Fig 2: Truncated Sez6L2 has decay accelerating activity for the alternative pathway C3 convertase but has only modest decay accelerating activity for the classical/lectin pathway C3 convertase. (A–D) Alternative C3 convertase assay: (A–C) A 96 well plate coated with C3b was incubated with Factor B and Factor D to form the C3 convertase C3bBb, then incubated with Sez6L2-MH or FH at various concentrations ranging from 0-6000 nM (equivalent to 0-500 µg/mL) to assess their decay accelerating activity. Factor B remaining bound to the plate (as C3bBb) was detected using an anti-Factor B antibody ELISA in (A) and Bb released into the supernatant is shown via western blot in (B) with Bb band densities quantified in (C). For the western blot and quantification, Sez6L2-MH and FH concentrations are listed in both nM and µg/mL. Quantification of Bb band densities were normalized to the 0 nM control lanes. N=3 samples per group. (D) The alternative C3 convertase decay ELISA described above was repeated comparing 0, 1uM, and 5uM of Sez6L2-MH, H-DAF, FH, and BSA. (E) Classical C3 convertase assay: A plate coated with C4b was incubated with C2 and C1s-enzyme to form the classical/lectin pathway C3 convertase C4b2b, then incubated with Sez6L2-MH, H-DAF at concentrations ranging from 0 to 7000 nM to assess their decay accelerating activity. C2 remaining bound to the plate (presumably as C4b2b) was detected using an anti-C2 antibody ELISA. For A, D, and E ELISAs: N=3 (1 experiment with 3 replicates; representative of 2-3 independent experiments). Statistics: one-way ANOVAS with Holms-Sidak multiple comparison’s tests to controls. *p < 0.05, **< 0.01.

Fig 3: Full Length Sez6L2, Sez6, and Sez6L inhibit C3b/iC3b opsonization of CHO cells by the classical pathway. (A, B) Sez6L2 inhibits C3b/iC3b opsonization at a range of serum concentrations. CHO cells were transfected with plasmids for GFP alone or with Myc-tagged Sez6L2 (M-Sez6L2) or His-tagged DAF (H-DAF). CHO cells were coated with antibodies and exposed to 0-20% C5-depleted human serum for one hour and then immuno-stained with anti-C3b/iC3b antibodies and analyzed by flow cytometry. One experiment is shown that is representative of two independent experiments. (B) C3b/iC3b on GFP transfected cells with or without M-Sez6L2 or H-DAF at 15% serum. ANOVA (P=0.0016; F(2,6)=22.51). N=3; one experiment with three replicates (representative of 3+ independent experiments). (C) Schematic of Sez6L2, Sez6, and Sez6L protein domain structures. (D–I) CHO cells were transfected with the indicated Myc-tagged cDNAs and processed as outlined in A with 15% C5 depleted serum, except that an anti-Myc antibody was used in place of GFP to identify transfected and expressing CHO cells. (D) 5% Contour plots of C3b/iC3b versus Myc fluorescence (top layer) and C3b/iC3b fluorescence histograms (bottom layer) of the same samples normalized to mode and compared to baseline cells not exposed to serum. For Contour plots, boxed regions highlight cells designated as Myc-positive (top box) and Myc-negative (lower box) populations. For C3b/iC3b histograms, dark grey, solid line population = Myc-positive cells; Light grey, dotted line population= Myc-negative cells; White, dashed grey line population = baseline. Representative of 4+ independent experiments. (E) Quantification of the average median C3b/iC3b fluorescence intensity from Myc-positive and Myc-negative cells within each sample. Statistics = t-tests. N=3 (one experiment with three replicates; Representative of 4+ independent experiments). (F) Average median C3b/iC3b fluorescence intensities after normalization to the Myc-negative cells from each experimental group. ANOVA between Myc-positive cell populations (p<0.001; F(4, 15)=64.53). Sez6L2 inhibits C3b/iC3b opsonization at a level comparable to positive control MCP. Sez6 is a stronger complement inhibitor than Sez6L2 and Sez6L is a weaker inhibitor. (F) Average median Myc fluorescence intensity from Myc-positive cells. ANOVA (p<0.001; F(4, 15)=36.79). (G) Average % of Myc-positive cells in each experimental group (ANOVA, p=0.115; F(4, 15)=2.224). For sections (F–H), N=4 (four independent experiments). (I) Sez6 blocks complement opsonization more efficiently than Sez6L2 and Sez6L even when comparing similar levels of Myc surface expression. Average C3b/iC3b median fluorescence intensity normalized to internal Myc-negative populations for M-Sez6, M-Sez6L2, and M-Sez6L samples shown relative to the Myc median fluorescence intensity. N=3 (one experiment with three replicates, Representative of three independent experiments). For all graphs *p < 0.05; **p < 0.01; #p < 0.001 for all Myc-positive groups compared to M-CR2.

Fig 4: Sez6 family expression in the hippocampus. (A) Sez6, Sez6L, and Sez6L2 are expressed by principal (excitatory, pyramidal) neurons of the mouse hippocampus at much higher levels than other known complement regulators (namely Crry, C4BP, CFH, C1-INH, DAF, and MCP). Expression data was obtained from Hipposeq: a comprehensive RNA-Seq database of gene expression in hippocampal principal neurons [http://hipposeq.janelia.org (60)]. The RNA samples used in this database were isolated from mouse hippocampal principal neurons micro-dissected from the CA1, CA3, or Dentate Gyrus (DG) cell layers of the hippocampus at Postnatal Day 25-32. Differential gene expression is shown in the heatmap with the relative units of FPKM (Fragments per Kilobase of Exon per Million Reads Mapped.) (B) Brain sections from adult WT mice or Sez6 triple knockout mice (TKO) were immuno-stained for Sez6L2 (green) and DAPI and imaged in the CA1 region of the hippocampus. Scale Bar= 27 µm. High density Sez6L2 staining occurs around cell bodies in the pyramidal layer, but significant Sez6L2 is also found in the stratum radiatum and stratum oriens. (C) Higher magnification images of sections immuno-stained for Sez6L2 (green) and the postsynaptic protein, Homer1 (red), shows a subset of Sez6L2 is found near or co-localized with synapses in the stratum radiatum. Scale bar = 1.8 µm.