Overview A brief account of nanoparticle contrast agents ...coilab.caltech.edu/epub/2013/PanD_2013_WIREs... · Overview A brief account of nanoparticle contrast agents for photoacoustic

Overview

A brief account of nanoparticlecontrast agents for photoacousticimagingDipanjan Pan,1,∗ Benjamin Kim,1 Lihong V. Wang2 andGregory M Lanza1

Photoacoustic imaging (PAI) is a hybrid, nonionizing modality offering excellentspatial resolution, deep penetration, and high soft tissue contrast. In PAI, signalis generated based on the absorption of laser-generated optical energy byendogenous tissues or exogenous contrast agents leading to acoustic emissionsdetected by an ultrasound transducer. Research in this area over the years hasshown that PAI has the ability to provide both physiological and molecularimaging, which can be viewed alone or used in a hybrid modality fashion to extendthe anatomic and hemodynamic sensitivities of clinical ultrasound. PAI may beperformed using inherent contrast afforded by light absorbing molecules such ashemoglobin, myoglobin, and melanin or exogenous small molecule contrast agentsuch as near infrared dyes and porphyrins. However, this review summarizes thepotential of exogenous nanoparticle-based agents for PAI applications includingcontrast based on gold particles, carbon nanotubes, and encapsulated coppercompounds. © 2013 Wiley Periodicals, Inc.

How to cite this article:WIREs Nanomed Nanobiotechnol 2013. doi: 10.1002/wnan.1231

INTRODUCTION

Photoacoustic imaging (PAI), also referred to asoptoacoustic imaging, is an emergent hybrid

optical and ultrasound diagnostic modality withhigh spatial resolution and excellent soft tissuecontrast.1–6 In a typical photoacoustic (PA) systemtissue is irradiated with a nonionizing short-pulsedlaser beam within the acceptable the AmericanNational Standards Institute (ANSI) defined limits.3

Endogenous proteins, e,g. hemoglobin, myoglobin,melanin, or exogenous PA contrast agents absorb theoptical energy, undergo thermoelastic expansion, andproduces PA waves that are received with a clinicalor nonclinical, wide-band ultrasonic transducer

∗Correspondence to: [email protected] of Medicine, Washington University School ofMedicine, St Louis, MO, USA2Department of Biomedical Engineering, Washington University,St Louis, MO, USA

Conflict of interest: L. V. W. has a financial interest inMicrophotoacoustics, Inc. and Endra, Inc.

(5–50 MHz). The PA images may be visualized aloneor hybridized with an interleaved ultrasound imageof the same field of interest. Because the PA imageis created only from ultrasound generated withinthe tissue, it lacks the typical background speckleof ultrasound images. Moreover, PAI offers contrastsensitivity to physiological parameters such as theconcentration and oxygenation of hemoglobin as wellas unique molecular imaging contrast information.

While inherent PA contrast afforded by endoge-nous substances can be utilized for PA imaging toassess levels of vascularization and oxygen saturationin tumors,1–5 particularly malignant cancers,7,8 amyriad of advances have been made in terms of exoge-nous contrast agent development for PA molecularimaging applications, particularly nanoparticle basedtechnologies. As with inherent protein based contrast,the optical absorptive of exogenous agents dictatestheir utility for PA imaging. Several optically activematerials have been explored as PA contrast agents,e.g. small molecule dyes, gold nanoparticles, singlewalled carbon nanotubes (SWNT), copper nanopar-ticles, etc.2,9,10 However, given the optical absorptive

© 2013 Wiley Per iodica ls, Inc.

Overview wires.wiley.com/nanomed

Photoacousticscontrast agents

Gold nanoparticlesCarbon nanotubes

Hybrid nanoparticlesCopper nanoparticle

Small moleculecontrast agents

Nanoparticlecontrast agents

NIR dyes

Porphyrins

Organic/polymeric

Inorganic

Liposomalmicelles

X = 2Hor

for example, Zn, Cu, Pd

-O3S SO

3Na

OO O

NN

N N

N

NH2

S

N⊕

FIGURE 1 | Different classes of photoacoustic contrast agents.

capacity of blood and muscle, exogenous PA contrastenhancement must significantly exceed inherentbackground in vivo contrast. Moreover, these agentsmust have outstanding biocompatibility, appropriatein vivo stability (i.e., particularly during circulatorytransit to targets and through the imaging period), anddesirable synthetic processing attributes, specificallytolerance to sterilization and prolonged shelf-life sta-bility. Although exogenous agents, e.g. near infrared(NIR) dye and porphyrin-based agents, have beenadministered directly into the blood pool, this reviewwill focus on nanoparticulate compositions (Figure 1).Examples of ‘nano’-contrast agents include differentclasses of gold nanoparticles of varying size and mor-phology (spherical vs rod, cages); carbon nanotubes(SWNTs), hybrid nanoparticles (gold coated SWNTs)and copper oleate encapsulated nanoparticles.

NANOPARTICLE-BASED CONTRASTAGENTS

Gold NanoparticlesGold nanoparticles are excitable in the NIR rangewithin the ‘optical transmission window’ of biologicaltissues (λ = 650–900 nm). This allows for deeper lightpenetration, lower autofluorescence, and reduced lightscattering. In addition to strong optical absorption,gold particles, unlike small molecule fluorophores,are resistant to photobleaching or fading due todye destruction.11–13 Gold nanoparticles exhibit aphenomenon known as localized surface plasmonresonance (LSPR) because of their inherent abilitiesto absorb and scatter light at specific wavelengths.Nanoparticle LSPR refers to the resonance established

when incident light photons match frequency ofparticle surface electrons vibrating against theopposing restoring force of positive nuclei. LSPR canbe manipulated by altering particle shape and surfacecoating. Several possibilities of PA contrast imaginginvolving gold nanoparticles have been explored bymultiple research groups.

Gold Nanocages as PA Contrast AgentsThe LSPR of gold nanoparticles typically must betuned from the visible electromagnetic range ofenergies, where blood and tissue attenuation impairslight penetration, into the NIR region bandwidth formaximum interrogation depth in vivo. This is theso-called NIR window. Conventional spherical goldparticles does places the LSPR in the visible lightrealm, which led Xia and co-workers to embark upondevelopment of a novel class of nanoparticles calledgold nanocages (AuNC). These particles were charac-terized by hollow interiors enclosed by a single-crystalultrathin outer structure with porous walls.14 Xiademonstrated that these AuNCs exhibited wavelengthplasticity and could be manipulated to any wave-length in the NIR region based on their modificationsof the nanocage sizes and thickness. Additionally,because of their novel properties, which included highabsorption, low cytotoxicity, and their ability to beeasily conjugated to tumor-specific ligands, AuNCsprovided highly effective PA contrast imaging.15–17

Golf Nanocage SynthesisAuNCs synthesis followed a galvanic replacementreaction using Ag solid as a template for the reductionof HAuCl4 because of their electrochemical potentialdifference to form Au atoms. In this reaction, the


WIREs Nanomedicine and Nanobiotechnology Contrast agents for PAI

1.0

0.8

0.6

0.4

0.2

Ext

inct

ion

0.0

400 1000600

Wavelength (nm)

800

0 mL 0.2 mL 0.5 mL 0.6 mL 0.8 mL 1.0 mL 1.6 mL

(a) (b)

(c)

FIGURE 2 | SEM images of (a) Ag nanocubesand (b) AuNCs. The inset shows thecorresponding TEM images of the same sample.(c) UV–vis spectra of the samples obtained bytitrating Ag nanocubes with different volumes of0.1 mM HAuCl4 solution. (Reprinted withpermission from Ref 21. Copyright 2007American Chemical Society)

growth of AuNCs is template over silver nanocubesthat are concurrently produced:

3Ag (s) + AuCl−4(aq

) → Au (s) + 3Ag+ (aq

)

+ 4Cl−(aq

)

Ag nanocubes were synthesized using AgNO3or CF3COOAg as a precursor to Ag, and a polyolreduction by using ethylene glycol as a solventand a source of reducing agent. This reaction wasfurther assisted with a capping agent, poly(vinylpyrrolidone) (PVP). Subsequently, the addition ofcatalytic amounts of NaHS and or Cl− ionsimparted rapid nucleation of single-crystal seeds whilesimultaneously, due to oxidative etching, discardedtwinned seeds, and ultimately formed single-crystal Agnanocubes. Additional titration of these Ag nanocubeswith HAuCl4 was followed at 100 ◦C in order toensure maximal growth of thin layered Au on thesurface of the Ag crystals and to evade precipitation ofAgCl. Cooling the reaction precipitates AgCl, whichis then dissolved in a saturated NaCl solution andeliminated, leaving AuNCs as the final remainingproduct (Figure 2). Interestingly, the formation ofAuNCs can be distinctively noted by observing theextinction spectra of the titration of Ag nanocubeswith varying volumes of HAuCl4. As more HAuCl4is added to the Ag nanocubes to promote epitaxialgrowth of AuNCs, the LSPR peak shifts from thevisible light into the NIR region.

PA Imaging of Cerebral Cortex via AuNCThe cranium and cerebral cortex of rodent lendthemselves to PA techniques, being permeable to laserlight and ultrasonic RF waves.18–20 Wang et al.21

imaged of the vasculature blood pool in rat brainusing gold nanoshells with >100 nm diameters thatwas further improved in collaboration with Xia et al.22

utilizing smaller AuNCs (<50 nm). Figure 3 shows thecomparative results of an ex vivo PA study producedfrom pure rat blood and a mixture of AuNC-rat blood,and an in vivo study of injections of AuNCs in thecerebral cortex of a rat model.

Sentinel Lymph Node Mapping Via AuNCsSentinel lymph node (SLN) imaging is an adjunctivetechnique used with minimally invasive breast cancersurgery, or lumpectory, to guide SLN identificationand resection for disease staging and prognosis.23–28

Typically, methylene blue dye or radioactive colloidare utilized to preoperatively localize the SLN.However, despite its widely used practice, radioactivecolloids have low precision and difficulty in local-ization due to very low lateral and negligible verticalresolution, which can accrue to potentially harmfuleffects when resecting the SLN during surgicaloperations. On the other hand, methylene blue staincan be difficult to place preoperatively or detect intra-operative, offering far less sensitivity than the nuclearcounterpart. Besides the imprecision of SLN methy-lene blue technique, potential adverse effects includingallergic reactions are frequent with the stain.29



(b) (c)

(a)

2 mm2 mm

760

Pho

toac

oust

icsi

gnal

am

plitu

de (

a.u.

)

770 780 790 800

Laser wavelength (nm)

810 820 830

Blood only

With cages

FIGURE 3 | (a) The measured PA signal amplitude generated withand without AuNCs in rat blood at several wavelengths. Noninvasive PAimaging of a rat’s cerebral cortex, (b) before the injection of AuNCs and(c) about 2 h after the final injection of nanocages, which is the peakenhancement point. (Reprinted with permission from Ref 22. Copyright2007 American Chemical Society; Reprinted with permission fromRef 21.)

Xia and co workers demonstrated that AuNCscan be injected subcutaneously, migrate to thedraining lymphnode to afford effective sentinel lymphnode mapping30 with high spatial resolution atimaging depths reaching 33 mm (Figure 4).

Imaging Melanoma with AuNCsAuNCs can be surface modified with functional,biologically relevant moieties, AuNC’s surfacegroups were modified with thiol-PEG groupsand then conjugated to [Nle4,D-Phe7]-a-melanocyte-stimulating hormone ([Nle4,D-Phe7]-a-MSH).31 Thehoming ligand, [Nle4,D-Phe7]-a-MSH, has the abilityto bind to a-MSH receptors over-expressed onmelanomas, thus acting as a method to retainand concentratie these particles at the tumorsite. PA signals increased significantly with timeas shown in Figures 5 and 6 for the targeted[Nle4,D-Phe7]-a-MSH-AuNCs when compared withnontargeted AuNC, resulting in almost 300% PAsignal enhancement. These results suggested that the

MSH-targeted AuNCs were undergoing ligand–receptor binding with the melanoma cells andsuggested further efficacy and applicability for ligand-modified AuNCs.

Carbon Nanotubes for Advanced ContrastAgentsBecause carbon nanotubes (CNT) are stronglyoptically active over a wide range of wavelengthsrelevant to PA imaging, many studies were devoted toexploring the potential of CNTs for PA applications.Following the work of Zharov et al.32 who initiallydemonstrated the potential of CNTs to function asexogenous PA contrast agents, others have extendedthis result by coupling targeting peptides containing anRGD (arginine–glycine–aspartate) motif to the tumorneovasculature in mice.10 Additional modificationsof CNT chemistry and coating for enhancementof contrast33–35 led to their use for SLN mappingand tumor cell characterization.35–40 Interestingly,CNTs also found utility for therapeutic hyperthermiaapplication, which was demonstrated in vitro withtumor and bacterial cells.41–49

CNT HybridsUnfortunately, CNTs have a relatively lowNIR absorption coefficient compared with goldnanoparticles,48 and therefore require very highlocal concentrations to produce useful PA contrastimaging.34,35 Single-walled nanotube (SWNT) havea very high surface area (1300 m2/g) for interactingwith optical radiation. To enhance this, a hybridgold-SWNT was devised wherein the SWNT corecoated with a gold layer (SWNT-Au)35 (Figure 7(b)).The hybrid SWNT-Au offered a more biocompatiblesurface that was better for chemical bioconjugationof homing ligands or drugs. Compared with SWNTalone, the gold layer greatly augmented LSPR theresponses of the SWNT-Au hybrids.

An alternative approach to the low NIRabsorption properties of SWNT was achieved bycoupling small molecule dyes onto the high surfacearea of the nanotubes. Liu et al.50 demonstratedthis effect by loading indocyanine green (ICG)molecules onto SWNTs via simple pi-pi bond stackingof hydrophobic benzene rings. Remarkably, hybridICG-SWNTs exhibited a 20-fold increase in opticalabsorption (Figure 7(c)).

Applications of CNT Geared Towards PAISWNT were used to enhance contrast from theneovasculature by coupling RGD peptides via a PL-PEG5400

51 . Integrin-targeted SWNTs’ were evaluated



(a) (b) (c)

(d) (e) (f)

BV BV

SLN

SLN SLN SLN

5mm 5mm

5mm 5mm 5mm

0.5 1.5 2.5 0.4 0.8 1.2 1.6 2

0.1

0.9

0.7

0.5

0.3

0.1

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.05 0.07 0.09 0.11 0.13 0.14

0 50 100Time (min)

Injection time

200150 250

PA a

mpl

itude

(A

.U.)

21

FIGURE 4 | PA images of the axillary region of a rat taken (a) before and (b) 28 min after the injection of AuNCs. (c) The changes of PA signalamplitude as a function of the post-injection time. After the injection, PA signals increased with time, which means gradual accumulations of thenanocages. (d–f) Depth capability of SLN mapping with AuNCs. The PA images were acquired after the injection of nanocages for: (d) 126 min with atotal imaging depth of 10 mm by placing a layer of chicken breast tissue on the axillary region; (e) 165 min with a total imaging depth of 21 mm byadding another layer of chicken breast tissue; and (f) 226 min with a total imaging depth of 33 mm by using three layers of chicken breast tissue. Thebars represent the optical absorption. BV, blood vessel. SLN, sentinel lymph node. (Reprinted with permission from Ref 30. Copyright 2009 AmericanChemical Society)

in mice bearing U87 gliobastoma xenograft tumorsand showed a significantly higher PA signal (sevenfoldincrease) in the tumor compared to that of theuntargeted SWNTs (Figure 8).

To further enhance SWNT-RGD contrast, ICG,a small molecule optical dye, was hydrophobicallycoupled to the surface structure. In vitro testing ofSWNT-ICG-RGD constructs had stability in serumand a 300-fold increase in PA contrast associatedwith a shift in the laser excitation wavelengthfrom 690 to 780 nm when compared to SWNT-RGD. However, despite the photoresistive nature ofSWNT particles, photobleaching was observed in thecyanine dye modified version. This optical fading ofPA contrast reflected the optical instability of ICGrather than structural degradation of the SWNT. Invivo, SWNT-ICG-RGD were shown to target tumorneovasculature33 (Figure 9).

Hybrid Magnetic/Gold NanoparticlesDevelopment of techniques to assay and characterizecirculating tumor cells (CTCs) are emerging as an area

of Nanomedicine research. Unfortunately, such assaysare limited by sampling errors, which relates to theprobability that adequate numbers of cancer cells arecontained with the minute blood volume measured.Typically, these methods have high sensitivities fordetecting cells when present, but limited samplevolume and sparse numbers of cells circulating,particularly in early stages of cancer, reduces theoverall sensitivity of the method. One approachto address the issue has been the development ofmultiplex targeting strategy utilizing a conjugatednanoparticle consisting of a magnetic (MNP) andgold nanoparticle (GNP) which have the propertiesof having an absorption spectra in the NIR regionand allows for detection of two colors at laserwavelengths of 639 and 900 nm as shown in Figure10(a) and (b). These conjugated nanoparticles werespecifically synthesized to evade normal blood cells,and to specifically target urokinase plasminogenactivator (ATF) and folate receptors.34 PA signalsindicated that the optimal NP concentration andsolution volume were 109 NPs/ml for GNTs and



[Nle4, D-Phe7]-α-MSH-AuNCs PEG-AuNCs

0 h 0 h

3 h

Tumor

Tumor

Tumor

Tumor

3 h

6 h 6 h

(a)

(b)

(c) (g)

(d) (h)

2 mm

0 10 20 30

Increase in PA amplitude (%)

40 50

(f)

(e)

FIGURE 5 | In vivo noninvasive PA time-course coronal MAP imagesof B16 melanomas using [Nle4, D-Phe7]-a-MSH- and PEG-AuNCs. (a, e)a schematic of the [Nle4, D-Phe7]-a-MSH- and PEG-AuNCs. Time-coursePA images of the B16 melanomas after intravenous injection with100 μL of 10 nM (b–d) [Nle4, D-Phe7]-a-MSH- and (f–h) PEG-AuNCsthrough the tail vein. The background vasculature images wereobtained using the PA microscope at 570 nm (ultrasonic frequency = 50MHz), and the melanoma images were obtained using the PAmacroscope at 778 nm (ultrasonic frequency = 10 MHz). (Reprinted withpermission from Ref 31. Copyright 2010 American Chemical Society)

10–20 μL, respectively. Also, there were no detectablePA signals in mouse ear microvessels above bloodbackground which suggested that PA signals were ableto differentiate between bound and unbound NPs,supporting its specificity. In order to detect CTCsin a primary tumor, photoacoustic flow cytometry

50[Nle4, D-Phe7]-α-MSH-AuNCs

[Nle4, D-Phe7]-α-MSH-AuNCs

PEG-AuNCs

PEG-AuNCs

40

30

20

10

0

6

5

4

3

2

1

0

0 1 2 3 4 5 6

Time (h)

Incr

ease

in P

A s

igna

l (%

)N

tum

or (

108

/g)

(a)

(b)

FIGURE 6 | (a) Increase of PA amplitude in the melanoma tumorsafter intravenous injection of [Nle4, D-Phe7]-a-MSH- and PEG-AuNCs(n = 4 mice for each group), respectively, for different periods of time.The PA signals increased up to 38 ± 6% for [Nle4, D-Phe7]-a-MSH-AuNCs while the maximum signal increase only reached 13 ± 2% forPEG-AuNCs at a post-injection time of 6 h (P < 0.0001). (b) The averagenumber of AuNCs accumulated in the melanomas dissected at 6 h postinjection for the two types of AuNCs as measured by ICP-MS. HereNtumor denotes the number of AuNCs per unit tumor mass (g).(Reprinted with permission from Ref 31. Copyright 2010 AmericanChemical Society)

(PAFC) was used to measure the progressive increasein the number of CTCs over a course of four weeks,and showed a high correlation with respect to theprogression of the tumor (Figure 10(c) and (d)). Suchmethod deemed as ’in vivo blood cancer test’, provesas an applicable and potential measure for earlyPA diagnosis of primary tumors and prevention ofmetastasis.

Gold NanobeaconsGold nanobeacons (GNB) were developed byentrapping numerous small gold nanoparticles(spherical, 2–4 nm) within a larger colloidal particle,



1 – 2 nm2 – 25 nm

SWNT

SWNT

SWNTcore Gold

layerMWNT

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.36 nm

0.1

0.0

1.0

20

15

10

5

0

20X

SWNT-RGD

PEG5020

-RGD

PEG5000

-RGD

SWNT-ICG-RGD

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0300 400 500 600

Wavelength (nm)

700 800 900 1000 1100

500 600

Wavelength (nm)

700 800 900 1000 1100

Rel

ativ

e ab

sorb

ance

Opt

ical

abs

orba

nce

(a.u

.)

Rel

ativ

e PA

sig

nal a

mpl

itude

ICGIC

G

ICG

(a) (b) (c)

FIGURE 7 | (a) Schematic of single-walled and multi-walled carbon nanotubes (SWNT and MWNT, respectively). (b) Absorption spectra of SWNTsand golden carbon nanotubes (GNTs), and photoacoustic (PA) spectra of GNTs. Lines represent normalized optical spectra (left vertical axis) of GNTsin water (red curve), SWNTs in water (black curve) and water only (green curve) and the dots represent normalized PA signal amplitude (blue dots,right axis) of GNTs in water. The concentration of the SWNTs is 35 times higher than that of GNTs; hence 85- to 100-fold enhanced NIR contrast isachieved by the hybrid GNTs. Reprinted with permission from Kim et al.35 (c) Optical absorption spectrum of SWNT-RGD (black curve) andindocyanine green-enhanced SWNT-RDG (SWNT-ICG-RGD, green curve). The optical absorbance spectrum of plain SWNT-RGD is relatively flat withslight gradual absorption decrease as the wavelength increase. However, by attaching a large number of ICG molecules to the SWNT surface, a20-fold increase in optical absorption results at 780 nm. (Reprinted with permission from Ref 33. Copyright 2010 American Chemical Society)

encapsulated with a biocompatible phospholipidscoating in our laboratories.52–55 It was presumedthat the entrapment of numerous small gold particleswould greatly amplify the signal for each bindingevent, effectively similar to a larger single goldparticle, but upon particle metabolism, the goldparticles, falling below the renal elimination threshold(6–10 nm), would be bioeliminated through thekidney and urine. Moreover, the interaction ofthe small particles within the matrix of the lipidencapsulated particle would lead to effectively creatinga randomly irregular shapes that would shift theoptical absorbance peak for visible into the NIRspectral region for optimal in vivo use.

The first generation gold nanobeacon (GNB1)were synthesized by suspending octylthiol-coatedspherical gold nanoparticles (AuNPs, 2 w/v%,) invegetable oil (20 vol%) matrix followed by homog-enization of the mixture with phospholipid surfac-tants mixture (2 w/v%) (Figure 11). The phospho-lipids mixture was comprised of phosphatidylcholine(PC, 91 mole%) and cholesterol (8 mole%) andincluded biotin-caproyl-PE (1 mole %) or a phos-pholipid anchored homing ligand (0.3–1.0 mole%)at the equimolar expense of PC dependent on thebiological target under question. The GNB1 parti-cles were 154 ± 10 nm with polydispersity and zetapotential of 0.08 ± 0.03 and −47 ± 7 mV, respec-tively (Brookhaven Instrument Co.). Gold content,determined by ICP-MS, was 1080 μg/g of the 20%colloid suspension. UV–vis spectroscopy confirmedthe absorbance at 520 nm and in the NIR win-dow (900 nm). GNB1 particles exhibited remarkable

stability as observed by the particle size and ζ poten-tial, which varied <5% over 100 days when stored at4 ◦C under argon in sealed serum vials.

The PA signal from a 20% GNB suspensionmixed 1:1 with rat blood within Tygon® tubing (I.D.250 μm, O.D. 500 μm) was 15 times stronger thanthat from pure rat blood.14 When monitored overthe entire 740–820 nm NIR window, the average PAsignal enhancement by GNB over blood was greaterthan 10-fold (Figure 12). While the agent offeredhigh contrast for typical blood pool applications, theintent of the agent was for vascular-targeted molecularimaging medical applications of clinical importanceand unmet need, such as intravascular thrombosis,neovascularization in cancer and atherosclerosis, andinflammation in cancer, arthritis, and more. Detectingintravascular targets requires a substantial signal overthe background contrast associated with circulatingerythrocyte hemoglobin.

The initial target for GNB1 was to detect fibrin inthrombosis as a means to diagnose ruptured carotidatherosclerotic plaques in patients at high risk forstroke. Today, ultrasound is used to assess carotidstenosis as clinical tool to guide medical decisions forsurgical carotid endarterectomy (CEA) versus medicaltherapy alone. While patients with greater than 70%stenosis benefit from CEA, the vast majority of highrisk individuals stroking over a two year periodhave less than 60% stenosis. In many patients withsevere disease, small and repeated microruptures ofthe plaque intima with intermittent healing preceedsa major event. To minimize the financial and humancosts of stroke, PA imaging could be added to standard



Plain SWNT

Tumorphotograph

Ultrasound

Photoacousticpre-injection

Pho

toac

oust

ic s

igna

l (a.

u.)

100

0

Photoacoustic4 h post-injection

Subtractionimage

SWNT-RGD

Tumor 5 mm Tumor 5 mm

3 mm3 mm

FIGURE 8 | Single-walled carbon nanotube argine–glyice–asparticacid (SWNT-RGD) tumor targeting in living mice. Ultrasound (gray) andphotoacoustic (PA) (green) images of a vertical slice (white dotted line)through the tumors of mice injected with SWNT-RGD (right column) andcontrol plain SWNTs (left column). Subtraction images were calculatedas 4 h post-injection minus pre-injection to remove tissue backgroundsignal from the PA image. Mice injected with SWNT-RGD showed anaveraged sevenfold PA signal increase in the tumor over mice injectedwith control untargeted SWNTs. The high PA signal in the mouseinjected with plain SWNTs (indicated by the white arrow) is not seen inthe subtraction image, suggesting that it is due to a large blood vesseland not SWNTs. (Reprinted with permission from Ref 51. Copyright2007 Nature Publishing Group)

diagnostic carotid imaging of high-risk patientsas a technique to identify microthrombi in smallendothelial microfissures, which could potentiallysupport earlier surgical intervention and a greaterreduction in stroke incidence.

As an initial proof of concept, GNB1 andthe control nanobeacons (i.e., containing no gold)were pretargeted to acellular fibrin clot phantomswith classic avidin–biotin coupling. The particleswere homed to fibrin using a well-characterizedbiotinylated antihuman fibrin-specific monoclonalantibody (NIB5F3).56 The clots were prepared in low-density polyethylene (LDPE) tube (1 cc volume, I.D.6 mm) and treated with biotinylated GNB and control(targeted, no gold nanobeacon). Figure 12(e) and (f)shows cross-sectional PAT images using a curved arrayPAT system57 with an 800 nm wavelength laser. Asevident from the images, high contrast was clearlyappreciated for the clots targeted with biotinylatedGNB1 in comparison to the control clot.

The cross-sectional PAT images were obtainedusing a PA breast scanner system,58 of the same con-trol and targeted plasma clots, respectively52 (Figure

12(g) and (h)). Similar to the previous experiments,the targeted plasma clot was clearly distinguishable(Figure 12(f)) in the PAT image, whereas the con-trol clot image was not appreciable (Figure 12(e)).This result was corroborated analytically by total goldcontent analyses using inductively couple plasmon res-onance (ICP-MS). The total gold content of the clotswith targeted GNB1, nontargeted GNB1 and targetednonmetallic nanobeacons were found to be 47 μg/g,not detected (ND, <0.02 μg/g) and ND, respectively,by ICP-MS. This experiment, for the first time, clearlyillustrated the concept of intravascular PA with GNB1.

Gold Nanobeacons (Rod) as Alternativeto Spherical NanoparticlesGold’s optical properties can further be tuned andoptimized to have more NIR absorption. Rod-shapedgold nanoparticles (i.e., nanorods) offer distinctoptical properties resulting from two surface plasmon(SP) bands corresponding to the transverse andlongitudinal SP bands in the visible (λ= 520 nm)and the NIR (λ = 900 nm) regions, respectively.11–13

Gold nanorods are frequently used as probes forfluorescence, light scattering, and two-photon lumi-nescence imaging due to a large intrinsic extinctioncoefficient of the longitudinal band. The longitudinalSP bands are very sensitive to the aggregation ofgold nanorods and the peak points of the SP bandspredominantly depend on the anisotropic shape ofeach gold nanorod. The aggregation of rods generatesisotropic (random) coupling of SP oscillations, andproduces dramatic changes of the longitudinal SPbands. With these unique properties, gold nanorodsinteracting as a random cluster within a phospholipidencapsulated nanoparticle were expected to provideexcellent contrast for PAI in the NIR region.

The self-assembled gold nanobeacons (rods)[GNBR] was based on the self-assembly of phospho-lipids in aqueous media to entrap multiple copies(i.e. hundreds) of gold atoms.54 Parent gold nanorods(polymer coated GNR; transmission electron micro-scope axial diameter: 25 ± 5 nm, length: 80 ± 8 nm;lLSPR: 750 nm, lTSPR: 530 nm) (Figure 13(a))were dispersed in chloroform and premixed withvegetable oil (20% v/v almond oil) core matrix beforeencapsulated within phospholipids coating.

This synthesis resulted in an encapsulationof 127 gold metal atoms (ICP-MS = 8.12 μg ofgold/g of 20% colloidal nanobeacon), markedlyless than GNB1. The GNBR particles had nominalhydrodynamic diameter of 129 ± 7 nm as measuredby dynamic light scattering (DLS). The polydispersityand zeta potential were measured as 0.06 ± 0.02and −41 ± 12 mV (Brookhaven Instrument Co.),



22 nm

3 mm

0 100

Photoacousticsignal

25 nm

7.4 nm

820 pM

820 pM 2.5 nM 7.4 nM

22 nM 67 nM 200 nM

SkinFIGURE 9 | Photoacoustic (PA) detection ofsingle-walled carbon nanotube indocyanine green(SWNT-ICG) in living mice. Vertical slices ofultrasound images (gray) and PA images (green) ofmice injected subcutaneously with SWNT-ICG-RGDat concentrations of 0.82-200 nm (dotted black line).The white dotted lines on the images illustrate theapproximate edges of each inclusion. Quantitativeanalysis of the images estimated that 170 pm ofSWNT-ICG-RGD gives the equivalent PA signal as thetissue background. (Reprinted with permission fromRef 33. Copyright 2010 American Chemical Society)

- Human ATF

MNP

Polymer coating

Polymer coating

Abs

orba

nce

/ PA

sig

nal,

a.u.

1

Blood

MNPs

GNTs

16

0

Tumor

Tumor

2 3Weeks

Week 2

Week 4

4

Cel

l num

ber

/ min

4

0

8

12

700600 800

Wavelength, nm

900 1000

1000

100

10

Goldlayer

Carbonnanotube

core

Folate

PEG

(a) (b)

(c) (d)

FIGURE 10 | In vivo multiplex two-color photoacoustic (PA)detection of circulating tumor cells (CTCs). (a) The 10 nm magnetic NPs(MNPs) coated with amphiphilic triblock polymers, polyethylene glycol(PEG) and the amino-terminal fragment of urokinase plasminogenactivator (ATF). The 12 × 98 nm GNTs coated with PEG and folic acid.(b) PA spectra of 70 μm veins in mouse ear (open circles). Absorptionspectra of the MNPs and GNTs (dashed curves) normalized to PA signalsfrom CTC labeled with MNPs (black cricle) and GNTs (open circle).(c) The size of the primary breast cancer xenografts at different timestages of tumor development. (d) Average rate of CTCs in mouse earvein. (Reprinted with permission from Ref 34. Copyright 2009 NaturePublishing Group)

respectively. The large negative zeta potentialimplied successful phospholipids encapsulation andhigh colloidal stability of these nanoparticles. Thedehydrated state diameter (Dav) and height parameters(Hav) of the GNBR were 105 ± 28 and 60 ± 24 nm, asmeasured by transmission electron microscopy (TEM)

and atomic force microscopy measurements (AFM),respectively (Figure 13(b) and (c)). Contrary to GNB1the encapsulation efficiency of the gold nanorods waslow per GNBR, which was anticipated due to the largerand irregular shape of the gold nanorods. Ultimately,GNBR had less PA contrast relative to blood poolcontrast (threefold) and the larger gold nanorodsize exceeded the renal threshold for bioelimination,bringing the issue of long-term safety in patientswho would receive and likely retain these rodsfor life throughout their bodies. With these resultsand concerns, subsequent studies from the groupmainly utilized spherical gold nanoparticles similarto GNB1.

Sentinel Lymphnode ImagingAs mentioned previously, breast cancer patientsoften undergo sentinel lymphnode biopsy (SLNB)in lieu of radical axillary dissection to excludemetastatic disease in patients undergoing lumpectomy.However, the technique can be complicated and therecovery of lymph nodes (LNs) is less adequate thanexpected and often the key tissue is missed.59,60

Moreover, SLNB has been associated with postbiopsy complications e.g. local seroma formation,lymphedema, nerve injury and numbness of arms.61

In this context, efforts to improve noninvasiveimaging of the axillary LNs to guide minimallyinvasive percutaneous fine-needle biopsy (FNAB) orselective node resection for staging for breast cancerpatients remains clinically relevant. PAI offers markedadvancements over typical preoperative methyleneblue dye or radioactive labeling techniques to enhanceintraoperative localization and resection. Contrastagents injected subcutaneously will migrate intolymphatics and travel to the sentinel LNs and canbe detected at depths of 4–5 cm with PA imaging. Ina series of studies, Pan et al. demonstrated that thelymphatic transport of the nanopartculate contrastagents was highly influenced by size and mass.



FIGURE 11 | (a) Preparation of goldnanobeacons from octanethiol-functionalized goldnanoparticles (AuNPs), x = 1–2 mol% phospholipidcoating. (b) TEM image of gold nanobeacons (dropdeposited over nickel grid, 1% uranyl acetate; scalebar: 100 nm). (c) AFM image of gold nanobeacons.Average height Hav = 10,151 nm. (d) UV/Visspectroscopic profile. Solid blue line: goldnanobeacons; purple dashed line:octanethiol-coated AuNPs. Spectra are notnormalized. (Reprinted with permission from Ref52. Copyright 2009 Wiley Interscience)

Octylthiol functionalizedgold nanoparticles (AuNPs)

Octylthiol functionalizedgold nanoparticles

2-4 nm S

H

Au

(a)

(b)

(d)

(c)

H

O

Ox

x= 1-2 mole% of phospholipid coatingGNB-M

(Gold Nanobeacon)

~150 nm

1.4

Lipid encapsulation

Gold cluster

100 nm

5 μM

1.2

1.0

0.8

0.6

Abs

orba

nce

0.4

0.2

0.0400 500 600 700

Wavelength (nm)

800 900 1000 1100

O

HN

NH

NH

SCH2(CH2)6-CH3

Homing ligands(e.g. biotin)

Phospholipidcoating

Caproyl spacer

Octanethiol coated AuNPs

GNB-M(Gold Nanobeacon)

The effectiveness of GNB1 distribution intosentinel lymph node was studied first following asubcutaneous forepaw injection in a rodent model.53

A sagittal maximum amplitude projection (MAP)PA image of the axillary area clearly delineated thevasculature by virtue of the light absorbing red bloodcells with high spatial resolution of 500 μm. Within5 min, GNB1 was found accumulated in the SLNand in adjacent secondary LN in the same chain.The signal clearly highlighted the lymphnode at thisshallow depth, however, inorder to achieve a morerobust contrast particularly for nodes farther fromthe transducer, further formulation approaches wereconsidered and studied.

The most common, ‘bigger is better’ mentalityled to the development and evaluation of alarger PA agent with substantially more goldper particle than GNB1. In this approach, GNBwas a polymer encapsulated gold nanobeacon(GNBP). In a typical synthesis of GNBP, a di-block copolymer (polystyrene-b-polyacrylic acid) PS-b-PAA62–65 (Mn×103: 0.8-b-29.3 PDI = 1.18, 0.0033mmoles) was dissolved in a mixture of methanoland CHCl3 (4:1) and subjected to controlledevaporation under reduced pressure to generate athin film of polymer. The thin film was dispersedin deionized water (0.2 μM) by probe sonication

at ambient temperature. Octanethiol coated AuNPs(2 w/v%) were suspended in polysorbate (sorbitanmonolaureate (5 vol%) and microfluidized withPS-b-PAA dispersion (0.5 vol%) yielding a 10%colloidal suspension of nanoparticles (Figure 14).The nanobeacons were purified by exhaustivedialysis against infinite sink of nanopure waterusing cellulosic dialysis membrane (20 kDa MWCO).GNBP (289 ± 24 nm) had a narrow distribution(polydispersity index: 0.15 ± 0.04). Following thisapproach, the gold content of GNB1 (6120 metalatoms/particle) was impressively enhanced to reach71,493 gold metal atoms per GNBP. As expected,GNBP produced very strong PA signal, which wassignificantly higher than GNB1.53

However, contrary to our expectations, GNBPwas unable to provide signal enhancement in vivo.Injection of GNBP into the rat forepaw followedby serial examination of the draining lymphnodeswith PA imaging demonstrated no detectable signalover the anticipated course of the imaging session(Figure 6(d)). The LN territories of the animalswere reimaged daily for the the next 3 days withno significant migration to and uptake of GNBP bythe sentinel lymphnode. Despite a tenfold increasein gold content and ultrahigh PA signal contrast invitro, GNBP was ineffective for in vivo LN imaging.



3

0

0.08

0.06

0.04

0.02

740 760

Wavelength (nmm)

GNB control

780 800 820

0.05

−0.05

0.01

−0.01

0

0

2 mm

2 mm

2 mm

2 mm

0

1

100

Time (μs)

200

PA s

igna

l (V

p-p)

PA s

igna

l (V

p-p)

2

2

−1

0

5

10

740 760 780

Wavelength (nm)

800 820

Rat

io

15

PA s

igna

l (V

)

1

Blood

Blood

GNBGNB

(a) (b)

(e) (f)

(g) (h)(c) (d)

FIGURE 12 | (a) PA signals generated from a Tygon® tube (I.D. 250 μm, O.D. 500 μm) filled with GNB and rat blood. The excitation light is of764 nm wavelength. (b) PA spectrum of GNB and blood over a 740–820 nm range of NIR wavelengths. (c) Ratio of the peak-to-peak PA signalamplitudes generated from GNB to those of blood. (d) PA signal to noise ratio for the GNB and the control (no gold). Cross-sectional PA image of alow density polyethylene (LDPE) tube (1 cc volume, I.D. 6 mm) filled with plasma clot: (e) control, (f) targeted with GNB using a curved array PAsystem (λ= 800 nm). (g) Control, (h) targeted with GNB using a photoacoustic breast scanner system (λ= 532 nm). (Reprinted with permission fromRef 52. Copyright 2009 Wiley Interscience)

TEM image TEM image

Self assemblymicrofluidization

GNB-R

GNB-R

0.70.60.50.40.30.20.1A

bsor

banc

e

0400 500 600

Wavelength (nm)

700 800 900

Emptyparticles

110 nm

AFM images

(a) (b) (c)

(d)

FIGURE 13 | (a) TEM image of gold nanorods dropdeposited over Ni-grid; (b) TEM images of GNBR dropdeposited over Ni-grid; (c) AFM images of GNBR dropdeposited over glass; (d) UV–vis spectrum of goldnanorods showing LSPR and TSPR bands at 750 nm and530 nm, respectively. (Reprinted with permission fromRef 54. Copyright 2009 American Scientific Publishers)

We hypothesized that the larger size (and or mass)of GNBP led to poor migration of the particles toand within the lymphatic channels, essentially leavingthe contrast agents essentially at the injection site.Since bigger particles with more gold was ineffective,we explored the potential that a smaller particle withless gold, smaller size, and less mass would have betterlymphatic migration. A third smaller gold nanobeacon(GNBS), was produced and examined.53

GNBS was prepared similar as GNB1 byretuning the core matrix to polysorbates (Figure 14).Briefly, oleate-coated AuNPs (2 w/v% of inner matrix,2–4 nm) were suspended in polysorbate (sorbitanmonolaureate, 20 vol%) and homogenized with asurfactant mixture, resuspended from a lipid film, at137.9 MPa (i.e., 20,000 PSI) for 4 min. The surfactantmixture was mainly comprised of phosphatidylcholine(PC) (92 mol% of lipid constitutents) and cholesterol



Polysorbatesself assembly

AuNP

2–4 nm

GNB-L

GNB-L

GNB-M

GNB-M

Phospholipidsmixture

PS-b-PAA

Veg oilself assembly

Polysorbatesself assembly

(Polymer-entrappedGold Nanobeacon)

m

n

CO2H

(Gold Nanobeacon)

GNB-S

Efficient lymph node detection

Phospholipids

Sorbitanmonolaureate Vegetable oil

Increasing size

MORE GOLD

PS-b-PAA

CO2H

m

n

GNB-S

(Lipid-entrappedGold Nanobeacon)

Dav = 289 ± 24 nm

Hav = 153 ± 31 nmPDI = 0.15 ± 0.04Gold atoms = 71493/NB

z = −35 ± 8 mV

Dav = 92 ± 13 nm


z = −55 ± 14 mV

Dav = 155 ± 11 nm


z = −48 ± 7 mV

Octanethiolcoated AuNP(2–4 nm)

AuNP

Dav = < 100 nm Dav = < 150 nm Dav = < 300 nm

FIGURE 14 | Top: Synthesis of GNBs(Dav = Number averaged (DLS); ζ =electrophoretic (zeta) potential; Hav = averageheight (AFM); PDI = polydispersity index (DLS)).Bottom: a general scheme showing the effect ofparticle size in facile lymphatic distribution.(Reprinted with permission from Ref 53.Copyright 2010 Elsevier BV)

(8 mole %). Hydrodynamic particle size for GNBS was92 ± 12 nm with a polydispersity index of 0.35 ± 0.05.The ζ potential was −35 ± 8 nm, confirmingappropriate particle encapsulation. Gold metal atom(nine gold metal atoms per GNBS) content per particlewas determined to be much less in comparisonto GNB1 and GNBP. The result with GNBS wasdramatically improved relative to GNB1 and GNBp.

In the rat sentinel lymph node model, the blood-filled microvasculature was clearly visible (markedwith red arrows) while the lymph node, devoid ofoptical absorbers, was not seen at baseline as before.Figure 6(f) and (g) presents two MAP images ofthe same axillary area 5 and 20 min after GNB-Ssubcutaneous injection into the mouse forepaw,respectively. The SLN is dramatically apparent as

a bright spot (green arrow) and the conspicuity of thelymph node far exceeds what achieved previously withGNB1 or GNBp. GNBS flowing within the lymphaticvessel toward and into the sentinal lymphnode wasobserved at 5 min post injection. Within 10 minspost GNBS injection, the nanobeacons completedmigration into LN and a persistent and robust signalwas appreciated over the next hour. The contrastof the sentinel LN to the surrounding blood vesselwas calculated to be 9:1 (ratio of the peak-to-peakPA signal amplitude obtained from blood vessel andSLN) after GNBS injection.53 In summary, the smallerGNBS, with less gold particles per nanoparticle, lessmass, and smaller size were more effective in detectingthe SLN in comparison to the larger GNB particles.



GNB-M

GNB-S GNB-S

GNB-L

Control

Control

Control5 min

5 min 20 min

60 min

0.8

0.6

0.4

0.2

(a) (b) (c) (d)

(e) (f) (g)

FIGURE 15 | In vivo noninvasive photoacousticimaging of sentinel lymph nodes in rat (λ = 767 nm). (a–g)Scale bar is 5 mm, 150 mL of nanobeacons were injectedintradermally in all the cases. GNB1: (a) Control PA image;(b) 5 min post-injection image of GNB1 (5 mM). GNBP: (c)Control PA image. (d) Lymph node is not visible in a 60 minpost-injection image of GNBP (680 nM). GNBS: (e) Sagittalmaximum amplitude projection (MAP)66 pre-injectioncontrol image: Bright parts represent optical absorptionfrom blood vessels, marked with red arrows. (f) PA image(MAP) acquired 5 min after GNBS injection (10 nM). SLNsare clearly visible, marked with green arrow. Lymphaticvessel is also visible, marked with blue arrow. (g) 20 minpost-injection PA image. (Reprinted with permission fromRef 53. Copyright 2010 Elsevier BV)

Moreover, the particles were retained adequately inthe LN for practical use (Figure 15).

Integrin-Specific PAI of AngiogenesisAngiogenesis is an essential process to expandthe vasculature during embryonic development andadult life. In many inflammation related diseases,the angiogenic neovasculature is a microanatomicalfeature of medically important diseases, includingcancer, atherosclerosis, and arthritis to name afew. Neovascularization reflects a multitude ofpathologic processes that include cytokine liberationby inflammatory cells to stimulate endothelialproliferation, neovessel sprouting and microvesseldevelopment, as well as the secretion of matrixdissolving enzymes, such as metalloproteinases,elastases, and collagenases, that facilitate endothelialcells and tubule penetration through the extracellularmatrix in cancer and arthritis progression.67–69 Thelist of diseases that have angiogenesis as an underlyingmechanism grows longer every year.

Although PAT generates high-resolution imagesof red blood cells in the microvasculature,1,70–75

hemoglobin imaging cannot discriminate immatureneovasculature from mature microvessels. Moreover,neovessel sprouting and bridging, which is asignificant component of the neovascular contrastsignal typically lacks any blood flow and isotherwise invisible to PA imaging. The αvβ3-integrin,a heterodimeric transmembrane glycoprotein,, isexpressed by nonpolarized neovasculular endothelialcells76 as well as numerous other cell types includingmacrophages,77 platelets,78 lymphocytes,79 smoothmuscle cells80 and tumor cells.81,82 Fortunately, the

steric constraint of GNB and other particles exceeding100 nm diameter within the vasculature precludessignificant interaction with nonendothelial integrin-expressing cells and greatly enhances neovasculartarget specificity.83

To date, PA contrast agents targeting angio-genesis have been designed around integrin-targeted indocyanine green (ICG)-fluorescent-peptideconjugates10,84–88 and coupled to SWNT. Unfortu-nately, small molecule agents easily diffuse from theneovascular vasculature and can bind the numer-ous cell types expressing ανβ̃3 integrin. SWNT,although interesting academically, pose challengesfor clinical translation in terms of bioeliminationand long term safety. ανβ̃3 GNB1 was vascularlyconstrained and could not be detected microscopi-cally beyond neovasculature of the Matrigel™ plug.This observation was further supported independentlyby other reports using copolymer and nanoemulsionparticles.83,89

αvβ3-targeted GNB1 (160 nm), as opposedto GNBS (90 nm) with a higher potential forextravasation or GNBP (290 nm) which was expectedto have a short circulatory half-life due to itsoverall mass and size. αvβ3-targeted GNB1 wasproduced by microfluidization as discussed beforeby incorporating an αvβ3-peptidomimetic antagonist.αvβ3-peptidomimetic antagonist was conjugatedto PEG2000-phosphatidylethanolamine (0.1 mole%,Kereos, Inc, St. Louis, MO, USA).90 This quinalonenonpeptide was developed by Lantheus MedicalImaging (Billerica, MA, US patent 6,511,648 andrelated patents) and produced by Kereos, Inc. Thevitronectin antagonist was initially reported and



Baseline 15 min 45 min 1.5 h 2.0 h

2.5 h

5.0 h

3.0 h 3.5 h 4.0 h 4.5 h

(a) (b) (c) (d) (e)

(f)

(k) (l) (m)

(g) (h) (i) (j)

FIGURE 16 | Matrigel™ (0.75 mL) was implanted subcutaneously in nude mouse. The mouse was imaged photoacoustically 8–20 days afterMatrigel™ implantation. (a) Photoacoustic (PA) maximum amplitude projection (MAP) image of the dotted area. This is a control image. After thecontrol image was taken targeted gold nanobeacons (αvβ3-GNB-M) were injected intravenously using the tail vein. In a time course study (b–k), PAimages were acquired with an interval of approximately 1/2 hour up to 5 h.(g) Three hour post-injection PA image. Red arrows point to the angiogenic sprout (not visible in A).(k) Five hour post-injection PA image. For all PA images λ = 767 nm, scale bar = 5 mm. (l) Digital photograph of a mouse implanted with Matrigel™plug. Blue arrow points to the plug. The black dotted area was imaged. The smallest tick: 1 mm. (m) Digital photograph of the sacrificed mouse afterall the image acquisition was over. The skin was removed to show the Matrigel™ plug (blue arrow). (Reprinted with permission from Ref 55.Copyright 2010 by the Federation of American Societies for Experimental Biology)

characterized as the 111In-DOTA conjugate RP478and cyan 5.5 homolog TA145.91 The specificityof the αvβ3-ligand mirrors that of the anti αvβ3-LM609 antibody92 (Chemicon International, Inc.,Temecula, CA, USA) as assessed by staining and flowcytometry. The IC50 for αvβ5, α5β1, and GP IIbIIIawas determined to be >10 μM (Lantheus MedicalImaging, unpublished data).

Neovascular imaging was performed in vivo in aMatrigel™ plug model of angiogenesis. Matrigel™(750 μL) enriched with fibroblast growth factor-2(500 nm/mL; Sigma Aldrich, St. Louis, MO) andheparin (64 U/mL) was implanted subcutaneouslyalong the flank of mice. PAT imaging was performedprior to treatment and serially imaged over 5 hpostinjection on days 16 and 17 post Matrigel™implantation (Figure 16). Animals were randomlydistributed into four groups and received: 1)αvβ3-gold nanobeacons (αvβ3-GNB1, n = 6) αvβ3-nanobeacons without gold (αvβ3-NB) followed byαvβ3-GNB1 (competitive blockage, n = 2) nontargetedgold nanobeacons (NT-GNB1, n = 3) or saline (n = 2).

For the first time, high spatial resolutionnoninvasive PAT imaging of angiogenesis was

demonstrated using a 10 MHz ultrasound receiverwhich clearly revealed the formation of nascentneovessel tubules, sprouts and bridges much of whichwas still without blood flow. αvβ3-GNB1 produceda 600% increase in signal in a Matrigel™ plugmouse model relative to the inherent hemoglobincontrast pretreatment. Competitive inhibition of αvβ3-GNB1 with αvβ3-NB (no gold) blocked contrastenhancement to pretreatment levels. Similar imagesin the saline control animals showed no changein vascular anatomy over the same time course.Indeed these images illustrate the genesis ofneovasculature in the Matrigel™ plug model. Asimilar effect was seen with saline treated animals.Nontargeted GNB passively accumulated in thetortuous neovascularity, but provided low (less thanhalf) contrast enhancement of the targeted agent. PAsignal changes in the Matrigel™ plug were monitoredserially over 5 hours or more (Figure 7).

The Matrigel angiogenesis study was supportedhistologically using a FGF Matrigel subcutaneousexplant from FVB/N-TgN(TIE2LacZ)182Sato micefollowing injection (IV) of αvβ3-targeted rhodaminelabeled GNB-M nanoparticles. These transgenic



Matrigel plugανβ3-Targetedrhodamine NP

PECAM-Pos.endothelium

Tie-2 Pos.endothelium

Muscle

Matrigel

Skin

1.0 mm 0.2 mm 0.2 mm

0.2 mm

0.2 mm

0.2 mm0.2 mm

(a)

(b) (c) (d)

(e) (f) (g)

FIGURE 17 | Microscopic examination of FGF Matrigel subcutaneous explant from FVB/N-TgN(TIE2LacZ)182Sato mice following injection (IV) ofαvβ3-targeted rhodamine labeled GNB-M nanoparticles. These transgenic mice carry a β-galactosidase reporter gene under the control of the murineTek (Tie2) promoter. LacZ is expressed specifically in vascular endothelial cells in embryonic and adult mice. (a) H&E staining of the excised implantproviding spatial orientation of the matrix with respect to skin and muscle. The red box region is further examined in panels b, c, and d. The blue boxregion is studied in more detail in panels e, f, and g. Panels b and e depict the accumulation of αvβ3-targeted rhodamine nanoparticles in the red andblue tissue regions respectively. Note the brilliant and dense accumulation of NPs in panel b (red arrows) and little to no accumulation of particles inthe panel e. Region. Panels c and f depict the staining of vascular endothelium for PECAM (CD34) in the red (c) and blue (f) regions of the matrigelplug. There was dense vascularity in both locations (Red arrows in c and turquoise arrows in f. Panels d and g depict the LacZ signal forβ-galactosidase under Tie2 promoter control. In panel d, no LacZ signal was appreciated, reflecting a paucity of mature microvessels. Incontradistinction, there is strong LacZ signal in panel g. These results indicated that the αvβ3-GNB nanoparticles were specifically targeted toangiogenic endothelial cells (PECAM +/Tie-2−) and not to more mature microvessels, which were PECAM +, Tie-2+. The data corroborated thatPAT imaging with αvβ3-targeted GNB specifically distinguished and enhanced the angiogenic neovasculature from new, but more matured anddifferentiated microvessels. (Reprinted with permission from Ref 55. Copyright 2010 by the Federation of American Societies for ExperimentalBiology)

mice carry a β-galactosidase reporter gene underthe control of the murine Tek (Tie2) promoter.LacZ is expressed specifically in vascular endothe-lial cells in embryonic and adult mice. Microscopydefinitively established that rhodamine-labeled αvβ3-GNB1 targeted specifically to immature neovascu-lature expressing PECAM+ but not Tie-2− local-ized along the implant periphery, but not to otherperipheral microvasculature expressing PECAM+ andTie-2+ (Figure 17). This experiment demonstratedthat αvβ3-GNB1 PA imaging sensitively and specif-ically discriminated angiogenesis from the adjacentmicrovasculature.

SMALL MOLECULE-BASED ANDOTHER EXOGENOUS PA AGENTS

NIR Cyanine Dyes for Real-Time SentinelLymph Node ImagingAs we discussed previously, identification of SLN isa critical component of breast cancer staging andmanagement.93–95 Another approach to this issue wastaken by Pan et al.96 wwith the development of a

sub 20 nm ‘soft’ polymeric nanoparticle (Figure 18).This new agent was designed for rapid intraoperativeadministration with real time PA imaging. Theintraoperative approach helps to eliminate anatomicplane displacement, which can complicate detectionbetween the preoperative images and the patientrepositioning in the OR. Moreover, by establishing arapid operative procedure, the surgeon can determinethe most direct approach to the node for resectionminimizing potential secondary complications of theaxillary dissection. The obvious key to this procedureis rapid signal generation with adequate intraoperativepersistence.

The synthesis used by Pan et al. utilized a onepot rapid synthetic pathway in which polysterene-b-poly(acrylic acid) (PS8-b-PAA400, Mnx103,0.8-b-33.0, polydispersity index: PDI = 1.18,0.00033 mmol, 0.5 mole%) was mixed with a co-surfactant, sorbitan monoleate, to restrict the particlediameter to 20 nm. A NIR cyanine dye (ADS832WS,λex = 824 nm, 1.90 × 105 L mol−1 cm−1) was incor-porated to act as a surrogate PA payload in this proofof concept study as shown in Figure 18.



m

A

A

B

B

Co-self assembly

NIR-polymer micelle

ADS-832-WS

ADS-832-WS

Drug

Dav 16 ± 2 nm

-b-PAA(n = 8; m = 400) Dialysis

(i)-(ii)

Rapid formation of micelles

Incorporates both hydrophilic

Sub 20 nm sized particle forextravascular application

Polysorbate

Fumagillin

OHO

O

O

OO

O

OO20

20

20

20

HOOH

OH

NH2

S

O

O

O

O O

OH

H

OMe

-O3S SO3Na

N N⊕

n

(practically finished in minutes)

and hydrophobic payloads

PS

FIGURE 18 | Synthesis and characterization of theranostic polymeric micelles: (1) co-self assembly of amphiphilic diblock copolymer and sorbitanmonooleate, ADS-832-WS (a) and/or fumagillin (b), sonication, 25 ◦C, 1 min; (2) dialysis 10 kDa cellulosic membrane, nanopure water (0.2 μM).(Reprinted with permission from Ref 96. Copyright 2011 Wiley Interscience)

Particle characterization was performed throughmany analytic techniques including TEM, AFMand DLS, which revealed the hydrodynamic particlediameter (Dav) to be 16 ± 2 nm with a polydispersityof 0.021. The anhydrous particle diameter was(Dah = 12 ± 0.4 nm) based on TEM images andpresented a height of 15 ± 0.4 nm based on AFMimages. Because of the incorporation of the NIR dye,the particle had high fluorescence with confirmedabsorbencies at 750–860 nm. Additionally, theseparticles were found to possess long shelf-life stability(40 days at ambient temperatures). Loading efficiencyof dye was approximately 97% (Figures 18 and19). Interestingly, the new particle also possessedtheranostic potential, incorporating fumagillin andproviding a prolonged release with dissolution testingin vitro.

PA tomographic imaging of the new agent as amarker for axillary lymph nodes was studied in rats.Intradermal injection of the nanoparticles showed

that the PA signals via maximum intensity projection(MAP) (Figure 20) produced a contrast enhancementof 510% almost immediately. The signal enhancementwas retained for 30 min at the SLNs and within110 min, was almost completely drained. Such resultssuggest that these nanoparticles were delayed in transitwithin the lymph nodes but maintained their structuralintegrity and migrated out of the lymph nodes intothe circulation system, which was attributed to theirminiature size.

Copper as a Contrast Agent in PA Imagingof Sentinel Lymph NodesIn an interesting study, Pan et al.97 has demonstratedhow copper could be utilized as an exogenous PAcontrast agents for the identification of SLNs. Thesynthesis followed a self-assembly of Cu-neodecanoatecomplexes with polysorbates (sorbitan mono-9-octadecenoate poly(oxy-1,2-ethanediyl)) which wereencapsulated within a phospholipid outer layer. As



TEM

AFM

102 12

10

8

6

4

2

0

100

98

96

94

Rabbit plasma

Fumagillin releaseHyman plasma albumin

0 1 2

Days

Rem

aini

ng %

Dru

g re

leas

e (%

)

3 0 1 2

Days

3 4

25 nm

2 μm

Dav (DLS/nm) = 16 ± 2

Dah (TEM/nm) = 12 ± 4

Dye encapsulation: 97%

Fumagillin release: 18% (3d)

Hav (AFM/nm) = 15 ± 4

z (Zeta/mV) = -10 ± 3

l (Uv-vis/nm) = 824lex/lem (Fluor/nm) 824/832

(a)(b)

(d) (e)

(c)

FIGURE 19 | (a) TEM image of the polymericmicelles drop desposited over nickel brid, scalebar = 100 nm; (b) AFM image of the micelles dropdeposited over glass; (c) physio-chemicalcharacterization table for a NIR-polymer micelle; (d)dissolution of ADS-832-WS from micelle whenincubated against rabbit plasma and human plasmaalbumin; (e) time-dependent release of fumagillinfrom micelle. (Reprinted with permission from Ref 96.Copyright 2011 Wiley Interscience)

MAP (top view)5

2 cm SLN

SLN removed

PA Imaging

(e)

(h)

(g)

(f)4

3

2

Opt

ical

abs

orpt

ion

1

0

6

5

PA s

igna

l enh

ance

men

t

4

3

2

10 20 40 60 80

Post-injection time (min)

100 120

Control(a) (b)

(c) (d)

BV BV

SLN

SLN

Wavelength =750 nm; Animal: rat; Head on the left.

BV: Blood Vessel; LV : Lymph Vessel; SLN: Sentinel Lymph Node

SLN

LV

LV

4 mm

10 mm

90 min30 min

FIGURE 20 | Noninvasive in vivo PA images of SLN in rat: For all PA images, the laser was tuned to 750 nm wavelength. (a) Control PA imageacquired before polymeric nanoparticle injection. Bright parts represent optical absorption, here, from blood vessels (BV). (b) PA image (MAP)acquired immediately after the polymeric nanoparticle injection. (c) Post-injection (35 min) PA image. (d) Post-injection (110 min) PA imageBloodvessel (BV), lymph vessel (LV) and sentinel lymph node (SLN) are marked with arrows, and the SLN is visible in (b)–(d), however, invisible in (a). (e)Photograph of the rat after the hair was removed from the scanning region before taking the PA images. The scanning region is marked with a blackdotted square. (f) Photograph of the rat with the skin removed after PA imaging. (g) Excised lymph node. Smallest tick: 1 mm. (h) PA signalenhancement in the SLN after the injection of polymeric nanoparticle as a function of post-injection time. For (a)–(d): FOV = 25 mm × 24 mm, stepsize along the X direction = 0.2 mm, step size along the Y direction = 0.4 mm, total scan time = − 23 min. No signal averaging was used. (Reprintedwith permission from Ref 96. Copyright 2011 Wiley Interscience)



FIGURE 21 | Synthesis andphysiochemical characterization ofself-assembled nanoparticles of copperneodecanoate. Schematic describing thepreparation of copper-enriched nanoparticles:(1) suspension of copper neodecanoate (1) insorbitan sesquioleate, vigorously vortex andmixing, filter using cotton bed, vortex; (2)dissolve phospholipids in anhydrouschloroform and preparation of phospholipidthin film by slow evaporation of solvent at45 ◦C under reduced pressure; (3)resuspension of the thin film in water (0.2μM); (4) self-assembly by high pressurehomogenization at 4 ◦C, 20000 psi (141 MPa),4 min; (5) dialysis (cellulosic membrane,MWCO 20 k); in box, characterization tablefor nanoparticles of copper neodecanoate.(Reprinted with permission from Ref 97.Copyright 2011 American Chemical Society)

Polysorbate

Phospholipids

80 nm

Copper nanoparticle

HO

CuO

OO

O

(i)–(v)

‘Self-assembly’

Cu(II)-Neodecanoate

Physico-Chemical Characterization Table

Hydrodynamic diameter (Dh) = 86 ± 06 nm

Polydispersity index (PDI) = 0.21 ± 0.02

Zeta potential (ζ ) = −12 ± 0.7 mV

Particle height (Hav) = 60 ± 14 nm

UV-vis spectroscopy = λ 603–746 nm

EDX = C 22 wt%, O 6 wt%, Cu 11 wt%

ICP-OES = 16.99 mg/L of Cu (20% colloidal suspension)

O

O O

O

O O

OOH

OHy

zx

w

~

synthesized nanoparticles were subsequently purifiedby exhaustive dialysis to remove unbound copper(Figure 21).

The copper-rich nanoparticles had a hydrody-namic diameter of 86 ± 0.6 nm (Dh) with low poly-dispersity of 0.21 ± 0.02. Negative electrophoreticpotential of −12 ± 07 mV indicated a complete encap-sulation of the copper-polysorbate core with thesurfactant. Further characterization of these nanopar-ticles was performed using TEM, AFM, SEM,and energy-dispersive X-ray spectroscopy (EDX)(Figure 22).

Cu-particles exhibited a PA signal that wassignificantly stronger and greater (the PA signal ampli-tude obtained from NanoCuN was 650 mV com-pared to that of 140 mV from blood) at 680 nm.Also, NanoCuN particles exhibited a strong absorp-tion in the NIR region compared to a weak onedisplayed by hemoglobin. In vivo studies were con-ducted in rats as previously described to image thesentinel lymph nodes via PA imaging. Figure 23shows a baseline image of the rat before NanoCuNadministration presented as a maximum intensity pro-jection (MIP) (Figure 23(b)). An image immediatelyafter NanoCuN injection into the paw was showed ini-tial uptake by the SLN (Figure 23(c)) which continuedto increase for 60 min (Figure 23(d)).

Porphysome Nanovesicles for MultimodalBiophotonic Contrast AgentsMany attempts at creating nanoparticles withoptically active inorganic molecules have beenstudied,98–101 but despite their drug loading efficiencyand questions regarding long-term safety,102–105

they have not been translated into the clinic.Organic nanoparticles have been shown to havemore biocompatibility.105 Based on these premises,Lovell et al.106 synthesized a supramolecularself-assembled porphysome using porphyrin-lipidconjugates. Polyethylene glycol (PEG)-lipids wereincluded to enhance in vivo pharmacokineticproperties as shown in Figure 24. These particleswere spherical about 100 nm in diameter as confirmedby TEM. With greater magnification, it was shownthat the porphysome structure contained two layersof higher-density material separated by 2 nm gaps. Theporphysome showed two distinct absorption peaks at400 and 680 nm, confirming the properties regardingthe NIR nature of the particle. By inserting metal ionsinto the structure, Lovell was able to demonstrate theshifts in optical density bands at 440 and 670 nm,which also suggested that these porphysomes retainedavailability for metal chelation.

Porphysomes were tested for its photochemicalproperties by exposing them to laser irradiation. Ther-mal energy was released, which was similar in efficacy



0 100

400C 200 nm

15 μm

O Cu

350

300

250

Cou

nts/

s

200

150

100

50

120 110

Saline

Human plasma albumin

106

Rem

aini

ng % 102

98

94

900 1 2 3

100

Dav

(DLS

/nm

)ζ(

/mv)

DLS

Zeta

80

60

40

200

0

0 50

Time (Days) Time (Days)

100−20

00 20 40 60 80 100

Energy (keV)

120 140 160

200 300 400 400 500 600 700

Wavelength (nm)

800500 600

Hydrodynamic diameter (nm)

Lipidcoating

500 nm 1 μm

Cu

700 800 900 1000

20 0.9

0.8

0.7

Abs

orba

nce

(a.u

.)

0.6

0.5

0.4

0.3

(a)

(c)

(f) (g)

(i)(h)

(d) (e)

(b)

15

10

Num

ber

aver

age

(%)

5

0

FIGURE 22 | (a) Hydrodynamic particle size distribution from DLS; (b) UV–vis spectrum of NanoCuN nanoparticles in water; (c) anhydrous stateTEM image; (d, e) SEM images; (f) EDX spectrum of the selected area from the image in (e); (g) AFM image (deposited on glass substrate); shelf lifestability of NanoCuN over 90 days from formulation; (i) dissolution of Cu over 3 days when incubated with saline and human plasma albumin.(Reprinted with permission from Ref 97 Copyright 2011 American Chemical Society)



(a)

(b)

BV

4 mm

0.5 1 1.5 2.5

Optical absorption

2 3

(c) (d)

(e)

2 cm

SLN

BV

SLNSLN

FIGURE 23 | Noninvasive in vivo PA images (maximum intensity projections, MIP) of the SLN in a rat. (a) Photograph of the rat with region ofinterest depilated before scanning. The scanning region is delineated by a black dotted square. (b) Baseline PA image acquired before NanoCuNinjection. Bright regions represent inherent optical absorption from a blood vessel (BV). (c) PA image (MIP) acquired almost immediately afterNanoCuN administration. (d) PA image 60 min post injection showing a marked signal enhancement corresponding to increased NanoCuN uptake bythe node: blood vessel (BV) and sentinel lymph node (SLN) are marked with arrows. The SLN is visible in both (c) and (d), however not apparent in (b).(e) Photograph of the rat with the skin excised after PA imaging. For (b)–(d), FOV = 25 × 24 mm, step size along the X direction = 0.2 mm, step sizealong the Y direction = 0.4 mm, total scan time = 23 min. No signal averaging was used. (Reprinted with permission from Ref 97. Copyright 2011American Chemical Society)

to that of gold nanorods. PA amplitudes were atleastsix fold (670 nm) higher for porphysomes than methy-lene blue, which is often used in the clinic for SLNdetection, as mentioned earlier. The addition of deter-gent to the porphysome decreased the PA signal by six-fold while it had no effect on the methylene blue, indi-cating the necessary self-quenching nature of the por-physome to give off its PA signals (Figure 25(a)–(c)).

To further show that porphysomes could beused concurrently with PA tomography, in vivostudies were implemented by injecting porphysomesintradermally in rats. Figure 25(d) shows that the locallymphatic network was clearly visible and detectablewithin 15 min post-injection. Porphysomes displayedvery strong PA signals affording distinct localizationof the lymph node distinct from surroundings vessels.After 2 days, porphysomes accumulated within thetumor area and exhibited high fluorescence emissiondue to receptor-mediated endocytosis or more likelysplenic uptake by macrophages which migrated to theinflammatory sites associated with the tumors, as iscommon for many particles.

CONCLUSION AND FUTURE OF PATOMOGRAPHY

For most of the preclinical work, the instrumentationis now commercially available; whereas the majority

of the work reported was conducted with laboratory-built robotic scanners. Visual Sonics, Inc., asubsidiary of SonoSite, Inc. was the first companyto commercialize PA systems. They integrated PAcapability into their existing Vevo LAZR™ platformto enhance high-resolution ultrasound-derived imageswith the sensitivity of optical imaging. This systemprovides real-time, in vivo imaging of deep tissuewithin a pebetration capability up to 1 cm. Theintegrated, 20Hz tunable laser can be operated within680–970 nm and a resolution down to 45 μm caouldbe achieved. High optical contrast is co-registered withhigh-resolution imaging and an advanced analysissoftware package provided the ability to capture2D and 3D images. Recently Endra Life Scienceslaunched Nexus128, a preclinical small animalimaging scanner with multimodal capabilities ofPAI and CT. This system is the only commerciallyavailable fully 3D PA CT scanner. The system iscapable of generating fast, noninvasive quantificationof physiological parameters such as tumor vasculaturewithout ionizing radiation. Using Endra’s Nexus128™, Dominique Van de Sompel et al. demonstratedimproving image quality by correcting temperaturechanges during long scans in both phantom and tissuesamples.107 Seno Medical Instruments, Inc. developedan Opto-acoustic system, which allows for theidentification of tumors 2 mm and has demonstrated



X = 2Hor

for example, Zn, Cu, Pd

100 nm 100 nm 5 nm

O

6

4

2

125

100

75

50

25

1

Sro

d/S

pyro

-por

ph

0

20

15

10

5

01 10 100 1,000

Size (nm)

1 10 100 1,000

Size (nm)

1 10 100 1,000

Size (nm)

% d

istr

obut

ion 25

20

1015

50%

dis

trob

utio

n 2520

1015

50%

dis

trob

utio

n

300 400 500 600Wavelength (nm)

Propheophorbideporphysomes

Zinc-pyropheophorbideporphysomes

Bacteriochlorophylporphysomes

Wavelength (nm)

ε(×1

04 M

−1 c

m−1

)

700400 500 600 700

800

O(a)

(b)

(c)

(e)

(d)

OO

OO

POO

ON

NN

N N

+

FIGURE 24 | Porphysomes are optically active nanovesicles formed from porphyrin bilayers. (a) Schematic representation of apyropheophorbide–lipid porphysome. The phospholipid headgroup (red) and porphyrin (blue) are highlighted in the subunit (left) and assemblednanovesicle (right). (b) Electron micrographs of negatively stained porphysomes (5% PEG-lipid, 95% pyropheophorbide-lipid). (c) Absorbance of theporphyrin–lipid subunits incorporated in porphysomes formed from pyropheophorbide (blue), zinc pyropheophorbide (orange) andbacteriochlorophyll (red) in PBS. (d) Resonance light scattering spectra ratio between gold nanorods and pyropheophorbide porphysomes. Nanorodand porphysome concentration was adjusted to have equal optical density at 680 nm. (e) Dynamic light scattering size profiles of indicatedporphysomes recorded in PBS. (Reprinted with permission from Ref 106. Copyright 2011 Nature Publishing Group)



PBS(a)

(d)

Pre-injection 15 min post i.d.

0 836Counts

15 min post i.v. 48 h post i.v.

(b) (c)

Gold nanorods Porphysomes

Porphysomes

Methylene blue

LiposomesPBS

Porphysomes

Porphysomes+ det.

PBS + det.

P.A. intensity0 1

5

(P.A

)/(P

.A. w

ith d

et.)

4

3

2

1

0745 760 775 790 805

Wavelength (nm)

55 Tem

perature (C°)

20

FIGURE 25 | Multimodal optical utility of porphysomes. (a) Photothermal transduction. Solutions were irradiated with a 673 nm laser andimaged with a thermal camera. (b) Ratio of photoacoustic amplitudes (P.A.) measured for porphysomes and methylene blue_0.5% Triton X-100(mean_ s.e.m. from 10 measurements); det., detergent. (c) Photoacoustic images of tubing containing porphysomes and PBS measured_0.5% TritonX-100. (d) Dual modality for photoacoustic contrast and activatable fluorescence. Top, lymphatic mapping. Rats were imaged using photoacoustictomography before and after intradermal (i.d.) injection of porphysomes (2.3 pmol). Secondary lymph vessels (cyan), lymph node (red), inflowinglymph vessel (yellow) and 5 mm scale bar are indicated. Bottom, fluorescence activation after i.v. injection of porphysomes (7.5 pmol) in a KBxenograft-bearing mouse. (Reprinted with permission from Ref 106. Copyright 2011 Nature Publishing Group)

the ability to see submillimeter structures. Recently,Seno Medical has licensed their technology toVisual Sonics. TomoWave Loboratories, Inc., aTexas based company is also working towards thecommercialization of this technique. As described

in TomoWave website, their optoacoustic systemLOUIS-3D™ consists of a Q-switched laser Spectra-Wave, generating wavelength 1064 nm, 532 nm andtunable in the NIR from 730 nm to 850 nm. Of note,the first PA scanners have been placed in clinics



as prototypes for augmenting ultrasound techniques,such as SLN dye injection guidance.

As discussed above, PAI has been demon-strated by several groups using agents, e.g.organic dyes,108 gold nanoparticles,52–55 copperneodecanoate nanoparticles,97 single-walled carbonnanotubes (SWNT)39,51 and others.96 These agentshave shown definite potential for PA SLN mapping,imaging angiogenesis and melanoma with a depthcapability of several centimeter, satisfactory spatialresolution. However, their use is not always outof questions. Organic dyes are extensively used inclinics but they are rather small (<2 nm), and thuscan easily be transported into the echelon lymphnodes, causing a high possibility of producing falsepositives.109 Often times they cause skin staining thatleads to irritation. Gold nanoparticles of different sizesand morphologies are utilized widely, however, theiroptical properties are highly dependent on expensiveand complicated surface chemistries. Their clinicaluse is also discouraged by the unpredictable high

cost. Copper-based approach could potentially pro-vide an innovative, inexpensive, and commerciallyviable approach. Although these particles were syn-thesized in a highly stable manner, copper is known topose neurotoxicity.110 SWNTs are inexpensive but thesafety of these materials is still an ongoing debate.111

With the constant advancement of PA technology, thefuture of this novel modality will also be dictatedby the evolution of safer and sensitive molecularlytargeted contrast agents.

The clinical opportunities garnered from theusage of PAT with approved dyes such as methyleneblue are significant, and may be further improvedwith some of the preclinical agents discussed whichwould afford more robust and effective PA imaging.Moreover, the addition of physiological and targetedmolecular imaging with PA alone and in concertwith the new preclinical agents offers a significantexpansion of the diagnostic utility of ultrasound,which remains one of the most cost effective andprotable imaging modalities in medicine today.

REFERENCES1. Wang XD, Pang YJ, Ku G, Xie XY, Stoica G,

Wang LHV. Noninvasive laser-induced photoacoustictomography for structural and functional in vivoimaging of the brain. Nat Biotechnol 2003,21:803–806.

2. Li ML, Oh JT, Xie XY, Ku G, Wang W, Li C, LunguG, Stoica G, Wang LHV. Simultaneous molecularand hypoxia imaging of brain tumors in vivo usingspectroscopic photoacoustic tomography. Proc IEEE2008, 96:481–489.

3. Wang XD, Xie XY, Ku G, Wang LHV. Noninvasiveimaging of hemoglobin concentration and oxygena-tion in the rat brain using high-resolution photoacous-tic tomography. J Biomed Opt 2006, 11:024015.

4. Lungu GF, Li ML, Xie XY, Wang LHV, Stoica G. Invivo imaging and characterization of hypoxia-inducedneovascularization and tumor invasion. Int J Oncol2007, 30:45–54.

5. Ku G, Wang XD, Stoica G, Wang LHV. Multiple-bandwidth photoacoustic tomography. Phys Med Biol2004, 49:1329–1338.

6. Ku G, Wang XD, Xie XY, Stoica G, Wang LHV.Imaging of tumor angiogenesis in rat brains invivo by photoacoustic tomography. Appl Opt 2005,44:770–775.

7. Schneider BP, Miller KD. Angiogenesis of breastcancer. J Clin Oncol 2005, 23:1782–1790.

8. Vaupel P, Mayer A, Briest S, Hockel M. OxygenTransport to Tissue XXVI 2005; 566:33–342.

9. Agarwal A, Huang SW, O’Donnell M, Day KC,Day M, Kotov N, Ashkenazi S. Targeted goldnano-rod contrast agent for prostate cancer detectionby photoacoustic imaging. J Appl Phys 2007,102:064701–064704.

10. de la Zerda A, Zavaleta C, Keren S, Vaithilingam S,Bodapati S, Liu Z, Levi J, Smith BR, Ma T-J, OralkanO, et al. Carbon nanotubes as photoacoustic molecularimaging agents in living mice. Nat Nanotechnol 2008,3:557–562.

11. Kojima H. Development of near infrared fluorescentprobes for near infrared imaging. Yakugaku Zasshi2008, 128:1653.

12. Ntziachristos V, Bremer C, Weissleder R. Fluorescenceimaging with nearinfrared light: new technologicaladvances that enable in vivo molecular imaging. EurRadiol 2003, 13:195.

13. Rostro-Kohanloo BC, Bickford LR, Payne CM,Day ES, Anderson LJ, Zhong M, Lee S, MayerKM, Zal T, Adam L, et al. The stabilization andtargeting of surfactant-synthesized gold nanorods.Nanotechnology 2009, 20:434005.

14. Pan D, Pramanik M, Senpan A, Yang X, Scott MJ,Song KH, Gaffney PJ, Wickline SA, Wang LV, LanzaGM. Molecular photo acoustic imaging (PAI) withligand-directed gold nanobeacons. Angew Chem IntEd 2009, 48:4170–4173.



15. Skrabalak SE, Chen J, Au L, Lu X, Li X, Xia Y. Goldnanocages for biomedical applications. Adv Mater2007, 19:3177–3184.

16. Skrabalak SE, Chen J, Sun Y, Lu X, Au L, CobleyC, Xia Y. Gold nanocages: synthesis, properties, andapplications. Acc Chem Res 2008, 41:1587–1595.

17. Chen J, Yang M, Zhang Q, Cho EC, Cobley CM,Claus C, Kim C, Wang LV, Welch MJ, Xia Y.Gold nanocages: a novel class of multifunctionalnanomaterials for theranostic applications. Adv FunctMater 2010, 20:3684–3694.

18. Hoelen CGA, de Mul FFM, Pongers R, Dekker A.Three-dimensional photoacoustic imaging of bloodvessels in tissue. Opt Lett 1998, 23:648–650.

19. Kruger RA, Reinecke DR, Kruger GA. Thermoacousticcomputed tomography-technical considerations. MedPhys 1999, 26:1832–1837.

20. Xu M, Wang LV. Photoacoustic imaging inbiomedicine. Rev Sci Instrum 2006, 77:1–22.

21. Wang Y, Xie X, Wang X, Ku G, Gill KL, O’Neal DP,Stoica G, Wang LV. Photoacoustic tomography of ananoshell contrast agent in the in vivo rat brain. NanoLett 2004, 4:1689–1692.

22. Yang X, Skrabalak S, Li Z, Xia Y, Wang L.Photoacoustic tomography of a rat cerebral cortex invivo with Au nanocages as an optical contrast agent.Nano Lett 2007, 7:3798–3802.

23. Morton DL, Wen DR, Wong JH, Economou JS, CagleLA, Storm FK, Foshag LJ, Cochran AJ. Intraoperativelymphatic mapping and selective lymphadenectomy:technical details of a new procedure for clinical stageI melanoma. Arch Surg 1992, 127:392–399.

24. Krag DN, Weaver DL, Alex JC, Fairbank JT. Surgicalresection and radiolocalization of the sentinel lymphnode in breast cancer using a gamma probe. SurgOncol 1993, 2:335–340.

25. Koyama Y, Talanov VS, Bernardo M, Hama Y,Regino CA, Brechbiel MW, Choyke PL, KobayashiH. A dendrimer-based nanosized contrast agentdual-labeled for magnetic resonance and opticalfluorescence imaging to localize the sentinel lymphnode in mice. J Magn Reson Imag 2007, 25:866–871.

26. Nakajima M, Takeda M, Kobayashi M, Suzuki S,Ohuchi N. Nano-sized fluorescent particles as newtracers for sentinel node detection: experimental modelfor decision of appropriate size and wavelength.Cancer Sci 2005, 96:353–356.

27. Michel SC, Keller TM, Frohlich JM, Fink D,Caduff R, Seifert B, Marincek B, Kubik-Huch RA.Preoperative breast cancer staging: MR imaging of theaxilla with Ultrasmall superparamagnetic iron oxideenhancement. Radiology 2002, 225:527–536.

28. Torchia MG, Nason R, Danzinger R, Lewis JM,Thliveris JA. Interstitial MR lymphangiography forthe detection of sentinel lymph nodes. J Surg Oncol2001, 78:151–156.

29. Krag DN, Anderson SJ, Julian TB, Brown AM,Harlow SP, Costantino JP, Ashikaga T, Weaver DL,Mamounas EP, Jalovec LM, et al. Sentinel-lymph-noderesection compared with conventional axillary-lymph-node dissection in clinically node-negative patientswith breast cancer: overall survival findings from theNSABP B-32 randomised phase 3 trial. Lancet Oncol2010, 11:927–933.

30. Song KH, Kim C, Cobley CM, Xia Y, Wang LV. Near-infrared gold nanocages as a new class of tracers forphotoacoustic sentinel lymph node mapping on a ratmodel. Nano Lett 2009, 9:183–188.

31. Kim C, Cho EC, Chen J, Song KH, Au L, Favazza C,Zhang Q, Cobley CM, Gao F, Xia Y, et al. In vivomolecular photoacoustic tomography of melanomastargeted by bioconjugated gold nanocages. ACS Nano2010, 4:4559–4564.

32. Zharov VP, Galanzha EI, Shashkov EV. KimJ-W, Khlebtsov NG, Tuchin VV. Photoacousticflow cytometry: principle and application for real-time detection of circulating single nanoparticles,pathogens, and contrast dyes in vivo. J Biomed Opt2007, 12:051503.

33. de la Zerda A, Liu Z, Bodapati S, Teed R, VaithilingamS, Khuri-Yakub BT, Chen X, Dai H, GambhirSS. Ultrahigh sensitivity carbon nanotube agents forphotoacoustic molecular imaging in living mice. NanoLett 2010, 10:2168–2172.

34. Galanzha EI, Shashkov EV, Kelly T, Kim J-W, Yang L,Zharov VP. In vivo magnetic enrichment and multiplexphotoacoustic detection of circulating tumour cells.Nat Nanotechnol 2009, 4:855–860.

35. Kim J-W, Galanzha EI, Shashkov EV, Moon H-M,Zharov VP. Golden carbon nanotubes as multimodalphotoacoustic and photothermal high-contrast molec-ular agents. Nat Nanotechnol 2009, 4:688–694.

36. Galanzha EI, Shashkov EV, Kokoshka MS, MyhillJA, Zharov VP. In vivo noninvasive detection ofmetastatic cells in vasculature and sentinel lymphnodes by photoacoustic cytomery. Laser Surg Med2008, S20:81–82.

37. Galanzha EI, Kokoska MS, Shashkov EV, Kim J-W,Tuchin VV, Zharov VP. In vivo fiber-based multicolorphotoacoustic detection and photothermal purgingof metastasis in sentinel lymph nodes targeted bynanoparticles. J Biophotonics 2009, 2:528–539.

38. Pramanik M, Song KH, Swierczewska M, Green D,Sitharaman B, Wang LV. In vivo carbon nanotube-enhanced non-invasive photoacoustic mapping ofthe sentinel lymph node. Phys Med Biol 2009,54:3291–3301.

39. Pramanik M, Swierczewska M, Green D, SitharamanB, Wang LV. Single-walled carbon nanotubes asa multimodal-thermoacoustic and photoacoustic-contrast agent. J Biomed Opt 2009, 14:034018.

40. Galanzha EI, Kim J-W, Zharov VP. Integrated nan-otechnology based photoacoustic and photothermal



flow cytometry platform for in vivo detection andkilling of circulating cancer stem cells. J Biophoton2009, 2:725–735.

41. Zharov VP, Galitovsky V, Viegas M. Photother-mal detection of local thermal effects during selec-tive nanophotothermolysis. Appl Phys Lett 2003,83:4897–4899.

42. Zharov VP, Galitovskaya EN, Jonson C, Kelly T.Synergistic enhancement of selective nanophotother-molysis with gold nanoclusters: potential for cancertherapy. Laser Surg Med 2005, 37:219–226.

43. Zharov VP, Letfullin RR, Galitovskaya EN.Microbubbles-overlapping mode for laser killing ofcancer cells with absorbing nanoparticle clusters.J Phys D: Appl Phys 2005, 38:257–258.

44. Zharov VP, Kim J-W, Everts M, Curiel DT. Self-assembling nanoclusters in living system applicationfor integrated photothermal nanodiagnostics andnanotherapy. Nanomedicine 2005, 1:326–345.

45. Zharov VP, Mercer KE, Galitovskaya EN, SmeltzerMS. Photothermal nanotherapeutics and nanodiag-nostics for selective killing of bacteria targeted withgold nanoparticles. Biophys J 2006, 90:619–628.

46. Kim J-W, Galanzha EI, Shashkov EV, Kotagiri N,Zharov VP. Photothermal antimicrobial nanotherapyand nanodiagnostics with self-assembling carbon nan-otube clusters. Lasers Surg Med 2007, 39:622–634.

47. Khlebtsov B, Zharov V, Melnikov A, Tuchin V,Khlebtsov N. Optical amplification of photothermaltherapy with gold nanoparticles and nanoclusters.Nanotechnology 2006, 17:5167–5179.

48. Hirsch LR, Stafford RJ, Bankson JA, Sershen SR,Rivera B, Price RE, Hazle JD, Halas NJ, West JL.Nanoshell-mediated thermal therapy of tumors undermagnetic resonance guidance. Proc Natl Acad Sci USA2003, 100:13549–13554.

49. Kam NW, O’Connell M, Wisdom JA, Dai H.Carbon nanotubes as multifunctional biologicaltransporters and near-infrared agents for selectivecancer cell destruction. Proc Natl Acad Sci USA 2005,102:11600–11605.

50. Liu Z, Sun X, Nakayama-Ratchford N, Dai H.Chemistry on watersoluble carbon nanotubes for drugloading and delivery. ACS Nano 2007, 1:50–56.

51. Liu Z, Cai W, He L, Nakayama N, Chen K, SunX, Chen X, Dai H. In vivo biodistribution and highlyefficient tumour targeting of carbon nanotubes in mice.Nat Nanotechnol 2007, 2:47–52.

52. Pan D, Pramanik M, Zhang H, Senpan A,Schmieder A, Wickline SA, Wang LV, LanzaGM. Photoacoustics tomography differentiates earlyvulnerable neoangiogenesis from mature vasculaturein vivo. JACC 2011, 57:E1595.

53. Pan D, Pramanik M, Senpan A, Ghosh S,Wickline SA, Wang LV, Lanza GM. Near infrared

photoacoustic detection of sentinel lymphnodes withgold nanobeacons. Biomaterials 2010, 31:4088–4093.

54. Pan D, Pramanik M, Senpan A, Wickline SA,Wang LV, Lanza GM. A facile synthesis of novelself-assembled gold nanorods designed for near-infrared imaging. J Nanosci Nanotech 2010,10:8118–8123.

55. Pan D, Pramanik M, Senpan A, Stacy A, Zhang H,Wickline SA, Wang LV, Lanza GM. Molecular pho-toacoustic imaging of angiogenesis with integrin tar-geted gold nanobeacons. FASEB J 2011, 25:378–388.

56. Raut S, Gaffney PJ. Evaluation of the fibrin bindingprofile of two anti-fibrin monoclonal antibodies.Throm Haemostasos 1996, 76:56–64.

57. Gamelin J, Aguirre A, Maurudis A, Huang F, CastilloD, Wang LV, Zhu Q. Curved array photoacoustictomographic system for small animal imaging.J Biomed Opt 2008, 13:024007.

58. Pramanik M, Ku G, Li CH, Wang LHV. Designand evaluation of a novel breast cancer detectionsystem combining both thermoacoustic (TA) andphotoacoustic (PA) tomography. Med Phys 2008,35:2218–2223.

59. Krag D, Weaver D, Ashikaga T, Moffat F, KlimbergVS, Shriver C, Feldman S, Kusminsky R, Gadd M,Kuhn J, et al. The sentinel node in breast cancer—amulticenter validation study. N Engl J Med 1998,339:941–946.

60. McMasters KM, Tuttle TM, Carlson DJ, Brown CM,Noyes RD, Glaser RL, Vennekotter DJ, Turk PS,Tate PS, Sardi A, et al. Sentinel lymph node biopsyfor breast cancer: a suitable alternative to routineaxillary dissection in multi-institutional practice whenoptimal technique is used. J Clin Oncol 2000, 18:2560–2566.

61. Chen SL, Iddings DM, Scheri RP, Bilchik AJ.Lymphatic mapping and sentinel node analysis. CACancer J Clin 2006, 56:292–309.

62. Laruelle G, Francios J, Billon L. Self-assembly inaqueous media of amphiphilic poly acrylic acid baseddi-block copolymers synthesized by direct nitroxide-mediated polymerization. Macromol Rapid Commun2004, 25:1839–1844.

63. Huang HY, Kowalewski T, Remsen EE, GertzmannR, Wooley KL. Hydrogel-coated glassy nanospheres:A novel method for the synthesis of shell cross-linkedknedels. J Am Chem Soc 1997, 119:11653–11659.

64. Wu J, Eisenberg A. Proton diffusion across membranesof vesicles of poly(styrene-b-acrylic acid) diblockcopolymers. J Am Chem Soc 2006, 128:2880–2884.

65. Xu J, Sun G, Rossin R, Hagooly A, Li Z, FukukawaK, Messmore BW, Moore DA, Welch MJ, HawkerCJ, et al. Labeling of polymer nanostructures formedical imaging: importance of charge density, spacerlength, and crosslinking extents. Macromolecules2007, 40:2971–2973.



66. Zhang HF, Maslov K, Stoica G, Wang LHV.Functional photoacoustic microscopy for high-resolution and noninvasive in vivo imaging. NatBiotechnol 2006, 24:848–851.

67. Brooks PC, Clark RAF, Cheresh DA. Requirement ofvascular integrin avb3 for angiogenesis. Science 1994,264:569–571.

68. Winter P, Neubauer A, Caruthers S, Harris T,Robertson J, Williams T, Schmieder A, Hu G, AllenJ, Lacy E, et al. Endothelial alpha(nu)beta(3)-integrintargeted fumagillin nanoparticles inhibit angiogenesisin atherosclerosis. Arterioscler Thromb Vasc Biol2006, 26:2103–2109.

69. Winter PM, Schmieder AH, Caruthers SD, Keene JL,Zhang H, Wickline SA, Lanza GM. Minute dosages ofα(ν)β3-targeted fumagillin nanoparticles impair Vx-2 tumor angiogenesis and development in rabbits.FASEB J 2008, 22:2758–2767.

70. Siphanto RI, Thumma KK, Kolkman RG, vanLeeuwen TG, de Mul FF, van Neck JW, van AdrichemLN, Steenbergen W. Serial noninvasive photoacousticimaging of neovascularization in tumor angiogenesis.Opt Express 2005, 13:89–95.

71. Maslov K, Zhang HF, Hu S, Wang LV. Optical-resolution photoacoustic microscopy for in vivoimaging of single capillaries. Optics Lett 2008,33:929–931.

72. Lao Y, Xing D, Yang S, Xiang L. Noninvasivephotoacoustic imaging of the developing vasculatureduring early tumor growth. Phys Med Biol 2008,53:4203–4212.

73. Zhang EZ, Laufer JG, Pedley RB, Beard PC. Invivo high-resolution 3D photoacoustic imaging ofsuperficial vascular anatomy. Phys Med Biol 2009,54:1035–1046.

74. Jose J, Manohar S, Kolkman RG, Steenbergen W,van Leeuwen TG. Imaging of tumor vasculature usingTwente photoacoustic systems. J Biophotonics 2009,2:701–717.

75. Hu S, Oladipupo S, Yao J, Santeford AC, Maslov K,Kovalski J, Arbeit JM, Wang LV. Opticalresolutionphotoacoustic microscopy of angiogenesis in a trans-genic mouse model. Proc SPIE 2010, 7564:756406.

76. Cheresh DA. Integrins in thrombosis, wound healingand cancer. Biochem Soc Trans 1991, 19:835–838.

77. Friedlander M, Theesfeld CL, Sugita M, Fruttiger M,Thomas MA, Chang S, Cheresh DA. Involvementof integrins anb3 and anb5 in ocular neovasculardiseases. Proc Natl Acad Sci USA 1996, 93:9764–9769.

78. De Nichilo M, Burns G. Granulocytemacrophage andmacrophage colonystimulating factors differentiallyregulate a v integrin expression on cultured humanmacrophages. Proc Natl Acad Sci USA 1993,90:2517–2521.

79. Helluin O, Chan C, Vilaire G, Mousa S, DeGradoWF, Bennett JS. The activation state of avb3 regulatesplatelet and lymphocyte adhesion to intact andthrombin-cleaved osteopontin. J Biol Chem 2000,275:18337–18343.

80. Itoh H, Nelson P, Mureebe L, Horowitz A, KentK. The role of integrins in saphenous vein vascularsmooth muscle cell migration. J Vasc Surg 1997,25:1061–1069.

81. Carreiras F, Denoux Y, Staedel C, Lehman M, SichelF, Gauduchon P. Expression and localization of a vintegrins and their ligand vitronectin in normal ovarianepithelium and in ovarian carcinoma. Gynecol Oncol1996, 62:260–267.

82. Kageshita T, Hamby CV, Hirai S, Kimura T, Ono T,Ferrone S. Differential clinical significance of a(v)B(3)expression in primary lesions of acral lentiginousmelanoma and of other melanoma histotypes. Int JCancer 2000, 89:153–159.

83. Hu G, Lijowski M, Zhang H, Partlow KC, CaruthersSD, Kiefer G, Gulyas G, Athey P, Scott MJ, WicklineSA, et al. Imaging of Vx-2 rabbit tumors with avb3-integrintargeted 111In nanoparticles. Int J Cancer2007, 120:1951–1957.

84. Wang LV, Xie X, Oh JT, Li MN, Ku G, Ke S,Similache S, Li C, Stoica G. Combined photoacousticand molecular fluorescence imaging in vivo. IEEE EngMed Biol Proc 2005, 7:190–192.

85. Xie X, Oh JT, Li ML, Ku G, Wang X, Ke S, LiC, Similache S, Stoica G, Wang LV. Photoacoustictomography and molecular fluorescence imaging: Dualmodality imaging of small animal brains in vivo. ProcSPIE 2005, 5697:107–110.

86. Welsher K, Liu Z, Sherlock SP, Robinson JT, Chen Z,Daranciang D, Dai H. A route to brightly fluorescentcarbon nanotubes for near-infrared imaging in mice.Nat Nanotechnol 2009, 4:773–780.

87. Xie X, Li ML, Oh JT, Ku G, Wang W, Li C, SimilacheS, Lungu GF, Stoica G, Wang LV. Photoacousticmolecular imaging of small animals in vivo. Proc SPIE2006, 6086:608606.

88. Xiang L, Yuan Y, Xing D, Ou Z, Yang S, ZhouF. Photoacoustic molecular imaging with antibody-functionalized single-walled carbon nanotubes forearly diagnosis of tumor. J Biomed Opt 2009,14:021008.

89. Weissleder R, Bogdanov A Jr, Tung CH, WeinmannHJ. Size optimization of synthetic graft copolymers forin vivo angiogenesis imaging. Bioconjug Chem 2001,12:213–219.

90. Harris TD, Rajopadhye M. Vitronectin receptorantagonist pharmaceuticals. US patent 6,511,648

91. Meoli DF, Sadeghi MM, Krassilnikova S, Bourke BN,Giordano FJ, Dione DP, Su H, Edwards DS, Liu S,Harris TD, et al. Noninvasive imaging of myocardial



angiogenesis following experimental myocardialinfarction. J Clin Invest 2004, 113:1684–1691.

92. Miyauchi A, Alvarez J, Greenfield EM, Teti A,Grano M, Colucci S, Zambonin-Zallone A, RossFP, Teitelbaum SL, Cheresh D, et al. Recognitionof osteopontin and related peptides by an alpha vbeta 3 integrin stimulates immediate cell signals inosteoclasts. J Biol Chem 1991, 266:20369–20374.

93. Cady B, Stone MD, Schuler JG. The new era inbreast cancer. Invasion, size, and nodal involvementdramatically decreasing as a result of mammographicscreening. Arch Surg 1996, 131:301–308.

94. Haffty BG, Ward B, Pathare P. Reappraisal of the roleof axillary lymph node dissection in the conservativetreatment of breast cancer. J Clin Oncol 1997,15:691–700.

95. Veronesi U, Paganelli G, Viale G. A randomizedcomparison of sentinel-node biopsy with routineaxillary dissection in breast cancer. N Engl J Med2003, 349:546–553.

96. Pan D, Cai X, Kim B, Stacy AL, Wang LV, LanzaGM. Rapid synthesis of near infrared polymericmicelles for real-time sentinel lymph node imaging.Adv Healthcare Mater 2012, 1:582–589.

97. Pan D, Cai X, Yalaz C, Senpan A, Omanakuttan K,Wickline SA, Wang LV, Lanza GM. Photoacousticsentinel lymph node imaging with self-assembledcopper neodecanoate nanoparticles. ACS Nano 2012,6:1260–1267.

98. Chan WC, Nie S. Quantum dot bioconjugates forultrasensitive nonisotopic detection. Science 1998,281:2016–2018.

99. Storhoff JJ, Lucas AD, Garimella V, Bao YP, MullerUR. Homogeneous detection of unamplified genomicDNA sequences based on colorimetric scatter ofgold nanoparticle probes. Nat Biotechnol 2004,22:883–887.

100. Oraevsky AA. Laser optoacoustic tomography formedical diagnostics: principles. Proc SPIE 1996,2676:22–31.

101. Lal S, Clare SE, Halas NJ. Nanoshell-enabledphotothermal cancer thereapy: impending clinicalimpact. Acc Chem Res 2008, 41:1842–1851.

102. Ghosh P, Han G, De M, Kim CK, Rotello VM. Goldnanoparticles in delivery applications. Adv Drug DelivRev 2008, 60:1307–1315.

103. Lewinski N, Colvin V, Drezek R. Cytotoxicity ofnanoparticles. Small 2008, 4:26–49.

104. Nel A, Xia T, Madler L, Li N. Toxic potentialof materials at the nanolevel. Science 2006,311:622–627.

105. Peer D, Karp JM, Hong S, Farokhzad OC, MargalitR, Langer R. Nanocarriers as an emerging platformfor cancer therapy. Nat Nanotechnol 2007, 2:751–760.

106. Lovell JF, Jin CS, Huynh E, Jin H, Kim C,Rubinstein JL, Chan WC, Cao W, Wang LV, ZhengG. Porphysome nanovesicles generated by porphyrinbilayers for use as multimodal biophotonic contrastagents. Nat Mater 2011, 10:324–332.

107. Van de Sompel D, Sasportas LS, Dragulescu-AndrasiA, Bohndiek S, Gambhir SS. Improving image qualityby accounting for changes in water temperature duringa photoacoustic tomography scan. PLoS One 2012,7:e45337.

108. Song KH, Stein EW, Margenthaler JA, Wang LV.Noninvasive photoacoustic identification of sentinellymph nodes containing methylene blue in vivo in arat model. J Biomed Opt 2008, 13:054033.

109. Sohrabnezhad S, Pourahmad A, Sadjadi MA. Newmethylene blue incorporated in mordenite zeoliteas humidity sensor material. Mater Lett 2007,61:2311–2314.

110. Brewer GJ. Copper toxicity in the general population.Clin Neurophysiol 2010, 121:459–460.

111. Yang S-T, Luo J, Zhou Q, Wang H. Pharmacokinetics,metabolism and toxicity of carbon nanotubesfor biomedical purposes. Theranostics 2012, 2:271–282.


Documents

Overview A brief account of nanoparticle contrast agents ...coilab.caltech.edu/epub/2013/PanD_2013_WIREs... · Overview A brief account of nanoparticle contrast agents for photoacoustic