Supplemental Data

Files in this Data Supplement:

• Supplemental Figure 1 - Figure S1. Effects of cycloheximide on membrane internalization, and bleedthrough controls. Treatment of plants expressing ACR4:GFP with cycloheximide for 3.5 hours, followed by a subsequent 30 minute incubation in cycloheximide and FM4-64 (A). Internalized vesicles are easily observable in these samples, suggesting that membrane internalization has not been disrupted. Cells from a control plant subjected to a 3.5h treatment in 0.1% ethanol and a 30 minute treatment in 0.1% ethanol and FM4-64 are shown in B. C shows bleedthrough of FM4-64 fluorescence into the GFP channel in wild-type FM4-64 treated plants using conditions similar to those in our other experiments. D shows an overlay of the GFP channel with FM4-64 signal. Bright bodies occasionally observed in wild-type plants using the GFP channel are indicated with a closed arrow (C and D). These do not co-localise with FM4-64 staining bodies (open arrow in D).
• Supplemental Figure 2 - Figure S2. Complementation of the acr4 mutant phenotype. Plate A shows seeds from a wild-type (Col-0) silique. Seeds from an acr4-2 mutant silique fully complemented by ΔTNFR are shown in B. C shows acr4-2 seeds with strong partial complementation by ΔTNFR. Weak partial complementation by ΔTNFR a lack of complementation by ACR4C180-Y:GFP are shown in D and E respectively. Weak partial complementation usually involves a decrease in ovule abortion only, with seed phenotypes remaining severe. Seeds from an acr4-2 homozygous plant are shown in F.
• Supplemental Figure 3 - Figure S3. Alignment of proteins predicted to show structural similarity to the ACR4 crinkly repeat domain. The optimal alignment between the target sequence of the crinkly repeat domain, ACR4 and the template structures, BLIP-II and Rcc1 as described in the Materials and Methods section. The seven putative disulfide bridges within the seven repeats, restrained during model building, are indicated. The extent of secondary structure, for the ACR4 model derived using DSSP (Kabsch and Sander, 1983) is shown on the sequence of ACR4, color coded sequentially according to Figure 4a. The model displays a predominantly β-domain fold, with one helical region within repeat 2, indicated within the dotted blue-shaded box in the alignment. Completely conserved residues between the target and templates are highlighted in red. The deleted loop in repeat 3 of the model, for which no template-derived restraints were available, is indicated in blue.
• Supplemental Figure 4 - Figure S4. Multiple sequence alignment of ACR4-related proteins containing the seven putative crinkly repeats. The extracellular seven crinkly repeat sequence from ACR4 (SwissProt/TrEMBL (Bairoch and Apweiler, 1997) Accession Number: Q9LX29) was used in a BLAST (Altschul et al., 1997) search against the ?non-redundant? protein sequence database using the NPS@ (Combet et al., 2000) server, to identify putative homologues, related to the query sequence using the default BLOSUM 62 matrix (Henikoff and Henikoff, 1992). Sequences containing the seven putative repeats, and having an E-Value less than 1 X 10-⁰⁶ detected by the BLAST search, were used in the construction of a multiple sequence alignment using the program MUSCLE version 3.41 (Edgar, 2004). The MUSCLE-derived multiple sequence alignment was manually edited to ensure the most plausible alignment. Sequence identifiers are referred to by their SwissProt/TrEMBL Accession Numbers followed by the organism name (Abbreviations used in the identifiers: AT = Arabidopsis thaliana, OS = Oryza sativa, ZM = Zea mays, MA = Musa acuminata, NT = Nicotiana tabacum, BB = Bdellovibrio bacteriovorus). The shading (50% agreement) for each repeat was derived using the program BOXSHADE version 3.21 (http://www.ch.embnet.org/software/BOX_form.html).
• Supplemental Table 1
• Supplemental Table 2
• Supplemental Table 3