Nanopore Whole Transcriptome Analysis and Pathogen Surveillance by a Novel Solid-Phase Catalysis Approach.

The requirement of a large input amount (500 ng) for Nanopore direct RNA-seq presents a major challenge for low input transcriptomic analysis and early pathogen surveillance. The high RNA input requirement is attributed to significant sample loss associated with library preparation using solid-phase reversible immobilization (SPRI) beads. A novel solid-phase catalysis strategy for RNA library preparation to circumvent the need for SPRI bead purification to remove enzymes is reported here. This new approach leverages concurrent processing of non-polyadenylated transcripts with immobilized poly(A) polymerase and T4 DNA ligase, followed by directly loading the prepared library onto a flow cell. Whole transcriptome sequencing, using a human pathogen Listeria monocytogenes as a model, demonstrates this new method displays little sample loss, takes much less time, and generates higher sequencing throughput correlated with reduced nanopore fouling compared to the current library preparation for 500 ng input. Consequently, this approach enables Nanopore low-input direct RNA-seq, improving pathogen detection and transcript identification in a microbial community standard with spike-in transcript controls. Besides, as evident in the bioinformatic analysis, the new method provides accurate RNA consensus with high fidelity and identifies higher numbers of expressed genes for both high and low input RNA amounts.

Type II Restriction of Bacteriophage DNA With 5hmdU-Derived Base Modifications.

To counteract bacterial defense systems, bacteriophages (phages) make extensive base modifications (substitutions) to block endonuclease restriction. Here we evaluated Type II restriction of three thymidine (T or 5-methyldeoxyuridine, 5mdU) modified phage genomes: Pseudomonas phage M6 with 5-(2-aminoethyl)deoxyuridine (5-NedU), Salmonella phage ViI (Vi1) with 5-(2-aminoethoxy)methyldeoxyuridine (5-NeOmdU) and Delftia phage phi W-14 (a.k.a. ΦW-14) with α-putrescinylthymidine (putT). Among >200 commercially available restriction endonucleases (REases) tested, phage M6, ViI, and phi W-14 genomic DNAs (gDNA) show resistance against 48.4, 71.0, and 68.8% of Type II restrictions, respectively. Inspection of the resistant sites indicates the presence of conserved dinucleotide TG or TC (TS, S=C, or G), implicating the specificity of TS sequence as the target that is converted to modified base in the genomes. We also tested a number of DNA methyltransferases (MTases) on these phage DNAs and found some MTases can fully or partially modify the DNA to confer more resistance to cleavage by REases. Phage M6 restriction fragments can be efficiently ligated by T4 DNA ligase. Phi W-14 restriction fragments show apparent reduced rate in E. coli exonuclease III degradation. This work extends previous studies that hypermodified T derived from 5hmdU provides additional resistance to host-encoded restrictions, in parallel to modified cytosines, guanine, and adenine in phage genomes. The results reported here provide a general guidance to use REases to map and clone phage DNA with hypermodified thymidine.

Solid-phase enzyme catalysis of DNA end repair and 3' A-tailing reduces GC-bias in next-generation sequencing of human genomic DNA.

The use of next-generation sequencing (NGS) has been instrumental in advancing biological research and clinical diagnostics. To fully utilize the power of NGS, complete, uniform coverage of the entire genome is required. In this study, we identified the primary sources of bias observed in sequence coverage across AT-rich regions of the human genome with existing amplification-free DNA library preparation methods. We have found evidence that a major source of bias is the inefficient processing of AT-rich DNA in end repair and 3' A-tailing, causing under-representation of extremely AT-rich regions. We have employed immobilized DNA modifying enzymes to catalyze end repair and 3' A-tailing reactions, to notably reduce the GC bias observed with existing library construction methods.

Enhancing Multistep DNA Processing by Solid-Phase Enzyme Catalysis on Polyethylene Glycol Coated Beads.

Covalent immobilization of enzymes on solid supports provides an alternative approach to homogeneous biocatalysis by adding the benefits of simple enzyme removal, improved stability, and adaptability to automation and high-throughput applications. Nevertheless, immobilized (IM) enzymes generally suffer from reduced activity compared to their soluble counterparts. The nature and hydrophobicity of the supporting material surface can introduce enzyme conformational change, spatial confinement, and limited substrate accessibility, all of which will result in loss of the immobilized enzyme activity. In this work, we demonstrate through kinetic studies that flexible polyethylene glycol (PEG) moieties modifying the surface of magnetic beads improve the activity of covalently immobilized DNA replication enzymes. PEG-modified immobilized enzymes were utilized in library construction for Illumina next-generation sequencing (NGS) increasing the read coverage across AT-rich regions.

Single-molecule visualization of a formin-capping protein 'decision complex' at the actin filament barbed end.

Precise control of actin filament length is essential to many cellular processes. Formins processively elongate filaments, whereas capping protein (CP) binds to barbed ends and arrests polymerization. While genetic and biochemical evidence has indicated that these two proteins function antagonistically, the mechanism underlying the antagonism has remained unresolved. Here we use multi-wavelength single-molecule fluorescence microscopy to observe the fully reversible formation of a long-lived 'decision complex' in which a CP dimer and a dimer of the formin mDia1 simultaneously bind the barbed end. Further, mDia1 displaced from the barbed end by CP can randomly slide along the filament and later return to the barbed end to re-form the complex. Quantitative kinetic analysis reveals that the CP-mDia1 antagonism that we observe in vitro occurs through the decision complex. Our observations suggest new molecular mechanisms for the control of actin filament length and for the capture of filament barbed ends in cells.

Probing homodimer formation of epidermal growth factor receptor by selective crosslinking.

Ligand binding promotes conformational rearrangement of the epidermal growth factor receptor (EGFR) leading to receptor autophosphorylation and downstream signaling. However, transient interactions between unstimulated EGFR molecules on the cell surface are not fully understood. In this report, we describe the investigation of homodimer formation of EGFR by means of an SNAP-tag based selective crosslinking approach (S-CROSS). EGFR homodimers were selectively captured in living cells and utilized for analysis of protein receptor interactions on the plasma membrane and ligand-induced activation. We showed that EGFR forms homodimers in unstimulated cells with efficiencies similar to those seen in cells treated with the epidermal growth factor ligand (EGF) supporting the existence of constitutive transient receptor-receptor interactions. EGFR crosslinked homodimers displayed a substantially increase in kinase activation upon ligand stimulation. Interestingly, in unstimulated cells the levels of spontaneous phosphorylation were found to correlate with the yields of the crosslinked homodimers species. In addition, we demonstrated that this crosslinking approach can be applied to interrogate the effect of small molecule inhibitors on receptor dimerization and kinase activity. Our crosslinking assay provides a new tool to dissect ligand-independent dimerization and activation mechanisms of receptor tyrosine kinases, many of which are important anticancer drug targets.

Alternative spliceosome assembly pathways revealed by single-molecule fluorescence microscopy.

Removal of introns from nascent transcripts (pre-mRNAs) by the spliceosome is an essential step in eukaryotic gene expression. Previous studies have suggested that the earliest steps in spliceosome assembly in yeast are highly ordered and the stable recruitment of U1 small nuclear ribonucleoprotein particle (snRNP) to the 5' splice site necessarily precedes recruitment of U2 snRNP to the branch site to form the "prespliceosome." Here, using colocalization single-molecule spectroscopy to follow initial spliceosome assembly on eight different S. cerevisiae pre-mRNAs, we demonstrate that active yeast spliceosomes can form by both U1-first and U2-first pathways. Both assembly pathways yield prespliceosomes functionally equivalent for subsequent U5·U4/U6 tri-snRNP recruitment and for intron excision. Although fractional flux through the two pathways varies on different introns, both are operational on all introns studied. Thus, multiple pathways exist for assembling functional spliceosomes. These observations provide insight into the mechanisms of cross-intron coordination of initial spliceosome assembly.

Three-color single molecule imaging shows WASP detachment from Arp2/3 complex triggers actin filament branch formation.

During cell locomotion and endocytosis, membrane-tethered WASP proteins stimulate actin filament nucleation by the Arp2/3 complex. This process generates highly branched arrays of filaments that grow toward the membrane to which they are tethered, a conflict that seemingly would restrict filament growth. Using three-color single-molecule imaging in vitro we revealed how the dynamic associations of Arp2/3 complex with mother filament and WASP are temporally coordinated with initiation of daughter filament growth. We found that WASP proteins dissociated from filament-bound Arp2/3 complex prior to new filament growth. Further, mutations that accelerated release of WASP from filament-bound Arp2/3 complex proportionally accelerated branch formation. These data suggest that while WASP promotes formation of pre-nucleation complexes, filament growth cannot occur until it is triggered by WASP release. This provides a mechanism by which membrane-bound WASP proteins can stimulate network growth without restraining it. DOI:http://dx.doi.org/10.7554/eLife.01008.001.

The formin Daam1 and fascin directly collaborate to promote filopodia formation.

Filopodia are slender cellular protrusions that dynamically extend and retract to facilitate directional cell migration, pathogen sensing, and cell-cell adhesion. Each filopodium contains a rigid and organized bundle of parallel actin filaments, which are elongated at filopodial tips by formins and Ena/VASP proteins. However, relatively little is known about how the actin filaments in the filopodial shaft are spatially organized to form a bundle with appropriate dimensions and mechanical properties. Here, we report that the mammalian formin Daam1 (Disheveled-associated activator of morphogenesis 1) is a potent actin-bundling protein and localizes all along the filopodial shaft, which differs from other formins that localize specifically to the tips. Silencing of Daam1 led to severe defects in filopodial number, integrity, and architecture, similar to silencing of the bundling protein fascin. This led us to investigate the potential relationship between Daam1 and fascin. Fascin and Daam1 coimmunoprecipitated from cell extracts, and silencing of fascin led to a striking loss of Daam1 localization to filopodial shafts, but not tips. Furthermore, purified fascin bound directly to Daam1, and multicolor single-molecule TIRF imaging revealed that fascin recruited Daam1 to and stabilized Daam1 on actin bundles in vitro. Our results reveal an unanticipated and direct collaboration between Daam1 and fascin in bundling actin, which is required for proper filopodial formation.

Substrates for improved live-cell fluorescence labeling of SNAP-tag.

The SNAP-tag labeling technology provides a simple, robust, and versatile approach to the imaging of fusion proteins for a wide range of experimental applications. Owing to the specific and covalent nature of the labeling reaction, SNAP-tag is well suited for the analysis and quantification of fused target protein using fluorescence microscopy techniques. In this report, we present our most recent findings on the labeling of SNAP-tag fusion proteins both in vitro and in cell culture with SNAP-tag substrates derived from single regioisomers of carboxyrhodamine dyes. Carboxyrhodamines are invaluable fluorescent dyes for biotechnology applications including DNA sequencing, detection on microarrays, and fluorescence in situ hybridization. We found that SNAP-tag reacts preferentially with the 6-positional regioisomer of carboxyrhodamine fluorescent dyes, whereas the 5-regioisomer predominantly contributes to background fluorescence. Our experimental study also indicates that benzylchloropyrimidine (CP) conjugates of 6-carboxyrhodamines exhibit a dramatic increase in the signal-to-noise ratio of fluorescently labeled cellular proteins compared to the benzylguanine (BG) conjugates, presumably due to higher cell permeability. These new SNAP-tag substrates based on pure 6-regioisomers can significantly improve fluorescence labeling in live cells and should become powerful tools for bioimaging applications.

Near-infrared fluorescence imaging of mammalian cells and xenograft tumors with SNAP-tag.

Fluorescence in the near-infrared (NIR) spectral region is suitable for in vivo imaging due to its reduced background and high penetration capability compared to visible fluorescence. SNAP(f) is a fast-labeling variant of SNAP-tag that reacts with a fluorescent dye-conjugated benzylguanine (BG) substrate, leading to covalent attachment of the fluorescent dye to the SNAP(f). This property makes SNAP(f) a valuable tool for fluorescence imaging. The NIR fluorescent substrate BG-800, a conjugate between BG and IRDye 800CW, was synthesized and characterized in this study. HEK293, MDA-MB-231 and SK-OV-3 cells stably expressing SNAP(f)-Beta-2 adrenergic receptor (SNAP(f)-ADRß2) fusion protein were created. The ADRß2 portion of the protein directs the localization of the protein to the cell membrane. The expression of SNAP(f)-ADRß2 in the stable cell lines was confirmed by the reaction between BG-800 substrate and cell lysates. Microscopic examination confirmed that SNAP(f)-ADRß2 was localized on the cell membrane. The signal intensity of the labeled cells was dependent on the BG-800 concentration. In vivo imaging study showed that BG-800 could be used to visualize xenograph tumors expressing SNAP(f)-ADRß2. However, the background signal was relatively high, which may be a reflection of non-specific accumulation of BG-800 in the skin. To address the background issue, quenched substrates that only fluoresce upon reaction with SNAP-tag were synthesized and characterized. Although the fluorescence was successfully quenched, in vivo imaging with the quenched substrate CBG-800-PEG-QC1 failed to visualize the SNAP(f)-ADRß2 expressing tumor, possibly due to the reduced reaction rate. Further improvement is needed to apply this system for in vivo imaging.

Two-color STED microscopy in living cells.

Diffraction-unlimited resolution provided by Stimulated Emission Depletion (STED) microscopy allows for imaging cellular processes in living cells that are not visible by conventional microscopy. However, it has so far not been possible to study dynamic nanoscale interactions because multicolor live cell STED microscopy has yet to be demonstrated and suitable labeling technologies and protocols are lacking. Here we report the first realization of two-color STED imaging in living cells. Using improved SNAP(f) and CLIP(f) technologies to label epidermal growth factor (EGF) and EGF receptor (EGFR), we report resolutions of 78 nm and 82 nm for 22 sequential two-color scans in living cells.

Development of SNAP-tag fluorogenic probes for wash-free fluorescence imaging.

The ability to specifically attach chemical probes to individual proteins represents a powerful approach to the study and manipulation of protein function in living cells. It provides a simple, robust and versatile approach to the imaging of fusion proteins in a wide range of experimental settings. However, a potential drawback of detection using chemical probes is the fluorescence background from unreacted or nonspecifically bound probes. In this report we present the design and application of novel fluorogenic probes for labeling SNAP-tag fusion proteins in living cells. SNAP-tag is an engineered variant of the human repair protein O(6)-alkylguanine-DNA alkyltransferase (hAGT) that covalently reacts with benzylguanine derivatives. Reporter groups attached to the benzyl moiety become covalently attached to the SNAP tag while the guanine acts as a leaving group. Incorporation of a quencher on the guanine group ensures that the benzylguanine probe becomes highly fluorescent only upon labeling of the SNAP-tag protein. We describe the use of intramolecularly quenched probes for wash-free labeling of cell surface-localized epidermal growth factor receptor (EGFR) fused to SNAP-tag and for direct quantification of SNAP-tagged ß-tubulin in cell lysates. In addition, we have characterized a fast-labeling variant of SNAP-tag, termed SNAP(f), which displays up to a tenfold increase in its reactivity towards benzylguanine substrates. The presented data demonstrate that the combination of SNAP(f) and the fluorogenic substrates greatly reduces the background fluorescence for labeling and imaging applications. This approach enables highly sensitive spatiotemporal investigation of protein dynamics in living cells.

Site-specific protein labeling by intein-mediated protein ligation.

Intein-mediated protein ligation (IPL) employs an intein to create a protein possessing a C-terminal thioester that can be ligated to a protein or peptide with an amino-terminal cysteine via a native peptide bond. Here we present a procedure to conduct isolation and labeling of recombinant proteins expressed in E. coli using synthetic short peptides possessing a fluorescent moiety. This approach can be readily utilized for site-specific conjugation of a fluorophore to the C-terminus of a protein of interest, without the drawback of non-specific chemical labeling. This chapter also gives a general review of the critical parameters of intein-mediated cleavage and ligation reactions.

Fluorescent site-specific labeling of Escherichia coli expressed proteins with Sfp phosphopantetheinyl transferase.

Fluorescent tagging of proteins has become a critical step in optical analysis of protein function in vitro and in living cells. Here we describe a two-tag system for expression and isolation of a protein of interest from Escherichia coli and subsequent site-specific fluorescent labeling with Sfp phosphopantetheinyl transferase (Sfp synthase). In the example presented, adenoviral protein E3-14.7 K (E14.7) was expressed as a tripartite fusion protein with a fluorophore-targeting peptide tag and a chitin-binding domain. This system allows for rapid isolation of the recombinant fusion protein from crude bacterial cell lysate via a single chitin column. Sfp synthase-mediated labeling with fluorophore conjugated to coenzyme A-SH (CoA-SH) resulted in covalent attachment of a fluorescent dye to a specific residue of the peptide tag via a phosphopantetheinyl linker. The fluorescently labeled E14.7 fusion protein was analyzed with a fluorescence imager and subsequently transfected into mammalian cells for imaging with a fluorescence microscope.

Protein cyclization enhanced thermostability and exopeptidase-resistance of green fluorescent protein.

A mutant of green fluorescent protein (GFPmut3*) from the jellyfish Aequorea victoria was cyclized in vitro and in vivo by the use of a naturally split intein from the dnaE gene of Synechocystis species PCC6803 (Ssp). Cyclization of GFPmut3* was confirmed by amino acid sequencing and resulted in an increased electrophoretic mobility compared with the linear GFPmut3*. The circular GFPmut3* was 5 degrees C more thermostable than the linear form and significantly more resistant to proteolysis of exopeptidase. The circular GFPmut3* also displayed increased relative fluorescence intensity. In addition, chemical stability of GFPmut3* against GdnHCl revealed more stability of the circular form compared with the linear form.

Purification of green fluorescent protein using a two-intein system.

A two-intein purification system was developed for the affinity purification of GFPmut3*, a mutant of green fluorescent protein. The GFPmut3* was sandwiched between two self-cleaving inteins. This approach avoided the loss of the target protein which may result from in vivo cleavage of a single intein tag. The presence of N- and C-terminal chitin-binding domains allowed the affinity purification by a single-affinity chitin column. After the fusion protein was expressed and immobilized on the affinity column, self-cleavage of the inteins was sequentially induced to release the GFPmut3*. The yield was 2.41 mg from 1 l of bacterial culture. Assays revealed that the purity was up to 98% of the total protein. The fluorescence and circular dichroism spectrum of GFPmut3* demonstrated that the purified protein retains the correctly folded structure and function.

Intein-mediated peptide arrays for epitope mapping and kinase/phosphatase assays.

Synthetic peptides are widely used for production and analysis of antibodies as well as in the study of protein modification enzymes. To circumvent the technical challenges of the existing techniques regarding peptide quantization and normalization, a new method of producing peptide arrays has been developed. This approach utilizes intein-mediated protein ligation that involves linkage of a carrier protein possessing a reactive carboxyl-terminal thioester to a peptide with an amino-terminal cysteine through a native peptide bond. Ligated protein substrates or enzyme-treated samples are arrayed on nitrocellulose membranes with a standard dot-blot apparatus and analyzed by immunoassay. This technique has improved sensitivity and reproducibility, and is suitable for various peptide-based applications. In this report, several experimental procedures including epitope mapping and the study of protein modifications were described.

Reconstitution of glyphosate resistance from a split 5-enolpyruvyl shikimate-3-phosphate synthase gene in Escherichia coli and transgenic tobacco.

A highly N-phosphonomethylglycine (glyphosate)-resistant Pseudomonas fluorescens G2 5-enolpyruvyl shikimate-3-phosphate synthase (EPSPS) was mapped to identify potential split sites using a transposon-based linker-scanning procedure. Intein-mediated protein complementation was used to reconstitute glyphosate resistance from the genetically divided G2 EPSPS gene in Escherichia coli strain ER2799 and transgenic tobacco.

Design, preparation and use of ligated phosphoproteins: a novel approach to study protein phosphatases by dot blot array, ELISA and Western blot assays.

The study of substrate specificity of protein phosphatases (PPs) is very challenging since it is difficult to prepare a suitable phosphorylated substrate. Phosphoproteins, phosphorylated by a protein kinase, or chemically synthesized phosphopeptides are commonly used substrates for PPs. Both types of these substrates have their advantages and limitations. Phosphoproteins mimic more closely the physiologically relevant PP substrates, but their preparation is technically demanding. Synthetic phosphopeptides present advantages over proteins because they can be easily produced in large quantity and their amino acid sequence can be designed to contain potential determinants of substrate specificity. However, short peptides are less optimal compared to in vivo PP substrates and often display poor and variable binding to different matrices, resulting in low sensitivity in analysis of PP activity on solid support. In this work we utilize the intein-mediated protein ligation (IPL) technique to generate substrates for PPs, combining the advantages of proteins and synthetic peptides in one molecule. The ligation of a synthetic phosphopeptide to an intein-generated carrier protein (CP) with a one-to-one stoichiometry results in the formation of a ligated phosphoprotein (LPP). Three widely used assays, dot blot array, Western blot and ELISA were employed to study the PP activity on LPP substrates. Dephosphorylation was measured by detection of the remaining phosphorylation, or lack of it, with a phospho-specific antibody. The data show the advantage of LPPs over free peptides in assays on solid supports. LPPs exhibited enhanced binding to the matrices used in the study, which significantly improved sensitivity and consistency of the assays. In addition, saturation of the signal was circumvented by serial dilution of the assay samples. This report describes detailed experimental procedures for preparation of LPP substrates and their use in PP assays based on immobilization on solid supports.