The molecular basis of prostate cancer cell escape from protoporphyrin IX-based photodynamic therapy

ARTICLE IN PRESS

Medical Laser Application 24 (2009) 237–246 www.elsevier.de/mla

The molecular basis of prostate cancer cell escape from protoporphyrin IX-based photodynamic therapy Robert Kammerera,b,, Patrick Pallucha, Konstantin Oboukhovskija, Mariana Toelgea, Thomas Pongratzc, Wolfgang Beyerc, Alexander Buchnera,d, Reinhold Baumgartnerc, Wolfgang Zimmermanna a

Tumor Immunology Laboratory, LIFE Center, Ludwig Maximilian University, Großhadern Medical Campus, Marchioninistrasse 23, 81377 Munich, Germany b Institute of Immunology, Friedrich-Loeffler-Institute, Paul-Ehrlich-Straße 28, 72076 Tu¨bingen, Germany c Laser Research Laboratory, LIFE Center, Ludwig Maximilian University, Großhadern Medical Campus, Marchioninistrasse 23, 81377 Munich, Germany d Department of Urology, Ludwig Maximilian University, Großhadern Medical Campus, Marchioninistrasse 15, 81377 Munich, Germany Received 14 July 2009; accepted 29 July 2009

Abstract Introduction: Photodynamic therapy (PDT) uses the combination of a photosensitizing drug and light to cause selective damage to solid tumors. For early or localized disease, PDT can be a curative therapy with many advantages over available alternatives. For more advanced or disseminated disease, curative therapy is usually not possible with current technologies of light application, nevertheless it can improve quality of life and lengthen survival. More recently, there has been increasing evidence that PDT can also induce systemic anti-tumor immunity, indicating that PDT can become a rational therapeutic option, even if not all tumor cells are primarily eliminated by PDT. One prerequisite to minimize tumor relapse in these settings is to elucidate how tumor cells can cope with the PDT-induced damage. Material and methods: Here we have characterized the rescue response of human PC-3 prostate cancer cells exposed in vitro to sublethal PDT after 5-aminolevulinic acid-induced protoporphyrin IX sensitization at the transcriptome level using Affymetrix HG U133 Plus 2.0 oligonucleotide microarrays. Cells were irradiated with laser light at a wavelength of 635 nm adjusted to an irradiance of 100 mW/cm2 with irradiations of 1.5 and 3 J/cm2. Results: The early response was characterized by the up-regulation of early response genes like FOS, JUN, EGR1, ATF3, DUSP, heat shock protein genes as well as histone genes, therefore resembling the early response of tumor cells to high dose PDT but without signs of irreversible cell damage. Twenty-four hours after PDT the cells still express high level of early response genes but 235 additional probe sets/genes were now significantly up-regulated (Z3x) that were not up-regulated 4 h after PDT. The most prominently up-regulated genes belong to gene families encoding the aldoketo reductases, fibroblast growth factors, and HSP40-related proteins. In terms of a possible anti-tumor immune response it is noteworthy that also a multitude of chemokine and interleukin genes were up-regulated by the tumor Corresponding author. Tel.: +49 7071 967 156; fax: +49 7071 967 305.

E-mail address: Robert.Kammerer@fli.bund.de (R. Kammerer). 1615-1615/$ - see front matter & 2009 Elsevier GmbH. All rights reserved. doi:10.1016/j.mla.2009.07.001

ARTICLE IN PRESS 238

R. Kammerer et al. / Medical Laser Application 24 (2009) 237–246

cells upon PDT. Most of them are involved in granulocyte attraction and activation and some are also important for angiogenesis. Conclusion: In conclusion, the global molecular characterization of the rescue response to PDT of tumor cells indicates that PDT rather favors an anti-tumor immune response than tumor immune escape reactions. Therefore, combining PDT and immunotherapy seem to be an attractive direction for the establishment of novel multimodal tumor therapies. & 2009 Elsevier GmbH. All rights reserved. Keywords: Photodynamic therapy; Prostate cancer; 5-Aminolevulinic acid; Oligonucleotide microarrays; Immunotherapy

Introduction Prostate cancer (PCa) is one of the most frequent cancers in males in Western industrialized countries. Its course is highly variable, from indolent to highly lethal. The number of patients with PCa is increasing and, according to the rates from 2003 to 2005, it has been estimated that 15.78%, or 1 in every 6 men, will be diagnosed with PCa at some point during his lifetime [1]. Depending on the disease stage, the primary treatment of PCa includes radiotherapy (RT), radical prostatectomy (RP), transperineal brachytherapy, cryotherapy, high-intensity focused ultrasound (HIFU), and androgen-deprivation therapy. Although significant improvement has occurred in the outcomes of many treatments of localized PCa, a number of significant adverse effects, which have a considerable effect on patients and their families, are associated with definitive treatment. New techniques are being developed that aim to combine similar treatment effects with reduced adverse effects. One candidate of such treatments is photodynamic therapy (PDT) [2]. The main goal of PDT of cancer is to kill all tumor cells by a lethal dose of PDT. However, due to the heterogeneity of tumor tissues, the light dose effectively delivered to a given tumor cell within solid tumors is highly variable and may not exceed the lethal dose threshold. In addition, in tumors with highly invasive tumor cell populations at the tumor–normal tissue interface, a significant number of tumor cells will not be treated with a lethal light dose. In such tumor cells rescue responses might be induced, which allow the tumor cells to cope with the damage inflicted by the reactive radicals and to escape PDT, giving rise to tumor relapse. However, although PDT is in principle a local therapy, more recently an impressive amount of evidence has accumulated that suggests that PDT has also systemic effects by eliciting an anti-tumor immune response [3,4]. It is well known that massive cell destruction by PDT leads to an innate immune response. More recently, it was demonstrated that PDT can also stimulate the adaptive immune system [5–7]. It is thought that PDT leads to an overwhelming presenta-

tion of danger signals due to the generation of heat shock proteins, release of intracellular proteins and degradation products of cellular membranes (lysophospholipids and arachidonic acid metabolites) [8–10]. The presence of danger signals may favor an adaptive immune response and the release of cellular proteins may facilitate cross presentation of tumor antigens by professional antigen-presenting cells. However, the direct effect of PDT may be harmful for immune cells within the tumor area [11], and surviving tumor cells may exhibit an immunosuppressive effect on newly invading immune cells by secretion of immune suppressive factors like IL-10 [12]. This study was conducted to give a better understanding of the possible role of tumor cells, which survive PDT, for tumor recurrence and tumor immunity. In particular, we were interested in identifying mechanisms which induce invasion and proliferation, as well as suppressing an anti-tumor immune response after PDT.

Material and methods Cell culture The prostate carcinoma cell line PC-3 was cultured in RPMI-1640 supplemented with 5% fetal calf serum (FCS ‘‘Gold’’; PAA Laboratories, Coelbe, Germany), 2 mM L-glutamine, 100 U/ml penicillin, 100 mg/ml streptomycin, non-essential amino acids and 1 mM sodium pyruvate (GIBCO/Invitrogen, Karlsruhe, Germany) at 37 1C with 5% CO2 in a humidified atmosphere. Absence of mycoplasma contamination was determined using the VenorGeMs Kit (Minerva Biolabs, Berlin, Germany).

Photodynamic treatment For determination of the light dose suitable for induction of a 15% and 25% loss of cell viability 4 and 24 h after PDT respectively, 1  104 cells per well were plated in flat bottom 96-well plates (Nunc GmbH, Wiesbaden, Germany) and grown for 6–8 h in medium

ARTICLE IN PRESS R. Kammerer et al. / Medical Laser Application 24 (2009) 237–246

with 5% FCS. Sensitization of the cells with 50 mg/ml 5-ALA (Medac GmbH, Hamburg, Germany) was performed for 16 h in fresh medium with 5% FCS. Generation of protoporphyrin IX (PpIX) in the tumor cells was followed by determination of the mean fluorescence in the FL3 photomultiplier tube (670 nm long pass filter) on a FACSCaliburs flow cytometer (BD Biosciences, Erembodegem, Belgium). The medium was replaced with RPMI-1640 without phenol red (GIBCO/Invitrogen, Karlsruhe, Germany) containing supplements and 5% serum as listed above. Specific light doses were delivered to the cells by irradiation (six wells at the time) with laser light (635 nm; Ceralas diode laser, CeramOptec GmbH, Bonn, Germany) at 100 mW/cm2 for varying times in a light-tight box with a 37 1C warm plate. For the isolation of RNA from PDT-treated cells, tumor cells (1  106) were grown in 6 cm diameter cell culture dishes (Nunc GmbH, Wiesbaden, Germany), treated with 5-ALA as described above and irradiated for 15 s (1.5 J/cm2) at room temperature. Control cells were treated identically but were kept in the dark. Identically set-up and treated cell cultures were used for cell viability determination (see below). After PDT, cells were incubated in the same medium at 37 1C for the indicated times.

Cell viability and survival After PDT, cell viability and cell growth were quantified using the CellTiter-BlueTM Cell Viability Assay (Promega, Mannheim, Germany) as recommended by the manufacturer. In brief, after the indicated times, 200 ml of CellTiter-BlueTM solution per ml were directly added to the cell culture medium, mixed gently and after 1 h, 1 ml of the cell culture supernatant were removed and stored in the dark at 4 1C (up to 24 h) until quantification of its fluorescence in 96-well cell culture plates (100 ml per well) using a FLUOstar OPTIMA microplate reader (BMG LABTECH, Jena, Germany). The background fluorescence values were obtained by adding CellTiter-BlueTM solution to medium without cells for 1 h.

RNA amplification and microarray hybridization PDT-treated cells or cells identically treated except for irradiation were harvested using trypsin/EDTA. Total RNA was isolated from the cell pellets using the RNeasy Mini Kit (Qiagen, Hilden, Germany). The RNA yield was quantified photometrically and the integrity was determined by capillary electrophoresis (Agilent 2100 Bioanalyzer and RNA 6000 Pico Kit; Agilent Technologies Deutschland, Bo¨blingen, Germany). The RNA integrity number (RIN) provided by the Agilent system allows a quantitative estimate of the RNA quality. The

239

RIN number (10 represents intact RNA) for all samples was 49.5 [13]. RNA amplification and biotin labeling was performed by reverse transcription of 5 mg of total RNA with an oligo-dT T7 promoter primer and linear in vitro transcription using a kit from Affymetrix (One Cycle Target Labeling Kit) according to the manufacturer’s instructions. Hybridization of biotin-labeled RNA to Affymetrix GeneChips HG U133 Plus 2.0 oligonucleotide microarrays was done for 16 h at 37 1C in a hybridization oven (Affymetrix). Then the arrays were washed according to the standard protocol and stained by addition of streptavidin-phycoerythrin using the Fluidics Station 450 (Affymetrix). Laser scanning of the arrays was performed using the GeneChips Scanner 3000 (Affymetrix). Raw hybridization data were collected using the GeneChips Operating Software (GCOS) released by Affymetrix and stored as CEL files.

Microarray data analysis CEL file data from three independent experiments were used to determine a model-based expression index (MBEI) with the approach of Li and Wong [14] that is implemented in the dChip 2007 software (http:// www.dchip.org; Dana-Farber Cancer Center and Harvard School of Public Health, Boston, MA, USA). This algorithm has shown good performance in processing of raw microarray data in a comparative study of seven methods [15]. The MBEI values were used for subsequent high-level analysis (e.g. hierarchical clustering). Additionally, the detection call for each transcript was calculated using the MAS5.0 algorithm that is implemented in the Expression ConsoleTM 1.1 software (Affymetrix). A transcript (gene/probe set) is called ‘‘present’’ if its expression is switched on, and ‘‘absent’’ if its expression is switched off. Functional classification was performed by using the Database for Annotation, Visualization and Integrated Discovery (DAVID, 2008) online tools [16,17].

Cytokine determination For the quantitation of cytokines in supernatants of PDT-treated cells, 1  106 cells were plated in 6 cm diameter cell culture dishes, sensitized with 5-ALA and PDT-treated as described above. Twenty four hours later, cells were removed by trypsin-treatment and 1  106 live (trypan blue-excluding) cells were re-plated and cultured for additional 24 h. Supernatants were collected, centrifuged to remove cell debris and stored at 20 1C until use. The concentrations of cytokines were determined using the BDTM Cytometric Bead Array system (BD, Heidelberg, Germany) in the FACSCaliburs flow cytometer (BD Biosciences, Erembodegem, Belgium) according to the manufacturer’s instruction.

ARTICLE IN PRESS 240

R. Kammerer et al. / Medical Laser Application 24 (2009) 237–246

Overview of transcriptional changes upon low-dose PDT

Results Prostate cancer cell viability and survival upon lowdose PDT using 5-ALA-induced PpIX as photosensitizer

For gene expression analysis, tumor cells were treated with low-dose PDT and total RNA was harvested from irradiated and non-irradiated cells 4 and 24 h after the end of photosensitization. Gene expression levels were determined using oligonucleotide microarrays (Affymetrix GeneChips HG U133 Plus 2.0 Array) containing more than 47,000 probe sets. Without treatment, cell lines expressed between 37% and 43% of all probe sets, which was not changed significantly 4 and 24 h after irradiation (Fig. 2A). We identified differentially expressed genes comparing the expression level of irradiated cell at a given time point with the expression of untreated cells from three independent experiments. Those genes were considered significantly modulated when a Z3 fold down- or up-regulation of their mean normalized expression values with a relative expression value difference of Z75 was observed. Only up-regulated genes with a ‘‘present call’’ for the irradiated group were accepted in at least two of the three independent experiments. Down-regulated genes had a ‘‘present call’’ in two of three experiments in the untreated group, were Z3 fold down-regulated and had a minimum difference of the relative expression value of 75. Four hours after PDT 232 probes/genes were Z3 fold up-regulated while only 11 probes/genes were Z3 fold down-regulated (Fig. 2B). Twenty-four hours after

To understand the survival strategies including immune escape mechanisms of prostate tumor cells following 5-ALA-based PDT, we have chosen a nonlethal irradiation protocol in which cells exhibit more than 70% viability, 24 h after light application. We chose the prostate cancer cell line PC-3 which has been derived from a bone metastasis of a prostate adenocarcinoma patient as a model. Cells were incubated with 5-ALA and accumulation of PpIX within the tumor cells was determined by FACS analysis as described in the ‘Material and methods’ section. All tumor cells showed a similar accumulation of 5-ALA-induced PpIX resulting in a Gaussian distribution of fluorescence intensity (Fig. 1A). PC-3 cells accumulated 90% of maximum levels of PpIX under the chosen conditions. Maximum accumulation of PpIX was determined for each cell line by variation of 5-ALA concentrations and incubation times (data not shown). Fig. 1B shows the sensitivity of the PC-3 cell line to the applied light dose of 1.5 J/cm2. Under these PDT conditions, re-growth of cells was observed at 48 h after irradiation (data not shown).

A

B 200 180

64

160

Events

Survival [%]

140 120 100 80 60 40

Control

20 0 100

PDT

0 101

102 FL3-H

103

104

0

10 20 time after PDT [h]

30

Fig. 1. PpIX accumulation and sensitivity to low-dose PDT of the prostate carcinoma cell line PC-3: (A) concentration of PpIX in individual cells as determined by flow cytometry. PC-3 cells were incubated with (open curve) or without 50 mg/ml 5-ALA (red filledin curve) for 16 h in medium with 5% FCS. Then their PpIX content was measured by flow cytometry in the FL-3 channel. Note the Gaussian distribution of PpIX which can differ more than 10-fold between individual cells. (B) Influence of PDT on the viability of PC-3 tumor cells. PpIX-sensitized PC-3 cells were irradiated (open circles) or not (closed circles) with 1.5 J/cm2 – 635 nm light from a laser diode. At the indicated time points, cell viability/survival was determined by measuring their metabolic capacity through reduction of resazurin to the fluorescent resorufin.

ARTICLE IN PRESS R. Kammerer et al. / Medical Laser Application 24 (2009) 237–246

241

PDT 442 probes/genes were Z3 fold up-regulated and 401 probes/genes were Z3 fold down-regulated. The 138 probes/genes were up-regulated after 4 h as well as after 24 h while no probe/gene was down-regulated after 4 h and after 24 h (Fig. 2B). During recovery between 4 and 24 h, 211 probes/genes were up-regulated Z3 fold and 235 probes/genes were down-regulated. Sixteen probes/ genes were already up-regulated Z3 fold in the first four hours and further Z3 fold up-regulated until 24 h after irradiation. Forty-one probes/genes, which were upregulated at 4 h, were Z3 fold down-regulated between 4 and 24 h after irradiation (Fig. 2C).

Transcriptional changes upon low-dose PDT

Fig. 2. Genome-wide transcriptional changes after PpIXbased low-dose PDT of PC-3 cells: (A) PC-3 tumor cells were treated with low-dose PDT (see legend to Fig. 1) and total RNA was isolated from irradiated and non-irradiated cells (‘‘untreated’’) 4 and 24 h after the end of sensitization with 5-ALA. Global gene expression was determined using oligonucleotide microarrays. The fraction of expressed genes/probe sets was calculated using the detection-call information provided by the AffymetrixsExpression ConsoleTM software. (B) The number and identity of genes/probe sets with more than 3 fold up- (Z3x) or down-regulated transcript levels (r3x) were determined (shown in brackets) and compared between samples obtained 4 and 24 h after irradiation. The numbers of genes with this defined change of expression shared by both samples are shown in the intersection of the two circles. The number of genes the expression of which was elevated Z3 fold 4 h after PDT (compared to non-irradiated cells) and further increased Z3 fold between 4 and 24 h after PDT are shown in C, left panel. The number of genes the expression of which first increased Z3 fold, but, declined between 4 and 24 h Z3 fold after irradiation is shown in C, right panel.

Genes that were identified as up-regulated at 4 h and at 24 h after PDT were used for hierarchical clustering. Untreated as well as treated samples formed separate clusters. No up-regulation of these genes was found in any of the untreated samples. Two obviously different groups of genes were identified, one of which comprises genes that were up-regulated within the first 4 h and a second contains genes which were up-regulated after 4 h (Fig. 3). More detailed analysis of the ‘‘early response’’ genes (already up-regulated after 4 h) by enrichment analysis using DAVID bioinformatics resources demonstrated that genes involved in response to DNA damage, like histone genes, genes involved in protein folding, like heat shock protein genes and various transcription factor genes characteristic for stress response are most prominently up-regulated. In addition, genes the products of which are able to suppress the MAP kinase pathway and genes with immune functions were slightly enriched within the upregulated genes (Tables 1 and 2). At 24 h after irradiation, up-regulated genes are still enriched for genes encoding proteins involved in protein folding, anti-apoptosis, nucleotide assembly, and stress response, but more genes involved in immune regulation, growth factor activity, chemotaxis and small GTPase-mediated signal transduction (Table 3). Down-regulated genes are functional during cell cycle, mitosis and DNA replication indicating that cell proliferation is highly suppressed in cells during recovery from PDT (Table 4).

Regulation of immunorelevant genes A number of cytokine and chemokine genes were found to be up-regulated in PC-3 cells upon PDT. Although first increase of cytokine/chemokine gene expression was observed after 4 h (Table 1) these genes were mainly up-regulated during the rescue phase between 4 and 24 h. After 24 h, a strong increase of the expression of chemokine genes, including CXCL1, CXCL2, CXCL3, CXCL8/IL-8 and CCL26 was

ARTICLE IN PRESS 242

R. Kammerer et al. / Medical Laser Application 24 (2009) 237–246

Fig. 3. Hierarchical clustering of genes with Z3 fold change of expression level induced by low-does PDT in PC-3 cells. Genes/ probe sets exhibiting a Z3 fold increase of expression 4 and 24 h after low-dose PDT (see ‘Materials and method’ section and legend to Fig. 1) were used for hierarchical clustering. Regions corresponding to genes with early (4 h (A)) and late (24 h (B)) activation by PDT are indicated by boxes (left panel) and shown in higher resolution in the right panel.

Table 1.

Transcriptional up-regulation of gene groups 4 h after irradiation (functional group enrichment).

Enrichment score

Number of genes in group

Example genes

Functional annotation

6.71 6.28 6.16 5.08 3.32

13 3 43 7 10

Nucleosome assembly Apoptosis Regulation of transcription Protein folding anti-apoptosis Stress response

2.15 1.39

4 5

Histone 1 H2BJ DNA-damage-inducible transcript 4 ATF3 Heat shock 60 kDa protein 1 (chaperonin) DNAJ (HSP40) homolog, subfamily b, member 1 DUSP1 CXCL2

observed. CXCL1, CXCL2, CXCL3 CXCL8/IL-8 interact with the chemokine receptors CXCR1 and CXCR2, which are specifically expressed by neutrophilic granulocytes. CCL26 binds to the CCR3 chemokine receptor, which is expressed by eosinophilic granulocytes, indicating that prostate cancer cells attract granulocytes during the rescue phase from PDT treatment. Indeed the supernatants harvested from PC-3 cells 24 h after low-dose PDT were able to enhance chemotaxis of peripheral blood granulocytes in an in vitro chemotaxis experiment (data not shown). Remarkably,

Negative regulation of MAP kinase pathway Cytokine/chemokine activity

the up-regulated CXCL chemokines favor also angiogenesis which may be further supported by the upregulation of vascular endothelial growth factor gene (Table 3). In PC-3 cells after PDT the expression of various cytokine genes, including IL-6, IL-11, GM-CSF, and IL-1 was also up-regulated. The encoded cytokines are in general pro-inflammatory, although IL-6 and especially IL-11 may have also anti-inflammatory properties. IL-1, GM-CSF and IL-6 are cytokines which play a pivotal role in the accumulation and activation of granulocytes in inflamed tissues. Expression of IL-6 was

ARTICLE IN PRESS R. Kammerer et al. / Medical Laser Application 24 (2009) 237–246

Table 2.

243

Transcriptional changes 4 h after irradiation (most up-regulated genes).

Gene

Function

Normalized expressiona

Fold changeb

FOS: v-fos FBJ murine osteosarcoma viral oncogene homolog HSPA6: heat shock 70 kDa protein 6 (HSP70B0 ) EGR1: early growth response 1 ATF3: activating transcription factor 3 DUSP1: dual specificity phosphatase 1

Transcription factor

1705

201

Chaperone Transcription factor Transcription factor Suppressor of MAP kinase activation Chaperone Unknown

3708 279 2451 707

108 66.3 54.4 41.8

1755 1317

40.8 40.2

Transcription factor Chaperone Nucleosome assembly

2233 700 857

35 23.8 21.1

HSPD1: heat shock 60 kDa protein 1 (chaperonin) ZFP36: zinc finger protein 36, C3H type, homolog (mouse) JUN: jun oncogene CRYAB: crystalline, alpha B HIST1H2BG: histone cluster 1, H2BG a

Irradiated sample. Fold change, signal intensity of irradiated sample divided by the signal intensity of the untreated control sample.

b

Table 3.

Transcriptional up-regulation of gene groups 24 h after irradiation (functional group enrichment).

Enrichment score

Number of genes in group

4.39 3.34

8 4

3.31 3.31 3.22 3.01 2.62 2.25 1.81 1.46

4 9 57 4 5 8 15 9

Table 4.

Example genes

Functional annotation

Heat shock 60 kDa protein 1 (chaperonin) DNAJ (HSP40) homolog, subfamily b, member 1 Interleukin 6 Histone 1 H2AD EGR1 VEGF CXCL2 Fatty acid 2-hydroxylase CDC-like kinase 1 Rho family GTPase 3

Protein folding anti-apoptosis Stress response Immune regulation; cytokine activity Nucleosome assembly Regulation of transcription Growth factor activity Cytokine/chemokine activity Endoplasmic reticulum Protein kinase activity Small GTPase-mediated signal transduction

Transcriptional down-regulation of gene groups 24 h after irradiation (functional group enrichment).

Enrichment score

Number of genes in group

Example genes

Functional annotation

7.79 3.86 1.43

37 8 28

Kinesin family members DNA replication complex GINS protein PSF1 Transcription factor 19 (SC1)

Mitosis/cell cycle DNA replication Transcription

further verified by detection of IL-6 protein in the supernatant of irradiated PC-3 cells using the cytometric bead assay. While less than 75 pg/ml IL-6 was detected in the supernatant of untreated PC-3 cells after 24 and 48 h of culture, the supernatants of PC-3 cell cultures treated by PDT contained 588 and 235 pg/ml IL-6 after 24 and 48 h of recovery, respectively. Interestingly, expression of cytokine receptor genes, including IL-6 receptor and IL-1 receptor-like receptor genes was also up-regulated in PC-3 cells (Table 5).

Discussion Although the main goal of PDT in cancer therapy is to eradicate cancer cells completely by the primary therapy, PDT is more recently also applied to cancer patients with more advanced-stage disease. For these patients a curative treatment based on the phototoxic effect alone is not expected because of infiltrative tumor growth and systemic tumor cell dissemination. Therefore, secondary antitumor effects (e.g. anti-tumor

ARTICLE IN PRESS 244

Table 5.

R. Kammerer et al. / Medical Laser Application 24 (2009) 237–246

Upregulated cytokines and chemokines.

Functional group

Gene

Cytokine

IL-6 IL-11 IL-1 GM-CSF

Cytokine receptor

IL-1 receptor-like IL-6 receptor

Chemokine

CXCL2 CXCL3 CCL26 CXCL1 CXCL8/IL-8

a

Fold changeb after 4 h recovery

Fold changeb after 24 h recovery

136 251 87 161

2.2 1.7 2.4 1.6

8.2 6.4 11.5 10.8

92 334

5 1.4

5.8 4.8

697 169 1041 283 1609

3.8 3.7 7.7 1.4 1.4

19.2 10.4 11.1 3.9 8.2

Normalized expressiona

Normalized expression of treated cultures after 24 h recovery. Fold change, signal intensity of irradiated sample divided by the signal intensity of the untreated control sample.

b

immunity) are very important. The present study was undertaken to better understand the influence of cancer cells, which may escape from PDT due to the application of an insufficient light dose and/or photosensitizer concentration, on the tumor microenvironment. Using a global approach, we analyzed the transcriptome of PC-3 cells, a commonly used PCa-cell line, after low-dose PDT treatment. Low-dose PDT was defined as an irradiation regime, which allows 75% of the cells to survive 24 h after PDT treatment based on cell viability measurements. The early response of the cells was dominated by an increase of gene transcripts the encoded proteins of which reduce the toxic effect of oxidative stress very similar to the transcriptional changes of high-dose PDT as previously reported [18]. In addition, similar results were reported using other cell lines and photosensitizers, indicating that there are only minor differences in the early response of cancer cells to various PDT regimes [19–23]. However, there are a significant number of genes (Fig. 2), the expression of which was not up-regulated after 4 h but strongly after 24 h post PDT treatment. These genes encode aldo-keto reductases (detoxification) and cyclin-dependent kinase inhibitors (cell cycle arrest), genes which may limit the toxic effects of PDT. Interestingly, transcripts of various interleukin, chemokine and growth factor genes also increased, which may induce or favor inflammation. It is well-known that tumors are infiltrated by a large number of granulocytes upon PDT [24]. More recently it was demonstrated that these infiltrating granulocytes enhance the anti-tumor immunity [25]. Here we show that low-dose PDT induces the expression of genes encoding granulocyte-attracting chemokines in PCacells, indicating that tumor cells which escape from PDT actively participate in the recruitment of granulocytes and thereby in the enhancement of anti-tumor immu-

nity. However, these chemokines belong to the ELR motif-containing CXCL chemokines which also have angiogenic properties. In addition we found that prostate cancer cells up-regulate the expression of the vascular endothelial growth factor (VEGF) gene. VEGF may synergize with the angiogenic chemokines, leading to a re-growth of tumor vasculature, thereby enhancing the nutrition for surviving tumor cells. On the other hand, VEGF is thought to have also immune suppressive properties in the tumor microenvironment [26]. In addition, pro-inflammatory cytokines which have the capability to activate granulocytes like IL-1, IL-6 and GM-CSF are also secreted by prostate cancer cells upon PDT. These cytokines may overcome the immunosuppressive effects mentioned above. Recent findings indicate that granulocytes play an important role for the transport of antigens from the periphery to lymph nodes [27]. There, granulocytes may activate dendritic cells which in turn could stimulate the generation of tumor antigen-specific T cells. Such a mechanism may be responsible for the findings of Kousis et al. [25] that neutrophils regulate antigen-specific anti-tumor immunity. It is intriguing that tumor cells favor anti-tumor immunity upon PDT, but possibly this is the price paid by the tumor cells to escape from cell death induced by oxidative stress. In an effort to appraise the overall effect of PDT on tumor cells, it is important to consider that tumor cells not only express cytokines like IL-6 but also its corresponding receptor. Therefore, a function of an autocrine growth factor of IL-6 has to be taken into consideration. However, genes which were down-regulated after 24 h are involved in cell cycle and mitosis indicating that although these cells can escape from PDT they were forced to stop cell proliferation at least as long as repair mechanisms are active. Taken together, our study lends further support to the idea that PDT

ARTICLE IN PRESS R. Kammerer et al. / Medical Laser Application 24 (2009) 237–246

may favor anti-tumor immunity and that combination of PDT and immunotherapy may be a potent new therapeutic strategy.

Acknowledgements We would like to thank T. Haferlach and his team at the MLL Mu¨nchner Leuka¨mielabor GmbH for hybridization and read-out of the oligonucleotide microarrays. The work was carried out with support of the Deutsche Krebshilfe (107320).

Zusammenfassung Molekulare Grundlagen der Resistenz von Prostatakarzinomzellen gegenu¨ber einer limitierten ProtoporphyrinIX vermittelten photodynamischen Therapie Einleitung: Bei der photodynamischen Therapie (PDT) von soliden Tumoren wird eine Kombination aus einer photosensibilisierenden Substanz und Licht benutzt, um eine selektive Zersto¨rung der Tumorzellen zu erzielen. Fu¨r Tumore im Fru¨hstadium oder lokalisierte Tumorerkrankungen, kann die PDT eine kurative Therapie darstellen, die viele Vorteile gegenu¨ber bestehenden Behandlungsmethoden hat. Bei fortgeschrittenen oder disseminierten Erkrankungen ist eine kurative Therapie in der Regel nicht mo¨glich, es kann aber durchaus eine Verbesserung der Lebensqualita¨t bzw. eine Lebensverla¨ngerung erreicht werden. In ju¨ngster Zeit ha¨ufen sich Hinweise, dass durch PDT auch systemische Tumorimmunita¨t induziert werden kann. Deshalb kann PDT durchaus eine sinnvolle Therapieoption darstellen, auch wenn es zuna¨chst nicht gelingt, die Tumorzellen vollsta¨ndig zu eliminieren. Um bei einem solchen Vorgehen die Gefahr einer Rezidivbildung zu minimieren, sollte bekannt sein, wie Tumorzellen mit einer durch PDT verursachten Scha¨digung umgehen. Material und Methoden: Wir haben die Antwort von humanen PC-3 Prostatakarzinomzellen auf eine sublethale Bestrahlung nach 5-Aminola¨vulinsa¨ure-induzierter Sensibilisierung mittels Protoporphyrin-IX auf Transkriptionsebene mit Affymetrix HG U133 Plus 2.0 Oligonukleotid-Mikroarrays charakterisiert. Die Zellen wurden mit Laserlicht einer Wellenla¨nge von 635 nm und einer Bestrahlungssta¨rke von 100 mW/cm2 und einer Lichtdosis von 1,5 bzw. 3 J/cm2 bestrahlt. Ergebnisse: Die fru¨he Antwort 4 h nach PDT war durch die Hochregulation von early response-Genen wie FOS, JUN, EGR1, ATF3, DUSP, sowie Hitzeschock- und Histongenen bestimmt, wie dies in a¨hnlicher Weise auch fu¨r die Hochdosis-PDT bekannt ist. Allerdings blieben dabei irreversible Zellscha¨den aus. 24 Stunden nach der PDT exprimierten die Zellen immer noch große Mengen

245

der early response-Gene, aber zusa¨tzlich waren 235 weitere probe sets-Gene signifikant hochreguliert (Z3x). Die am sta¨rksten hochregulierten Gene geho¨rten zu Genfamilien, die fu¨r Aldo-keto-Reduktasen, Fibroblasten-Wachstumsfaktoren und HSP40-verwandte Proteine kodieren. Im Hinblick auf eine mo¨gliche Anti-Tumorimmunreaktion ist es bemerkenswert, dass die Expression einer Vielzahl von Chemokin- und Zytokingenen durch PDT gesteigert wurde, die meisten davon spielen eine Rolle bei der Granulozyten-Rekrutierung und -Aktivierung und einige sind fu¨r die Angiogenese wichtig. Schlussfolgerung: Zusammenfassend kann man sagen, dass die umfassende molekulare Charakterisierung der Tumorzellreaktion auf PDT andeutet, dass PDT eher eine Anti-Tumorimmunantwort als Tumor-EscapeMechanismen unterstu¨tzt. Deshalb ko¨nnte die Kombination von PDT und Immuntherapie eine vielversprechende multimodale Tumortherapie werden. Schlu¨sselwo¨rter: Photodynamische Therapie; Prostatakarzinom; 5-Aminola¨vulinsa¨ure; Oligonukleotid-Mikroarrays; Immuntherapie

References [1] Gomella LG, Johannes J, Trabulsi EJ. Current prostate cancer treatments: effect on quality of life. Urology 2009;73(5 Suppl.):S28–35. [2] Moore CM, Pendse D, Emberton M. Photodynamic therapy for prostate cancer – a review of current status and future promise. Nat Clin Pract Urol 2009;6(1):18–30. [3] Castano AP, Mroz P, Hamblin MR. Photodynamic therapy and anti-tumour immunity. Nat Rev Cancer 2006;6(7):535–45. [4] van Duijnhoven FH, Aalbers RI, Rovers JP, Terpstra OT, Kuppen PJ. The immunological consequences of photodynamic treatment of cancer, a literature review. Immunobiology 2003;207(2):105–13. [5] de Graaf YG, Kennedy C, Wolterbeek R, Collen AF, Willemze R, Bouwes Bavinck JN. Photodynamic therapy does not prevent cutaneous squamous-cell carcinoma in organ-transplant recipients: results of a randomizedcontrolled trial. J Invest Dermatol 2006;126(3):569–74. [6] Gollnick SO, Evans SS, Baumann H, Owczarczak B, Maier P, Vaughan L, et al. Role of cytokines in photodynamic therapy-induced local and systemic inflammation. Br J Cancer 2003;88(11):1772–9. [7] Kabingu E, Vaughan L, Owczarczak B, Ramsey KD, Gollnick SO. CD8+ T cell-mediated control of distant tumours following local photodynamic therapy is independent of CD4+ T cells and dependent on natural killer cells. Br J Cancer 2007;96(12):1839–48. [8] Andrews NW. Membrane repair and immunological danger. EMBO Rep 2005;6(9):826–30.

ARTICLE IN PRESS 246

R. Kammerer et al. / Medical Laser Application 24 (2009) 237–246

[9] Beg AA. Endogenous ligands of Toll-like receptors: implications for regulating inflammatory and immune responses. Trends Immunol 2002;23(11):509–12. [10] Seong SY, Matzinger P. Hydrophobicity: an ancient damage-associated molecular pattern that initiates innate immune responses. Nat Rev Immunol 2004;4(6):469–78. [11] Casas A, Perotti C, Fukuda H, del C Batlle AM. Photodynamic therapy of activated and resting lymphocytes and its antioxidant adaptive response. Lasers Med Sci 2002;17(1):42–50. [12] Gollnick SO, Liu X, Owczarczak B, Musser DA, Henderson BW. Altered expression of interleukin 6 and interleukin 10 as a result of photodynamic therapy in vivo. Cancer Res 1997;57(18):3904–9. [13] Schroeder A, Mueller O, Stocker S, Salowsky R, Leiber M, Gassmann M, et al. The RIN: an RNA integrity number for assigning integrity values to RNA measurements. BMC Mol Biol 2006;7:3. [14] Li C, Wong WH. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci USA 2001;98(1):31–6. [15] Shedden K, Chen W, Kuick R, Ghosh D, Macdonald J, Cho KR, et al. Comparison of seven methods for producing Affymetrix expression scores based on false discovery rates in disease profiling data. BMC Bioinform 2005;6:26. [16] Dennis Jr G, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, et al. DAVID: database for annotation, visualization, and integrated discovery. Genome Biol 2003;4(5):P3. [17] Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 2009;4(1):44–57. [18] Ruhdorfer S, Sanovic R, Sander V, Krammer B, Verwanger T. Gene expression profiling of the human carcinoma cell line A-431 after 5-aminolevulinic acidbased photodynamic treatment. Int J Oncol 2007;30(5): 1253–1262.

[19] Buytaert E, Matroule JY, Durinck S, Close P, Kocanova S, Vandenheede JR, et al. Molecular effectors and modulators of hypericin-mediated cell death in bladder cancer cells. Oncogene 2008;27(13):1916–29. [20] Cekaite L, Peng Q, Reiner A, Shahzidi S, Tveito S, Furre IE, et al. Mapping of oxidative stress responses of human tumor cells following photodynamic therapy using hexaminolevulinate. BMC Genomics 2007;8:273. [21] Prasmickaite L, Cekaite L, Hellum M, Hovig E, Høgset A, Berg K. Transcriptome changes in a colon adenocarcinoma cell line in response to photochemical treatment as used in photochemical internalisation (PCI). FEBS Lett 2006;580(24):5739–46. [22] Verwanger T, Sanovic R, Aberger F, Frischauf AM, Krammer B. Gene expression pattern following photodynamic treatment of the carcinoma cell line A-431 analysed by cDNA arrays. Int J Oncol 2002;21(6):1353–9. [23] Wild PJ, Krieg RC, Seidl J, Stoehr R, Reher K, Hofmann C, et al. RNA expression profiling of normal and tumor cells following photodynamic therapy with 5-aminolevulinic acid-induced protoporphyrin IX in vitro. Mol Cancer Ther 2005;4(4):516–28. [24] Cecic I, Stott B, Korbelik M. Acute phase responseassociated systemic neutrophil mobilization in mice bearing tumors treated by photodynamic therapy. Int Immunopharmacol 2006;6(8):1259–66. [25] Kousis PC, Henderson BW, Maier PG, Gollnick SO. Photodynamic therapy enhancement of antitumor immunity is regulated by neutrophils. Cancer Res 2007;67(21):10501–10. [26] Frumento G, Piazza T, Di Carlo E, Ferrini S. Targeting tumor-related immuno-suppression for cancer immunotherapy. Endocr Metab Immune Disord Drug Targets 2006;6(3):233–7. [27] van Gisbergen KP, Geijtenbeek TB, van Kooyk Y. Close encounters of neutrophils and DCs. Trends Immunol 2005;26(12):626–31.