Journal ofMaterials Chemistry B

PAPER

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article OnlineView Journal | View Issue

A dual gold nano

aDepartment of Biomedical Engineering, U

78712, USAbDepartment of Electrical and Computer En

Austin, TX 78712, USA. E-mail: laura.suggs@

Cite this: J. Mater. Chem. B, 2014, 2,8220

Received 16th June 2014Accepted 8th September 2014

DOI: 10.1039/c4tb00975d

www.rsc.org/MaterialsB

8220 | J. Mater. Chem. B, 2014, 2, 822

particle system for mesenchymalstem cell tracking

L. M. Ricles,a S. Y. Nam,ab E. A. Trevino,a S. Y. Emelianovab and L. J. Suggs*a

Stem cell-based therapies have demonstrated improved outcomes in preclinical and clinical trials for

treating cardiovascular ischemic diseases. However, the contribution of stem cells to vascular repair is

poorly understood. To elucidate these mechanisms, many have attempted to monitor stem cells

following their delivery in vivo, but these studies have been limited by the fact that many contrast agents,

including nanoparticles, are commonly passed on to non-stem cells in vivo. Specifically, cells of the

reticuloendothelial system, such as macrophages, frequently endocytose free contrast agents, resulting

in the monitoring of macrophages instead of the stem cell therapy. Here we demonstrate a dual gold

nanoparticle system which is capable of monitoring both delivered stem cells and infiltrating

macrophages using photoacoustic imaging. In vitro analysis confirmed preferential labeling of the two

cell types with their respective nanoparticles and the maintenance of cell function following nanoparticle

labeling. In addition, delivery of the system within a rat hind limb ischemia model demonstrated the

ability to monitor stem cells and distinguish and quantify macrophage infiltration. These findings were

confirmed by histology and mass spectrometry analysis. This work has important implications for cell

tracking and monitoring cell-based therapies.

Introduction

Stem cell therapy shows great potential to treat a large variety ofdiseases, including cardiovascular diseases, which are thenumber one cause of death globally.1 In particular, bone-marrow derived mesenchymal stem cells (MSCs) are advanta-geous in that they possess angiogenic properties, are easilyobtained in large numbers and expandable in culture, and arepart of the ischemic response.1 Numerous preclinical andclinical trials have investigated the therapeutic benets of stemcell therapy for cardiovascular diseases. However, advances inthe eld of stem cell therapy have been limited by the inabilityto track administered cells,2 which would provide informationconcerning cell engrament and persistence, mechanisms ofvascular regeneration, and the role of MSCs in vascular repair.

Conventional methods for assessing the biological mecha-nisms underlying disease states and potential therapies rely onpostmortem histology, which only offers endpoint measure-ments and requires a large number of animals to be sacricedin order to produce statistically signicant results. A more idealcell tracking method would involve using noninvasive longitu-dinal imaging to monitor cells. Towards this end, many contrastagents are currently being investigated to label cells for cell

niversity of Texas at Austin, Austin, TX

gineering, University of Texas at Austin,

mail.utexas.edu

0–8230

tracking purposes, including reporter genes,3–6 radionuclides,6–8

uorescent probes,9–11 and nanoparticles.4,8,12–14 Nanoparticles,such as quantum dots, iron oxide nanoparticles, and plasmonicnanoparticles (gold and silver), offer many advantages overother contrast agents. For example, nanoparticles can be opti-mized to promote cellular uptake through shape, size, andsurface coating modication12,15–19 and allow for long-termmonitoring of cells.12–14,20 However, viable and non-viable cellscannot be distinguished using nanoparticle contrast agents. Asa result, it is not possible to detect if a cell is dead and has beenendocytosed by macrophages, leading to a transfer of contrastagent from the labeled cells to macrophages. Other investiga-tors have found nanoparticle transfer to macrophages,4,21,22

resulting in the monitoring and tracking of macrophagesinstead of the stem cell therapy.

Thus, the goal of this study is to develop a nanoparticlesystem which is capable of tracking stem cells following deliveryin vivo and also monitoring macrophage inltration andtransfer of contrast agents from stem cells to macrophages as aresult of macrophage endocytosis. Macrophages are known tohave key roles in wound healing and vascular regeneration23–26

and to be inuenced by and exert paracrine effects on stemcells, including MSCs.27–29 The nanoparticle system will consistof gold plasmonic nanoparticles. Gold nanoparticles can besynthesized in various shapes and sizes and their absorptioncharacteristics can be tuned to maximally absorb in the near-infrared region, where the absorption from endogenous tissueis the lowest. Gold nanoparticles are also non-toxic to cells in

This journal is © The Royal Society of Chemistry 2014

Fig. 1 Outline of the dual nanoparticle system for labeling MSCs withgold nanorods and macrophages with spherical gold nanospheres.The nanoparticles exhibit different optical absorption characteristicsfollowing endocytosis by cells and can thus be distinguished whenimaged within the tissue optical window (highlighted wavelengthregion).

Paper Journal of Materials Chemistry B

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

certain formulations12,30,31 and exhibit surface plasmon reso-nance, which contributes to their superior optical absorptionproperties,32,33 making them ideal contrast agents for photo-acoustic imaging.20,34 Fig. 1 shows the outline of the dualnanoparticle system consisting of gold nanorods to label MSCsand gold nanospheres to label macrophages. This system isdelivered within a 3D PEGylated brin gel, which promotes theangiogenic potential of MSCs and leads to tubular networkformation within the gels, as demonstrated by previous work inour lab.35 The gold nanorods were selected because theseparticles maximally absorb in the near-infrared region. On theother hand, gold nanospheres maximally absorb in the visiblelight region (520 nm), but plasmon coupling following nano-sphere endocytosis by cells leads to peak broadening. Thus, thegold nanospheres will only be detected using photoacousticimaging when they are endocytosed by macrophages andimaged within the tissue optical window of 650–900 nm. Toevaluate this nanoparticle system, various in vitro assays wereperformed, including labeling of cells with the nanoparticlesand the assessment of cell function and viability followingnanoparticle labeling. In addition, in vitro and in vivo photo-acoustic imaging experiments were performed to assess thefeasibility of monitoring the two cell types. Histological analysisand mass spectrometry were also performed to verify the in vivophotoacoustic imaging results. This study has importantimplications for cell tracking and the role of MSCs andmacrophages in vascular regeneration.

Experimental sectionCell culture

Rat mesenchymal stem cells were isolated from the bonemarrow of Lewis rats (200–300 g). The femoral marrow cavitywas ushed and adherent cells were collected and cultured inDulbecco's Modied Eagle Media (DMEM) (Invitrogen, Carls-bad, CA) supplemented with 10% fetal bovine serum (FBS), 1%glutamax, and 1% penicillin–streptomyocin. Non-adherent cells

This journal is © The Royal Society of Chemistry 2014

were removed aer 24 hours by replacing the media. The mediawas then changed every two days and the cells were passagedonce reaching 80% conuency. The cells were cultured understandard cell culture conditions (37 �C, 5% CO2). Passage 3–7cells were used in all studies.

Murine macrophages (J774A.1, ATCC) were cultured understandard cell culture conditions (37 �C, 5% CO2) in DMEM with10% FBS. The media was changed every two days and passagedonce reaching 80% conuency. Passage 10–20 cells were used inall studies.

Nanoparticle synthesis

Spherical gold nanoparticle seeds (20 nm in diameter) weresynthesized by heating 100 mL of distilled (DI) water to 100 �Cand adding 1 mL of 10 mg mL�1 gold(III) chloride hydrate(HAuCl4). Then, 5mL of 10mgmL�1 sodium citrate dissolved inDI water was added and the reaction was allowed to cool toroom temperature. Gold nanoparticles 50 nm in diameter weresynthesized by mixing 7.5 mL of 10 mg mL�1 HAuCl4, 15.61 mLof 0.2 M ammonium hydroxylamine, 750 mL of DI water, and 50nm of seed solution while stirring. The surface of the goldnanoparticles was coated with methoxy-polyethylene glycol–thiol (mPEG–SH) at a concentration of 0.1 mM and allowed toreact for 30 minutes on a shaker, followed by centrifugalltration (Amicon ultra-15, Millipore) to remove residual PEG.

Gold nanorods (NRs) were synthesized via seed mediatedgrowth as previously described.36,37 Briey, the seed solutionwas synthesized by mixing 5 mL of cetyl trimethylammoniumbromide (CTAB) (0.2 M) with 5 mL of HAuCl4 (0.5 mM) whilestirring vigorously. Then 0.6 mL of ice cold sodium borohydride(NaBH4) was added to the solution. The growth solution wasprepared by mixing 380 mL of 0.2 M CTAB, 8 mL of 0.01 M silvernitrate (AgNO3), and 40 mL of 0.01 M HAuCl4 while stirring at30 �C. Then 4.4 mL of 0.1 M ascorbic acid (C6H8O6) was addeddrop wise, resulting in a color change from yellow to colorless.To produce NRs, 0.958 mL of gold nanoseeds was added to thegrowth solution and allowed to stir for 3 minutes. The solutionwas allowed to age overnight, centrifuged twice at 14 000 rcf for45 minutes, and resuspended in ultraltrated (18 MU cm,Thermo Scientic Barnstead Diamond water puricationsystems) deionized water.

The NR surface was modied by replacing CTAB with mPEG–SH through ligand exchange. Briey, an equal volume of mPEG–SH (0.2 mM) was added to the gold NR solution, sonicated for 1minute, and allowed to react overnight. Excess PEG wasremoved by centrifugal ltration (Amicon ultra-15, Millipore) at2000 rcf for 10 minutes. Finally, a layer of silica was depositedon the surface of the PEG modied NRs through a modiedStober method, as previously described.38–40 Under vigorousstirring, 2 mL of PEGylated NRs was added to 3 mL of iso-propanol, followed by 1.2 mL of tetraethyl orthosilica (TEOS)(3%) in isopropanol and 1.2mL of ammonium hydroxide (8.4%)in isopropanol. The reaction was allowed to mix for 2 hours,followed by centrifugal ltration at 500 rcf for 15minutes, twice.The NRs were then coated with a layer of poly-L-lysine (5 kD;Sigma-Aldrich) at a concentration of 0.833 mM and allowed to

J. Mater. Chem. B, 2014, 2, 8220–8230 | 8221

Journal of Materials Chemistry B Paper

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

react for 2 hours, followed by centrifugation at 3000 rcf for 10minutes and then 5000 rcf for 7 minutes.

The size of the nanoparticles was characterized usingtransmission electron microscope (TEM) analysis and dynamiclight scattering (DLS), and the surface charge was characterizedusing zeta potential measurements. For TEM, an FEI Tecnaitransmission electron microscope tted with a top mount AMTAdvantage HR 1k � 1k digital camera and operating at 80 kVwas used. A drop of nanoparticle solution was placed on carboncoated copper 300 mesh grids (Electron Microscopy Sciences,Hateld, Pa) which had been glow discharged. ImageJ analysiswas used to quantify the size of the NRs and thickness of thesilica coating (n ¼ 100). A DelsaNano (Beckman Coulter, Inc.,Brea, CA) was used for DLS analysis of spherical nanoparticles.Nanoparticle solutions were loaded into a cuvette and readingswere taken at a temperature of 25 �C. For size measurements, 5repetitions were performed for each nanoparticle solution with60 readings per repetition. A Zetasizer Nano ZS (Malvern) wasused to measure the zeta potential of nanoparticle solutions. Atotal of 3 measurements, with 15 repetitions, were collected at atemperature of 25 �C.

Nanoparticle labeling of cells

Gold nanosphere media was made following PEGylation of 50nm nanospheres. The nanospheres were sterile ltered using a0.22 mm lter, centrifuged in a centrifugal lter tube (Amiconultra-15, Millipore) at 3000 rcf for 5 minutes, and resuspendedin cell culture media at a concentration of approximately 1012

NPs mL�1. Gold NR media was made following PLL coating ofsilica NRs. The NRs were sterile ltered using a 0.22 mm lterand centrifuged at 3000 rcf for 10 minutes. The supernatant wasthen centrifuged at 5000 rcf for 7 minutes. The NRs wereresuspended in cell culture media at a concentration ofapproximately 1012 NRs mL�1. Macrophages were incubatedwith nanosphere media (200 mL cm�2) and MSCs were incu-bated with NR media (200 mL cm�2) for 24 hours.

Bright eld and dark eld microscopy was used to qualita-tively analyze nanoparticle uptake by macrophages and MSCs.Following nanoparticle incubation, the media was removed andthe cells were washed with phosphate buffered saline (PBS) toremove excess particles. The cells were then incubated withDAPI (1 mg mL�1) for 15 minutes, followed by PBS washes. ALeica DMI2000B microscope, equipped with a Leica DFC290camera, was used to obtain bright eld and dark eld images.Controls consisted of cells not incubated with nanoparticles.

Inductively coupled plasma mass spectrometry (ICP-MS) wasused to quantitatively analyze nanoparticle uptake by macro-phages and MSCs in 2D, as well as by macrophages in 3DPEGylated brin gels. The PEGylated brin gels were synthe-sized as described below and contained 50 000 cells per mL and1012 NPs mL�1. Following nanoparticle loading in 2D, the cellswere collected using trypsinization and counted. To collect thecells in 3D, the gels were digested by incubation with 500 mL ofbovine pancreatic trypsin (5 mg mL�1 in 0.9% sodium chloride)per 1 mL of gel for 1 hour. The reaction was quenched withserum containing media and the solution was ltered with a 70

8222 | J. Mater. Chem. B, 2014, 2, 8220–8230

mm and 40 mm cell strainer and centrifuged at 600 rcf for 7minutes to collect the cells. To prepare the samples for ICP-MS,all of the solutions were incubated in 300 mL of 70% nitric acidat 60 �C overnight. The solutions were then diluted to 3.7%nitric acid using ultrapure water prior to running the sampleson the ICP-MS machine. The samples containing MSCs incu-bated with 50 nm gold nanospheres were prepared for ICP-MSanalysis by heating the samples at 150 �C and bringing them todryness. The samples were then resuspended in 200 mL of aquaregia and further diluted with 2% HCl. Nanoparticle loadingwas quantied using a standard curve. Standard solutions weremade by diluting AAS gold standard (Sigma-Aldrich, St. Louis,MO) in the same background solution as the samples. Thenumber of nanoparticles per cell was calculated based on theNP dimensions and cell number.

MSCs and macrophages were co-cultured in a 3D PEGylatedbrin gel to investigate the nanoparticle system in a 3D envi-ronment. PEGylated brin gels were synthesized as describedbelow and contained 25 000 MSCs mL�1; 50 000 macrophagesper mL; and 1014 NPs mL�1. The gels were made in 12 welltranswell plates with 8.0 mm pore sizes and were incubated at37 �C in MSC growth media.

Cell viability

Cell viability was assessed following nanoparticle labeling usingLIVE/DEAD viability staining and an MTS proliferation assay.LIVE/DEAD staining was performed by incubating the cells witha working solution of calcein AM (4 mm) and ethidium homo-dimer-1 (1 mm) for 45 minutes at 37 �C. The cells were washedwith PBS and imaged using uorescence microscopy (LeicaDMI2000B microscope equipped with a Leica DFC 290 camera).

Cell proliferation following nanoparticle labeling wasassessed using MTS. Cells were incubated with DMEM mediacontaining 20% MTS solution for 4 hours at 37 �C. The super-natant was collected and the absorbance value measured at 490nm (n ¼ 3). A blank sample containing no cells was subtractedfrom all measurements.

The maintenance of cell function in 3D PEGylated brin gelswas assessed aer laser irradiation by examining cellmorphology and tubular network formation. Difunctional suc-cinimidyl glutarate PEG (8 mg mL�1 in PBS without calcium;NOF America) was added to brinogen (80 mg mL�1 in PBSwithout calcium; Sigma) in a 1 : 1 volume ratio, and the reactionwas allowed to take place at room temperature for 3–5 minutes.An equal volume of MSCs as the brinogen and PEG mixturewas then added at a concentration of 200 000 cells per mL. Gelswith either NR labeled MSCs or unlabeled MSCs were made.The reaction then underwent gelation by adding an equalvolume solution of thrombin (25 U mL�1 in 40 mM CaCl2;CalBiochem). The nal concentrations in the gel were 10 mgmL�1 of brinogen; 1 mg mL�1 of SG-PEG-SG; 50 000 cells permL; and 12.5 U mL�1 of thrombin. The gels were made in 4 wellchambered coverglass slides (Laboratory-Tek II; Nalge Nunc)and incubated at 37 �C in MSC growth media.

The gels were irradiated at various time points (day 1, 4, and7) using a 800 nm laser with a uence of 10 mJ cm�2 for 100

This journal is © The Royal Society of Chemistry 2014

Paper Journal of Materials Chemistry B

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

pulses. Control conditions consisted of the same gel formula-tions without laser irradiation. At day 7, the cells were stainedwith calcein AM (10 mM) and 3D z-stacks were taken using aLeica SP2 AOBS confocal microscope (10�magnication; 512�512 resolution). The images covered approximately 1 mm with astep size of approximately 3 mm. The 3D images were recon-structed using ImageJ.

In vitro photoacoustic imaging

A tissue mimicking phantom with varying ratios of NR labeledMSCs and nanosphere labeled macrophages was prepared forcombined ultrasound and photoacoustic (US/PA) imaging. Thebottom layer of the phantom consisted of gelatin (8% by weight;Sigma-Aldrich) with 15 mm diameter silica particles (0.2% byweight; Sigma-Aldrich) for ultrasonic scattering. The cellinclusions (20 mL) were placed on top of the bottom layer andconsisted of gelatin mixed with cells. Each cell inclusion had atotal of 10 000 cells.

Photoacoustic signals were captured using the Vevo LAZRsystem (VisualSonics, Inc.) including a tunable laser as well as a40 MHz array transducer combined with an optical ber bundleto deliver the laser pulses. RF signals were acquired for laserwavelengths ranging from 680–970 nm and with a laser uenceof 5–15 mJ cm�2. The transducer was mechanically scannedalong the tissue mimicking phantom to obtain 3D photo-acoustic information. The signals exported from the imagingsystem were post-processed for beamforming, laser uencecompensation, image reconstruction, and signal quantication.For spectral analysis, the quantied photoacoustic signals wereused to calculate the ratio of the NR labeled MSCs and thenanosphere labeled macrophages in each voxel by comparisonto weighted sums of their optical absorption spectrum. Then,the estimated ratios in the voxels were averaged to calculate theoverall cell ratio for whole inclusions.

Ischemic injury and in vivo ultrasound and photoacousticimaging

Animal handling and care followed the recommendations of theNational Institutes of Health (NIH) guidelines for the care anduse of laboratory animals. All protocols were approved by theAnimal Care Committee at the University of Texas at Austin.Lewis rats (11 weeks) weighing 250–300 g were used. A femoralartery ligation was performed in Lewis rats (11 weeks, male) toinduce hind limb ischemia. Through a small incision on themedial side of the thigh, the femoral artery of a single hind limbwas separated from the nearby nerve and vein and ligatedimmediately distal to the inferior epigastric artery and proximalto the branch point of the popliteal and saphenous arteriesusing Prolene 5-0 sutures. The ligated segment was then excisedand the skin incision closed with interrupted sutures. Theanimal was allowed to recover overnight and the following day(about 24 hours later), MSCs were injected intramuscularly intothe gastrocnemius muscle of the ligated limb. PEGylated brininjections were prepared by combining difunctional succini-midyl glutarate PEG (4 mg mL�1 in PBS without calcium; NOFAmerica) to brinogen (40 mg mL�1 in PBS without calcium;

This journal is © The Royal Society of Chemistry 2014

Sigma) in a 1 : 1 volume ratio. An equal volume of MSCs labeledwith gold NRs was mixed with the PEGylated brinogen solu-tion in a 1 : 1 volume ratio at a concentration of 13 � 106 cellsper mL. Next, 50 nm PEGylated gold nanospheres was added tothe cell/PEGylated brin solution at a concentration of 1014 NPsmL�1. The solution was then loaded into a 23 G needle syringe,followed by an equal volume of thrombin (25 U mL�1 in 40 mMCaCl2). The solution was mixed thoroughly within the syringeand the gel solution (300 mL) was injected into the gastrocne-mius of the rat. The nal concentrations in the gel were 5 mgmL�1 of brinogen; 0.5 mg mL�1 of SG-PEG-SG; 3.33 � 106 cellsper mL; and 12.5 U mL�1 of thrombin. Prior to delivery, MSCswere uorescently labeled with CellTracker CM-DiI (Invitrogen).The cells were incubated with CM-DiI (15 mM) at 37 �C for 8minutes and then 4 �C for 15 minutes, washed with PBS, andresuspended in DMEM.

Ultrasound and photoacoustic signals were captured usingthe Vevo LAZR (VisualSonics, Inc.) system. A 40 MHz arraytransducer incorporated with a ber bundle was submerged in awater bath, and the rat hind limb was coupled through atransparent lm at the bottom of the water bath. RF signalswere acquired for laser wavelengths ranging from 680–970 nmand with a laser uence of 5–10 mJ cm�2. Post-processing wasthe same as that for the in vitro photoacoustic imaging. Theprocessed photoacoustic and spectroscopic images were over-laid with ultrasound images using a user-dened threshold ofthe photoacoustic signals.

Histology

At the terminal endpoint of the study, animals were sacricedand the gastrocnemius muscles were isolated, embedded inoptimal cutting temperature (OCT) compound, submerged inliquid nitrogen-cooled isopentane, and stored at �80 �C untilfurther processing. The samples were cut into 12 mm thicksections using a cryostat and placed onto positively chargedmicroscope slides. The tissue sections were xed in 10%formalin for 15 minutes. Immunohistochemical and histo-chemical staining were subsequently performed.

For immunostaining, the slides were incubated in 0.125%trypsin solution for 20 minutes at 37 �C. They were then per-meabilized in 0.5% Triton X-100 in TBST for 30 minutes andblocked in 10% normal goat serum for 1 hour. The slides wereincubated overnight at 4 �C in the primary antibody (1 : 50dilution in 2.5% normal goat serum) (ED1; Millipore), followedby incubation with a uorescence-conjugated secondary anti-body (1 : 200 in 2.5% normal goat serum) (Alexa Fluor 488 GoatAnti-Mouse Immunoglobulin G; Life Technologies). As a nega-tive control, normal immunoglobulin G was used instead of theprimary antibody. The samples were counterstained with DAPI(5 mg mL�1) for 15 minutes, mounted, and viewed using auorescence microscope (DMI2000B; Leica).

TUNEL staining was performed with the APO-BrdU TUNELAssay Kit (Invitrogen). The slides were incubated in 0.125%trypsin for 20 minutes at 37 �C, followed by permeabilization in0.5% Triton X-100 for 30 minutes. The slides were then incu-bated with TdT and BrdUTP for 2 hours at 37 �C. Alexa Fluor 488

J. Mater. Chem. B, 2014, 2, 8220–8230 | 8223

Journal of Materials Chemistry B Paper

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

dye-labeled anti-BrdU antibody was applied for 30 minutes. Theslides were then counterstained with DAPI (5 mg mL�1) for 15minutes, mounted, and viewed using a uorescencemicroscope(DMI2000B; Leica).

Fluorescence-activated cell sorting of macrophages

Seven days following intramuscular injection of nanorodlabeled or unlabeled MSCs and nanospheres within PEGylatedbrin, animals were sacriced and the gastrocnemius musclewas isolated. The muscle samples were nely minced andincubated in 1% collagenase in DMEM for 1 hour at 37 �C underconstant agitation. Samples were then ltered through cellstrainers (40 mm and 70 mm), washed with PBS, and blocked for30 minutes in uorescence-activated cell sorting (FACS) buffer(1% BSA in PBS + 1% sodium azide). Samples were then incu-bated with primary antibody (1 : 50; ED1; Millipore) in FACSbuffer for 45 minutes, washed, and incubated in secondaryantibody (1 : 50; Alexa Fluor 488 Goat Anti-Mouse Immuno-globulin G; Life Technologies) for 45 minutes. Next, sampleswere xed in 4% neutral buffered formalin for 15 minutes,washed, and resuspended in PBS. Samples were then sorted formacrophages using a FACSAria ow cytometer. Following sort-ing, samples were digested for ICP-MS analysis. The cells werecollected by centrifuging at 270 � g for 5 minutes and resus-pended in a small volume of aqua regia solution. The sampleswere heated at 150 �C and brought to dryness. They were thenresuspended in 250 mL of aqua regia and further diluted withultrapure water to give an HCl concentration of 5%. A standardcurve was used to quantify the amount of gold in the cells bydiluting AAS gold standard (Sigma-Aldrich, St. Louis, MO) in thesame background solution as that of the cell solutions. The datawas quantied by calculating the mass of gold per macrophage.

Fig. 2 Characterization of (A) gold nanorods and (B) nanospheres usingzeta potential analysis.

8224 | J. Mater. Chem. B, 2014, 2, 8220–8230

ResultsNanoparticle characterization

The size, shape, and surface charge of gold NRs and nano-spheres were characterized using ultraviolet-light spectroscopy(UV-vis), TEM analysis, DLS, and zeta potential measurements.The absorbance spectra of gold NRs had a peak around 750 nm,which red shied 20 nm following silica coating and an addi-tional 8 nm following poly-L-lysine (PLL) coating (Fig. 2A). Zetapotential measurements conrmed the sequential coatingswere deposited on the NRs (Fig. 2A), as the surface charge was+40 mV for CTAB coated NRs, �0.61 mV for PEG coated NRs,�30.37 mV for silica coated NRs, and +26 mV for PLL coatedNRs. Using TEM analysis, the size of the NRs was measured as55.63 � 9.14 � 15.59 � 3.45 nm, with a silica coating thicknessof 25.04 � 3.06 nm (Fig. 2A). The 50 nm spherical nanoparticleshad an absorbance peak around 520 nm (Fig. 2B), with a redshi of a few nm following PEG coating (data not shown). TEManalysis conrmed the spherical shape of the particles and DLSmeasured the hydrodynamic diameter of the particles as being50.4 nm with a polydispersity index of 0.044 (Fig. 2B). Zetapotential measurements demonstrated that the nanosphereshad a negative surface charge (�46.07 mV), which becamecloser to neutral (�8.41 mV) following PEG coating (Fig. 2B).

Nanoparticle labeling of cells

MSCs were incubated with PLL silica NRs and the uptake by thecells was qualitatively analyzed using microscopy and quanti-tatively analyzed using ICP-MS. Dark eld microscopy demon-strated nanoparticle labeling of MSCs as evident by the yellow/green color of the cells as compared to control cells (Fig. 3A andC). In addition, dark nanoparticle aggregates within cells werevisible under bright eld (Fig. 3B and D). ICP-MS analysis

ultraviolet-light spectroscopy, transmission electron microscopy, and

This journal is © The Royal Society of Chemistry 2014

Fig. 3 Assessment of nanoparticle labeling of cells using dark field andbright field microscopy. Control MSCs are shown in (A) and (B) andnanorod labeled MSCs are shown in (C) and (D). Images are overlaidwith DAPI fluorescence images. Control macrophages are shown in (E)and (F) and nanosphere labeledmacrophages are shown in (G) and (H).MSCs incubated with 50 nm PEGylated nanospheres are shown in (I)and (J).

Fig. 4 Quantification of (A) nanorod (NR) labeling of MSCs and (B)nanosphere (NP) labeling of macrophages (MPH) and MSCs usinginductively coupled plasma mass spectrometry.

Paper Journal of Materials Chemistry B

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

quantied the number of nanorods per MSC as 1.89 � 0.877 �105 nanorods per cell (Fig. 4A). This large extent of cellularuptake of the nanorods can be attributed in part to both thesilica coating, providing a more spherical shape to the particlewhich is more conducive for cellular uptake,16,17,41 and the PLLsurface coating, which is commonly used as a transfectionagent and has been used to promote cellular uptake of ironoxide nanoparticles.42–44

Macrophages were incubated with 50 nm PEGylated goldnanospheres and dark eld and bright eld microscopydemonstrated a large extent of uptake by the cells (Fig. 3G andH) as compared to control cells (Fig. 3E and F). In addition,MSCs were incubated with the same nanoparticles and therewas no evident uptake by the cells (Fig. 3I and J), which wasveried by ICP-MS measurements (Fig. 4B). The reason MSCsdid not uptake these particles can be attributed to the PEGsurface coating, which passivates the particles and gives thema neutral charge, leading to minimal detection and uptake bycells.41,45 Macrophages, on the other hand, are characterizedby their high rate of nonspecic uptake, and thus endocytosethese particles to a large extent.46 Over time the mPEG–SHlayer is slowly displaced by serum proteins, leading to pref-erential uptake by macrophages.47,48 Thus, 50 nm PEGylatedgold nanospheres can be used as a contrast agent to prefer-entially label macrophages and identify the inltration ofmacrophages in vivo. ICP-MS measurements quantied thenumber of gold nanospheres per macrophage as 6.76 � 1.68

This journal is © The Royal Society of Chemistry 2014

� 103 nanospheres per cell in 2D culture (Fig. 4B). Themacrophages were also cultured in 3D PEGylated brin gelscontaining gold nanospheres in order to more closelyresemble the in vivo scenario in which macrophages willinltrate the ischemic area and come in contact with nano-spheres delivered within a 3D hydrogel. The number of goldnanospheres per macrophage in 3D culture as measured withICP-MS was 4.38 � 1.63 � 103 nanospheres per cell. Thesemeasurements took into account the fact that gold nano-spheres not endocytosed by cells could still be entrapped inthe gels. As a result, gels containing only gold nanosphereswere used as a blank and subtracted from the 3D cellmeasurements.

Co-culture of the two cell types within a 3D PEGylatedbrin gel showed that nanorod labeled MSCs formedtubular networks in the gels aer 7 days. Macrophagesmaintained a rounded morphology within the gel and star-ted to endocytose nanoparticles. Fig. 5 shows the co-culturesystem, with nanoparticle labeled macrophages circled inyellow.

Cell viability and function following nanoparticle labeling

LIVE/DEAD staining and an MTS proliferation assay were usedto assess cell viability following nanoparticle labeling. Nano-particle cytotoxicity is dependent upon the size, shape, andsurface coating of the nanoparticles. In particular, CTAB, asurfactant used during nanorod synthesis and which forms abilayer on the surface of the nanorod, is especially cytotoxic tocells, and thus removing or displacing CTAB is necessary toreduce the cytotoxic effects.31,49,50 Following labeling of MSCswith silica coated NRs, MSCs were viable up to 5 days andcontinued to proliferate, with no signicant reduction in cellproliferation compared to the control cells (Fig. 6A and B).Labeling of macrophages with 50 nm PEGylated gold

J. Mater. Chem. B, 2014, 2, 8220–8230 | 8225

Fig. 5 (A) Schematic showing a 3D co-culture system of nanorodlabeled MSCs, macrophages, and nanospheres within a PEGylatedfibrin gel. (B) Bright field microscopy demonstrating nanoparticleuptake by macrophages within the gels (yellow circles).

Fig. 6 LIVE/DEAD staining of (A) control and nanorod labeled MSCsand (C) control and 50 nm PEGylated nanosphere labeled macro-phages at day 5. MTS assay of (B) control and nanorod labeled MSCsand (D) control and 50 nm PEGylated nanosphere labeled macro-phages to assess cell proliferation.

Fig. 7 3D stacks of tubular network formation within PEGylated fibringels for (A and C) unlabeled and (B and D) NR labeled MSCs with (A andB) no laser irradiation or (C and D) laser irradiation.

Journal of Materials Chemistry B Paper

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

nanospheres did not affect cell proliferation or viability (Fig. 6Cand D) compared to controls.

The maintenance of MSC tubular network formation within3D PEGylated brin gels was assessed following nanorodlabeling and laser irradiation. This was performed to verify theprocess of photoacoustic imaging did not affect the cells' abilityto form tubular networks. MSCs without NR labeling or laserirradiation formed extensive tubular networks within the gels,and this network formation was maintained following NRlabeling (Fig. 7A and B). The gels were also irradiated with a 800nm laser to ensure photoacoustic imaging did not affect cellfunction. Both non-labeled cells and NR labeled cells continuedto form networks following laser irradiation (Fig. 7C and D).These results demonstrate that the cells maintain their functionof forming tubular networks following NR labeling and photo-acoustic imaging.

8226 | J. Mater. Chem. B, 2014, 2, 8220–8230

In vitro photoacoustic imaging of nanoparticle labeled cells

In vitro photoacoustic imaging of nanosphere labeled macro-phages and NR labeled MSCs was performed to evaluate thefeasibility of distinguishing and quantifying the two cell typesusing photoacoustic imaging. The two cell types were mixed invarious ratios and Fig. 8 shows the spectroscopic results inwhich the 100%macrophage inclusion was identied as having50 nm PEGylated nanospheres (red) and the 100% MSC inclu-sion was identied as having NRs (green). A transition fromnanospheres to NRs is evident as the concentration of macro-phages decreases and of MSCs increases. These resultsdemonstrate that the two cell types can be distinguished basedon labeling with different nanoparticles, even when they are inclose proximity. In addition, the ratios of the two cell types werecalculated based on the photoacoustic imaging results, with thepredicted cell ratios being a very good estimate of the actual cellratios in the inclusions. Thus, photoacoustic imaging can beused in combination with the dual nanoparticle system pre-sented here to quantify the ratios of cells within a region. Thisdata would be important for in vivo applications to quantify theinltration of macrophages and the persistence of MSCsfollowing delivery.

In vivo photoacoustic imaging of nanoparticle labeled cells

The dual nanoparticle system (consisting of NR labeled MSCsand nanospheres) was intramuscularly injected within PEGy-lated brin gel into the gastrocnemius of a rat following anischemic injury. Combined ultrasound and photoacousticimaging (US/PA) obtained at 800 nm shows a high photo-acoustic signal located within the gastrocnemius muscle(Fig. 9A), which persisted until day 7. Spectroscopic analysisattributes the photoacoustic signal to NRs (shown in green)immediately aer the injection on day 0. Over time, thecontribution from 50 nm PEGylated nanospheres increases(shown in red) (Fig. 9B). In addition, there is a 20% relativeincrease in signal attributed to the nanospheres from day 0 today 7 (Fig. 9C). This increase in signal from nanospheres can beattributed to inltration of macrophages following the ischemicinjury and subsequent endocytosis of the PEGylated nano-spheres by the cells. The endocytosis of the nanospheres bymacrophages leads to plasmon coupling of the particles,46 asdemonstrated by the peak shi in Fig. 1, and thus an increase inphotoacoustic signal within the range of wavelengths whichwere used for imaging (680–970 nm). A majority of the inl-trating macrophages are localized around the delivered gel

This journal is © The Royal Society of Chemistry 2014

Fig. 8 (A) Spectroscopic in vitro photoacoustic imaging of nanosphere labeled macrophages (red) and nanorod labeled MSCs (green) mixed invarious ratios. (B) Quantification of the ratio of the two cell types based on the photoacoustic imaging results.

Fig. 9 (A) Longitudinal ultrasound/photoacoustic imaging (l ¼ 800nm) and spectroscopic photoacoustic imaging of NR labeled MSCsand nanospheres injected within PEGylated fibrin gel into thegastrocnemius. (B) Ratio of macrophages/MSCs over time. (C) Quan-tification of the relative photoacoustic signal increase attributed tonanospheres compared to day 0.

Paper Journal of Materials Chemistry B

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

system. This could be attributed in part to the fact that thenanospheres were delivered within the gel system containingthe MSCs, and thus the nanoparticles which were available forphagocytosis by themacrophages were already localized in closeproximity to the stem cells. In addition, others have shown thatmacrophages have important roles in wound healing andangiogenesis,24,51,52 with stem cells andmacrophages potentially

This journal is © The Royal Society of Chemistry 2014

undergoing some sort of synergistic crosstalk to furtherpromote the regeneration process. MSCs have been shown tosecrete not only angiogenic factors, but also factors whichinuence macrophage function.53–55 Furthermore, others havefound that macrophages are frequently associated with endo-thelial cells during the process of angiogenesis.56 In addition,chemotactic signals secreted from apoptotic stem cells couldalso promotemacrophagemigration to the area. Stem cell deathfollowing delivery for therapeutic purposes is a common issue,and thus it would not be surprising if macrophages are enteringthe area to scavenge apoptotic MSCs. Thus, one of the advan-tages of this dual nanoparticle system is the ability to distin-guish the two cell types in vivo, providing information about theinteraction between stem cells and macrophages. Such infor-mation could include if macrophages are phagocytosing MSCsleading to a transfer of contrast agent from MSCs to macro-phages. Based on our imaging system, we would designatetransfer of contrast agent from MSCs to macrophages if there isco-localization of the nanosphere and nanorod signal.

Histological analysis of the muscle sample following sacri-ce at day 7 shows the presence of injected MSCs (uorescentlypre-labeled with CM-DiI) distributed throughout the musclebers (Fig. 10A). In addition, the presence of nanoparticleswithin the cells is evident under bright eld. TUNEL stainingwas performed to assess cell apoptosis following injection. TheTUNEL staining in Fig. 10B shows a large extent of cell deathwithin the injection region corresponding to the delivered

J. Mater. Chem. B, 2014, 2, 8220–8230 | 8227

Fig. 10 (A) Fluorescent microscopy of muscle sections demonstratingCM-DiI labeled MSCs which were intramuscularly injected. Bright fieldmicroscopy shows the localization of nanoparticles within the cells. (B)TUNEL staining demonstrates MSC death following injection. (C)Immunostaining for ED1 cell surface marker demonstrates macro-phage infiltration. Scale bars ¼ 50 mm.

Journal of Materials Chemistry B Paper

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

MSCs. It has been largely reported in the literature that only asmall proportion of stem cells survive following delivery.55,57,58

Immunostaining for macrophages (ED1) in Fig. 10C showsmacrophage inltration within the muscle and in particularlyclose proximity to the injected MSCs, which were uorescentlylabeled with CM-DiI prior to delivery. The ED1 immunostainingdata supports the in vivo photoacoustic imaging results, whichshows the presence of macrophages on day 7 surrounding theinjected MSCs.

Fluorescence-activated cell sorting of macrophages

Macrophages were sorted from gastrocnemius muscle 7days following injection. ICP-MS analysis conrmed thatmacrophages endocytosed gold nanoparticles within themuscle (Fig. 11). To conrm that macrophages were not justendocytosing MSCs labeled with gold NRs, a control condi-tion consisting of unlabeled MSCs was used. In this conditiongold was also detected within isolated macrophages, con-rming that macrophages are endocytosing the goldnanospheres.

Fig. 11 Inductively coupled plasma mass spectrometry analysis toquantify nanoparticle labeling of macrophages isolated fromgastrocnemius. Test conditions included nanorod labeled and unla-beled MSCs delivered in combination with nanospheres.

8228 | J. Mater. Chem. B, 2014, 2, 8220–8230

Conclusions

This study demonstrated the use of a dual nanoparticle systemto monitor stem cells following delivery and to detect thepresence of inltrating macrophages to a wound area. Cellularuptake of the nanoparticles was qualitatively and quantitativelyassessed in vitro using microscopy and mass spectrometry,respectively. In addition, the viability and function of the cellswas maintained following nanoparticle labeling. An in vitrophantom imaging experiment veried that the two cell typescould be distinguished and quantied using photoacousticimaging. The stem cells were also successfully monitoredfollowing intramuscular injection into a hind limb ischemiamodel, and the increased inltration of macrophages into thearea was quantied over time. Histological analysis of macro-phages conrmed the photoacoustic imaging results, and massspectrometry analysis conrmed that the macrophages hadendocytosed gold nanoparticles, and specically gold nano-spheres. These results represent important advancements inmonitoring stem cells for therapeutic purposes and dis-tinguishing the delivered cells from inltrating immune cells,which also have important roles in the wound healing response.

Acknowledgements

The TEM images were acquired at the Microscopy and ImagingFacility of the Institute for Cellular and Molecular Biology at theUniversity of Texas at Austin. The ICP-MS data was acquired atthe Mass Spectrometry Facility in the Department of Chemistryat the University of Texas at Austin and at the Quadrupole ICP-MS lab in the Jackson School of Geosciences at the University ofTexas at Austin. The FACS samples were run at the Microscopyand Imaging Facility of the Institute for Cellular and MolecularBiology at the University of Texas at Austin. This work wassupported in part by a National Science Foundation GraduateResearch Fellowship awarded to Laura M. Ricles and by NIHunder grant EB015007.

Notes and references

1 I. B. Copland, Mesenchymal stromal cells for cardiovasculardisease, J. Cardiovasc. Dis. Res., 2011, 2, 3–13.

2 S. Taghavi, J. M. Duran and J. C. George, Stem cell therapy forperipheral arterial disease: a review of clinical trials, StemCell Stud., 2011, 1, 105–114.

3 F. Cao, S. Lin, X. Xie, P. Ray, M. Patel, X. Zhang, et al., In vivovisualization of embryonic stem cell survival, proliferation,and migration aer cardiac delivery, Circulation, 2006, 113,1005–1014.

4 Z. Li, Y. Suzuki, M. Huang, F. Cao, X. Xie, A. J. Connolly, et al.,Comparison of reporter gene and iron particle labeling fortracking fate of human embryonic stem cells anddifferentiated endothelial cells in living subjects, StemCells, 2008, 26, 864–873.

5 L. W. van Laake, R. Passier, J. Monshouwer-Kloots,A. J. Verkleij, D. J. Lips, C. Freund, et al., Humanembryonic stem cell-derived cardiomyocytes survive and

This journal is © The Royal Society of Chemistry 2014

Paper Journal of Materials Chemistry B

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

mature in the mouse heart and transiently improve functionaer myocardial infarction, Stem Cell Res., 2007, 1, 9–24.

6 I. M. Barbash, P. Chouraqui, J. Baron, M. S. Feinberg,S. Etzion, A. Tessone, et al., Systemic delivery of bonemarrow-derived mesenchymal stem cells to the infarctedmyocardium - Feasibility, cell migration, and bodydistribution, Circulation, 2003, 108, 863–868.

7 M. Hofmann, K. C. Wollert, G. P. Meyer, A. Menke,L. Arseniev, B. Hertenstein, et al., Monitoring of bonemarrow cell homing into the infarcted humanmyocardium, Circulation, 2005, 111, 2198–2202.

8 D. L. Kraitchman, M. Tatsumi, W. D. Gilson, T. Ishimori,D. Kedziorek, P. Walczak, et al., Dynamic imaging ofallogeneic mesenchymal stem cells trafficking tomyocardial infarction, Circulation, 2005, 112, 1451–1461.

9 H. E. Daldrup-Link, M. Rudelius, S. Metz, G. Piontek,B. Pichler, M. Settles, et al., Cell tracking with gadophrin-2:a bifunctional contrast agent for MR imaging, opticalimaging, and uorescence microscopy, Eur. J. Nucl. Med.Mol. Imaging, 2004, 31, 1312–1321.

10 T. Xiong, Z. Zhang, B. F. Liu, S. Zeng, Y. Chen, J. Chu, et al.,In vivo optical imaging of human adenoid cystic carcinomacell metastasis, Oral Oncol., 2005, 41, 709–715.

11 H. Kawada, J. Fujita, K. Kinjo, Y. Matsuzaki, M. Tsuma,H. Miyatake, et al., Nonhematopoietic mesenchymal stemcells can be mobilized and differentiate intocardiomyocytes aer myocardial infarction, Blood, 2004,104, 3581–3587.

12 L. M. Ricles, S. Y. Nam, K. Sokolov, S. Y. Emelianov andL. J. Suggs, Function of mesenchymal stem cells followingloading of gold nanotracers, Int. J. Nanomed., 2011, 6, 407–416.

13 R. Guzman, N. Uchida, T. M. Bliss, D. He,K. K. Christopherson, D. Stellwagen, et al., Long-termmonitoring of transplanted human neural stem cells indevelopmental and pathological contexts with MRI, Proc.Natl. Acad. Sci. U. S. A., 2007, 104, 10211–10216.

14 S. Lin, X. Xie, M. R. Patel, Y. H. Yang, Z. Li, F. Cao, et al.,Quantum dot imaging for embryonic stem cells, BMCBiotechnol., 2007, 7, 67.

15 E. Chung, L. M. Ricles, R. S. Stowers, S. Y. Nam,S. Y. Emelianov and L. J. Suggs, Multifunctional nanoscalestrategies for enhancing and monitoring blood vesselregeneration, Nano Today, 2012, 7, 514–531.

16 B. D. Chithrani and W. C. Chan, Elucidating the mechanismof cellular uptake and removal of protein-coated goldnanoparticles of different sizes andshapes, Nano Lett.,2007, 7, 1542–1550.

17 B. D. Chithrani, A. A. Ghazani and W. C. Chan, Determiningthe size and shape dependence of gold nanoparticle uptakeinto mammalian cells, Nano Lett., 2006, 6, 662–668.

18 S. Kubinova and E. Sykova, Nanotechnologies inregenerative medicine, Minim. Invasive Ther. Allied Tech.,2010, 19, 144–156.

19 C. W. Lu, Y. Hung, J. K. Hsiao, M. Yao, T. H. Chung, Y. S. Lin,et al., Bifunctional magnetic silica nanoparticles for highly

This journal is © The Royal Society of Chemistry 2014

efficient human stem cell labeling, Nano Lett., 2007, 7,149–154.

20 S. Y. Nam, L. M. Ricles, L. J. Suggs and S. Y. Emelianov, Invivo ultrasound and photoacoustic monitoring ofmesenchymal stem cells labeled with gold nanotracers,PLoS One, 2012, 7, e37267.

21 Y. Amsalem, Y. Mardor, M. S. Feinberg, N. Landa, L. Miller,D. Daniels, et al., Iron-oxide labeling and outcome oftransplanted mesenchymal stem cells in the infarctedmyocardium, Circulation, 2007, 116, I38–I45.

22 K. M. Dupont, K. Sharma, H. Y. Stevens, J. D. Boerckel,A. J. Garcia and R. E. Guldberg, Human stem cell deliveryfor treatment of large segmental bone defects, Proc. Natl.Acad. Sci. U. S. A., 2010, 107, 3305–3310.

23 C. Sunderkotter, K. Steinbrink, M. Goebeler, R. Bhardwajand C. Sorg, Macrophages and angiogenesis, J. LeukocyteBiol., 1994, 55, 410–422.

24 B. Chazaud, M. Brigitte, H. Yacoub-Youssef, L. Arnold,R. Gherardi, C. Sonnet, et al., Dual and benecial roles ofmacrophages during skeletal muscle regeneration, ExerciseSport Sci. Rev., 2009, 37, 18–22.

25 B. Mahdavian Delavary, W. M. van der Veer, M. van Egmond,F. B. Niessen and R. H. Beelen, Macrophages in skin injuryand repair, Immunobiology, 2011, 216, 753–762.

26 T. A. Wynn, A. Chawla and J. W. Pollard, Macrophage biologyin development, homeostasis and disease, Nature, 2013, 496,445–455.

27 J. Kim and P. Hematti, Mesenchymal stem cell-educatedmacrophages: a novel type of alternatively activatedmacrophages, Exp. Hematol., 2009, 37, 1445–1453.

28 K. Anton, D. Banerjee and J. Glod, Macrophage-associatedmesenchymal stem cells assume an activated, migratory,pro-inammatory phenotype with increased IL-6 andCXCL10 secretion, PLoS One, 2012, 7, e35036.

29 L. Chen, E. E. Tredget, P. Y. Wu and Y. Wu, Paracrine factorsof mesenchymal stem cells recruit macrophages andendothelial lineage cells and enhance wound healing, PLoSOne, 2008, 3, e1886.

30 E. Chung, S. Y. Nam, L. M. Ricles, S. Y. Emelianov andL. J. Suggs, Evaluation of gold nanotracers to trackadipose-derived stem cells in a PEGylated brin gel fordermal tissue engineering applications, Int. J. Nanomed.,2013, 8, 325–336.

31 A. M. Alkilany and C. J. Murphy, Toxicity and cellular uptakeof gold nanoparticles: what we have learned so far?, J.Nanopart. Res., 2010, 12, 2313–2333.

32 P. K. Jain and M. A. El-Sayed, Plasmonic coupling in noblemetal nanostructures, Chem. Phys. Lett., 2010, 487, 153–164.

33 M. J. A. Hore and R. J. Composto, Nanorod Self-Assembly forTuning Optical Absorption, ACS Nano, 2010, 4, 6941–6949.

34 S. Mallidi, T. Larson, J. Tam, P. P. Joshi, A. Karpiouk,K. Sokolov, et al., Multiwavelength photoacoustic imagingand plasmon resonance coupling of gold nanoparticles forselective detection of cancer, Nano Lett., 2009, 9, 2825–2831.

35 G. Zhang, C. T. Drinnan, L. R. Geuss and L. J. Suggs, Vasculardifferentiation of bone marrow stem cells is directed by a

J. Mater. Chem. B, 2014, 2, 8220–8230 | 8229

Journal of Materials Chemistry B Paper

Publ

ishe

d on

10

Sept

embe

r 20

14. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

Lib

rari

es o

n 20

/07/

2015

23:

29:2

6.

View Article Online

tunable three-dimensional matrix, Acta Biomater., 2010, 6,3395–3403.

36 N. R. Jana, L. Gearheart and C. J. Murphy, Seed-mediatedgrowth approach for shape-controlled synthesis ofspheroidal and rod-like gold nanoparticles using asurfactant template, Adv. Mater., 2001, 13, 1389–1393.

37 B. Nikoobakht and M. A. El-Sayed, Preparation and growthmechanism of gold nanorods (NRs) using seed-mediatedgrowth method, Chem. Mater., 2003, 15, 1957–1962.

38 W. Stober, A. Fink and E. Bohn, Controlled Growth ofMonodisperse Silica Spheres in Micron Size Range, J.Colloid Interface Sci., 1968, 26, 62–69.

39 Y. Lu, Y. D. Yin, B. T. Mayers and Y. N. Xia, Modifying thesurface properties of superparamagnetic iron oxidenanoparticles through a sol-gel approach, Nano Lett., 2002,2, 183–186.

40 Y. S. Chen, W. Frey, S. Kim, K. Homan, P. Kruizinga,K. Sokolov, et al., Enhanced thermal stability of silica-coated gold nanorods for photoacoustic imaging andimage-guided therapy, Opt. Express, 2010, 18, 8867–8878.

41 Arnida, M. M. Janat-Amsbury, A. Ray, C. M. Peterson andH. Ghandehari, Geometry and surface characteristics ofgold nanoparticles inuence their biodistribution anduptake by macrophages, Eur. J. Pharm. Biopharm., 2011, 77,417–423.

42 S. Albukhaty, H. Naderi-Manesh and T. Tiraihi, In vitrolabeling of neural stem cells with poly-L-lysine coatedsuper paramagnetic nanoparticles for green uorescentprotein transfection, Iran. Biomed. J., 2013, 17, 71–76.

43 N. Nayerossadat, T. Maedeh and P. A. Ali, Viral and nonviraldelivery systems for gene delivery, Adv. Biomed. Res., 2012, 1,27.

44 D. Horak, M. Babic, P. Jendelova, V. Herynek, M. Trchova,K. Likavcanova, et al., Effect of different magneticnanoparticle coatings on the efficiency of stem celllabeling, J. Magn. Magn. Mater., 2009, 321, 1539–1547.

45 S. Mallidi, T. Larson, J. Aaron, K. Sokolov and S. Emelianov,Molecular specic optoacoustic imaging with plasmonicnanoparticles, Opt. Express, 2007, 15, 6583–6588.

46 B. Wang, E. Yantsen, T. Larson, A. B. Karpiouk,S. Sethuraman, J. L. Su, et al., Plasmonic intravascularphotoacoustic imaging for detection of macrophages inatherosclerotic plaques, Nano Lett., 2009, 9, 2212–2217.

47 T. A. Larson, P. P. Joshi and K. Sokolov, Preventing proteinadsorption and macrophage uptake of gold nanoparticlesvia a hydrophobic shield, ACS Nano, 2012, 6, 9182–9190.

8230 | J. Mater. Chem. B, 2014, 2, 8220–8230

48 C. D. Walkey, J. B. Olsen, H. Guo, A. Emili and W. C. Chan,Nanoparticle size and surface chemistry determine serumprotein adsorption and macrophage uptake, J. Am. Chem.Soc., 2012, 134, 2139–2147.

49 A. M. Alkilany, P. K. Nagaria, C. R. Hexel, T. J. Shaw,C. J. Murphy and M. D. Wyatt, Cellular Uptake andCytotoxicity of Gold Nanorods: Molecular Origin ofCytotoxicity and Surface Effects, Small, 2009, 5, 701–708.

50 T. Niidome, M. Yamagata, Y. Okamoto, Y. Akiyama,H. Takahashi, T. Kawano, et al., PEG-modied goldnanorods with a stealth character for in vivo applications,J. Controlled Release, 2006, 114, 343–347.

51 L. Arnold, A. Henry, F. Poron, Y. Baba-Amer, N. van Rooijen,A. Plonquet, et al., Inammatory monocytes recruited aerskeletal muscle injury switch into antiinammatorymacrophages to support myogenesis, J. Exp. Med., 2007,204, 1057–1069.

52 M. L. Novak and T. J. Koh, Phenotypic transitions ofmacrophages orchestrate tissue repair, Am. J. Pathol., 2013,183, 1352–1363.

53 L. W. Chen, E. E. Tredget, P. Y. G. Wu and Y. J. Wu, ParacrineFactors of Mesenchymal Stem Cells Recruit Macrophagesand Endothelial Lineage Cells and Enhance WoundHealing, PLoS One, 2008, 3, e1886.

54 J. Kim and P. Hematti, Mesenchymal stem cell-educatedmacrophages: A novel type of alternatively activatedmacrophages, Exp. Hematol., 2009, 37, 1445–1453.

55 X. Y. Hu, S. P. Yu, J. L. Fraser, Z. Y. Lu, M. E. Ogle, J. A. Wang,et al., Transplantation of hypoxia-preconditionedmesenchymal stem cells improves infarcted heart functionvia enhanced survival of implanted cells and angiogenesis,J. Thorac. Cardiovasc. Surg., 2008, 135, 799–808.

56 S. Nucera, D. Biziato and M. De Palma, The interplaybetween macrophages and angiogenesis in development,tissue injury and regeneration, Int. J. Dev. Biol., 2011, 55,495–503.

57 C. Toma, M. F. Pittenger, K. S. Cahill, B. J. Byrne andP. D. Kessler, Human mesenchymal stem cells differentiateto a cardiomyocyte phenotype in the adult murine heart,Circulation, 2002, 105, 93–98.

58 M. Zhang, D. Methot, V. Poppa, Y. Fujio, K. Walsh andC. E. Murry, Cardiomyocyte graing for cardiac repair:Gra cell death and anti-death strategies, J. Mol. Cell.Cardiol., 2001, 33, 907–921.

This journal is © The Royal Society of Chemistry 2014