Herpes Simplex Virus (HSV) Amplicon-mediated Codelivery of Secondary Lymphoid Tissue Chemokine and CD40L Results in Augmented Antitumor Activity

INTRODUCTION

The elaboration of functional immune responses requires fine-tuned interactions of multiple immune effector cell types, including APCs 3 and different subsets of T cells. Because of the low progenitor frequency and short half-life of any particular APC, the process of APC/T-cell interaction is designated to occur at specific anatomical locations, the secondary lymphoid organs (1 , 2) . To maximize the probability that a particular antigen-loaded APC will encounter its cognate antigen-specific T cell, an organizational hierarchy has evolved to couple the migratory abilities of different cellular subsets to their functional role in an immune response. This is mediated through the capacity of different tissues either constitutively or in response to extrinsic stimuli to release an array of chemokines and a highly flexible capacity to up- or down-regulate receptors corresponding to these chemokines (3 , 4) .

Antigen capture, processing, and presentation to naïve T cells within the LN are prime examples of this highly coordinated trafficking process. Important components of this process are immature DCs that survey peripheral tissues and capture antigen before honing to secondary lymphoid tissues (5) . Efficient retrieval and transport of DCs into T cell-rich areas of the LN are central to the development of an immune response and are primarily mediated by two chemokines, SLC (6) and ELC (7) , which engage a single receptor, CCR7. During the transition from peripheral tissue to the LN, DCs undergo maturation and become more efficient at presenting antigen to T cells. This maturation process includes enhanced expression of the CCR7 receptor and renders mature DCs highly responsive to the SLC gradient (8, 9, 10) . In the T-cell area of the LN, DCs encounter two types of T cells also expressing the CCR7 receptor, naïve T cells and a subset of memory T cells known as central memory (T_cm) that can be readily activated on exposure to recall antigens captured by DCs (11) . The SLC/CCR7 axis is crucial for the T-cell/DC interaction in T-cell areas of the LN and generation of an immune response. This has been confirmed in mice deficient for either SLC (12) or CCR7 expression (13) .

We hypothesized that local elaboration of SLC would be effective in augmenting antitumor immune responses, in part by promoting the recruitment of naïve T cells and colocalization of such T cells with mature APCs. In this report, we studied the antitumor activity of SLC alone and in combination with the CD40L costimulatory ligand. CD40L, a member of the TNF superfamily, is a type II transmembrane glycoprotein transiently expressed on antigen-activated CD4⁺ T cells. Acting through its receptor CD40, CD40L plays a central role in activating APCs including DCs, monocytes, and B cells, and initiation of antigen-specific immune responses (14) . SLC and CD40L were administered to mice using an HSV amplicon vector, a gene delivery platform shown previously to be amenable for tumor immunotherapy (15 , 16) . Using two different murine tumor models, we examined the immune effector cell infiltrate, cytokine mRNA expression within the tumor bed, and the antitumor responses that resulted from high-level SLC and/or CD40L expression within the tumor environment.

MATERIALS AND METHODS

Cell Culture and Amplicon Packaging.

A20 and CT-26 cells were grown in vitro in RPMI 1640 (DMEM for CT-26) medium supplemented with 10% FCS, penicillin (50 units/ml), streptomycin (50 μg/ml), l-glutamine (2 mm/ml), and β-mercaptoethanol (50 μm/ml), and maintained in culture at 37°C at 5% CO₂ atmosphere. The cDNA for murine SLC and CD40 were cloned into the HSV amplicon plasmid, HSVPrPUC (17) . HSVlac, a β-galactosidase-expressing control amplicon, has been described previously (18) . HSV amplicon vectors were packaged and titered as described previously (15) .

Detection of SLC Produced by A20 Tumor Cells.

A20 cells were transduced with HSVlac, HSV-SLC at an MOI of 3, or left nontransduced, and 48 h later supernatant was collected for ELISA using polyclonal rabbit antimouse SLC as a capture antibody and biotinylated polyclonal rabbit antimouse SLC antibody for detection (both from R&D Systems, Minneapolis, MN). The biological activity of secreted SLC was confirmed in a two-chamber transmigration assay using naïve T cells purified from BALB/c mice spleens. T cells (10⁷) were suspended in RPMI 1640 with 10% FCS and added to the top chambers of triplicate wells of a Transwell plate (Costar, Cambridge, MA). Supernatants from untreated A20 cells, or A20 cells transduced with HSVlac or HSV-SLC at an MOI of 3, as well recombinant SLC protein at 100 ng/ml were added in triplicates to the bottom chamber of the Transwell plate. After a 4-h incubation, transmigrating T cells were collected from the bottom chamber and enumerated.

IHC for CD4⁺ and CD8⁺ T Cells.

Established A20 tumors were left nontransduced or transduced with HSVlac or HSV-SLC (3.5 × 10⁶ amplicon particles), and 96 h later the mice were sacrificed and tumors dissected, embedded in OCT, and snap-frozen in liquid nitrogen. Tissue sections (10 μm) were fixed in acetone and incubated in 1% H₂O₂ for 15 min before adding goat serum to block nonspecific binding. The slides were washed in PBS three times before adding specific hamster antimouse CD4- or CD8-specific primary antibody (BD PharMingen, San Diego, CA) for 1 h. This incubation was followed by addition of biotinylated goat antihamster secondary antibody (Vector Laboratories, Burlingame, CA), and developed by streptavidin-peroxidase (ABC kit; Vector Laboratories). Enzymatic reaction conditions used to detect antibody staining were performed as suggested by the manufacturer.

Tumor Establishment and Vector Injection.

A20 and CT-26 cells were cultured in vitro as described above, and subsequently 10⁶ cells were injected s.c. into the right and/or left flank of 6–8-week-old BALB/c mice (Jackson Laboratory). Developing tumors were monitored twice weekly by measuring tumor diameter. Tumors were injected with HSV-SLC, HSV-CD40L, or a combination of both amplicons, or control HSVlac when the tumors reached a diameter of 5 mm. HSV amplicon stocks were diluted in PBS before injection, and 100 μl containing 3.5 × 10⁶ amplicons was used in the direct intratumoral inoculations. Mock treatment controls received intratumoral injections of sterile PBS.

FACS Analysis of Tumor-infiltrating T Cells and DCs.

Established A20 tumors were either left nontransduced or transduced with HSVlac or HSV-SLC (3.5 × 10⁶ amplicon particles), and 96 h later the mice were sacrificed. Each tumor was dissociated into a single cell suspension and washed twice in PBS. Fc receptors were blocked using Fc-Block CD16/CD32 antibody (BD PharMingen), and cells were stained with phycoerythrin-labeled anti-CD4 (GK1.5), anti-CD8a (53–5.8), and anti-CD11c (HL3; all from BD PharMingen). Labeled cells were studied by FACS analysis for T cells and DC infiltration using a FACScan flow-cytometer and CellQuest software (BD Bioscience, Mountain View, CA).

CTL Assay.

Splenocytes were harvested from mice that had eradicated their tumors in response to delivery of HSV-SLC, HSV-CD40L, or a combination of both amplicons, as well as from control mice treated with HSVlac or mock-treated. Red blood cells were lysed using AKC buffer, and the splenocytes were washed before being incubated with irradiated A20 cells at a ratio of 4:1 for 6 days in RPMI 1640 supplemented with 10% FCS and glutamine penicillin streptomycin (no IL-2 added). CTL activity was assayed using a standard chromium release assay by incubating the splenocytes with ⁵¹Cr-labeled A20 cells in triplicates wells of a V-shaped 96-well plate at the following E:T ratios (60:1, 20:1, 6:1, 2:1, and 1:1). After a 4-h incubation, the supernatant was collected, and radioactivity was measured using a gamma counter.

T-Cell Subset Depletion.

For CD4⁺ and CD8⁺ T-cell depletion, mice were injected i.p. with 100 μg of anti-CD4 (clone Gk 1.5) or anti-CD8 (clone 53–6.7) in 100 μl of PBS (both antibodies from PharMingen) 5 days before tumor inoculation. This resulted in >95% depletion of the respective T-cell subset for >2 weeks (data not shown).

RT-PCR Analysis of Intratumoral Cytokine Gene Expression.

Regressing A20 tumors obtained from animals treated with local injection of HSV-SLC, HSV-CD40L, or a combination of HSV-SLC plus HSV-CD40L amplicons, and nonregressing tumors from control animals treated with either HSVlac or untransduced tumors were dissected and lysed for RNA extraction, and whole tumor RNA was isolated. After DNase treatment (Life Technologies, Inc.), 5 μg of total RNA were used to generate first strand cDNA using oligodeoxythymidylic acid (Pharmacia, Uppsala, Sweden) as a primer and Superscript II (Life Technologies, Inc.) for reverse transcription. Two μl of the reverse transcription reaction was used as a PCR template using the following set of primers: β-actin: 5′ GTTGCTATCCAGGCTGTGCT, 3′ CGGATGTCCACGTCACAC TT; T-cell receptor δ chain: 5′ CCTGCAATACCAGCGTCATGC, 3′ GAACCTCAGCAGCCC CAGAAG; perforin: 5′: CAGAATGCAAGCAGAAGCACA AG, 3′: GGTGGAGTGGAGGTTTTTGTA CC; IFN-γ: 5′: CATTGAAAGCCTAGAAAG TCT G, 3′: CTCATGAATGCATCCTTTTTC G; IL-12 p35: 5′: GATCATGAAGACATCACACGG, 3′: AGAATGATCTGCTGATGGTTG; and IL-12 p40: 5′: ATGTGTCCTCAGAAGCTAACCATC, 3′: AACCGTCCGGAGTAATTTGGTGCT. PCR was performed under the following conditions: 94°C for 5 min followed by 30 cycles (25 for β-actin) of 1 min denaturation at 94°C, 30 s annealing at 59°C, and 1 min amplification at 72°C followed by 10 min extension at 72°C.

RESULTS

We hypothesized that local elaboration of SLC expression in the tumor bed would result in enhanced antitumor immune responses because of recruitment of APC influx. To confirm sensitivity of tumor cells to transduction by HSV amplicons, A20 cells were transduced in vitro with HSV amplicons, and expression of SLC was confirmed by ELISA (Table 1) ⇓ and that of CD40L by flow-cytometry (data not shown). The chemotactic activity of amplicon-expressed SLC was demonstrated in a two-chamber transmigration assay (Fig. 1) ⇓ . Two ml of supernatant from transduced A20 cultures (containing roughly 10 ng/ml of amplicon-derived SLC) was added to each well, whereas recombinant SLC (100 ng/ml; R&D Systems) was used as a positive control. We noted consistently that SLC generated by amplicon-transduced A20 cells was more biologically active in the two-chamber transmigration assay than the reconstituted recombinant protein that was produced in Escherichia coli. This observation may reflect differential post-translational modification of the two SLC sources.

To examine the immune cell recruitment activity of HSV-SLC in vivo, pre-established A20 and CT-26 tumors in BALB/c mice were injected with HSVlac, HSV-SLC, or were mock transduced, and 48 and 96 h later, animals were sacrificed and studied by IHC for infiltrating CD4⁺ and CD8⁺ T cells. IHC analysis illustrated that qualitatively higher numbers of CD4⁺ and CD8⁺ T cells were recruited into tumors injected with HSV-SLC than observed in the mock and HSVlac-injected tumors (Fig. 2) ⇓ . A modest increase in CD4⁺ and CD8⁺ T-cell density was seen in HSVlac-injected tumors as compared with mock-infused tumors possibly because of an inflammatory reaction generated by HSV amplicon infusion (19) . Quantitation of infiltrating DCs, and CD4⁺ and CD8⁺ T-cell numbers was performed using flow cytometric analysis of single-cell suspensions generated from these treated tumors (Table 2) ⇓ . We observed significantly higher numbers of CD11c⁺ DCs in HSV-SLC-injected tumors as compared with mock and HSVlac-injected tumors. Additionally, we confirmed the increase in CD4⁺ and CD8⁺ T-cell infiltration as visualized by IHC for HSV-SLC (Table 2) ⇓ .

Antitumor Activity of HSV-SLC and HSV-CD40L.

The ability of HSV-SLC to enhance antitumor immune responses was subsequently assessed using two in vivo tumor types, the A20 B-cell lymphoma and the CT-26 adenocarcinoma models. As a point of reference, we compared the antitumor activity of SLC with a previously described amplicon expressing the CD40L costimulatory ligand (HSV-CD40L). CD40L plays a central role in activating APCs including DCs, monocytes, and B cells, and initiation of antigen-specific immune responses. Pre-established A20 and CT-26 tumors were injected directly with HSVlac, HSV-SLC, or HSV-CD40L twice a week, or were left nontreated. Tumor size was monitored twice a week, and mice were sacrificed when tumor size exceeded 20 mm in one diameter or if the animal showed signs of distress. In mice bearing A20 tumors, 4 of 5 mice treated with HSV-SLC eradicated their tumors, whereas 5 of 6 mice eradicated their A20 tumor after treatment with HSV-CD40L (Fig. 3A) ⇓ . In contrast, of the mice receiving HSVlac or mock treatment exhibited progressive tumor growth resulting in their eventual sacrifice (Fisher’s exact test, P < 0.05). Similarly, in the CT-26 tumor model treatment with HSV-SLC or HSV-CD40L resulted in a significantly reduced growth rate as compared with that observed in mice receiving HSVlac or untreated mice (Fig. 3B ⇓ ; Fisher’s exact test, P < 0.05).

Cooperative Antitumor Activity of Combined HSV-SLC and HSV-CD40L Administration.

Because a primary function of SLC is to recruit DCs, and CD40L activates APCs including DCs, we reasoned that local expression of CD40L on tumor cells via an amplicon vector would activate recruited DCs. To test this hypothesis, we treated A20 and CT-26 tumor-bearing mice with HSVlac, HSV-SLC, HSV-CD40L, or with a combination of HSV-SLC and HSV-CD40L amplicons in a 1:1 ratio. In mice with unilateral A20 tumors, given the high frequency of tumor eradication in response to intratumoral injection of HSV-SLC or HSV-CD40L, we could not demonstrate an added benefit of the SLC/CD40L combination in unilaterally implanted and injected tumors (Fig. 4A ⇓ ; χ² = 0.433 (2 degrees of freedom), P = 0.81). In mice that had cleared the A20 tumors after amplicon infusion, a contralateral tumor rechallenge was performed. This rechallenge represented one means to examine elicitation of systemic antitumoral immunity. On day 40 of the experiment, mice were rechallenged contralaterally with 10⁶ A20 cells. Again, we were unable to demonstrate an added benefit of the SLC/CD40L combination.

In a separate study paradigm, bilateral A20 tumors were established, and mice subsequently received unilateral intratumoral injections of HSV-SLC and/or HSV-CD40L. Animals treated with a combination of HSV-SLC plus HSV-CD40L exhibited significantly improved survival profiles as compared with treatment with either HSV-SLC or HSV-CD40L alone (Fig. 4B ⇓ ; Fisher’s exact test, P = 0.03). A statistically significant difference was also observed between treatment with either HSV-SLC or HSV-CD40L as compared with either of the two control arms (Fig. 4B ⇓ ; Fisher’s exact test, P < 0.05). In mice bearing CT-26 tumors, the combination of HSV-SLC plus HSV-CD40L was successful at eradicating tumor in 3 of 6 tumor-bearing mice, whereas the single amplicon vector treatments reduced the tumor size but did not eradicate any pre-established tumors (Fig. 5) ⇓ . This CT-26 tumor survival data trended toward improvement with the combination of HSV-SLC and HSV-CD40L, as compared with either HSV-SLC or HSV-CD40L alone, but did not reach statistical significance (Fisher’s exact test, P = 0.1).

To investigate the roles of CD4⁺ and CD8⁺ T-cell subsets in HSV-SLC/HSV-CD40L-mediated tumor regression, groups of BALB/c mice were depleted of CD4⁺ or CD8⁺ T-cell subsets, and 5 days later were inoculated with A20 tumors. A20 tumors exhibited similar growth kinetics in CD4⁺ and CD8⁺ T cell-depleted mice as seen in normal BALB/c mice (Fig. 6 ⇓ ; data not shown). When tumor size reached 5 mm, groups of mice (4 per arm) were treated with HSVlac, HSV-SLC, HSV-CD40L, a combination of HSV-SLC/HSV-CD40L, or were mock treated, and tumor size was measured twice weekly. As Fig. 6A ⇓ illustrates, CD4⁺ T-cell depletion did not negatively impact the antitumor activity of HSV-SLC, HSV-CD40L, or the combination of HSV-SLC/HSV-CD40L, where 3 of 4 mice had eradicated their tumors. However, CD8⁺ T-cell depletion significantly diminished antitumor activity, as only 1 of 4 mice eradicated their tumor after treatment with HSV-SLC, HSV-CD40L, or the combination of HSV-SLC/HSV-CD40L (Fig. 6B) ⇓ . In aggregate, these data indicate that the antitumor effects exhibited by amplicon-mediated SLC and/or CD40L treatments require CD8⁺ T-cell function.

Generation of CTL Activity from Splenocytes of Responding Mice.

Because the T-cell depletion studies indicated that CD8⁺ T cells were responsible for the observed antitumor activity, we subsequently determined whether tumor-specific CTL function was present. To that end, splenocytes from A20 and CT-26 tumor-bearing mice receiving HSV-SLC and/or HSV-CD40L, as well as from mice in the two control arms, were assayed for CTL activity in a standard chromium release assay. As shown in Fig. 7 ⇓ , mice receiving HSV-SLC, HSV-CD40L, or both amplicons exhibited enhanced CTL activity as compared with those mice that had received HSVlac or mock treatment.

Study of Cytokine mRNA Expression Profiles in Regressing Tumors.

To additionally characterize the in situ-generated immune response, we purified RNA from tumors excised from mice in each of the treatment arms and performed RT-PCR for the detection of several cytokines. We were able to amplify the δ-chain of CD3 in all of the tumor samples indicating the presence of infiltrating T cells and also detected mRNA encoding the p35 subunit of IL-12, which is constitutively expressed by most cells including APCs and natural killer cells (Fig. 8) ⇓ . Only regressing tumors treated with HSV-SLC, HSV-CD40L, or a combination of both amplicons expressed detectable mRNA levels for IFN-γ, perforin, and the inducible p40 subunit of IL-12 (Fig. 8) ⇓ . In aggregate, these data indicate that local elaboration of SLC and/or CD40L via an amplicon vector leads to the induction of a Th1-like cascade and activation of T cells that culminates in the generation of a potent antitumor response.

DISCUSSION

Generation of an antitumor immune response is a complex process dependent on coordinate interaction of different subsets of effector cells. DCs play a central role in this event through a series of functions including antigen capture, processing, up-regulation of costimulatory signal, and migration to lymphoid tissue that culminates in antigen presentation to the T cell (20) . This process is orchestrated through timely expression of different subsets of chemokines and their cognate receptors on DCs, which is closely linked to the DC maturation stage (21 , 22) . The latter is influenced by signals arising from bacterial and viral by-products of inflammation, cytokines, and costimulatory signals derived from monocytes, T cells, stromal cells, and DCs within the local microenvironment (22, 23, 24) .

Previous work demonstrated that SLC chemotactic activity for DCs and T cells could be harnessed to generate antitumor immune responses (25, 26, 27) . We reasoned that such an activity could be additionally improved on by in situ activation of recruited DCs via local expression of CD40L. Physiologically, SLC serves to guide mature antigen-loaded DCs from peripheral tissues to the T-cell area of the LN where two subsets of T cells, naïve and central memory T cells, reside. Local elaboration of SLC within the tumor bed could theoretically reproduce the LN conditions shown to be conducive for effective T-cell priming. We asked specifically whether DC-mediated CD8⁺ T-cell priming within the tumor could be enhanced additionally by the extrinsic provision of CD40L.

It is well established that DCs require CD4⁺ T-cell help to properly cross-prime CD8⁺ T cells, a process dependent on stimulation of the CD40 receptor on DCs by transiently expressed CD40L from CD4⁺ T cells (28) . This “licensing” step is crucial in determining the outcome of the DC/CD8⁺ T-cell interaction: whether it leads to priming or the induction of tolerance. CD40L also provides the primary stimulus for induction of IL-12 by maturing DCs (29, 30, 31) , a feature that sets CD40L apart from most other maturation stimuli such as TNF-α, IL-1, or FasL. Animals injected with HSV-SLC alone successfully mounted an antitumor immune response although to a lesser degree than that observed with the combination treatment of HSV-SLC and HSV-CD40L. In the absence of HSV-CD40L infusion, DCs are dependent on native proinflammatory stimuli (TNF-α and IL-1) within the tumor for maturation, possibly arising from the codelivered HSV helper virus that has been shown previously to induce local inflammation (32) .

On the basis of these results, we suggest the following model to illustrate the sequence of events triggered by transduction with HSV amplicons encoding SLC and/or CD40L. Initial treatment of the A20 or CT-26 tumors with the HSV amplicon triggers an inflammatory response possibly mediated by the virus itself. This is reflected in the modest increase in numbers of infiltrating CD4⁺ and CD8⁺ T cells in tumors injected with HSVlac over nontransduced tumors (Fig. 2) ⇓ . Proinflammatory cytokines (TNF-α and IL-1) and chemokines released in that setting can up-regulate adhesion molecules on endothelial cells that recruit neutrophils, macrophages, and immature DCs (33, 34, 35) . In tumors treated with HSV-SLC, incremental numbers of CCR7⁺ T cells and mature DCs are recruited as well (Fig. 2 ⇓ ; Table 2 ⇓ ). Although mature DCs are less efficient at capturing antigen, they are an abundant source of inflammatory chemokines including macrophage inflammatory protein 1α, macrophage inflammatory protein 1β, MCP-1, -2, -4, and RANTES (36) . These chemokines may establish a positive feed forward loop that can additionally recruit immature DCs, monocytes, and natural killer cells beyond what is initially mediated by viral by-products. The ensuing positive reinforcement could provide for adequate capture and presentation of tumor antigens.

As mentioned above, two subsets of T cells express the CCR7 receptor and are responsive to SLC recruitment, namely naïve T cells and central memory T cells. Whereas the CCR7⁺ central memory T cells are readily activated after brief interaction with APCs into effector T cells, naïve T cells require more prolonged T-cell receptor signaling and costimulation to develop into effector T cells (11) . It is plausible that the antitumor activity of HSV-SLC is derived primarily from priming and expansion of the central memory T-cell subset. In mice treated with the combination of HSV-SLC and HSV-CD40L, the availability of high-level CD40L expression optimizes DC maturation and partially reproduces the LN conditions necessary for naïve T-cell transition into effector cells, as the latter might contribute to the augmented antitumor effect seen with combination treatment.

In summary, our work outlines a comprehensive immune therapy strategy applicable for both hematologic and solid tumors based on cooperative interaction of multiple effector cell types. This strategy may overcome functional immune impairment observed in patients with advanced malignancy by facilitating the recruitment, activation, and expansion of naïve and central memory T cells.