ABSTRACT
Autologous serological screening of a cDNA expression library (SEREX) derived from childhood neuroblastoma led to the identification of 10 different antigens, including 6 novel gene products. The novel antigen 018INX was derived from a small open reading frame in a region of alpha-internexin mRNA that was previously described as 3' untranslated region. 018INX thus represents a novel type of tumor antigen. Five novel gene products were derived from NNP-1 (NNP3) and Hu genes (HuC-L, HuD3, HuDY, HuD1pro(c)). As indicated by sequence analysis, these antigens were generated by alternative splicing and/or alternative promoter usage or allelic polymorphism. mRNA expression analyses revealed different tissue restrictions of novel compared to known HuD and NNP-1 transcripts in normal and malignant tissues. The expressions patterns of distinct transcripts indicated potential clinical meanings as diagnostic and/or prognostic tissue markers. When kinetics of serum antibody titres against SEREX-defined antigens were compared to tumor load over time in our patient with neuroblastoma, we found 100-fold increases of anti-Hu and anti-018INX antibody titres preceding the clinical diagnosis of recurrent tumor growth after 2 years. When sera of pediatric patients with cancer (30) and healthy controls (30) were tested for humoral responses to SEREX-defined neuroblastoma antigens, we detected antibodies against all known antigens and NNP3 with low frequencies and titres in control sera, while anti-018INX and anti-Hu antibodies were found in cancer patients only. Our findings indicate that SEREX-defined tumor antigens might provide novel tools for understanding and treatment of this aggressive childhood malignancy.
Subject(s)
Carrier Proteins/genetics , Nerve Tissue Proteins/genetics , Neuroblastoma/genetics , Nuclear Proteins/genetics , RNA-Binding Proteins/genetics , Adult , Amino Acid Sequence , Base Sequence , Blotting, Southern , Case-Control Studies , Child, Preschool , DNA, Complementary/metabolism , ELAV Proteins , ELAV-Like Protein 3 , ELAV-Like Protein 4 , Female , Gene Library , Humans , Intermediate Filament Proteins , Models, Genetic , Molecular Sequence Data , RNA, Messenger/metabolism , Reverse Transcriptase Polymerase Chain Reaction , Sequence Homology, Amino Acid , Time FactorsABSTRACT
Medulloblastoma is an embryonal childhood malignancy with poor prognosis. By screening 4 medulloblastoma cDNA expression libraries (SEREX) with autologous sera, 15 different antigens were identified. These antigens were encoded by 3 novel genes, genes of unknown function (KIAA0445, KIAA1853, KIAA0665, FLJ13942, HSPC213), a proto-oncogene (rab18), candidate tumor suppressor genes (BAP1, PRDM13) and genes encoding a motor protein (kinesin-2), a histone (H2A1.2), the ankyrin residue-rich nasopharyngeal cancer susceptibility protein (NZ16) and the transcription factor TZP, which is homologous to the tumor-associated antigens HCA58 and GLEA2. In a consecutive analysis of serum antibody titers and tumor load, a more than 10-fold increase in serum antibodies against PRDM13 preceded the clinical diagnosis of recurrent tumor growth in a patient with aggressive large cell medulloblastoma. When sera of pediatric patients with cancer (n = 40) and healthy controls (n = 40) were tested for humoral responses against the SEREX-defined antigens, 5 antigens were exclusively recognized by sera from cancer patients. These antigens included a novel rab18 gene product translated from mRNA sequences formerly described as 3' untranslated region. Humoral responses against 2 of the remaining 10 antigens were found preferentially in cancer patients. Antibodies against these antigens were detected in 8/40 and 12/40 cancer patients, respectively, but in only 1 healthy control. The 2 antigens were characterized by a tumor-specific deletion and a tumor-specific mutation, respectively. These findings indicate that the humoral immune response against medulloblastoma is directed against diverse antigens that may be useful as diagnostic markers or targets for immunotherapy.