Aging differentially modulates the Wnt pro‐survival signalling pathways in vascular smooth muscle cells

Abstract We previously reported pro‐survival effects of Wnt3a and Wnt5a proteins in vascular smooth muscle cells (VSMCs). Wnt5a achieved this through induction of Wnt1‐inducible signalling pathway protein‐1 (WISP‐1) consequent to β‐catenin/CREB‐dependent, TCF‐independent, signalling. However, we found that as atherosclerosis advances, although Wnt5a protein was increased, WISP‐1 was reduced. We hypothesized this disconnect could be due to aging. In this study, we elucidate the mechanism underlying Wnt3a pro‐survival signalling and demonstrate the differential effect of age on Wnt3a‐ and Wnt5a‐mediated survival. We show Wnt3a protein was expressed in human atherosclerotic coronary arteries and co‐located with macrophages and VSMCs. Meanwhile, Wnt3a stimulation of primary mouse VSMCs increased β‐catenin nuclear translocation and TCF, but not CREB, activation. Wnt3a increased mRNA expression of the pro‐survival factor WISP‐2 in a TCF‐dependent manner. Functionally, β‐catenin/TCF inhibition or WISP‐2 neutralization significantly impaired Wnt3a‐mediated VSMC survival. WISP‐2 was upregulated in human atherosclerosis and partly co‐localized with Wnt3a. The pro‐survival action of Wnt3a was effective in VSMCs from young (2 month) and old (18–20 month) mice, whereas Wnt5a‐mediated rescue was impaired with age. Further investigation revealed that although Wnt5a induced β‐catenin nuclear translocation in VSMCs from both ages, CREB phosphorylation and WISP‐1 upregulation did not occur in old VSMCs. Unlike Wnt5a, pro‐survival Wnt3a signalling involves β‐catenin/TCF and WISP‐2. While Wnt3a‐mediated survival was unchanged with age, Wnt5a‐mediated survival was lost due to impaired CREB activation and WISP‐1 regulation. Greater understanding of the effect of age on Wnt signalling may identify targets to promote VSMC survival in elderly patients with atherosclerosis.

day, slides were washed three times in PBS then incubated with biotinylated secondary antibodies (either biotinylated goat-anti-mouse IgG (BA-9200, VECTOR LABORATORIES, Peterborough, UK) or goat-anti-rabbit IgG (B7389, SIGMA ALDRICH, Dorset, UK) as appropriate) diluted 1:200 in 1% (w/v) BSA/PBS for 45 minutes at room temperature. Sections were then washed three times in PBS then treated with ExtrAvidin-peroxidase (E2886, SIGMA ALDRICH, Dorset, UK) diluted 1:200 in 1% (w/v) BSA/PBS for 45 minutes at room temperature. Sections were then washed three times in PBS and incubated with SigmaFast 3,3'-diaminobenzidine (DAB: D4293, SIGMA ALDRICH, Dorset, UK) for 3-10 minutes at room temperature. Sections were then washed in distilled water and stained with the nuclear stain Mayer's hematoxylin for 30 seconds. Slides were washed with tap water to 'blue' the hematoxylin until the water appeared clear. Sections were then dehydrated by one 5 minute wash in 70% (v/v) alcohol, two 5 minute washes in 100% (v/v) alcohol and then three 5 minute washes in 100% (v/v) clearene. Slides were then mounted in DPX mounting medium (10050080, THERMO FISHER SCIENTIFIC, Massachusetts, USA) and cured overnight at room temperature. The following day the sections were imaged using an Olympus BX41 microscope and Q-capture pro 6.0 software.

Quantitative PCR
To investigate changes in mRNA expression with age, young and old VSMCs were quiesced for 48 hours. RNA was extracted from VSMCs and purified using the PureLink RNA mini-kit (12183025, AMBION, Paisley, UK). 500ng of RNA and 4µL of SuperScript VILO Master Mix (11755-050, INVITROGEN, Paisley, UK) was then added to RNase-free PCR tubes and the samples were made up to 20µL with RNasefree water. The samples were heated to 25°C for 10 minutes, then 42°C for 90 minutes and finally 85°C for 5 minutes (C1000™ thermocycler, BIO-RAD, Dorset, UK). The resultant cDNA was then cooled to 4˚C. To enable simultaneous analysis of 48 Wnt related genes, 384-well Taqman Array Micro-Fluidic Cards were designed and purchased from Applied Biosystems. 20µL of cDNA, 35µL of nuclease-free water and 55µL of Taqman Universal Master Mix-II with UNG (4440042, APPLIED BIOSYSTEMS, Paisley, UK) were mixed, then 100µL/sample was added to each reservoir of the array cards. Cards were centrifuged twice at 1200rpm for 1 minute to evenly distribute the cDNA between two rows of the array card and cards were sealed. Comparative CT was then analysed using the ViiA 7 Real-Time PCR System and ViiA 7 RIO software. The card layout and thermocycling conditions were imported using the Gene expression micro-fluidic card CD (4480358, APPLIED BIOSYSTEMS, Paisley, UK) provided with the array cards.
For genes that were not included on the array plate, or to analyse the effect of Wnt on mRNA expression, 96-well QPCR was used. cDNA was synthesised using the Transcriptor First Strand cDNA Synthesis Kit (04896866001, ROCHE, West Sussex, UK) and QPCR was performed using the LightCycler 480 SYBR Green I Master kit (04707516001, ROCHE, West Sussex, UK) and LightCycler 480 II (ROCHE, West Sussex, UK). Primer sequences are detailed in Supplementary Table 2. mRNA levels were normalised to 18s mRNA, except for AXIN-2, IGF-1, TCF-7 and WISP-2 which were normalised to 36B4 mRNA, and WISP-1 which was quantified from a standard curve.    Figure 1 of the manuscript.

Supplementary Figure 2: Wnt3a protein induced nuclear translocation of βcatenin in the absence and presence of H2O2.
Representative images of β-catenin immunofluorescence on young TOPGAL mouse aortic VSMCs stimulated with 100μM H2O2, in the presence or absence of 400ng/mL recombinant Wnt3a protein, for 30 minutes. Non-immune mouse IgG was used as a negative control. Red arrows indicate β-catenin nuclear translocation, defined as perinuclear β-catenin staining (green). Blue nuclei: DAPI. For the top and bottom row the scale bar represents 25µm, for the middle row the scale bar represents 5µm.

Supplementary Figure 4: WISP-2 and IGF-1 protein rescued VSMCs from H2O2induced apoptosis, which could be inhibited by neutralising antibodies.
Apoptosis was quantified in young TOPGAL mouse aortic VSMCs stimulated with 100µM H2O2, in the absence and presence of 100ng/mL recombinant IGF-1 protein and 10µg/mL of IGF-1 neutralising antibody (nAb) (A) or in the absence and presence of 500ng/mL recombinant WISP-2 protein and 10µg/mL of WISP-2 nAb (B), for 24 hours using CC3 immunofluorescence. The number of CC3 positive cells were counted and expressed as a percentage of the total number of cells viewed. * indicates p<0.05 vs. control, $ indicates p<0.05 vs. H2O2, repeated measures ANOVA and Student Newman Keuls post hoc test, n=5.

Supplementary Figure 5: The divergent signalling pathways induced by Wnt3a and Wnt5a cannot be explained by the use of different Fzd receptors.
Apoptosis was quantified in young TOPGAL mouse VSMCs transfected with either Allstars negative control siRNA (A) or Fzd1 siRNA (B) or Fzd6 siRNA (C). Six hours after transfection, VSMCs were quiesced in serum free conditions overnight, then 24 hours after transfection cells were stimulated with 100μM H2O2, in the presence or absence of 400ng/mL recombinant Wnt3a protein, for a further 24 hours. Apoptosis was quantified using CC3 immunofluorescence.

Supplementary Figure 6: Wnt3a protein inhibited H2O2-induced apoptosis in VSMCs from young and old mice, whereas Wnt5a-mediated rescue was lost with age.
Apoptosis was quantified in aortic VSMCs isolated from young (A, C) and old mice (B, D) and stimulated with 100μM H2O2, in the presence or absence of 400ng/mL recombinant Wnt3a (A, B) Wnt5a protein (C, D) for 24 hours using cleaved caspase-3 (CC3) immunofluorescence. Representative images are shown, red arrows indicate positive cells (green). Blue nuclei:DAPI. The scale bar represents 25µm and applies to all images.

Supplementary Table 3: Changes in mRNA expression of Wnt signalling components with age.
mRNA expression of Wnt signalling components was analysed in aortic VSMCs isolated from young and old mice and quiesced for 48 hours. Genes investigated included Wnt ligands, frizzled receptors (Fzd), low density lipoprotein receptor related protein (LRP) co-receptors, N-cadherin (NCAD) and Wnt inhibitors secreted frizzledrelated proteins (sFRPs), Dickkopf (Dkk) proteins and Wnt-inhibitory factor-1 (WIF-1). For each gene, the average CT value generated from VSMCs from young mice and the fold change in gene expression in cells from old mice is shown. All genes were normalised to 18s mRNA levels. Blue highlighting indicates mRNAs that were significantly reduced with age, whereas pink highlighting indicates one mRNA that was significantly increased with age (p<0.05 vs. young, one sample t-test, n=3-5). ND represents not detected, NS represents non-significant.  Representative images of β-catenin immunofluorescence on aortic VSMCs isolated from young (A) and old (B) mice and stimulated with 400ng/mL recombinant Wnt5a protein for 30 minutes. Non-immune mouse IgG was used as a negative control. Red arrows indicate β-catenin nuclear translocation, defined as perinuclear β-catenin staining (green). Blue nuclei: DAPI. The scale bar represents 25µm on the top rows, and 5µm on the bottom rows.