Multiplexed in vivo fluorescence optical imaging of the therapeutic efficacy of photodynamic therapy

Biomaterials 34 (2013) 10075e10083

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Multiplexed in vivo fluorescence optical imaging of the therapeutic efficacy of photodynamic therapyq Katja Haedicke a, *, Susanna Gräfe b, Frank Lehmann c, Ingrid Hilger a, ** a Department of Experimental Radiology, Institute of Diagnostic and Interventional Radiology I, Jena University Hospital, Friedrich-Schiller University Jena, Erlanger Allee 101, Jena D-07747, Germany b Biolitec Research GmbH, Research & Development, Jena D-07745, Germany c Dyomics GmbH, Jena D-07745, Germany

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 June 2013 Accepted 27 August 2013 Available online 17 September 2013

In our study we wanted to elucidate a time frame for in vivo optical imaging of the therapeutic efficacy of photodynamic therapy (PDT) by using a multiplexed imaging approach for detecting apoptosis and vascularization. The internalization of the photosensitizer FoslipÒ into tongue-squamous epithelium carcinoma cells (CAL-27) was examined in vitro and in vivo. For detecting apoptosis, annexin V was covalently coupled to the near-infrared dye DY-734 and the spectroscopic properties and binding affinity to apoptotic CAL-27 cells were elucidated. CAL-27 tumor bearing mice were treated with PDT and injected 2 days and 2 weeks thereafter with DY-734-annexin V. PDT-induced changes in tumor vascularization were detected with the contrast agent IRDyeÒ 800CW RGD up to 3 weeks after PDT. A perinuclear enrichment of FoslipÒ could be seen in vitro which was reflected in an accumulation in CAL-27 tumors in vivo. The DY-734-annexin V (coupling efficiency 30e50%) revealed a high binding affinity to apoptotic compared to non-apoptotic cells (17.2% vs. 1.2%) with a KD-value of 20 nM. After PDT-treatment, the probe showed a significantly higher (p <0.05) contrast in tumors at 2 days compared to 2 weeks after therapy (2e8 h post injection). A reduction of the vascularization could be detected after PDT especially in the central tumor areas. To detect the therapeutic efficacy of PDT, a multiplexed imaging approach is necessary. A detection of apoptotic cells is possible just shortly after therapy, whereas at later time points the efficacy can be verified by investigating the vascularization. Ó 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

Keywords: Multiplexed imaging Apoptosis Annexin V Tumor vascularization Fluorescence optical imaging

1. Introduction Non-invasive optical imaging of molecular processes in vivo is a promising technology for the therapeutic monitoring in personalized medicine [1]. Among the different oncologic therapeutic approaches available so far, like chemotherapy or radiation, photodynamic therapy (PDT) of tumors uses light along with the administration of a photosensitizing agent as a therapeutic tool [2]. In effect, a direct cell death occurs as a result of the production of reactive oxygen species. Besides the induction of apoptosis in tumor cells, the vasculature is damaged, which leads to an

q This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited. * Corresponding author. Tel.: þ49 3641 9324896; fax: þ49 3641 9325922. ** Corresponding author. Tel.: þ49 3641 9325921; fax: þ49 3641 9325922. E-mail addresses: [email protected] (K. Haedicke), Ingrid.Hilger@ med.uni-jena.de (I. Hilger).

undersupply of the tumor with nutrients and ultimately determines the therapeutic success [3,4]. A central process in the effective elimination of tumor cells is apoptosis [5]. Cells externalize phosphatidylserine on their outer membrane leaflet in the early phase of apoptosis [6]. This process makes the phospholipid accessible to targeted probes where it serves as target for detecting apoptotic cells. Typical examples of phosphatidylserine targeted probes are annexin V based probes [7,8]. In recent studies the use of different modified 99mTc-labeled annexin V for single photon emission computed tomography (SPECT) or autoradiography have been reported [9,10]. Also targeting of apoptotic cells with biotinylated annexin V followed by administration of 64Cu-labeled streptavidin allowed apoptotic tissues to be imaged by positron emission tomography (PET) [11]. Although the utilization of near-infrared fluorescent (NIRF) probes for the in vivo optical imaging of apoptosis is also very suitable and non-invasive, only few data demonstrated the use of NIRF probes in imaging apoptosis. After conjugation of annexin V to Cy5.5, such probes were shown to image the effect of cyclophosphamide chemotherapy on Lewis lung carcinoma at 24 h post treatment

0142-9612/$ e see front matter Ó 2013 The Authors. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biomaterials.2013.08.087

10076

K. Haedicke et al. / Biomaterials 34 (2013) 10075e10083

† (n=7)

IRDye® 800CW RGD

DY-734annexin V

IRDye® 800CW RGD

IRDye® 800CW DY-734RGD annexin V

IRDye® 800CW RGD

1week

2weeks

3weeks

IRDye® 800CW RGD

IRDye® DY-734800CW annexin V RGD

IRDye® 800CW RGD

Foslip® + PDT start

IRDye® 800CW RGD

DY-734annexin V

† (n=3)

dissection

non-PDT-treated group

PDT-treated group

† (n=3)

† (n=7)

Fig. 1. Schematic presentation of the in vivo experimental design for imaging tumor vascularization and apoptotic cells after PDT. The time points demonstrate injection of the respective probes in PDT-treated and non-PDT-treated mice. y: animals were sacrificed.

[12,13]. In contrast, a failure of annexin-based in vivo apoptosis imaging was demonstrated at day 4 of an anti-angiogenic therapy because of a significant breakdown of the vasculature [14]. These reports reveal that the use of multiplexed methods which enable simultaneous detection of apoptosis and vascularization are inevitable in the reliable monitoring of therapeutic efficacy. Apoptosis is important for detecting cell death as a consequence of the therapeutic efficacy of PDT and vascularization also verifies probe accessibility. The advantage of fluorescence optical imaging over PET is that it allows multiplexed approaches by means of a simultaneous detection of several fluorescent probes which differ in their emission maxima. In this study, we therefore sought to elucidate the appropriate time frame for in vivo imaging of the therapeutic efficacy of PDT via the use of a multiplexed imaging approach which addresses both the phosphatidylserine molecules on cells at the tumor region as well as the expression of avb3 integrin on the surface of proliferating endothelial cells as a marker for vascularization. We first examined the internalization of the photosensitizer formulation FoslipÒ into the tumor cells to corroborate a basic prerequisite for a successful PDT. Then, the binding affinity of a self-designed NIRF annexin V probe to the cells was evaluated in vitro. We subsequently used the probe for detection of apoptosis in mice xenografts together with an integrin-targeted probe to assess the level of vascularization at defined time points after PDT. 2. Materials and methods 2.1. Cells and animals Tongue-squamous epithelium carcinoma cells (CAL-27, DSMZ, Braunschweig, Germany) were cultured in DMEM-GlutaMAXÔ (GibcoÒ, Darmstadt, Germany) containing 10% FBS, at 37  C in a humidified 5% CO2 atmosphere. In vivo experiments were carried out with female athymic nude-Foxn1nu mice (20e25 g, Harlan Laboratories GmbH, Venray, The Netherlands). During experimentation, the animals were anesthetized with isofluorane (Actavis, Munich, Germany). All procedures were approved by the regional animal committee and were in accordance with international guidelines on the ethical use of animals. Mice were housed under standard conditions with food and water ad libitum. 2.2. In vitro characterization and biodistribution of the photosensitizer formulation FoslipÒ The photosensitizer formulation FoslipÒ (liposomal formulation of meta tetrahydroxyphenyl chlorine e mTHPC) was provided by biolitec research GmbH (Jena,

Germany) as a lyophilized powder that was reconstituted with sterile water to 1.5 mg mTHPC ml1. To verify the spectroscopic features of FoslipÒ, both the absorption and the emission spectrum were determined via spectrophotometry and spectrofluorometry. To investigate the cellular uptake of FoslipÒ, CAL-27 cells were incubated with 50 mM of the photosensitizer for 24 h. Nuclei were counterstained with 0.2 mg/ml Hoechst-33258 (Applichem, Darmstadt, Germany). Cells were analyzed using a confocal laser scanning microscope (LSM 510 Meta, Zeiss, Jena, Germany). The effect of laser treatment alone was investigated by illuminating CAL-27 cells in vitro with different light dosages up to 100 J/cm2 and measuring caspase 3/7 activity (important enzyme of apoptosis) as well as cell viability by XTT assay at 24 h after illumination. To study the biodistribution of FoslipÒ, 2  106 CAL-27 cells in MatrigelÔ (BD, Heidelberg, Germany) were injected subcutaneously into 8 mice. When the tumors reached a diameter of approximately 5 mm, mice were injected with FoslipÒ (50 mg/ kg, n ¼ 5). 3 mice (control group) were left without any photosensitizer to monitor autofluorescence of the organs. 24 h post injection, the animals were sacrificed and the fluorescence intensity of the tumors and organs was measured ex vivo using the MaestroÔ in vivo fluorescence imaging system (excitation filter: 570e610 nm, emission filter: 645 nm longpass, CRi Inc., Woburn, MA, USA). The generated composite spectral data were unmixed into the spectrum of the autofluorescence of the mice and the spectrum of FoslipÒ. For semiquantitative analysis, regions of interest (ROIs) were drawn over the spectrally unmixed image of tumor and organs. For every ROI an averaged fluorescence signal was determined, which represents the area under the curve of the unmixed FoslipÒ spectrum. Average counts were normalized to exposure time and pixel number. 2.3. Labeling of annexin V with the NIRF dye DY-734 To detect apoptotic cells after PDT via fluorescence optical imaging, annexin V (Abcam, Cambridge, UK) was covalently coupled to the near-infrared dye DY-734 (using a NHS-ester derivative, abs: 736 nm, em: 759 nm, Dyomics GmbH, Jena, Germany): 0.1 mg annexin V was mixed with DY-734-NHS in coupling buffer (100 mM NaHCO3, 500 mM NaCl, pH 8) and incubated for 2 h (22  C and 900 rpm). Unbound dye molecules were separated by ultrafiltration on a 10 kDa amicon column (Merck Millipore, Billerica, MA, USA). After washing 3 times with PBS, the fluorescent annexin V probe (DY-734-annexin V) was eluted and the success of coupling as well as the annexin V to DY-734 ratio were determined spectroscopically. Hereto, the absorption maxima of both the protein and the dye were used and the ratio was determined via the LamberteBeer law. 2.4. Determination of the binding affinity of DY-734-annexin V to apoptotic cells To induce apoptosis, CAL-27 cells were incubated with etoposide (30 mM, Sigmae Aldrich, Munich, Germany) for 24 h. Afterward, the cells were split into 4 different groups and stained as follows: 1) native control (without staining), 2) apoptosis (staining with 10 mg/ml DY-734-annexin V), 3) necrosis (staining with propidium iodide (PI), Annexin-V-FLUOS Staining Kit, Roche, Grenzach, Germany), and 4) apoptosis and necrosis (staining with DY-734-annexin V and PI). The same procedure was applied to cells without etoposide to control the autofluorescence and the

K. Haedicke et al. / Biomaterials 34 (2013) 10075e10083

1000

*

0.000

er

600 800 wavelength [nm]

0.001

liv

B

400

*

0.002

dn ey

0.0 200

native

ki

0.5

*

*

gu t

1.0

0.003

Foslip®

lu ng

1.5

0.004

sp le en

2.0

C

he ar t

absorption emission

tu m or

2.5

Average Signal (counts/s)

absorption/emission [a.u.]

A

10077

Fig. 2. Characterization of FoslipÒ and cellular uptake of FoslipÒ by CAL-27 cells. (A) Absorption and emission spectrum of FoslipÒ in sterile water. (B) Uptake of FoslipÒ into CAL-27 cells. Cells with (left) and without (right) FoslipÒ incubation. Blue: cell nuclei stained with Hoechst. Red: FoslipÒ. Bar: 50 mm. (C) Biodistribution of FoslipÒ in CAL-27 tumor bearing mice 24 h after intravenous administration (black bars, n ¼ 5), in comparison to mice of the native group without FoslipÒ administration (autofluorescence of the organs, white bars, n ¼ 3). (*p <0.05). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

unspecific binding to non-apoptotic cells. Flow cytometry was carried out using a FC500 flow cytometer (Beckman Coulter, Krefeld, Germany). The differently treated cells were also examined under an Olympus BX 50 microscope (Olympus Deutschland GmbH, Hamburg, Germany) to see changes within the morphology. To elucidate cellular localization of DY-734-annexin V, etoposide (10 mM) induced apoptotic CAL-27 cells were incubated with the optical probe (1 mM) for 10 min. Afterward, the cell membrane was stained using wheat germ agglutinin (WGA) Alexa 555 (10 mg/ml, Invitrogen, Darmstadt, Germany). Nuclei were counterstained with 0.2 mg/ml Hoechst 33258. Stained cells were imaged using an EVOS fluorescence microscope (AMG, Bothell, WA, USA). To quantitatively characterize the binding affinity of the optical probe, etoposide (30 mM) induced apoptotic CAL-27 cells were incubated with increasing concentrations of DY-734-annexin V (0 up to 60 nM) for 10 min. Thereafter, the fluorescence intensity of the pellets was measured with the in vivo fluorescence imaging system. The dissociation constant KD was calculated indirectly with respect to the fluorescence intensity with SigmaPlot 12.0. The specificity of the binding of DY-734-annexin V was investigated in vitro by incubating etoposide induced apoptotic CAL-27 cells with a 10-fold excess (50 nM) or a 100-fold excess (500 nM) of unlabeled annexin V to block binding sites before adding 5 nM of DY-734-annexin V. The fluorescence intensity of the cell pellets was measured with the in vivo fluorescence imaging system.

with the MaestroÔ system as described above (excitation filter: 670e710 nm, emission filter: 750 nm). 2.7. In vivo imaging of tumor vascularization after PDT The effect of PDT on tumor vascularization was analyzed with the contrast agent IRDyeÒ 800CW RGD Optical Probe (abs: 776 nm, em: 792 nm, LI-CORÒ Biosciences, Lincoln, NE, USA). This near-infrared imaging agent specifically targets avb3 integrins on the surface of endothelial cells in the tumor. The contrast agent was injected into the PDT-treated animals (n ¼ 10, 40 nmol/kg) directly before PDT, as well as at 1, 2, and 3 weeks after PDT to image changes of the tumor vascularization over a longer time period after treatment and to assess its influence on imaging of apoptosis as well as probe accessibility (Fig. 1). Contrast agent accumulation was compared to the non-PDT-treated group (n ¼ 10). Imaging was carried out by acquisition of nearinfrared fluorescence signals at 24 h after probe injection with the MaestroÔ system as described above (excitation filter: 670e710 nm, emission filter: 750 nm). The relative fluorescence intensity was related to the fluorescence of the native mice (before PDT). Macroscopic changes were depicted by light images of the mice and tumor volumes were measured using a digital caliper and the equation V ¼ p/ 6  (length  width  height). The relative tumor volume after therapy was related to the tumor volume of the native mice (before PDT).

2.5. PDT of tumors

2.8. Biodistribution of DY-734-annexin V and IRDyeÒ 800CW RGD

For in vivo PDT studies, tumors were implanted in 20 mice subcutaneously as described above. The animals were divided into 2 groups. The PDT-treated group (n ¼ 10) was injected with FoslipÒ (40 mg/kg weight) and 24 h later the tumors were illuminated with a 2 W 652 nm Ceralas PDT laser (Biolitec AG, Jena, Germany). The power density at the tumor was 0.48 W/cm2 and the total light dose 40 J/cm2. The non-PDT-treated control group (n ¼ 10) was left without photosensitizer and without laser illumination.

Biodistribution of DY-734-annexin V was compared in the PDT-treated and nonPDT-treated groups (n ¼ 3 animals/group) 5 days after PDT (3 days after injection, Fig. 1). All other mice were sacrificed 3 weeks after PDT (17 (n ¼ 4) and 8 (n ¼ 10) days after DY-734-annexin V injection, Fig. 1). Cell nuclei of cryo-frozen tumor slices were stained with 2 mg/ml Hoechst-33258 and imaged using an EVOS fluorescence microscope. The biodistribution of IRDyeÒ 800CW RGD Optical Probe was analyzed 3 weeks after PDT and 24 h after last IRDyeÒ 800CW RGD injection (n ¼ 14). The fluorescence intensity of the tumors and the organs were measured ex vivo as described above.

2.6. In vivo imaging of apoptotic cells after PDT To detect apoptotic cells in the tumors, DY-734-annexin V was injected into the PDT-treated animal group 2 days (short term therapy monitoring, n ¼ 5) and 2 weeks (long term therapy monitoring, n ¼ 5) after therapy (4 nmol/kg weight, Fig. 1). To control for unspecific enrichment, the probe was injected into the animals of the non-PDT-treated group (n ¼ 10). The near-infrared fluorescence intensity of the mice was monitored before probe injection (1 h), and 0, 2e48 h after application

2.9. Statistical analysis Data are expressed as means  standard error. Statistical significance was analyzed by a two-tailed Student’s t test using SigmaPlot 12.0. If the equal variance test failed, a ManneWhitney Rank Sum test was run. A p-value of 0.05 or less was considered statistically significant.

10078

1.0

0.5

0.6 0.4 0.2

102 101 100

101 100

103

100 101 102 propidium iodide

DY-734-annexin V

101 100

100 101 102 propidium iodide

103

102 101 100

103

100 101 102 propidium iodide

102 101 100

100 101 102 propidium iodide

101 100

100 101 102 propidium iodide

E

95 0

90 0

101 100

100 101 102 propidium iodide

103

100 101 102 propidium iodide

103

103

102

103

+ DY-734-annexin V + PI

102

103

103

Average Signal (counts/s)

103

+ PI

103

103

102

85 0

wavelength [nm]

+ DY-734-annexin V

102

80 0

1000

DY-734-annexin V

103

103

+ etoposide

0.8

DY-734-annexin V

native

100 101 102 propidium iodide

DY-734-annexin V

DY-734-annexin V DY-734

70 0

600 800 wavelength [nm]

DY-734-annexin V

DY-734-annexin V

103

1.0

0.0 400

DY-734-annexin V

C - etoposide

emission [a.u.]

DY-734

0.0 200

D

B

DY-734-annexin V

75 0

1.5

DY-734-annexin V

absorption [a.u.]

A

K. Haedicke et al. / Biomaterials 34 (2013) 10075e10083

103

102 101 100

0.012 0.010 0.008 0.006 0.004 KD = 20 nM

0.002 0.000 0

10

40 50 20 30 DY-734-annexin V [nM]

60

Fig. 3. Spectroscopic analysis of DY-734-annexin V and its binding affinity to apoptotic cells. (A) Absorption spectrum of DY-734-annexin V and the free dye DY-734. (B) Emission spectrum of DY-734-annexin V and the free dye DY-734. Small inserts show fluorescence of the probes in small tubes. (C) Binding of DY-734-annexin V to apoptotic CAL-27 cells (flow cytometry). Unstained CAL-27 cells (native) show no autofluorescence. Upper row: cells without etoposide and thus without induced apoptotic cell death. Lower row: cells treated with etoposide to induce apoptosis. PI ¼ propidium iodide. (D) Cellular localization of DY-734-annexin V in apoptotic CAL-27 cells. Blue: cell nuclei (Hoechst). Green: cell membrane stained with WGA Alexa 555. Red: DY-734-annexin V. Bar: 100 mm. (E) Saturation curve and dissociation constant (KD) of DY-734-annexin V (n ¼ 3). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

K. Haedicke et al. / Biomaterials 34 (2013) 10075e10083

2 days after PDT 2 weeks after PDT max

min max

min max

B contrast (tumor-muscle)

A

*

0.002

min max

min max

min max

Average Signal (counts/s)

max

0.001

*

*

*

E

*

0.000 -1h

0h

2h

4h

6h

8h

24h 48h

PDT (I) non-PDT (II)

0.003 0.002

*

0.001 0.000

spleen kidney liver

D Average Signal (counts/s)

min

PDT 2 d non-PDT 2 d PDT 2 w non-PDT 2 w

* *

C min max

10079

0.003 0.002

PDT 17d p.i. (I) non-PDT 17d p.i. (II) PDT 8d p.i. (III) non-PDT 8d p.i. (IV)

*

0.001

*

0.000

min

spleen kidney liver

Fig. 4. In vivo detection of PDT-induced apoptotic cells in tumors 2 days after therapy and biodistribution of DY-734-annexin V. (A) Representative composite images of CAL-27 tumor bearing mice injected with DY-734-annexin V 2 days (left column) or 2 weeks (right column) after PDT. (B) Fluorescence contrast (tumor e muscle) of DY-734-annexin V in CAL-27 tumors in PDT-treated or non-treated mice at 2 days (d) or 2 weeks (w) after PDT or without PDT (up to 48 h post injection respectively). Each bar represents the mean of fluorescence contrast  SEM (n ¼ 5). (*p <0.05) (C) Biodistribution of DY-734-annexin V 5 days after PDT and 3 days after probe injection in PDT-treated (black, n ¼ 3) and non-PDT-treated (white, n ¼ 3) mice. The images show ex vivo fluorescence signals of the organs. (*p <0.05) (D) Biodistribution of DY-734-annexin V 3 weeks after PDT and 17 days (black bars, n ¼ 2) and 8 days (dark gray bars, n ¼ 5) after probe injection in PDT-treated mice or 17 days (white bars, n ¼ 2) and 8 days (light gray bars, n ¼ 5) after probe injection in non-PDT-treated mice. The images show associated fluorescence of the organs. (*p <0.05) (E) Cell nuclei of the cryo-frozen tumor slices 5 days (upper images) and 3 weeks (lower images) after PDT and without PDT, respectively. Arrow depicts condensed chromatin which is visible after DNA stain with Hoechst.

3. Results 3.1. In vitro and in vivo studies on the photosensitizer FoslipÒ The absorption spectrum of FoslipÒ showed a peak at 418 nm and another smaller peak at 650 nm (Fig. 2 A). The emission peak was at 650 nm after an excitation at 405 nm. After incubating CAL-27 cells with FoslipÒ, a strong uptake of the photosensitizer into the cytoplasm could be observed, especially in the peri-nuclear region (Fig. 2B). Cells without FoslipÒ incubation did not show any background fluorescence. Laser treatment alone using light dosages up to 100 J/cm2 did not affect viability of CAL-27 cells in vitro as could be shown via caspase 3/7 activity and XTT assay (data not shown). After injection of FoslipÒ into mice, the tumors showed a high ex vivo fluorescence intensity (0.0026 counts/s) 24 h post injection just like the other organs (Fig. 2C), such as the lungs (0.0018 counts/ s) and gut (0.0026 counts/s). The heart, spleen, kidney, and liver featured virtually no fluorescence. The organs of the mice without

any photosensitizer showed nearly no autofluorescence, with a significantly lower tumor fluorescence (0.00059 counts/s) compared to mice with FoslipÒ. 3.2. Spectroscopic analysis of DY-734-annexin V The spectroscopic analysis of DY-734-annexin V showed a high absorption peak at 718 nm and a smaller protein peak at 280 nm (Fig. 3A). The DY-734 alone revealed an absorption peak at 715 nm. The dye to protein ratio was in a range of 0.3 up to 0.5 dye molecules per annexin V protein. Using the in vivo fluorescence imaging system, DY-734-annexin V revealed a comparable emission spectrum to DY-734 alone with a maximum at 750 nm (Fig. 3B). 3.3. Binding affinity of DY-734-annexin V to apoptotic cells Flow cytometry illustrated a specific binding of DY-734-annexin V to apoptotic cells (Fig. 3C). This was evident in the observation that unstained cells (native) showed little or no autofluorescence,

K. Haedicke et al. / Biomaterials 34 (2013) 10075e10083

native

1week

2weeks

3weeks

B

1.5

1.0

PDT non-PDT

0.5

PDT

A

relative fluorescence intensity

10080

0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0

3

PDT non-PDT

0

Average Signal (counts/s)

D non-PDT

2 time [w]

relative tumor volume

C

1

5

10 15 time [d]

0.006

20

25

PDT (I) non-PDT (II)

0.005 0.004 0.003 0.002 0.001 0.000 tumor heart spleen lung

low

gut kidney liver

high signal intensity

Fig. 5. Reduction of the tumor vascularization and the tumor volume after PDT and biodistribution of IRDyeÒ 800CW RGD. (A) Representative light pictures (upper row) and composite images (lower row) of CAL-27 tumor bearing mice injected with IRDyeÒ 800CW RGD before (native), as well as 1, 2 and 3 weeks after PDT. Upper two panels: PDT-treated mice. Lower two panels: non-PDT-treated mice. (B) Semiquantitative analysis of the relative fluorescence intensity of IRDyeÒ 800CW RGD in CAL-27 tumors of PDT-treated (black, n ¼ 7) and non-PDT-treated (gray, n ¼ 7) mice with time. (C) Relative volume of CAL-27 tumors in PDT-treated (black) and non-PDT-treated (gray) mice with time. Arrow: time point of PDT. (D) Biodistribution of IRDyeÒ 800CW RGD in PDT-treated (black, n ¼ 7) and non-PDT-treated (white, n ¼ 7) mice 3 weeks after PDT and 24 h after last probe injection. The images show associated fluorescence of the organs.

irrespective of etoposide treatment. Apoptotic cells exhibited a specific binding of DY-734-annexin V (17.2% of the population) in comparison to non-apoptotic cells (1.2% of the population). The incubation of the CAL-27 cells with both DY-734-annexin V and propidium iodide revealed a clear differentiation between apoptotic and necrotic cells. Especially after etoposide treatment, many apoptotic (10.2% of the population) as well as some necrotic (7.5% of the population) cells could be observed. Altogether, DY734-annexin V revealed a high binding capability to phosphatidylserine on apoptotic cells, which were induced by etoposide. There were no differences between untreated and DY-734-annexin V stained cells (10 mg/ml) in terms of granularity and cell morphology (data not shown). Fluorescence microscopy revealed a co-localization of DY-734annexin V with WGA-Alexa-555-stained cell membrane in apoptotic CAL-27 cells (Fig. 3D). In addition, a fluorescence signal of the probe could also be detected in the cell nuclei. The estimated dissociation constant of the probe was approximately 20 nM (Fig. 3E). The specificity of the binding of DY-734annexin V could be shown in vitro by reduced fluorescence after incubating with a 10-fold excess (fluorescence reduction of 21%) or

a 100-fold excess (fluorescence reduction of 89%) of unlabeled annexin V (data not shown). 3.4. Detection of PDT-induced apoptosis via fluorescence imaging Among all time points investigated, PDT-induced apoptosis could be detected only at 2 days after treatment. In this case, fluorescence signals of DY-734-annexin V were obtained in PDTtreated tumors in mice with a decrease in signal with time after injection (Fig. 4A). 48 h after probe application, no optical fluorescence signal could be detected anymore. In contrast to the 2 day time frame after PDT, at 2 weeks after therapy no accumulation of the probe could be observed in the tumor over the whole measurement period (0 up to 48 h post injection, Fig. 4A). Macroscopically, a stimulated growth of the tumor periphery could be observed at 2 weeks after PDT. The corresponding accumulation kinetics at 2 days after therapy showed a higher contrast (tumor-to-muscle) of DY-734-annexin V in PDT-treated mice compared to non-PDT-treated mice, especially as of 4 h post injection (Fig. 4B, black and white bars). This effect was observed even after 48 h, indicating that the probe is

K. Haedicke et al. / Biomaterials 34 (2013) 10075e10083

specifically retained in PDT-treated tumors in contrast to the nonPDT-treated ones. The signals in both animal groups decreased with time, with a faster decrease in non-PDT-treated mice. At 2 weeks after PDT, imaging with DY-734-annexin V revealed lower signal intensities than at 2 days after PDT (Fig. 4B, gray bars). The enrichment of DY-734-annexin V in tumors in PDT-treated mice at 2 days after therapy was significantly higher (p <0.05) from 2 to 8 h post injection than in mice which received PDT 2 weeks before probe injection. Furthermore, the signals were lower in the 2-week PDT-treated mice than in the non-PDT-treated ones. Biodistribution analysis of DY-734-annexin V at 5 days after PDT and 3 days after probe injection showed the highest fluorescence intensity in the kidneys in PDT-treated as well as in non-PDTtreated animals (Fig. 4C). The liver of PDT-treated mice revealed a significantly lower (p <0.05) fluorescence signal compared to nonPDT-treated ones, whereas the spleen showed no signal. 3 weeks after PDT and 17 days after probe injection the fluorescence signals in the kidney and liver were comparably low in both animal groups (Fig. 4D, black and white bars). Also 3 weeks after PDT but 8 days after probe injection, the fluorescence intensities in these organs were higher (Fig. 4D, gray bars). Interestingly, the fluorescence signal in the kidneys of PDT-treated mice was significantly lower (p <0.05) than in the kidneys of non-PDT-treated mice at this time point. The liver fluorescence was also lower in PDT-treated animals whereas the spleen revealed no fluorescence. The stained nuclei of cryo-frozen tumor slices showed an altered morphology with condensed chromatin especially 5 days after PDT (Fig. 4E). 3.5. Detection of PDT-induced reduction of the tumor vascularization via fluorescence imaging After PDT, a loss of the tumor vascularization with time could be detected via whole body in vivo fluorescence optical imaging. Initially, the avb3 integrin targeting probe (IRDyeÒ 800CW RGD) showed a homogeneous intratumoral distribution in PDT-treated animals, whereas with increasing time after PDT, this fluorescence signal could only be detected in the outer areas of the tumor (Fig. 5A). In contrast, non-PDT-treated animals showed homogeneous fluorescence signals within the tumors over the whole experimental period. These observations were substantiated by the light images of mice, which clearly depicted the macroscopic destruction of the tumors after PDT in comparison to the tumors without therapy (Fig. 5A). Simultaneously, an increased growth of the outer tumor area could be seen at 2 weeks after PDT. The semiquantitative analysis of the relative fluorescence intensities of the tumors was in agreement with these findings (Fig. 5B). The fluorescence intensity of the PDT-treated tumors decreased as early as 1 week after PDT, remained constant up to 2 weeks after therapy and increased afterward. In comparison, the relative fluorescence intensity of the non-PDT-treated tumors showed a constant slight decrease with time. Concomitant to the decrease of the tumor fluorescence in PDT-treated mice, the tumor volume of the animals diminished after therapy (Fig. 5C). On the contrary, in mice without PDT, the tumor volume steadily increased or remained unchanged with time. The biodistribution of IRDyeÒ 800CW RGD 3 weeks after PDT and 24 h after the last injection showed the highest fluorescence signal in the tumors of PDT-treated (0.0044 counts/s) as well as of non-PDT-treated mice (0.0025 counts/s, Fig. 5D) compared to the other organs. The ex vivo images of the tumors showed that 3 weeks after PDT the fluorescence which reflects accumulation of the IRDyeÒ 800CW RGD probe could only be detected in the outer areas and not in the central area of the tumor, whereas without PDT the fluorescence is much higher in the tumor center. Beside the

10081

tumors, the lungs, gut, kidneys, and liver revealed fluorescence signals.

4. Discussion Using an in vivo multiplexed fluorescence optical imaging approach we were able to demonstrate that shortly after treatment a detection of the therapeutic efficacy of PDT was possible via apoptosis, whereas at longer times after PDT the therapeutic outcome could only be detected via tumor vascularization. The photosensitizer formulation FoslipÒ was shown to accumulate in CAL-27 cells with a strong enrichment in the peri-nuclear region and a diffuse distribution in the cytoplasm and other organelles. This cellular distribution pattern seems to be typical for photosensitizers irrespective of the cell line [15,16]. Especially the Golgi apparatus and the endoplasmic reticulum are the primary intracellular sites for localization of mTHPC and, to a lesser extent, also lysosomes and mitochondria [17,18]. The ability of the used photosensitizer to internalize into target cells fulfills an important prerequisite for a successful therapy. In this work FoslipÒ also revealed the ability to accumulate in CAL-27 xenografts in mice. Since this formulation is characterized by the encapsulation of the photosensitizer Temoporfin in liposomes, the high water solubility allowed for a good accessibility to the tumor region and a lower formation of aggregates [19]. The accumulation of the photosensitizer in the other organs, especially the lung, can be ascribed to the high perfusion of these organs. Therefore, a well-directed illumination on the tumor will locally activate the photosensitizing agent and also protect the other organs. In this context, the high enrichment of gut and lungs with the photosensitizer makes it clear that the treatment of tumors in these regions is rather challenging. For in vivo detection of apoptotic cell death after PDT, we successfully designed a fluorescence optical annexin V probe. Covalent coupling of the dye to the annexin V was accomplished, since a NHS-ester derivative of the DY-734 was used. NHS-esters are known to be highly reactive with primary amino groups, for example those which were present on the annexin V protein. This kind of reaction is widely known to produce an amide bond between the reactants [20]. Spectroscopic analysis demonstrated the efficient coupling of the NIRF dye DY-734 to annexin V by the simultaneous appearance of a protein peak and a dye peak. The absorption and emission spectra of DY-734 were not altered after coupling to annexin V. The cellular localization of the probe on the cell membrane coincided with the localization of phosphatidylserine after treating of the cells with the anticancer agent etoposide [21]. This means that a specific detection of a changed plasma membrane structure is feasible with the annexin V probe. Moreover, the probe featured a high binding affinity and selectivity to apoptotic CAL-27 cells. Since the target phosphatidylserine is highly accessible after externalization to the outer leaflet of the plasma membrane, our probe will mostly bind apoptotic cells. In contrast, binding to necrotic cells can just occur at later stages, when the membrane is already disrupted and phosphatidylserine is accessible at the inner leaflet [22]. Additionally, flow cytometry of apoptotic cells showed that binding to cellular membrane fragments of necrotic cells was comparatively lower than binding to phosphatidylserine on apoptotic cells. This means that our probe will mainly detect apoptotic cells in the in vivo situation. The semiquantitative assessment of the binding affinity of the annexin V optical probe revealed a dissociation constant of 20 nM. The determined KD of DY-734-annexin V was nearly of the same magnitude as corresponding values in other reports [23e27]. This indicates that the DY-734 dye molecule in the probe did not

10082

K. Haedicke et al. / Biomaterials 34 (2013) 10075e10083

interfere with the binding sites of the annexin protein and reflects a high affinity of the optical probe to apoptotic cells. Our probe is expected to elicit a low cytotoxic potential. It could be shown that hemicyanine dyes, such as DY-734, are not cytotoxic on endothelial and macrophage cells because of its low lipophilicity and thus high hydrophilicity [28,29]. The same applies for the annexin V protein. Annexin is a naturally occurring protein and it has already been widely used in the clinics as target-affine probe in nuclear medicine applications [22]. As we could show by flow cytometry, the granularity of the cells after staining with DY-734annexin V (10 mg/ml) was comparable to native cells. Furthermore there were no changes of the morphology after staining compared to untreated cells detectable. For in vivo imaging, a dose of 4 nmol/kg body weight was applied. This suggests that our optical annexin probe is not cytotoxic and highly biocompatible in the in vivo situation. The enrichment of DY-734-annexin V in tumor models in vivo was seen 2 days after PDT, which is a real effect of the therapy since we could show that laser treatment alone is safe using in vitro experiments. It is known that 1 h after PDT defined apoptosis markers, such as DNA fragments, externalization of phosphatidylserine etc., are evident in tumor tissue and increase with time after treatment [6,30,31]. This means that the annexin V probe accumulation in tumors at 2 days after PDT coincides with the known pathobiological mechanisms of apoptosis, particularly with the alteration of the morphology of cell nuclei. This could be shown with the chromatin staining of treated tumor tissues. In previous studies using a radiolabeled annexin V probe, signaling in tumors were observed only at 2, 4 and 7 h after PDT [9]. The fact that the annexin probe enrichment was detected at 2 days after PDT proposes that the time frame for a reliable detection of therapeuticallybased apoptotic effects is 48 h post PDT. The avb3 integrin targeting probe (IRDyeÒ 800CW RGD) showed a homogeneous intratumoral distribution before PDT and only a slight fluorescence signal in the outer areas of the tumor 2 weeks after PDT. The inability to detect apoptosis with DY-734-annexin V 2 weeks after PDT is partly attributed to the PDT-induced damage of the tumor vascularization. Consistent with this damage of tumor vessels, access of the optical probe to apoptotic cells in the tumor is aborted. The decreased signaling of the RGD probe in tumors at 1 week after therapy also reflects the therapeutic efficacy of PDT. Furthermore, the PDT-based destruction of the blood vessels also results in a decrease of the tumor volume with time, due to limited availability of nutrients. This mechanism, which is associated with the tumor destruction through PDT [32,33], has to the best of our knowledge, never been monitored via fluorescence optical imaging over a time frame of 3 weeks. A cytotoxic effect of the IRDyeÒ 800CW RGD probe can be excluded because radiolabelled RGD peptides are already widely used in clinical nuclear medicine [34] and the cyanine fluorophore IRDye 800CW does not evidence any pathological effect or toxicity after intravenous or intradermal injection in rats at a much higher concentration of the dye (18 mmol/ kg) [35] compared to our study (40 nmol/kg). The intratumoral distribution pattern of the IRDyeÒ 800CW RGD probe in PDT-treated tumors was clearly different from non-PDTtreated tumors. PDT-treated tumors showed fluorescence signals only in their outer areas, whereas non-treated tumors with intact vascularization revealed fluorescence especially at the tumor center. Also the macroscopically observed morphology of PDT-treated tumors shows that therapy may stimulate tumor growth at the periphery. The vascularization and thus the perfusion in the tumor become altered after therapy as evident in the inhomogeneous distribution of the avb3 integrin targeting probe in the tumors. This heterogeneous distribution reflects a partially poor perfusion which can lead to hypoxia and acidosis, both known to promote

different aspects of carcinogenesis, like induction of chromosomal strand breaks or metastasis, increased resistance to anticancer therapies, the development of a more aggressive tumor phenotype and the induction of proangiogenic factors [36]. Therefore, particular attention should be devoted to the treatment plan or multiple treatments should be organized to avoid these incidents [32]. The reason for the presence of suboptimal treated regions at the outer tumor range is unknown and might be ascribed to the inhomogeneous distribution of FoslipÒ [37] or a non-uniform laser illumination of the tumor area. The biodistribution data of DY-734-annexin V 5 days after PDT and 3 days after application showed the highest fluorescence signals in the kidneys and lower signals in the liver. This biodistribution pattern is similar to that found for 99mTc labeled annexin V [9,38]. This implies that regardless of which dye or radionuclide is coupled to annexin V, the biodistribution of the probe remains almost unaltered. The observed kidney and liver fluorescence signals illustrate the fast renal excretion of the optical probe, suggesting a nonspecific uptake by the proximal renal tubule cells, which has been shown to be a typical event for low-molecularweight proteins [9,38]. It also implies a rapid blood clearance. The half-life of annexin V after intravenous injection is known to be less than 7 min [23,39,40]. Due to the high affinity (KD ¼ 20 nM) and high specificity (reduction of fluorescence around 89% in a blocking experiment), our DY-734-annexin V probe was still able to efficiently and specifically detect the therapeutic efficacy of PDT at 2 days after treatment as our in vivo imaging data show. The comparatively low fluorescence signals in the liver peaked at 8 days after DY-734-annexin V injection. This finding can be attributed to fluorescence labeled apoptotic cells, which are removed out of the tumor by phagocytic macrophages [23] and transferred to the liver. On the other hand, it also suggests that the fluorescence signals are due to probe degradation processes, which lead to the accumulation of the hydrophilic dye in the kidney and liver [41]. The reason for the higher accumulation of the annexin probe or free dye in the liver of non-PDT-treated compared to PDTtreated animals is not clear. Presumably, the systemic injection of the photosensitizing agent modifies distinct functions in the liver. A systemic effect of PDT of tumors (i.e. use of photosensitizer and laser light) is unlikely since the photosensitizer is selectively activated at the tumor area. In this context, DY-734 Fab fragments were also shown to be accumulated in the kidney in contrast to the whole antibody counterparts, which demonstrates the particular renal elimination of low-molecular-weight probes [42]. Additionally the hydrophilic properties of DY-734 imply a low affinity for plasma proteins, which further promotes renal excretion. The biodistribution of the avb3 integrin targeting probe showed the highest fluorescence signal in the tumors and thus a specific detection of the tumor vascularization via binding to integrins. Fluorescence signals of the kidneys reflect the renal excretion of this small peptide. On the other hand, a cleavage of the IRDyeÒ 800CW from the RGD-peptide could be responsible for the probe accumulation in the kidneys [35]. Our data show that multiplexed fluorescence optical imaging offers the possibility of a simultaneous detection of several targets by using different spectroscopic properties of the used probes. This allows us to distinctly monitor the molecular changes that result due to therapy. Considering the limited penetration depth of light into tissues, the application of the proposed methodology is restricted to pathologies being easily accessible to light. 5. Conclusion In summary, we demonstrated the feasibility of a multiplexed approach for the detection of the therapeutic efficacy of PDT by

K. Haedicke et al. / Biomaterials 34 (2013) 10075e10083

means of the simultaneous detection of phosphatidylserine in early apoptosis with annexin V and the tumor vascularization. We show that the detection of PDT-based apoptosis induction is feasible at least 2 days after treatment. Furthermore, we demonstrated that to avoid false negative findings, the inclusion of a second probe targeting the tumor vasculature is indispensable. Taken together, our observations provide new insights into tumor response to PDT as well as into the mechanisms of therapy monitoring.

Acknowledgments The study was supported by the Thüringer Aufbaubank (Erfurt, Germany). We gratefully acknowledge Kathrin Hornung, Susann Burgold, Julia Göring, Yvonne Ozegowski and Doreen May for valuable technical assistance.

References [1] Massoud TF, Gambhir SS. Molecular imaging in living subjects: seeing fundamental biological processes in a new light. Genes Dev 2003;17(5):545e 80. [2] Agostinis P, Berg K, Cengel KA, Foster TH, Girotti AW, Gollnick SO, et al. Photodynamic therapy of cancer: an update. CA Cancer J Clin 2011;61(4): 250e81. [3] Dome B, Hendrix MJ, Paku S, Tovari J, Timar J. Alternative vascularization mechanisms in cancer: pathology and therapeutic implications. Am J Pathol 2007;170(1):1e15. [4] Miller JC, Pien HH, Sahani D, Sorensen AG, Thrall JH. Imaging angiogenesis: applications and potential for drug development. J Natl Cancer Inst 2005;97(3):172e87. [5] Moor AC. Signaling pathways in cell death and survival after photodynamic therapy. J Photochem Photobiol B 2000;57(1):1e13. [6] Fadok VA, Voelker DR, Campbell PA, Cohen JJ, Bratton DL, Henson PM. Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J Immunol 1992;148(7): 2207e16. [7] van Engeland M, Nieland LJ, Ramaekers FC, Schutte B, Reutelingsperger CP. Annexin V-affinity assay: a review on an apoptosis detection system based on phosphatidylserine exposure. Cytometry 1998;31(1):1e9. [8] Zhang G, Gurtu V, Kain SR, Yan G. Early detection of apoptosis using a fluorescent conjugate of annexin V. Biotechniques 1997;23(3):525e31. [9] Subbarayan M, Hafeli UO, Feyes DK, Unnithan J, Emancipator SN, Mukhtar H. A simplified method for preparation of 99mTc-annexin V and its biologic evaluation for in vivo imaging of apoptosis after photodynamic therapy. J Nucl Med 2003;44(4):650e6. [10] Vangestel C, Peeters M, Oltenfreiter R, D’Asseler Y, Staelens S, Van Steenkiste M, et al. In vitro and in vivo evaluation of [99mTc]-labeled tricarbonyl His-annexin A5 as an imaging agent for the detection of phosphatidylserine-expressing cells. Nucl Med Biol 2010;37(8):965e75. [11] Cauchon N, Langlois R, Rousseau JA, Tessier G, Cadorette J, Lecomte R, et al. PET imaging of apoptosis with (64)Cu-labeled streptavidin following pretargeting of phosphatidylserine with biotinylated annexin-V. Eur J Nucl Med Mol Imaging 2007;34(2):247e58. [12] Ntziachristos V, Schellenberger EA, Ripoll J, Yessayan D, Graves E, Bogdanov Jr A, et al. Visualization of antitumor treatment by means of fluorescence molecular tomography with an annexin V-Cy5.5 conjugate. Proc Natl Acad Sci U S A 2004;101(33):12294e9. [13] Petrovsky A, Schellenberger E, Josephson L, Weissleder R, Bogdanov Jr A. Near-infrared fluorescent imaging of tumor apoptosis. Cancer Res 2003;63(8): 1936e42. [14] Lederle W, Arns S, Rix A, Gremse F, Doleschel D, Schmaljohann J, et al. Failure of annexin-based apoptosis imaging in the assessment of antiangiogenic therapy effects. EJNMMI Res 2011;1(1):26. [15] Lassalle HP, Wagner M, Bezdetnaya L, Guillemin F, Schneckenburger H. Fluorescence imaging of Foscan and Foslip in the plasma membrane and in whole cells. J Photochem Photobiol B 2008;92(1):47e53. [16] Kiesslich T, Berlanda J, Plaetzer K, Krammer B, Berr F. Comparative characterization of the efficiency and cellular pharmacokinetics of Foscan- and Foslip-based photodynamic treatment in human biliary tract cancer cell lines. Photochem Photobiol Sci 2007;6(6):619e27. [17] Teiten MH, Bezdetnaya L, Morliere P, Santus R, Guillemin F. Endoplasmic reticulum and Golgi apparatus are the preferential sites of Foscan localisation in cultured tumour cells. Br J Cancer 2003;88(1):146e52.

10083

[18] Teiten MH, Marchal S, D’Hallewin MA, Guillemin F, Bezdetnaya L. Primary photodamage sites and mitochondrial events after Foscan photosensitization of MCF-7 human breast cancer cells. Photochem Photobiol 2003;78(1):9e14. [19] Svensson J, Johansson A, Grafe S, Gitter B, Trebst T, Bendsoe N, et al. Tumor selectivity at short times following systemic administration of a liposomal temoporfin formulation in a murine tumor model. Photochem Photobiol 2007;83(5):1211e9. [20] Mentinova M, McLuckey SA. Covalent modification of gaseous peptide ions with N-hydroxysuccinimide ester reagent ions. J Am Chem Soc 2010;132(51): 18248e57. [21] Schonn I, Hennesen J, Dartsch DC. Cellular responses to etoposide: cell death despite cell cycle arrest and repair of DNA damage. Apoptosis 2010;15(2): 162e72. [22] Yang TJ, Haimovitz-Friedman A, Verheij M. Anticancer therapy and apoptosis imaging. Exp Oncol 2012;34(3):269e76. [23] Blankenberg FG, Tait J, Ohtsuki K, Strauss HW. Apoptosis: the importance of nuclear medicine. Nucl Med Commun 2000;21(3):241e50. [24] Yen TC, Wey SP, Liao CH, Yeh CH, Shen DW, Achilefu S, et al. Measurement of the binding parameters of annexin derivative-erythrocyte membrane interactions. Anal Biochem 2010;406(1):70e9. [25] Andree HA, Reutelingsperger CP, Hauptmann R, Hemker HC, Hermens WT, Willems GM. Binding of vascular anticoagulant alpha (VAC alpha) to planar phospholipid bilayers. J Biol Chem 1990;265(9):4923e8. [26] Reutelingsperger CP, van Heerde WL. Annexin V, the regulator of phosphatidylserine-catalyzed inflammation and coagulation during apoptosis. Cell Mol Life Sci 1997;53(6):527e32. [27] van Heerde WL, Poort S, van’t Veer C, Reutelingsperger CP, de Groot PG. Binding of recombinant annexin V to endothelial cells: effect of annexin V binding on endothelial-cell-mediated thrombin formation. Biochem J 1994;302(Pt 1):305e12. [28] Pauli J, Brehm R, Spieles M, Kaiser WA, Hilger I, Resch-Genger U. Novel fluorophores as building blocks for optical probes for in vivo near infrared fluorescence (NIRF) imaging. J Fluoresc 2010;20(3):681e93. [29] Pauli J, Vag T, Haag R, Spieles M, Wenzel M, Kaiser WA, et al. An in vitro characterization study of new near infrared dyes for molecular imaging. Eur J Med Chem 2009;44(9):3496e503. [30] Agarwal ML, Clay ME, Harvey EJ, Evans HH, Antunez AR, Oleinick NL. Photodynamic therapy induces rapid cell death by apoptosis in L5178Y mouse lymphoma cells. Cancer Res 1991;51(21):5993e6. [31] Zaidi SI, Oleinick NL, Zaim MT, Mukhtar H. Apoptosis during photodynamic therapy-induced ablation of RIF-1 tumors in C3H mice: electron microscopic, histopathologic and biochemical evidence. Photochem Photobiol 1993;58(6): 771e6. [32] Dolmans DE, Kadambi A, Hill JS, Flores KR, Gerber JN, Walker JP, et al. Targeting tumor vasculature and cancer cells in orthotopic breast tumor by fractionated photosensitizer dosing photodynamic therapy. Cancer Res 2002;62(15):4289e94. [33] Fingar VH, Wieman TJ, Wiehle SA, Cerrito PB. The role of microvascular damage in photodynamic therapy: the effect of treatment on vessel constriction, permeability, and leukocyte adhesion. Cancer Res 1992;52(18): 4914e21. [34] Gaertner FC, Kessler H, Wester HJ, Schwaiger M, Beer AJ. Radiolabelled RGD peptides for imaging and therapy. Eur J Nucl Med Mol Imaging 2012;39(Suppl. 1):S126e38. [35] Marshall MV, Draney D, Sevick-Muraca EM, Olive DM. Single-dose intravenous toxicity study of IRDye 800CW in SpragueeDawley rats. Mol Imaging Biol 2010;12(6):583e94. [36] Gillies RJ, Schornack PA, Secomb TW, Raghunand N. Causes and effects of heterogeneous perfusion in tumors. Neoplasia 1999;1(3):197e207. [37] Maeda H, Nakamura H, Fang J. The EPR effect for macromolecular drug delivery to solid tumors: improvement of tumor uptake, lowering of systemic toxicity, and distinct tumor imaging in vivo. Adv Drug Deliv Rev 2013;65(1): 71e9. [38] Blankenberg FG, Katsikis PD, Tait JF, Davis RE, Naumovski L, Ohtsuki K, et al. Imaging of apoptosis (programmed cell death) with 99mTc annexin V. J Nucl Med 1999;40(1):184e91. [39] Tait JF, Cerqueira MD, Dewhurst TA, Fujikawa K, Ritchie JL, Stratton JR. Evaluation of annexin V as a platelet-directed thrombus targeting agent. Thromb Res 1994;75(5):491e501. [40] Wen X, Wu QP, Ke S, Wallace S, Charnsangavej C, Huang P, et al. Improved radiolabeling of PEGylated protein: PEGylated annexin V for noninvasive imaging of tumor apoptosis. Cancer Biother Radiopharm 2003;18(5):819e27. [41] Hamann FM, Brehm R, Pauli J, Grabolle M, Frank W, Kaiser WA, et al. Controlled modulation of serum protein binding and biodistribution of asymmetric cyanine dyes by variation of the number of sulfonate groups. Mol Imaging 2011;10(4):258e69. [42] Busch C, Passon M, Wenzel M, Socher I, Kaiser WA, Hilger I. Detection of leukotriene receptor CysLT1R in inflammatory diseases by molecular imaging with near-infrared fluorescence-based contrast agents. Mol Imaging 2011;10(2):81e90.