physiol. Rev. 73, 1–78.
Belisle, C., Sainte-Marie, G., 1990. Blood vascular network of the rat lymph node: tridimensional studies by light and scanning electron microscopy. Am. J. Anat. 189, 111–126. Boardman, K.C., Swartz, M.A., 2003. Interstitial flow as a guide for lymphangiogenesis. Circ. Res. 92, 801–808. Cunnick, G.H., Jiang, W.G., Gomez, K.F., Mansel, R.E., 2001. Lymphangiogenesis quantification using quantitative PCR and breast cancer as a model. Biochem. Biophys. Res. Commun. 288, 1043–1046. Gale, N.W., Thurston, G., Hackett, S.F., Renard, R., Wang, Q., McClain, J., Martin, C., Witte, C., Witte, M.H., Jackson, D., Suri, C., Campochiaro, P.A., Wiegand, S.J., Yancopoulos, G.D., 2002. Angiopoietin-2 is required for postnatal angiogenesis and lymphatic patterning, and only the latter role is rescued by Angiopoietin-1. Dev. Cell 3, 411–423. Gray, J.H., 1939. Studies of the regeneration of the lymphatic vessels. J. Anat. 74, 309–335. Harfouche, R., Gratton, J.P., Yancopoulos, G.D., Noseda, M., Karsan, A., Hussain, S.N., 2003. Angiopoietin-1 activates both anti- and proapoptotic mitogen-activated protein kinases. FASEB J. 17, 1523–1525. Heath, T., Brandon, R., 1983. Lymphatic and blood vessels of the popliteal node in sheep. Anat. Rec. 207, 461–472. Hong, Y.K., Harvey, N., Noh, Y.H., Schacht, V., Hirakawa, S., Detmar, M., Oliver, G., 2002. Prox1 is a master control gene in the program specifying lymphatic endothelial cell fate. Dev. Dyn. 225, 351–357. Kanda, S., Miyata, Y., Mochizuki, Y., Matsuyama, T., Kanetake, H., 2005. Angiopoietin 1 is mitogenic for cultured endothelial cells. Cancer Res. 65, 6820–6827. Kim, I., Kim, H.G., So, J.N., Kim, J.H., Kwak, H.J., Koh, G.Y., 2000a. Angiopoietin-1 regulates endothelial cell survival through the phosphatidylinositol 3′-Kinase/Akt signal transduction pathway. Circ. Res. 86, 24–29. Kim, I., Kim, J.H., Moon, S.O., Kwak, H.J., 2000b. Angiopoietin-2 at high concentration can enhance endothelial cell survival through the phosphatidylinositol 3′-Kinase/Akt signal transduction pathway. Oncogene 19, 4549–4552. Kim, C., Li, B., Papaiconomou, C., Zakharov, A., Johnston, M., 2003. Functional impact of lymphangiogenesis on fluid transport after lymph node excision. Lymphology 36, 111–119. Kontos, C.D., Stauffer, T.P., Yang, W.P., York, J.D., Huang, L., Blanar, M.A., Meyer, T., Peters, K.G., 1998. Tyrosine 1101 of Tie2 is the major site of association of p85 and is required for activation of phosphatidylinositol 3kinase and Akt. Mol. Cell. Biol. 18, 4131–4140. Morisada, T., Oike, Y., Yamada, Y., Urano, T., Akao, M., Kubota, Y., Maekawa, H., Kimura, Y., Ohmura, M., Miyamoto, T., Nozawa, S., Koh, G.Y., Alitalo, K., Suda, T., 2005. Angiopoietin-1 promotes LYVE-1-positive lymphatic vessel formation. Blood 105, 4649–4656. Pain, S.J., Barber, R.W., Ballinger, J.R., Solanki, C.K., Mortimer, P.S., Purushotham, A.D., Peters, A.M., 2004. Local vascular access of radioprotein injected subcutaneously in healthy subjects and patients with breast cancer-related lymphedema. J. Nucl. Med. 45, 789–796. Papp, M., Makara, G.B., Hajtman, B., 1971. The resistance of in situ perfused lymph trunks and lymph nodes to flow. Experientia 27, 391–392. Piller, N.B., Clodius, L., 1985. Experimental Lymphoedema: Its Applicability and Contribution to our Clinical Understanding. Elsevier Science Publishers, Amsterdam, pp. 189–230. Rockson, S.G., 2001. Lymphedema. Am. J. Med. 110, 288–295. Rommel, C., Clarke, B.A., Zimmermann, S., Nunez, L., Rossman, R., Reid, K., Moelling, K., Yancopoulos, G.D., Glass, D.J., 1999. Differentiation stagespecific inhibition of the Raf-MEK-ERK pathway by Akt. Science 286, 1738–1741. Saaristo, A., Veikkola, T., Tammela, T., Enholm, B., Karkkainen, M.J., Pajusola, K., Bueler, H., Yla-Herttuala, S., Alitalo, K., 2002. Lymphangiogenic gene therapy with minimal blood vascular side effects. J. Exp. Med. 196, 719–730. Salvador, A.C., Pereira, A.S., Grande, N.R., 1992. Microanatomy of the blood vasculature of lymph node follicles in the dog. Lymphology 25, 37–41. Swartz, M.A., Boardman Jr., K.C., 2002. The role of interstitial stress in lymphatic function and lymphangiogenesis. Ann. N. Y. Acad. Sci. 979, 197–210. Szuba, A., Skobe, M., Karkkainen, M.J., Shin, W.S., Beynet, D.P., Rockson,
A. Jila et al. / Microvascular Research 73 (2007) 214–223 N.B., Dakhil, N., Spilman, S., Goris, M.L., Strauss, H.W., Quertermous, T., Alitalo, K., Rockson, S.G., 2002. Therapeutic lymphangiogenesis with human recombinant VEGF-C. FASEB J. 16, 1985–1987. Tammela, T., Saaristo, A., Lohela, M., Morisada, T., Tornberg, J., Norrmen, C., Oike, Y., Pajusola, K., Thurston, G., Suda, T., Yla-Herttuala, S., Alitalo, K., 2005. Angiopoietin-1 promotes lymphatic sprouting and hyperplasia. Blood 105, 4642–4648. Van Slyke, P., Coll, M.L., Master, Z., Kim, H., Filmus, J., Dumont, D.J., 2005. Dok-R mediates attenuation of epidermal growth factor-dependent mitogenactivated protein kinase and Akt activation through processive recruitment of c-Src and Csk. Mol. Cell. Biol. 25, 3831–3841. Wick, M.J., Dong, L.Q., Hu, D., Langlais, P., Liu, F., 2001. Insulin receptormediated p62dok tyrosine phosphorylation at residues 362 and 398 plays distinct roles for binding GTPase-activating protein and Nck and is essential
223
for inhibiting insulin-stimulated activation of Ras and Akt. J. Biol. Chem. 276, 42843–42850. Wigle, J.T., Oliver, G., 1999. Prox1 function is required for the development of the murine lymphatic system. Cell 98, 769–778. Wigle, J.T., Harvey, N., Detmar, M., Lagutina, I., Grosveld, G., Gunn, M.D., Jackson, D.G., Oliver, G., 2002. An essential role for Prox1 in the induction of the lymphatic endothelial cell phenotype. EMBO J. 21, 1505–1513. Witte, M.H., Bernas, M.J., Martin, C.P., Witte, C.L., 2001. Lymphangiogenesis and lymphangiodysplasia: from molecular to clinical lymphology. Microsc. Res. Tech. 55, 122–145. Yoffey, J.M., Courtice, F.C., 1970. Lymphatics, Lymph and the Lymphomyeloid Complex. Academic Press, London. Zimmermann, S., Moelling, K., 1999. Phosphorylation and regulation of Raf by Akt (protein kinase B). Science 286, 1741–1744.
