Lind P, Larsson K, Spira J, Sydow-Bäckman M, Almstedt A, Gray E, Sandberg H
Pharmacia AB, Biopharmaceuticals, Stockholm, Sweden.
Eur J Biochem. 1995 Aug 15;232(1):19-27. doi: 10.1111/j.1432-1033.1995.tb20776.x.
Recombinant molecules similar to the smallest active plasma-derived factor VIII molecule, a complex of an 80-kDa and a 90-kDa polypeptide chain lacking the B domain, have been produced using various factor VIII cDNA constructs in order to obtain primary translation products which were efficiently processed into the 80 + 90-kDa complex. Three types of single-chain cDNAs encoding B-domain-deleted derivatives factor VIII were designed, taking account of sites at Arg740 and Glu1649, assumed to be important for processing factor VIII. In the type 1 constructs, either Arg747, Arg752, or Arg776 in the N-terminal region of factor VIII B domain was fused to the N-terminus (Glu1649) of the 80-kDa subunit. In the type 2 construct r-VIII SQ, Ser743 was fused to Gln1638, creating a link of 14 amino acids between the C-terminus (Arg740) of the 90-kDa chain and N-terminus of the 80-kDa chain, whereas in type 2 r-VIII RH, Arg747 was fused to His1646. In the type 3 constructs, the B-domain was completely removed or replaced with 1-4 Arg residues. After expression in Chinese hamster ovary cells, the type 1 derivatives and the type 3 derivatives with 0-2 Arg residues inserted were found to be only partially processed and contained a large amount of the 170-kDa primary translation product. In contrast, most of the type 2 derivatives r-VIII SQ and r-VIII RH and the type 3 derivatives r-VIII R4 and r-VIII R5 containing three or four extra Arg residues preceding the N-terminus of the 80-kDa chain were processed into the desired 80 + 90-kDa chain complexes. The feature common to the most efficiently processed factor VIII deletion derivatives was that they contained the recognition motif for proteolytic cleavage by the membrane-bound subtilisin-like protease furin, which is expressed in most types of cells; that is, basic amino acid residues at positions -1 and -4 relative to the cleavage site at Glu1649. Biochemical studies of r-VIII SQ and r-VIII R5, two of the most effectively processed factor VIII derivatives, showed that both proteins had a normal factor VIII cofactor function, and had N- and C-termini of the 80-kDa and 90-kDa chains corresponding to those found in plasma-derived factor VIII.
为了获得能有效加工成80 + 90 kDa复合物的初级翻译产物，已使用各种因子VIII cDNA构建体生产了与最小活性血浆来源的因子VIII分子（一种由80 kDa和90 kDa多肽链组成且缺少B结构域的复合物）相似的重组分子。考虑到假定对因子VIII加工很重要的Arg740和Glu1649位点，设计了三种编码B结构域缺失的因子VIII衍生物的单链cDNA。在1型构建体中，因子VIII B结构域N端区域的Arg747、Arg752或Arg776与80 kDa亚基的N端（Glu1649）融合。在2型构建体r-VIII SQ中，Ser743与Gln1638融合，在90 kDa链的C端（Arg740）和80 kDa链的N端之间形成了14个氨基酸的连接，而在2型r-VIII RH中，Arg747与His1646融合。在3型构建体中，B结构域被完全去除或被1 - 4个Arg残基取代。在中国仓鼠卵巢细胞中表达后，发现1型衍生物和插入0 - 2个Arg残基的3型衍生物仅被部分加工，并且含有大量170 kDa的初级翻译产物。相比之下，大多数2型衍生物r-VIII SQ和r-VIII RH以及在80 kDa链N端之前含有三个或四个额外Arg残基的3型衍生物r-VIII R4和r-VIII R5被加工成了所需的80 + 90 kDa链复合物。加工效率最高的因子VIII缺失衍生物的共同特征是它们含有膜结合的枯草杆菌蛋白酶样蛋白酶弗林蛋白酶（furin）进行蛋白水解切割的识别基序，弗林蛋白酶在大多数类型的细胞中都有表达；也就是说，相对于Glu1649处切割位点的 - 1和 - 4位的碱性氨基酸残基。对加工最有效的两种因子VIII衍生物r-VIII SQ和r-VIII R5的生化研究表明，这两种蛋白都具有正常的因子VIII辅因子功能，并且80 kDa和90 kDa链的N端和C端与血浆来源的因子VIII中的对应端相同。
