Protein solubility and folding enhancement by interaction with RNA

Abstract

While basic mechanisms of several major molecular chaperones are well understood, this machinery has been known to be involved in folding of only limited number of proteins inside the cells. Here, we report a chaperone type of protein folding facilitated by interaction with RNA. When an RNA-binding module is placed at the N-terminus of aggregation-prone target proteins, this module, upon binding with RNA, further promotes the solubility of passenger proteins, potentially leading to enhancement of proper protein folding. Studies on in vitro refolding in the presence of RNA, coexpression of RNA molecules in vivo and the mutants with impaired RNA binding ability suggests that RNA can exert chaperoning effect on their bound proteins. The results suggest that RNA binding could affect the overall kinetic network of protein folding pathway in favor of productive folding over off-pathway aggregation. In addition, the RNA binding-mediated solubility enhancement is extremely robust for increasing soluble yield of passenger proteins and could be usefully implemented for high-throughput protein expression for functional and structural genomic research initiatives. The RNA-mediated chaperone type presented here would give new insights into de novo folding in vivo.

Figure 1. Development of RBPs as solubility enhancers.

(a) Proposed model for RNA binding-mediated protein folding. Both the folded RBD at N-terminal position and bound RNA prevent inter-molecular interactions among folding intermediates, leading to soluble expression and favoring kinetic network into productive folding. The number of black bars (| and ∥) represents the extent of aggregation inhibition. (b) The comparison of solubility-enhancing ability by RBP with that of MBP. E. coli lysyl tRNA synthetase (LysRS) and influenza virus nucleoprotein (NP) were used as RBP to monitor the soluble expression of tobacco etch virus (TEV) protease. The solubility-enhancing ability of RBP was compared to that of MBP. The fusion proteins were expressed at 37°C and their solubility was analyzed by SDS-PAGE. M, T, S, and P represent molecular weight marker, total lysates, soluble fraction, and insoluble fraction, respectively. (c) Autocatalytic cleavage of LysN-TEV containing TEV cleavage sequence between LysN and TEV protease in E. coli cytosol. Non-induced (−) and IPTG induced (+) cell extracts were analyzed by SDS-PAGE. The uncleaved LysN-TEV was not detected clearly on SDS-PAGE due to efficient cleavage. (d) Cell proliferation assay of GCSF expressed as LysN-GCSF. The purified TEV protease described in Figure 1c was used to cleave the purified LysN-GCSF. The purified LysN, TEV protease, and LysN-GCSF before and after cleavage with TEV protease were compared with the GCSF standard in the cell proliferation assay as described in Methods.

Figure 2. RNA-mediated protein folding in vitro.

(a) In vitro refolding of LysRS. The refolding of LysRS was performed in the presence of E. coli tRNA^Lys (2 µM), yeast tRNA^Phe (2 µM) or yeast total RNA (the amount equivalent to 2 µM of E. coli tRNA^Lys), and then the enzymatic activity of refolded LysRS was investigated using aminoacylation assay, as described in Methods. (b) In vitro refolding of LysN-EGFP. Refolding of LysN-EGFP was performed in vitro in the presence of E. coli tRNA^Lys or yeast tRNA^Phe. The fluorescence emission of refolded LysN-EGFP was continuously monitored. As a control, MBP-EGFP was tested under the same condition. (c) The effects of tRNAs on the refolding yield at 100 min in (b) were compared and summarized. The fluorescence intensity in the absence of tRNA was set to 100%.

Figure 3. Correlation between RNA binding and solubility enhancement.

(a) Expression of unfused wt LysRS, LysRS(K130A), and LysRS(T133A) at 37°C. These proteins have hexa-histidine tag at their C-termini. Mutants were constructed using PCR overlapping mutagenesis. (b) The effects of point mutations on the solubility of LysRS fusion proteins in vivo. Three passenger proteins GNB2L1, ANGPTL4 and FAM3D were fused to the C-termini of wt LysRS, LysRS(K130A), and LysRS(T133A). The expression temperature was 37°C (30°C in case of FAM3D fusion proteins). The representative SDS-PAGE data are shown in left panel. The solubility of fusion proteins obtained by three independent experiments is summarized in right panel. (c) RNA binding analysis of LysRS and its mutants using gel-retardation assay. The binding affinity of wt LysRS, LysRS(K130A), and LysRS(T133A) to 5′-³²P-labeled tRNA^Lys was analyzed by gel-retardation assay as described in Methods. For the competition assay, the cold tRNA^Lys (middle) and tRNA^Phe (right) of various concentrations (0, 0.46, 1.16, and 2.3 µM) was used. Arrow indicates the LysRS-tRNA^Lys complexes. Note that the relative amounts of tRNA^Lys binding to LysRS, LysRS(K130A), and LysRS(T133A) are 1, 0.35, and 1.17, respectively. (d) The effect of tRNA coexpression on the solubility of LysRS fusion proteins in vivo. GNB2L1 as a C-terminal passenger protein was fused to wt LysRS and LysRS(K130A), and the fusion proteins were expressed at 37°C.

Figure 4. Comparison of the solubility of MBP and LysRS fusion proteins.

(a) SDS-PAGE corresponding to the total, soluble and insoluble fractions of fusion proteins. (b) The summary of the solubility of directly expressed proteins and their fusion proteins. The proteins were expressed either at 37°C or at 30°C (AP1M2, CXX1, FAM3D, HPR, mVDUP1). The bands of MBP-CXX1 and MBP-LECT2 were not detectable on the SDS-PAGE. The expression cassette of fusion proteins comprises LysRS (or MBP)-D6-SG-ENLYFQ-MCS-H6 where the sequence of ENLYFQ, MCS, and H6 are TEV protease recognition site, multi-cloning site of KpnI-BamHI-EcoRV-SalI-HindIII, and C-terminal hexahistidine tag, respectively.

Figure 5. Generation of target proteins from the fusion proteins.

Twenty two LysRS fusion proteins purified on one-step Ni-affinity chromatography were incubated at 30°C for 2 h in 30 µl containing 50 mM Tris–HCl, pH 8.0, 0.5 mM EDTA, 1 mM DTT with 1 unit of TEV protease (Invitrogen). All samples were analyzed by SDS-PAGE. The arrow indicates the released target proteins.