Altered Expression of Interleukin 6 and Interleukin 10 As a Result of Photodynamic Therapy in Vivo1

[CANCERRESEARCH57.3904-3909. September 5. 997) Advances in Brief

Sandra 0. Gollnick, Xiaonan Liu, Barbara Owczarczak, David A. Musser, and Barbara W. Henderson2

Departments ofMolecular Medicine (S. 0. G.J, Radiation Biology (B. 0.. X. L. B. W. H.), and Dermatology (D. A. M.), Roswell Park Cancer Institute, Buffalo, New York 14263

Abstract hines. Here we report that PDT can alter the expression of IL-6 and IL- 10 in tumor and normal tissues in vivo and propose that these Photodynamic therapy (PDT), which can effectively destroy malignant changes are mechanistically involved in the observed effects of PDT tissue, also induces a complex Immune response that potentiates antitumor on host immune responses. immunity but also inhibits skin contact hypersensitivity (CHS) and pro longs skin graft survival. The underlying mechanisms responsible for Materials and Methods these effects are poorly understood but are likely to involve mediation by cytokines. We demonstrate in a BALB/c mouse model that PDT delivered Animals and Tumor System. BALB/cJ mice, obtained pathogen-free to normal and tumor tissue in vivo causes marked changes in the expres from The Jackson Laboratory (Bar Harbor, ME), were used for all experi sion of cytokines Interleukin (IL)-6 and IL-b but not tumor necrosis ments, either tumor-free or carrying the EMT6 mammary tumor (8). Animals factor a. IL.6 mRNA and protein are strongly enhanced in the PDT were housed in microisolator cages in a laminar flow unit under ambient light. treated EMT6 tumor. PDT also Increased IL-ti mRNA in exposed spleen Six- to 12-week-oldanimals were inoculatedintradermallyon the shoulder and skin. These data suggest that the general inflammatory response to with 2 X l0@ tumor cells harvested from exponentially growing cultures. PDT may be mediated at least in part by IL-6. In addition, IL-6 may Althoughthe EMT6 cell line is highly antigenic and significantly immuno modulate the local antitumor immune response. In contrast, IL-lO mRNA genic (9), no spontaneous tumor regressions were observed in this study. Prior in the tumor decreases following PDT. Most importantly, IL-lO is mark to tumor inoculation and/or light treatment, all hair was removed from the edly induced in the skin of mice exposed to a PDT regime that strongly prospective treatment site by shaving and depilation. Tumors were used for Inhibits the CHS response, and the kinetics of IL-b induction coincide experimentationabout10days afterinoculation,when they hadreacheda size with the known kinetics of CHS inhibition. We propose that the enhanced of 6â€”8mmin diameter. IL-b expression plays a role in the observed suppression of celi.mediated Photosensitizer. Clinical-grade, pyrogen-free Photofrmn (QLT PhotoThera responses seen following PDT. peutics Inc., Vancouver, British Columbia, Canada) was reconstituted to 2.5 mg/ml in pyrogen-free5% dextrose(D5W; BaxterCorp.,Deerfield, IL). Introduction In Vivo PDT Treatment. Three treatment protocols were used: (a) Animals were given iv. injections via the tail vein of 5 mg/kg Photofrmn. PDT3 has obtained regulatory approvals in the United States and Twenty-four h later, 630 nm light was delivered with an argon-dye laser elsewhere as a cancer treatment modality that can selectively destroy system (Spectra Physics, Mt. View, CA). A treatment field I cm in diameter, malignant, premalignant, and certain benign lesions in patients (I). containing either tumor or normal skin, was illuminated at 75 mW/cm2 for a The cytotoxic effect of this therapy is mediated primarily by the total light dose of 100 J/cm2(approximateTCD@,IJ).Atselected time intervals formation of singlet oxygen, which is generated by energy transfer following light exposure, tumors, spleens, and/or normal skin were harvested. from a light-activated, tissue-localized photosensitizer to ground state Tissues were kept frozen at â€”70Â°Cuntilfurther processed. Tumors and spleens for flow cytometric analysis were harvested and processed immediately. oxygen, and which can cause oxidative damage to numerous cellular (b) Animals were treated as above except that a subcurative light dose (34 components at or near the site of its generation. Tumor eradication is i/cm2) was given. a consequence of direct photodamage to the malignant cells and of (c) Animals were injected as above with 10 mg/kg Photofrin. Six h later, damage of the tumor-supporting microvasculature (2, 3). In its acute their shaved backs were illuminated with a bank of blue fluorescent lights phase, the PDT response is of a pronounced inflammatory nature, (Sylvania FI5T8IB) emitting between 350 and 650 nm (with a maximum at characterized by the release of vasoactive, inflammatory mediators â€”450nm) at a distance of I 8 cm. A 2-mm glass plate was placed 10 cm above and accumulation of host immune cells at the treatment site (2â€”4). the mice to removeany UV radiation.Theanimalsreceiveda lightdose of I.3 Long-term tumor control appears to be aided by a potentiation of J/cm2 (0.745 mW/cm2 for 30 mm). Mice were sacrificed at selected time points antitumor immunity (4). However, preclinical evidence also demon after illumination, and treated skin was removed, frozen in liquid nitrogen, and stored at â€”70Â°Cuntilprocessed. All experiments included control animals strates that PDT exposure can significantly suppress certain immune without treatment, with photosensitizer only, and with light only. functions, such as the CHS response (5, 6) and murine skin graft In Vitro PDT Treatment. EMT6 cells were grown in basal media eagle rejection in allogeneic recipients (6, 7), possibly by interfering with (BME) supplemented with 15% FBS and antibiotics (all from Life Technol the function of antigen presenting cells. It is likely that these dichot ogies, Inc., Grand Island, NY) in a humidified atmosphere of 5% CO2 in air. omous responses involve the mediation of immune-modulating cyto Exponentially growing cells were exposed to 10 @&g/mIPhotofrmnincomplete medium for 24 h, followed by exposure to drug-free complete medium for 3 h. Received 7/2/97; accepted7/29/97. Cells were removed from culture plates with trypsin-EDTA (Life Technolo The costs of publicationof this articlewere defrayedin partby the paymentof page gies, Inc.), washed once with complete medium, and placed in Petri dishes to charges. This article must therefore be hereby marked advertisement in accordance with avoid attachment. They were then exposed to laser light as above for a total 18 U.S.C. Section 1734 solely to indicate this fact. light dose of 0.75 J/cm2 (â€”LD@).At selected time intervals thereafter, cells I This work was supported by NIH Grant CA5579lb (to B. W. H.). S. 0. G. was partially supported by NIH Grant HDl7Ol 3 and a grant from the Elisabeth and Peter werecollected.Cell viabilitywas assessedby trypanblueexclusionandranged Tower Foundation.D. A. M. was supportedby NIH GrantCA5579lc. from50 to 70%.Cells werespundown,andpelletswerekeptfrozenat â€”70Â°C. 2 To whom requests for reprints should be addressed, at Department of Radiation RNAIsolationand RT-PCR. TotalRNAwasisolatedfromeithercellsor Biology, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263. tissues with Ultraspec (Biotecx Laboratories, Inc., Houston, TX) using a 3 The abbreviations used are: PDT, photodynamic therapy; CHS, contact hypersensi tivity; IL, interleukin; RT-PCR, reverse transcription-PCR; TNF, tumor necrosis factor; modification of the method of Chomczynski and Sacchi (10). RNA was FACS, fluorescence-activatedcell sorting;MoAb, monoclonalantibody. quantified spectrophotometrically and diluted to a final concentration of 2 3904

@.tg/ml.RT-PCRwas performed as described earlier, with slight modifications gen Gr-l (PharMingen). Appropriate immunoglobulin isotypes were used as (11). Briefly, first-strand cDNA synthesis was performed with 2 @goftotal controls. Staining was performed on ice. Twenty p1 of a mouse IgG (Sigma) RNA in 50 mMTris-HC1(pH 8.3), 75 mMKC1,3 mMMgCl2, 10mMDTF, 500 solution (0.1 mg/mI) was added for 20 mm to block Fc receptor. Ten @dof p.M deoxynucleotide triphosphate (Perkin-Elmer, Foster City, CA), 50 @g/ml appropriately titrated MoAb solutions were added for 15 mm. Cells were then random hexamer (Promega Corp., Madison, WI), and 10,000 units/ml Super washed once and resuspended in 500 p1 Ho-PAB (2 @.tWmlHoechst3342 in Script II reverse transcriptase (Life Technologies, Inc.) in a total of 20 @lfor PBS, pH 7.0, 0.5%azide,and I%BSA containing1%ultrapureformaldehyde 1 h at 37Â°C.Followingfirst-strand synthesis, the reaction was placed on ice, (Polysciences, Inc., Warrington, PA) until flow cytometric analysis. and 5 pi were used for PCR analysis ofcytokine RNA. PCR analysis was done For flow cytometric analysis a two-laser FACStar Plus (Becton Dickinson, using 1 @.LMofcytokine-specific primers (Stratagene, La Jolla, CA) in 10 mr@i San Jose, CA) flow cytometer was used, operating in the UV and at 488 nm. Tris-HC1, 50 mM MgCl2, 0.2 m@vtdeoxynucleotide triphosphates, 0.1 Four colors and light-scattering properties could be resolved employing 420/ @Ci[a-32PJdCTP, and 2 units of Taq polymerase (Boehnnger Mannheim, 20, 530/30, and 575/30 band pass filters and a 640 long pass filter. Data were Indianapolis, IN). Following a 5-mm denaturation step at 95Â°C and a 5-mm acquired from 5,000 cells, stored in collateral list mode, and analyzed using the annealing step at 60Â°C,atotal of 20â€”30cycleswas performed with each cycle WinList processing program (Verity Software House, Inc., Topsham, ME). consisting of 1.5 mm at 72Â°C,45 s at 94Â°C,and 45 s at 60Â°C,with a final Results extension of 10 mm at 72Â°C.Allamplifications were done in the linear range of the assay; actin and TNF-a reactions were amplified for 20 cycles, whereas PDT Alters Cytokine mRNA Expression in EMT6 Cells in IL-6 and IL-b reactions were amplified for 30 cycles. The reaction products Vitro, EMT6 Tumors, and Spleen. Initial analysis showed that of were separated on a 6% acrylamide gel, dried, and exposed to KOdak X the cytokines examined in EMT6 cells in vitro (IL- 1, IL-3, IL-6, OMATfilm. ControlPCRreactionswereperformedwithâ€œcDNAâ€•synthesized IL-lO, IFN-'y, granulocyte/macrophage-colony-stimulating factor, in the absence of reverse transcriptase. Reagents were routinely checked for contamination. Results were confirmed in at least three replicate experiments. TNF-a, and transforming growth factor @3;data not shown) and in Western Analysis. Single-cell suspensions of tumors were prepared by tumors in vivo (same cytokines as above, IL-2 and IL-4; data not enzymatic digestion of the excised tumor (8), and IL-6 protein was determined shown), mRNA levels for IL-6 and IL-lO were most markedly as follows. Cell suspensions were lysed in 0.2% NP4O, 10 miviHEPES (pH changed by PDT. Therefore, this work was focused primarily on IL-6 7.9), 1 mM EDTA, 60 mM KC1, 1 mr@iDTF, S pg/mi each pepstatin, aprotinin, and IL-lO. TNF-a was investigated in parallel because it is a recog and leupeptin for 10 mm on ice (all from Life Technologies, inc.). A total of nized IL-6 inducer and acts synergistically with IL-6 to induce anti 2.5 @gofprotein per sample was separated on a 12.5% SDS polyacrylamide tumor responses in mice (14, 15). gel in the presence of @3-mercaptoethanolaccordingto Laemmli (12). Follow Fig. 1 shows the mRNA expression of IL-6, IL-lO, and TNF-a as ing transfer to nitrocellulose, Western analysis was performed using biotiny a function of time after PDT. EMT6 cells in vitro responded with a lated antimouse IL-6 (PharMingen, San Diego, CA) and developed using an transient enhancement of IL-6 mRNA expression, which peaked at Immuno-Stain streptavidin-biotin-alkaline phosphatase kit according to the manufacturer's instructions (Bio-Rad, Hercules, CA). Results were confirmed 1â€”3h after PDT exposure (Fig. 1A). IL-lO mRNA was not expressed in at least three experiments. in the cells in vitro, either constitutively or PDT induced. In contrast ELISA Assays. IL-b protein in PDT-treatedskin was determinedby to IL-6, TNF-a mRNA remained largely unchanged after PDT in ELISA as recommended by the manufacturer (PharMingen). Briefly, skin vitro. samples were frozen in liquid nitrogen, mechanically ground, and prepared for In the tumor, IL-6 mRNA was not or was minimally present in protein extraction according to Liu et a!. (13). Protein was extracted as untreated control tumors but was found to be slightly enhanced after described above. Wells of a 96-well microtiter plate were coated with 2 @g/ml exposure to photosensitizer alone (Fig. 1B). Between 1 and 6 h after capture antibody (rat antimouse IL-lO, JES5â€”l6E3; PharMingen) in 0.1 M treatment (100 i/cm2), IL-6 mRNA expression was strongly and NaHCO3 (100 s.d/well) and incubated overnight at 4Â°C.Plateswere washed consistently enhanced, and this enhancement lasted for at least 24 h. twice, 5 mm/wash, with 0.02% Tween 20 in PBS and blocked for 2 h at room The results with IL-lO were more variable, with occasional control temperature with 10% FCS in PBS (blocking buffer). Duplicate samples (10 samples not expressing any IL-lO mRNA. The majority of tumors, ILg of total protein) or standards (recombinant IL-b) were added in 100 @.dof blocking buffer and incubated overnight at 4Â°C.Plateswere washed twice and however, showed IL-lO mRNA decreasing with time after PDT, as incubated with 2 @.tg/mldetectionantibody (biotinylated rat antimouse IL-lO, shown in Fig. lB. TNF-a mRNA was expressed at high levels in SXC-l; PharMingen),100s.d/well,andincubatedfor2 h at roomtemperature. control tumors but showed no changes with time after PDT. Very Plates were washed four times, and 100 pi/well streptavidin-horseradish per similar results were obtained after subcurative treatment (data not oxidase (Bio-Rad) was added for 30 mm. Plates were washed six times, shown). and 100 g.tl/wellsubstrate (2,2'-azino-di[3-ethylbenzthiazoline sulfonate(6)J Spleen cell populations harvested from animals after exposure to diammonium salt; Boehringer Mannheim) were added. Absorbance was read at Photofrin only or PDT treatment of their shoulder tumors also showed 495 nm. The assay was carried out at least twice, using duplicate wells. weak, transient IL-6 mRNA induction (Fig. 1C). IL-lO mRNA was Flow Cytometry. The cell populations present in EMT6 tumors and also transiently induced by PDT (lOOJ/cm2 to shoulder tumor), peak spleens before and after PDT were characterized through FACS analysis, using ing at 3 h after treatment. Again, TNF-a mRNA was unchanged. Hoechst 33342 staining to assess ploidy (all EMT6 cells are >diploid) and panels of MoAbs to detect specific cell surface antigens. Briefly, tumors were Very similar results were obtained in spleens from animals after disaggregated in 25 ml of HBSS containing 50 mg collagenase, type II PDT was given to a 1-cm diameter area of skin at the shoulder without (Worthington Biochemical Corp., Freehold, NJ), and 500 mg of BSA. Spleens the presence of a tumor (data not shown). When a comparable area of were pushed through a tissue sieve, and cells were collected in PAB buffer skin at the right hind thigh was exposed to PDT, no cytokine induction (HBSSw/oCa2@Mg2@containing5 mg/mlBSAand 1mg/mlsodiumazide). was observed in the spleen. All cell preparations were carried out in subdued light to avoid unwanted PDT Induces IL.6 Protein in EMT6 Tumors. These experiments photodynamic cell damage. Cell viability, as determined by trypan blue cx used subcurative treatment (34 J/cm2) to avoid rapid tumor destruc clusion, was >90%. Cells were pelleted and resuspended in prewarmed tion and allow for tissue to be harvested at delayed time points. IL-6 complete medium without (for assessment of autofluorescence) and with protein was not detectable by Western blotting in untreated control Hoechst 33342 (Sigma Chemical Co., St. Louis, MO; final concentration, 3 tumors or tumors exposed to Photofrin alone in vivo (Fig. 2). A weak @.&g/ml).MoAbsconjugatedwithfluorescein,phycoerythrin,orRED613 were IL-6 band can be observed for the sample exposed to light treatment used to directly quantify cells expressing the common leukocyte antigen CD45 (Life Tecnologies, Inc.), T cell antigens CD4 and CD8 (Caltag, San Francisco, only and harvested 24 h later. Strong induction of IL-6 protein was CA; PharMingen), granulocyte and macrophage antigen Mad (Caltag), evident by 3 h after PDT and persisted for at least 48 h but had macrophage antigen F4/80 (Serotec, Washington, DC), and granulocyte anti disappeared by 6 days after PDT. 3905

CYTOKINE EXPRESSION AFTER PDT

small percentage of lymphocytes (mostly CD4@ T cells) remaining A Timepost-PDT(h) largely unchanged. The flow cytometric analysis in Fig. 3B illustrates for a representative sample the marked accumulation of granulocytes C Ph 0 1 3 6 24 (Mad @,Gr-l @)rising from â€”3%to over 50% of recovered cells, IL-6 and the approximate tripling in the percentage of macrophages (Mad @,Gr-F) over a 24-h period following PDT. No changes were IL-I 0 observedin the compositionof spleencell populationsafter PDT to the shoulder region. TNF- cx PDT Induces Expression of IL-lO mRNA and Protein in Skin. @ .S Skin samples examined after PDT with 630-nm light exposure also showed transient induction of IL-6 and IL-lO. Although the IL-6 Actin ...... induction kinetics were similar to the tumor, IL-lO induction appeared delayed, with IL-lO mRNA becoming detectable between 24 and 48 h (data not shown). Because of the known involvement of IL-lO in the immunosuppressive effects of UV irradiation (16) and the observed suppression by PDT of the skin CHS responses (5, 17), IL-lO expres sion was also tested after a PDT protocol (10 mg/kg Photofrmn, blue B light, 1.3 J/cm2), which has been shown to induce CHS suppression in mice (17). IL-lO mRNA in control skin samples was absent but was strongly induced at 72 and 96 h after light exposure (Fig. 4A). IL-lO @ C Ph0136 18 24 protein also was detectable at 72 h and progressively increased up to IL-6 at least 120 h after treatment(Fig. 4B).

iL-lO Discussion This study demonstrates for the first time in an experimental in vivo TNF- a systemthat PDT can significantly affect the expressionof the cyto kines IL-6 and IL-lO. We confirm the enhancement of IL-6 in cells Actin afterPDTinvitrodescribedearlierbyKicketa!.(18)andshowthat IL-6 mRNA levels are strongly enhanced in tumor and normal tissues after PDT in vivo. This is accompanied, at least in tumor (other tissues were not tested), by a prolonged increase in IL-6 protein. As sug gested by Kick et a!. (18), TNF-ct does not seem to play a role in the IL-6 induction by PDT because the changes in that cytokine are C neither preceded nor accompanied by similar changes in TNF-a. TNF-a has been found induced by PDT in murine peritoneal macro C Ph 0 1 3 6 24 phages in vitro (19), and a recent study by Anderson et a!. (20) has IL-6 demonstratedup-regulation of TNF-a in keratinocytesin vitro by PDT using a phthalocyanine-derived photosensitizer. The absence of @ â€”.- TNF-a changes in our study might be related to the mouse strain used IL-i 0 astheregulatoryregionoftheTNF-csgenehasbeenshowntohave allelic differences (21). The TNF-a allele expressed in BALB/c mice @ â€”. . . is resistant to up-regulation by UV radiation and may also be resistant to alterations induced by PDT. Actin It remains to be determined whether the enhanced generation of Fig. I. Modulation of IL-6, IL-b, and TNF-a mRNA expression by PDT. In A, EMT6 IL-6 plays a role in the PDT tumor response. Intratumoral injection of cells in vitro were treated with 10 @g/mlPhotofrinfor 24 h in complete medium, followed IL-6 or transduction of the JL-6 gene into tumor cells can enhance by 3 h of drug efflux in complete medium, and illumination at 630 nm with 0.75 J/cm2. tumor immunogenicity and inhibit tumor growth in experimental In B, EMT6 tumors were treated in vivo at 630 nm with 100 J/cm2 after animals were exposed to 5 mg/kg Photofrin for 24 h. In C, spleens were harvested after exposure of shoulder tumors as in B. Total RNA was isolated from cells or tissues harvested immediately or at 1, 3, 6, 18, and/or 24 h after PDT, and 2 @zgfrom each sample were subjected to RT-PCR analysis as described in â€œMaterialsandMethods.â€•Controls(C) were Time post-PDT(h) untreated samples; Photofrin controls (Ph) were samples exposed to drug only. C Ph LtiO 3 6 24 486d1 kDa

@ Changes in Tumor and Host Cell Populations in EMT6 Tumors ...,,;@ .@ â€˜ .. â€”31.0 @215 after PDT. Because both tumor cells and tumor-infiltrating host cells @@ I L - 6 â€”0'. 7 â€œ @4 .. may contribute to the overall changes in cytokine expression, we â€˜ .. . -@ â€” 14.4 examined by FACS analysis the time-dependent changes in the var ious cellular tumor components after PDT (5 mg/kg Photofrin, 100 Fig. 2. Western analysis of IL-6 in EMT6 tumor by PDT as a function of time after i/cm2). Fig. 3A illustrates the changes in overall tumor composition treatment. Tumors were exposed to 5 mg/kg Photofrin and 34 J/cm2 and excised at 0, 3, 6, 24, 48 h, and 6 days after treatment Cell suspensions were prepared by enzymatic occurring between the time immediately after PDT exposure (control digestion, and whole-cell lysates were subjected to Western blotting; 2.5 @gproteinper tumors showed virtually identical results) and 24 h after PDT. The sample were resolved on 12.5% SDS-PAGE, transferred to a nitrocelluiose membrane, most obvious changes are the relative expansions of macrophages and, and probed with a biotinyiated antimouse IL-6 antibody. Controls included untreated samples (C'), samples exposed to Photofrmn alone (Ph), and light alone (Ii; harvested 24 h especially, granulocytes, at the expense of tumor cells, with the very after exposure). 3906

tumorcells @ macrophages EffOgranulocytes 3% D lymphocytes . diploid,CD45- 5minpostPDT 24hpostPDT

Fig. 3. Changes in the cell composition of EMT6 tumors determinedby harvesting the tissue after B PDT.TumorsweretreatedwithPDTas described Control 5 Minutes 5 Hours 24 Hours in Fig. I and analyzed by FACS analysis. In A, mean values from three independent experiments of the percentages of the different cell phenotypes comprisingthe tumorisolate were plotted.Tumor cells were identified as >diploid, CD45, macro phages as MacI', Gr-1, granulocytes as MacI, .3 Gr-i@, and lymphocytes as Maci +, CD4@, or CD8'. Diploid, CD45@cells represent fibroblasts and endotheial cells. In B, scattergrams of repre sentative tumor samples obtained before or at 5 miis,5 h, or 24 h after PDT as described in Fig. 1, showing PE-MacI versus FI'FC-Gr-l staining (up per panels) and the appropriate isotype controls (lower panels), are shown. I

murine tumor systems (15, 22, 23). Thus, PDT may enhance local the dark. Kick et a!. (18, 25) have not found dark-induction of either antitumor immunity by up-regulating IL-6. The mechanisms by which IL-6 or c-fos and c-jun in their in vitro human cell system. Our this is achieved are not yet clear. Dougherty et al. (23) have suggested experiments support the observation of a Photofrin dark-effect. This that IL-6 may further the recruitment of tumoricidal macrophages into may be of some significance for the hematopoiesis-stimulating effects the tumor bed. FACS analysis of PDT-treated tumors in the present of Photofrmnobserved by others (26) and us,4 because IL-6 can act as study does demonstrate an approximate 2â€”3-foldincrease in the a cofactor with other cytokines in the growth of early hematopoietic percentage of macrophages over a 24-h period. On the other hand, stem cells (14). Mule et aL (15) have shown that IL-6-mediated tumor regression IL-lO was not expressed constitutively in EMT6 cells, nor could it could be abrogated by in vivo depletion of either CD4@ or CD8@ be induced by PDT in vitro. It was, however, present in most, albeit T-cell subsets. Although this study did not detect any T-cell accumu not all, untreated tumors. When it was present, PDT caused a reduc lations in PDT-treated tumors, changes in T-cell function might be tion in IL-lO mRNA over time after treatment. Differences in IL-lO present, which are presently being analyzed. expression among EMT6 tumors have been described by Kurt et a!. Noteworthy is the induction of IL-6 mRNA with exposure to (9), who have demonstrated higher IL-lO levels in regressing than Photofrin in the absence of light activation in spleen (Fig. 1, B and C). progressing tumors. The inconsistency of mRNA levels in our tumors Earlier, Luna et a!. (24) have described in murine RIF cells in vitro the may reflect differences in their immune status. In most tumors, the induction by Photofrmn of the early-response genes c-los and c-jun, mRNA expression of IL-6 and IL-lO appeared inversely related, which are related to IL-6 transcription via the AP-l transcription showing a marked decrease at the same time point (1 h after PDT) at factor (14, 18, 25). Although some ofthis effect was artifactual due to

medium changes during the in vitro experiments, there clearly ap 4 x. Liu, S. 0. Gollnick, B. Owczarczak, and B. W. Henderson, unpublished obser peared to be additional gene induction due to Photofrmnincubation in vations. 3907

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1997 American Association for Cancer Research. CYTOKINEEXPRESSIONAFTERPDT A extended areas of skin in vivo. Keratinocyte-derived IL-lO has been â€” .@ .c found responsible for the UV-induced systemic suppression of CHS @ .c@ (0 0 Q...J Cd)C@@ and delayed-type hypersensitivity reactions (30). The kinetics of IL-lO induction in the skin after PDT, with IL-lO becoming prominent .@ â€˜4 at 72 h after exposure, coincide with the development of suppression @ - â€˜@ Actin of CHS after Photofrmn PDT (17), suggesting a mechanistic role in the observed immunosuppressive effects. This is further supported by the fact that IL-lO is known to down-regulate MHC class II and costimu latory molecule expression, resulting in suppression of Langerhans cell antigen-presenting function (31). Such down-regulation has been B observed recently by Obochi et at. (32) after low-dose PDT of skin allografts.

C Acknowledgments We are gratefulto Dr. CarletonC. Stewart,Flow CytometryFacility,Roswell 0 5.. Park Cancer Institute, for assistance and advice, to Dr. Allan Oseroff for support and suggestions, and to Dr. Thomas B. Tomasi, Molecular Medicine, Roswell Park Cancer institute, for his continued support and helpful suggestions.

References 0 I. Fisher, A. M. R., Murphree, A. L., and Gomer, C. J. Clinical and preclinical photodynamic therapy. Lasers Surg. Med., 17: 2â€”32,1996. 2. Henderson,B. W., and Dougherty,T. J. How does photodynarnictherapywork? Photochem. Photobiol., 55: 145â€”157,1992. C 3. Fingar, V. H. Vascular effects of photodynamic therapy. J. Clin. Laser Med. Surg., 14: 323â€”328,1996. 4. Korbelik, M. Induction of tumor immunity by photodynamic therapy. J. Clin. Laser Control Ph Light 24h 48h 72h 96h 120h Med. Surg., 14: 329â€”334,1996. 5. Elmets, C. A., and Bowen, K. D. Immunological suppression in mice treated with Fig. 4. RT-PCR (A) and ELISA (B) analysis of IL-lO in murine skin exposed to PDT hematoporphyrin derivative photoradiation. Cancer Res., 46: 1608â€”1611,1986. (10 mg/kg Photofrin, 6 h later I.3 J/cm2 of blue light). In A, 2 @goftotal RNA from each 6. Simkin, G., Obochi, M., Hunt, D. W. C., Chan, A. H., and Levy, J. 0. Effect of sample were subjected to RT-PCR analysis as described in â€œMaterialsandMethods.â€•In photodynamic therapy using benzoporphyrin derivative on the cutaneous immune B. 10 @.tgof total protein from each sample or appropriate standards (rIL-lO) were response.Proc.tnt. Soc. OpticalEng., 2392: 23â€”33,1995. analyzedby ELISAas describedin â€œMaterialsandMethods.â€•Untreatedcontrolsamples. 7. Gruner, S., Meffert, H., Volk, H. D., Oninow, R., and Jahn, S. The influence of samples exposed to Photofrin only (Ph), and samples exposed to light only (harvested at haematoporphyrin derivative and visible light on murine skin graft survival, epider 72 h after light) were analyzed in parallel. Data are the mean of two independent mal Langerhans cells and stimulation of the allogeneic mixed leucocyte reaction. experiments, each with duplicate wells; bars, SE. Scand.J. Immunol.,21: 267â€”273,1985. 8. Henderson, B. W., Waldow, S. M., Mang, T. S., Potter, W. R., Malone, P. B., and Dougherty, T. J. Tumor destruction and kinetics of tumor cell death in two experi mental mouse tumors following photodynamic therapy. Cancer Res., 45: 572â€”576, which IL-6 was markedly up-regulated. IL-lO is known to inhibit the 1985. IL-6 mRNA and protein expression by macrophages by down-regu 9. Kurt, R. A., Park, J. A., Panelli, M. C., Schiuter, S. F., Marchalonis, J. J., Carolus, B., and Akporiaye. E. T. T lymphocytes infiltrating sites of tumor rejection and progres lating the transcriptional activity of the IL-6 gene through inhibition of sion display identical V@usage but different cytotoxic activities. J. Immunol., 154: AP-l binding activity (27, 28). 3969â€”3974,1995. The cells responsible for altered overall cytokine mRNA and pro 10. Chomczynski, P., and Sacchi, N. Single step method and RNA isolation by acid guanidinium thiocyanate-phenol chloroform extract. Anal. Biochem., 162: 156â€”159, tein expression in tumors after PDT are not yet identified. Although 1987. IL-6 and/or TNF-a could have derived from EMT6 tumor cells, ii. Speliman, J. E., Gollnick, S. 0., Zhang, P. J., and Tomasi, T. B. Cytokine production by soft tissue sarcomas: implications for immunosuppression within the tumor bed. macrophages, lymphocytes, fibroblasts, and endothelial cells, the lack Surg.Oncoi.. 5: 237â€”244,1996. of IL-lO expression in EMT6 cells in vitro indicates that tumor IL-b 12. Laemmli, U. K. Cleavage and sti-uctural proteins during the assembly of the head of must have been generated by host cells such as macrophages and bacteriophage T4. Nature (Lond.), 227: 680â€”685,1970. 13. Liu, K., Kasper. M., Bierhaus, A., Langer, S., Muller, M., and Trott, K. Connexin 43 lymphocytes. The significance of the pronounced granulocyte accu expressionin normaland irradiatedmouseskin. Radiat.Res., 147: 437â€”441,1997. mulation in tumors after PDT is also not yet clear. Such accumulation 14. Kishimoto, T. The biology of interleukin-6. Blood, 74: 1â€”10,1989. has been observed by others, and experiments have shown that de 15. Mule, J. J., Custer, M. C., Travis, W. D., and Rosenberg, S. A. Cellular mechanisms of the antitumor activity of recombinant IL-6 in mice. J. Immunol., 148: 2622â€”2629, pletion of granulocytes in animals in vivo may retard the PDT tumor 1992. response (4, 29). 16. Ullrich,S. E. Mechanisminvolved in the systemic suppressionof antigen-presenting In contrast to tumor, both IL-6 and IL-lO were induced simulta cell function by UV irradiation: keratinocyte-derived IL-b modulates antigen-pre senting cell function of splenic adherent cells. J. Immunol., 152: 3410â€”3416, 1994. neously in the spleen. Induction was independent of the presence of 17. Musser, D. A., and Fiel, R. J. Cutaneous photosensitizing and immunosuppressive tumor and appears to be related to the penetration of some therapeutic effects of a series of tumor localizing porphyrins. Photochem. Photobiol., 53: 119â€” 123, 1991. light to the spleen. This hypothesis is supported by the results from 18. Kick, G., Messer, G., Goetz, A., Plewig, 0., and Kind, P. Photodynamic therapy animals treated with subcurative light doses (34 J/cm2) in which induces expression of interleukin 6 by activation of AP-l but not NF-kB DNA cytokine alterations in the tumor were similar to those seen with binding. Cancer Res., 55: 2373â€”2379,1995. 19. Evans, S., Matthews, W., Perry, R., Fraker, D., Norton, J., and Pass, H. I. Effect of higher light doses but where little or no change was seen in cytokine photodynamic therapy on tumor necrosis factor production by murine macrophages. levels in spleen (data not shown). This observation demonstrates that J. Natl.CancerInst.,82: 34â€”39,1990. care must be exercised both in the delivery of PDT to small animals 20. Anderson, C., Hrabovsky, S., McKinley, Y., Tubesing, K., Tang, H., Dunbar, R., Mukhtar, H., and Elmets, C. A. Phthalocyanine photodynamic therapy: disparate and in the interpretation of the results. effects of pharmacologic inhibitors on cutaneous photosensitivity and on tumor The IL-b induction in the PDT-treated skin may be of particular regression. Photochem. Photobiol., 65: 895â€”901,1997. 21. Vincek, V., Kurimoto, I., Medema, J. P., Prieto, E., and Streilein, J. W. Tumor biological significance because it may explain, at least in part, the necrosis factor a polymorphism correlates with deleterious effects of ultraviolet B transient suppression of the CHS reaction after PDT exposure of light on cutaneous immunity. Cancer Res., 53: 728â€”732,1993. 3908

22. Mullen, C. A., Coale, M. M., Levy, A. T., Stetler-Stevenson, W. 0., Liofta, L. A., carcinomacolon26-inducedcachexiabyinterleukin10genetransfer.CancerRes., Brandt, S., and Blaese, R. M. Fibrosarcoma cells transduced with the JL-6 gene 57: 94â€”99,1997. exhibit reduced tumorigenicity, increased immunogenicity, and decreased metastatic 28. Dokter, W. H. A., Koopmans, S. B., and Vellenga, E. Effects of IL-b and IL-4 on potential. Cancer Res., 52: 6020â€”6024,1992. LPS-induced transcription factors (AP-l, NF-1L6 and NF-kB) which are involved in 23. Dougherty. G. J., Thacker, J. D., Lavey, R. S., Belldegrun, A., and McBride, W. H. IL-6 regulation. Leukemia (Baltimore), 10: 1308â€”1316,1996. Inhibitory effect oflocaily produced and exogenous interleukin-6 on tumor growth in 29. deVree, W. J. A., Essers, M. C., DeBruijn, H. S., Star, W. M., Koster, J. F., and vivo. Cancer Immunol. Immunother., 38: 339â€”345,1994. Sluiter, W. Evidence for an important role of neutrophils in the efficacy of photody 24. Lana, M. C., Wong, S., and Gomer, C. J. Photodynamic therapy mediated induction namic therapy in Vivo. Cancer Res., 56: 2908â€”2911, 1996. of earlyresponsegenes.CancerRca.,54: 1374â€”1380,1994. 30. Ullrich, S. E. The role of epidermal cytokines in the generation of cutaneous immune 25. Kick, 0., Messer, 0., Plewig, 0., Kind, P., and Goetz, A. E. Strong and prolonged reactions and ultraviolet radiation-induced immune suppression. Photochem. Photo injunctionof c-jun and c-fos proto-oncogenesbyphotodynamictherapy.Br. J. biol., 62: 389â€”401, 1995. Cancer, 74: 30â€”36,1996. 31. Enk, A. H., Angeloni, V. L., Udey, M. C., and Katz, S. I. Inhibition ofLangerhans cell 26. Canti, 0., Franco,P., Marelli,0., Ricci,L, and Nicolin,A. Hematoporphyrin antigen-presentingfunctionby IL-lO.A role for IL-lOin inductionof tolerance. derivativerescuefromtoxicitycausedby chemotherapyorradiationin a murine J. Immunol., 151: 2390â€”2398, 1993. leukemia model (L12l0). Cancer Res., 44: 1551â€”1556,1984. 32. Obochi, M. 0. K., Ratkay, L. G., and Levy, J. G. Prolonged murine skin allograft 27. Fujiki, F., MUkaida, N., Hirose, K., Ishida, H., Harada, A., Ohno, S., Bluethmann, H., survival after photodynamic therapy (PDT) associated with modification of donor Kawakami, M., Akiyama, M., Sone, S., and Matsushima, K. Prevention of adeno skin antigenicity. Photochem. Photobiol. 63S: 645, 1996.

3909

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 1997 American Association for Cancer Research. Altered Expression of Interleukin 6 and Interleukin 10 As a Result of Photodynamic Therapy in Vivo

Sandra O. Gollnick, Xiaonan Liu, Barbara Owczarczak, et al.

Cancer Res 1997;57:3904-3909.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/57/18/3904

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/57/18/3904. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.