Skip to main content
NCBI home page
Protein Science : A Publication of the Protein Society logoLink to Protein Science : A Publication of the Protein Society
. 2014 Dec 30;24(3):310–318. doi: 10.1002/pro.2615

Specific ion effects on macromolecular interactions in Escherichia coli extracts

Ciara Kyne 1,*, Brian Ruhle 2, Virginie W Gautier 3, Peter B Crowley 1,*
PMCID: PMC4353357  PMID: 25492389

Abstract

Protein characterization in situ remains a major challenge for protein science. Here, the interactions of ΔTat-GB1 in Escherichia coli cell extracts were investigated by NMR spectroscopy and size exclusion chromatography (SEC). ΔTat-GB1 was found to participate in high molecular weight complexes that remain intact at physiologically-relevant ionic strength. This observation helps to explain why ΔTat-GB1 was not detected by in-cell NMR spectroscopy. Extracts pre-treated with RNase A had a different SEC elution profile indicating that ΔTat-GB1 predominantly interacted with RNA. The roles of biological and laboratory ions in mediating macromolecular interactions were studied. Interestingly, the interactions of ΔTat-GB1 could be disrupted by biologically-relevant multivalent ions. The most effective shielding of interactions occurred in Mg2+-containing buffers. Moreover, a combination of RNA digestion and Mg2+ greatly enhanced the NMR detection of ΔTat-GB1 in cell extracts.

Keywords: arginine, cell extracts, in-cell NMR, protein interactions, size exclusion chromatography

Introduction

As early as 1930, it was surmised that biomolecules assemble into “organized networks” that form “a three dimensional mosaic extending throughout the cell.”1 Despite such prescience, the nature of intracellular macromolecular organization remains poorly understood.2–10 This knowledge gap is a result of the reductionist approach of investigating pure proteins under dilute, artificial conditions.7,8 While the study of pure proteins remains important it is now appreciated that the biophysical properties of proteins inside cells are modulated by the supercrowded, heterogeneous cytoplasm.4–6,9,10 Though daunting, the need for intracellular protein characterization is apparent. Over the past decade or so, in-cell NMR spectroscopy has emerged to address this challenge.5,11–26 Interestingly, in-cell NMR has confirmed the presence of abundant non-specific interactions inside cells, which can hamper the detection of proteins by 1H-15N HSQC and related experiments.5,13,19–26

Size exclusion chromatography (SEC) of whole cell extracts provides a simple means to study macromolecular interactions under native-like conditions.18,27 Importantly, the elution buffer can be selected to test the effects of biologically-relevant ions, ionic strength and pH. Recently, SEC was used to characterize the interactions of cytochrome c (12.8 kDa) and GB128 (6.2 kDa) in concentrated Escherichia coli extracts.18 GB1 eluted from the SEC column at a volume corresponding to the pure protein, indicating that GB1 does not interact with E. coli cytosolic macromolecules. Under the same conditions, cytochrome c eluted with an apparent molecular weight >150 kDa indicating its participation in high molecular weight complexes. When the SEC buffer contained 200 mM NaCl, cytochrome c eluted at its expected position suggesting that charge-charge interactions drive the complexation process. Similarly, SEC revealed that point and multiple charge-inverted mutants (Arg/Lys to Glu) of cytochrome c eluted with lower apparent molecular weights (i.e., reduced complexation) than the wild type. This difference in protein “stickiness” tallied with observations by in-cell NMR spectroscopy. While in-cell 1H-15N HSQC spectra can be obtained on GB1,13,14,16,18,19 cytochrome c and its mutants were undetectable.18 However, the triple cytochrome c mutant was detectable by NMR in E. coli extracts. Taken together these data show that extensive complexation rendered cytochrome c undetectable by in-cell NMR.18 Moreover, the utility of combining SEC and NMR for dissecting the interaction mechanisms of proteins in cell extracts was demonstrated.

Here, we have investigated the influence of the nuclear localization signal (NLS) of HIV-1 Tat29 on protein interactions in E. coli cell extracts. An 11 residue Arg-rich sequence based on the Tat NLS was fused to the N-terminus of GB1 to form ΔTat-GB1. Despite high expression levels it was not possible to obtain a 1H-15N HSQC spectrum of ΔTat-GB1 either in cells or in untreated cell extracts. SEC studies revealed that ΔTat-GB1 interacted extensively with E. coli macromolecules. To determine the nature of these interactions different SEC buffers were employed. Combinations of 16 ions were tested at the physiologically-relevant ionic strength of 100 mM.30–33 Previously reported differences in the “shielding” properties of physiological and laboratory ions18,32,34,35 prompted us to investigate both types and we have analyzed our results in light of the Collins model.35–37 Hofmeister effects, which occur at ionic strengths >200 mM, were not relevant to the present study.35,38 The digestion of nucleic acids revealed that ΔTat-GB1 was predominantly bound to RNA and these interactions could be further disrupted by Mg2+ or succinate (Succ−) at an ionic strength of 100 mM (≈35 mM MgCl2 or Na2Succ). Interestingly, ΔTat-GB1 could be detected by NMR in extracts that were treated with RNase A and 10 mM MgCl2. This supports recent evidence that the NMR detection of proteins in cells is hampered by non-specific interactions5,12,18–26 akin to those expected to promote intracellular organization.7–10,39 Herein, cell extract NMR is demonstrated as a viable alternative to in-cell methods, especially for the investigation of macromolecular interactions.

Results

ΔTat-GB1 is undetectable by in-cell NMR

Figure 1(A) shows the in-cell 1H-15N HSQC spectrum of over-expressed ΔTat-GB1 in the E. coli cytosol. The cell suspension contained ∼1 mM ΔTat-GB1, which is ∼50-fold greater than the detection limit. Unlike GB1,13,14,16,18,19 ΔTat-GB1 was not detected in cells or cell extracts even with sensitivity enhanced measurements40 on a 600 MHz spectrometer equipped with a cold probe. The peaks observed in the in-cell spectrum for ΔTat-GB1 are due to mobile side chain amides and/or 15N- metabolites.10 Purified ΔTat-GB1 yielded a high quality spectrum similar to that of GB1 [Fig. 1(B)]. We, and others, have shown the deleterious consequences of high molecular weight complex formation for in-cell NMR.5,12,18–26 The lack of an in-cell spectrum for ΔTat-GB1 suggests that it interacts with macromolecules in the E. coli cytosol.

Figure 1.

Figure 1

Open in a new tab

1H-15N HSQC spectra of (A) a suspension of E. coli cells containing over-expressed ΔTat-GB1 and (B) purified ΔTat-GB1.

Interactions of ΔTat-GB1 with cytosolic macromolecules

To investigate the nature of ΔTat-GB1 interactions, SEC was performed on cell extracts containing the over-expressed protein. The total macromolecular concentration of the extracts was comparable to that of the E. coli cytosol, thus preserving the effects of macromolecular crowding on protein interactions.18 Pure ΔTat-GB1, a 7.7 kDa protein, eluted from the Superdex G75 column at a volume of ∼80 mL, in agreement with molecular weight standards. However, the SEC elution profile for ΔTat-GB1 in cell extracts revealed a different pattern.

ΔTat-GB1 interactions with nucleic acids were probed by SEC studies on cell extracts that were digested with (i) RNase A or (ii) DNase I (the cell lysis step involved sonication, which also fragments DNA). When extracts were treated with RNase A, the SEC elution profile revealed ΔTat-GB1 in many fractions with the greatest amount in fractions 65–75 close to where small proteins (20–30 kDa) typically elute (Fig. 2). In contrast, when extracts were treated with DNase I, the majority of ΔTat-GB1 eluted in the void volume (45 mL fraction) corresponding to high molecular weight complexes >150 kDa (Fig. 2, 100 mM NaCl data). It can be concluded that a substantial portion of ΔTat-GB1 binds to RNA and/or ribosomes (the most abundant ribonucleoprotein, containing ∼60% ribosomal RNA) in E. coli. Although this system lacks physiological Tat partners, ΔTat-GB1 interactions with RNA still occurred. This is consistent with Tat NLS binding to the HIV “TAR” RNA.41–43 Notably, the interactions remained intact after cell lysis and exposure to the SEC column, suggesting a relatively high affinity complex. Hereafter, all SEC experiments were performed on cell extracts treated with DNase I only. This reduced the viscosity of the samples (necessary for SEC) without compromising prominent ΔTat-GB1 interactions with RNA.

Figure 2.

Figure 2

Open in a new tab

SEC elution profiles of E. coli cell extracts containing over-expressed ΔTat-GB1. The extracts were treated with RNase A or DNase 1 and the SEC buffers contained 50, 100, or 200 mM NaCl. The gel lanes are labelled; MM: molecular weight marker; CE: cell extract; 45–80: fraction volume (mL). The arrow marks the migration position of ΔTat-GB1.

Effect of ionic strength on ΔTat-GB1 interactions

To test the effects of biologically relevant ionic strengths, including those observed under osmotic stress,4 SEC was performed at 50, 100, and 200 mM NaCl. As the NaCl concentration was increased, the amount of ΔTat-GB1 in the 45 mL fraction decreased and there was a concomitant increase of ΔTat-GB1 in the 50–60 mL fractions. At 200 mM NaCl, fraction 60 contained the largest amount of ΔTat-GB1 (Fig. 2). Although the ΔTat-GB1 interactions with RNA were sensitive to ionic strength the protein was never observed in the 80 mL fraction, where pure ΔTat-GB1 eluted. Thus, even at increased concentrations, NaCl was incapable of completely liberating ΔTat-GB1 from its interaction partners. This observation suggests that the complexes are not simply charge-charge in nature. It is likely that additional non-covalent interactions stabilize complexes containing ΔTat-GB1. This is consistent with the fact that Tat NLS is rich in arginine,29 a uniquely versatile residue in its contribution to protein interactions.43–46 Apart from charge-charge interactions, the guanidinium group is a pentadentate hydrogen bond donor that forms cation-π bonds with aromatic groups including the nucleotide bases in RNA.44,46

Specific ion effects on complexes containing ΔTat-GB1

To identify which ions, if any, disrupt the interactions of ΔTat-GB1 with E. coli RNA, SEC buffers containing different salts were tested. For the majority of the salts investigated, at an ionic strength of 100 mM [KGlu (Fig. 3), Na2HPO4, CH3COONa, Na2SO4, NaF, guanidine hydrochloride (GuanCl), KI, NaSCN (Supporting Information Fig. S1), guanidine thiocyanate (GuanSCN, Supporting Information Fig. S2), (NH4)2SO4, NH4Cl, (data not shown)], the elution profiles were closely similar to the 100 mM NaCl data (Fig. 2). In many profiles the amount of ΔTat-GB1 in fraction 45 was identical, within error, to the cell extract (Figs. 2, 3 and Supporting Information Figs. S1 and S2). Even buffers containing 100 mM Guan+ salts, which might be expected to compete with the arginine-driven interactions, failed to dissociate the complexes (Supporting Information Figs. S1 and S2). Interestingly, disruption did occur at 200 mM GuanSCN (Supporting Information Fig. S2). Notably, at this ionic strength the Hofmeister effect may contribute to the observed decrease in ΔTat-GB1 complexation. Guan+ and SCN− are Hofmeister chaotropes capable of unfolding macromolecules38 and thus, altering their interaction propensities.

Figure 3.

Figure 3

Open in a new tab

Elution profiles for SEC buffers containing potassium glutamate, sodium succinate, or magnesium chloride at an ionic strength of 100 mM. The arrow marks the migration position of ΔTat-GB1.

KGlu is an abundant salt present at concentrations upwards of 100 mM in E coli.4,8,47 In buffers containing 100 (Fig. 3) or 200 mM KGlu (not shown), ΔTat-GB1 eluted in the high molecular weight fractions. The lack of dissociation in the presence of an abundant cellular ion supports the idea that ΔTat-GB1 interacts extensively with E. coli macromolecules inside cells.18 The most effective shielding of ΔTat-GB1 interactions occurred with divalent cations. In Mg2+-containing buffers, ΔTat-GB1 eluted in the 75–85 mL fractions corresponding to the pure protein (Fig. 3). The identity of the protein was confirmed by NMR spectroscopy; the 75–85 mL fractions yielded a 1H-15N HSQC spectrum similar to that in Figure 1(B). A similar result was observed in Ca2+-containing buffers (data not shown). The elution profiles obtained in buffers containing di/tri-carboxylates were also noteworthy for the absence/reduction of ΔTat-GB1 in the 45 mL fraction (See data for Succ−, Fig. 3). Perhaps di/tri-carboxylates are effective at disrupting ΔTat-GB1 interactions because they compete with the backbone phosphates of nucleic acids for the Tat NLS.

Herein, the effects of both physiological (e.g., K+, Mg2+, Glu−, Succ2−) and laboratory ions (e.g., Guan+, Na+, Cl−, F−, I−, SCN−) were tested by SEC. Although common laboratory ions are often used to adjust the ionic strength of buffers in biological studies, recent investigations have shown that they do not accurately mimic the effects of physiological ions.18,32,33,35 The laboratory ions tested by SEC were ineffective at shielding ΔTat-GB1 interactions with E. coli RNA. Although similar results were observed for the major cellular ions K+ and Glu−, the multivalent ions Mg2+, Ca2+, Succ2− and citrate (Cit3−) shielded ΔTat-GB1 interactions. Thus, it is important to consider the effects of numerous (biological) ions in the assessment of macromolecular interactions.

Cell extract NMR studies recapitulate the nature of ΔTat-GB1 interactions

While SEC can reveal the individual mechanisms governing a protein's interaction propensities in extracts, NMR can reveal how these mechanisms interrelate and manifest. Thus, ΔTat-GB1 interactions with nucleic acids and specific ions were also investigated in cell extracts by NMR (Fig. 4). The extracts contained over-expressed ΔTat-GB1 and were untreated, treated with a nuclease (DNase I or RNase A) and/or spiked with a salt (MgCl2, Na2Succ or Na3Cit) at an ionic strength of 100 mM. Careful extract preparation was implemented to avoid sample precipitation (see materials and methods). Owing to the predominance of ΔTat-GB1 interactions with RNA, the effect of RNase A incubation times on ΔTat-GB1 interactions was also assessed by NMR. Spectra were obtained on extracts after incubation with RNase A for 30 and 60 min at 20°C (Supporting Information Fig. S3). The results were closely similar suggesting that the effects of RNase A digestion had culminated by 30 min.

Figure 4.

Figure 4

Open in a new tab

1H-15N HSQC spectra of E. coli cell extracts containing over-expressed ΔTat-GB1. The left column (A–D) shows spectra of extracts that were untreated or spiked with a salt. The right column (E–H) are extracts that were RNase A treated and spiked with a salt. All of the salts were added to an ionic strength of 100 mM.

Figure 4(A) shows the 1H-15N HSQC spectrum for ΔTat-GB1 in an untreated cell extract. Compared to the in-cell data [Fig. 1(A)], a marked improvement in spectral quality was evident in the extract. However, the spectrum did not resemble that of purified ΔTat-GB1 [Fig. 1(B)] because many amide resonances were broadened beyond detection. The spectra of extracts after RNA digestion or spiking with MgCl2, Na2Succ or Na3Cit [Fig. 4(B–E)] were comparable to the spectrum from the untreated extract. This indicates that ΔTat-GB1 interacts with RNA despite digestion or the addition of “shielding” ions. Spectra were also obtained on extracts treated with RNase A and MgCl2, Na2Succ or Na3Cit [Fig. 4(F–H)]. The presence of MgCl2 resulted in a dramatic improvement in spectral quality, yielding a HSQC spectrum similar to purified ΔTat-GB1 [Fig. 1(B)]. Importantly, 10 mM MgCl2 was sufficient to effect this substantial improvement in the spectrum (Supporting Information Fig. S4). The carboxylate anions Cit−, and to a lesser extent Succ−, also gave rise to spectral improvements with the appearance of additional ΔTat-GB1 resonances and narrower line widths. Together, these data confirm that ΔTat-GB1 interactions with RNA (and the formation of high molecular weight complexes) render it undetectable by in-cell NMR.

Discussion

The interaction propensity of ΔTat-GB1 in extracts

The grand challenge of protein science is to decipher how macromolecular interactions drive subcellular architectures and biological processes. Such knowledge relies on in vivo studies. However, semi-reductionist approaches are also helpful to determine the physicochemical basis of macromolecular interactions in biological milieus.14,18–23,25,26,32,39,48,49 Herein, SEC and NMR were used to investigate the interaction behavior of ΔTat-GB1 in E. coli extracts. Importantly, ΔTat-GB1 and E. coli RNA are not natural partners. This facilitated the assessment of pervasive, non-specific interactions involving this model RNA-binding motif41,50 in a biological environment. For example, the stickiness of the Tat NLS was demonstrated in that its interactions were dissociated in the presence of Mg2+, which interacts strongly with biological oxyanions (i.e., phosphates and carboxylates).8,36,37 This emphasizes the importance of Arg-oxyanion salt bridges in driving ΔTat-GB1 interactions and suggests that Arg-phosphate salt bridges drive complexation between ΔTat-GB1 and RNA in E. coli. Together with Mg2+ addition, RNA digestion was necessary for the NMR detection of ΔTat-GB1 in cell extracts. This supports the idea that multiple RNA phosphates neutralize the disordered, poly-cationic Tat NLS.51 Notably, this binding mode differs from that reported for the specific Tat-TAR complex.41,42,52 Although Arg/Lys-phosphate contacts have been linked with “non-specific” protein-RNA binding,43,46 recent evidence shows that some RNA-binding proteins interact at sites lacking defined recognition elements.53

Preferential ion pairing in cell extracts

To identify whether the specific ion effects are corroborated by established ion-pairing rules at physiological ionic strengths, our findings are assessed in light of the Collins model.35–37 Hofmeister effects (which occur at ionic strengths >200 mM) are not relevant to the data presented here and will not be discussed.35,38 Collins' “Law of matching water affinities” classifies ions as kosmotropes (small, charge dense and strongly hydrated) and chaotropes (larger, less charge dense and weakly hydrated by rapidly exchanging water molecules). Preferential interactions are observed between species of equivalent hydration and opposite charge. Kosmotrope-kosmotrope interactions are more favorable than kosmotrope-water interactions due to the high charge density and small ionic radii of the kosmotropic pair. Chaotropes interact weakly with water and so, preferentially pair, allowing for the formation of more energetically favorable water-water interactions. Kosmotrope-chaotrope pairs are unfavorable compared to their hydrated ionic counterparts and therefore tend not to interact.

As per the Collins model, the guanidinium groups in the Arg-rich Tat NLS are chaotropic (weakly hydrated) and should preferentially interact with weakly hydrated anions. I− and SCN− might therefore be expected to dissociate complexes containing ΔTat-GB1. However, at 100 mM ionic strength these anions failed to compete with E. coli RNA for binding to ΔTat-GB1 (Supporting Information Figs. S3 and S4). Similarly, efforts to break ΔTat-GB1 interactions using simple physiological anions8,33,47 such asInline graphic, Ac−, (Supporting Information Fig. S1) and Glu− (Fig. 3) were also ineffective. By comparison, di/tricarboxylate anions such as Succ− (Fig. 3) disrupted a substantial proportion of the ΔTat-GB1 complexes. Correspondingly, the majority of E. coli macromolecules contain negatively charged patches that, like polycarboxylate anions, are kosmotropic and multivalent.8,9,47,54 Mg2+, the most strongly hydrated of all biological cations, is a “Collins counterion” compatible with biological phosphate and carboxylate oxyanions.8,37 The disruptive effect of Mg2+ is therefore consistent with the idea that the Tat NLS binds RNA phosphates in E. coli extracts.While the preferential Mg2+-phosphate interaction is consistent with the Collins model, the Arg-oxyanion salt bridge represents a chaotrope-kosmotrope ion pair that is deemed energetically unfavorable.36–38 We surmise that the high effective concentration55 of positive charge (8 Arg and Lys residues) renders the NLS an effective oxyanion (i.e., kosmotrope) binder. Our data are part of a growing repertoire that suggests that preferential ion interactions between macromolecular charge clusters deviate from those predicted by the Collins model, which holds well for simple systems.32 A more advanced model that incorporates not only the ion properties, but also the charge density/distribution at the protein surface is necessary to describe systems of biological complexity.

Broader implications: Macromolecular interactions in cells and extracts

We and others have demonstrated that extensive, non-specific macromolecular interactions persist in cell extracts.14,18–23,25,26,39,49 The physicochemical mechanisms governing macromolecular assembly in cellulo are likely similar to those in extracts. It is therefore instructive to consider our data in light of current models for intracellular assembly, which emphasise the importance of screened electrostatic interactions and macromolecular crowding.7–9,33 For instance, the E. coli cytosol is rife with acidic proteins.54 In addition, the anionic groups of nucleic acid and lipid bilayers generate negatively charged microenvironments throughout the cytoplasm.56 Positively charged proteins and counter-ions partially neutralize the negative charge to form supercrowded “clusters” which are bathed in dilute, electrolyte pools and channels.9 SEC and NMR revealed that the positively charged proteins cytochrome c18 and ΔTat-GB1 interact extensively with negatively charged macromolecules in cell extracts at physiological ionic strengths. This provides fresh evidence for cluster formation between oppositely charged macromolecules in cells and their extracts. Although cytochrome c interactions are predominantly charge-charge in nature, the ΔTat-GB1 interactions are governed by additional forces (e.g. cation-π, chelate effect). The difference in interaction propensities likely arise due to the distinct binding sites of the proteins: the Tat NLS is a disordered, Arg-rich motif while the cytochrome c binding patch is Lys-rich and relatively ordered.

Studies of cell extracts containing ΔTat-GB1 revealed differences between the shielding effects of biologically relevant ions. This disparity coincides with the bipartite function of ions described by the electrochemical model for subcellular organization.8,33 Ions that preferentially interact with macromolecules are expected to reside in a cluster. For example, Mg2+ is likely sequestered by phosphates in nucleic acid clusters.8 This model is supported by the shielding effect of Mg2+ identified by our cell extract studies and the low concentration of free Mg2+ (1 mM) in E. coli.57 Other ions (e.g., K+, Glu−, Ac−,Inline graphic, and Cl−) are expected to reside in electrolyte reservoirs and channels to assist with metabolite transfer by producing electrochemical gradients.8 The inability of these ions to disrupt ΔTat-GB1 interactions invokes their role in gradient formation. The TCA cycle intermediates Succ2− and Cit3− are expected to reside in the electrolyte system for transportation to the TCA machinery. These anions disrupted some ΔTat-GB1 interactions. This may reflect the predicted role for metabolites in the regulation and rearrangement of the “metabolon.”2,58

Materials and Methods

Mutagenesis

The 11 amino acid sequence GRKKRRERRRA (based on the NLS of HIV-1 Tat)29 was introduced at the N-terminus of GB1 between Met1 and Gln2 to yield ΔTat-GB1. The PCR product was digested with NdeI and SacI, cloned into a modified pET3a vector28 and sequence verified to yield pΔTat-GB1.

Protein expression

E. coli BL21(DE3) was transformed with the vector pΔTat-GB1, which contains the ampicillin resistance gene. Cultures were grown as described previously.18 15N-enriched ΔTat-GB1 was produced using a two-step expression procedure.10,18

Cell extract preparation for SEC

Cell extracts were prepared as described18 with the following changes: the cell pellet from 50 mL of culture was resuspended in 1 mL of 20 mM Tris-HCl, 50 mM NaCl, pH 7.0 and frozen overnight. Lysis was completed by thawing the cell suspension, adding a few crystals of DNase I or RNase A, and sonicating.18 The extract was obtained by centrifugation (20,000 rcf for 20 min). The stability of the extracts was temperature dependent. Precipitation occurred after 30 min incubation at 30°C while the extracts remained stable for >6 h at 20°C. Sample stability was optimized by preparing the extracts at 20°C directly prior to use.

Size exclusion chromatography

SEC was performed on an Äkta FPLC at 21°C using an XK 16/70 column (1.6 cm diameter, 65 cm bed height) packed with Superdex 75 (GE Healthcare).18 A continuous flow rate of 1.5 mL min−1 was employed and the column was equilibrated with ∼150 mL buffer prior to sample injection. Cell extract samples (850 µL) were injected onto the column and 1 mL fractions were collected. SEC experiments were performed on cell extracts grown on rich or 15N-labelled medium. Sample elution was monitored at 280 nm. The elution buffer was 20 mM Tris-HCl plus a salt at 100 mM ionic strength (unless otherwise indicated). The pH was adjusted to 7.0 and the buffers were filtered and degassed. To confirm reproducibility, the SEC experiments were performed at least three times for each condition tested.

Cell extract preparation for NMR

The extracts used for the NMR experiments were produced from a 1 L stock culture. Here, the cell pellet was resuspended in 20 mM KH2PO4, 50 mM NaCl, pH 6. Owing to the temperature-dependent stability of the extracts, a fresh sample was used for each HSQC measurement (acquisition time ∼20 min at 30°C) and precipitation was avoided during data acquisition. Nucleic acid degradation was achieved by adding a few crystals of DNase I and/or RNase A to the extracts and incubating for 30 min at 20 (±1)°C prior to spectral acquisition. The extracts were spiked with the following salts: MgCl2, Na2Succ or Na3Cit at an ionic strength of 100 mM. A 1.5–3.0 M stock solutions of the salts were used so that <15 µL volumes were added and the extract was diluted by <3%. Extracts typically had a pH of 6.5 (± 0.05 pH units).

NMR spectroscopy

1H-15N HSQC (watergate) spectra40 were acquired at 30°C with 8 scans and 64 increments on a Varian 600 MHz Spectrometer equipped with a HCN probe or a cold probe. Data processing was performed in Biopack (with linear prediction in the 15N dimension) and NMRPipe.59 The spectra were analyzed using CCPN.60 All spectra were contoured identically (Figs. 1, 4, and Supporting Information Figs. S3 and S4). In-cell 1H-15N HSQC spectra were acquired as described.18 To avoid leakage, careful sample handling was required to prepare the cell slurry.5,14,17,18

Gel electrophoresis

SDS 15 % polyacrylamide gel electrophoresis (SDS-PAGE) with Coomassie blue staining was used to analyze the protein content of the cell extracts and the SEC fractions. Cell extract samples were diluted fourfold while the SEC fractions were concentrated threefold (miVac Duo Centrifugal Concentrator, Genevac) prior to loading.

Acknowledgments

Dr. M. Ismail and Dr. A. Ní Chorrduibh are acknowledged for technical assistance.

Glossary

Abbreviations:

Ac−

acetate

Cit−

citrate

GB1

B1 domain of protein G

Glu−

glutamate

Guan+

guanidinium

HIV

human immunodeficiency virus

NLS

nuclear localization signal

NMR

nuclear magnetic resonance

Ox−

oxalate

SEC

size exclusion chromatography

Succ−

succinate

Tat

transactivator of transcription

ΔTat-GB1

N-terminal fusion protein of GB1 and the NLS of HIV-1 Tat.

Supporting Information

Additional Supporting Information may be found in the online version of this article.

Supplementary Information

pro0024-0310-sd1.tif (2.9MB, tif)

Supplementary Information

pro0024-0310-sd2.tif (1.1MB, tif)

Supplementary Information

pro0024-0310-sd3.tif (1.4MB, tif)

Supplementary Information

pro0024-0310-sd4.tif (264.4KB, tif)

Supplementary Information

pro0024-0310-sd5.doc (26KB, doc)

References

  1. Peters RA. Surface structure in the integration of cell activity. Trans Faraday Soc. 1930;26:797–807. [Google Scholar]
  2. Srere PA. Macromolecular interactions: tracing the roots. Trends Biochem Sci. 2000;25:150–153. doi: 10.1016/s0968-0004(00)01550-4. [DOI] [PubMed] [Google Scholar]
  3. Ellis RJ. Macromolecular crowding: an important but neglected aspect of the intracellular environment. Curr Opin Struct Biol. 2001;11:114–119. doi: 10.1016/s0959-440x(00)00172-x. [DOI] [PubMed] [Google Scholar]
  4. Cayley S, Record TM., Jr Large changes in cytoplasmic biopolymer concentration with osmolality indicate that macromolecular crowding may regulate protein-DNA interactions and growth rate in osmotically stressed Escherichia coli K-12. Biochemistry. 2003;42:12596–12609. doi: 10.1002/jmr.695. [DOI] [PubMed] [Google Scholar]
  5. Li C, Charlton LM, Lakkavaram A, Seagle C, Wang G, Young GB, Macdonald JM, Pielak GJ. Differential dynamical effects of macromolecular crowding on an intrinsically disordered protein and a globular protein: implications for in-cell NMR spectroscopy. J Am Chem Soc. 2008;130:6310–6311. doi: 10.1021/ja801020z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Gierasch LM, Gershenson A. Post-reductionist protein science, or putting Humpty Dumpty back together again. Nat Chem Biol. 2009;5:774–777. doi: 10.1038/nchembio.241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Laue TM, Demeler B. A postreductionist framework for protein biochemistry. Nat Chem Biol. 2011;7:331–334. doi: 10.1038/nchembio.575. [DOI] [PubMed] [Google Scholar]
  8. Spitzer JJ. From water and ions to crowded biomacromolecules: in vivo structuring of a prokaryotic cell. Microbiol Mol Biol Rev. 2011;75:491–506. doi: 10.1128/MMBR.00010-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Spitzer JJ, Poolman B. How crowded is the prokaryotic cytoplasm? FEBS Lett. 2013;587:2094–2098. doi: 10.1016/j.febslet.2013.05.051. [DOI] [PubMed] [Google Scholar]
  10. Serber Z, Dötsch V. In-cell NMR spectroscopy. Biochemistry. 2001;40:14317–14323. doi: 10.1021/bi011751w. [DOI] [PubMed] [Google Scholar]
  11. Serber Z, Straub W, Corsini L, Nomura AM, Shimba N, Craik CS, De Montellano PO, Dötsch V. Methyl groups as probes for proteins and complexes in in-cell NMR experiments. J Am Chem Soc. 2004;126:7119–7125. doi: 10.1021/ja049977k. [DOI] [PubMed] [Google Scholar]
  12. Reckel S, Löhr F, Dötsch V. In-cell NMR spectroscopy. ChemBioChem. 2005;6:1601–1606. doi: 10.1002/cbic.200500076. [DOI] [PubMed] [Google Scholar]
  13. Reardon PN, Spicer LD. Multidimensional NMR spectroscopy for protein characterization and assignment inside cells. J Am Chem Soc. 2005;127:10848–10849. doi: 10.1021/ja053145k. [DOI] [PubMed] [Google Scholar]
  14. Serber Z, Selenko P, Hänsel R, Reckel S, Löhr F, Ferrell JE, Wagner G, Dötsch V. Investigating macromolecules inside cultured and injected cells by in-cell NMR spectroscopy. Nat Protoc. 2007;1:2701–2709. doi: 10.1038/nprot.2006.181. [DOI] [PubMed] [Google Scholar]
  15. Selenko P, Frueh DP, Elsaesser SJ, Haas W, Gygi SP, Wagner G. In situ observation of protein phosphorylation by high-resolution NMR spectroscopy. Nat Struct Mol Biol. 2008;15:321–329. doi: 10.1038/nsmb.1395. [DOI] [PubMed] [Google Scholar]
  16. Inomata K, Ohno A, Tochio H, Isogai S, Tenno T, Nakase I, Takeuchi T, Futaki S, Ito Y, Hiroaki H, Shirakawa M. High-resolution multi-dimensional NMR spectroscopy of proteins in human cells. Nature. 2009;458:106–109. doi: 10.1038/nature07839. [DOI] [PubMed] [Google Scholar]
  17. Pielak GJ, Li C, Miklos AC, Schlesinger AP, Slade KM, Wang GF, Zigoneanu IG. Protein NMR under physiological conditions. Biochemistry. 2009;48:226–234. doi: 10.1021/bi8018948. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Crowley PB, Chow E, Papkovskaia T. Protein interactions in the Escherichia coli cytosol: an impediment to in-cell NMR spectroscopy. ChemBioChem. 2011;12:1043–1048. doi: 10.1002/cbic.201100063. [DOI] [PubMed] [Google Scholar]
  19. Wang Q, Zhuravleva A, Gierasch LM. Exploring weak, transient protein–protein interactions in crowded in vivo environments by in-cell NMR spectroscopy. Biochemistry. 2011;50:9225–9236. doi: 10.1021/bi201287e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Barnes CO, Monteith WB, Pielak GJ. Internal and global protein motion assessed with a fusion construct and in-cell NMR. ChemBioChem. 2011;12:390–391. doi: 10.1002/cbic.201000610. [DOI] [PubMed] [Google Scholar]
  21. Arnesano F, Banci L, Bertini I, Felli IC, Losacco M, Natile G. Probing the interaction of cisplatin with the human copper chaperone Atox1 by solution and in-cell NMR spectroscopy. J Am Chem Soc. 2011;133:18361–18369. doi: 10.1021/ja207346p. [DOI] [PubMed] [Google Scholar]
  22. Luh LM, Hänsel R, Löhr F, Kirchner DK, Krauskopf K, Pitzius S, Schäfer B, Tufar P, Corbeski, Güntert P, Dötsch V. Molecular crowding drives active pin1 into nonspecific complexes with endogenous proteins prior to substrate recognition. J Am Chem Soc. 2013;135:13796–13803. doi: 10.1021/ja405244v. [DOI] [PubMed] [Google Scholar]
  23. Amata I, Maffei M, Igea A, Gay M, Vilaseca AR, Nebreda M, Pons M. Multi-phosphorylation of the intrinsically disordered unique domain of c-Src studied by in-cell and real-time NMR spectroscopy. ChemBioChem. 2013;14:1820–1827. doi: 10.1002/cbic.201300139. [DOI] [PubMed] [Google Scholar]
  24. Banci L, Barbieri L, Bertini I, Luchinat E, Secci E, Zhao Y, Aricescu AR. Atomic-resolution monitoring of protein maturation in live human cells by NMR. Nat Chem Biol. 2013;9:297–299. doi: 10.1038/nchembio.1202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Bertini I, Felli IC, Gonnelli L, Kumar V, Pierattel MVR. 13C Biomolecular NMR spectroscopy in living cells. Angew Chem Int Ed. 2011;50:2339–2341. doi: 10.1002/anie.201006636. [DOI] [PubMed] [Google Scholar]
  26. Wang Y, Li C, Pielak GJ. Effects of proteins on protein diffusion. J Am Chem Soc. 2010;132:9392–9397. doi: 10.1021/ja102296k. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Liu X, Yang W, Gao Q, Regnier F. Toward chromatographic analysis of interacting protein networks. J Chromatog A. 2008;1178:24–32. doi: 10.1016/j.chroma.2007.10.067. [DOI] [PubMed] [Google Scholar]
  28. Lindman S, Xue WF, Szczepankiewicz O, Bauer MC, Nilsson H, Linse S. Salting the charged surface: pH and salt dependence of protein G B1 stability. Biophys J. 2006;90:2911–2921. doi: 10.1529/biophysj.105.071050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Ruben S, Perkins A, Purcell R, Joung K, Sia R, Burghoff R, Haseltine WA, Rosen CA. Structural and functional characterization of human immunodeficiency virus tat protein. J Virol. 1989;63:1–8. doi: 10.1128/jvi.63.1.1-8.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. André I, Kesvatera T, Jönssonz B, Linse S. Salt enhances calmodulin-target interaction. Biophys J. 2006;90:2903–2910. doi: 10.1529/biophysj.105.068718. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Raghuraman H, Ganguly S, Chattopadhyay A. Effect of ionic strength on the organization and dynamics of membrane-bound melittin. Biophys Chem. 2006;124:115–124. doi: 10.1016/j.bpc.2006.06.011. [DOI] [PubMed] [Google Scholar]
  32. Buell AK, Hung P, Salvatella X, Welland ME, Dobson CM, Knowles TP. Electrostatic effects in filamentous protein aggregation. Biophys J. 2013;104:1116–1126. doi: 10.1016/j.bpj.2013.01.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Poolman B, Spitzer JJ, Wood JM. Bacterial osmosensing: roles of membrane structure and electrostatics in lipid–protein and protein–protein interactions. Biochim Biophys Acta. 2004;1666:88–104. doi: 10.1016/j.bbamem.2004.06.013. [DOI] [PubMed] [Google Scholar]
  34. Kontur WS, Capp MW, Gries TJ, Saecker RM, Record MT., Jr Probing DNA binding, DNA opening, and assembly of a downstream clamp/jaw in Escherichia coli RNA polymerase-λPR promoter complexes using salt and the physiological anion glutamate. Biochemistry. 2010;49:4361–4373. doi: 10.1021/bi100092a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Gokarn YR, Fesinmeyer RM, Saluja A, Razinkov V, Chase SF, Laue TM, Brems DN. Effective charge measurements reveal selective and preferential accumulation of anions, but not cations, at the protein surface in dilute salt solutions. Prot Sci. 2011;20:580–587. doi: 10.1002/pro.591. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Collins KD. Sticky ions in biological systems. Proc Natl Acad Sci USA. 1995;92:5553–5557. doi: 10.1073/pnas.92.12.5553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Collins KD. Ion hydration: implications for cellular function, polyelectrolytes, and protein crystallization. Biophys Chem. 2006;119:271–281. doi: 10.1016/j.bpc.2005.08.010. [DOI] [PubMed] [Google Scholar]
  38. Zhang Y, Cremer PS. The inverse and direct Hofmeister series for lysozyme. Proc Natl Acad Sci USA. 2009;106:15249–15253. doi: 10.1073/pnas.0907616106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sarkar M, Smith AE, Pielak GJ. Impact of reconstituted cytosol on protein stability. Proc Natl Acad Sci USA. 2013;110:19342–19347. doi: 10.1073/pnas.1312678110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mori S, Abeygunawardana C, Johnson MO, van Zijl P. Improved sensitivity of HSQC spectra of exchanging protons at short interscan delays using a new fast HSQC (FHSQC) detection scheme that avoids water saturation. J Magn Reson B. 1996;108:94–98. doi: 10.1006/jmrb.1995.1109. [DOI] [PubMed] [Google Scholar]
  41. Davidson A, Patora-Komisarska K, Robinson JA, Varani G. Essential structural requirements for specific recognition of HIV TAR RNA by peptide mimetics of Tat protein. Nucleic Acids Res. 2011;39:248–256. doi: 10.1093/nar/gkq713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Gautier VW, Gu L, O'Donoghue N, Pennington S, Sheehy N, Hall WW. In vitro nuclear interactome of the HIV-1 Tat protein. Retrovirology. 2009;6:47–64. doi: 10.1186/1742-4690-6-47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Austin RJ, Xia T, Ren J, Takahashi TT, Roberts RW. Designed arginine-rich RNA-binding peptides with picomolar affinity. J Am Chem Soc. 2002;124:10966–10967. doi: 10.1021/ja026610b. [DOI] [PubMed] [Google Scholar]
  44. Crowley PB, Golovin A. Cation–π interactions in protein–protein interfaces. Proteins. 2005;59:231–239. doi: 10.1002/prot.20417. [DOI] [PubMed] [Google Scholar]
  45. Moreira IS, Fernandes PA, Ramos MJ. Hot spots—a review of the protein–protein interface determinant amino-acid residues. Proteins. 2007;68:803–812. doi: 10.1002/prot.21396. [DOI] [PubMed] [Google Scholar]
  46. Bahadur RP, Zacharias M, Janin J. Dissecting protein–RNA recognition sites. Nucleic Acids Res. 2008;36:2705–2716. doi: 10.1093/nar/gkn102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Bennett BD, Kimball EH, Gao M, Osterhout R, Van Dien SJ, Rabinowitz JD. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nat Chem Biol. 2009;5:593–599. doi: 10.1038/nchembio.186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Martorell G, Adrover M, Kelly G, Temussi PA, Pastore A. A natural and readily available crowding agent: NMR studies of proteins in hen egg white. Proteins. 2011;79:1408–1415. doi: 10.1002/prot.22967. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Monteith WB, Pielak GJ. Residue level quantification of protein stability in living cells. Proc Natl Acad Sci USA. 2014;111:11335–11340. doi: 10.1073/pnas.1406845111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Ivanyi-Nagy R, Davidovic L, Khandjian EW, Darlix J-L. Disordered RNA chaperone proteins: from functions to disease. Cell Mol Life Sci. 2005;62:1409–1417. doi: 10.1007/s00018-005-5100-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Shojania S, O'Neil JD. HIV-1 Tat is a natively unfolded protein: the solution conformation and dynamics of reduced HIV-1 Tat-(1–72) by NMR spectroscopy. J Biol Chem. 2006;281:8347–8356. doi: 10.1074/jbc.M510748200. [DOI] [PubMed] [Google Scholar]
  52. Ye X, Kumar RA, Patel DJ. Molecular recognition in the bovine immunodeficiency virus Tat peptide-TAR RNA complex. Chem Biol. 1995;2:827–840. doi: 10.1016/1074-5521(95)90089-6. [DOI] [PubMed] [Google Scholar]
  53. Guenther UP, Yandek LE, Niland CN, Campbell FE, Anderson D, Anderson VE, Harris ME, Jankowsky E. Hidden specificity in an apparently nonspecific RNA-binding protein. Nature. 2013;502:385–388. doi: 10.1038/nature12543. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Link AJ, Robison K, Church GM. Escherichia coli. 1997;18:1259–1313. doi: 10.1002/elps.1150180807. Comparing the predicted and observed properties of proteins encoded in the genome of K-12 Electrophoresis. [DOI] [PubMed] [Google Scholar]
  55. Page MI, Jencks WP. Entropic contributions to rate accelerations in enzymic and intramolecular reactions and the chelate effect. Proc Natl Acad Sci USA. 1971;68:1678–1683. doi: 10.1073/pnas.68.8.1678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Theillet F-X, Binolfi A, Frembgen-Kesner T, Hingorani K, Sarkar M, Kyne C, Li C, Crowley PB, Gierasch LM, Pielak GJ, Elcock A, Gershenson A, Selenko P. Physicochemical properties of cells and their effects on intrinsically disordered proteins (IDPs) Chem Rev. 2014;114:6661–6714. doi: 10.1021/cr400695p. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Tyrrell J, McGinnis JL, Weeks KM, Pielak GJ. The cellular environment stabilizes adenine riboswitch RNA structure. Biochemistry. 2013;52:8777–8785. doi: 10.1021/bi401207q. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Winkle BS. Metabolic channeling in plants. Annu Rev Plant Biol. 2004;55:85–107. doi: 10.1146/annurev.arplant.55.031903.141714. [DOI] [PubMed] [Google Scholar]
  59. Delaglio F, Grzesiek S, Vuister GW, Zhu G, Pfeifer J, Bax A. NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J Biomol NMR. 1995;6:277–293. doi: 10.1007/BF00197809. [DOI] [PubMed] [Google Scholar]
  60. Vranken WF, Boucher W, Stevens TJ, Fogh RH, Pajon A, Llinas M, Ulrich EL, Markley JL, Ionides J, Laue ED. The CCPN data model for NMR spectroscopy: development of a software pipeline. Proteins. 2005;59:687–696. doi: 10.1002/prot.20449. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information

pro0024-0310-sd1.tif (2.9MB, tif)

Supplementary Information

pro0024-0310-sd2.tif (1.1MB, tif)

Supplementary Information

pro0024-0310-sd3.tif (1.4MB, tif)

Supplementary Information

pro0024-0310-sd4.tif (264.4KB, tif)

Supplementary Information

pro0024-0310-sd5.doc (26KB, doc)

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of The Protein Society

RESOURCES