异常O-糖基化的粘蛋白在肿瘤高度表达，以MUC1 量为最高，因其抑制肿瘤细胞凋亡和抑制免疫杀伤细胞，是肿瘤治疗的靶点。申请人发表了生物信息学预测糖肽序列的理论，得到了特异性识别依靠理论预测到的糖肽的单克隆抗体。为了探索这些抗体识别MUC1糖肽的精确分子结构，本项目将 1）在分子水平分别合成PDTR，PGSTA，GVTSA三个抗原表位区的多肽骨架，以三种肿瘤糖链Tn，STn，ST加以修饰，检测抗体的特异性和亲和力；2）在细胞水平，对表达MUC1糖蛋白的肿瘤细胞，用基因转染和基因敲除方法改造糖基转移酶基因，获得特异表达Tn，STn或ST糖链结构的肿瘤细胞系，检测一系列细胞系对抗体的结合和内吞； 3）在临床标本水平研究抗体对肺癌组织和血清外泌体的识别；在模式动物模型水平进行抗体药物偶联物和CAR NK 细胞治疗肿瘤的成药性研究。

Abnormally O-glycosylated mucins are highly expressed in lung cancer, with MUC1 as the most abundant one. They inhibit the apoptosis of tumor cells and the cytotoxity function of immune cells, thus serve as ideal candidates for therapy targets. We published a bioinformatics approach to predict glycopeptide sequences, synthesize such sequences and to generate monoclonal antibodies which bind to such predicted epitopes. For all the five O-glycosylation sites of Tn-modified MUC1 tandem repeat (TR) sequence, site-specific monoclonal antibodies have been generated.In order to characterize the antibody specificity at precise molecular level, and to discover MUC1 glycopeptide epitopes for cancer diagnosis and therapy, we will 1) chemically synthesize the PDTR，PGSTA，GVTSAbackbone, modify them with Tn，STn，STglycans, and study their binding specificity and affinity to antibodies; 2) engineer MUC1-expressing tumor cells by gene transfection and gene knockdown, to generate tumor cell lines which specifically express Tn, STn, or ST glycans, and study the antibody recognition and lysosomal internalization by such cell lines; 3) study the binding to clinical tumor tissue sections by antibodies and establish novel pathological classification of tumor according to glycopeptide epitopes; 4) study the expression of MUC1 glycopeptide epitopes by serum exosome from cancer patients; 5) study the treatment efficacy of antibody-drug conjugate and CAR NK cell therapy targeting MUC1 glycopeptides.