supporting information - pnas · 8/3/2013 · supporting information awojoodu et al....

Supporting InformationAwojoodu et al. 10.1073/pnas.1221309110SI MethodsFabrication of Poly(Lactic-Coglycolic Acid) Thin Films and Encapsulationof Sphingosine 1-Phosphate Receptor-Targeted Compounds. Poly(lac-tic-coglycolic acid) (PLAGA) thin films were fabricated by usinga solvent-casting technique as described (34). A total of 350 mg ofPLAGA was combined with 2 mL of methylene chloride (FisherScientific) in a borosilicate liquid scintillation vial (20-mL capacity;Fisher Scientific) and vortexed until completely dissolved. For 1:200(drug weight:polymer weight) FTY720 (Cayman Chemical),VPC01091, or Compound 26 (Kevin Lynch, University of Virginia)loaded films, 1.75 mg of drug was added to the solution. Thepolymer/drug solution was quickly poured into a P35 Petri dish(Nunc; area = 8.8cm2) lined with Bytac Teflon paper. Films wereallowed to dry at −20 °C for 7 d, and then were stored at roomtemperature in a desiccator until needed. All films were lyophilized(Labconco FreeZone 2.5; Labconco) for 24 h before being used forexperiments to remove any excess solvent. For implantation invivo, films with a diameter of 1 mm were extracted by using a 1-mmbiopsy punch (Acuderm) and rinsed in 70% (vol/vol) ethanol for∼30 s and then washed in sterile Ringer’s solution (137.9 mMNaCl, 4.7 mM KCl, 1.2 mM MgSO4, 1.9 mM CaCl2, and 23 mMNaHCO3) for an additional 30 s. Films had an average thickness of517 ± 41 μm, as measured with calipers (L. S. Starrett Co.).

Fabrication of Nanofiber Scaffolds and Encapsulation of FTY720.A 1:1mixture of polycaprolactone (PCL; Sigma) and PLAGA (Lake-shore Biomaterials) was dissolved in a 3:1 (vol/vol) chloroform:methanol solution. The final concentration of polymer solutionwas 18% (wt/vol). The solution was agitated at room temperatureuntil the polymer dissolved, followed by loading into a 3-mLrubber-free syringe. Electrospinning was performed at a flow rateof 1.0 mL/h, an applied voltage of 19 kV, and a working distanceof 10 cm. Nanofibers were collected on a stationary aluminumplate and then stored in a desiccator until use. To make drug-loaded nanofibers, FTY720 (Cayman Chemical) was dissolved in3:1 chloroform:methanol solution, and 1:1 PCL/PLAGA wasadded at a concentration of 20% (wt/vol). The final drug:polymerratio was 1:200.

Generation of Bone Marrow Chimeric Mice. To generate bone mar-row (BM) chimeric mice, donor mouse tibia were harvested andflushed of marrow. Recipient mice were lethally irradiated witha total dosage of 10.5 gray (Gy) (5.5- and 5-Gy doses, 3 h apart)before transplantation of 2 × 106 donor cells in 150 μL of PBS viatail vein injection. The cells were allowed to engraft for 6 wkbefore experimentation.

Dorsal Skinfold Window Chamber Surgical Procedure. Mice wereimplanted with dorsal skinfold window chambers (APJ TradingCo.).Mice were treated with a preanesthetic of atropine (0.08mg/kg IP) and further anesthetized by using i.p. injections of ketamine(80 mg/kg) and xylazine (8 mg/kg). For nanofiber implantation,anesthesia was induced with isoflurane gas in a chamber (2–3%),and the surgical plane was maintained with a nose cone (1–2%)equipped with a scavenging apparatus for the procedure. Dorsalskin was shaved, depilated, and sterilized by using triplet washesof 70% ethanol and iodide. A double-layered skin fold was el-evated off the back of the mouse and pinned down for surgicalremoval. The titanium frame of the window chamber was sur-gically fixed to the underside of the skinfold. The epidermisand dermis were removed from the top side of the skinfold ina circular area (diameter = ∼12 mm) to reveal the underlying

vasculature. Exposed tissue was kept hydrated with sterileRinger’s solution (137.9 mM NaCl, 4.7 mM KCl, 1.2 mMMgSO4, 1.9 mM CaCl2, and 23 mM NaHCO3). The titaniumframe was then mounted on the top side of the tissue and at-tached to its underlying counterpart. The dorsal skin was suturedto the two titanium frames, and the exposed tissue was sealed witha protective glass window. Mice were allowed to recover in heatedcages and subsequently were administered buprenorphine via s.c.injection (0.1–0.2 mg/kg) as a postoperative analgesic. All micereceived a laboratory diet and water ad libitum throughout thetime course of the experiment.

Implantation of Thin Films and Intravital Image Acquisition. PLAGAthin films were implanted into the window chamber 7 d aftersurgical implantation, hereafter referred to as day 0. Mice wereanesthetized via 2% isoflurane mixed with 1 mL/min O2. Sub-sequently, the glass window was removed to expose the thin layerof vessel networks. The window chamber was flooded with 1 mMadenosine in Ringer’s solution (3 × 5 min) to maximally dilate allvessels and maintain tissue hydration. Following the last ad-ministration of adenosine, the solution was aspirated, and twofilms (either both loaded or both unloaded) were placed equi-distant from one another and from each edge of the window. Themouse was then mounted to a microscope stage and imagednoninvasively by using a 4× objective on an Axioskope 40 micro-scope (Carl Zeiss). Images were captured by using an OlympusMicroFire color digital camera and PictureFrame image acquisi-tion software (Optronics). Individual images were later photo-merged into a single image of the entire microvascular network byusing Adobe Photoshop CS. Mice were initially imaged on day0 following film implantation and again on days 3 and 7.

Quantitative Microvascular Metrics. Intravital microscopy montagesof entire vascular windows at days 0, 3, and 7 were analyzed byusing a combination of Adobe Photoshop CS and ImageJ (http://rsb.info.nih.gov/ij/) software packages. First, circles with a di-ameter of 5 mm (or 2-mm concentric radius from outer edge ofone film) were cropped around each film, with no overlap of thetwo circles.

Changes in Microvascular Length Density. For microvascular lengthdensity measurements, vessels located within these croppedimages were traced by using Photoshop and skeletonized by usingImageJ. These binary images were analyzed by counting the totalnumber of pixels, representing the total length of all tracedvessels. Pixels were converted to millimeters by using the con-version factor (350 pixels = 1 mm) calculated from an image ofa micrometer at the same imaging conditions (4×). Pixel lengthwas divided by total area of the region of interest (a circle with di-ameter of 5 mm; A = 19.63 mm2). The limit of resolution of vesselsthat can be visualized in this manner is 10 μm.

Changes in Microvascular Diameter. To assess changes in luminaldiameter, arteriole–venule pairs were identified in the 2-mmcropped images. In these intravital images, vessels are visualizedby blood column width. Therefore, arterioles and venules wereidentified on the basis of size only; venule diameters were largerthan arteriolar diameters on day 0. Identical vessel segmentswere labeled on day-0, -3, and -7 images at the bisection of eachvessel segment in between branch points. Internal diametersbased on blood column widths were measured by using ImageJ.Day-0 diameter measurements were used to bin vessels by sizeat the onset of the experiment to track how initial size affects

Awojoodu et al. www.pnas.org/cgi/content/short/1221309110 1 of 7

http://rsb.info.nih.gov/ij/http://rsb.info.nih.gov/ij/www.pnas.org/cgi/content/short/1221309110

luminal expansion potential. This metric is limited to vessels thatare visible at all three time points, and the limit of resolution is∼10 μm.

BM Harvest and Cell Sorting.Mice tibia and femurs were harvestedwith scissors and forceps. The ends of the bone were cut with bonecutters, and marrow was flushed with sterile saline by using asyringe and 25-gauge needle. BM was collected, filtered, andresuspended in sterile PBS + 10% FBS for flow cytometrystaining. SSCloCD45+CD11b+Gr1+Ly6C+ inflammatory subtypeof monocytes (IM) and SSCloCD45+CD11b+Gr1−Ly6C− anti-inflammatory subtype of monocytes (AM) were sorted for invitro and adoptive transfer experiments.

Adoptive Transfer. SSCloCD45+CD11b+Gr1+Ly6C+ IMs andSSCloCD45+CD11b+Gr1−Ly6C− AMs were sorted from mouseBM and stained with 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindo-carbocyanine perchlorate (DiI) for 20 min. Host mice received800,000 IMs or 175,000 AMs i.v. via retroorbital sinus 24 h be-fore backpack surgery. Blood was collected immediately beforesurgery and at 1 and 3 d after surgery. Dorsal tissue was alsoharvested 3 d after surgery for flow cytometric analysis.

Dorsal Tissue Immunohistochemistry. Five-micrometer-thick sec-tions were cut from paraffin-embedded s.c. dorsal tissue. Sectionswere dewaxed with 100% xylenes, 100% ethanol, 95% ethanol,70% ethanol, and 50% ethanol in series for 3 min each. Antigenswere retrieved by high-temperature pressure cooking of slides for15 min in 10 mM sodium citrate buffer (pH 6). Endogenousperoxidase activity was blocked by incubation of specimens with3% hydrogen peroxide in 40% methanol for 10 min followed bynonspecific blocking with 2.5% horse serum. Slides were in-cubated with 1:50 rabbit anti-mouse MHCII (Novus), CD206(Bioss), and F4/80 (AbCam) overnight at 4 °C in a humidifiedchamber. Slides were washed and fluorescent secondary anti-bodies or anti-rabbit peroxidase secondary antibodies, and dia-minobenzidine–peroxidase substrate kit (Vector Lab) were usedto stain slides. Hematoxylin counterstaining was performed for5 min to stain tissue and cells on specimens.

Dorsal Tissue Digestion for Single-Cell Suspensions and Protein. Tomeasure the recruitment of cells to inflamed tissue, dorsal skinfoldwindow chambers were implanted on CX3CR1–eGFPmice. Ninety-five percent of CX3CR1, also known as fractalkine, can be attrib-uted to monocytes. Two FTY720-loaded or unloaded PLAGAfilms were implanted on the day of surgery. Three days after, two8-mm biopsy punches were taken around each implant in eachanimal. This dorsal tissue was treated with 1 mg/mL collagenase(Sigma) at 37 °C and further disaggregated with a cell strainer tocreate a single-cell suspension. For protein analysis, single-cellsuspensions were placed in 1× radioimmunoprecipitation assaybuffer with protease and phosphatase inhibitors for 5 min on ice.The lysed cells were then centrifuged to pellet out debris, and thelysate was stored at −80 °C until analysis.

Spinotrapezius Ligation. Spinotrapezius ligation was performedsimilarly to previous studies (13). In summary, DsRed–NG2CX3CR1–eGFP mice were anesthetized via i.p. injection of ket-amine/xylazine/atropine (60/4/0.2 mg/kg). The position where thefat pad ends (∼1–2 cm from the head posteriorly) was found, andan incision (∼0.5 cm, parallel to the spine) was made 1 cm laterallyfrom the spine at this location. Under a dissecting microscope,blunt manipulation of the tissue was performed to locate the spi-notrapezius. The main feeding arteriole was located and followedupstream until it exited the spinotrapezius and entered the fat pad.Blunt dissectionwas used to isolate the arteriole from its associatedvenule, and care was taken to minimize mechanical manipulationof themuscle and tissue. Two ligatures were placed on the arteriole

by using 10-0 sutures followed by ligation of the arteriole betweenthe ligatures. Tissue was returned to its original position. Asmall 1-mm disk of PLAGA (either empty or loaded withFTY720) was placed between the fascia and spinotrapeziusbefore closing the wound with 8-0 nonresorbable sutures. Oneweek later, the mouse was euthanized, and the whole spino-trapezius was harvested.

Spinotrapezius Tissue Immunohistochemistry. Both spinotrapeziusmuscles were harvested from each mouse and permeabilized with0.2% saponin in PBS overnight (18 h) at 4 °C. Muscles wereblocked for 1 h (5% mouse serum, 0.2% saponin, 0.5% BSA inPBS) before being treated with primary antibody overnight at 4 °C.Spinotrapezius tissues were immunolabeled for smooth muscleα-actin [mural cell stain, IA4-Cy3 (Sigma-Aldrich), 1:300] andisolectin [endothelial cell (EC) stain, IB4–Alexa 647 (In-vitrogen), 1:200]. Muscles were washed five times at 20 min perwash (0.2% saponin and 0.5% BSA in PBS) before being wholemounted on a coverslip.

Spinoptrapezius Imaging and Data Analysis. Whole-mounted spi-notrapezius tissue was imaged by using confocal microscopy(Nikon; model TE200-E2; 4×, 10×, 20×, and 60× objectives).Analyses of the confocal images were conducted by using ImageJimaging software to quantify vascular lengths and diameters.Tortuosity of arterioles was determined by tracing the patha vessel took and dividing it by the absolute distance of the startand end point. All tortuosity data were acquired in regions of thevessel where no branching occurred. The 10× images were takenof the primary watershed, i.e., the immediate downstream regionof the ligated arteriole. Green fluorescent positive cells werecounted. Additionally, cells touching or aligned with vessels wereconsidered to be associated with the vessel. All vascular analysiswas performed by a single blinded person who had no knowledgeof the treatment groups.

Intravital Microscopy of Rolling in Dorsal Skinfold Window Chamber.Intravital microscopy of dorsal skinfold window chambers wasperformed on a Nikon Eclipse 80i microscope equipped with anEXFO XCite 120 xenon light source for epifluorescence mi-croscopy; Nikon 32 B2E/C, G2E/C, and UV2E/C filter cubes; and10× and 20× objectives. Video was taken by using a high-speedPhotometrics HQ2 CCD camera cooled to −30 °C controlled byNikon NIS Elements Advanced Research software with a 2Dobject tracking package (Version 3.22.00; Laboratory Imaging)running on an HP DC7900 PC. Mice were anesthetized by usingisoflurane for the duration of the imaging. The glass coverslipwas removed from the window chamber before video acquisition.One-minute videos were taken of perfused vessels near implantsimmediately after surgery and 1 d after surgery. Green fluores-cent positive cells captured in the video were separated intothree categories, flowing, rolling, and adhered, based on thespeed of the cells. Flowing cells had no abnormal changes invelocity. Rolling cells exhibited a stop–start/variable velocitycharacteristic, whereas adhered cells remained stationary for atleast 10 s. In addition, the highest intensity for cells in focuswas recorded. One person did all of the video acquisitionand analysis.

Flow Cytometry. Immunostaining and flow cytometry analyseswere performed according to standard procedures and analyzedon a FACSCanto flow cytometer (BD Biosciences). Monoclonalantibodies to CD45, CD11b, Ly6C, and Gr1 (Abcam, Biolegend,and eBiosciences) were used to detect and sort inflammatory andanti-inflammatory monocytes and macrophages.

Cell Culture.Human primary isolated pericyte capillary cells werecultured in Pericyte Growth Medium (Angio-Proteomie) in the


www.pnas.org/cgi/content/short/1221309110

presence of 1 μM sphingosine 1-phosphate (S1P) receptorcompounds for 160 h with medium change every 48 h. Humanumbilical vein ECs (HUVECs) were cultured in EBM-2 medium(Lonza) with 5% FBS and 1% penicillin/streptomycin (Pen/Strep) in the presence of 1 μM S1P receptor compounds for 24 h.Human THP-1 monocytes were differentiated to macrophageswith 100 nM phorbol myristate acetate treatment for 2 d andpolarized to M1 and M2 macrophages with 20 ng/mL IL-4 or5 ng/mL TNF-α + 20 ng/mL IFN-α, respectively. Murine primaryisolated AMs and IMs were serum starved for 2 h before 1-htreatment with 12.5 nM S1P receptor compounds and chemo-taxis assays in transwell chambers. For transwell migration as-says, cells were plated at the top of 8-μm pore-size transwellinserts and allowed to migrate toward 12.5 nM SDF-1 for 4 h.Murine RAW 264.7 cells (ATCC) were cultured in high-glucoseDMEM supplemented with 10% FBS, 2 mM L-glutamine, 1%Pen/Strep, and 1 mM sodium pyruvate. Macrophages were po-larized to M1 cells by treatment with 1 μg/mL LPS and 20 ng/mLIFN-γ, or M2 cells by treatment with 10 ng/mL IL-4 for 16 h.Mouse WEHI–274.1 monocytes between passages 25 and 30were cultured in low-glucose DMEM supplemented with 5%FBS and 1% Pen/Strep. For monocyte polarization, LPS (Sigma)was dissolved in the medium at a concentration of 100 ng/mLMonocytes were incubated in this medium at 37 °C in 5% CO2just before seeding. PLAGA thin films were placed in 96-wellplates, and 2.5 × 105 cells per mL were seeded onto them andcultured for as long as 14 d. The films were recovered at severaltime points: 2 h after incubation on day 0 and on days 1–7, 10,and 14, for a total of 30 films.

Protein Analysis. For Western blotting analysis, total cellularprotein was isolated, quantified, and electrophoresed by standardmethods. Proteins were transferred to low-fluorescence PVDFmembrane, blocked with blocking buffer for fluorescent Westernblotting (Rockland Immunochemicals). Membranes were in-cubated with primary antibodies overnight for S1P1 (NovusBiologicals; NB120-11424), S1P3 (Novus Biologicals; NBP1-95141), and β-actin (Sigma Aldrich). After washing with Tris-buffered saline with Tween-20 (TBST), blots were incubatedwith the appropriate infrared-conjugated secondary for 45 min,washed in TBST, and imaged on the LI-COR Odessey infraredimaging system.

Real-Time PCR. Total RNA was isolated by Ribozol extraction, andcDNA was generated from 500 ng of total RNA by reversetranscription with random primers (Applied Biosystems). Quan-titative PCR with SYBR green was performed on a StepOne Plusthermocycler (Applied Biosystems). The S1P3 receptor mRNAexpression (F: 5′-GTTACTTCAACAGTCCACGAGA-3′; R: 5′-AGATGCGCCTTGCAGAA-3′) was normalized to GAPDH asan internal control (F: 5′-ACCACAGTCCATGCCATCAC-3′;

R: 5′-TCCACCACCCTGTTGCTGTA-3′) (IDT). Quantitationwas performed by the ΔΔCT method.

Cytokine Measurement. Protein isolated from backpack tissue andcell culture-conditioned medium was assayed for cytokine releaseby using Luminex bead arrays (Millipore). Panels consisting ofinterferon-γ (IFN-γ), interleukin 1-α (IL-1α), interleukin 1-β(IL-1β), interleukin 6 (IL-6), interleukin 12 (IL-12 - p40), in-terleukin 17 (IL-17), tumor necrosis factor α (TNF-α), interleukin12 (IL-12 - p70), granulocyte colony stimulating factor (G-CSF),granulocyte macrophage colony stimulating factor (GM-CSF),macrophage colony stimulating factor (M-CSF), interleukin 5 (IL-5), interleukin 7 (IL-7), interleukin 8 (IL-8), interleukin 9 (IL-9),interleukin 2 (IL-2), interleukin 3 (IL-3), interleukin 4 (IL-4),interleukin 10 (IL-10), interleukin 13 (IL-13), interleukin 15 (IL-15), eotaxin, interferon-inducible protein 10 (IP-10), macrophageinflammatory protein 2 (MIP-2), keratinocyte chemoattractant(KC), leukemia inhibitory factor (LIF), monocyte chemo-attractant protein 1 (MCP-1), macrophage inflammatory protein1α (MIP-1α), macrophage inflammatory protein 1β (MIP-1β),monokine induced by gamma interferon (MIG), epidermal growthfactor (EGF), fibroblast growth factor 2 (FGF-2), transforminggrowth factor α (TGF-α), Fms-related tyrosine kinase 3 ligand(Flt-3L), fractalkine, interferon α2 (IFNα2), interferon γ (IFNγ),growth-related oncogene (GRO), monocyte chemoattractantprotein 3 (MCP-3), macrophage derived chemokine (MDC),platelet derived growth factor AA (PDGF-AA), platelet derivedgrowth factor BB (PDGF-BB), soluble cluster of differentiation 40ligand (sCD40L), interleukin 17a (IL-17a), interleukin 1 receptorantagonist A (IL-1RA), vascular endothelial growth factor(VEGF), tumor necrosis factor β (TNF-β) and regulated on ac-tivation, normal T-cell expressed and secreted (RANTES) wereused to quantify the cytokine secretion profile of monocytes,HUVECs, and in tissue protein ELISA for SDF-1α and MCP-1(R&D Systems) were used to quantify protein levels from tissueand HUVEC lysate.

Statistical Analyses. All statistical analyses were performed byusing Minitab 15 statistical software (Minitab). Results are pre-sented as mean ± SEM, unless otherwise noted. Comparisonswere made by using a one-way ANOVA, followed by Tukey’s testfor pairwise comparisons. Diameter analysis was performed byusing a general linear model (GLM) ANOVA with an un-balanced nested design, followed by Tukey’s test for pairwisecomparisons. The model for the GLM ANOVA was as follows:drug group film number (drug group), where drug group rep-resents S1P, for example, and film number represents eachn number (i.e., mouse 5 left film). Significance was asserted atP < 0.05. Where specified, power calculations were performedwith α = 0.05 and power = 0.80 to determine statistically sig-nificant sample size.



Fig. S1. FTY720 promotes the generation of a regenerative module after injury and implantation. Microvascular injury and biomaterial implant leads to theproduction of proinflammatory cytokines and chemokines (e.g., MCP-1) from local and recruited cells. This process enhances the recruitment of CX3CR1 lowIMs (orange cells), which clear debris and pathogens in the inflamed tissue (Left). FTY720, through S1P3 agonism, enhances the production of proregenerativechemokines (e.g., SDF-1α) by ECs and the chemotaxis of CX3CR1 high AMs (green cells), resulting in the recruitment of AMs, which contribute to neo-vascularization and arteriogenesis (Right).

FTY7

20

FTY7

20

Infla

mm

ator

y G

row

th F

acto

r

Th1/

Th2

Che

mok

ine

Ud. 0

1

2

3

4

>5

PLAG

A

PLAG

A

*

* * *

*

*

*

Sha

m

Acu

te P

LAG

A A

cute

FTY

720

A B

Fig. S2. (A) FTY720 reduces inflammatory cytokine concentration in tissue. Cytokine quantification (fold increase over sham) in tissue surrounding FTY720-loadedimplants shows significant reduction of inflammatory cytokines with FTY720 3 d after implantation. (B) Microvascular networks surrounding PLAGA implants at0 and 7 d after implantation when implanted on day of surgery (acute). Red arrows indicate arterioles enlarged around PLAGA films. *P < 0.05 compared withPLAGA. (Scale bars: 100 μm.) Ud., undetectable by luminex.



FSC

SS

C

CD45

CD

11b

GR1

Ly6C

PLAG

A S1

P

PLAGA S1P

A

CD

11b+

cel

l per

ve

ssel

leng

th

0

10

20

30

40

50

FTY720 PLAGA

0

10

20

30

40

50

AM IM

CD45 CX3CR1

CD45 CX3CR1 C

D11

b

CD

11b

CD

11b

CD

11b

IM AM

IM AM

PLAGA

FTY720

CD45

CD

11b

CX3CR1

CD

11b

CX3CR1

Ly6C

B D

C

Spleen BM %

CD

45+ C

D11

b+ c

ells

% C

D45

+ CD

11b+

cel

ls

Spleen BM

E

IM

AM

FSC

SSC CD45

CD

11b

Gr1

Ly6C

AM

IM

FSC

SSC CD45

CD

11b

Gr1

Ly6C

AM

IM

Blood

Tissue

FSC

SSC CD45

CD

11b

Gr1

Ly6C

AM

IM

Bone Marrow

Dorsal Tissue

Dorsal Tissue

Fig. S3. S1P regulates tissue monocytes without mobilizing from spleen or BM. (A) Reduction of CD11b+ cells (arrows) lining lumen of vessels (arrows) aroundS1P-loaded implants. (B) Gating strategy for AMs/IMs from blood, dorsal tissue, and BM in wild-type mice. IM were SSCloCD45+CD11b+Ly6C+Gr1+ and AM wereSSCloCD45+CD11b+Ly6C−Gr1−. (C) Gating strategy to distinguish AMs and IMs in tissue in CX3CR1–eGFP mouse model. IM were CD45+CD11b+CX3CR1loLy6C+

and AM were CD45+CD11b+CX3CR1hiLy6C− AM. (D) FTY720 enhances proportion of AMs in dorsal tissue of CX3CR1–eGFP mice. (E) Local FTY720 delivery didnot alter the proportion of AMs or IMs in spleen or BM (Scale bar: 100 μm).

0 50 100

150

200

250

300

350

400

450

500

550

600

650

0

10

20

30

40

010

020

030

040

050

060

070

080

090

010

0011

0012

0013

000

10

20

30

40

Anti-Inflammatory Monocyte Adoptive Transfer

PLAGA FTY720 A B

Distance from nearest vessel (µm) Distance from nearest vessel (µm)

Freq

uenc

y

Freq

uenc

y

Fig. S4. FTY720 recruits anti-inflammatory monocytes from blood to tissue surrounding implants. (A and B) After AM adoptive transfer, median DiI+ celldistance from vessels is significantly lower in FTY720-treated tissue (B) than PLAGA-treated tissue (A). Red line represents median.



Fig. S5. Inflammatory and regenerative cytokine secretion from HUVECs, AMs, and IMs treated with FTY720/SEW2871 for 1, 6, and 24 h. (A–C) Inflammatory(Left) and regenerative (Right) cytokine quantification (fold increase over vehicle) in conditioned medium from HUVECs treated with 1 μM FTY720 or SEW2871for 1 (A), 6 (B), and 24 (C) h. S1P1 activation promotes regenerative cytokine secretion early and maintains a balance late. S1P3 activation enhances secretion ofregenerative cytokines from 1 to 24 h treatment. (D and E) Inflammatory cytokine quantification (fold increase over vehicle) in conditioned media from AMs(D) and IMs (E) treated with 1 μM FTY720 or SEW2871 for 1 h. S1P1 and S1P3 regulate the secretion of distinct cytokines from AMs and IMs and decrease mostinflammatory cytokines relative to vehicle. *P < 0.05 compared with SEW2871.

0 50 100 150 2000

2

4

6

8

10

* * *

* * *

Cel

ls/F

OV

Hours

Vehicle

FTY720

SEW2871

Fig. S6. S1P1 activation enhances pericyte proliferation. A concentration of 1 μM SEW2871 and FTY720 enhance proliferation of primary isolated pericytesafter 48 h of treatment. *P < 0.05 compared with vehicle.



A B

Medial

Caudal

Lateral

20 40 60 80 100

% C

hang

e in

Tortu

osity

CX3CR1+ cells

PLAGA FTY720

0

10 20 30 40 50

0

Fig. S7. Tortuosity of ischemic vessels correlated with CX3CR1+ cell recruitment with FTY720. (A) Diagram of smooth muscle actin-stained whole-mountedspinotrapezius muscle with ligation (red cross) and film implantation (blue circle). (B) Enhanced percentage change in tortuosity (actual vessel length divided byend-to-end length) of vessels was observed in the remodeling watershed of FTY720-treated muscles along with higher CX3CR1+ content. Tortuosity waspositively correlated with CX3CR1+ cell content in watershed. FTY720 significantly enhanced tortuosity (average = 17% increase) relative to unloaded PLAGA(average = 10% increase).

FTY7

20

PLAG

A

0

20

40

60

80A B

1mm

C

FTY720

PLAGA

Con

tact

ang

le

PLAGA PLAGA+Cells

0 1 2 30

10

20

30

0 1 2 30

10

20

30

0 1 2 30

10

20

30

# of beads # of beads # of beads

# of

cel

ls

# of

cel

ls

# of

cel

ls

PLA

GA

FTY

720

PLAGA+FTY720+Cells

M0 M1 M2

D

E

Fig. S8. FTY720 reduces the hydrophobicity of PLAGA and monocyte spreading but not phagocytosis. (A and B) FTY720 lowers the contact angle of water on,and increases hydrophilicity of, PLAGA films. (C) LPS stimulated WEHI monocytes 7 d after being seeded on unloaded (Upper) PLAGA polymer films are morespread out than those seeded on FTY720-loaded (Lower) films. (D) PLAGA degrades at a much faster rate than FTY720-loaded films in the presence of cells. (E)FTY720 does not alter the phagocytosis of M0, M1, or M2 macrophages. P < 0.05 compared with PLAGA (Scale bar: 20 μm).


supporting information - pnas · 8/3/2013 · supporting information awojoodu et al....

Documents