Microscale measurements of protein complexes from single cells.

Proteins execute numerous cell functions in concert with one another in protein-protein interactions (PPI). While essential in each cell, such interactions are not identical from cell to cell. Instead, PPI heterogeneity contributes to cellular phenotypic heterogeneity in health and diseases such as cancer. Understanding cellular phenotypic heterogeneity thus requires measurements of properties of PPIs such as abundance, stoichiometry, and kinetics at the single-cell level. Here, we review recent, exciting progress in single-cell PPI measurements. Novel technology in this area is enabled by microscale and microfluidic approaches that control analyte concentration in timescales needed to outpace PPI disassembly kinetics. We describe microscale innovations, needed technical capabilities, and methods poised to be adapted for single-cell analysis in the near future.

Summit: Automated Analysis of Arrayed Single-Cell Gel Electrophoresis.

New pipelines are required to automate the quantitation of emerging high-throughput electrophoretic (EP) assessment of DNA damage, or proteoform expression in single cells. EP cytometry consists of thousands of Western blots performed on a microscope slide-sized gel microwell array for single cells. Thus, EP cytometry images pose an analysis challenge that blends requirements for accurate and reproducible analysis encountered for both standard Western blots and protein microarrays. Here, we introduce the Summit algorithm to automate array segmentation, peak background subtraction, and Gaussian fitting for EP cytometry. The data structure storage of parameters allows users to perform quality control on identically processed data, yielding a ~6.5% difference in coefficient of quartile variation (CQV) of protein peak area under the curve (AUC) distributions measured by four users. Further, inspired by investigations of background subtraction methods to reduce technical variation in protein microarray measurements, we aimed to understand the trade-offs between EP cytometry analysis throughput and variation. We found an 11%-50% increase in protein peaks that passed quality control with a subtraction method similar to microarray "average on-boundary" versus an axial subtraction method. The background subtraction method only mildly influences AUC CQV, which varies between 1% and 4.5%. Finally, we determined that the narrow confidence interval for peak location and peak width parameters from Gaussian fitting yield minimal uncertainty in protein sizing. The AUC CQV differed by only ~1%-2% when summed over the peak width bounds versus the 95% peak width confidence interval. We expect Summit to be broadly applicable to other arrayed EP separations, or traditional Western blot analysis.

Measuring expression heterogeneity of single-cell cytoskeletal protein complexes.

Multimeric cytoskeletal protein complexes orchestrate normal cellular function. However, protein-complex distributions in stressed, heterogeneous cell populations remain unknown. Cell staining and proximity-based methods have limited selectivity and/or sensitivity for endogenous multimeric protein-complex quantification from single cells. We introduce micro-arrayed, differential detergent fractionation to simultaneously detect protein complexes in hundreds of individual cells. Fractionation occurs by 60 s size-exclusion electrophoresis with protein complex-stabilizing buffer that minimizes depolymerization. Proteins are measured with a ~5-hour immunoassay. Co-detection of cytoskeletal protein complexes in U2OS cells treated with filamentous actin (F-actin) destabilizing Latrunculin A detects a unique subpopulation (~2%) exhibiting downregulated F-actin, but upregulated microtubules. Thus, some cells may upregulate other cytoskeletal complexes to counteract the stress of Latrunculin A treatment. We also sought to understand the effect of non-chemical stress on cellular heterogeneity of F-actin. We find heat shock may dysregulate filamentous and globular actin correlation. In this work, our assay overcomes selectivity limitations to biochemically quantify single-cell protein complexes perturbed with diverse stimuli.

Microparticle Delivery of Protein Markers for Single-Cell Western Blotting from Microwells.

Immunoblotting confers protein identification specificity beyond that of immunoassays by prepending protein electrophoresis (sizing) to immunoprobing. To accurately size protein targets, sample analysis includes concurrent analysis of protein markers with known molecular masses. To incorporate protein markers in single-cell western blotting, microwells are used to isolate individual cells and protein marker-coated microparticles. A magnetic field directs protein-coated microparticles to >75% of microwells, so as to 1) deliver a quantum of protein marker to each cell-laden microwell and 2) synchronize protein marker solubilization with cell lysis. Nickel-coated microparticles are designed, fabricated, and characterized, each conjugated with a mixture of histidine-tagged proteins (42.3-100 kDa). Imidazole in the cell lysis buffer solubilizes protein markers during a 30 s cell lysis step, with an observed protein marker release half-life of 4.46 s. Across hundreds of individual microwells and different microdevices, robust log-linear regression fits (R2 > 0.97) of protein molecular mass and electrophoretic mobility are observed. The protein marker and microparticle system is applied to determine the molecular masses of five endogenous proteins in breast cancer cells (GAPDH, ß-TUB, CK8, STAT3, ER-α), with <20% mass error. Microparticle-delivered protein standards underpin robust, reproducible electrophoretic cytometry that complements single-cell genomics and transcriptomics.

Joule Heating-Induced Dispersion in Open Microfluidic Electrophoretic Cytometry.

While protein electrophoresis conducted in capillaries and microchannels offers high-resolution separations, such formats can be cumbersome to parallelize for single-cell analysis. One approach for realizing large numbers of concurrent separations is open microfluidics (i.e., no microchannels). In an open microfluidic device adapted for single-cell electrophoresis, we perform 100s to 1000s of simultaneous separations of endogenous proteins. The microscope slide-sized device contains cells isolated in microwells located in a ∼40 µm polyacrylamide gel. The gel supports protein electrophoresis after concurrent in situ chemical lysis of each isolated cell. During electrophoresis, Joule (or resistive) heating degrades separation performance. Joule heating effects are expected to be acute in open microfluidic devices, where a single, high-conductivity buffer expedites the transition from cell lysis to protein electrophoresis. Here, we test three key assertions. First, Joule heating substantially impacts analytical sensitivity due to diffusive losses of protein out of the open microfluidic electrophoretic (EP) cytometry device. Second, increased analyte diffusivity due to autothermal runaway Joule heating is a dominant mechanism that reduces separation resolution in EP cytometry. Finally, buffer exchange reduces diffusive losses and band broadening, even when handling single-cell lysate protein concentrations in an open device. We develop numerical simulations of Joule heating-enhanced diffusion during electrophoresis and observe ∼50% protein loss out of the gel, which is reduced using the buffer exchange. Informed by analytical model predictions of separation resolution (with Joule heating), we empirically demonstrate nearly fully resolved separations of proteins with molecular mass differences of just 4 kDa or 12% (GAPDH, 36 kDa; PS6, 32 kDa) in each of 129 single cells. The attained separation performance with buffer exchange is relevant to detection of currently unmeasurable protein isoforms responsible for cancer progression.

Profiling protein expression in circulating tumour cells using microfluidic western blotting.

Circulating tumour cells (CTCs) are rare tumour cells found in the circulatory system of certain cancer patients. The clinical and functional significance of CTCs is still under investigation. Protein profiling of CTCs would complement the recent advances in enumeration, transcriptomic and genomic characterization of these rare cells and help define their characteristics. Here we describe a microfluidic western blot for an eight-plex protein panel for individual CTCs derived from estrogen receptor-positive (ER+) breast cancer patients. The precision handling and analysis reveals a capacity to assay sparingly available patient-derived CTCs, a biophysical CTC phenotype more lysis-resistant than breast cancer cell lines, a capacity to report protein expression on a per CTC basis and two statistically distinct GAPDH subpopulations within the patient-derived CTCs. Targeted single-CTC proteomics with the capacity for archivable, multiplexed protein analysis offers a unique, complementary taxonomy for understanding CTC biology and ascertaining clinical impact.

Single cell-resolution western blotting.

This protocol describes how to perform western blotting on individual cells to measure cell-to-cell variation in protein expression levels and protein state. Like conventional western blotting, single-cell western blotting (scWB) is particularly useful for protein targets that lack selective antibodies (e.g., isoforms) and in cases in which background signal from intact cells is confounding. scWB is performed on a microdevice that comprises an array of microwells molded in a thin layer of a polyacrylamide gel (PAG). The gel layer functions as both a molecular sieving matrix during PAGE and a blotting scaffold during immunoprobing. scWB involves five main stages: (i) gravity settling of cells into microwells; (ii) chemical lysis of cells in each microwell; (iii) PAGE of each single-cell lysate; (iv) exposure of the gel to UV light to blot (immobilize) proteins to the gel matrix; and (v) in-gel immunoprobing of immobilized proteins. Multiplexing can be achieved by probing with antibody cocktails and using antibody stripping/reprobing techniques, enabling detection of 10+ proteins in each cell. We also describe microdevice fabrication for both uniform and pore-gradient microgels. To extend in-gel immunoprobing to gels of small pore size, we describe an optional gel de-cross-linking protocol for more effective introduction of antibodies into the gel layer. Once the microdevice has been fabricated, the assay can be completed in 4-6 h by microfluidic novices and it generates high-selectivity, multiplexed data from single cells. The technique is relevant when direct measurement of proteins in single cells is needed, with applications spanning the fundamental biosciences to applied biomedicine.

Effect of Polymer Hydration State on In-Gel Immunoassays.

Applications as diverse as drug delivery and immunoassays require hydrogels to house high concentration macromolecular solutions. Yet, thermodynamic partitioning acts to lower the equilibrium concentration of macromolecules in the hydrogel, as compared to the surrounding liquid phase. For immunoassays that utilize a target antigen immobilized in the hydrogel, partitioning hinders introduction of detection antibody into the gel and, consequently, reduces the in-gel concentration of detection antibody, adversely impacting assay sensitivity. Recently, we developed a single-cell targeted proteomic assay with polyacrylamide gel electrophoresis of single cell lysates followed by an in-gel immunoassay. In the present work, we overcome partitioning that both limits analytical sensitivity and increases consumption of costly detection antibody by performing the immunoassay step after dehydrating the antigen-containing polyacrylamide gel. Gels are rehydrated with a solution of detection antibody. We hypothesized that matching the volume of detection antibody solution with the hydrogel water volume fraction would ensure that, at equilibrium, the detection antibody mass resides in the gel and not in the liquid surrounding the gel. Using this approach, we observe (compared with antibody incubation of hydrated gels): (i) 4-11 fold higher concentration of antibody in the dehydrated gels and in the single-cell assay (ii) higher fluorescence immunoassay signal, with up to 5-fold increases in signal-to-noise-ratio and (iii) reduced detection antibody consumption. We also find that detection antibody signal may be less well-correlated with target protein levels (GFP) using this method, suggesting a trade-off between analytical sensitivity and variation in immunoprobe signal. Our volume-matching approach for introducing macromolecular solutions to hydrogels increases the local in-gel concentration of detection antibody without requiring modification of the hydrogel structure, and thus we anticipate broad applicability to hydrogel-based assays, diagnostics, and drug delivery.

The differential extension in dsDNA bound to Rad51 filaments may play important roles in homology recognition and strand exchange.

RecA and Rad51 proteins play an important role in DNA repair and homologous recombination. For RecA, X-ray structure information and single molecule force experiments have indicated that the differential extension between the complementary strand and its Watson-Crick pairing partners promotes the rapid unbinding of non-homologous dsDNA and drives strand exchange forward for homologous dsDNA. In this work we find that both effects are also present in Rad51 protein. In particular, pulling on the opposite termini (3' and 5') of one of the two DNA strands in a dsDNA molecule allows dsDNA to extend along non-homologous Rad51-ssDNA filaments and remain stably bound in the extended state, but pulling on the 3'5' ends of the complementary strand reduces the strand-exchange rate for homologous filaments. Thus, the results suggest that differential extension is also present in dsDNA bound to Rad51. The differential extension promotes rapid recognition by driving the swift unbinding of dsDNA from non-homologous Rad51-ssDNA filaments, while at the same time, reducing base pair tension due to the transfer of the Watson-Crick pairing of the complementary strand bases from the highly extended outgoing strand to the slightly less extended incoming strand, which drives strand exchange forward.

Tension on dsDNA bound to ssDNA-RecA filaments may play an important role in driving efficient and accurate homology recognition and strand exchange.

It is well known that during homology recognition and strand exchange the double stranded DNA (dsDNA) in DNA/RecA filaments is highly extended, but the functional role of the extension has been unclear. We present an analytical model that calculates the distribution of tension in the extended dsDNA during strand exchange. The model suggests that the binding of additional dsDNA base pairs to the DNA/RecA filament alters the tension in dsDNA that was already bound to the filament, resulting in a non-linear increase in the mechanical energy as a function of the number of bound base pairs. This collective mechanical response may promote homology stringency and underlie unexplained experimental results.

RecA homology search is promoted by mechanical stress along the scanned duplex DNA.

A RecA-single-stranded DNA (RecA-ssDNA) filament searches a genome for sequence homology by rapidly binding and unbinding double-stranded DNA (dsDNA) until homology is found. We demonstrate that pulling on the opposite termini (3' and 5') of one of the two DNA strands in a dsDNA molecule stabilizes the normally unstable binding of that dsDNA to non-homologous RecA-ssDNA filaments, whereas pulling on the two 3', the two 5', or all four termini does not. We propose that the 'outgoing' strand in the dsDNA is extended by strong DNA-protein contacts, whereas the 'complementary' strand is extended by the tension on the base pairs that connect the 'complementary' strand to the 'outgoing' strand. The stress resulting from different levels of tension on its constitutive strands causes rapid dsDNA unbinding unless sufficient homology is present.

Probing the mechanical stability of DNA in the presence of monovalent cations.

We examine the interaction between monovalent cations and DNA using several different assays that measure the stability of double-stranded DNA (dsDNA). The thermal melting of dsDNA and the mechanical separation of dsDNA into two single strands both depend on the stability of dsDNA with respect to ssDNA and are sensitive to the interstrand phosphate repulsion. We find that the experimentally measured melting temperatures and unzipping forces are approximately the same for all of the ions considered in this study. Likewise, the force required to transform B-DNA into the overstretched form is also similar for all of the ions. In contrast, for a given ion concentration, the force at which the overstretched state fully relaxes back to the canonical B-DNA form depends on the cation; however, for all cations, the overstretching force decreases with decreasing ion concentration, suggesting that this force is sensitive to screening. We observe a general trend for smaller ions to produce more efficient relaxation. Finally, for a given cation, the relaxation can also depend on the anion.