Fig. 4. (A) Map-based cloning of COV1 based on data obtained from recombinants identified in the testcross screen. The number of recombinants identified between COV1 and closely linked markers are shown in brackets. Subclones used to determine complementation are shown as unbroken black lines. The cov1-1 mutant was partially complemented with pGDP1 and fully complemented with pGDP6 and pGDP7. Other genes in the region are indicated by numbers (1, At2g20160; 2, At2g20150; 3, At2g20140; 4, At2g20130; 5, At2g20110). The intron-exon structure and position of COV1 and the closely related gene LCV1 are also shown (bottom). The predicted translation initiation sites are indicated by arrows. (B) The predicted topology of COV1. The protein contains three predicted membrane-spanning domains; the N-terminal end of the protein is predicted to be intracellular and the C-terminal end is predicted to be in the wall. The mutation in cov1-1 (*1) is in exon 3, which forms part of the first membrane-spanning domain and results in an amino acid change from proline to serine (P⁷⁶S). The mutation in cov1-2 (*2) is located in exon 4, which is predicted to lie just after the second membrane domain and results in an amino acid change from glycine to arginine (G¹³⁰R).