INTRODUCTIONThe type II interferon, interferon γ

(IFN-γ), has recently been implicated incancer immunosurveillance in additionto its well-known function in promot-ing host responses to microorganisms(reviewed in [1]). According to a num-ber of different experiments in mice,IFN-γ participates in immunoediting,which describes the interaction of thehost’s immune system with emergingtumors (2). Experimental data gener-ated with IFN-γ–insensitive mice,

deficient for either an IFN-γ receptorsubunit or the respective signal trans-duction pathway (3,4), indicated that inwild-type mice endogenously producedIFN-γ not only promotes rejection oftransplantable tumors, but also pre-vents development of primary tumors.Although natural killer (NK) cells, NK T cells and apoptosis-mediating lig-ands such as TRAIL (tumor necrosisfactor–related apoptosis-inducing lig-and) have been shown to be involved inthis process (2), the detailed cellular

and molecular levels of the cancer- eliminating phase still remain to be elu-cidated. In this respect, Kundu and col-leagues (5,6) previously reportedinterleukin (IL)-10–induced antitumoralactivity in a mammary cancer mousemodel. Follow-up studies revealed asubstantial role of IFN-γ in this process(7). Interestingly, during these experi-ments, the upregulation of two IFN-γ–inducible genes, mig-1 and gbp-1,within the transplanted tumor cells wasdemonstrated. Mig-1, the monokine in-duced by IFN-γ or CXCL9, is known toattract activated T and NK cells andwas recently shown to at least partiallycontribute to the observed limitedmetastasis within the investigatedtumor model (8). The role of the IFN-γ–induced guanylate binding pro-tein (GBP), however, remained unclear.

M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0 | L I P N I K E T A L . | 1 7 7

Interferon γ–Induced Human Guanylate Binding Protein 1Inhibits Mammary Tumor Growth in Mice

Karoline Lipnik,1,2 Elisabeth Naschberger,3 Nathalie Gonin-Laurent,3 Petra Kodajova,1 Helga Petznek,1

Stefanie Rungaldier,1 Simonetta Astigiano,4 Silvano Ferrini,4 Michael Stürzl,3 and Christine Hohenadl1*

1Institute of Virology, Department of Pathobiology, University of Veterinary Medicine Vienna, Vienna, Austria; 2Centre of Biomolecular Medicine and Pharmacology, Department of Vascular Biology and Thrombosis Research, Medical University Vienna,Vienna, Austria; 3Division of Molecular and Experimental Surgery, Department of Surgery, University of Erlangen-Nuremberg, Erlangen, Germany; 4Transgenic Unit and Immunotherapy Unit, Instituto Nazionale per la Ricerca sul Cancro, Genoa, Italy

Interferon γ (IFN-γ) has recently been implicated in cancer immunosurveillance. Among the most abundant proteins inducedby IFN-γ are guanylate binding proteins (GBPs), which belong to the superfamily of large GTPases and are widely expressed in var-ious species. Here, we investigated whether the well-known human GBP-1 (hGBP-1), which has been shown to exert antiangio-genic activities and was described as a prognostic marker in colorectal carcinomas, may contribute to an IFN-γ–mediatedtumor defense. To this end, an IFN-independent, inducible hGBP-1 expression system was established in murine mammary carci-noma (TS/A) cells, which were then transplanted into syngeneic immune-competent Balb/c mice. Animals carrying TS/A cellsthat had been given doxycycline for induction of hGBP-1 expression revealed a significantly reduced tumor growth comparedwith mock-treated mice. Immunohistochemical analysis of the respective tumors demonstrated a tightly regulated, high-levelexpression of hGBP-1. No signs of an enhanced immunosurveillance were observed by investigating the number of infiltrating Band T cells. However, hemoglobin levels as well as the number of proliferating tumor cells were shown to be significantly reducedin hGBP-1–expressing tumors. This finding corresponded to reduced amounts of vascular endothelial growth factor A (VEGF-A)released by hGBP-1–expressing TS/A cells in vitro and reduced VEGF-A protein levels in the corresponding mammary tumors invivo. The results suggest that hGBP-1 may contribute to IFN-γ–mediated antitumorigenic activities by inhibiting paracrine effectsof tumor cells on angiogenesis. Consequently, owing to these activities GBPs might be considered as potent members in an in-nate, IFN-γ–induced antitumoral defense system.© 2010 The Feinstein Institute for Medical Research, www.feinsteininstitute.orgOnline address: http://www.molmed.orgdoi: 10.2119/molmed.2009.00172

Address correspondence and reprint requests to Christine Hohenadl, Institute of Virology,

Department of Pathobiology, University of Veterinary Medicine Vienna, Veterinaerplatz 1,

A-1210 Vienna, Austria. Phone: +431250772330; Fax: +431250772390; E-mail:

christine.hohenadl@vetmeduni.ac.at.

Submitted November 20, 2009; accepted for publication February 4, 2010; Epub

(www.molmed.org) ahead of print February 5, 2010.

GBPs were originally identified inhuman fibroblasts as the most abundantproteins induced by IFN-γ treatment (9).The GBPs are part of a subfamily withinthe protein superfamily of large GTPasesincluding dynamins and Mx proteins(for a review see [10]). To date, 7 and 11highly homologous GBPs with a relativemolecular mass of 65–71 kDa have beenidentified in humans and mice, respec-tively (11–13). Human GBP-1 (hGBP-1),with respect to structure and function, isthe best-characterized member withinthis family of proteins (14,15). Despite adetailed knowledge of biochemical char-acteristics, however, the biological func-tions of GBPs remain obscure. Recentstudies focusing on hGBP-1 suggestedthat the protein mediates the inhibitoryeffects of inflammatory cytokines onproliferation, migration and invasive-ness of endothelial cells (16–19). The an-tiangiogenic activity of hGBP-1 has beenrecently demonstrated for colorectal car-cinomas. In a study investigating morethan 380 tumor samples, it was shownthat tumors expressing hGBP-1 exhib-ited reduced angiogenic activity, andthis was associated with a significantlyimproved prognosis for the respectivepatients (20). In addition, an effect onproliferation of fibroblasts has beendemonstrated for murine GBP-2 (21).These observations indicate that GBPscan alter cell growth and are able tomodulate cellular interactions with theirenvironment.

Thus, in the present study, the effectsof hGBP-1 on tumor growth were in-vestigated using an immune-competentmouse model of mammary cancer. Tothis aim, a tetracycline-regulated, IFN- independent hGBP-1 expression systemwas established in the highly malig-nant mouse mammary cancer cell lineTS/A, followed by generation of re-spective tumors in syngeneic Balb/cmice. The impact of hGBP-1 on tumorcell proliferation, immune cell extrava-sation and angiogenesis was analyzed,revealing a highly significant tumorgrowth–inhibiting effect of hGBP-1,which appeared to be due to a defect in

angiogenesis rather than to enhancedtumor immunosurveillance.

MATERIALS AND METHODS

Cells and PlasmidsThe murine adenocarcinoma cell line

TS/A, which was derived from a sponta-neous mammary cancer of Balb/c mice(22), was cultured in RPMI 1640 (Invitro-gen Life Technologies, Carlsbad, CA,USA) supplemented with 1% nonessen-tial amino acids, 2 mmol/L glutamineand 10% fetal bovine serum (FBS) (all In-vitrogen Life Technologies). Cells weremaintained at 37°C with 5% CO2 and95% humidity. To establish an inducibleexpression system, cells were transducedwith the murine leukemia virus (MLV)-derived retroviral vector pLib-rtTA-M2-IRES-TRSID-IRES-Puror containing atetracycline on/off regulatory system(23,24). The coding sequence of hGBP-1was isolated from pMCV1.4 + GBP-1(25) by restriction digestion with EcoRIand inserted into the linearized tetracy-cline response plasmid pUHD10.3 (24).In addition, a SV40 promoter–driven hygromycin selection cassette (24) wasinserted via the restriction sites PciI andNarI, finally giving rise to the inducibleexpression vector pUHD10.3-GBP-Hyg(pUHD-GBP). As a control, the plasmidpUHD10.3-Hyg lacking the hGBP-1 en-coding sequence (pUHD) was also gen-erated. Stably transfected cell clonesand populations were selected by addi-tion of hygromycin (200 μg/mL) andpuromycin (10 μg/mL), respectively. Expression of hGBP-1 was induced byaddition of 1–2.5 μg/mL doxycycline(Sigma-Aldrich, St. Louis, MO, USA), atetracycline homolog.

Proliferation AssayA population of TS/A cells as well as

isolated single cell clones, stably trans-fected with pUHD-GBP, and a popula-tion of cells transfected with the controlvector pUHD were seeded in six-wellplates (duplicates) at a density of 1 × 105

cells/well. The next day, 1 μg/mL (low-level hGBP-1 expression) or 2.5 μg/mL

(high-level hGBP-1 expression) doxycy-cline was added to the cell culture me-dium. For cells that were treated withdoxycycline for more than 24 h, freshdoxycycline was added daily. Cells wereincubated for 24, 48, 72 and 96 h. Atthese time points, cells were harvested,and total cell numbers (cells/well) weredetermined by use of a CasyTT® cellcounter (Schärfe System, Reutlingen,Germany). For determination of poten-tial effects of hGBP-1 expression on pro-liferation, cell numbers were comparedwith cells that did not receive doxycy-cline (nontreated control). Mean values ±SD were calculated from three indepen-dent experiments.

Soft Agar AssayTo investigate tumorigenicity of TS/A

cells (nontransfected and transfectedwith the control vector pUHD) in com-parison to pUHD-GBP–transfected cellclones (clones 14 and 18, mixed 1:1) invitro, growth in soft agar was evaluated.To this end, the wells of a six-well platewere coated with a solution of RPMI1640/10% FBS containing 0.5% agar(Difco, Detroit, MI, USA) as well as either1 or 2.5 μg/mL doxycycline. Once thislayer was set, 2 × 104 cells/well (tripli-cates) embedded in RPMI 1640/10%FBS/0.3% agar containing either 1 or 2.5 μg/mL doxycycline were seeded ontop. Cells were allowed to grow for 8 d,at which point they were stained with0.005% crystal violet and the averagenumber of colonies determined by ana-lyzing 10 different optical fields (ZeissAxiovert 200M, Carl Zeiss GmbH, Vi-enna, Austria; magnification 100×). Meanvalues and SD were calculated.

Western Blot AnalysisProteins extracted from cells or tissues

were quantified by applying a modifiedLowry assay (Bio-Rad DC protein assay,Bio-Rad Laboratories, Vienna, Austria)using bovine serum albumin (Promega,Mannheim, Germany) as a referencestandard. Equal amounts of protein werethen separated under reducing condi-tions in 10% SDS-polyacrylamide gels

1 7 8 | L I P N I K E T A L . | M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0

A N T I T U M O R I G E N I C A C T I V I T Y O F h G B P - 1

and transferred onto a Hybond-P poly -vinyldifluoride (PVDF) membrane (GEHealthcare Technologies, Vienna, Aus-tria). Membranes were blocked for 1 h in5% organic skim milk powder (HeilerCenovis, Radolfzell, Germany) dissolvedin phosphate-buffered saline (PBS)(VWR International, West Chester, PA,USA) containing 0.1% Tween-20 (Bio-Rad). Detection of hGPP-1 was per-formed as described (26). Briefly, mem-branes were incubated for 1 h inPBS/0.1% Tween-20/5% milk powdercontaining a 1:500 diluted rat mono-clonal anti–hGBP-1 antibody. Bound an-tibody was detected by applying aspecies-specific horseradish peroxidase(HRP)-conjugated secondary antibody(Dako Cytomation, Vienna, Austria) (di-luted 1:10,000) followed by enhancedchemiluminescence (GE Healthcare). Forstaining of murine vascular endothelialgrowth factor (VEGF), a rabbit poly-clonal antiserum recognizing the differ-ent variants of mouse and human VEGF-A (Santa Cruz Biotechnology, Heidelberg, Germany) was applied in adilution of 1:400. As an internal controlallowing evaluation of loaded proteinamounts, glyceraldehyde-3-phosphatedehydrogenase (GAPDH) was detectedusing a monoclonal mouse anti-GAPDHantibody (Millipore-Chemicon Interna-tional, Vienna, Austria) (diluted1:50,000). For quantification of proteinlevels, linear exposures of blotted mem-branes were scanned using a fluoroim-ager (Storm 860; Amersham Biosciences[GE Healthcare Europe, Freiburg, Ger-many]), and bands were evaluated applying the respective analysis soft-ware (ImageQuant TL, GE HealthcareEurope).

Tumor Mouse ModelCells (1 × 106) of stably transfected/

infected TS/A cells harboring either thedoxycycline-inducible expression vectorpUHD10.3-GBP-Hyg (cell clones 14 and18, mixed 1:1) or the parental vectorpUHD10.3-Hyg (cell population), wereinjected subcutaneously into the pectoralmammary fat pad of female, 8-wk-old

Balb/c mice (40 animals per cell type)(Jackson Laboratories, Bar Harbor, ME,USA). All animals were implanted sub-cutaneously with a microchip transpon-der (BackHome; Virbac, Vienna, Austria)for individual identification. At d 7 afterinjection, mice were randomly distrib-uted into two groups. One group(treated) was injected intraperitoneallywith doxycycline at a concentration of 50 mg/kg body weight (27) every 12 hfor the first 72 h and every 24 h there-after. The other group (nontreated) wasinjected in the same way with physio-logical saline solution (0.9% NaCl). Tu-mors were measured three times a wkin two dimensions, using a caliper, andthe tumor volume was calculated ac-cording to the formula L × w × w/2.Evaluation of tumor growth was per-formed in a blinded manner, knowingonly the identification number but notthe group affiliation of the respectiveanimals. The experiment was stopped17 d after onset of induction, whentumor sizes exceeded 1500 mm3. Micewere killed by cervical dislocation anddissected. Tumor tissue was frozen inliquid nitrogen and stored at –80°C untiluse or was fixed in 10% neutral bufferedformalin (Sigma-Aldrich) for histologi-cal examination. Tumor growth in micewas performed according to the rulesand regulations governing animal ex-perimentation in Austria (BMBWK-68.205/0126-BrGT/2005).

Statistical AnalysisTumor growth rates (doubling times)

of individual animals in each experimen-tal group were determined by nonlinearregression analysis of the obtainedgrowth curves, shown as mean tumorvolumes ± SEM. A two-tailed, unpairedStudent t test (GraphPad Prism® soft-ware, GraphPad Software Inc., La Jolla,CA, USA) was used to determine the sig-nificance of differences between meantumor doubling times of two experimen-tal groups (doxycycline-treated versusnontreated control). The difference wasconsidered statistically significant whenP < 0.05.

ImmunohistochemistryTumor samples fixed in 10% neutral

buffered formalin (Sigma-Aldrich) or zincfixative (BD Pharmingen, part of BD Bio-sciences, Heidelberg, Germany) were em-bedded in paraffin (Histo-Comp®;Sanova, Vienna, Austria) using automaticembedding equipment (Tissue Tek; MilesScientific, Vienna, Austria). Subsequently,5-μm sections were prepared, dewaxed inxylol (ACM Handels, Vienna, Austria)and rehydrated in a descending ethanolseries (100%, 96% and 70%). After antigenretrieval by microwave pretreatment (3 ×7 min, 580 W) in target retrieval solution,pH 9 (TRS9; Dako), slides were treatedwith 7.5% hydrogen peroxide for 10 minat room temperature to block endoge-nous peroxidases. For staining of hGBP-1,slides were incubated for 1 h with mono-clonal rat anti–hGBP-1 (1:300) at roomtemperature. Subsequently, sections wereincubated for 30 min with biotinylatedhorse antirat IgG, followed by 30 min ofincubation with an avidin-biotin com-plex (ABC kit; Vector Laboratories,Burlingame, CA). The reaction was devel-oped with 0.1% 3,3′- diaminobenzidine(DAB) hydrochloride (Sigma-Aldrich) in0.03% hydrogen peroxide and counter-stained with Mayer’s hemalum (VWR International). After dehydration for 2min in ethanol (96% and 100%) and2-min incubation in xylol, sections weremounted in DPX mounting medium(Sigma-Aldrich) and examined by lightmicroscopy (Zeiss Axiovert 200M). Stain-ing for the mouse endothelial cell– specific marker CD31 and the prolifera-tion marker Ki-67 was performed onzinc-fixed tissue sections (28,29) usingmonoclonal rat antimouse CD31 (cloneMEC 13.3; Santa Cruz) diluted 1:250) andrat antimouse Ki-67 antibodies (Tec-3;Dako Cytomation) (diluted 1.50). Stainingwas visualized using a species-specificAPAAP (alkaline phosphatase anti– alkaline phosphatase) detection systemapplying fuchsin (Dako Cytomation) as asubstrate. Sections were counterstainedwith hematoxylin Gill-III (Merck, Darm-stadt, Germany) and mounted. Quantita-tive evaluation of proliferating cells

R E S E A R C H A R T I C L E

M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0 | L I P N I K E T A L . | 1 7 9

within tumor sections (GBP+, n = 11;GBP–, n = 10) was performed using theAxioVision version 1.4 software (Zeiss).Therefore, sections were photographed(Zeiss Axiovert 200M, magnification100×) and aligned to consecutive sectionsstained for hGBP-1 expression. The re-spective tumor area was divided into in-dividual squares to facilitate counting ofKi-67–positive cells/100,000 μm2 in re-gions with high or low hGBP-1 expres-sion. Subsequently, mean values and re-spective SDs were calculated. Resultswere statistically analyzed as describedabove, applying either an unpaired Stu-dent t test or, if more than two groupswere compared, a one-way ANOVA.

For detection of B and T cells in tumorsections, slides were dewaxed and hy-drated as described above. Antigen re-trieval was performed by boiling sam-ples in citrate buffer pH 6.0 (3 × 5 min ina microwave oven). As primary antibod-ies, a rabbit monoclonal anti-CD3 (cloneSP7; Lab Vision-Neomarkers, Fremont,CA, USA), diluted 1:50, and a rat mono-clonal anti-CD45R/B220 (BD Pharmin-gen), diluted 1:25, were used. Species-specific secondary antibodies werebiotinylated (Dako Cytomation), andantigen reaction was revealed with HRP-conjugated streptavidin (Bio SPA, Milan,Italy) and DAB as a substrate (Lab Vision).

Positively stained cells in eight differ-ent sections were counted blindly byevaluating 10 different randomly chosenoptical fields at a magnification of 400×.Results are given as mean values ± SD.

Determination of HemoglobinContents

To determine the level of perfusion inexplanted tumors (30), snap-frozen tis-sue samples (TS/A-GBP + doxycycline,n = 18; TS/A-GBP – doxycycline andTS/A-Hyg + doxycycline, n = 18) wereweighed and pulverized under liquidnitrogen using a pestle and mortar.Powdered tissue was suspended inwater for injection (100 μL/10 mg tis-sue) and centrifuged at 16110g at 4°C for5 min (Eppendorf centrifuge 5415R),

and total protein contents were deter-mined for normalization as describedabove. To quantify the hemoglobin con-tent, 100 μL of the tumor suspensionwas mixed with 900 μL Drabkin’sreagent (Sigma-Aldrich) and incubatedin the dark at room temperature for 15min. As a reference, human hemoglobin(30–160 mg/mL) (Sigma-Aldrich) wasused. Absorbance was measured using aspectrophotometer at 540 nm. Statisticalevaluation was performed by use of anunpaired Student t test as describedabove.

VEGF-ELISAFor analysis of synthesis of murine

VEGF, tetracycline (Tet)-regulated TS/Acells stably transfected with pUHD10.3-hGBP-Hyg (cell clones 14 and 18) as wellas cells transfected with the control vec-tor pUHD10.3-Hyg were seeded in dupli-cates into six-well plates at a density of1 × 105 cells/well. The next day, and dailythereafter, either 1 or 2.5 μg/mL doxycy-cline was added to the cell culture me-dium (RPMI/5% FBS). Non–doxycycline-treated cells were used as a control. Allcells were cultured for 96 h, after whichculture supernatants were harvested andfiltered (Millipore, 0.45 μm), and theamount of secreted VEGF-A was deter-mined using the Quantikine MouseVEGF enzyme-linked immunosorbentassay (ELISA) (R&D Systems, Minneapo-lis, MN, USA), according to supplier’s instructions. Samples were analyzed at450 nm using an ELISA plate reader(Tecan GENios; Tecan, Männedorf, Aus-tria). For normalization, cell numbers ofeach sample were determined (CasyTT®;Schärfe System, Reutlingen, Germany)and protein measurements (pg/mL) cal-culated according to 1 × 106 cells. Resultsare given as mean values ± SD.

RESULTS

Establishment and Characterization ofTS/A Mouse Mammary Tumor Cellswith Inducible hGBP-1 Expression

To generate an inducible hGBP-1– expressing tumor cell line, the mouse

mammary adenocarcinoma cell lineTS/A was first transduced with an MLV-derived retroviral vector encoding atetracycline (Tet) on/off regulator system(23,24). The resulting cell population wassubsequently transfected with the corre-sponding Tet-responsive expression vec-tor pUHD10.3-GBP-Hyg (pUHD-GBP),giving rise to single cell clones whichwere then tested for inducible hGBP-1expression upon treatment with doxycy-cline, a commonly used tetracycline de-rivative. Western blot analysis was per-formed to identify distinct cell clones,showing a tightly regulated high-levelexpression of hGBP-1 exclusively upondoxycycline treatment (Figure 1A).

To investigate potential effects ofhGBP-1 expression in vitro, single cellclones as well as a population of TS/Acells stably transfected with pUHD-GBPwere analyzed for their proliferationcharacteristics (Figure 1B). As a control,cells stably transfected with the expres-sion vector lacking the hGBP-1 encodingsequence (pUHD) were analyzed. Cellswere either left untreated (data notshown, 100% values) or incubated for 24to 96 h with 1 or 2.5 μg/mL doxycycline.Finally, cell numbers were determinedand calculated relative to untreated con-trol cells (Figure 1B). The results showthat hGBP-1 expression does not affectproliferation of the investigated cells invitro. Similarly, no effects of hGBP-1 ex-pression were observed on the potentialof transfected TS/A cells to formcolonies in soft agar (Figure 1C), indicat-ing that hGBP-1 alters neither tumori-genicity nor metastatic potential of thesecells.

hGBP-1 Expression Inhibits Growth ofTS/A Cell–Derived Tumors in Balb/cMice

After analysis of the growth charac-teristics of the established stably trans-fected TS/A cells in vitro, two tightlyregulated cell clones that expressed highlevels of hGBP-1 upon doxycyclinetreatment (see Figure 1A) were equallymixed and injected subcutaneously intothe mammary fat pad of immune-

1 8 0 | L I P N I K E T A L . | M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0

A N T I T U M O R I G E N I C A C T I V I T Y O F h G B P - 1

competent syngeneic Balb/c mice to in-vestigate the effects of hGBP-1 expres-sion on TS/A cell–derived tumors invivo. As a control, a population of cells

stably transfected with the vectorpUHD10.3-Hyg (pUHD) was also in-jected. After 7 d, when tumor volumeshad reached 100 mm3, mice were split

into two groups (n = 20) and injected in-traperitoneally with either doxycycline(+ dox) or saline (– dox). Tumor growthwas regularly measured, and the experi-ment was stopped when tumor volumesexceeded 1500 mm3. Mice that did notdevelop tumors at the end of the experi-ment (n = 4) or died of other causes (n = 2) were excluded from the analyses.Significantly reduced tumor growth (P = 0.025) was shown in mice that wereinjected with pUHD-GBP–transfectedcells and had been given doxycyclinecompared with animals that were in-jected with saline (Figure 2A). An even more pronounced effect (P =0.009) was observed when tumorgrowth of hGBP-1–expressing cells(TS/A pUHD-GBP + dox) was com-pared with that of cells transfected withthe parental vector (see Figure 2A). Ap-plication of doxycycline did not haveany effect on tumor growth as demon-strated by equal growth rates of pUHD-transfected cells in both doxycycline-and saline-treated mice (see Figure 2A).

At the end of the experiment, animalswere killed, and tumors were excisedfor further analyses. Expression ofhGBP-1 was shown by immunohisto-chemical analyses of excised tumor tis-sue using a monoclonal antibody spe-cific for the human protein (26). Alltumors derived from doxycycline-treated mice that had been injected withpUHD-GBP- transfected TS/A cellsshowed strong expression of hGBP-1 intumor cells (Figure 2B, a and b). Tumorsderived from pUHD-GBP–transfectedTS/A cells that had been treated withsaline, on the other hand, showed onlysingle tumor cells positively stained for the presence of hGBP-1 protein (Fig-ure 2B, c). No unspecific staining wasobserved in tumors derived from pUHD-transfected TS/A cells (Figure 2B, d).

Tumor Cell Proliferation Is Reduced inHigh-Level hGBP-1–Expressing Tumors

To elucidate the cause of hGBP-1 expression–associated reduced tumorgrowth, the excised tissue was analyzedfor the presence of Ki-67, a common cellu-

R E S E A R C H A R T I C L E

M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0 | L I P N I K E T A L . | 1 8 1

Figure 1. Characterization of TS/A mammary carcinoma cells inducibly expressinghGBP-1. (A) Western Blot analysis of TS/A cell clones 13–18 harboring the Tet on/off regulatory system and the vector pUHD10.3-GBP-Hyg, mediating tightly regulated, in-ducible expression of hGBP-1 in response to doxycycline (+ dox). As a control for analy-sis of equal amounts of protein, a Coomassie-stained SDS-polyacrylamide gel with therespective samples is depicted (lower panel). (B) Analysis of proliferation characteristicsof pUHD10.3-GBP–transfected cells (cell population, gray bars; cell clone, white bars) incomparison to pUHD10.3-Hyg–transfected cells (black bars). Cell numbers (mean val-ues ± SD) from three independent experiments are shown in % relative to nontreated (– dox) samples. (C) Soft agar assay. Nontransfected TS/A cells and TS/A cells trans-fected with pUHD10.3-Hyg or the hGBP-1–encoding vector (pUHD10.3-GBP) were ana-lyzed for their ability to form colonies in soft agar. Expression of hGBP-1 was induced byaddition of 1 (gray bars) or 2.5 (black bars) μg/mL doxycycline and the number ofcolonies (mean values ± SD) obtained by counting 10 different optical fields comparedwith nontreated samples (white bars).

lar marker indicating cell proliferation(31). To this end, hGBP-1–expressing (n =11) and nonexpressing (n = 10) tumorswere immunohistochemically stained with

a specific antibody detecting murine Ki-67(Figure 3A). The number of Ki-67– positivecells was evaluated by analyzing at leastfour different areas of a given section

using the Zeiss AxioVision software (Fig-ure 3A, area report). By this means, a sig-nificantly (P = 0.0175) reduced number ofproliferating tumor cells/100,000 μm2 wasshown in hGBP-1– expressing tumors (Fig-ure 3B). Analysis of consecutive sectionsfor hGBP-1 expression and Ki-67 (see Fig-ure 3A) allowed an assignment of cell pro-liferation with hGBP-1 expression. Thisapproach showed that inhibition of cellproliferation is predominantly observed inareas with high GBP-1 expression (see Fig-ure 3B). In contrast, reduced cell prolifera-tion was hardly observed in tissues with alow level of hGBP-1 expression and wasnot observed in tumors lacking hGBP-1expression (see Figure 3B). Statistical eval-uation comparing numbers of Ki-67–posi-tive cells in high-level hGBP-1– expressingareas to all other investigated samples re-vealed a highly significant decrease (P <0.0001) (see Figure 3B).

Inhibition of tumor cell proliferation asa result of hGBP-1 expression, however,was not encountered in the respective invitro studies (see Figure 1B). Thus, basedon previous reports showing that hGBP-1 exerts antiangiogenic effects in vitroand in vivo (17–19) and is robustly in-duced by IFN-γ, which in turn is impli-cated in tumor immunosurveillance(1,32), a possible contribution of theseprocesses was investigated.

hGBP-1–Mediated Tumor GrowthInhibition Is Not due to a SpecificImmune Reaction

Because an IFN-γ–stimulated immuneresponse has been previously implicatedin inhibition of mammary cancercell–derived tumor growth in Balb/cmice (7,8), the hGBP-1–expressing TS/Acell–derived tumors generated in thepresent study were analyzed for evi-dence of immune stimulation by stainingtissue sections for the presence of thecell-surface marker molecules CD3 andB220, indicating infiltration of tumorswith T and B cells, respectively. CD3+

cells were detected in the periphery ofanalyzed sections, predominantly in thevicinity of larger vessels, whereas B220+

B cells were barely detectable (Figure 4,

1 8 2 | L I P N I K E T A L . | M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0

A N T I T U M O R I G E N I C A C T I V I T Y O F h G B P - 1

Figure 2. Effect of hGBP-1 expression on tumor growth in a syngeneic mammary carci-noma mouse model. (A) Tumor growth was monitored in Balb/c mice injected with 1 × 106

TS/A cells transfected with either pUHD10.3-GBP (�, �) or the parental vector pUHD10.3-Hyg (�, �). At d 7 after injection, animals were divided into two groups (n = 20) given ei-ther doxycycline (+ dox) or saline (– dox). Mean tumor volumes ( ± SEM) of individual ani-mals are indicated. Animals that did not develop a tumor at the end of the experiment ordied due to unrelated reasons were excluded. Statistical significance (P values) was de-termined applying a Student t test comparing the tumor doubling times of doxycycline-treated versus nontreated animals. (B) hGBP-1–specific immunohistochemical analysis oftumors derived from pUHD-GBP–transfected TS/A cells treated with doxycycline (a, b) orsaline (c). No unspecific staining was observed in tumors from doxycycline-treated ani-mals that had been injected with TS/A cells harboring the control vector pUHD (d).

inserts). Altogether, no significant differ-ences in lymphocyte cell numbers wereobserved in hGBP-1– expressing tumorsin comparison to the respective saline-treated controls (see Figure 4). Moreover,evaluation of hematoxylin/eosin–stained

sections did not reveal any signs of in-flammation. Thus, these results indicatethat the observed growth reduction inhGBP-1– expressing tumors in this syn-geneic mouse model is not due to an ele-vated immune response.

Reduced Tumor Growth Correspondswith Decreased Hemoglobin Contentsand Reduced VEGF-A Levels in vitroand in vivo

Finally, to investigate vascularizationin TS/A cell–derived tumors, correspon-ding tumor sections were stained for ex-pression of hGBP-1 and murine CD31, amolecular marker of endothelial cells(33). By this means, all tumors wereshown to be well vascularized with fewnecrotic areas (Figure 5A). Quantitativeanalyses, which were performed to eval-uate microvessel densities in hGBP- expressing tumors compared with non-expressing tumors, did not reveal anystatistically significant differences (datanot shown). To determine the qualitativecapacity of the newly formed tumor ves-sels to supply the respective tissue withoxygen and nutrients, the amount of he-moglobin was determined in tumorswith and without hGBP-1 expression(Figure 5B). This well-establishedmethod is applied to quantify the actualstate of perfusion within a given tumortissue (34,35). For each group of tumors,18 samples were analyzed. Interestingly,a significantly lower amount of hemo-globin was determined in tumor sam-ples expressing hGBP-1 in comparison tononexpressing tumors (P = 0.0032), indi-cating that hGBP-1 expression in tumorcells may impair their capabilities topromote tissue vascularization. Thus, toaddress potential indirect effects ofhGBP-1 on angiogenesis, VEGF releasefrom hGBP-1–expressing and nonex-pressing tumor cells was determined byapplication of an ELISA specific for thedetection of murine VEGF-A. Two inde-pendent cell clones (14 and 18) were in-vestigated and showed a 20–60% de-crease of secreted VEGF-A when hGBP-1expression was induced by doxycycline(Figure 6A). In comparison, mock- transfected cells (TS/A pUHD) underthe applied conditions showed a 5–25%reduced release of VEGF-A. To investi-gate the VEGF-A protein levels in corre-sponding mammary tumors, 12 differentsamples, derived from animals trans-planted with pUHD-GBP–transfected

R E S E A R C H A R T I C L E

M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0 | L I P N I K E T A L . | 1 8 3

Figure 3. Evaluation of effects of hGBP-1 expression on tumor cell proliferation. (A) Immunohistochemistry. Consecutive sections of hGBP-1–expressing (TS/A-GBP +dox) and nonexpressing (TS/A-Hyg + dox) tumors were stained for expression of hGBP-1 (α-hGBP-1, brown cytoplasmic staining) and murine Ki-67 (α-mKi-67, red nu-clear staining), a cellular proliferation marker. Using the Zeiss AxioVision analysis soft-ware, sections were divided into different areas correlating with a high level (a, b, d, e,g, h, i) or low level (c, f) hGBP-1 expression. (B) The number of Ki-67–positive TS/A cellswas calculated and expressed as mean values per 100.000 μm2. Statistical analysis re-vealed significant differences (P = 0.017) in the number of Ki-67–positive cells (meanvalues ± SD) in hGBP-1–expressing (total GBP+, n = 11) compared with nonexpressingtumors (total GBP–, n = 10). Highly significant differences (P < 0.0001, one-way ANOVA)were observed when Ki-67–positive cell numbers were correlated with hGBP-1 expres-sion levels.

TS/A cells that were treated with eitherdoxycycline or saline, were subjected toquantitative Western blot analysis (Fig-ure 6B). After normalization to loadedprotein amounts as reflected by GAPDH-specific staining, VEGF-A protein levelsin hGBP-expressing tumors in averagewere shown to be significantly lower (P = 0.0286) than in the respective mock-treated controls (see Figure 6B). This re-sult indicates that hGBP-1 may inhibitVEGF-A release from tumor cells, whichmay in turn result in decreased tumorvascularization and inhibition of tumorcell proliferation in vivo.

DISCUSSIONIFN-γ coordinates diverse cellular pro-

grams through transcriptional regulationof immunologically relevant genes. Mul-tiple cellular effects of IFN-γ have beendescribed, including upregulation ofpathogen recognition, antigen processingand presentation, immune modulationand leukocyte trafficking, and inductionof the antiviral state as well as inhibitionof cellular proliferation and effects on

apoptosis (36). The latter effects havebeen implicated in the more recently ac-knowledged antitumoral activities ofIFN-γ, in addition to an IFN-γ–mediatedtumor angiostasis and initiation of anti-tumoral innate and adoptive immune responses (1). Within the broad spec-trum of IFN-γ–induced genes, those en-coding GBPs are among the most abun-dant. In humans, hGBP-1 is the majorIFN-γ–induced protein (32,37). Initial ob-servations indicated an antiviral activity(38) of this molecule; however, within thelast few years a major role of hGBP-1 inthe regulation of endothelial cell functionhas been demonstrated. Thus, hGBP-1was shown to inhibit proliferation (17)and migration (19) as well as capillaryformation and invasiveness (16) of en-dothelial cells and is required to transferthe antiangiogenic activities of inflamma-tory cytokines such as IFN-γ, IL-1 andtumor necrosis factor (TNF)-α onto en-dothelial cells (16,17,19). In addition, re-cent evidence is accumulating showingthat hGBP-1 is involved in antitumoralresponse processes. In a study involving

388 patients with colorectal carcinomas(20), robust hGBP-1 expression was de-tected in 30% of the tumors in endothe-lial cells and monocytes of the desmo-plastic stroma. Most interestingly, thisfinding was associated with a reducedtumor angiogenesis and a highly signifi-cantly improved prognosis for the pa-tients (20). Although GBPs from otherspecies, mice in particular, are less wellstudied with respect to their biologicalfunction, an antitumorigenic activity ofmurine GBP-1 has been recently sug-gested by a significant upregulation ofthe respective gene in prostate tumors,which regressed during IFN-α therapy(39). This finding corresponds to previ-ous work implicating IFN-γ–inducedmurine GBP-1 in the regression of mam-mary carcinomas in a Balb/c-derivedtumor mouse model (7). Moreover, anantiproliferative effect of IFN-γ has beendescribed for normal human mammaryepithelial cells, and interestingly, mam-mary cancer cell lines with missing or reduced GBP-1 expression have beendemonstrated to be resistant to IFN-γ– induced growth arrest (40).

In the present study, using an in-ducible, IFN-γ–independent expressionsystem, we could demonstrate thathGBP-1 significantly inhibits the growthof highly malignant TS/A mammary car-cinoma cells in syngeneic immune-com-petent Balb/c mice in vivo. In light of pre-vious data described above, this findingmight suggest concordant functions ofmurine and human GBP-1. The hGBP-1–mediated antitumorigenic effect in thismodel was associated with reducedtumor cell proliferation, as demonstratedby a significantly decreased number ofKi-67–positive cells in TS/A cell–derivedtumors expressing high levels of hGBP-1(see Figure 3B). Because hGBP-1 expres-sion in vitro affected neither tumor cellproliferation nor tumorigenicity (see Fig-ure 1B, C), the effects observed in vivoare most likely due to more indirect,paracrine effects. In this respect, effectsresulting from interactions of tumor cellswith the surrounding stroma or potential immune reactions are of particular inter-

1 8 4 | L I P N I K E T A L . | M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0

A N T I T U M O R I G E N I C A C T I V I T Y O F h G B P - 1

Figure 4. Quantification of tumor infiltration. Sections of TS/A-pUHD10.3-GBP–derived tu-mors from four different animals that had been treated with doxycycline (TS/A-GBP + dox)or saline (TS/A-GBP – dox) were analyzed by immunohistochemistry for the presence ofCD3+ T cells or B220+ B cells. In both experiments, 10 different optical fields were evalu-ated. Results are indicated as mean values ± SD. No statistically significant differenceswere observed. Examples of successful staining are shown (inserts).

est. The immune response in the presentstudy was investigated at the T- andB-cell level. Evaluation of immunohisto-chemical analyses revealed neither signsof inflammation nor elevated levels of infiltrating lymphocytes (see Figure 4).Interestingly, analysis of tumor angiogen-esis revealed significantly reduced con-centrations of hemoglobin in tumors expressing hGBP-1 compared with non-expressing samples (see Figure 5B). Be-cause microvessel density was not af-fected (data not shown), however, thisresult indicates that quality rather thanquantity of tumor vasculature might beimpaired. Unlike normal blood vessels,tumor vessels are often immature and ofirregular shape and diameter, and revealan increased permeability. This may resultin an increased interstitial fluid pressureand decreased perfusion (41,42). One ofthe major molecules regulating blood ves-sel formation and function is VEGF-A. Itcontrols proliferation, survival and migra-tion of endothelial cells as well as angio-genic sprouting (43). By analyzing the re-lease of VEGF-A from hGBP-1–expressingTS/A cells in vitro (see Figure 6A), amarked decrease (up to 60%) was ob-served in comparison to nonexpressingcells. The slight decrease (5–25%) ob-served in mock-treated TS/A cells lack-ing the hGBP-1 expression constructmight reflect reduced cellular metabolicactivities under the applied conditions.Quantitative analysis of VEGF-A proteinlevels in tumor samples by means ofWestern blotting (see Figure 6B) consis-tently revealed a significantly reducedamount of VEGF-A in hGBP-1– expressingtumors compared with mock-treated con-trols. Significantly lower hemoglobin lev-els, indicating less tissue perfusion in ac-cordance with a significantly reducedtumor cell proliferation, as observed inhGBP-1 expressing tumors, suggests thatan hGBP-1–mediated decrease in VEGF-A synthesis may account for reduced an-giogenesis and tumor growth in vivo. Theunderlying detailed mechanisms andhow IFN-mediated effects might add tothis antitumorigenic activity will be thesubject of further investigations.

R E S E A R C H A R T I C L E

M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0 | L I P N I K E T A L . | 1 8 5

Figure 5. Analysis of tumor angiogenesis. (A) Immunohistochemistry. Consecutive sections ofhGBP-1–expressing (TS/A-GBP + dox) and nonexpressing (TS/A-Hyg + dox, TS/A-GBP – dox)tumors were stained for the presence of hGBP-1 (α-hGBP-1, brown cytoplasmic staining)and CD31 (α-mCD31), a specific marker of endothelial cells (red staining). (B) Biochemicaldetermination of hemoglobin contents. The amount of hemoglobin, normalized to totalprotein amounts, was determined in frozen tissue (n = 18) of hGBP-1–expressing (TS/A-GBP+)and nonexpressing (TS/A-GBP–) tumors. Statistical significance (P = 0.0032) of the encoun-tered differences was shown by applying an unpaired Student t test.

In summary, we have demonstratedthat hGBP-1 may significantly contributeto an antitumoral response reaction of thehost. Although the ectopic high-level ex-pression in the present study might notreflect a physiological situation, recent re-ports have demonstrated hGBP-1 expres-sion in human prostate cancers in stroma(including endothelial and inflammatorycells), as well as in tumor cells (39). Simi-lar findings were obtained analyzingsamples of human cholangiocellular carci-nomas (our unpublished data). The mech-anisms underlying this antitumorigeniceffect may include autocrine processes affecting cell proliferation, but alsoparacrine activities affecting the expres-sion and release of protumorigenic fac-tors such as VEGF. Thus, owing to theseactivities, GBPs may represent potentmembers of an innate, IFN-γ–induced an-titumoral defense system.

ACKNOWLEDGMENTSWe acknowledge the excellent techni-

cal assistance of Doris Rosenfellner andMelanie Nurtsch in immunohistochem-istry. We also thank Daniel Portsmouthfor critical reading of the manuscript.This work was partially funded by theZIT program co-operate Vienna 2003 (toC Hohenadl and M Stürzl) and by grantsof the Interdisciplinary Center of ClinicalResearch (IZKF) of the University of Er-langen, the Deutsche Forschungsgemein-schaft (DFG 317/2-1) and the FederalMinistry of Education and Research(BMBF, Polyprobe Study) to M Stürzl.

DISCLOSUREThe authors declare that they have no

competing interests as defined by Molec-ular Medicine, or other interests thatmight be perceived to influence the re-sults and discussion reported in thispaper.

REFERENCES1. Dunn GP, Koebel CM, Schreiber RD. (2006) Inter-

ferons, immunity and cancer immunoediting.Nat. Rev. Immunol. 6:836–48.

2. Kim R, Emi M, Tanabe K. (2007) Cancer immu-noediting from immune surveillance to immuneescape. Immunology 121:1–14.

1 8 6 | L I P N I K E T A L . | M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0

A N T I T U M O R I G E N I C A C T I V I T Y O F h G B P - 1

Figure 6. Effect of hGBP-1 expression on VEGF-A synthesis. (A) The Tet-regulated hGBP-1– ex-pressing TS/A cell clones 14 and 18, which were used for generation of subcutaneous tu-mors in Balb/c mice, were analyzed for VEGF-A synthesis applying a mouse-specific ELISA.As a control, cells harboring the parental vector pUHD were used. Cells (duplicates) werecultured for 96 h in the presence of doxycycline (1 and 2.5 μg/mL), and the respective cellculture supernatants were analyzed. Obtained values (mean ± SD) were normalized to cellnumbers (1 × 106) and set in relation to values obtained with nontreated cells (0 μg/mLdox). Corresponding cell extracts were analyzed for hGBP-1 expression by Western blotting(insert). (B) Quantitative Western blot analysis of VEGF-A in TS/A cell–derived tumor sam-ples. Frozen tissue samples (n = 6) of tumors grown in mice that had received doxycyclinefor induction of hGBP-1 expression (TS/A-GBP+) or were injected with saline (TS/A-GBP–)were subjected to Western blot analysis to detect murine VEGF-A protein. GAPDH wasstained as an internal control to monitor the amount of loaded protein. Quantification ofband intensities was performed with the ImageQuant software. Background correctedsingle values were normalized to measurements of GAPDH-specific bands and summa-rized, and the mean values (± SEM) were compared for statistically significant differencesby applying an unpaired Student t test. The calculated P value is indicated.

3. Dighe AS, Richards E, Old LJ, Schreiber RD.(1994) Enhanced in vivo growth and resistance torejection of tumor cells expressing dominant neg-ative IFN gamma receptors. Immunity 1:447–56.

4. Kaplan DH, et al. (1998) Demonstration of an in-terferon gamma-dependent tumor surveillancesystem in immunocompetent mice. Proc. Natl.Acad. Sci. U. S. A. 95:7556–61.

5. Kundu N, Beaty TL, Jackson MJ, Fulton AM.(1996) Antimetastatic and antitumor activities ofinterleukin 10 in a murine model of breast can-cer. J. Natl. Cancer Inst. 88:536–41.

6. Kundu N, et al. (1998) Interleukin-10 gene trans-fer inhibits murine mammary tumors and ele-vates nitric oxide. Int. J. Cancer 76:713–9.

7. Sun H, Jackson MJ, Kundu N, Fulton AM. (1999)Interleukin-10 gene transfer activates interferon-gamma and the interferon-gamma-induciblegenes Gbp-1/Mag-1 and Mig-1 in mammary tu-mors. Int. J. Cancer 80:624–9.

8. Walser TC, et al. (2007) Immune- mediated mod-ulation of breast cancer growth and metastasisby the chemokine Mig (CXCL9) in a murinemodel. J. Immunother. 30:490–8.

9. Cheng YS, Colonno RJ, Yin FH. (1983) Interferoninduction of fibroblast proteins with guanylatebinding activity. J. Biol. Chem. 258:7746–50.

10. Vestal DJ. (2005) The guanylate-binding proteins(GBPs): proinflammatory cytokine-inducedmembers of the dynamin superfamily withunique GTPase activity. J. Interferon Cytokine Res.25:435–43.

11. Kresse A, et al. (2008) Analyses of murine GBPhomology clusters based on in silico, in vitro andin vivo studies. BMC Genomics 9:158.

12. Degrandi D, et al. (2007) Extensive characteriza-tion of IFN-induced GTPases mGBP1 to mGBP10involved in host defense. J. Immunol. 179:7729–40.

13. Olszewski MA, Gray J, Vestal DJ. (2006) In silicogenomic analysis of the human and murineguanylate-binding protein (GBP) gene clusters.J. Interferon Cytokine Res. 26:328–52.

14. Prakash B, Renault L, Praefcke GJ, Herrmann C,Wittinghofer A. (2000) Triphosphate structure ofguanylate-binding protein 1 and implications fornucleotide binding and GTPase mechanism.EMBO J. 19:4555–64.

15. Prakash B, Praefcke GJ, Renault L, Wittinghofer A,Herrmann C. (2000) Structure of human guanylate-binding protein 1 representing a unique class ofGTP-binding proteins. Nature 403:567–71.

16. Guenzi E, et al. (2003) The guanylate bindingprotein-1 GTPase controls the invasive and an-giogenic capability of endothelial cells throughinhibition of MMP-1 expression. EMBO J.22:3772–82.

17. Guenzi E, et al. (2001) The helical domain ofGBP-1 mediates the inhibition of endothelial cellproliferation by inflammatory cytokines. EMBOJ. 20:5568–77.

18. Naschberger E, Bauer M, Stürzl M. (2005)Human guanylate binding protein-1 (hGBP-1)characterizes and establishes a non-angiogenic

endothelial cell activation phenotype in inflam-matory diseases. Adv. Enzyme Regul. 45:215–27.

19. Weinländer K, et al. (2008) Guanylate bindingprotein-1 inhibits spreading and migration of en-dothelial cells through induction of integrinalpha4 expression. FASEB J. 22:4168–78.

20. Naschberger E, et al. (2008) Angiostatic immunereaction in colorectal carcinoma: impact on sur-vival and perspectives for antiangiogenic ther-apy. Int. J. Cancer 123:2120–9.

21. Gorbacheva VY, Lindner D, Sen GC, Vestal DJ.(2002) The interferon (IFN)-induced GTPase,mGBP-2: role in IFN-gamma-induced murine fi-broblast proliferation. J. Biol. Chem. 277:6080–7.

22. Nanni P, de Giovanni C, Lollini PL, Nicoletti G,Prodi G. (1983) TS/A: a new metastasizing cellline from a BALB/c spontaneous mammary ade-nocarcinoma. Clin. Exp. Metastasis 1:373–80.

23. Ausserlechner MJ, Obexer P, Deutschmann A,Geiger K, Kofler R. (2006) A retroviral expressionsystem based on tetracycline-regulated tri-cistronic transactivator/repressor vectors forfunctional analyses of antiproliferative and toxicgenes. Mol. Cancer Ther. 5:1927–34.

24. Agu CA, et al. (2006) The cytotoxic activity of thebacteriophage lambda-holin protein reduces tu-mour growth rates in mammary cancer cellxenograft models. J. Gene Med. 8:229–41.

25. Naschberger E, et al. (2006) Human guanylatebinding protein-1 is a secreted GTPase present inincreased concentrations in the cerebrospinalfluid of patients with bacterial meningitis. Am. J.Pathol. 169:1088–99.

26. Lubeseder-Martellato C, et al. (2002) Guanylate-binding protein-1 expression is selectively in-duced by inflammatory cytokines and is an acti-vation marker of endothelial cells duringinflammatory diseases. Am. J. Pathol. 161:1749–59.

27. Zabala M, et al. (2004) Optimization of the Tet-onsystem to regulate interleukin 12 expression inthe liver for the treatment of hepatic tumors.Cancer Res. 64:2799–804.

28. Beckstead JH. (1994) A simple technique forpreservation of fixation-sensitive antigens inparaffin-embedded tissues. J. Histochem. Cy-tochem. 42:1127–34.

29. Beckstead JH. (1995) A simple technique forpreservation of fixation-sensitive antigens inparaffin-embedded tissues: addendum. J. His-tochem. Cytochem. 43:345.

30. Zeng G, Gao L, Birkle S, Yu RK. (2000) Suppres-sion of ganglioside GD3 expression in a rat F-11tumor cell line reduces tumor growth, angiogen-esis, and vascular endothelial growth factor pro-duction. Cancer Res. 60:6670–76.

31. Urruticoechea A, Smith IE, Dowsett M. (2005)Proliferation marker Ki-67 in early breast cancer.J. Clin. Oncol. 23:7212–20.

32. Naschberger E, et al. (2004) Nuclear factor- kappaB motif and interferon-alpha-stimulated re-sponse element co-operate in the activation ofguanylate-binding protein-1 expression by in-flammatory cytokines in endothelial cells.Biochem. J. 379:409–20.

33. Baluk P, McDonald DM. (2008) Markers for mi-croscopic imaging of lymphangiogenesis and an-giogenesis. Ann. N. Y. Acad. Sci. 1131:1–12.

34. Russo A, et al. (2005) Inhibition of granuloma- associated angiogenesis by controlling mast cellmediator release: role of mast cell protease-5. Br.J. Pharmacol. 145:24–33.

35. Norrby K. (2006) In vivo models of angiogenesis.J. Cell Mol. Med. 10:588–612.

36. Schroder K, Hertzog PJ, Ravasi T, Hume DA.(2004) Interferon-gamma: an overview of signals,mechanisms and functions. J. Leukoc. Biol.75:163–89.

37. Tripal P, Bauer M, Naschberger E, et al. (2007)Unique features of different members of thehuman guanylate-binding protein family. J. Inter-feron Cytokine Res. 27:44–52.

38. Anderson S, Carton J, Lou J, Xing L, Rubin B.(1999) Interferon-induced guanylate bindingprotein-1 (GBP-1) mediates an antiviral effectagainst vesicular stomatitis virus and en-cephalomyocarditis virus. Virology 256:8–14.

39. Persano L, et al. (2009) Interferon-alpha counter-acts the angiogenic switch and reduces tumorcell proliferation in a spontaneous model of pro-static cancer. Carcinogenesis 30:851–60.

40. Harvat BL, Jetten AM. (1996) Gamma-interferoninduces an irreversible growth arrest in mid-G1in mammary epithelial cells which correlateswith a block in hyperphosphorylation ofretinoblastoma. Cell Growth Differ. 7:289–300.

41. Jain RK. (2005) Normalization of tumor vascula-ture: an emerging concept in antiangiogenic ther-apy. Science 307:58–62.

42. Jain RK. (2003) Molecular regulation of vesselmaturation. Nat. Med. 9:685–93.

43. Ellis LM, Hicklin DJ. (2008) VEGF-targeted ther-apy: mechanisms of anti-tumour activity. Nat.Rev. Cancer 8:579–91.

R E S E A R C H A R T I C L E

M O L M E D 1 6 ( 5 - 6 ) 1 7 7 - 1 8 7 , M A Y - J U N E 2 0 1 0 | L I P N I K E T A L . | 1 8 7