*2.1. Mice*

This study was performed in compliance with Dutch governmen<sup>t</sup> guidelines and the Directive 2010/63/EU of the European Parliament. All animal experiments were approved by the animal welfare committee of the Leiden University Medical Center. Male ApoE3\*Leiden mice, crossbred in our own colony on a C57BL/background, 8 to 16 weeks old, were fed a diet containing 15% cacao butter, 1% cholesterol and 0.5% cholate (100193, Triple A Trading, Tiel, The Netherlands) for three weeks prior to surgery until sacrifice.

#### *2.2. Vein Graft Surgery*

Vein graft surgery was performed by donor mice caval vein interpositioning in the carotid artery of recipient mice as previously described [5,18]. Briefly, thoracic caval veins from donor mice were harvested. In recipient mice, the right carotid artery was dissected and cut in the middle. The artery was everted around the cu ffs that were placed at both ends of the artery and ligated with 8.0 sutures. The caval vein was sleeved over the two cu ffs, and ligated. On the day of surgery and on the day of sacrifice mice were anesthetized with midazolam (5 mg/kg, Roche Diagnostics, Basel, Switzerland), medetomidine (0.5 mg/kg, Orion, Espoo, Finland) and fentanyl (0.05 mg/kg, Janssen Pharmaceutical Beerse, Belgium). The adequacy of the anesthesia was monitored by keeping track of the breathing frequency and the response to toe pinching of the mice. After surgery, mice were antagonized with atipamezol (2.5 mg/kg, Orion Espoo, Finland) and fluminasenil (0.5 mg/kg, Fresenius Kabi, Bad Homburg vor der Ho¨he, Germany). Buprenorphine (0.1 mg/kg, MSD Animal Health, Keniworth, NJ, USA) was given after surgery to relieve pain.

#### *2.3. Carbogen Treatment*

Acute reoxygenation was investigate in ApoE3\*Leiden mice on day 28 after vein graft surgery. Mice were randomized in two groups, a control group (*n* = 13) and a carbogen treated group (*n* = 12) and exposed for 90 min to air (21% O2) or carbogen gas (95% O2, 5% CO2) respectively. Halfway during the treatment, the mice received intraperitoneal injection of hypoxia specific marker pimonidazole (100 mg/kg, hypoxyprobe Omni kit, Hypoxyprobe Inc., Burlington, MA, USA) and anesthesia. Directly after the end of the treatment, mice were sacrificed after 5 min of in vivo perfusion-fixation under anesthesia. Vein grafts were harvested, fixated in 4% formaldehyde, dehydrated and para ffin-embedded for histology.

Chronic reoxygenation was investigated in ApoE3\*Leiden mice starting on day 7 after vein graft surgery. The decision for this timepoint was based on our previous finding that intraplaque angiogenesis is detectable in ApoE3\*Leiden mice starting from day 14 after vein graft surgery [5]. Mice were randomized based on their plasma cholesterol levels (Roche Diagnostics, kit 1489437, Basel, Switzerland) and body weight in two groups, a control group (*n* = 16) and a carbogen treated group (*n* = 16) and exposed daily for 90 min to air (21% O2) or carbogen (95% O2, 5% CO2) respectively, until the day of sacrifice. On day 28 after surgery, mice received the last treatment and halfway during this last treatment they received intraperitoneal injection of hypoxia specific marker pimonidazole (100 mg/kg, hypoxyprobe Omni kit, Hypoxyprobe Inc., Burlington, MA, USA) and anesthesia. Immediately after the end of the treatment, mice were sacrificed as previously described for the acute reoxygenation experiment.

#### *2.4. Histological and Immunohistochemical Assessment of Vein Grafts*

Vein graft samples were embedded in para ffin, and sequential cross-sections (5 μm thick) were made throughout the embedded vein grafts. To quantify the vein graft thickening (vessel wall area), MOVAT pentachrome staining was performed. Total size of the vein graft and lumen were measured. Thickening of the vessel wall (measured as intimal thickening + media thickening) was defined as the area between lumen and adventitia and determined by subtracting the luminal area from the total vessel area. The optimal lumen area was calculated by converting the luminal circumference, measured as the luminal perimeter, into luminal area.

Intraplaque angiogenesis was measured as the amount of CD31+ vessels in the vessel wall area and intraplaque hemorrhage (IPH) was monitored by the amount of erythrocytes outside the (neo)vessels and scored as either not present, low, moderate or high.

Antibodies directed at alpha smooth muscle cell actin ( αSMActin, Sigma, Santa Clara, CA, USA), Mac-3 (BD Pharmingen, Franklin Lakes, NJ, USA), Pimonidazole (mouse IgG1 monoclonal antibody, clone 4.3.11.3, Hypoxyprobe Inc., Burlington, MA, USA), 8OHdG (bs-1278R, Bioss antibodies, Woburn, MA, USA), CD31 (sc-1506-r, Santa Cruz, Dallas, TX, USA), Ter119 (116202, Biolegend, San Diego, CA, USA), Ki67 (ab16667, Abcam, Cambridge, UK) and cleaved caspase 3 (9661-S, Cell SignalingDanvers, MA, USA) were used for immunohistochemical staining. Sirius red staining (80115, Klinipath, Amsterdam, The Netherlands) was performed to quantify the amount of collagen present in the vein grafts. The immuno-positive areas are expressed as a total area or percentage of the lesion area. Stained slides were photographed using microscope photography software (Axiovision, Carl Zeiss Microscopy, White Plains, NY, USA) or Ultrafast Digital Pathology Slide Scanner and associated software (Philips, Cambridge, MA, USA) and image analysis softwares were used to quantify the vein graft intimal hyperplasia and composition (Qwin, Leica, Wetzlar, Germany and Imagej, Bethesda, MD, USA).

#### *2.5. RNA Isolation, cDNA Synthesis and qPCR*

Total RNA was isolated from 10 (20 μm thick) para ffin sections (at least *n* = 6/group) following the manufacture's protocol (FFPE RNA isolation kit, Qiagen, Venlo, the Netherlands). cDNA was synthesized using the Superscript IV VILO kit according to the manufacture's protocol (TermoFisher, Waltham, MA, USA).

Commercially available TaqMan gene expression assays for the housekeeping gene hypoxanthine phosphoribosyl transferase (Hprt) (Mm01545399\_m1), and selected genes were used (Applied Biosystems, Foster City, CA, USA); Vegfa (Mm03015193\_m1), Hif1a (Mm00468869\_m1), Cxcl12 (Mm00445553\_m1), Epas1 (Mm01236112\_m1), Il6 (Mm00446190\_m1), Tnf (Mm00443258\_m1) and Ccl2 (Mm00441242\_m1). Q-PCRs were performed on the ABI 7500 Fast system (Applied Biosystems). The 2-ΔΔCt method was used to analyze the relative changes in gene expression.

#### *2.6. Bone Marrow Derived Macrophages Isolation and In Vitro Experiments*

Monocytes were isolated from bone marrow of tibias and femurs of male ApoE3\*Leiden mice (*n* = 4) and cultured in RPMI 1640 medium (52400-025, ThermoFisher, GIBCO, Waltham, MA, USA,) supplemented with 25% heat inactivated fetal calf serum (Gibco ® by Life Technologies), 100 U/mL Penicillin/Streptomycin (ThermoFisher, GIBCO, Waltham, MA, USA ) and 0.1 mg/mL macrophage colony-stimulating factor (, 14-8983-80, ThermoFisher, E-Bioscience Waltham, MA, USA). After eight days the derived macrophages were seeded in a 12 wells plate for RNA isolation and in a chamber slide for immunocytochemistry (ICC) (NUNC LAB-TEK II, 154534, ThermoFisher, Waltham, MA, USA). 24 h later, when BMM were fully attached, BMM were stimulated with either 200 or 400 μm tert-butylhydroperoxide, t-BHP, (Luperox, 458139, Sigma Aldrich, St. Louis, Missouri, USA) as a ROS mimic for 6 h.

RNA was isolated according to standard protocol using TRIzol ® (Ambion ®, ThermoFisher,Waltham, MA, USA ) after which sample concentration and purity were examined by nanodrop (Nanodrop Technologies, ThermoFisher, Waltham, MA, USA). Complementary DNA (cDNA) was prepared using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, ThermoFisher, Waltham, MA, USA) according to manufacturer's protocol. For qPCR, commercially available TaqMan gene expression assays for the selected genes were used as explained above.

For ICC cells were fixated in 4% formaldehyde and antibodies directed at Mac-3 (BD Pharmingen, Franklin Lakes, NJ, USA), 8OHdG (bs-1278R, Bioss antibodies, Woburn, MA, USA) and cleaved caspase 3 (9661-S, Cell Signaling, Danvers, MA, USA) were used for immunocytochemical staining. Tile-scans of stained slides were photographed using a fluorescent microscope (Leica AF-6000, Leica, Wetzlar, Germany) and Fiji image analysis software was used to quantify the mean grey value expression of the targets (Imagej, Bethesda, MD, USA).

#### *2.7. Statistical Analysis*

Results are expressed as mean ± SEM. A 2-tailed Student's t-test was used to compare individual groups. Non-Gaussian distributed data were analyzed using a Mann-Whitney U test using GraphPad Prism version 6.00 for Windows (GraphPad Software). Probability-values < 0.05 were regarded significant.
