Skip to main content
. 2009 Dec 15;11(2):265–276. doi: 10.1111/j.1364-3703.2009.00603.x

Figure 4.

Figure 4

Schematic representation of Soybean mosaic virus (SMV) constructs used for the introduction of helper component‐proteinase (HC‐Pro) mutations. (a) SMV genome organization. HC‐Pro is processed by P1 and by HC‐Pro by itself in an in cis manner. Amino acid sequences of the peptide cleavage sites recognized by either P1 or HC‐Pro are underlined, and arrowheads indicate the location of the cleaved peptide bond. (b) Construction of an available insertion cassette between P1 and P3. A cloning site (XbaI) and the NIa‐Pro cleavage site (ESVSLQ/S) were inserted into the polyprotein open reading frame (ORF) between the P1 and P3 cistrons. The amino acid sequences of the peptide cleavage sites recognized by either P1 or NIa‐Pro are underlined, and arrowheads indicate the location of the cleaved peptide bond. (b) Schematic representation of HC‐Pro processing from the reconstructed SMV genome. HC‐Pro, which is inserted into pSMVdHC utilizing the XbaI site, is processed by P1 in cis and by NIa‐Pro in an in cis manner, and is predicted to have two additional amino acids (SR) at its N‐terminus and eight amino acids (SRESVSLQ) at its C‐terminus.