OUP user menu

Role of rat α adducin in angiogenesis: Null effect of the F316Y polymorphism

Claudia Cappuzzello, Roberta Melchionna, Antonella Mangoni, Grazia Tripodi, Patrizia Ferrari, Lucia Torielli, Diego Arcelli, Mauro Helmer-Citterich, Giuseppe Bianchi, Maurizio C. Capogrossi, Monica Napolitano
DOI: http://dx.doi.org/10.1016/j.cardiores.2007.04.020 608-617 First published online: 1 August 2007


Objective Rat α adducin point mutation (F316Y) has been associated with primary systemic arterial hypertension. As microcirculatory abnormalities are present in most forms of hypertension, the aim of the present study was to investigate whether rat α adducin may regulate endothelial cell (EC) functions in vitro and in vivo.

Methods and results The overexpression of rat wild type α adducin (WT-Add1) in ECs induced capillary-like structure development in Matrigel in vitro and enhanced capillary formation in Matrigel implants in vivo in CD1 mice. In contrast, the overexpression of the mutated form (MUT-Add1) of rat α adducin had a Null effect in vitro and lacked any significant activity in vivo. Further, adenovirus-mediated rat WT-Add1 but not MUT-Add1 gene transfer to murine ischemic hindlimb enhanced capillary formation in skeletal muscles.

Gene profiling of human umbilical vein endothelial cells overexpressing α adducin was performed in order to identify putative effector molecules of α adducin-mediated activities on ECs. Interestingly, among a number of genes involved in angiogenesis regulation, retinoic acid-induced protein (RAI17) was found to be upregulated in WT-Add1 vs MUT-Add1 overexpressing cells, possibly representing a key molecule/axis for the functional Add1-induced effect.

Conclusions Rat WT α adducin enhanced EC functions both in vitro and in vivo. The expression of the F316Y variant, associated with the hypertensive phenotype, had a Null effect and might contribute to endothelial rarefaction/dysfunction in hypertension. RAI17 was found to be a putative effector molecule differentially regulated by the overexpression of the two forms of Add1 in endothelial cells.

1 Introduction

Adducins are a family of cytoskeletal proteins that associate with F-actin and spectrin, favour their assembly [1–3] and modulate Na–K pump activity thereby affecting ion transport and renal tubule Na reabsorption [4–6]. The role of adducins in hypertension has been demonstrated both in animal and human studies. Specifically, the point mutation of the rat α adducin gene (F316Y) is responsible for approximately 50% of hypertension of Milan Hypertensive rats (MHS), an animal model of primary hypertension [7]. Further, the association between the human α adducin (G460W) polymorphism and essential hypertension has been demonstrated by linkage analysis and association studies [8,9], although not in all populations investigated [10,11].

Despite the fact that rat and human α adducin polymorphisms are on different codons, α mutated adducin of both species similarly affect cell and protein functions. In particular, in transfected renal cells, mutated adducin variants increase the Na–K pump function [4], reduce Na–K pump endocytosis [5], favour actin bundling and activate focal adhesion sites [4]; in a cell-free system, they increase actin polymerization [4] and stimulate the Na–K ATPase activity with higher affinity than the wild type variants [12]. These data therefore support the role of mutated adducin in enhancing renal tubular sodium reabsorption through an increased function of the Na–KATPase and a possible involvement in organ damage through cytoskeleton remodelling. Indeed, in hypertensive patients carrying the adducin mutation, alone or in combination with that of ACE, a positive association with stroke [13] coronary heart disease [14–16], renal [17] or vascular dysfunctions [18,19] has been reported. Among all the possible organ complications associated to hypertension, microcirculatory abnormalities are present in both essential and secondary hypertension [20–22]. They are mainly constituted by an increase in wall-to-lumen ratio of small vessels and by their rarefaction as a functional consequence of vasoconstriction and of vascular remodeling, ultimately resulting in increased peripheral resistance [21]. Actin cytoskeleton dynamics have a primary role in determining cell shape and motility and are involved both in physiological and pathological processes including angiogenesis [23]. Since adducin is a key regulator of the actin polymerization process and controls cell-to-cell contact formation and cell migration [24,25], and α adducin mutations increase actin polymerization and bundling affecting cytoskeleton remodelling [4], became important to investigate whether adducin polymorphisms may affect the molecular mechanisms underlying angiogenesis.

The aim of the present study was to analyse whether rat α adducin may be involved in EC function regulation and angiogenesis, and whether the point mutation of α adducin, associated with the hypertensive phenotype, may have functional effects on the vasculature.

2 Methods

2.1 Cell culture

Human umbilical vein endothelial cells (HUVECs) (Clonetics, CA, USA) were grown in complete medium (EBM-2) (Cambrex Corporation, New Jersey, USA) as previously described [26]. Experiments were performed on passage 4–6 subcultures.

2.2 Mice

CD1 mice were purchased from Charles River (Milan, Italy). The experimental animal work, described below, conforms with the Guide for Care and use of Laboratory Animals published by The US National Institute of Health (NIH Publication No. 85-23, revised 1996).

2.3 Construction of Add1-expressing adenovirus vectors

Recombinant adenovirus vectors expressing either the wild type (Ad.CMV.WT-Add1), or mutated (Ad.CMV.MUT-Add1) rat-α-adducin cDNA were generated by homologous recombination in bacteria following cloning into ptrackCMV GFP-expressing vectors as described [27].

2.4 Western blot analysis

HUVECs were infected with Add1-expressing and Null adenovirus vectors at 100 M.O.I. for 2 h, washed and resuspended in culture medium. After 48 h incubation, cell lysates were obtained and 50 μg/lane were loaded onto 7% SDS-PAGE and transferred onto a nitrocellulose membrane. Blots were probed with anti-α-adducin monoclonal antibodies (1:2000, kindly provided by Prof. Giuseppe Bianchi, HSR, Milan) and normalized with anti-tubulin antibodies (1:1000, Santa Cruz Biotechnology, CA, USA).

2.5 Proliferation assay

HUVEC were infected either with Ad.CMV.WT-Add1, Ad.CMV.MUT-Add1 or Ad.CMV.Null. 24 h post-infection and cells were plated at 6.0×104 cells/well in 24 well plates in duplicates. After 2 and 5 days of culture, viable cell number was determined by trypan blue exclusion and cells were counted by hemacytometer. Results were expressed as mean cell number ±SE.

2.6 Matrigel morphogenic assay

WT-Add1-, MUT-Add1- or Null-transduced HUVECs, at 6.0×104 cells, were resuspended in 1 ml of 2% FCS EBM-2 onto 200 μl of Matrigel (BD Biosciences Clontech, CA, USA) in 24-well plates in duplicates. After 2–4 h incubation at 37 °C, cells were fixed and photographed. Capillary-like structure formation was quantified by counting the number of branching points (magnification 40×) of five random fields ±S.E.

2.7 In vivo Matrigel angiogenesis assay

CD1 mice were anesthetized with intraperitoneal injection of 2% tribromomethyl alcohol diluted in tert-amyl alcohol (880 mmol/kg body weight, Sigma Immunochemical, MO, USA). HUVEC (1×105 cells) infected either with WT-Add1, MUT-Add1 or Null viruses, were resuspended in 100 μl PBS, mixed with 400 μl Matrigel and injected into the abdominal subcutaneous tissue of CD1 mice (n=7 each group). Matrigel plugs were extracted after 7 days, fixed in 10% buffered formalin and embedded in paraffin. Sections were cut, Masson Trichromic-stained and capillary density determined under a microscope, by two independent investigators, as previously described [28]. Arterioles length density, ranges 4–10.99 μm and 4–41 μm, was measured by morphometric analysis as mentioned in the material and methods section [29,30].

2.8 Mouse hindlimb ischemia model

CD1 mice were anesthetized and unilateral hindlimb ischemia was induced by removal of the left femoral artery as previously described [31,32]. Immediately after ischemia induction, animals (n=5 in each group) were injected with 1×108 pfu Null, WT-Add1 or MUT-Add1 in 25 μl saline buffer. At 14 days following ischemia, muscle sections from the three groups of mice were cut and stained with H…E for capillary counting.

2.9 Immunofluorescence

Serial sections (4 μm) were preincubated with PBS/BSA 10% to then be incubated with either rabbit anti-GFP-(Invitrogen CA, USA) or mouse anti-smooth muscle actin-(SMA) (Sigma) antibodies. A swine FITC-conjugated anti-rabbit or rabbit TRITC-conjugated anti-mouse was used as secondary antibodies.

Further, sections were double stained with both rat anti-CD34 (Abcam, USA) and rabbit anti-GFP antibodies overnight at 4 °C. Swine anti-rabbit FITC-conjugated and donkin anti-rat TRITC-conjugated (Jackson, USA) Abs were used as secondary antibodies. Images were obtained with a Zeiss microscope at 400× magnification.

2.10 ELISA

Supernatants of HUVECs differentiated onto Matrigel were collected at 6 h. VEGF, b-FGF and TGF-β levels were measured by ELISA kit (R…D System, USA) according to manufacturer's specifications.

2.11 Affymetrix array screening and analysis

Total RNA was isolated from WT-Add1-, MUT-Add1- and Null-infected HUVECs by Trizol reagent (Invitrogen, CA, USA). Preparation of labeled cRNA and hybridization (GeneChip Human Genome U133A Array, Affymetrix, CA, USA) was obtained following manufacturer's instructions. Affymetrix GeneChip scanning was analysed by a customized R language-based script [33] which utilizes the Bioconductor packages [34] (see www.bioconductor.org) for quality control analysis, data normalization, unsupervised two-way-hierarchical cluster and identification of differentially expressed transcripts. Three independent experiments were performed, for control (C1–C3), WT adducin (W1–W3)- and mutated adducin (M1–M3)-overexpressing HUVECs. Gene expression profiles of the three groups of samples were generated using Prediction Analysis of Microarrays (PAM) [35].

2.12 qRT-PCR validation

cDNA synthesis for quantitative real time PCR (qRT-PCR) was obtained using the DNA Synthesis In Vitro Transcription Kit (Invitrogen, CA, USA) according to the manufacturer's protocol. The sequences of forward and reverse primers for target genes (THBD, EFNB2, RAI17, NRP1, NRP2) and housekeeping gene (GAPDH) were selected based on published sequence data from NCBI database.

All reactions were performed in 96-well format in the Perkin-Elmer ABI PRISM 7000 Sequence detection system (Perkin-Elmer, MA, USA). For each gene of interest, qRT-PCR was performed as follows: each RNA sample was tested in duplicate and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used to normalize transcript abundance. mRNA expression levels were calculated by Comparative Ct Method by using the Applied Biosystem software (Applied Biosystem, CA, USA) and were presented as fold induction of transcripts for target genes. Fold change above 1 denotes upregulated expression, and fold change below 1 denotes downregulated expression vs Null-infected cells. From five to seven cDNAs obtained from independent infections were tested.

2.13 Statistical analysis

Continuos variables were analysed by the Student's t test. Data are expressed as mean±S.E. A value of P≥0,05 was considered statistically significant.

3 Results

3.1 Rat WT α adducin overexpression induces EC differentiation

In order to achieve high expression levels of α adducin in ECs, recombinant adenovirus vectors expressing either rat WT (Ad.CMV.WT-Add1) or F316Y mutated form (Ad.CMV.MUT-Add1) of α adducin in a bicistronic green fluorescent protein (GFP)-expressing vector were engineered. Similar levels of both forms of exogenously-expressed Add1 in HUVEC were obtained, as shown by Western Blot analysis (Fig. 1A), while cells infected with control vector (Ad.CMV.Null) had very low levels of endogenous α adducin levels, visible only in overexposed autoradiographies (not shown). Under the experimental conditions of the present study, >80% of cells were infected, as assessed by GFP-expression (Fig. 1B–D). We first analysed whether a known hypertensive stimuli, i.e. Angiotensin II (Ang II), would be able to modulate Add1 mRNA levels, and found that treatment with Ang II upregulated Add1 levels, as assessed by real time PCR (Fig. 1E).

Fig. 1

Adenovirus-mediated gene transfer of α adducin in human ECs. A) Western Blot analysis of Add1 expression, of HUVEC infected with Null, WT-Add1 and MUT-Add1 viruses. B–D) Representative GFP-positive pictures of HUVEC infected with Null, WT-Add1 and MUT-Add1 viruses, respectively. Bars=90 μm. E) Add1 mRNA expression, by qRT-PCR, following 24 h Ang II treatment (upper panel: EtBr staining; bottom line: ΔCt values).

It was then analysed whether Add1 overexpression may affect EC functions in vitro and found that there was no significant difference in cell number at 2 and 5 days among WT-Add1-, MUT-Add1- and Null-infected HUVECs (Fig. 2A) nor in % cell survival, as measured by propidium iodide (PI) staining of BrdU incorporation, or in migration assay (not shown). In contrast, WT-Add1 overexpression induced a significant increase in HUVEC differentiation in the Matrigel morphogenic assay vs control cells, as shown in Fig. 2B. Interestingly, the mutated form of Add1 did not show an enhancement of EC differentiation, differently from the WT form (Fig. 2B). The fold increase in intersection point number, a measure of EC differentiation into capillary-like structures, in WT- or MUT-Add1-infected cells, were: 2.2±0.2 and 1.3±0.2 fold of Null-infected cells, respectively. Further, it was examined whether the WT-adducin effect to enhance HUVEC differentiation may be due to augmented growth factor expression. Thus, VEGF, FGF-2, TGF-β levels were measured by ELISA in the supernatants of Matrigel-overlaid HUVEC overexpressing the two forms of adducin vs control but no significant differences were found (Fig. 2C).

Fig. 2

Rat WT α adducin overexpression in primary endothelial cells induced a differentiated phenotype. A) Cell number (×103) at 2 and 5 days following adenovirus infection. HUVEC were infected either with Null, WT-Add1, or MUT-Add1 viruses. B) In vitro Matrigel assay: average number of branching points/field is shown (n=3). Asterisks (*) indicate statistical significance (p<0.01) vs Null-infected sample; (#) indicate p<0.01 of WT- vs MUT-Add1-infected samples. C) Expression of VEGF, bFGF and TGF-β by ELISA, in the supernatants of Matrigel-differentiated transduced ECs.

3.2 Rat WT- but not MUT-Add1 overexpression, induces in vivo angiogenesis in Matrigel plugs and in the mouse hindlimb ischemia model

To assess whether the WT α adducin effects observed in vitro were also exerted in vivo, CD1 mice were subcutaneously implanted with Matrigel plugs containing HUVEC transduced with WT-Add1, MUT-Add1 or Null viruses. Overexpression of WT-Add1 in ECs entrapped in Matrigel promoted the formation of an increased number of capillaries in the plug (Fig. 3B, D) vs control cells (Fig. 3A, D), i.e. 17.2±6.4 vs 6.1±1.4 (p<0.05). In contrast, the forced overexpression of the mutated form of Add1 in ECs did not enhance capillary formation (Fig. 3C, D), i.e. 8.3±2.5 vs 6.1±1.4 (p>0.05). Arterioles length density, analysed by morphometric analysis of SMA staining of serial sections of the plug, did not statistically differ in the three experimental groups as shown in Fig. 3E. In order to support the hypothesis that Add1-transduced ECs may participate to vessel formation in vivo, we performed immunofluorescence experiments on serial sections, in the in vivo Matrigel assay, and found vWF-positive adenovirus-transduced EC (Fig. 3F).

Fig. 3

Rat WT but not MUT α adducin overexpression, induces in vivo neoangiogenesis in Matrigel plugs. Capillary count (×mm2) of in vivo subcutaneously-implanted Matrigel plugs containing HUVEC transduced with 100 M.O.I. Null, WT-Add1, or MUT-Add1 in CD1 mice. Matrigel plugs were analyzed at 7 days post-implant. A–C) Representative pictures of histological sections. Bars=25 μm. D) Average capillary density in Matrigel plugs (n=7 each group). Asterisks (*) indicate statistical significance (p<0.05) of WT-Add1 sample vs both Null and MUT-Add1 samples. E) Arterioles length density/mm3±S.E., as measured by morphometric analysis of smooth muscle actin staining (p>0.05). F) Immunofluorescence experiments on serial sections of adenovirus-transduced in vivo Matrigel experiments. Left panel: Hoechts staining, middle panel: GFP immunofluorescence, right panel: vWF staining. Arrows indicate GFP-positive ECs.

We then analysed the Add1-induced angiogenic effect in a mouse model of hindlimb ischemia as an independent in vivo assay for evaluating vessel formation. Skeletal adductor muscles of CD1 mice were injected with 1×108 pfu recombinant adenoviruses, and hindlimb ischemia was induced at the time of adenovirus injection. The different groups of animals were sacrificed at day 14 and vessel count analysed. Interestingly, animals treated with WT-Add1 showed a small, yet statistically significant, increase in capillary number when compared to Null-treated animals as shown in Fig. 4A. Arterioles number, on the contrary, was similar in all the experimental groups of treated mice (not shown). Further, we tested whether ECs were indeed transduced by adenovirus in vivo and this was the case (Fig. 4B), differently from smooth muscle cells (Fig. 4C), thus supporting the hypothesis that adducin-transduced ECs may participate to vessel formation following ischemia.

Fig. 4

Rat WT- but not MUT-Add1 overexpression enhances capillary formation in skeletal muscles in a mouse model of hindlimb ischemia. A) Capillary density at 14 days in adductor skeletal muscles of CD1 mice following hindlimb ischemia. After femoral artery removal, mice were injected with 1×108 pfu Null, WT-Add1, or MUT-Add1 (n=5 each group). Asterisks (*) indicate statistical significance vs Null sample (p<0.05). B) Adenovirus transduction of ECs in skeletal muscles. Left panel: GFP immunofluorescence, middle panel: CD34 immunofluorescence, right panel: Merge of GFP-CD34 immunofluorescence. White arrows indicate a transduced vessel (upper right) and a transduced endothelial cell (lower left). C) Immunofluorescence on serial sections of adenovirus-transduced skeletal muscles. Left panel: GFP immunofluorescence, middle panel: smooth muscle actin (SMA) immunofluorescence, right panel: Hoechst staining. Bars=25 μm.

3.3 Differential gene expression profiling in Add1-transduced HUVEC vs control cells

In order to identify effector molecules of Add1-mediated biological effects on ECs, gene profiling of HUVEC transduced either with Null, WT-Add1 or MUT-Add1 was performed. Briefly, mRNAs extracted from the three sets of samples were processed using the Affymetrix chip technology and hybridized to human microchips (U133A). Data were then subjected to bioinformatical analysis that allowed to identify a very limited number of genes whose expression levels were statistically different among the three experimental groups and that clusterized for expression and function (Fig. 5). Prediction analysis of microarrays (PAM) identified a number of “best candidate genes” that were differentially expressed among groups (Table 1).

Fig. 5

Differential gene expression profiling in α adducin-transduced HUVEC vs control cells. Hierarchical cluster of gene expression in control (C1-3), WT-Add1 (W1-3) and MUT-Add1 (M1-3)-overexpressing samples. Each row represents one gene and each column represents one sample. Expression levels greater that mean are shaded in red, and those below mean are shaded in blue.

View this table:
Table 1

PAM-analysis output in rat WT α adducin-, rat MUT α adducin-overexpressing ECs vs control cells

SymbolLocus linkDescriptionMUT-Add1 scoreNull scoreWT-Add1 score
Add1118Adducin 1 (alpha)0.2584−1.10390
SAMSN164092SAM domain, SH3 domain and nuclear localization signals, 100.5952−0.136
TCF7L26934Transcription factor 7-like 2 (T-cell specific, HMG-box)0.0259−0.55070
ENC18507Ectodermal-nuclear cortex (with BTB-like domain)00.53340
CDC42998Cell division cycle 42 (GTP binding protein, 25kDA)00.3452−0.4077
FLJ1280664853Chromosome 1 open reading frame 800−0.38570.097
RAB40B10966RAB40B, member RAS oncogene family0.384400
DKFZp76N1910221092Heterogeneous nuclear ribonucleoprotein U-like 2000.3802
ZCCHC1423174Zinc finger, CCHC domain containing 140−0.3630
RAI1757178Retinoic acid induced 170−0.33460
DND1373863Dead end homolog 1 (zebrafish)0−0.28040
ABCG19619ATP-binding cassette, sub-family G (WHITE), member 10−0.27280
JMJD1B51780Jumonji domian containing 1B0−0.16630
PBX25089Pre-B-cell leukemia transcription factor 20−0.15130
ARHGDIA396Rho GDP dissociation inhibitor (GDI) alpha00−0.147
AMFR267Autocrine motility factor receptor0.12280−0.1112
PAICS10606Phosphoribosylaminoimidazole carboxylase, phosphoribosylaminoimidazole succinocarboxamide synthetase00−0.108
RKHD1399664Ring finger and KH domain containing 10−0.10530
PPFIBP18496PTPRF interacting protein, binding protein 1 (liprin beta 1)00.09630
ATP2A2488ATpase, Ca++ transporting, cardiac muscle, slow twitch 20−0.09170
SQSTM18878Sequestosome 100.0870
TFAM7019Transcription factor A, mitochondrial00−0.0827
NID4811Nidogen (enactin)0−0.07760
SMARCA26595SWI/SNF related, matrix associated, actin dependent regulator of chromatin subfamily a, member 20−0.06620
NRP18829Neuropilin 10−0.05930
  • Positivity or negativity in respect to Null score (upper right) indicates relative up- or downregulation of MUT-Add1-(MUT-Add1 score) or WT-Add1-(WT-Add1 score) expressing cells. Values are expressed in probability scores.

We next validated the differential expression of few selected genes (Table 2), identified by PAM analysis, and based on their role in EC function, by means of qRT-PCR, in the three groups of samples. The upregulation of three genes involved in angiogenesis, i.e. neuropilin1 (NRP1), thrombomodulin (THBH) and retinoic acid induced 17 (RAI17) and downregulation of the ephrin B2 gene (EFNB2) in Add1-overexpressing samples vs control samples were validated by qRT-PCR (Fig. 6).

Fig. 6

Validation of differential expression of selected genes by qReal Time PCR. Validation of differential gene expression in Null, WT-Add1, and MUT-Add1-infected HUVEC by qReal Time-PCR. Upper (histograms) and lower (fold) panels: shown is the mRNA fold increase of WT-Add1-(Gray bar) and MUT-Add1-(Black bar) vs Null-infected HUVECs (White bar). Asterisks (*) indicate statistical significance of WT-Add1 samples vs MUT-Add1 samples (p<0.05).

View this table:
Table 2

Real time PCR primer sequences

GENEForward primerReverse primer

Specifically, the gene expression of Retinoic Acid-Induced protein 17 (RAI17) showed a 3.9±1.2 fold increase (p<0.05) and 1.8±0.4 fold increase (p<0.05), respectively, in WT and MUT vs control samples. Interestingly, a statistically significant difference was found between the WT vs MUT samples. Among the other tested genes, NRP1 and THBH were upregulated in Add-1 overexpressing vs control cells (p≥0.05) while EFNB2 was downregulated (p≥0.05). Further, also Neuropilin-2 (NRP2) was similarly found to be upregulated vs control samples. However, no differences were observed in the expression of NRP1, NRP2, THBH and EFNB2 between the WT vs the MUT samples.

4 Discussion

Microcirculatory abnormalities are present in both essential and secondary hypertension. Patients carrying the α adducin mutation show hypertension and an association with stroke, coronary heart disease, or vascular dysfunctions. Add1 is involved in actin cytoskeleton dynamics that play an important role in many processes, including angiogenesis.

Phosphorylation of adducin by protein kinase A and C may affect calmodulin binding [36] and phosphorylation by Rho-kinase may regulate cell motility [25,37]. Further, the src family kinase fyn associates and phosphorylates β adducin promoting colocalization with actin filaments [38].

Rat α adducin F316Y mutation is involved in primary hypertension in the MHS rat strain [39]. The regulatory effect on cytoskeleton assembly, the altered distribution of αV integrins within focal adhesion proteins, and the modulation of Na–K ATPase activity may be important determinants of adducin-mediated biological functions that lead to the regulation of blood pressure [4].

It is still debated whether capillaries and arterioles rarefaction may be the cause or the effect of hypertension [21]. In fact, several studies have been performed to clarify this issue. Evidences in favour of the latter hypothesis come from the increase in wall-to-lumen ratio of small vessels in most forms of hypertension as a putatively adaptative response to increased blood pressure. Further, capillary rarefaction is a hallmark of hypertension and may be the consequence of increased reactivity to vasoconstrictive stimuli and/or vessel obliteration following reduced perfusion [20]. On the other hand, both in young spontaneous hypertensive rats (SHR) [40] as well as in humans, for example in young non-hypertensive patients with a family history of hypertension, capillary rarefaction may precede the onset of hypertension [41].

In the present study we showed that Add1 overexpression regulates EC function in vitro and in vivo. In fact, WT-Add1-overexpressing HUVEC showed increased differentiation into capillary-like structures in the Matrigel assay and an increased number of capillaries in vivo following the implant in Matrigel plugs. Further, WT-Add1-overexpression in mice subjected to hindlimb ischemia also showed an increased number of capillaries in the injected muscles at day 14, when compared to control adenovirus-injected muscles. In the present study it was described as a novel role for Add1 in regulation of EC function and angiogenesis. The F316Y polymorphism of Add1 is thought to be responsible for 50% of hypertension in MHS rats; similarly, the polymorphism G460W of α adducin was associated to hypertension in a number of studies in humans [6].

It was therefore investigated whether the rat F316Y polymorphism may affect EC functions and angiogenesis, similarly to the WT form. Interestingly, instead, the overexpressed Add1-mutated form had a Null effect on EC differentiation.

These data suggest that Add1 mutation may contribute to EC dysfunction associated to hypertension. Further, the Null effect that we showed in vitro in mutated Add1-overexpressing ECs suggests that Add1-mediated signals may affect EC functions even in the absence of hypertension, therefore supporting the hypothesis that modulation of EC function may also precede hypertension.

In order to identify key molecules involved in Add1-mediated functions in ECs, we have characterized the gene expression profiles of HUVEC overexpressing WT- or MUT-Add1 vs control cells. We validated few among the most significantly-regulated genes that are also involved in EC function regulation. Retinoic acids have been previously shown to modulate angiogenesis, specifically RA as been described as a potent inducer of microvascular EC differentiation into capillary-like network in vitro [28,42] and it was found that RA can induce angiogenesis in vivo [28,43]. Interestingly, we found that RAI17, a Retinoic Acid-Induced protein 17 gene, was strongly upregulated by WT-Add1, and to a lesser extent by MUT-Add1 overexpression. Interestingly, a statistically significative difference in RAI17 expression levels was observed in WT-Add1 overexpressing cells vs MUT-Add1 overexpressing cells thus suggesting a putative role of this signaling pathway in the Add1-induced functional effects.

Neuropilins (NRPs) are cell surface molecules involved in neuronal guidance, vascular development and angiogenesis [44,45]. We found that the two forms of neuropilins were similarly upregulated in both WT- and MUT-overexpressing cells in respect to control cells.

Further, we found that thrombomodulin, an anticoagulant, EC membrane glycoprotein recently described as an angiogenic factor that enhances formation of new vessels both in vitro and in vivo, [46] showed increased mRNA expression levels in Add1-overexpressing cells. EphrinB2, a molecule involved in angiogenesis, [47–49] is instead mildly, but statistically significantly reduced in Add1-overexpressing cells. As NRP1 and 2, thrombomodulin and EPNB2 are similarly expressed in WT- vs MUT-Add1 overexpressing ECs, they are unlikely involved in the functional differences observed among WT- vs MUT-Add1.

Although in our experimental system we analysed the biological effects of Add1 following its overexpression, we may hypothesise that, in vivo, different stimuli, such as Ang II, may increase WT α adducin expression levels, and that Add1 may play a role in angiogenesis. Further, the mutated form of Add1, that has a Null effect both in vitro and in vivo, may contribute to impairment of EC function in hypertension.

In conclusion we have shown that α adducin is an inducer of EC functions both in vitro and in vivo and that the F316Y point mutation may contribute to EC dysfunction.


We are grateful to D. Carlini and S. Truffa for their technical contribution and M. Inzillo for the artwork. This work was partially supported by a grant FIRB from Ministero della Salute to G. Bianchi and M. C. Capogrossi.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
  46. [46]
  47. [47]
  48. [48]
  49. [49]
View Abstract