VEGF isoforms have differential effects on permeability of human pulmonary microvascular endothelial cells
- 1.3k Downloads
Alternative splicing of Vascular endothelial growth factor-A mRNA transcripts (commonly referred as VEGF) leads to the generation of functionally differing isoforms, the relative amounts of which have potentially significant physiological outcomes in conditions such as acute respiratory distress syndrome (ARDS). The effect of such isoforms on pulmonary vascular permeability is unknown. We hypothesised that VEGF165a and VEGF165b isoforms would have differing effects on pulmonary vascular permeability caused by differential activation of intercellular signal transduction pathways.
To test this hypothesis we investigated the physiological effect of VEGF165a and VEGF165b on Human Pulmonary Microvascular Endothelial Cell (HPMEC) permeability using three different methods: trans-endothelial electrical resistance (TEER), Electric cell-substrate impedance sensing (ECIS) and FITC-BSA passage. In addition, potential downstream signalling pathways of the VEGF isoforms were investigated by Western blotting and the use of specific signalling inhibitors.
VEGF165a increased HPMEC permeability using all three methods (paracellular and transcellular) and led to associated VE-cadherin and actin stress fibre changes. In contrast, VEGF165b decreased paracellular permeability and did not induce changes in VE-cadherin cell distribution. Furthermore, VEGF165a and VEGF165b had differing effects on both the phosphorylation of VEGF receptors and downstream signalling proteins pMEK, p42/44MAPK, p38 MAPK, pAKT and peNOS. Interestingly specific inhibition of the pMEK, p38 MAPK, PI3 kinase and eNOS pathways blocked the effects of both VEGF165a and VEGF165b on paracellular permeability and the effect of VEGF165a on proliferation/migration, suggesting that this difference in cellular response is mediated by an as yet unidentified signalling pathway(s).
This study demonstrates that the novel isoform VEGF165a and VEGF165b induce differing effects on permeability in pulmonary microvascular endothelial cells.
KeywordsVascular permeability Vascular endothelial growth factor (VEGF) Cell signalling
Alveolar capillary membrane
AKT8 virus oncogene cellular homolog
Analysis of variance
Acute respiratory distress syndrome
Bovine serum albumin
Electrical cell impedance system
Endothelial nitric oxide synthase
Human pulmonary microvascular endothelial cell
Human umbilical vein endothelial cell
NG-Nitro-L-arginine Methyl Ester
Mitogen-activated protein kinase
Mitogen/extracellular signal-regulated kinase
p38 mitogen-activated protein kinase
Sodium dodecyl sulphate polyacrylamide gel electrophoresis
Standard error of the mean
V-scr sarcoma (Schimidt-Ruppin A-2) viral oncogene homolog
Transendothelial electrical resistance
United States of America
Vascular endothelial cadherin
Vascular endothelial growth factor
Vascular endothelial growth factor –receptor 1
Vascular endothelial growth factor –receptor 2
VEGF was originally identified by its properties as both a permogen and a mitogen, key elements in the function of the alveolar-capillary membrane, leading to interest in its role in many forms of lung disease particularly ARDS [1, 2, 3]. We and others found that VEGF levels were compartmentalised between the alveolar space and the vascular bed [4, 5]. Low levels of intrapulmonary VEGF were found in patients with ARDS with increasing intrapulmonary VEGF levels associated with recovery . In contrast, plasma levels in patients with ARDS were elevated compared with normal, at-risk, or ventilated control subjects, with falling levels associated with recovery . These data suggest that VEGF is beneficial in the alveolar space but detrimental in the vascular space. To explore the significance of these observations, it is necessary to understand the mechanisms that regulate VEGF bioactivity. VEGF exerts its biological effect through specific receptors, VEGF-R1 and VEGF-R2 and co-receptors, neuropilin-1 and neuropilin-2 . In addition, alternative splicing of VEGF transcripts leads to the generation of several functionally different isoforms [8, 9]. We have previously explored changes in VEGFxxx-isoforms and receptor expression as mechanisms for regulating VEGF bioactivity and suggested that both these factors may contribute  but do not fully explain the reported contradictory findings. The VEGFxxxb isoform family consists of peptides of the same length as other forms but with a different C-terminal six amino acids-SLTRKD rather than CDKPRR . The receptor binding and dimerisation domains are intact, but VEGFxxxb stimulates a unique pattern of VEGF-R2 tyrosine residue phosphorylation, contrasting with those activated by conventional isoforms . Two specific isoforms, VEGF165a and VEGF165b isoforms were shown to have contrasting effects on the epithelial and endothelial sides of the alveolar-capillary membrane . These data suggest a pneumotropic effect which could be beneficial within the alveolar space following ARDS. However, the effect of these isoforms on vascular permeability another key element of ARDS is unknown.
We hypothesised that VEGF165a and VEGF165b activate different signalling pathways mediating cell permeability, a potential explanation for the conflicting observations on effects in the vascular space. To explore this theory, we used three methods of assessing vascular barrier function and found contrasting effects with VEGF165a increasing permeability and VEGF165b decreasing permeability. We then explored the relationship of downstream pathways to these functional differences. We compared the effects of specific signalling pathway inhibitors of MEK/p38MAPK/PI3K and eNOS on permeability, cell migration and proliferation to identify a mechanism by which increased permeability could be resolved whilst maintaining beneficial cell proliferation and migration.
A detailed description of materials and methods is given in the online data supplement.
Primary cell culture
Human Pulmonary microvascular endothelial cell (HPMEC) cryopreserved from passage 2 (PromoCell, Heidelberg, Germany) were cultured in endothelial cell basal medium MV2 (C-22221, PromoCell, Germany) complemented with supplement pack (C-39221, PromoCell, Germany) according to manufacturer’s instructions.
For all experiments cells were grown to 80% confluence, quiesced (MV2 media only) and stimulated with combinations of VEGF165a and VEGF165b (20 ng/ml as considered physiologically relevant in circulating plasma) [4, 6] in the presence or absence of specific signalling pathway inhibitors (U0126, SB203580, LY294002 (Cell Signalling, UK) or L-NAME (Calbiochem, UK).
Measurement of TEER by Endohm
Measurement of trans-endothelial electrical resistance (TEER) of the cell monolayer was performed using an Endohm 12 electrode chamber and an endothelial volt/ohm meter EVOM2 (World precision Instruments, USA) as previously described by Bevan and al .
Cells were plated at 20000 cells/cm2 into 8-well arrays (8W10E+; Wolf laboratories Ltd). Data was automatically and continuously collected every 2 min and recorded by computer. Experiments were performed after cells reached confluence with basal TEER values > 1500 Hz.
Transendothelial permeability to macromolecules was assessed by the passage of FITC-conjugated BSA (relative molecular mass 66,000) across cell monolayers in tissue culture inserts as previously described .
Scratch assay (Migration and proliferation)
Cells were seeded with 100 μl of cell suspension (5 × 105 cells/ml) in an Ibidi culture-chamber (Ibidi GmbH Munich, Germany). Cells were pre-incubated with or without inhibitor for 1 hour before removal of the chamber. Cells were then incubated in MV2 medium alone or MV2 medium with 20 ng/ml of recombinant protein VEGF165a or VEGF165b. Images were captured and analysed at 0 and 24 h.
Western blotting analysis
Cell lysates were separated on sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotted. Blots were blocked with 5% bovine serum albumin (BSA) (Fischer Scientific UK,) and incubated overnight at 4 °C with primary antibodies
HPMEC were stimulated with 100 ng/ml of VEGF165a, VEGF165b, VEGF165a + b or without any stimulation (control) for 10 min. They were then fixed, permeabilised and immunostained for VE-cadherin (Sigma, UK) and Alexa Fluor® 568 Phalloidin (Invitrogen, UK) for staining actin structures.
VEGF165a increases and VEGF165b decreases permeability in HPMEC
ECIS was utilised as another method to evaluate the effect of VEGF isoforms on cell permeability. A similar response to that seen using the Endohm system was observed for HPMEC. VEGF165a induced an increase in cell monolayer permeability (***p < 0.001) in contrast to VEGF165b which induced a significant decrease (*p < 0.05) in the cell monolayer permeability (Fig. 1b).
The last experimental technique used to assess permeability was FITC-coupled BSA passage. The permeability of the HPMEC monolayer for FITC-BSA was monitored every 15 min up to 90 min by measuring the fluorescence intensity of the medium in the lower compartment. For unstimulated control cells, the level of BSA in the lower compartment increased slowly over time in a comparable manner to those treated with VEGF165b; no significant difference was observed between them in HPMEC (Fig. 1c). When the cells were cultured in the presence of VEGF165a, a significant (*p < 0.05) time-dependant increase in BSA permeability was detected after 45 min and persisted for at least 90 min, suggesting an increase in the cell monolayer permeability.
VEGF165a induces changes in VE-cadherin distribution pattern and actin stress fibres in HPMEC compared to VEGF165b
VEGF-R2 has been reported to induce most downstream signalling effects through the tyrosine sites tyr1175 and tyr1214 [21, 22]. These two tyrosine sites play a crucial and direct role in the recruitment of adaptor proteins that activate multiple signalling pathways such as proliferation, survival, migration and permeability [23, 24]. Therefore, we studied VEGF isoform induced phosphorylation of those tyrosine sites in HPMECs.
VEGF165a and VEGF165b induce differential phosphorylation of the VEGF receptors at tyrosine 1175 and 1214 in HPMEC
Having identified these phosphorylation differences we wanted to identify specific pathways leading to cell permeability which is less well characterised than those of other VEGF functional effects . Further Western blotting of HPMEC was undertaken, to investigate the effect of VEGF isoforms on the phosphorylation of several proteins previously suggested to be involved in specific functional downstream cellular effects. In parallel, we continued to use the Endohm assay to assess paracellular permeability and utilised the scratch assay combining both migration and proliferation processes [25, 26] to assess the specificity of the functional effect of individual signalling pathway inhibitors.
Specific inhibition of pMEK1/2, p42/44MAPK and p38MAPK induction by VEGF165a and VEGF 165b in HPMEC does not have a differential effect on permeability and migration/proliferation pathways
The MEK and mitogen activated protein kinase (MAPK) pathways are among the most widely studied in VEGF biology and considered to have critical roles in cell proliferation and cell growth and differentiation [27, 28]. We initially sought to determine whether inhibition of these proteins would have a specific functional effect i.e. inhibit proliferation/migration but not permeability as previously suggested .
U0126 is a compound reported to be a highly selective inhibitor of MEK1 and MEK2 and utilised to block the classical MAPK cascade in cells that leads to cell proliferation . U0126 inhibited the permeability effects of both VEGF165a and VEGF165b on HPMEC (Fig. 4c and d respectively). VEGF165a, but not VEGF165b, induced a significant increase in the cell migration into the “scratch” (p = 0.004) in comparison to control (Fig. 4e) that was inhibited byU0126. These data suggest that the MEK pathway is activated by both VEGF165a and VEGF 165b but any functional divergence must occur downstream.
Having identified significant differences in the activation kinetics of MEK, p42/44MAPK and p38MAPK by VEGF165a compared to VEGF165b (that reflected changes in permeability but not specific to it alone), we then looked at the effect of VEGF isoforms on AKT also known as protein kinase B (PKB) protein. AKT has been shown to promote the proposed cell survival pathway mediated by the activation of PI3-kinase protein kinase  but it has been also associated with eNOS production which is closely associated with permeability .
Specific inhibition of AKT induction by VEGF165a and VEGF165b in HPMEC does not have a differential effect on permeability and survival pathways
VEGF165b induces a rapid and robust phosphorylation of pAKT at 5 min and 10 min (p < 0.01) in HPMEC with maximal AKT phosphorylation detected at 5 min (Fig. 5e). Stimulation with VEGF165a also induced a significant phosphorylation of AKT at 5 (p < 0.05) and 10 min (p < 0.01) and returned to control levels at 60 min. LY294002 compounds is an inhibitor of phosphatidylinositol 3 kinase (PI3K) and inactivation of PI3K have been reported to lead to dephosphorylation of Akt that subsequently stop G1 cycle in cell growth and ultimately lead to cell apoptosis [34, 35]. Co-culture with LY294002 to block AKT protein inhibited both the permeability effects of VEGF165a and VEGF165b on HPMEC (Fig. 5 f and g). In contrast with the permeability assay where the cell monolayer was stable over the experiment (2 h), inhibition of AKT for long periods (24 h) induced cell death so we were unable to undertake the scratch assay as a migration/proliferation model.
Finally, we investigated the potential of the eNOS pathway for differential effects. There is a significant body of evidence to suggest that endothelial nitric oxide (synthesised by eNOS) may have a crucial role in causing hyperpermeability in response to pro-inflammatory agents such as VEGF [36, 37]. Therefore, the effects of the different VEGF isoforms on the eNOS phosphorylation in HPMEC were studied.
Specific inhibition of eNOS induction by VEGF165a and VEGF165b in HPMEC does not have a differential effect on permeability and nitric oxide signalling pathways
The pulmonary endothelium is crucial to the regulation of the passage of solutes and molecules between the blood and the interstitial space of the lung, enabling close proximity of the vascular bed to the alveolar space for gaseous exchange to occur. Despite this, there is a very limited understanding of the mechanisms involved in the regulation of pulmonary endothelial cell (EC) barrier function integrity, which is so essential for maintaining this critical function of the lung.
Vascular endothelial growth factor (VEGF) was originally described as both an angiogenic and a permeability factor  and its effects have previously been studied using the human umbilical vein endothelial cell (HUVEC) as the archetypal EC. Large organ functional differences are reflected in the variability of endothelial cell junction structure and composition particularly relevant in the functional differences between the pulmonary and systemic circulation [40, 41]. To explore our hypothesis and its relationship to previous clinical studies (5, 6) it was important to study the response of human pulmonary microvascular endothelial cell (HPMEC) to VEGF isoforms.
Among all the pathological processes involved in. ARDS increases in lung vessel permeability are critical and non-redundant . The measurement of permeability in this study has been undertaken only in-vitro models with self-evident limitations . Transport of plasma proteins, cell and solutes across monolayers occurs paracellularly via specialised endothelial cell-cell junctions, or transcellularly by special transport mechanism including transcytosis, via transcellular channels or cell membrane transporter proteins.
Two types of inter-endothelial junction are present in the endothelium, adherens and tight junctions, the former being dominant in most vascular beds. The integrity of the adherens junction is particularly critical for regulating paracellular permeability via homophilic adhesions between VE-cadherin molecules [19, 20]. Disruptions of these domains lead to downstream events that result in organisational changes in the actin cytoskeleton . The transcellular pathway is responsible for the transport of larger molecules such as albumin across endothelial cell monolayers, classically via transcellular pores associated with caveolae and lipid rafts . Traditionally these pathways have been considered independent but there is now a body of evidence showing interdependence . Following on from this, the in-vitro methods of measuring permeability that we have used, TEER (thought to reflect only paracellular permeability) and FITC-BSA (thought to only reflect transcellular mechanism) are recognised to have influences from crosstalk between both pathways .
We have demonstrated for the first time that VEGF165a and VEGF165b induce differing effects on the permeability of pulmonary microvascular endothelial cells. Specifically, VEGF165a induced an increase and VEGF165b a decrease in permeability. The receptor binding and dimerisation domains are intact in the VEGFxxxb family of VEGF isoforms. However, in porcine aortic endothelial cells, VEGF165b has been shown to stimulate a unique pattern of VEGF-R2 tyrosine residue phosphorylation, contrasting with those activated by conventional isoforms suggesting activation of differing downstream signalling pathways in addition to partial agonist activity and changes in neuroplin-1 binding [22, 47, 48]. In this study, differing phosphorylation kinetics were clearly observed following stimulation by VEGF165a and VEGF165b using what we considered to be physiologically relevant concentrations of VEGF.
The differential effects of VEGF165a and VEGF165b on the vascular permeability in addition to those we have previously shown and published on proliferation in HPMEC (also repeated in the scratch experiments) and human alveolar epithelial cells offer a potential paradigm to explain the apparent compartmentalisation of VEGF between the alveolar and vascular space and the apparent disparity of data relating to the role of VEGF in ARDS [5, 6, 12].
We identified that VEGF165a and VEGF165b lead to differential functional outcomes with VEGF165a increasing cell permeability in methods suggested to reflect both para and transcellular permeability and VEGF165b reducing paracellular permeability only. This suggested that the differences were due to divergence of signalling pathways and therefore potential targets for amelioration of outcome e.g. reducing permeability whilst preserving a pneumotropic effect. To verify this hypothesis, different protein inhibitors have been used to look at their effect on the change in resistance reflecting the paracellular permeability pathway of pulmonary microvascular ECs.
The VEGF165a signalling pathways have been studied extensively in HUVEC although studies in HPMEC are limited . We chose to use inhibitors of pMEK, P38 MAPK, PI3 kinase and eNOS proteins as these proteins have been suggested to be involved in VEGF signalling pathways and look at both permeability and proliferation/migration in an attempt to identify a divergence in functionality and thus an opportunity for selective inhibition.
The use of L-NAME (eNOS inhibitor) and LY294002 (PI3K inhibitor) on HPMEC inhibited the effect of both VEGF165a and VEGF165b. Being part of the same signalling pathway these results suggest that VEGF cell paracellular permeability involves the phosphoinositide 3-kinase–AKT pathway, which then further phosphorylates and activates endothelial nitric oxide synthase (eNOS) . Also, the inhibition of pMEK and p38 MAPK did not affect VEGF isoforms activity on the proliferation and paracellular permeability pathway in HPMEC. In summary, the inhibitors chosen did not allow for the identification of specific differential pathways between VEGF165a and VEGF165b. Further studies of other pathways are required in order to unravel the molecules responsible for the differential permeability effects of VEGF isoforms such as Src kinase pathway and its role in the regulation of endothelial-barrier integrity as demonstrated recently by Gao et al. .
HPMEC have specific signalling characteristics that are probably adaptations to the unique pulmonary environment. In the context of ARDS, VEGF (from the alveolus) may be a triggering component leading to further endothelial dysfunction and failure of the ACM. The data presented has shown that exposure of HPMEC to exogenous VEGF165a significantly increased endothelial cell monolayer permeability which would lead to failure of the ACM barrier function, of particular significance in the lung environment. On the other hand, depending on the isoforms present, VEGF may also be a protective agent, the data presented showing a decrease in permeability with VEGF165b exposure. Further work will need to be undertaken to clearly identify the divergence in the permeability signalling pathway induced by both isoforms and the potential protective proprieties of VEGF165b.
This study was funded by an unrestricted Novartis PhD Studentship to KO and a Welcome Clinical training fellowship to SL. The funding organisations had no role in the design of the study, collection, analysis, and interpretation of the data, or writing of the manuscript.
Availability of data and materials
All data generated during this study are included in this published article [and its supplementary information files].
KO, GIW and ABM designed the experiments; KO and TB carried out experiments and collected data. CJ, SBO helped with analysis and interpretation of the data. KO, GIW and ABM contributed to data analysis and prepared the manuscript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 13.Bevan HS, van den Akker NM, Qiu Y, Polman JA, Foster RR, Yem J, Nishikawa A, Satchell SC, Harper SJ, Gittenberger-de Groot AC, Bates DO. The alternatively spliced anti-angiogenic family of VEGF isoforms VEGFxxxb in human kidney development. Nephron Physiol. 2008;110:p57–67.CrossRefPubMedPubMedCentralGoogle Scholar
- 19.Wallez Y, Huber P. Endothelial adherens and tight junctions in vascular homeostasis, inflammation and angiogenesis. Biochim Biophys Acta. 1778;2008:794–809.Google Scholar
- 36.Fukumura D, Gohongi T, Kadambi A, Izumi Y, Ang J, Yun CO, Buerk DG, Huang PL, Jain RK. Predominant role of endothelial nitric oxide synthase in vascular endothelial growth factor-induced angiogenesis and vascular permeability. Proc Natl Acad Sci U S A. 2001;98:2604–9.CrossRefPubMedPubMedCentralGoogle Scholar
- 50.Gao F, Sabbineni H, Artham S, Somanath PR. Modulation of long-term endothelial-barrier integrity is conditional to the cross-talk between Akt and Src signaling. J Cell Physiol. 2017. doi: 10.1002/jcp.25791. [Epub ahead of print]
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.