Supplemental Material

This article contains the following supporting material:

• Supplemental Figure 2 - Alternative splicing may create a brain-specific isoform of Sec16L.
  Shown is the exon/intron structure predicted by the Ensembl genome browser for the human Sec16L gene (KIAA0310). Our liver cDNA clone matches the predicted exon/intron structure, and encodes a protein of 2154 amino acids. By contrast, the previously described cDNA from brain (Nagase et al. , 1997; Nakajima et al. , 2002) includes the predicted 75-bp intron between the 28th and 29th exons, and therefore encodes a 2179-amino acid protein. Numbers indicate the nucleotides or amino acids relative to the start codon. Alternating light yellow shading marks codons, which are blue for exons or black for introns. Dark yellow shading marks 3’ untranslated sequence.
• Supplemental Figure 1-1 - Sequence alignment showing the central conserved domains of Sec16 homologs from various eukaryotes.
  Residues matching the consensus are highlighted in yellow. For this alignment, the boundaries of the central conserved domain of S. cerevisiae Sec16 were adjusted from our previous estimate of residues 992-1420 (Connerly et al. , 2005) to residues 984-1420.
• Supplemental Figure 1-2 - Sequence alignment showing the central conserved domains of Sec16 homologs from various eukaryotes.
  Residues matching the consensus are highlighted in yellow. For this alignment, the boundaries of the central conserved domain of S. cerevisiae Sec16 were adjusted from our previous estimate of residues 992-1420 (Connerly et al. , 2005) to residues 984-1420.
• Supplemental Figure 3 - FLAG-tagged Sec16L and Sec16S colocalize with Sec23A.
  A FLAG epitope was placed at the N terminus of Sec16L or Sec16S, and HeLa cells were transfected with the resulting plasmids. At 24 h after transfection, the cells were processed for immunofluorescence with monoclonal anti-FLAG antibody plus polyclonal anti-Sec23A antibody. Scale bar, 20 μm.
• Supplemental Figure 4 - tER sites become smaller and more numerous after prolonged expression of GFP-Sec16L but not of GFP-Sec16S.
  HeLa cells were transfected with a plasmid encoding either GFP Sec16L or GFP Sec16S. At 12, 24, or 36 h after transfection, cells were fixed and subjected to flourescence microscopy, and the average number of tER sites was estimated by counting 10 randomly chosen cells expressing either GFP Sec16L or GFP Sec16S. Bars represent standard errors of the mean.
• Supplemental Figure 5 - RNAi-mediated double knockdown of Sec16L and Sec16S disrupts tER sites in a manner similar to knockdown of either individual protein.
  The experiment was performed as in Figure 3, except that cells were treated with RNAi duplexes against both Sec16L and Sec16S. RT-PCR analysis (not shown) indicated that mRNA levels were reduced to approximately 21% of control levels for Sec16L and to approximately 12% of control levels for Sec16S. Scale bar, 20 μm.
• Supplemental Figure 6 - The central conserved domains of Sec16L and Sec16S do not confer tER localization.
  GFP was fused to the central conserved domain of Sec16L or Sec16S (see Figure 5), and the indicated protein was visualized at 24 h after transfection. Scale bar, 20 μm.