Sequence-specific 1H, 13C, and 15N resonance assignment of the autophagy-related protein Atg8
- First Online:
- Cite this article as:
- Schwarten, M., Stoldt, M., Mohrlüder, J. et al. Biomol NMR Assign (2009) 3: 137. doi:10.1007/s12104-009-9159-3
- 96 Views
The autophagy-related protein Atg8 is important for the formation of autophagosomes as it mediates membrane fusion. To elucidate the solution structure of Atg8 backbone and side chain chemical shifts of Atg8 were assigned as far as possible.
KeywordsAutophagyAtg8Heteronuclear NMRResonance assignment
Autophagy is an important process within eukaryotic cells for intracellular protein degradation (Mizushima 2005). During autophagy double membrane-bound structures called autophagosomes are formed around a subcellular volume that may contain diverse proteins and organelles which will ultimately be subject to degradation. Atg8 is an ubiquitin-like protein required for this process in Saccharomyces cerevisiae. Atg8 is a substrate for conjugation with the lipid phosphatidylethanolamine (PE) by an ubiquitin-like system. It mediates the tethering and hemifusion of membranes during the formation of autophagosomes (Nakatogawa et al. 2007). Atg8 is a 14 kDa (117 aa) monomeric single-chain protein, which shares 73% of sequence identity with already assigned Apg8a from Arabidopsis thaliana (Chae et al. 2005). Further it is a homologue to the mammalian GABARAP [56% sequence identity] (Stangler et al. 2001) and LC3 [37%] (Kouno et al. 2004). In order to obtain a better understanding of the processes during autophagosome formation we want to study the solution structure of Atg8 and characterise its dynamical behaviour in solution as well as its membrane association. Here we report the 1H, 13C, 15N sequence-specific backbone and side chain resonance chemical shift assignment of Atg8.
Methods and experiments
Expression and purification of isotope labelled Atg8
The gene coding for Atg8 was cloned into the vector pGEX-4T-2 (GE Healthcare, Munich, Germany) using BamHI and NotI restriction sites introduced by PCR. The sequence was verified by DNA sequencing. E. coli strain BL21 DE3 Rosetta cells were transformed with the resulting plasmid DNA and grown at 37°C in LB medium before being transferred to M9 medium supplemented with 13C glucose and 15N ammonium chloride. Expression was induced over night by adding 1 mM IPTG. The glutathione-S-transferase (GST)-Atg8 fusion protein was purified from the soluble extract by affinity chromatography on glutathione sepharose 4B (Amersham Bioscience). Thrombin (Merck) cleavage yielded full-length Atg8 with additional glycine and serine residues at the N-terminus. For final purification, the sample was applied to a Superdex 75 prepgrade size exclusion chromatography column (Amersham Biosciences) and the Atg8 containing fractions were pooled and concentrated by ultrafiltration to a final concentration of 500 μM.
Uniformly enriched 13C, 15N samples of 500 μM Atg8 were prepared using 20 mM sodium phosphate buffer, pH 6.4, 150 mM sodium chloride, 5 mM Dithiothreitol, 1 mM EDTA, 7% (v/v) D2O, and 0.05% (v/v) sodium azide.
NMR experiments were performed at 298 K on Varian Unity INOVA instruments, equipped with a cryogenically cooled Z-axis pulse-field-gradient (PFG) triple resonance probe and a triple-axis PFG triple resonance probe respectively, at proton frequency of 800 and 600 MHz. To obtain backbone resonance assignments 2D (1H-15N)-HSQC, 3D BEST-HNCA, BEST-HNCACB and BEST-HNCO spectra were recorded (Schanda et al. 2006; Lescop et al. 2007). Side chain assignments were obtained from 2D ct-(1H-13C)-HSQC and 3D H(C)CH-COSY, HNHA, (1H-1H-15N)-NOESY-HSQC and (1H-13C-1H)-HSQC-NOESY spectra.
Assignments and data deposition
Secondary structure prediction based on the amino acid sequence of Atg8 and the obtained chemical shifts by Protein energetic conformational analysis from NMR chemical shifts [PECAN] (Eghbalnia et al. 2005) lead to the prediction of three α-helical and four β-strand segments, whereas α-helices comprise residues A16-A22, G58-I68 and S92-Q97 and β-strands consist of residues I29-E34, K48-V51, I76-V80 and L105-S110. This is in excellent agreement with those reported for Apg8a from Arabidopsis thaliana (Chae et al. 2005). A minor difference for the apparent start of the very N-terminal α-helix is presumably due to the different extent in the backbone assignments of the N-terminal regions of both proteins.
The data of the 1H, 13C and 15N chemical shifts have been deposited in the BioMagResBank (http://www.bmrb.wisc.edu) under accession number 16120.
We would like to thank Sven Schünke for helpful discussions. This study was supported by a research grant from the Deutsche Forschungsgemeinschaft (DFG) to D. W. (Wi1472/5) and a fellowship from the International Helmholtz Research School on Biophysics and Soft Matter (“BioSoft”) to M.Sch.