Antigen-specific cancer immunotherapy is a promising strategy for improving cancer treatment. Recently, many tumor-associated antigens and their epitopes recognized by cytotoxic T lymphocytes (CTLs) have been identified. However, the density of endogenously presented antigen-derived peptides on tumor cells is generally sparse, resulting in the inability of antigen-specific CTLs to work effectively. We hypothesize that increasing the density of an antigen-derived peptide would enhance antigen-specific cancer immunotherapy. Here, we demonstrated that intratumoral peptide injection leads to additional peptide loading onto major histocompatibility complex class I molecules of tumor cells, enhancing tumor cell recognition by antigen-specific CTLs. In in vitro studies, human leukocyte antigen (HLA)-A*02:01-restricted glypican-3144–152 (FVGEFFTDV) and cytomegalovirus495–503 (NLVPMVATV) peptide-specific CTLs showed strong activity against all peptide-pulsed cell lines, regardless of whether the tumor cells expressed the antigen. In in vivo studies using immunodeficient mice, glypican-3144–152 and cytomegalovirus495–503 peptides injected into a solid mass were loaded onto HLA class I molecules of tumor cells. In a peptide vaccine model and an adoptive cell transfer model using C57BL/6 mice, intratumoral injection of ovalbumin257–264 peptide (SIINFEKL) was effective for tumor growth inhibition and survival against ovalbumin-negative tumors without adverse reactions. Moreover, we demonstrated an antigen-spreading effect that occurred after intratumoral peptide injection. Intratumoral peptide injection enhances tumor cell antigenicity and may be a useful option for improvement in antigen-specific cancer immunotherapy against solid tumors.

The overexpression of secreted protein acidic and rich in cysteine (SPARC) is associated with increased aggressiveness and poor prognosis in malignant melanoma. Its roles and underlying mechanisms on melanoma cell growth, however, are not fully clarified. To validate the potential of SPARC as a therapeutic target, we examined the effect of the knockdown of SPARC with SPARC-specific siRNA on the growth of human melanoma cell lines. SPARC siRNAs exerted a potent knockdown effect. Silencing of SPARC resulted in growth inhibition with G 1 arrest accompanied by accumulation of p21, a G 1 cyclin-dependent kinase inhibitor, in MeWo and CRL1579 cells. Moreover, the induction of p53 was observed in MeWo cells, but not in CRL1579 cells. Conditioned media containing SPARC from MeWo cells could not restore the growth of SPARC-silenced MeWo cells. This result suggests that intracellular SPARC, but not secreted SPARC, is involved in cell proliferation. In addition, silencing of SPARC induced apoptosis in MeWo and CRL1579 cells. Furthermore, when MeWo cells in which SPARC expression was transiently knocked down by SPARC siRNA were implanted in nude mice, the tumor growth was suppressed. Our findings suggest that SPARC contributes to cell growth and could be a potential target molecule for melanoma therapy. (Cancer Sci 2010; 101: 913-919)

Glypican-3 (GPC3) is overexpressed in human hepatocellular carcinoma (HCC) but not expressed in normal tissues except for placenta and fetal liver and therefore is an ideal target for cancer immunotherapy. In this study, we identified an H2-Kb or H2-Db restricted and murine GPC3 (mGPC3)-derived cytotoxic T-lymphocyte (CTL) epitope peptide in C57BL/6 (B6) mice, which can be used in the design of preclinical studies of various therapies with GPC3-target immunotherapy in vivo. First, 11 types of 9- to 10-mer peptides predicted to bind with H2-Kb or H2-Db were selected from the mGPC3 amino acid sequence based on the binding score as calculated by the BIMAS software. We evaluated the peptide-binding affinity and confirmed that all peptides were able to bind to H2-Kb or H2-Db by in vitro cellular binding assay. Subsequently, a mixed peptide vaccine and single peptide vaccine were given to B6 mice to evaluate immunogenic potential of the 11 selected peptides. Using the splenocytes from peptide-vaccinated mice, interferon (IFN)-γ enzyme-linked immunospot (ELISPOT) assays showed that mGPC3-1127–136 (AMFKNNYPSL) peptide was the most efficient for inducing CTLs among the 11 peptides. Next, we demonstrated that the mGPC3-1 peptide-specific CTL line could recognize mGPC3-expressing cancer cells, suggesting that mGPC3-1 peptide was an endogenously presented peptide. In conclusion, we identified mGPC3-1 as an H2-Kb or H2-Db restricted, mGPC3-derived CTL epitope peptide.

A gene of the Epstein-Barr virus (EBV), BamHI-C fragment rightward reading frame 1 (BCRF1), codes viral interleukin-10 (vIL-10), which is a close homolog to human IL-10. EBV strain variations are known at EBV latent membrane protein 1 (LMP1), and the distinct forms of LMP1 have been identified. In order to further elucidate the variations of EBV strains, the BCRF1 (vIL-10) gene was analyzed using PCR-direct sequencing in African Burkitt's lymphoma (BL) cell lines Raji, P3HR-1, EB1 and Daudi, Japanese BL cell line Akata, lymphoblastoid cell line OB and 22 wild EBV isolates from eight gastric carcinoma tissues and 14 throat washes. We found only five variations of the vIL-10 gene in them with one silent mutation and three non-silent mutations. Raji had no mutation to the prototype gene of B95-8. EB1 and P3HR-1 had non-silent mutations in the sequences leading to the arginine/serine and threonine/proline interchanges at residues 4 and 166, respectively. The silent mutation was detected at valine 102 in Daudi and also in the Japanese cell lines Akata, OB and 20 (90.9%) of the wild EBV isolates. The type of variations in the vIL-10 gene had a common relationship with those in the LMP1 gene. All of the variants of valine 102 had China1-type LMP1 sequences except for Daudi with Med-type LMP1 and other minorities with B95-8 type LMP1. The conservativeness of vIL-10 with a few variations suggests the indispensability of the vIL-10 gene in EBV and that the variations of the vIL-10 gene may depend upon the geographical prevalence of the EBV strains. This is the first report regarding the variations of the vIL-10 gene in cell lines and other wild isolates.

A device for the capture and recollection of live target cells is described. The platform was a silicon (Si) wafer modified with an anti-HEL antibody (anti-HEL-IgG, HEL = hen egg lysozyme) through a photocleavable 3-amino-3-(2-nitrophenyl)propionic acid (ANP) linker. The modification processes of the Si wafer surface were monitored by Fourier transform infrared spectroscopy-attenuated total reflection (FTIR-ATR) and fast-scanning atomic force microscopy (FS-AFM). The attachment of IgG and its release reaction on the Si surface via the photochemical cleavage of the ANP linker were observed directly by FS-AFM. The results of an enzyme-linked immunosorbent assay (ELISA) indicated that the photorelease of the complex of anti-HEL-IgG with the secondary antibody-alkaline phosphatase hybrid (secondary IgG-AP) from the Si surface occurs with minimum damage. Furthermore, it was possible to collect SP2/O cells selectively that express HEL on their cell membranes (SP2/O-HEL) on the Si wafer device. Photochemical cleavage of the ANP linker facilitated the effective release of living SP2/O cells whose viability was verified by staining experiments using tripan blue. Moreover, it was possible to reculture the recovered cells. This methodology represents an effective strategy for isolating intact target cells in the biological and medicinal sciences and related fields.