Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Dec 1.
Published in final edited form as: Structure. 2020 Aug 6;28(12):1337–1343.e4. doi: 10.1016/j.str.2020.07.009

A Single-Molecule Surface-Based Platform for Detecting the Assembly and Function of the Human RNA Polymerase II Transcription Machinery

Sang Ryul Park 1,2, Jesse Hauver 3,4, Yunxiang Zhang 5, Andrey Revyakin 1,2, Robert A Coleman 2, Robert Tjian 2, Steven Chu 5, Alexandros Pertsinidis 3,6
PMCID: PMC7710921  NIHMSID: NIHMS1616983  PMID: 32763141

SUMMARY

Single-molecule detection and manipulation is a powerful tool for unraveling dynamic biological processes. Unfortunately, success in such experiments is often challenged by tethering the bio-molecule(s) of interest to a biocompatible surface. Here we describe a robust surface passivation method by dense polymer-brush grafting, based on optimized polyethylene-glycol (PEG) deposition conditions, exactly at the lower critical point of an aqueous biphasic PEG-salt system. The increased biocompatibility achieved, compared to PEG deposition in sub-optimal conditions away from the critical point, allowed us to successfully detect the assembly and function of a large macro-molecular machine, a fluorescent-labeled multi-subunit, human RNA Polymerase II Transcription Pre-Initiation Complex, on single, promoter-containing, surface-immobilized DNA molecules. This platform will enable probing the complex biochemistry and dynamics of large, multi-subunit macromolecular assemblies, such as during the initiation of human RNA Pol II transcription, at the single-molecule level.

Graphical Abstract

graphic file with name nihms-1616983-f0001.jpg

Park et al. develop optimized surface preparation procedures for single-molecule experiments. The increased biocompatibility achieved enables visualizing the assembly and function of a large, multi-component molecular machinery, the human RNA Polymerase II transcription pre-initiation complex, on a natural promoter, at the single-molecule level.

INTRODUCTION

Single-molecule techniques represent arguably the ultimate sensitivity limit in biological assays and have proven very powerful in elucidating molecular mechanisms. Surface-based assays, based on tethering individual bio-molecules to a surface, e.g. to the cover glass, enable prolonged observation using fluorescence imaging, while dynamics can be manipulated by fast stopped-flow changes in solution conditions as well as delivery of various ligands. Also, attachment to a cantilever tip or a micro-sphere facilitates mechanical manipulation experiments with AFM or optical/magnetic tweezers. An even broader and diverse range of biological technologies - chromatography, micro-fluidics, protein and DNA micro-arrays, next-generation sequencing platforms – also relies on tethering bio-molecules to a functionalized and biocompatible surface. In all these surface-based experiments, inhibiting non-specific interactions with the surface while maintaining biological activity is essential (Ha et al., 2002; Roy et al., 2008). The earliest surface-based single-molecule experiments (Ha et al., 1999) relied on electrostatic repulsion of nucleic acids from a negatively charged silica-water interface. When proteins were first introduced in surface-tethered single molecule experiments, the gold standard became silica or glass surfaces with grafted (e.g. through an amino-silane interlayer) kDa-sized PEG chains. Such protocols have been rather successful for relatively simple biological systems (e.g. single protein enzymes) and for up to low-nanomolar working concentrations. Requirements for specificity over a larger range of bio-molecular sizes or chemical properties or working at higher concentrations (Elting et al., 2013; Li et al., 2019) indicate the limitations of the conventional, well-solvated linear PEG grafting condition. Given the lack of a robust surface passivation protocol that can be applied to a variety of biological systems, experimenters have often relied on additional (fortuitous) treatments e.g. chemically capping un-reacted amines in the underlying silane layer [A.P. unpublished observations], blocking putative “sticky” sites by physisorption of small “inert” nucleic acids and/or proteins (Blanchard et al., 2004) or exploring alternative surface chemistries such as self-assembled surfactants (Hua et al., 2014) or lipid bilayers (Roy et al., 2008), to further enhance the repellent properties of surfaces used in single-molecule experiments.

The physics of surface grafted polymers and the mechanism of protein repulsion by PEG have attracted considerable interest. Self-assembled-monolayers of short thiolated n=3–7 monomer PEG on gold reveal that entropic forces drive proteins off the surface (Prime and Whitesides, 1991, 1993). The adsorption mechanism of proteins on semiconductors or glass covered with higher MW PEG (2000–25000, n~100) involves competing contributions of steric repulsion, van der Walls attraction, and hydrophobic effects (Jeon et al., 1991). A general paradigm emerging is that higher grafting density (and to some extend longer chain length) results in more efficient repulsion (Sofia et al., 1998). For deposition of individual entropic chains from a good solvent however the density of grafting sites does not dictate the maximum grafted polymer density. Rather, inter-chain steric repulsion results in grafting density of ~Rg−2 (Rg: PEG chain radius of gyration), with each grafted chain being in a “mushroom” configuration (Degennes, 1980). Increasing the PEG density by using branched rather than linear molecules, achieved increased performance in single-molecule protein folding-unfolding (Groll et al., 2004) and DNA-nucleosome interactions studies (Koopmans et al., 2008) (although each system has its own peculiarities; earlier protein folding experiments (Zhuang et al., 2000) had utilized less elaborate surface passivation schemes). Alternative approaches can take advantage of the limited solubility of PEG in solutions of chaotropic salts (Ananthapadmanabhan and Goddard, 1987) above a critical point in temperature (Emoto et al., 1996) and/or salt (Kingshott et al., 2002a; Kingshott et al., 2002b). The marginally-solvated PEG chains presumably occupy smaller volume at such conditions and thus higher grafting densities can be achieved, resulting in a brush-like PEG layer (Emoto et al., 1996).

Notably, although increased protein repulsion was demonstrated for surface deposition of marginally solvated PEG vs. well-solvated PEG, above and below the solubility limit respectively, ultra-sensitive probing by secondary ion mass spectrometry (Kingshott et al., 2002a) could still detect very low levels of adsorbed small proteins (estimated at <10 ng/cm2 or ~4,000 adsorbed molecules/μm2 for 100 μg/mL 14kDa lysozyme). Assuming these results can extrapolate to 100- to 1000-fold higher dilution, in reconstituted in vitro transcription reaction conditions (ñg/μL or ~10 nM of protein factors) one would expect non-specific binding of 4–40 molecules/μm2. Interestingly, no adsorption of a larger protein (320 kDa fibronectin at 10 μg/mL) could be detected, leading the authors to hypothesize the existence of microscopic defects and stressing the need for a nanometer-scale characterization tool in order to optimize the deposition protocols and create a truly inert surface. Thus, although very promising, due to the limitations in sensitivity and resolution of typical characterization tools (ng/cm2 and ~100 μm for mass- and photo-electron spectroscopy (Wagner et al., 2002)) , it is impossible to confirm whether these earlier results would have applied to the 2 orders-of-magnitude more stringent conditions required in typical single-molecule experiments (0.01–0.1 adsorbed molecules/μm2 at up to 20 nM of fluorescent transcription factors for the optimized surfaces we report here). An essential point is that surface adsorption measurements using single-molecule fluorescence detection are at least 105 to 106 times more sensitive than mass- and photo-electron spectroscopies. Furthermore, additional metrics, essential for single-molecule functional assays, such as accessibility and biological activity of molecules tethered on such surfaces were not validated. Finally, the exact location of the deposition conditions in the phase diagram as well as the requirements to achieve robust dense PEG coverings for demanding single-molecule applications were not thoroughly characterized.

Here, we explore the properties of PEG surfaces prepared at different regimes of the phase diagram of biphasic PEG-K2SO4 systems. We discover that PEG coatings grafted at conditions just above the critical (cloud) point exhibit significantly improved resistance to non-specific protein adsorption, compared to PEG grafted below and well above the critical point. We further show that conditions (Emoto et al., 1996; Kingshott et al., 2002a; Kingshott et al., 2002b) well into the phase-separation regime produce surfaces that are inhomogeneous on the microscopic scale and not appropriate for single-molecule assays. Interestingly, the optimization by fine-tuning deposition conditions in the phase diagram applies not only to linear PEG but also to branched 4-arm PEG polymers, with 4-arm PEG providing an additional ~5-fold improvement compared to linear PEG. Using these optimized cloud-point grafted PEG systems, we detect the assembly and function of the human RNA Polymerase II transcription system at the single-molecule level, in a surface-based assay, using fluorescent purified factors.

RESULTS

Fine-tuning PEG grafting conditions at the phase boundary of a biphasic system is crucial for optimum surface properties in single-molecule experiments

Our exploration of improving the bio-passivation of surfaces with long-chain polymer molecules is motivated by our work to visualize the dynamics of individual polymer molecules embedded within a dense polymer brush (Degennes, 1980) [S.C. and W.T. Juan, private communication]. This grafting scheme is shown in Figure 1A–B. To achieve robust dense PEG grafting, we find that it is essential to carefully characterize the phase behavior of a bi-phasic PEG-K2SO4 system. The room-temperature phase diagram for 10% w/v solutions of varying M.W. PEG is shown in Figure 1C. We note that the transition point is sensitive to pH, salt concentration, type of PEG, as well as temperature, so accurate determination of the lower solubility-limit of each particular system is essential to achieve reproducible results (see below).

Figure 1. Biphasic PEG-K2SO4 phase diagram and cloud-point PEG grafting scheme.

Figure 1.

Open in a new tab

(A) Steric repulsion between end-grafted PEG coils limits area coverage and leaves some reactive sites un-occupied. (B) Above the transition point, compaction of PEG chains results in higher grafting density. Changing solution conditions to a good solvent, results in swelling of PEG chains resulting in a “brush” like structure. (C) Experimentally determined phase diagram for the PEG- K2SO4 system, in sodium carbonate/bi-carbonate buffers with various pH valus, at 298 K. As the PEG solubility boundary is crossed, the conformations of individual chains are expected to become more compact (shown schematically for the 8 kDa boundary). Below the phase boundary PEG chains adopt an equilibrium random coil configuration. For each PEG M.W. and pH value, the phase boundary was determined by preparing solutions of increasing K2SO4 concentration and determining the point at which the solution changes from clear to turbid. Data show mean± S.D.. See also Figure S1 and Tables S1, S4.

To quantitatively evaluate the performance of PEG surfaces deposited at different conditions, we prepare coverslips coated with 5 kDa PEG chains at varying K2SO4 concentrations. Representative snapshots after 5 minute exposure to 20 nM Alexa647-labelled transcriptional activator Gal4-VP16 are shown in Figure 2A. As the concentration of K2SO4 present in the deposition buffer is increased above the transition (cloud-point), we observe a striking decrease in the amount of non-specific binding. This general feature, showing a >10-fold decrease in non-specific binding at the onset of cloudiness (Figure 2B and Table S1), applies to several other proteins and fluorescent probes we studied: human TBP (TATA-box Binding Protein), TFIIB, TFIIF, or even multi-component complexes (e.g. the 15 sub-unit, 1.2 MDa TFIID complex), labeled with Qdot-565 (Figure 2B). Interestingly, we discover that surfaces with uniform properties over ~mm2 are obtained only very close to the transition point only. Grafting PEG in conditions with higher salt, well above the critical point and into the phase-separation regime, often results at inhomogeneous surfaces, with micron-sized patches showing excellent protein repulsion, enclosed by regions with much higher “stickiness” (e.g. last panel in Figure 2A). These results significantly extend previous bulk investigation of cloud-point PEG grafting (Emoto et al., 1996; Kingshott et al., 2002a; Kingshott et al., 2002b), and show that understanding the phase-behavior of the PEG- K2SO4 system is crucial in order to achieve homogeneous and optimal PEG-surface properties.

Figure 2. Fine-tuning PEG deposition conditions at the cloud-point phase transition point is crucial for optimum single-molecule surface properties.

Figure 2.

Open in a new tab

(A) Non-specific adsorption of 20 nM Alexa647-labelled Gal4-VP16. Linear PEG deposition in a carbonate/bi-carbonate buffer at pH 9.5 with a transition (cloud) point of 0.5 M K2SO4. Note the inhomogeneous surface properties for PEG grafted above the transition point (last two panels, 0.64 M). Scale bars: 1.5 μm. (B) Quantification of non-specific sticking vs. [K2SO4], for Gal4-VP16, TBP and TFIID. Linear PEG deposition in 0.1 M Na2HCO3 pH 9–9.5 (red symbols) or a carbonate/bi-carbonate buffer at pH 9.5 (black symbols). Vertical dotted lines: transition points, 0.4 M and 0.5 M K2SO4 respectively. (C) Performance of 4-arm star-PEG surfaces for TFIIB and TFIIF. The performance of the best linear PEG surfaces at 0.55 M K2SO4 is shown for comparison. In (B, C) symbols and error bars represent mean ± S.D. from 3 independent experiments. See also Figure S2 and Tables S2, S3.

We further perform physical characterization of linear PEG surfaces using X-ray Photoelectron Spectroscopy (XPS). In accordance to the single-molecule non-specific adsorption results, XPS results are also consistent with optimum properties obtained for PEG grafted exactly at the critical point. Surfaces well above the critical point are often heterogeneous, with average properties in-between the no- K2SO4 and optimum K2SO4 conditions (Figure S1, Tables S2–S3).

Similarly to linear PEG, we also discover that optimum surface properties are obtained exactly at the critical point of a multi-arm (“star”) PEG system (Figure 2C). Notably, fine-tuning the deposition conditions in the phase diagram of multi-arm (“star”) PEG can provide an additional 5-fold average improvement over the best cloud-point linear PEG surfaces (Table S4).

Optimized cloud-point PEG surfaces enable single-molecule nanometer co-localization assays of the human Pol II machinery

The reduction in non-specific interactions, as shown by the dramatic decrease in adsorbed proteins for both linear and star-PEG surfaces exactly at the cloud-point, is not the only consideration for surface-based single-molecule assays. Rather, in single-molecule assays of bio-molecular recognition and assembly of multi-component complexes, molecules must be (stereo)-specifically tethered to the surface, must retain their native conformation and activity, and must be accessible for binding of free-molecules from the solution. To evaluate these more stringent requirements, we prepared “cloudy” point PEG coverslips with 1:5 ratio of biotinylated PEG. After streptavidin coverage, dsDNA molecules, end-modified with biotin and Cy3, were attached to the surface through highly specific biotin-streptavidin interactions. The majority of surface tethered DNA molecules appear to retain their native conformation and are accessible for enzymes recognizing particular DNA sequences. Using single-molecule multi-color nanometer co-localization (Pertsinidis et al., 2010) (Figure 3A), we first evaluate binding of Cy5-labeled hTBP on a 300bp Cy3-labelled DNA construct that is immobilized on the surface. In experiments with TATA-box containing DNA, after 30min in situ incubation with 2nM Cy5-hTBP, 22% of surface bound Cy5-hTBP (65/298) is localized within 50 nm (Figure 3B) of Cy3-DNA molecules, while only 1.3% (4/306) co-localization is observed for mutant DNA, indicating efficient and specific recognition of the TATA-box DNA by individual Cy5-hTBP (consistent with our biochemical gel-shift and foot-printing assays determining KD’s for specific vs. non-specific DNA).

Figure 3. Optimized cloud-point PEG surfaces enable single-molecule nanometer co-localization assays with the human Pol II machinery.

Figure 3.

Open in a new tab

(A) Nanometer co-localization of Cy3 and Cy5 attached to the same DNA duplex. Each point is the coordinate of an individual Cy5 molecule mapped to the Cy3 channel and regions of 200 nm × 200 nm around each Cy3 are aligned and plotted. (B) Nanometer co-localization of Cy5-hTBP and TATA-box Cy3-DNA. The coordinates of each Cy5-hTBP spot were mapped to the Cy3 channel and regions of 200 nm × 200nm around each Cy3-DNA are aligned and plotted. (C) Nanometer co-localization of A647-hTFIIA and Cy3-DNA in presence of unlabelled hTBP. The coordinates of each A647-TFIIA spot were mapped to the Cy3 channel and regions of 200 nm × 200 nm around each Cy3-DNA are aligned and plotted. (A-C) Scale bar: 50 nm. (D, E) Snapshots of single Cy3-DNA molecules (left) immobilized on the surface and A647-labelled hTFIIA binding form the solution to the hTBP-TATA complex (right). D: efficient binding of TFIIA in presence of unlabelled hTBP, E: omission of hTBP abolishes TFIIA binding. (D, E) scale bars: 1.5 μm. See also Figure S2.

Individual surface-tethered TATA-DNA/hTBP complexes can be efficiently recognized by the general transcription factor TFIIA (Figure 3C) and can nucleate assembly of higher-order structures. We examine Alexa647-labelled hTFIIA, with and without unlabelled TBP (Figure 3D–E). Whereas 45% (351/775) of TFIIA is co-localized with Cy3 labeled DNA in the presence of 4nM TBP, there are only a few spots detected in the absence of TBP, resulting in 0.7% (7/956) co-localization, consistent with the notion that TBP is an essential component to form a stable complex with TFIIA and DNA.

Optimized cloud-point PEG surfaces enable visualizing the assembly and function of a human PIC on a natural promoter at the single-molecule level.

The nanometer co-localization approach can also dissect the assembly pathway of even higher-order complexes, potentially up to the complete Transcription Pre-Initiation Complex (PIC), comprised of promoter DNA, TFIID/IIA, RNA Polymerase II and the General Transcription factors (GTFs) TFIIB/IIF/IIE/IIH, a total of >40 polypeptides. In the following we demonstrate assembly of a minimal human PIC and accurate transcription initiation and from a natural promoter at the single molecule level. The minimal RNA Polymerase II system serves as a stringent test of the passivation chemistry and the single-molecule detection methodology as such high-resolution co-localization assays are obscured on non-optimal surfaces: at high non-specific binding, spots start to overlap and individual molecules cannot be accurately resolved (Figure S2).

To probe PIC assembly, we use a DNA template that contains the natural Adenovirus Major Late Promoter (AdMLP), followed by a G-less transcription cassette, so that in the presence of ATP, UTP and CTP only, Pol II elongation is stalled at +33nt. The DNA contains a single stranded bubble at positions −14 to −3, to allow transcription without TFIIH/TFIIE (Figure 4A).

Figure 4. Optimized cloud-point PEG surfaces enable visualizing the assembly and function of a human PIC on a natural promoter at the single-molecule level.

Figure 4.

Open in a new tab

(A) Schematic of minimal Pol II initiation system. (B-D) Pair-wise co-localizations: (B) Cy3-anti-CTD Pol II, (C) Cy3-hTFIIF and (D) Cy3-mRNA probe, co-localized with Cy5-AdMLP DNA. (E) Fraction of templates that show mRNA probes localized to within 64nm, from ≥2 independent experiments. Error bars: standard deviation. Panel (D) is obtained on linear cloud-point PEG, while panels (B, C) on 4-arm cloud-point PEG. See also Figures S2, S3 and S4.

To demonstrate assembly of a minimal PIC, TBP, TFIIA, TFIIB, TFIIF and Pol II are incubated with Cy5-AdMLP templates. Probing binding of Pol II with a monoclonal antibody against the Carboxyl-terminal-domain (CTD) of the largest subunit shows positive hits on 15% (29/192) of the templates (Figure 4B) while 0.0–1.6% background co-localization was detected in the absence of Pol II. Cy3-labeled TFIIF shows significant non-specific sticking, limiting the highest working concentration on the best surfaces obtained (4-arm cloudy PEG), to ~10 nM, at which an estimated 3–5% (56–101/1879) of templates show co-localization with Cy3-labeled TFIIF, while <2% co-localization is obtained in the absence of Pol II (Figure 4C and Figure S3).

As a further test of the stringency of the optimized PEG surfaces, we test the transcriptional activity of single PICs assembled on surface-tethered DNA templates. Upon incubation with ATP/CTP/UTP nucleotides, productive initiation from these minimal PICs is detected using a Cy3-labelled hybridization probe (Friedman and Gelles, 2012; Revyakin et al., 2012) complementary to the nascent mRNA (Figure 4A). About 6.2% (283/4551, total from two independent experiments) of the Cy5-AdMLP templates show positive hits for mRNA, as indicated by the number of Cy3 spots that co-localized to within 64 nm from the Cy5 templates (Figure 4D). This value is consistent with the well-known relatively low template utilization of the in vitro Pol II system on natural promoters and the existence of dead-end, non-productive initiation pathways (Murakami et al., 2013).

We performed several control experiments to test the background binding of the mRNA probe, as well as the validate the activity and specificity of the single-molecule Pol II transcription system: omitting nucleotides or ATP, removing individual or all of the general factors, deleting the TATA box or using a DNA template without a bubble (Figure 4E and Figure S4). The negative control experiments show <1% co-localization (~0.5% average for no NTPs/ATP/Pol II and Pol II only controls), originating from mostly random non-specific binding to the surface, without a discernible cluster of mRNA localizations within 64nm of the DNA templates. These results support the notion that the Cy3 probe detects productive full-length mRNA synthesis. Moreover, on non-optimized surfaces (PEG deposited in absence of K2SO4) only 2.6% of templates show co-localized mRNA probes, significantly less than on optimized PEG surfaces (6.2%). The reduced transcription efficiency suggests adverse effects on the Pol II PIC activity due to non-optimal surface properties. Taken together, these results indicate that optimized PEG surfaces can enable successful single-molecule detection of promoter- and factor-specific, productive transcription initiation by the RNA Polymerase II system on a low utilization efficiency natural promoter, at the single molecule level.

DISCUSSION

Here, we describe an optimized surface-based single-molecule assay that can detect the assembly and function of the human RNA Polymerase II system. Importantly, our system enables efficient detection of Pol II initiation from natural promoters, which typically feature template utilization efficiencies well below 10%. For comparison, our previous work (Revyakin et al., 2012) had relied exclusively on the use of the synthetic “super-core” promoter (Juven-Gershon et al., 2006), which features an extremely high template utilization efficiency of close to 40%; false positive rates of a few % precluded robust analysis of low efficiency natural promoters. We also demonstrate visualizing the assembly of human Pol II PICs via co-localization of fluorescent components, including Pol II and TFIIF, whose incorporation into human PICs had not been visualized before with single-molecule fluorescence. Mitigating non-specific surface interactions of TFIIF in particular required the most stringent passivation conditions, achieved by the optimized 4-arm PEG “cloudy” point surfaces. We expect the robust platform for human RNA Polymerase II transcription demonstrated in this paper will form the basis for further single-molecule experiments to eventually elucidate the complex molecular mechanisms of eukaryotic transcription regulation.

Successful mitigation of unwanted interactions of the molecules and the dyes with the surface is essential to not only detect ordered binding events, but also to address the conformational rearrangements and structural intermediates during PIC assembly. The enhanced biocompatibility of these “cloudy” point PEG surfaces lays the groundwork for successfully applying a variety of single-molecule assays, including Fluorescence Resonance Energy Transfer, super-resolved distance measurements along the DNA template, single-DNA-tethered particle tracking assays and single-molecule target-locking STED nanoscopy at elevated concentrations (Li et al., 2019; Wang et al., 2016)(S.R.P., A.P. et al., manuscript in preparation). Mitigating adverse effects due to non-specific surface interactions is increasingly important as the resolution and sensitivity of single-molecule assays is pushed further, e.g. tracking the real-time movement of RNA Polymerase with nanometer accuracy as it is transcribing a surface-tethered DNA (Wang et al., 2016), or using some of the best-performing dyes for STED nanoscopy, such as Atto 647N, to study weak and transient interactions at elevated concentrations (Li et al., 2019). We thus expect the improved surface properties of “cloudy” PEG to further enable using increasingly sophisticated single-molecule assays to studies of the human Pol II system.

The approach we described here to achieve a robust PEG coating of glass coverslips can readily be extended to PEG deposition onto other types of substrate materials, e.g. semiconductors or elastomers. We envision a variety of other ultra-sensitive bio-nanotechnological applications such as analytical chromatography, micro-fluidics, protein/DNA micro-arrays and next-generation sequencing, will benefit from the >10-fold improved resistance against non-specific interactions. The requirement appears to be careful optimization of the PEG deposition conditions ensuring that one works very close to the limit of PEG solubility for a biphasic PEG-salt system in order to achieve homogenous dense PEG coverage over macroscopic areas.

STAR METHODS

Resource Availability

Lead Contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Alexandros Pertsinidis (pertsina@mskcc.org).

Materials Availability

This study did not generate new unique materials.

Data and Code Availability

The datasets supporting the current study are available from the corresponding author upon request. The study did not generate new code.

Experimental Model and Subject Details

Cell culture conditions

All cell culture was performed at 37°C in 5% v/v CO2 atmosphere. HeLa S3 cells were grown in Spinner-MEM (AppliChem A1338,9050 or Sigma, M-4767), supplemented with 10% fetal bovine serum (Gemini Bio, 100–106) or 5% Newborn Calf serum (Gemini Bio, 100–504), 2 mM L-alanyl-L-glutamine (Thermo Fisher Scientific 35050079), 1 × Non-essential Amino Acids Solution (Termo Fisher Scientific 11140050), 1mM Sodium Pyruvate (Thermo Fisher Scientific 11360070) and 100U/mL Penicillin-Streptomycin (Thermo Fisher Scientific, 15140122).

Cell lines

HeLa S3 were obtained from ATCC (ATCC CCL-2.2; human female – no additional cell line authentification was performed).

Microbes

E. coli BL21-Codon-Plus(DE3)-RP/-RIL and BL21(DE3)pLysS cells were cultured at 23°C to 37°C in LB or TB media with appropriate antibiotics.

Method Details

Mapping of the PEG-K2SO4 phase diagram:

Solutions of 10% (w/v) PEG of 3 different molecular weights were prepared in either 0.1 M Na2HCO3 (pH 9–9.5) or in sodium carbonate/bi-carbonate buffers (pH 9.2, 9.5, 10.0 and 10.5) and with increasing K2SO4 concentration. The phase behavior was assessed visually by the appearance (cloudy or clear) of each solution. The critical K2SO4 concentration decreases with increasing pH, while the phase boundary shifts at higher [K2SO4] for shorter PEG chain (decreasing M.W.).

Preparation of PEG-coated glass surfaces:

Glass coverslips or slides (VWR) were flamed first to remove a thin oily film that prevents the coverslips from self-sticking and soaked in piranha solution (sulfuric acid: hydrogen peroxide=1:3) for 30 min to remove any residual organic materials. After thorough rinsing with deionized water, the coverslips are immersed in a 0.5 M KOH solution. The samples were treated with 2% 3-Aminopropyltriethoxysilane (3-APTES, Sigma) in acetone for 15 minutes following by rinsing with deionized water. After cleaning with acetone and water, the coverslips were dried with nitrogen. Then 10% (w/v) amine-reactive PEG (M.W. 5000 Da, M-SPA-PEG, Nektar or MPEG-SVA-5000, Laysan) or 4-arm Star PEG succinimidyl succinate (Mn~10,000Da, 4arm-PEG-SC-10K, Laysan) in 0–0.64 M K2SO4 and 0.1 M Na2HCO3 (or 0.1 M sodium carbonate/bi-carbonate) is applied to coverslips and incubated for 1 hr in a humidified chamber to prevent solution drying-out. For experiments with surface-tethered DNA, amine-reactive biotin-PEG (M.W. 3400 Da, SPA-PEG-biotin, Nektar or Biotin-PEG-SVA, M.W. 5,000, Laysan) is added to the solution in 1:5 biotin-PEG:PEG molar ratio. The coverslips can be kept at 4°C for longer than 6 months without deterioration, after successive rinsing with deionized water.

XPS analysis of PEG-passivated surfaces:

All XPS spectra were taken on a Surface Science Instruments S-probe spectrometer. This instrument has a monochromatized Al Kα X-ray and a low energy electron flood gun for charge neutralization of non-conducting samples. The samples were fastened to the sample holder using double sided tape and run as insulators. X-ray spot size for these acquisitions was approximately 800 μm. Pressure in the analytical chamber during spectral acquisition was less than 5 × 10−9 Torr. Pass energy for survey spectra (to calculate composition) was 150 eV and pass energy for high resolution scans was 50 eV. The take-off angle (the angle between the sample normal and the input axis of the energy analyzer) was ~55° (55° take-off angle ≅ 50 Å sampling depth).

Single-molecule surface assays:

Sample cells were assembled with a PEG-modified coverslip and a slide using double-sided tape (3M) as a spacer. Solutions were introduced using drilled holes on the slide. Imaging was performed in an enzymatic oxygen scavenging system (glucose oxidase/catalase, 1 mM methyl viologen, 1 mM ascorbic acid).

Preparation of labeled proteins:

Recombinant hTBP, hTFIIA, hTBIIF, hTFIIB and Gal4-VP16-were expressed in E. Coli. His-tagged hTBP was purified using Ni-NTA resin. His-tagged hTFIIAαβ and hTFIIAγ subunits were expressed separately, purified using NiNTA resin under denaturing conditions in the presence of 6M GuHCl, refolded at 1:1: ratio by sequential dialysis in 3M and 0M GuHCl and finally assembled hTFIA was purified by ion-exchange chromatography in a MonoQ column. hTFIIF subunits were purified separately (Rap30 by 6M Urea extraction and His-tagged Rap74 by Ni-NTA resin in presence of 4M Urea), mixed together at excess Rap30, refolded by sequential dialysis in 4M, 2M and 0M Urea, and assembled hTFIIF was then purified with Ni-NTA resin. hTFIIB was purified using multiple column chromatography steps: phosphocellulose (Whatman P11), followed by strong cation exchange (POROS HS) and heparin affinity (POROS HE). His-tagged Gal4-VP16 F150C was purified with Ni-NTA resin, followed by size-exclusion chromatography (Superdex 75).. Site-specific labeling of single engineered (hTBP-KKCK, KKCK-hTFIIB, Gal4-VP16 F150C) or native (hTFIIAαβ C339 and Rap74 C130) reactive Cysteine residues on each protein were achieved using Alexa or Cyanine thiol-reactive dyes. Native hTFIID and Pol II were immuno-purified from HeLa cell nuclear extracts. hTFIID was labeled using Qdots (Coleman et al., 2017).

Experimental Design:

Independent experiments were performed with different batches of coverslips, on different days, with reproducible results. Comparisons of surface stickiness for different PEG deposition conditions were performed side-by-side, with the same protein aliquots. For surface transcription assays, experiments and controls were performed side-by-side, on the same coverslip, and individual sample chambers were assigned randomly. Experimentalists were not blind to the experimental parameters during the study. No sample size calculations were performed; sample size was determined to be adequate based on the magnitude and consistency of the measurable effects. No data were excluded.

Quantification and Statistical Analysis

XPS analysis of PEG-passivated surfaces:

The Service Physics Hawk Data Analysis Software was used to determine peak areas, to calculate the elemental compositions from peak areas above a linear background, and to peak fit the high resolution spectra. The binding energy scales of the high-resolution spectra were calibrated by assigning the lowest binding energy C1s high-resolution peak a binding energy of 285.0 eV.

Three spots were analyzed on each sample. Analysis of the samples included survey spectra of all three spots, and high resolution spectra of the C1s peak from one spot. The analysis of the spectra can be found in Tables S2 and S3.

Quantification of non-specific sticking:

We quantified the ability of the various PEG surfaces to prevent non-specific biomolecule adsorption by exposing the surfaces to various solutions of fluorescently-labeled molecules. After incubation, snapshots of non-specifically surface-bound molecules were obtained with an objective-type TIR imaging configuration. The surface stickiness was quantified by directly measuring the density of non-specifically adsorbed fluorescent molecules (number of fluorescent spots per unit area). For each experiment at least 5 separate regions on the coverslip were quantified and each experiment was repeated 3 times. The statistical details of the experiments can be also found in the legend of Figure 2.

Single-molecule multi-color co-localization:

The actively-stabilized two-color imaging set-up has been described elsewhere (Pertsinidis et al., 2010). To register the two colors we imaged short DNA duplexes containing both Cy3 and Cy5. The pairs of Cy3-Cy5 coordinates were used to estimate the coefficients of a bi-linear transformation. Typically N~100 pairs were used to cover the field of view, achieving <10 nm typical registration accuracy over an area of 16 μm in diameter. 0.5 μm biotinylated polystyrene beads were attached on the coverslip and imaged on a separate CCD detector under blue LED illumination. The position of a bead on the second CCD was tracked in real-time and used in a closed feedback loop to actively lock the xyz position of the sample stage. This active stabilization scheme ensured nanometer long-term stability of the microscope setup, as well as reproducible placement of different areas across the coverslip at the same focal plane. To create the scatter plots in Figures 3–4 and Figures S2–S4, the coordinates of individual Cy5 or Alexa 647 molecules were mapped to the Cy3 channel and regions-of-interest around each molecule are aligned and plotted by placing the Cy3 or Cy5 DNA template coordinates at the origin. Typically a few hundreds of DNA templates were analyzed in each experiment from multiple regions on the coverslip and each experiment was repeated at least 2 times. The statistical details of each experiment can be found in the individual Figures and Figure legends.

Supplementary Material

1

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
Mouse monoclonal RNA Polymerase II antibody (8WG16) Neoclone Cat#WP011
Chemicals, Peptides, and Recombinant Proteins
3-Aminopropyltriethoxysilane Sigma-Aldrich Cat#440140 or Cat#A3648
Amine-reactive PEG, 5000 Da Laysan Bio. Cat#MPEG-SVA-5000
4-arm Star PEG succinimidyl succinate (Mn~10000 Da) Laysan Bio. Cat#4arm-PEG-SC-10K
Amine-reactive biotin-PEG, 5000 Da Laysan Bio. Cat#BIOTIN-PEG-SVA-5000
Experimental Models: Cell Lines
HeLa S3 ATCC Cat#CL-2.2
Experimental Models: Microbes
E. coli BL21-Codon-Plus(DE3)-RP Agilent Cat#230255
E. coli BL21-Codon-Plus(DE3)-RIL Agilent Cat#230245
E. coli BL21(DE3)pLysS EMD Millipore Cat#70236
Oligonucleotides
AdMLP Transcription template no bubble, Top strand: 5’- /5BioTEG/GAA GGG GGG CTA TAA AAG GGG GTG GGG GCG CGT TCG TCC TCA CTC TCT TCC TCA TCA CTT TCT CCT ACC TCC CGG GCT GTT GGC TGC −3’ This paper N/A
AdMLP Transcription template −14/−3 bubble, Top strand:
5’- /5BioTEG/GAA GGG GGG CTA TAA AAG GGG GTG GGG CAT GCA AGT AGA TCA CTC TCT TCC TCA TCA CTT TCT CCT ACC TCC CGG GCT GTT GGC TGC −3’		 This paper N/A
AdMLP Transcription template −14/−3 bubble no TATA-box, Top strand:
5’- /5BioTEG/GAA GGG GGG CCT GCG GGC GGG GTG GGG CAT GCA AGT AGA TCA CTC TCT TCC TCA TCA CTT TCT CCT ACC TCC CGG GCT GTT GGC TGC −3’		 This paper N/A
AdMLP Transcription template, Bottom strand:
5’- /5Cy5/GCA GCC AAC AGC CCG GGA GGT AGG AGA AAG TGA TGA GGA AGA GAG TGA GGA CGA ACG CGC CCC CAC CCC CTT TTA TAG CCC CCC TTC –3’		 This paper N/A
AdMLP Transcription template no TATA-box, Bottom strand:
5’- /5Cy5/GCA GCC AAC AGC CCG GGA GGT AGG AGA AAG TGA TGA GGA AGA GAG TGA GGA CGA ACG CGC CCC CAC CCC GCC CGC AGG CCC CCC TTC –3’		 This paper N/A
mRNA detection oligo, 5’Cy3:
5’- /5Cy3/GAT GAG GAA GAG AGT −3’		 This paper N/A
mRNA detection oligo, scrambled, 5’Cy3:
5’- /5Cy3/AAA GGT AAG GAT GGG −3’		 This paper N/A
Recombinant DNA
pET-19b hTBP C262T KKCK This paper N/A
pET-11c KKCK-hTFIIB This paper N/A
pJL2 Gal4-VP16 F150C This paper N/A
pET-11c hTFIIγ (Ozer et al., 1994) N/A
pET-19b hTFIIAαβ (DeJong and Roeder, 1993) N/A
pET-11d hRap30 (Wang et al., 1993) N/A
pET-23d hRap74 (Wang et al., 1993) N/A
Software and Algorithms
IDL 6.4 Harris Geospatial Solutions https://www.harrisgeospatial.com/Software-Technology/IDL
Origin 8.5.0 OriginLab https://www.originlab.com/index.aspx?go=PRODUCTS/Origin
Open in a new tab

DeJong, J., and Roeder, R.G. (1993). A single cDNA, hTFIIA/alpha, encodes both the p35 and p19 subunits of human TFIIA. Genes Dev 7, 2220–2234.

Ozer, J., Moore, P.A., Bolden, A.H., Lee, A., Rosen, C.A., and Lieberman, P.M. (1994). Molecular cloning of the small (gamma) subunit of human TFIIA reveals functions critical for activated transcription. Genes Dev 8, 2324–2335.

Wang, B.Q., Kostrub, C.F., Finkelstein, A., and Burton, Z.F. (1993). Production of human RAP30 and RAP74 in bacterial cells. Protein Expr Purif 4, 207–214.

  • Optimized PEG grafting by fine-tuning in the phase diagram of binary PEG-K2SO4 systems

  • >10-fold reduction in non-specific surface adsorption vs. conventional PEG passivation

  • Single-molecule detection of assembly and function of the human Pol II machinery

  • Nanometer co-localization of fluorescent Pol II and GTFs on surfaced-tethered DNA

ACKNOWLEDGEMENTS

A.P. and S.R.P. thank Carla Inoue for assistance with preparation of transcription factors, M. Vrljic and S. Sivasankar for multiple useful discussions and Wen-Tau Juan for sharing unpublished results on grafting DNA polymer brushes. We also thank Gerry Hammer for XPS analysis of the PEG-passivated surfaces. The surface analysis experiments done at NESAC/BIO were supported by NIBIB grant EB-002027. This work was funded by the National Institutes of Health with a grant to R.T. and S.C. (5P01CA112181-05), by DARPA (W911NF-06-1-0122), NSF (0647161) and NASA (NNX07AK54G) with grants to S.C., and the Louis V. Gerstner, Jr. Young Investigators Fund (A.P.), a National Cancer Institute grant (P30 CA008748) and a National Institutes of Health (NIH) Director’s New Innovator Award (1DP2GM105443-01; A.P.).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

DECLARATION OF INTERESTS

The authors declare no competing interests.

REFERENCES

  1. Ananthapadmanabhan KP, and Goddard ED (1987). Aqueous Biphase Formation in Polyethylene Oxide - Inorganic Salt Systems. Langmuir 3, 25–31. [Google Scholar]
  2. Blanchard SC, Gonzalez RL, Kim HD, Chu S, and Puglisi JD (2004). tRNA selection and kinetic proofreading in translation. Nat Struct Mol Biol 11, 1008–1014. [DOI] [PubMed] [Google Scholar]
  3. Coleman RA, Qiao Z, Singh SK, Peng CS, Cianfrocco M, Zhang Z, Piasecka A, Aldeborgh H, Basishvili G, and Liu WL (2017). p53 Dynamically Directs TFIID Assembly on Target Gene Promoters. Mol Cell Biol 37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Degennes PG (1980). Conformations of Polymers Attached to an Interface. Macromolecules 13, 1069–1075. [Google Scholar]
  5. Elting MW, Leslie SR, Churchman LS, Korlach J, McFaul CM, Leith JS, Levene MJ, Cohen AE, and Spudich JA (2013). Single-molecule fluorescence imaging of processive myosin with enhanced background suppression using linear zero-mode waveguides (ZMWs) and convex lens induced confinement (CLIC). Opt Express 21, 1189–1202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Emoto K, Harris JM, and VanAlstine JM (1996). Grafting poly(ethylene glycol) epoxide to amino-derivatized quartz: Effect of temperature and pH on grafting density. Analytical Chemistry 68, 3751–3757. [DOI] [PubMed] [Google Scholar]
  7. Friedman LJ, and Gelles J (2012). Mechanism of Transcription Initiation at an Activator-Dependent Promoter Defined by Single-Molecule Observation. Cell 148, 679–689. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Groll J, Amirgoulova EV, Ameringer T, Heyes CD, Rocker C, Nienhaus GU, and Moller M (2004). Biofunctionalized, ultrathin coatings of cross-linked star-shaped poly(ethylene oxide) allow reversible folding of immobilized proteins. Journal of the American Chemical Society 126, 4234–4239. [DOI] [PubMed] [Google Scholar]
  9. Ha T, Rasnik I, Cheng W, Babcock HP, Gauss GH, Lohman TM, and Chu S (2002). Initiation and re-initiation of DNA unwinding by the Escherichia coli Rep helicase. Nature 419, 638–641. [DOI] [PubMed] [Google Scholar]
  10. Ha T, Ting AY, Liang J, Caldwell WB, Deniz AA, Chemla DS, Schultz PG, and Weiss S (1999). Single-molecule fluorescence spectroscopy of enzyme conformational dynamics and cleavage mechanism. Proc Natl Acad Sci U S A 96, 893–898. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hua B, Han KY, Zhou R, Kim H, Shi X, Abeysirigunawardena SC, Jain A, Singh D, Aggarwal V, Woodson SA, et al. (2014). An improved surface passivation method for single-molecule studies. Nat Methods 11, 1233–1236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Jeon SI, Lee JH, Andrade JD, and Degennes PG (1991). Protein Surface Interactions in the Presence of Polyethylene Oxide.1. Simplified Theory. J Colloid Interf Sci 142, 149–158. [Google Scholar]
  13. Juven-Gershon T, Cheng S, and Kadonaga JT (2006). Rational design of a super core promoter that enhances gene expression. Nat Methods 3, 917–922. [DOI] [PubMed] [Google Scholar]
  14. Kingshott P, McArthur S, Thissen H, Castner DG, and Griesser HJ (2002a). Ultrasensitive probing of the protein resistance of PEG surfaces by secondary ion mass spectrometry. Biomaterials 23, 4775–4785. [DOI] [PubMed] [Google Scholar]
  15. Kingshott P, Thissen H, and Griesser HJ (2002b). Effects of cloud-point grafting, chain length, and density of PEG layers on competitive adsorption of ocular proteins. Biomaterials 23, 2043–2056. [DOI] [PubMed] [Google Scholar]
  16. Koopmans WJA, Schmidt T, and van Noort J (2008). Nucleosome immobilization microscopy. Chemphyschem 9, 2002–2009. [DOI] [PubMed] [Google Scholar]
  17. Li J, Dong A, Saydaminova K, Chang H, Wang G, Ochiai H, Yamamoto T, and Pertsinidis A (2019). Single-Molecule Nanoscopy Elucidates RNA Polymerase II Transcription at Single Genes in Live Cells. Cell. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Murakami K, Calero G, Brown CR, Liu X, Davis RE, Boeger H, and Kornberg RD (2013). Formation and Fate of a Complete 31-Protein RNA Polymerase II Transcription Preinitiation Complex. Journal of Biological Chemistry 288, 6325–6332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Pertsinidis A, Zhang YX, and Chu S (2010). Subnanometre single-molecule localization, registration and distance measurements. Nature 466, 647–U611. [DOI] [PubMed] [Google Scholar]
  20. Prime KL, and Whitesides GM (1991). Self-assembled organic monolayers: model systems for studying adsorption of proteins at surfaces. Science 252, 1164–1167. [DOI] [PubMed] [Google Scholar]
  21. Prime KL, and Whitesides GM (1993). Adsorption of Proteins onto Surfaces Containing End-Attached Oligo(Ethylene Oxide) - a Model System Using Self-Assembled Monolayers. Journal of the American Chemical Society 115, 10714–10721. [Google Scholar]
  22. Revyakin A, Zhang Z, Coleman RA, Li Y, Inouye C, Lucas JK, Park SR, Chu S, and Tjian R (2012). Transcription initiation by human RNA polymerase II visualized at single-molecule resolution. Genes Dev 26, 1691–1702. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Roy R, Hohng S, and Ha T (2008). A practical guide to single-molecule FRET. Nat Methods 5, 507–516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Sofia SJ, Premnath V, and Merrill EW (1998). Poly(ethylene oxide) grafted to silicon surfaces: Grafting density and protein adsorption. Macromolecules 31, 5059–5070. [DOI] [PubMed] [Google Scholar]
  25. Wagner MS, McArthur SL, Shen MC, Horbett TA, and Castner DG (2002). Limits of detection for time of flight secondary ion mass spectrometry (ToF-SIMS) and X-ray photoelectron spectroscopy (XPS): detection of low amounts of adsorbed protein. J Biomat Sci-Polym E 13, 407–428. [DOI] [PubMed] [Google Scholar]
  26. Wang G, Hauver J, Thomas Z, Darst SA, and Pertsinidis A (2016). Single-Molecule Real-Time 3D Imaging of the Transcription Cycle by Modulation Interferometry. Cell 167, 1839–1852 e1821. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Zhuang XW, Ha T, Kim HD, Centner T, Labeit S, and Chu S (2000). Fluorescence quenching: A tool for single-molecule protein-folding study. P Natl Acad Sci USA 97, 14241–14244. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

1
NIHMS1616983-supplement-1.pdf (827.4KB, pdf)

Data Availability Statement

The datasets supporting the current study are available from the corresponding author upon request. The study did not generate new code.

RESOURCES