Anal. Chem. 2010, 82, 4637–4643
Production of Ribosome-Released Nascent Proteins with Optimal Physical Properties David R. Ziehr, Jamie P. Ellis,† Peter H. Culviner, and Silvia Cavagnero* Department of Chemistry, University of Wisconsin-Madison, 1101 University Avenue, Madison, Wisconsin 53706 The growing interest in protein folding under physiologically relevant conditions has prompted investigations requiring direct comparisons between ribosome-bound and ribosome-released nascent proteins. Such studies, involving the ad hoc release of newly synthesized proteins from stalled ribosomes, demand a release agent able to produce nonaggregated native proteins and preserve the overall nature of the medium. Here, we explore hydroxylamine, a reactant rarely used to release nascent chains, and compare it to other ribosome-release agents: puromycin, RNase A/EDTA, and sodium hydroxide. Ribosome-bound nascent chains corresponding to the sequence of apoHmpH, the Escherichia coli N-terminal domain of Hmp, were used as a model system. Fluorescence anisotropy decays were employed to probe the selfassociation and overall physical properties of nascent proteins. Gel electrophoresis and RNA chip microfluidic capillary electrophoresis yielded information on the integrity of nascent peptidyl-tRNAs and ribosomes, respectively. Of the four reagents examined, only hydroxylamine releases nascent apoHmpH without causing extensive aggregation or degradation of the ribosome. Hydroxylamine does not introduce large hydrophobic C-terminal modifications and functions at nearly physiological pH. It is therefore a suitable reagent for the ad hoc release of nascent proteins from the ribosome. While our understanding of protein folding has grown considerably since the earliest studies on in vitro renaturation from chemically unfolded states,1 there is still a critical lack of insight into the process by which proteins achieve their native state in the cell. Stalled ribosome-bound nascent chain complexes (RNCs) in cell-free systems and other biologically relevant media provide a model environment to explore cotranslational protein dynamics and folding and the role of cotranslationally active chaperones. Antibody binding and enzymatic activity assays of RNCs demonstrate cotranslational folding.2-11 Fluorescence and nuclear magnetic resonance offer additional biophysical insights into the * To whom correspondence should be addressed. Phone: 608-262-5430. Fax: 608-262-9918. E-mail:
[email protected]. † Current address: Department of Molecular Biology, The Scripps Research Institute and Skaggs Institute for Chemical Biology, 10550 North Torrey Pines Road, MB-2, La Jolla, CA 92037. (1) Bartlett, A. I.; Radford, S. E. Nat. Struct. Mol. Biol. 2009, 16, 582–588. (2) Hartl, F. U.; Hayer-Hartl, M. Nat. Struct. Mol. Biol. 2009, 16, 574–581. (3) Zhang, G.; Hubalewska, M.; Ignatova, Z. Nat. Struct. Mol. Biol. 2009, 16, 274–280. 10.1021/ac902952b 2010 American Chemical Society Published on Web 04/16/2010
structure and dynamics of nascent polypeptides and proteins.12-18 Many of these investigations, as well as other studies addressing the mechanism of translation,19,20 rely critically on comparisons between ribosome-bound and ribosome-released nascent proteins and peptides. The generation of ribosome-released species is best achieved in situ, to ensure that the same environment is sampled by a given protein chain before and after its departure from the ribosome. The three most common reagents employed to induce the release of ribosome-bound nascent proteins are puromycin, ribonuclease A in the presence of sodium ethylenediamine tetraacetate (RNase A/EDTA), and sodium hydroxide. Puromycin, a well-studied inhibitor of translation,21 is structurally similar to the 3′ end of Tyr-tRNATyr. Puromycin releases nascent proteins from the ribosome by nucleophilic attack of the free amine on the ester linking the nascent protein to the tRNA. Previous reports caution that puromycin may induce aggregation of released proteins.22,23 Therefore, (4) Lakshmipathy, S. K.; Tomic, S.; Kaiser, C. M.; Chang, H. C.; Genevaux, P.; Georgopoulos, C.; Barral, J. M.; Johnson, A. E.; Hartl, F. U.; Etchells, S. A. J. Biol. Chem. 2007, 282, 12186–12193. (5) Kaiser, C. M.; Chang, H. C.; Agashe, V. R.; Lakshmipathy, S. K.; Etchells, S. A.; Hayer-Hartl, M.; Hartl, F. U.; Barral, J. M. Nature 2006, 444, 455– 460. (6) Evans, M. S.; Clark, T. F.; Clark, P. L. Protein Pept. Lett. 2005, 12, 189– 195. (7) Evans, M. S.; Sander, I. M.; Clark, P. L. J. Mol. Biol. 2008, 383, 683–692. (8) Kudlicki, W.; Odom, O. W.; Kramer, G.; Hardesty, B. J. Mol. Biol. 1994, 244, 319–331. (9) Kudlicki, W.; Odom, O. W.; Kramer, G.; Hardesty, B. J. Biol. Chem. 1994, 269, 16549–16553. (10) Clark, P. L.; King, J. J. Biol. Chem. 2001, 276, 25411–25420. (11) Clark, P. L.; King, J. Mol. Biol. Cell 2000, 11, 103. (12) Rutkowska, A.; Beerbaum, M.; Rajagopalan, N.; Fiaux, J.; Schmieder, P.; Kramer, G.; Oschkinat, H.; Bukau, B. FEBS Lett. 2009, 583, 2407–2413. (13) Hsu, S. T. D.; Cabrita, L. D.; Fucini, P.; Christodoulou, J.; Dobson, C. M. J. Am. Chem. Soc. 2009, 131, 8366. (14) Woolhead, C. A.; Johnson, A. E.; Bernstein, H. D. Mol. Cell 2006, 22, 587– 598. (15) Woolhead, C. A.; McCormick, P. J.; Johnson, A. E. Cell 2004, 116, 725– 736. (16) Hsu, S. T.; Fucini, P.; Cabrita, L. D.; Launay, H.; Dobson, C. M.; Christodoulou, J. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 16516–16521. (17) Ellis, J. P.; Bakke, C. K.; Kirchdoerfer, R. N.; Jungbauer, L. M.; Cavagnero, S. ACS Chem. Biol. 2008, 3, 555–566. (18) Ellis, J. P.; Culviner, P. H.; Cavagnero, S. Protein Sci. 2009, 18, 2003– 2015. (19) Youngman, E. M.; McDonald, M. E.; Green, R. Annu. Rev. Microbiol. 2008, 62, 353–373. (20) Zaher, H. S.; Green, R. Cell 2009, 136, 746–762. (21) Nelson, D. L.; Cox, M. M. In Lehninger Principles of Biochemistry; W. H. Freeman: New York, 2005; pp 1034-1080. (22) Reid, B. G.; Flynn, G. C. J. Biol. Chem. 1996, 271, 7212–7217. (23) Fedorov, A. N.; Baldwin, T. O. Methods Enzymol. 1998, 290, 1–17.
Analytical Chemistry, Vol. 82, No. 11, June 1, 2010
4637
use of this reagent may not be ideal, especially in studies requiring the generation of native-like ribosome-released proteins. RNase A mediates hydrolysis of RNA’s P-O5′ bonds24 and EDTA chelates Mg2+, thereby destabilizing the ribosome25 and facilitating its cleavage. Therefore, RNase A and EDTA are often used in combination. RNase A is expected to cleave rRNA, mRNA, and tRNA. Upon hydrolysis, the 3′ adenosine of the peptidyl-tRNA stays linked to the released nascent protein,26,27 introducing an extrinsic C-terminal moiety. Sodium hydroxide promotes the release of nascent proteins without modifying the C terminus.28,29 However, as shown here, the high pH required to produce efficient nascent chain release exceeds the physiological range. While prior studies suggest that the above reagents may not be suitable for biophysical applications requiring the generation of folded species, to date there has been no critical comparison of the effectiveness of different ribosome-release agents, particularly in terms of the quality of the newly synthesized protein. This work demonstrates the shortcomings of puromycin, RNase A/EDTA, and sodium hydroxide and details the favorable properties of hydroxylamine, a reagent known in the translation field19 but rarely used to produce newly synthesized proteins. We show that hydroxylamine promises to be an excellent substitute for the above ribosome-release agents in biophysical studies addressing co- and post-translational protein folding. This potent nucleophile, with a pKa in the physiological range (6.0 at 25 °C),30 releases minimally modified, soluble, and well-behaved nascent proteins from an intact ribosome. Unlike puromycin and RNase A/EDTA, hydroxylamine adds only a small nonnative moiety to the nascent chain. Furthermore, unlike sodium hydroxide, it is effective at a physiologically relevant pH. EXPERIMENTAL SECTION Materials. The plasmid carrying the gene for the three-domain E. coli protein flavohemoglobin (Hmp), subcloned into the pET11 vector (EMD Biosciences, Inc.), was a generous gift from Alberto Boffi. The GrpE and DnaJ chaperones (Stressgen) were used without further purification. Custom deoxyoligonucleotides were generated by the DNA Synthesis Facility of the UW-Madison Biotechnology Center and resuspended in RNase/DNase-free water (Mediatech, Inc.). Hydroxylamine (50% in water, Sigma-Aldrich) and EDTA (0.5 M, pH 8.0, Ambion) solutions were used as purchased. Puromycin (Sigma-Aldrich) was resuspended in RNase/DNase-free water and used as a 10 mM stock solution. RNase A was purchased from Sigma-Aldrich. As recommended by the manufacturer, the saltfree lyophilized protein powder was resuspended in 10 mM sodium acetate (pH 5.2) to 10 mg/mL. The solution was then (24) Raines, R. T. Chem. Rev. 1998, 98, 1045–1066. (25) Klein, D. J.; Moore, P. B.; Steitz, T. A. RNA 2004, 10, 1366–1379. (26) Zachau, H. G.; Acs, G.; Lipmann, F. Proc. Natl. Acad. Sci. U.S.A. 1958, 44, 885–889. (27) Hecht, L. I.; Stephenson, M. L.; Zamecnik, P. C. Proc. Natl. Acad. Sci. U.S.A. 1959, 45, 505–518. (28) Marquez, V.; Wilson, D. N.; Tate, W. P.; Triana-Alonso, F.; Nierhaus, K. H. Cell 2004, 118, 45–55. (29) Rutkowska, A.; Mayer, M. P.; Hoffmann, A.; Merz, F.; Zachmann-Brand, B.; Schaffitzel, C.; Ban, N.; Deuerling, E.; Bukau, B. J. Biol. Chem. 2008, 283, 4124–4132. (30) Bissot, T. C.; Parry, R. W.; Campbell, D. H. J. Am. Chem. Soc. 1957, 79, 796–800.
4638
Analytical Chemistry, Vol. 82, No. 11, June 1, 2010
heated at 100 °C for 15 min, cooled to room temperature, and adjusted to pH 7.4 with 0.1 volume of 1 M Tris-HCl to eliminate any DNase activity. Generation and Purification of Fluorescently Labeled RNCs. RNCs bearing the sequence of the N-terminal hemebinding domain of flavohemoglobin (apoHmpH)31 were generated in E. coli cell-free systems by procedures similar to those employed for the production of other nascent proteins.17 The 5′TTCGTTATAGATTTCCGCCTCGCGATTGATAAATACAT-3′ deoxyoligonucleotide (0.15 µg/µL final concentration), complementary to the 3′ end of the gene for the Hmp N-terminal domain, was added to cell-free reactions to generate the desired truncated mRNA. The latter species is generated by deoxyoligonucleotidedirected mRNA cleavage catalyzed by RNase H, an enzyme endogenously present in the cell-free system. The truncated mRNA creates a homogeneous population of stalled RNCs bearing tRNA-linked apoHmpH. The apoHmpH nascent chains were fluorescently labeled with the succinimidyl ester of 4,4-difluoro5,7-dimethyl-4-bora-3a,4a-diaza-s-indacene-3-propionic acid (BODIPY FL, SE, Invitrogen). Site-specific labeling of apoHmpH nascent chains was achieved at the N-terminal methionine by the addition of BODIPY-Met-tRNAMet to the cell-free reaction mixf ture. BODIPY-Met-tRNAMet was generated from BODIPY FL, f 17,32,33 SE, and tRNAMet (Sigma-Aldrich) as described. f 17 RNCs were isolated by ultracentrifugation. Ribosomal pellets were then dissolved in resuspension buffer (10 mM Tris, 10 mM Mg(OAc)2, 60 mM NH4Cl, 0.5 mM EDTA, 1 mM DTT, at pH 7.0) upon agitation on an orbital shaker (250 rpm) for 1 h at 4 °C. The resuspension buffer volume used for each pellet was one-fifth of that of the original cell-free reaction. Following resuspension, ATP (0.5 µM), KCl (100 mM), and the GrpE (0.4 µM) and DnaJ (0.4 µM) chaperones were added to all RNCs. Ad hoc Release of Nascent Proteins from the Ribosome. Resuspended ribosomes containing tRNA-linked nascent polypeptides were treated with hydroxylamine, puromycin, RNase A/EDTA, or sodium hydroxide as described below. The initial pH of each reaction mixture was measured with Hydrion Vivid pH paper (Micro Essential Laboratory, Brooklyn, NY; resolution, ±0.1 pH units). Nascent Chain Release via Hydroxylamine or Sodium Hydroxide. Three concentrations of each reagent in resuspended RNC solution were tested. The hydroxylamine concentrations were 0.5, 1.0, and 1.5 M, yielding pH 7.2, 7.6, and 7.8, respectively. The sodium hydroxide concentrations were 15, 75, and 100 mM, yielding pH 7.8, 9.5, and 10.0, respectively. Each reaction proceeded for 60 min at 37 °C with brief stirring every 20 min. Additional reactions with 1.5 M hydroxylamine (final pH 7.8) were conducted for 90 and 120 min at 37 °C with stirring every 20 min. Ribosome release reactions were quenched by readjusting the solution pH to 6.4-6.8 by addition of acetic acid. The final volume was rendered identical in each reaction by addition of RNase/ DNase-free H2O, as needed. (31) Ilari, A.; Bonamore, A.; Farina, A.; Johnson, K. A.; Boffi, A. J. Biol. Chem. 2002, 277, 23725–23732. (32) Gite, S.; Mamaev, S.; Olejnik, J.; Rothschild, K. Anal. Biochem. 2000, 279, 218–225. (33) Hardesty, B.; Kramer, G.; Ramachandiran, V. Recent Res. Dev. Anal. Biochem. 2001, 1, 39–48.
Nascent Chain Release via Puromycin or RNase A/EDTA. Ribosome release of nascent proteins was performed upon addition of either puromycin or RNase A/EDTA at pH 6.4-6.6. The final concentration of puromycin was 1 mM, and the final concentrations of RNase A and EDTA were 50 µg/mL and 50 mM, respectively. Addition of the reactants produced insignificant pH variations. As above, reactions proceeded for 60 min at 37 °C with brief stirring every 20 min. Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) Analysis of Ribosome-Released Nascent Chains. The release of nascent protein from the ribosome was analyzed by SDS-PAGE in a pH 6.8 Tris · MES buffer system (see the Supporting Information). The fluorescence intensity of the gel bands (peptidyl-tRNA and ribosome-released protein) was measured with the software ImageJ.34 Assay for the Presence of Insoluble Aggregates. Each of the methods analyzed here was tested for the presence of insoluble aggregates arising from the ribosome-released protein, as described.35 Briefly, reaction aliquots (15 µL) were ultracentrifuged for 15 min at 29 000 rpm (Beckman TL-100 bench ultracentrifuge with TLA-120.1 rotor), and supernatants were removed. The pellets were washed (without resuspension) in 40 µL of 10 mM Tris · HCl (pH 6.9) followed by ultracentrifugation and removal of the supernatant. The pellets were then dissolved in 15 µL of resuspension buffer on an orbital shaker (250 rpm) for 60 min at 4 °C. Aliquots from each step were examined by SDS-PAGE followed by ImageJ analysis. Time-Resolved Fluorescence Anisotropy. The relative molecular size of ribosome-released nascent apoHmpH bearing an N-terminal BODIPY fluorophore was examined via fluorescence anisotropy decays monitored as described17 by dynamic fluorescence depolarization in the frequency domain on a Chronos spectrofluorimeter (ISS) equipped with calcite prism polarizers and a laser diode (λex ) 473 nm, λem ) 510 nm, with a 51294 emission channel filter, Newport Corp.-Oriel). The fluorescence lifetime of all samples was measured with vertically polarized excitation and 54.7° polarized emission. Fluorescence lifetime values were included in the anisotropy data fitting. Raw depolarization data were fit with the GLOBALS software package.36 One- and two-component multiexponential depolarization decay models were considered. In all cases, the twocomponent decay model yielded significantly better fits, with a lower reduced χ2 by a factor >10, according to the expression
[ ( )+f e (
r(t)obs ) r(0) fl e-
t τc,l
-
g
t t + τc,l τc,g
)]
(1)
where r(t)obs is the experimentally observed fluorescence anisotropy decay, r(0) is the fundamental anisotropy, equivalent to the zero-time anisotropy, fl and fg are the fractional anisotropy decay amplitudes, and τc,l and τc,g are the rotational correlation times, for the local and global motions, respectively. The parameter of primary interest in this study is τc,g, corresponding (34) Abramoff, M. D.; Magelhaes, P. J.; Ram, S. J. Biophoton. Int. 2004, 11, 36–42. (35) Bakke, C. K.; Jungbauer, L. M.; Cavagnero, S. Protein Expression Purif. 2006, 45, 381–392. (36) Beecham, J. M.; Gratton, E.; Ameloot, M.; Knutson, J. R.; Brand, L. In Topics in Fluorescence Spectroscopy; Lakowicz, J. R., Ed.; Plenum Press: New York, 1991; p 241.
to the rotational correlation time for the global tumbling of the ribosome-released protein. For hydroxylamine-released apoHmpH, τc,l was 0.6 ± 0.1 ns, i.e., significantly shorter than the corresponding τc,g (see Results). Under these conditions, eq 1 is well approximated by
[ ( ) + f e ( )]
r(t)obs ≈ r(0) fl e-
t τc,l
-
g
t τc,l
(2)
At least three independent experiments were performed and average values are reported, with uncertainties expressed as ±1 standard deviation of the mean. In the GLOBALS curve fitting procedure, pre-exponential factors, rotational correlation times, and time-zero anisotropy were allowed to vary. RNA Microchip Electrophoresis. The integrity of the covalent structure of the ribosome was tested by RNA microchip technology for each of the ribosome-release methods. The rRNA of samples containing ribosome-released proteins was analyzed on an Agilent 2100 Bioanalyzer (Agilent Technologies) using the RNA 6000 Nano Assay Kit (see the Supporting Information). RNA chip electropherograms display prominent 23S and 16S rRNA signals for unhydrolyzed prokaryotic ribosomes. The extent of 23S and 16S rRNA signal reduction and the presence of signals corresponding to degradation products provide a qualitative assessment of ribosomal integrity.37 RESULTS This study examines RNCs derived from apoHmpH, a 140residue protein with a typical all-R-helical globin fold.38 Sitespecifically labeled apoHmpH nascent proteins bearing an N-terminal BODIPY fluorophore were used. RNCs were reacted with hydroxylamine, puromycin, RNase A/EDTA, and sodium hydroxide. We used puromycin and RNase A/EDTA concentrations typically employed in the literature.10,17,39 Given that few explicit protocols exist for the release of nascent proteins via sodium hydroxide and hydroxylamine, a range of concentrations was examined. Unless otherwise stated, ribosome-release reactions were carried out for 60 min. Each method was analyzed by gel electrophoresis followed by fluoroimaging to determine release efficiency, by RNA chip microfluidic capillary electrophoresis to examine postreaction ribosomal integrity, by ultracentrifugation to identify insoluble aggregates, and by time-resolved fluorescence anisotropy to identify soluble aggregates. Table 1 summarizes the results. Efficiency of Nascent Chain Release from the Ribosome. As illustrated in Table 1, hydroxylamine is expected to generate a newly synthesized protein with a C-terminal hydroxamic acid. Figure 1 and Table 2 show that the efficiency of nascent protein release by hydroxylamine depends on both reagent concentration (leading to different pH values) and reaction time. For 60 min reactions, increasing the hydroxylamine concentration from 0.5 to 1.0 to 1.5 M leads to higher yields of released protein. Increases in reaction time from 60 up to 120 min at constant hydroxylamine concentration produce only moderate yield enhancements. (37) Schroeder, A.; Mueller, O.; Stocker, S.; Salowsky, R.; Leiber, M.; Gassmann, M.; Lightfoot, S.; Menzel, W.; Granzow, M.; Ragg, T. BMC Mol. Biol. 2006, 7, 3. (38) Eun, Y. J.; Kurt, N.; Sekhar, A.; Cavagnero, S. J. Mol. Biol. 2008, 376, 879–897. (39) Schaffitzel, C.; Ban, N. J. Struct. Biol. 2007, 158, 463–471.
Analytical Chemistry, Vol. 82, No. 11, June 1, 2010
4639
Table 1. Summary of Reaction Conditions and Products
a
60 min reaction. b N6-dimethyladenine. c Adenine.
Figure 1. Comparison of different ribosome-release methods. (A) SDS-PAGE analysis displaying the extent of ribosome-release undergone by RNCs. Bands corresponding to unreleased peptidyltRNA (t) and released peptides (p) are indicated. Ribosome release of nascent chains was induced by NH2OH, puromycin (puro), RNase A/EDTA (R/E), or NaOH. Addition of hydroxylamine consistently leads to nearly quantitative release. The small fractions of unreleased peptidyl-tRNA decrease as the hydroxylamine concentration (and reaction pH) is increased. Puromycin and RNase A/EDTA yield nearly quantitative and quantitative release, respectively. Sodium hydroxideinduced nascent chain release is concentration-dependent. (B) Extent of insoluble aggregation of newly synthesized ribosome-released proteins. Any fluorescently labeled protein in the gel arises from large insoluble aggregates. Neither hydroxylamine nor puromycin gives rise to insoluble aggregates. In contrast, RNase A/EDTA and sodium hydroxide at pH 9.5 and 10 consistently produce detectable aggregates. The aggregates primarily consist of ribosome-released protein; only minor quantities of peptidyl-tRNA are present.
Puromycin and RNase A/EDTA cause nearly complete and complete protein release, respectively. Both reagents induce protein release quickly (within 30 s; unpublished data and refs 40 and 41). Puromycin and RNase A are expected to append 290 and 261 Da C-terminal moieties on released peptides, respectively (40) Brunelle, J. L.; Youngman, E. M.; Sharma, D.; Green, R. RNA 2006, 12, 33–39. (41) Katunin, V. I.; Muth, G. W.; Strobel, S. A.; Wintermeyer, W.; Rodnina, M. V. Mol. Cell 2002, 10, 339–346.
4640
Analytical Chemistry, Vol. 82, No. 11, June 1, 2010
(Table 1). These increases in molecular weight are too moderate to be detected by gel electrophoresis. As expected, nascent protein release by sodium hydroxide is strongly NaOH-concentration-dependent. At pH 7.8, treatment with NaOH releases less protein than hydroxylamine at equivalent reaction time. Sodium hydroxide is not expected to modify the C terminus of the ribosome-released protein (Table 1). Aggregation of Ribosome-Released Nascent Chains. Ultracentrifugation assays (Figure 1) and time-resolved fluorescence anisotropy (Figure 2) were used to test for the presence of large insoluble aggregates and smaller self-associated soluble species, respectively, arising from the ribosome-released protein. Ultracentrifugation. Large insoluble aggregates containing the newly released nascent protein were detected upon mild ultracentrifugation of resuspended RNC solutions. This procedure leads to the precipitation of any insoluble nascent protein present in the medium after ribosome release. Nascent protein release induced by hydroxylamine does not produce any detectable protein aggregates. The C-terminal hydroxamic acid resulting from treatment with hydroxylamine is not expected to significantly modify the solubility of the released protein. Similarly, treatment with puromycin produces no detectable insoluble aggregate. In contrast, nascent proteins generated by treatment with RNase A/EDTA consistently produce insoluble aggregates, as testified by the formation of a white precipitate at early release times (
