|
Fig 2. In-vitro substrate cleavage by xAtg4B fragments.A, Schematic representation of the protease substrates xLC3B-MBP (top) and xGATE16-MBP (bottom). Both fusion proteins contain an N-terminal polyHis-tag, a protease recognition site (xLC3B or xGATE16) and MBP (E. coli maltose-binding protein, MBP) as a model target protein. To ensure a comparable accessibility, the scissile bond is followed by the identical tri-peptide (AGT; Ala-Gly-Thr) in both substrate proteins. For simplicity, substrate names do not contain the polyHis-tag. B, In-vitro cleavage assay. 100 μM substrate (xLC3B-MBP (left) or xGATE16-MBP (right)) was incubated for 20 h at 37°C with defined concentrations of the indicated xAtg4B protease fragments. Cleavage products were separated by SDS-PAGE and stained with Coomassie G250. Shown are full-length substrate proteins (fl) and the C-terminal cleavage products (ccp). To estimate the completeness of cleavage, the band intensities were compared to a cleavage standard (see S3 Fig). C, Activity of xAtg4B fragments at 0°C. Indicated concentrations of xAtg4B protease fragments were incubated for 1 h at 0°C with 100 μM of the substrates sketched in A. For a similar comparison of xAtg4B fragments at 25°C see S4 Fig For examples of complete gels see S5 Fig.
|
|
Fig 3. Competitive binding of xAtg4B fragments to immobilized xLC3B and xGATE16.An equimolar mixture of full-length xAtg4B and indicated fragments (10 μM each) was incubated with immobilized xLC3B or xGATE16. A resin without bait protein (right panel) served as a specificity control. Bound proteins were analyzed by SDS-PAGE. xAtg4B degradation products lacking parts of the C-terminal extension are marked with an asterisk (*) in the input fractions. Note that binding is markedly reduced for protease fragments harboring C-terminal deletions. The pull-down efficiency is generally higher when using xLC3B instead of xGATE16 as bait. For complete SDS-PAGE gels see S6 Fig.
|
|
Fig 4. Cleavage efficiency at limiting substrate concentrations.The concentration of indicated protease fragments and the substrates xLC3B-MBP (left) or xGATE16-MBP (right) was titrated at constant protease/substrate ratio (1:1000 or 1:2000, respectively). After cleavage (1 h at 0°C), a fraction of each reaction corresponding to 1.2 μg (≈20 pmol) of substrate protein was analyzed by SDS-PAGE. Due to the different substrate concentrations, the absolute volume of the cleavage reaction analyzed by SDS-PAGE had to be adjusted accordingly.
|
|
Fig 5. Thermal stability of xAtg4B fragments.A, Long-term thermal stability. Indicated xAtg4B fragments were pre-incubated for 16 h at indicated temperatures in the presence of 20 mM DTT under argon to protect the active site cysteines from oxidation. The remaining activity was then assayed by treating 100 μM of xLC3B-MBP or xGATE16-MBP substrate with each protease fragment for 1 h at 0°C. B, Dynamic light scattering (DLS) analysis. xAtg4B fragments were diluted to a final concentration of 10 μM and assayed by DLS. The temperature was automatically raised by 1°C every 10 min. DLS signals were acquired just before each temperature step.
|
|
Fig 6. In-vitro cleavage characteristics of xAtg4B14-384.A, Time course. Substrates (100 μM) were incubated at 0°C with 500 nM of xAtg4B14-384. At indicated time points, aliquots were withdrawn. Cleavage products were separated by SDS-PAGE and stained with Coomassie G250. Shown are full-length substrate proteins (fl) and the C-terminal cleavage products (ccp). For a side-by side comparison of selected protease fragments see S4 Fig. B, Temperature dependence of substrate cleavage. 100 μM of xLC3B-MBP (left) or xGATE16-MBP (right) were incubated with xAtg4B14-384 for 1 h at defined temperatures. Note that in comparison to the xGATE16-MBP substrate, twice as much protease was used for cleavage of the xLC3B-MBP substrate. For a comparison of selected protease fragments see S8 Fig. C, Salt sensitivity. 100μM of each substrate was incubated for one hour at 0°C with 500 nM xAtg4B14-384 at NaCl concentrations ranging from 0.2 to 1.5 M. For a comparison of selected protease fragments see S8 Fig. D, P1' preference. Protease substrates used to analyze the P1' preference of xAtg4B14-384 followed the general outline shown in Fig 2A. Here, however, the P1' position of the P1-P1' scissile bond had been mutated to the potentially non-preferred residues methionine (Met), tyrosine (Tyr), arginine (Arg), glutamic acid (Glu), or proline (Pro). Solution cleavage assays were performed with indicated concentrations of xAtg4B14-384 for 1 h at 0°C. Bands marked with an asterisk (*) refer to the protease.
|
|
Fig 7. On-column cleavage using xAtg4B14-384.A, Schematic representation of substrate proteins used in (B)—(E). The N-terminal domain of E. coli IF2 (IF2d1 [65,66]) serves as a spacer. B and C, A silica-based Ni2+ chelate resin was pre-loaded with similar amounts of His14-bdNEDD8-mCherry and either His14-IF2d1-xLC3B-GFP (B) or His14-IF2d1-xGATE16-GFP (C). 50 μl aliquots were treated with indicated concentrations xAtg4B14-384 for 1 h at 4°C. Control incubations were performed with 4 μM bdNEDP1 or with buffer containing 400 mM imidazole. Resins and eluates were photographed while illuminated at 366 nm. GFP and mCherry in the eluate fractions were quantified via their specific absorptions. Quantification results are given below the respective eluate fractions. D and E, Protein purification using on-column cleavage by xAtg4B14-384. Indicated substrates were over-expressed in E. coli. After lysis and ultracentrifugation, the soluble material was incubated with a Ni2+ chelate resin. The resin was washed and treated with 500 nM xAtg4B14-384 at 4°C. At indicated time points, the concentration and purity of the released MBP was determined using the calculated absorption coefficient at 280 nm (OD280) and SDS-PAGE, respectively. Proteins remaining on the resin after 60 min were eluted by 500 mM imidazole. The time course of elution is shown in (D), the OD280 reading at 60 min elution time was set to 100%. Relevant steps of the purifications are shown in (E).
|
|
Fig 8. In-vitro cross-reactivity with other tag-cleaving proteases.A, Schematic representation of substrates used for (B) and (C). The TEV protease substrate contains an N-terminal His10-ZZ tag preceding the TEV protease recognition site. All other substrates follow the scheme described in Fig 2A, the protease recognition site, however, is replaced by the respective ubiquitin-like protein (UBL). B, Cross-reactivity between recombinant tag-cleaving proteases. bd, Brachypodium distachyon; tr, Triticum aestivum (summer wheat); xUb, Xenopus ubiquitin. Bands marked with an asterisk (*) originate from the respective protease. For complete gels see S6 Fig. C, Detailed titration analysis of cross-reactivity between Xenopus laevis (x), S. cerevisiae (sc) and wheat (tr) Atg4 homologs.
|
|
Fig 9. Stability of UBL fusions in eukaryotic lysates and in S. cerevisiae.A, Schematic representation of substrates used for (B). B, Stability of protease substrates in cell extracts. Note that in wheat germ extract no proteolytic fragments originating from SUMOstar-, xLC3B- or xGATE16-containing substrates can be detected. For complete blots and stained membranes see S10 Fig. C, Schematic representation of substrates used for expression in S. cerevisiae (D) harboring an N-terminal ZZ-tag, an ubiquitin-like protein (UBL) and a C-terminal Citrine. D, In-vivo stability of protease substrates in S. cerevisiae. Indicated protease substrates were over-expressed in a S. cerevisiae strain constitutively expressing H2B-CFP. Total cell lysates were analyzed by Western blot with antibodies recognizing the ZZ-tag (upper panel) or Citrine and CFP (middle panel). Equal loading was confirmed by staining the membrane after blotting (lower panel). Bands marked with an asterisk (*) originate from ZZ-tagged proteins cross-reacting with the anti-Citrine/CFP antibody. For complete original blots and stained membranes see S11 Fig. E, Cleavage of UBL substrates in extracts and in S. cerevisiae. ++, highly efficient cleavage; +, cleavage;–, traces cleaved;––, no cleavage; n.d.: not determined; 1, data not shown.
|
|
Fig 10. One-step protein purification from S. cerevisiae.ZZ-UBL-Citrine fusions sketched in (A) were over-expressed in S. cerevisiae. Cells were lysed and the soluble material was incubated with an anti-ZZ affinity resin. After washing off unbound material, highly pure Citrine was eluted by treatment with 0.1 μM SUMOstar protease (B), 1 μM xAtg4B14-384 (C) or 1 μM bdNEDD8 (D) for 1 h at 4°C. Material remaining on the resin was analyzed after elution with SDS sample buffer. The asterisk (*) denotes the full-length xLC3B fusion protein. The filled circle (•) marks band partially corresponding to low levels of free Citrine originating from in-vivo cleavage of the respective SUMOstar and bdNEDD8 fusion proteins.
|
|
Fig 1. Bacterial expression and purification of xAtg4B protease fragments.
A, Schematic illustration of expression constructs used for (B) and (C). B, Exemplary purification of xAtg4B14-384. His14-TEV-xAtg4B14-384 was over-expressed in E. coli strain NEB Express. After cell lysis and centrifugation, the soluble material was applied to a Ni2+ chelate resin. Bound proteins were eluted with imidazole and treated with polyHis-tagged TEV protease over night at 4°C before loading on a Superdex 200 gel filtration column. The pooled peak fractions mainly containing cleaved xAtg4B14-384 and TEV protease were subjected to a reverse Ni2+ chromatography step (rev. Ni2+). Here, the polyHis-tagged TEV protease bound to the resin while pure xAtg4B14-384 was found in the unbound fraction. Purification of other xAtg4B fragments was done identically. Minor amounts of degradation bands (*) originate from cleavage within the flexible C-terminus. C, Purity of xAtg4B protease fragments. 40 pmol (≈1.6 μg) of purified protease fragments were separated by SDS-PAGE and stained with Coomassie G250.
|
