in vivo antitumor activity of a recombinant il7/il15 ...in vivo antitumor activity of a recombinant...

Large Molecule Therapeutics

In Vivo Antitumor Activity of a RecombinantIL7/IL15 Hybrid Cytokine in MiceYinhong Song1,2, Yalan Liu1, Rong Hu1, Min Su1, Debra Rood1, and Laijun Lai1,3

Abstract

Both IL7 and IL15 have become important candidate immu-nomodulators for cancer treatment. However, IL7 or IL15 usedalone suffers from shortcomings, such as short serumhalf-life andlimited antitumor effect. We have cloned and expressed a recom-binant (r) IL7/IL15 fusion protein in which IL7 and IL15 arelinked by a flexible linker. We then compared the antitumor effectof rIL7/IL15 with the individual factors rIL7 and/or rIL15. Weshow here that rIL7/IL15 has a higher antitumor activity than thecombination of the individual factors in both murine B16F10

melanoma and CT-26 colon cancer models. This was associatedwith a significant increase in tumor infiltration of T cells, DCs, andNK cells and a decrease in regulatory T cells (Tregs). In addition,rIL7/IL15-treated DCs had higher expression of costimulatorymolecules CD80 and CD86. The higher antitumor activity ofrIL7/IL15 is likely due to its longer in vivo half-life and differenteffects on immune cells. Our results suggest that rIL7/IL15 mayoffer a new tool to enhance antitumor immunity and treat cancer.Mol Cancer Ther; 15(10); 2413–21. �2016 AACR.

IntroductionBoth IL7 and IL15 are the common cytokine receptor g–chain

(gc) family cytokines. IL7 plays a central role in the developmentandmaintenance of T cells (1–5). The IL7 receptor (R) consists oftwo subunits, the IL7Ra and gc (1–5), the latter also being acomponent of the receptors for IL2, IL4, IL9, IL15 and IL21. IL7Rais expressed by T cells and DCs, etc. (1–6). Several studies haveshown that IL7 has antitumor activity (7–12). For example, tumorcell lines that were transfected to produce rIL7 locally reducedtumorigenicity in vivo, which was dependent on CD4þ or CD8þ Tcells (7–9, 11). Local or systemic administration of rIL7 also hadantitumor effects (7, 10), especiallywhen rIL7was combinedwithcancer vaccines (10, 12).

IL15 induces the differentiation and proliferation of T andNK cells, enhances the cytolytic activity of CD8þ T cells, andinduces the maturation of DCs (13). The IL15 receptor iscomposed of a unique a subunit (IL15Ra), a b subunit(IL2R/15Rb) that is shared with the IL2 receptor, and the gc.IL15 can bind to IL15Rb via cis- or trans-presentation byIL15Ra. IL15Rb is expressed by multiple lymphoid popula-tions, such as T cells, DCs, NK cells, and NKT cells, etc. (14). Invivo administration of IL15 has antitumor effects in severalmouse tumor models (15–22); however, it has been shown thatadministration of IL15 alone is not optimal (13). Various

combination strategies have been explored to increase theefficacy of IL15 immunotherapy, including coadministrationof other cytokines or inhibitory antibodies against immune-suppression molecules (17–19, 22). These approaches pro-duced greater antitumor responses than did IL15 alone.

Because both rIL7 and rIL15 have a low molecular weight,they undergo a rapid renal clearance in vivo, thereby having ashort plasma half-life (23–26), which diminishes their in vivoantitumor effects (24, 25). Linking the coding sequences ofIL15 with other proteins has increased the IL15's plasma half-life. Some of the fusion proteins also increase IL15 signaling,thereby increasing its efficacy (24, 25, 27–29).

We have previously described a single-chain rIL-7/HGFbhybrid cytokine liking IL7 and HGFb by a flexible linker(30). The in vivo half-life of rIL7/HGFb was significantly longerthan that of rIL7 (31). In addition, rIL7/HGFb has differenteffects on immune cells (30–32), resulting in a higher antitumoractivity (33), as compared with the individual factors rIL7and/or rHGFb. Here we sought to determine whether a single-chain recombinant hybrid cytokine containing IL7 and IL15might have a higher antitumor effect than the combination ofthe individual factors rIL7 and rIL15.

Materials and MethodsAnimals and cell lines

Murine B16F10 melanoma and CT-26 colon cancer cellswere obtained from the ATCC and the National Cancer Insti-tute in 2008. ATCC characterized the cells by using karyotypingand cytochrome c oxidase I testing. We passaged the cells forless than 4 months before storing them in liquid nitrogen.Before the cancer cells were injected into mice, they werecultured in DMEM (Invitrogen) supplemented with 10%FBS at 37°C and 5% CO2 humidified air. Cell viability wasassessed using a Trypan blue exclusion dye method. Cellswith a viability >95% were resuspended in PBS for injection.BALB/c, C57BL/6 mice, homozygous C57BL/6 IFNg null mice(IFN-g�/�), and homozygous TNFa-null mice (TNF-a�/�)

1Department of Allied Health Sciences, University of Connecticut,Storrs, Connecticut. 2Medical College, Three Gorges University,Yichang, China. 3University of Connecticut Stem Cell Institute, Uni-versity of Connecticut, Storrs, Connecticut.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Laijun Lai, Department of Allied Health Sciences,University of Connecticut, 1390 Storrs Road, Storrs, CT 06269. Phone: 860-486-6073; Fax: 860-486-0534; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-16-0111

�2016 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org 2413

on October 17, 2020. © 2016 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst July 29, 2016; DOI: 10.1158/1535-7163.MCT-16-0111

http://mct.aacrjournals.org/

were purchased from The Jackson Laboratory. Mice werehoused, treated, and handled in accordance with protocolsapproved by the Institutional Animal Care and Use Committeeof the University of Connecticut.

Construction, expression, and purification of human single-chain IL7/IL15 hybrid cytokine

The human IL7/IL15 gene was constructed by adapting aprotocol we previously used to construct the human IL7/HGFagene (32). Briefly, the human IL7 cDNA was amplified withprimers A [that contains a secret sequence (SS)] and B, and thehuman IL15 cDNA with primer C and D (Supplementary TableS1). The PCR products of IL7 and IL15 were combined andsubjected to an additional round of PCR with primers A andD. Because primers B and C contained the linker sequenceencoding (Gly4Ser)2, the IL7 and IL15 genes (IL7/IL15) wereconnected by a flexible linker after the overlap extension PCR.The IL7/IL15 gene was then cloned into a pOptiVEC mammalianexpression vector (Invitrogen; Fig. 1A) that was then transfectedinto Chinese hamster ovary cell–derived DG44 cells (Invitrogen)to produce rIL7/IL15 protein.

rIL7/IL15 protein was then purified from the supernatant ofthe transfected DG44 cells. Briefly, the supernatant was con-centrated by a prep/scale–tangential flow filter cartridge with10-kDa molecular weight cut-off (Millipore) and diafilteredinto washing buffer. The sample was then applied to seriallylinked columns of CM and DEAE sepharose (GE Health CareBiosciences); after washing, the linked columns were separated.rIL7/IL15 protein was eluted from the DEAE column in thewashing buffer containing NaCl gradient, and further purifiedby a gel filtration column (16/60 Sephacryl S-100 high reso-lution, GE). The purified protein was analyzed by SDS-PAGEand Western blotting using antibodies against human IL7 andIL15. For controls, we also cloned and expressed human IL7(32) or IL15 (using primers E and D) gene individually, andpurified rIL7 and rIL15 proteins from the expression system,respectively.

Evaluation of local tumor growth and pulmonary metastasisTo induce localized tumors, 1 � 105 B16F10 melanoma

cells or 2 � 105 CT-26 colon cancer cells were injectedsubcutaneously into the flank of syngeneic C57BL/6 orBALB/c mice, respectively. The indicated doses of rIL7/IL15,rIL7, and/or rIL15 (or PBS) were then injected into the tumorinjection site at 2-day intervals over the indicated time period.Tumor size (volume) was determined every other day bycaliper measurements of the shortest (A) and longest (B)diameter, using the formula V ¼ (A2B)/2. Some mice werealso injected intraperitoneally with 450 mg anti-IL7Ra anti-body (clone A7R34; ref. 31), 200 mg anti-IL15Rb antibody(clone TMb1, from Biolegend), or isotype controls (31, 34) 1day before the cytokine injection.

For in vivo cell depletions (CD8 T cells, CD4 T cells, or NKcells), mice received the following antibodies via intraperi-toneal injections: anti-CD8 (clone 2.43, from BioXCell), 500mg/injection; anti-CD4 (clone GK1.5), 200 mg/injection; oranti-NK1.1 (clone PK136), 300 mg/injection on days �3, �1,and þ4 of the cancer cell injection (35). Depletions wereconfirmed by flow cytometry analysis of blood samples.

To induce pulmonary metastases, 1 � 105 B16F10 melanomacells were injected into the tail vein of syngeneic mice, and rIL7/15, rIL7, and rIL15 or PBS were injected intravenously at 2-dayintervals from days 2 to 12. The animals were euthanized at theindicated times after tumor inoculation. Metastatic tumornodules in the subpleural regions of the lungs were countedunder a dissecting microscope.

Confocal microscopyFrozen sections of tumor tissue were prepared as described

previously (36). The sectionswere stainedwith antibodies includ-ing PE-conjugated anti-mouse CD11c, APC-conjugated anti-mouse CD4, and FITC-conjugated anti-mouse CD8 or FITC-conjugated anti-mouse NK1.1 (BioLegend, or BD Biosciences).All of the sectionswere then counterstainedwith 40, 60-diamidino-2-phenylindole (DAPI; Sigma) and observed under a Nikon A1R

IL7

IL7 IL15

PCR PCR

IL7 IL15

pOptiVEC Vector

A B

pOptiVEC Vector containing human IL7/IL15 gene

Primer A containing SS

Primer B containing linker

Primer C containing linker

IL15

Primer D

IL7 IL15

IL7

IL15

Primer A

Primer D PCR

Ligation

Figure 1.

Strategy for molecular cloning ofIL7/IL15 and characterization of purifiedrIL7/IL15 protein. A, IL7, linker, and IL15cDNAs were constructed byoverlapping PCR. IL7/IL15 cDNA wasthen ligated into the pOptiVEC vector.B, lane 1, MW markers; L2, Coomassieblue-stained SDS-PAGE; L3, Westernblot analysis with anti-human IL7antibody; L4, Western blot analysiswith anti-human IL15 antibody.

Song et al.

Mol Cancer Ther; 15(10) October 2016 Molecular Cancer Therapeutics2414




Spectral Confocal microscope (Nikon). Aminimum of 6 sectionsfrom each tumor tissue were used to evaluate CD4þ, CD8þ,CD11cþ, and NK1.1þ cells.

Flow cytometrySingle-cell suspensions of tumor tissues and spleen cells were

stained with one or more of the following fluorochrome-conju-gated antibodies: CD3, CD4, CD8, CD25, FoxP3, CD11c, CD80,CD86, NK1.1, IFNg , and TNFa (BioLegend, BD Biosciences, oreBioscience). For intracellular staining, cells were labeledwith cellsurface antibodies, permeabilized with a Cytofix/Cytoperm solu-tion (BD Biosciences), and then stained with anti-FoxP3, IFNg , orTNFa antibody. The samples were analyzed on a FACSCaliburflow cytometer (BD Biosciences), and data analysis was per-formed using FlowJo software.

Evaluation of T cells producing IFNg or TNFaSplenocytes were treated with red blood cells lysis buffer,

washed, and cultured with irradiated CT-26 colon cancer cells(1 � 105/well) or B16F10 melanoma cells (0.5 � 105/well) for 2days. Suspended cells were collected, stained, with antibodiesagainst CD4, CD8, and IFNg or TNFa, and analyzed by flowcytometry.

ELISA for detection of IL7, IL15, and rIL7/IL15The concentration of IL7, IL15, and rIL7/IL15 in serum was

determined by Human IL-7 ELISA Kit (Abcam) and IL15 ELISAReady-DET-Go kit (eBioscience) according to the manufacturer'sinstructions. The serumhalf-life of the cytokines was calculated asdescribed previously (31).

Limulus amebocyte lysate assayThe endotoxin level in purified proteins was determined by the

endpoint chromogenic LAL test according to the manufacturer'sinstructions (Lonza).

Statistical analysisStudent two-tailed unpaired t test was used for comparisons

between two groups, and one-way ANOVA for more thantwo groups. The results were considered significantly differentat P < 0.05.

ResultsExpression and purification of human single-chain rIL7/IL15protein

We have successfully constructed the human IL7/IL15 gene inwhich the IL7 and IL15 cDNAs were connected by a flexible linker(Fig. 1A). The IL7/IL15 gene was transfected and expressed in thepOptiVEC/DG44 mammalian expression system, and rIL7/IL15protein was then purified by ion exchanges and gel filtration. Asshown in Fig. 1B, a relatively high purity of rIL7/IL15 protein wasobtained, as determined by Coomassie blue-stained SDS-PAGE(lane 2). The identity of the protein was verified by Western blotusing IL7 and IL15 antibodies (lane 3 and lane 4). The actualmolecular weight (MW) of the rIL7/IL15 was higher than thepredicted MW, suggesting that the recombinant protein wasglycosylated. The endotoxin level was less than 0.01 EU/mL of1mg of purified rIL7/IL15. In vitro assays show that rIL7/IL15 had aslightly lower activity than equimolar amounts of unfused rIL7and rIL15 in stimulating the proliferation of responding lympho-

cytes (Supplementary Fig. S1), probably because the fusion affectsthe binding of rIL7/IL15 to its receptors.

In vivo administration of rIL7/IL15 inhibits local tumorgrowth

To determine the antitumor activity of rIL7/IL15, B16F10melanoma cells were injected subcutaneously into the flank ofsyngeneicC57BL/6mice. Themicewere then injected at the tumorsite with different doses of rIL7/IL15 (2.5, 5, 10, and 20 mg/injection) or control vehicle (PBS) at 2-day intervals from days 2to 12. Tumor volumes in each groupweremeasured over time andcompared statistically. As shown in Fig. 2A, tumor growth wasinhibited by rIL7/IL15 in a dose-responsive manner, with about60% inhibition at the 2.5-mg level and greater than 80% inhibi-tion at the 20-mg level.

To compare the antitumor effect of rIL7/IL15 with its compo-nent cytokines, C57BL/6 mice injected with B16F10 melanomacells were treated with equimolar amounts of rIL7/IL15 (20 mg/injection), rIL7 (11.6 mg/injection), and/or rIL15 (8.4 mg/injec-tion), according to the above schedule. As shown in Fig. 2B,although individual factors rIL7 and/or rIL15 inhibited localtumor growth, the single-chain rIL7/IL15 hybrid cytokine hadsignificantly higher antitumor activity than rIL7 and/or rIL15.

To test whether the antitumor effect of rIL7/IL15 ismediated byboth IL7 and IL15, some of themice were also injected IL7 or IL15receptor-blocking antibody (anti-IL7Ra or anti-IL15Rb anti-body), or isotype control. As shown in Fig. 2C, either IL7 or IL15receptor-blocking antibody partly abolished the antitumor activ-ity of rIL7/IL15. The data suggest that the antitumor effect is IL7and IL15-specific.

To determine whether rIL7/IL15 has an antitumor effect onother cancers, BALB/c mice were injected subcutaneously withmurine CT-26 colon cancer cells, then injected at the tumor sitewith equimolar amounts of rIL7/IL15 (20 mg/injection), rIL7(11.6 mg/injection) and/or rIL15 (8.4 mg/injection) at 2-dayintervals from days 2 to 20. Again, rIL7/IL15 had a higherantitumor activity than rIL7 and/or rIL15 (Fig. 2D).

To determine whether rIL7/IL15, rIL7, and/or rIL15 directlyaffect the growth of tumor cells, CT-26 colon and B16F10 mel-anoma cancer cells were cultured in vitro for 2 to 7 days in thepresence of 20 to 150 ng/mL rIL7/IL15, rIL7, and/or rIL15, or PBS.The rate of tumor cell growth was not significantly different at anydose level of rIL7/IL15 from those observed in cultures containingPBS, or rIL7, and/or rIL15 (Supplementary Fig. S2). Hence, themechanism by which rIL7/IL15 inhibits the growth of B16F10melanoma or CT-26 colon cancer in vivo would not appear toinvolve direct cytotoxic or cytostatic activities.

rIL7/IL15 induces significant infiltration of T cells, NK cells,NKT cells, and DCs, but decreases infiltration of Tregs into thetumors

To determine the mechanisms by which rIL7/IL15 has antitu-mor activity, we assessed the percentage of immune cells in tumortissues. On day 14, after B16F10 cancer cell inoculation, single-cell suspensions of tumor tissues from mice treated with equi-molar amounts of rIL7/IL15, rIL7, and/or rIL15 (Fig. 2B) wereanalyzed for CD4þ and CD8þ T cells, CD4þCD25þFoxp3þ Tregs,CD3�NK1.1þ NK cells, CD3þNK1.1þ NKT cells, and CD11cþ

DCs by flow cytometry. As shown in Fig. 3A and B, rIL7 increasedthe tumor infiltration of bothCD8þ andCD4þ T cells, while rIL15increased the percentage of CD8þ T cells, but not that of CD4þ T

In Vivo Antitumor Activity of a Recombinant IL7/IL15

www.aacrjournals.org Mol Cancer Ther; 15(10) October 2016 2415




cells. The combination of rIL7 and rIL15 significantly increasedthe percentage both CD4þ and CD8þ T cells. However, rIL7/IL15hybrid cytokine hadhigher activity than rIL7 and/or rIL15 (Fig. 3Aand B).

Because Tregs can inhibit immune functions and enhance thegrowth of cancer cells in vivo (37), we determined whether rIL7/IL15 treatment affected the immune-suppressive cells intumors. rIL7 and/or rIL15 did not significantly affect infiltrationof CD4þCD25þFoxp3þ Tregs in the tumors (Fig. 3A). In con-trast, rIL7/IL15 treatment significantly reduced the percentageof Tregs (Fig. 3A).

Similar to the effect on CD8þ T cells, rIL7 and/or rIL15increased tumor infiltration of NK cells and NKT cells, andrIL7/IL15 had a higher activity than the individual factors rIL7and/or rIL15 on NK cells (Fig. 3A and C). However, rIL7/IL15 didnot have a higher activity than rIL7 and rIL15 onNKT cells (Fig. 3Aand C).

Because DCs play a critical role in the activation of T cells,we analyzed CD11cþ DCs in the tumors. As shown in Fig. 3A,rIL7 did not significantly affect tumor infiltration of DCs. Incontrast, rIL15 increased the percentage of DCs in the tumors.rIL7/IL15 had a higher effect in the tumor infiltration of DCsthan rIL15 used alone or mixed with rIL7 (Fig. 3A). We alsoexamined the expression of CD80 and CD86 on the DCs.Again, rIL7 did not significantly affect DCs, and rIL15increased the expression of CD80 and CD86 on the DCs.rIL7/IL15 treatment resulted in a higher expression of CD80and CD86 than rIL15 used alone or mixed with rIL7 (Fig. 3D),indicating that the DCs had undergone activation and matu-ration after rIL7/IL15 treatment.

The increased infiltration of CD4þ and CD8þ T cells, DCs, andNK cells in the rIL7/IL15-treated tumors was confirmed by con-focal microscopic observations (Fig. 3E and F).

In addition to the tumors themselves, there was a parallelincrease in the numbers of CD4þ and CD8þ T cells, NK cells,and activated DCs, and a decrease in the number of Tregs in thespleen of the rIL7/IL15-treated melanoma-bearing mice (Supple-mentary Fig. S3).

To determine whether CD4 and CD8 T cells and NK cellsmediated the antitumor activity, some of the B16F10 bearing-mice were also injected with anti-CD8, anti-CD4, or anti-NK1.1antibody. As shown in Fig. 3G, depletion ofCD4orCD8T cells, orNK cells partly abrogated the antitumor effect. The data suggestthat CD4 and CD8 T cells, and NK cells mediated the antitumoractivity of rIL7/IL15.

rIL7/IL15 induces a tumor-specific immunologic responseTo determine whether a tumor-specific immunologic response

has been generated, we examined the percentage of IFNg-pro-ducing CD4þ and CD8þ T cells from the spleens of cytokine-treated tumor-bearing mice after in vitro stimulation with homol-ogous or heterologous tumor cells. To this end, single-cell suspen-sions from the spleens of day 22, cytokine-treated, CT-26 coloncancer–bearing mice were stimulated with CT-26 cells. Cellsstimulated with B16F10 melanoma cells served as specificitycontrols. Two days after the stimulation, the percentage of IFNg-or TNFa-producingCD4þ andCD8þT cellswere analyzedbyflowcytometry.

As shown in Fig. 4A–C, the percentages of IFNg-producingCD4þ andCD8þ T cells among the cultured splenocytes from rIL7and/or rIL15-treated mice were 2.3- to 3.7- and 1.5- to 1.7-fold,respectively, higher than those from PBS-treated mice after stim-ulation with CT-26 cells. In contrast, the percentages of IFNg-producing CD4þ and CD8þ T cells among the cultured spleno-cytes from rIL7/IL15-treated mice were elevated 8- and 2.3-fold,respectively. We also analyzed the percentage of TNFa producing

Figure 2.

rIL7/IL15 inhibits the growth of localizedmelanoma and colon cancer. A–C,C57BL/6 mice were injectedsubcutaneously with 1 � 105 B16F10melanoma cells followed byintratumoral injections with rIL7/IL15(2.5, 5, 10, or 20 mg) or PBS (A);equimolar doses of rIL7/IL15 (20 mg),rIL7 (11.6 mg) and/or rIL15 (8.4 mg) orPBS at 2-day intervals between days 2and 12 after tumor inoculation (B); oranti-IL7Ra antibody, anti-IL15Rbantibody, or isotype control on day 1,and rIL7/IL15 (20 mg) or PBS at 2-dayintervals between days 2 to 12 aftertumor inoculation (C). D, BALB/c micewere injected subcutaneously with 2 �105 CT-26 colon cancer cells, followedby intratumoral injectionswith rIL7/IL15(20 mg), rIL7 (11.6 mg), and/or rIL15 (8.4mg) or PBS at 2-day intervals betweendays 2 and 20 after tumor inoculation.Tumors were measured every otherday. The mean tumor volume (mm3) �SD at the indicated time points isshown. These data are representative oftwo independent experiments with 3–5mice per group with similar results.

Song et al.





Figure 3.

Treatment of localized tumors with rIL7/15 increases the percentages of CD4þ and CD8þ T cells, NK, NKT cells, and DCs, but decreases the percentages of Tregs in thetumors. A–F, C57BL/6 mice were injected subcutaneously with B16F10 melanoma cells and treated with equimolar doses of rIL7/IL15, rIL7 and/or rIL15 asin Fig. 2B. Fourteen days after tumor inoculation, the tumors were harvested. A–D, single-cell suspensions of tumor tissues were analyzed for immune cells by flowcytometry. A, the percentages of CD4þ and CD8þ T cells, CD4þCD25þFoxp3þ Tregs, CD3�NK1.1þ NK cells, CD3þNK1.1þ NKT cells, and CD11þ DCs. B and C,representativeflowcytometricprofiles showing thepercentageofCD4þCD8� andCD4�CD8þTcells (B) andCD3�NK1.1þNKcells andCD3þNK1.1þNKTcells in the rIL7/IL15-treated tumors (C). D, relative fluorescence intensity (MFI) of CD80 and CD86 on the DCs. The mean � SD is shown. These data are representative of twoindependent experimentswith 5mice per groupwith similar results. � , P <0.05 as comparedwith the PBS-treated group; �� , P <0.05 as comparedwith the rIL7þ rIL15-treated groups. E and F, tumor sections were analyzed for presence of CD11cþ cells, CD4þ T cells, CD8þ T cells (E) and NK1.1þ cells by immunofluorescence (F).E, DAPI (blue), CD11cþ (red), CD4þ (yellow), and CD8þ (green). F, DAPI (blue) and NK1.1þ (green). Original magnification, 200�. Representative tumor sections areshown. G, C57BL/6 mice were injected subcutaneously with B16F10 melanoma cells and treated with rIL7/IL15 or PBS as in Fig. 2B. The mice were also injectedintraperitoneally with anti-CD8, anti-CD4, or anti-NK1.1 on days�3,�1, andþ4 of the cancer cell injection. The mean tumor volume (mm3)� SD at the indicated timepoints is shown. These data are representative of two independent experiments with 3–5 mice per group with similar results.






CD4þ and CD8þ T cells among the cultured splenocytes from thecytokine-treated mice. The percentages of TNFa producing CD4þ

and CD8þ T cells from rIL7/IL15-treated mice were also signif-icantly higher than those of the T cells from rIL7 and/or rIL15-treated mice (Fig. 4D–F). These results suggest rIL7/IL15-treat-ment greatly enhances systemic immunologic responses totumor-specific antigens in vivo.

To determine whether IFNg and TNFa play a role in theantitumor effects of rIL7/IL15, IFNg�/� or TNFa�/� mice wereused as the recipients for B15F10melanoma cells. As shown in Fig.4G and H, rIL7/IL15 treatment resulted in less than 40% tumorgrowth inhibition in IFNg�/�or TNFa�/�mice. As comparedwithmore than 85% tumor inhibition in the wild-type mice (Fig. 2Band D), the data suggest that IFNg and TNFa play a role in theantitumor effect of rIL7/IL15.

rIL7/IL15 inhibits pulmonary metastases of melanoma andcolon cancer

Having established that rIL7/IL15 inhibited local tumorgrowth, we wanted to assess whether rIL7/IL15 could also inhibitmetastatic tumors. To this end, C57BL6 mice were injectedintravenously with B16F10 melanoma cells to establish pulmo-narymetastases. Themice were then treatedwith equimolar dosesof rIL7/IL15 (20 mg/injection), rIL7 (11.6 mg/injection) plus rIL15(8.4 mg/injection), or PBS at 2-day intervals fromdays 2 to 12. Themice were euthanized on day 14 after tumor cell inoculation.The lungs were removed and weighed, and tumor colonies on thesurface of the lung were counted. rIL7/IL15 treatment reduced thenumbers of metastatic nodules on the lungs by approximately 8-fold, as compared with 2.7-fold after rIL7 and rIL15 treatment(Fig. 5). Similar antimetastatic patternswere observed in the lungs

of BALB/c mice after intravenous injection of CT-26 colon cancercells and treated with rIL7/IL15, or rIL7 plus rIL15 (Supplemen-tary Fig. S4). These results indicate that rIL7/IL15 also has higherantimetastatic activity than the combination of the individualfactors rIL7 and rIL15.

19.2%

11.4%

20.6%

11.5% 44.7%

49.7%

34.0%

21.1%

CD8

0

0

10

20

30

40

50

60

0 00

Mea

n t

um

or

volu

me

(mm

3 )M

ean

tu

mo

r vo

lum

e (m

m3 )

Days after transplantation

rlL7/IL15

rlL7/IL15

PBS

PBS

*P < 0.05

TNFa–/–

IFNg–/–

200

400600

800

1,000

1,200

1,400

1,600

1,800

0

200

400600

800

1,000

1,200

1,400

1,600

1,800

2 4 6 8 10 12 14

0 2 4 6 8 10 12 14

10

20

30

40

50

60

PBS

A B C G

HDE F

rIL7 rIL15 rIL7+rIL15 rIL7/IL15 PBS rIL7 rIL15 rIL7+rIL15 rIL7/IL15

PBS rIL7 rIL15 rIL7+rIL15 rIL7/IL15 PBS rIL7 rIL15 rIL7+rIL15 rIL7/IL15

5

10

15

20

25

30

0

5

10

15

20

25

30

35CT-26

B16F10

CT-26

B16F10

CT-26

B16F10

CT-26

B16F10

CD8

IFNγIF

Nγ

% IF

Nγ-

secr

etin

gce

lls in

CD

3+C

D4+

T c

ells

% T

NF

α-se

cret

ing

cells

in C

D3+

CD

4+ T

cel

ls

% T

NF

α-se

cret

ing

cells

in C

D3+

CD

8+ T

cel

ls

% IF

Nγ-

secr

etin

gce

lls in

CD

3+C

D8+

T c

ells

TN

Fα

TNFα

CD

4

CD8

CD8

CD

4

CD

4C

D4

Figure 4.

rIL7/IL15 treatment results in tumor-specific T-cell responses. A–F, BALB/c mice were injected subcutaneously with CT-26 colon cancer cells and treated withequimolar doses of rIL7/IL15, rIL7 and/or rIL15 as in Fig. 2D. Twenty-two days after tumor inoculation, splenocytes were cocultured with irradiated CT-26colon cancer cells or B16F10melanoma cells at 37°C for 2 days. CD4þ andCD8þT cells that produce IFNg or TNFawere analyzed by flowcytometry.A, representativeflow cytometric profiles showing the percentage of T cells in splenocytes and the percentage of IFNg-producing CD4þ and CD8þ T cells. Mean � SD of thepercentage of IFNg-producing CD4þ (B) and CD8þ T cells (C). D, representative flow cytometric profiles showing the percentage of T cells in splenocytes and thepercentageof TNFa-producingCD4þandCD8þTcells.Mean�SDof thepercentageof TNFa-producingCD4þ (E) andCD8þT cells (F).G andH, IFNg�/�or TNFa�/�

and their wild-typemice were injected subcutaneously with B16F10melanoma cells and treated with rIL7/IL15 or PBS as in Fig. 2B. Themean tumor volume (mm3)�SD at the indicated time points is shown. These data are representative of two independent experiments with 3–5 mice per group with similar results.� , P < 0.05; �� , P < 0.01, as compared with the B16F10-treated groups. #, P < 0.05, as compared with the rIL7 þ rIL15-treated groups.

Figure 5.

rIL7/15 inhibits the formation of pulmonary metastases by melanoma cells.C57BL/6 mice were injected intravenously with 1� 105 B16F10 cells, followed byintravenously injections with equimolar doses of rIL7/IL15 (20 mg), rIL7 (11.6 mg),and rIL15 (8.4 mg) or PBS at 2-day intervals between days 2 and 12. Themice were euthanized on day 14 after tumor cell inoculation. The total tumornodules visible at the surface of the lungs were counted under a dissectingmicroscope. These data are representative of two independent experiments,with 5 mice per group; �� , P < 0.01.

Song et al.





rIL7/IL15 has a longer serum half-life than rIL7 and/or rIL15It has been shown that both rIL7 and rIL15 have a short serum

half-life due to their low molecular weights (23–26). BecauserIL7/IL15 has a higher molecular weight, we reasoned that itshould have a longer half-life than rIL7 or rIL15. Indeed, theserum half-life of rIL7/IL15 was about 6-fold or 18-fold longerthan that of rIL7 or rIL15, respectively (Fig. 6).

DiscussionAntitumor immunotherapy is designed to stimulate the

immune system to reject and destroy cancers (13). Althougheither rIL7 or rIL15 can enhance antitumor immunity, bothcytokines have a short in vivo half-life, which limit theirantitumor activity. We show here that rIL7/IL15 hybridcytokine has higher antitumor activity than the individualfactors rIL7 and/or rIL15 in both melanoma and coloncancer mouse models. This is related to the higher tumorinfiltration of CD4 and CD8 T cells, DCs, NK and NKT cells,the maturation of DCs, and the inhibition of Tregs inducedby rIL7/IL15.

It is possible the higher antitumor activity of rIL7/IL15 is due toits longer in vivo half-life, broader activity, and different effects onimmune cells, as comparedwith the individual factors rIL7 and/orrIL15. Either rIL7 or rIL15 has a low molecular weight and canundergo a rapid renal clearance in vivo, resulting in a short serumhalf-life (23–26). The in vivo half-life of rIL7was increasedwhen itwas administered in a liposome-encapsulated form or as an IL7/anti-IL7 antibody complex (23, 38). Our previous data had alsoshown that the in vivo half-life of another hybrid cytokine rIL7/HGFbwas 7-fold greater than rIL7 (31). Similarly, IL15 fusedwithother proteins increased its in vivo half-life (24, 25, 27–29).Consistent with these data, we show here that rIL7/IL15 has 6-fold and 18-fold longer in vivo half-life than rIL7 and rIL15,respectively.

Our data also indicate that rIL7/IL15 has different effects onimmune cells, as compared with the individual factors. Forexample, although rIL7 and/or rIL15 treatment did not sig-nificantly affect the percentage of Tregs, rIL7/IL15 significantlydecreased the percentage or number of Tregs in the tumors

and the spleens. Our results are consistent with previousreports that rIL7 or rIL15 alone has little or no effect onTregs (13, 39–41). Although the mechanisms by which rIL7/IL15 inhibits Tregs remain to be investigated, it is possiblethat rIL7/IL15 cross-linked both the receptors on Tregs, whichresulted in signal cross-talk downstream of the receptors,leading to novel functional readouts, such as the inhibitionof the survival and/or growth of Tregs.

It has been reported that IL15 has both paracrine and autocrineactions onDCs, resulting in thematuration ofDCs (41), while IL7maintains the immature phenotype of DCs (41, 42). Our datashow that rIL7 treatment neither significantly affects the percent-age of DCs in the tumors, nor the expression of CD80 and CD86by the DCs. Although rIL15 increased both the intratumor infil-tration and the expression of CD80 and CD86 on DCs, rIL7/IL15hybrid cytokine had a higher activity than rIL7 and/or rIL15 onDCs in the tumors and the spleen.

NK cells are innate lymphoid cells involved in the immu-nosurveillance of cancers. Our data show that rIL15 signifi-cantly increased NK cells in the tumors and spleen, and thatrIL7 modestly increases NK cells. These results are consistentwith previous reports that IL15 has a central role in thedevelopment, homeostasis, and activation of NK cells (13)and IL7 only plays a minor role in NK homeostasis (6, 43, 44).Again, rIL7/IL15 has a higher activity than rIL7 and/or rIL15 onNK cells. However, although both rIL7 and rIL15 increased thepercentage or number of NKT cells in the tumors and in thespleen, rIL7/IL15 did not have a higher effect than rIL7 þ rIL15on NKT cells.

Both IL7 and IL15 play an important role in the differenti-ation, survival, and proliferation of T cells. We have shown thatrIL7 treatment increases the percentages of both CD4 and CD8T cells. rIL15 increases the percentage of CD8 T cells and onlymodestly affects CD4 T cells. Although the combination of rIL7and rIL15 had a synergistic effect in increasing CD4 and CD8 Tcells, rIL7/IL15 had a higher effect than rIL7 þ rIL15 on CD4and CD8 T cells.

The higher activities of rIL7/IL15 on DCs, NK, and T cellscould be due to its longer in vivo half-life. In addition, it ispossible that rIL7/IL15 also induced receptor cross-linking onthese cells and subsequent signal cross-talk. This might onlyhappen in DCs, NK, and T cells, but not in NKT cells as rIL7/IL15 did not have a significantly higher effect on NKT cells thanrIL7 þ rIL15.

In summary, although the precise mechanisms by whichrIL7/IL15 has higher antitumor effect remain to be furtherdetermined, this hybrid cytokine has the potential to be usedin enhancing antitumor immunity, thereby treating a variety oftumors.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: L. LaiDevelopment of methodology: Y. Song, L. LaiAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Y. Song, Y. Liu, R. Hu, M. Su, L. LaiAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): Y. Song, Y. Liu, L. LaiWriting, review, and/or revision of the manuscript: Y. Song, Y. Liu, L. Lai

Figure 6.

rIL7/IL15 has a longer serum half-life than rIL7 or rIL15. Mice were injectedintraperitoneally with optimal equimolar amounts of rIL7/IL15 (20 mg), rIL7 (11.6mg), or rIL15 (8.4 mg), and then bred over time (0.25, 0.5, 1, 2, 4, 8, 16, 22, 44, 80,and 120 hours after treatment). The concentration of rIL7/IL15, rIL7, and rIL15 inmouse serum was examined by IL7- or IL15-specific ELISA. These data arerepresentative of two independent experiments with 3 mice per group.






Administrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): Y. Song, Y. Liu, D. RoodStudy supervision: L. Lai

Grant SupportThis work was partly supported by a grant from the Connecticut Biomedical

Research Program (#2011-0145; to L. Lai).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received February 24, 2016; revised June 30, 2016; accepted July 15, 2016;published OnlineFirst July 29, 2016.

References1. Appasamy PM. Biological and clinical implications of interleukin-7 and

lymphopoiesis. Cytokines Cell Mol Ther 1999;5:25–39.2. Fry TJ, Mackall CL. Interleukin-7: from bench to clinic. Blood 2002

;99:3892–904.3. GoodwinRG, Lupton S, Schmierer A,Hjerrild KJ, JerzyR, ClevengerW, et al.

Human interleukin 7: molecular cloning and growth factor activity onhuman and murine B-lineage cells. Proc Natl Acad Sci U S A 1989;86:302–6.

4. von Freeden-Jeffry U, Vieira P, Lucian LA, McNeil T, Burdach SE, Murray R.Lymphopenia in interleukin (IL)-7 gene-deleted mice identifies IL-7 as anonredundant cytokine. J Exp Med 1995;181:1519–26.

5. Widmer MB, Morrissey PJ, Goodwin RG, Grabstein KH, Park LS, WatsonJD, et al. Lymphopoiesis and IL-7. Int J Cell Cloning 1990;8:168–70;discussion 71–2.

6. Boyman O, Krieg C, Homann D, Sprent J. Homeostatic maintenance of Tcells and natural killer cells. Cell Mol Life Sci 2012;69:1597–608.

7. Andersson A, Yang SC, Huang M, Zhu L, Kar UK, Batra RK, et al. IL-7promotes CXCR3 ligand-dependent T cell antitumor reactivity in lungcancer. J Immunol 2009;182:6951–8.

8. Aoki T, Tashiro K,Miyatake S, Kinashi T, Nakano T, Oda Y, et al. Expressionof murine interleukin 7 in a murine glioma cell line results in reducedtumorigenicity invivo. Proc Natl Acad Sci U S A 1992;89:3850–4.

9. Hock H, Dorsch M, Diamantstein T, Blankenstein T. Interleukin 7 inducesCD4þ T cell-dependent tumor rejection. J Exp Med 1991;174:1291–8.

10. Li B, VanRoeyMJ, Jooss K. Recombinant IL-7 enhances the potency of GM-CSF-secreting tumor cell immunotherapy. Clin Immunol 2007;123:155–65.

11. Murphy WJ, Back TC, Conlon KC, Komschlies KL, Ortaldo JR, Sayers TJ,et al. Antitumor effects of interleukin-7 and adoptive immunotherapy onhuman colon carcinoma xenografts. J Clin Invest 1993;92:1918–24.

12. Pellegrini M, Calzascia T, Elford AR, Shahinian A, Lin AE, DissanayakeD, et al. Adjuvant IL-7 antagonizes multiple cellular and molecularinhibitory networks to enhance immunotherapies. Nat Med 2009;15:528–36.

13. Steel JC, Waldmann TA, Morris JC. Interleukin-15 biology and its thera-peutic implications in cancer. Trends Pharmacol Sci 2012;33:35–41.

14. Sim GC, Radvanyi L. The IL-2 cytokine family in cancer immunotherapy.Cytokine Growth Factor Rev 2014;25:377–90.

15. MungerW,DeJoy SQ, JeyaseelanRSr, Torley LW,Grabstein KH, EisenmannJ, et al. Studies evaluating the antitumor activity and toxicity of interleukin-15, a new T cell growth factor: comparison with interleukin-2. CellImmunol 1995;165:289–93.

16. Evans R, Fuller JA, Christianson G, Krupke DM, Troutt AB. IL-15 mediatesanti-tumor effects after cyclophosphamide injection of tumor-bearingmice and enhances adoptive immunotherapy: the potential role of NKcell subpopulations. Cell Immunol 1997;179:66–73.

17. Yu P, Steel JC, ZhangM,Morris JC, Waldmann TA. Simultaneous blockadeof multiple immune system inhibitory checkpoints enhances antitumoractivity mediated by interleukin-15 in a murine metastatic colon carcino-ma model. Clin Cancer Res 2010;16:6019–28.

18. Zeng R, Spolski R, Finkelstein SE, Oh S, Kovanen PE, Hinrichs CS, et al.Synergy of IL-21 and IL-15 in regulating CD8þ T cell expansion andfunction. J Exp Med 2005;201:139–48.

19. Zhang M, Yao Z, Dubois S, Ju W, Muller JR, Waldmann TA. Interleukin-15combined with an anti-CD40 antibody provides enhanced therapeuticefficacy for murine models of colon cancer. Proc Natl Acad Sci U S A2009;106:7513–8.

20. Klebanoff CA, Finkelstein SE, Surman DR, Lichtman MK, Gattinoni L,Theoret MR, et al. IL-15 enhances the invivo antitumor activity of tumor-reactive CD8þ T cells. Proc Natl Acad Sci U S A 2004;101:1969–74.

21. Kobayashi H, Dubois S, Sato N, Sabzevari H, Sakai Y, Waldmann TA, et al.Role of trans-cellular IL-15 presentation in the activation of NK cell-mediated killing, which leads to enhanced tumor immunosurveillance.Blood 2005;105:721–7.

22. Habibi M, Kmieciak M, Graham L, Morales JK, Bear HD, Manjili MH.Radiofrequency thermal ablation of breast tumors combined with intrale-sional administration of IL-7 and IL-15 augments anti-tumor immuneresponses and inhibits tumor development and metastasis. Breast CancerRes Treat 2009;114:423–31.

23. Bui T, Faltynek C, Ho RJ. Differential disposition of soluble and liposome-formulated human recombinant interleukin-7: effects on blood lympho-cyte population in guinea pigs. Pharm Res 1994;11:633–41.

24. Ochoa MC, Fioravanti J, Rodriguez I, Hervas-Stubbs S, Azpilikueta A,Mazzolini G, et al. Antitumor immunotherapeutic and toxic propertiesof an HDL-conjugated chimeric IL-15 fusion protein. Cancer Res2013;73:139–49.

25. Van den Bergh JM, Van Tendeloo VF, Smits EL. Interleukin-15: new kid onthe block for antitumor combination therapy. CytokineGrowth Factor Rev2015;26:15–24.

26. Sportes C, Babb RR, Krumlauf MC, Hakim FT, Steinberg SM, Chow CK,et al. Phase I study of recombinant human interleukin-7 administration insubjects with refractory malignancy. Clin Cancer Res 2010;16:727–35.

27. ChangCM, LoCH, Shih YM,Chen Y,WuPY, TsuneyamaK, et al. Treatmentof hepatocellular carcinoma with adeno-associated virus encoding inter-leukin-15 superagonist. Hum Gene Ther 2010;21:611–21.

28. Kaspar M, Trachsel E, Neri D. The antibody-mediated targeted delivery ofinterleukin-15 and GM-CSF to the tumor neovasculature inhibits tumorgrowth and metastasis. Cancer Res 2007;67:4940–8.

29. Zhao DX, Li ZJ, Zhang Y, Zhang XN, Zhao KC, Li YG, et al. Enhancedantitumor immunity is elicited by adenovirus-mediated gene transfer ofCCL21 and IL-15 in murine colon carcinomas. Cell Immunol 2014;289:155–61.

30. Lai L, Zeff RA, Goldschneider I. A recombinant single-chain IL-7/HGFbetahybrid cytokine induces juxtacrine interactions of the IL-7 andHGF (c-Met)receptors and stimulates the proliferation of CFU-S12, CLPs, and pre-pro-Bcells. Blood 2006;107:1776–84.

31. Jin J, Goldschneider I, Lai L. In vivo administration of the recombinant IL-7/hepatocyte growth factor beta hybrid cytokine efficiently restores thymo-poiesis and naive T cell generation in lethally irradiated mice after synge-neic bone marrow transplantation. J Immunol 2011;186:1915–22.

32. Lai L, Jin J, Hodio J, Goldschneider I. A human recombinant IL-7/HGFal-pha hybrid cytokine enhances T-cell reconstitution in mice after syngeneicbone marrow transplantation. Transplantation 2011;92:516–22.

33. Lai L, Jin J, Goldschneider I. In vivo antitumor activity of a recombinant IL-7/HGFbeta hybrid cytokine in mice. Cancer Res 2011;71:61–7.

34. Hannani D, Vetizou M, Enot D, Rusakiewicz S, Chaput N, Klatzmann D,et al. Anticancer immunotherapy by CTLA-4 blockade: obligatory contri-bution of IL-2 receptors and negative prognostic impact of soluble CD25.Cell Res 2015;25:208–24.

35. Nagato T, Lee YR, Harabuchi Y, Celis E. Combinatorial immunotherapy ofpolyinosinic-polycytidylic acid and blockade of programmed death-ligand1 induce effective CD8 T-cell responses against established tumors. ClinCancer Res 2014;20:1223–34.

36. Lai L, Jin J. Generation of thymic epithelial cell progenitors by mouseembryonic stem cells. Stem Cells 2009;27:3012–20.

37. Morales JK, Kmieciak M, Graham L, Feldmesser M, Bear HD, Manjili MH.Adoptive transfer of HER2/neu-specific T cells expanded with alternatinggamma chain cytokines mediate tumor regression when combined withthe depletion of myeloid-derived suppressor cells. Cancer ImmunolImmunother 2009;58:941–53.


Song et al.




38. Boyman O, Ramsey C, Kim DM, Sprent J, Surh CD. IL-7/anti-IL-7 mAbcomplexes restore T cell development and induce homeostatic T cellexpansion without lymphopenia. J Immunol 2008;180:7265–75.

39. Berger C, Berger M, Hackman RC, Gough M, Elliott C, Jensen MC, et al.Safety and immunologic effects of IL-15 administration in nonhumanprimates. Blood 2009;114:2417–26.

40. Sportes C, Hakim FT, Memon SA, Zhang H, Chua KS, Brown MR, et al.Administration of rhIL-7 in humans increases invivo TCR repertoire diver-sity by preferential expansion of naive T cell subsets. J Exp Med2008;205:1701–14.

41. Rochman Y, Spolski R, Leonard WJ. New insights into the regulation of Tcells by gamma(c) family cytokines. Nat Rev Immunol 2009;9:480–90.

42. Guimond M, Veenstra RG, Grindler DJ, Zhang H, Cui Y, Murphy RD,et al. Interleukin 7 signaling in dendritic cells regulates the homeostaticproliferation and niche size of CD4þ T cells. Nat Immunol2009;10:149–57.

43. Ranson T, Vosshenrich CA, Corcuff E, RichardO,MullerW, Di Santo JP. IL-15 is an essential mediator of peripheral NK-cell homeostasis. Blood2003;101:4887–93.

44. Vosshenrich CA, Ranson T, Samson SI, Corcuff E, Colucci F, RosmarakiEE, et al. Roles for common cytokine receptor gamma-chain-dependentcytokines in the generation, differentiation, and maturation of NK cellprecursors and peripheral NK cells invivo. J Immunol 2005;174:1213–21.






2016;15:2413-2421. Published OnlineFirst July 29, 2016.Mol Cancer Ther Yinhong Song, Yalan Liu, Rong Hu, et al. Cytokine in Mice

Antitumor Activity of a Recombinant IL7/IL15 HybridIn Vivo

Updated version

10.1158/1535-7163.MCT-16-0111doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2016/07/29/1535-7163.MCT-16-0111.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/15/10/2413.full#ref-list-1

This article cites 44 articles, 23 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/15/10/2413To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-16-0111

http://mct.aacrjournals.org/content/suppl/2016/07/29/1535-7163.MCT-16-0111.DC1

http://mct.aacrjournals.org/content/15/10/2413.full#ref-list-1

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/15/10/2413


in vivo antitumor activity of a recombinant il7/il15 ...in vivo antitumor activity of a recombinant...

Documents