Differential adhesion of microspheres mediated by DNA hybridization I: experiment

Shear-driven rolling of DNA-adhesive microspheres

Many biologically important cell binding processes, such as the rolling of leukocytes in the vasculature, are multivalent, being mediated by large numbers of weak binding ligands. Quantitative agreement between experiments and models of rolling has been elusive and often limited by the poor understanding of the binding and unbinding kinetics of the ligands involved. Here, we present a cell-free experimental model for such rolling, consisting of polymer microspheres whose adhesion to a glass surface is mediated by ligands with well-understood force-dependent binding free energy-short complementary DNA strands. We observe robust rolling activity for certain values of the shear rate and the grafted DNA strands' binding free energy and force sensitivity. The simulation framework developed to model leukocyte rolling, adhesive dynamics, quantitatively captures the mean rolling velocity and lateral diffusivity of the experimental particles using known values of the experimental parameters. Moreover, our model captures the velocity variations seen within the trajectories of single particles. Particle-to-particle variations can be attributed to small, plausible differences in particle characteristics. Overall, our findings confirm that state-of-the-art adhesive dynamics simulations are able to capture the complex physics of particle rolling, boding well for their extension to modeling more complex systems of rolling cells.

Reviews on Current Liquid Biopsy for Detection and Management of Pancreatic Cancers

Pancreatic cancer is the fourth leading cause of cancer death in the United States. Pancreatic cancer presents dismal clinical outcomes in patients, and the incidence of pancreatic cancer has continuously increased to likely become the second most common cause of cancer-related deaths by as early as 2030. One of main reasons for the high mortality rate of pancreatic cancer is the lack of tools for early-stage detection. Current practice in detecting and monitoring therapeutic response in pancreatic cancer relies on imaging analysis and invasive endoscopic examination. Liquid biopsy-based analysis of genetic alterations in biofluids has become a fundamental component in the diagnosis and management of cancers. There is an urgent need for scientific and technological advancement to detect pancreatic cancer early and to develop effective therapies. The development of a highly sensitive and specific liquid biopsy tool will require extensive understanding on the characteristics of circulating tumor DNA in biofluids. Here, we have reviewed the current status of liquid biopsy in detecting and monitoring pancreatic cancers and our understanding of circulating tumor DNA that should be considered for the development of a liquid biopsy tool, which will greatly aid in the diagnosis and healthcare of people at risk.

Deposition of sticky spheres in channel flow: Modeling of surface coverage evolution requires accurate sphere-sphere collision hydrodynamics

We analyzed the role of hydrodynamic interactions in a microfluidic channel flow containing a dilute suspension of micron-scale colloidal spheres (0.03%, 0.1%, 0.3% volume fraction) engineered to adhere onto a collector patch on the channel wall at wall shear rates of 9.3-930 s^-1. Particle-wall adhesion was mediated by single-stranded DNA oligomers grafted onto the spheres and the glass channel wall, producing well-defined interactions via DNA strand base pairing. Particle positions in the flow were evolved using Brownian dynamics simulations in which hydrodynamic interactions between moving particles and the channel walls and/or adhered particles were computed off-line using a series of local simulations that explicitly resolve the fluid flow at the particle scale. By systematically varying the nature of hydrodynamic interactions captured in the Brownian dynamics simulations, we find that the interactions between moving and adhered particles represents the single most important physical element in such models. Once captured sufficiently accurately, the resulting models are able to predict coarse variables such as the overall particle coverage evolution, as well as more subtle characteristics, such as the microstructural distribution of the adhered particles.

Analysis of in Situ LNA and DNA Hybridization Events on Microspheres

The hybridization activity of single-stranded DNA and locked nucleic acid (LNA) sequences on microspheres is quantified in situ using flow cytometry. In contrast to conventional sample preparation for flow cytometry that involves several wash steps for posthybridization analysis, the current work entails directly monitoring hybridization events as they occur between oligonucleotide-functionalized microspheres and fluorescently tagged 9 or 15 base-long targets. We find that the extent of hybridization between single-stranded, immobilized probes and soluble targets generally increases with target sequence length or with the incorporation of LNA nucleotides in one or both oligonucleotide strands involved in duplex formation. The rate constants for duplex formation, on the other hand, remain nearly identical for all but one probe-target sequence combination. The exception to this trend involves the LNA probe and shortest perfectly matched DNA target, which exhibit a rate constant that is an order of magnitude lower than any other probe-target pair, including a mismatched duplex case. Separate studies entailing brief heat treatments to suspensions generally do not consistently yield appreciable differences in associated target densities to probe-functionalized microspheres.

Engineering nanoscale surface features to sustain microparticle rolling in flow

Nanoscopic features of channel walls are often engineered to facilitate microfluidic transport, for instance when surface charge enables electro-osmosis or when grooves drive mixing. The dynamic or rolling adhesion of flowing microparticles on a channel wall holds potential to accomplish particle sorting or to selectively transfer reactive species or signals between the wall and flowing particles. Inspired by cell rolling under the direction of adhesion molecules called selectins, we present an engineered platform in which the rolling of flowing microparticles is sustained through the incorporation of entirely synthetic, discrete, nanoscale, attractive features into the nonadhesive (electrostatically repulsive) surface of a flow channel. Focusing on one example or type of nanoscale feature and probing the impact of broad systematic variations in surface feature loading and processing parameters, this study demonstrates how relatively flat, weakly adhesive nanoscale features, positioned with average spacings on the order of tens of nanometers, can produce sustained microparticle rolling. We further demonstrate how the rolling velocity and travel distance depend on flow and surface design. We identify classes of related surfaces that fail to support rolling and present a state space that identifies combinations of surface and processing variables corresponding to transitions between rolling, free particle motion, and arrest. Finally we identify combinations of parameters (surface length scales, particle size, flow rates) where particles can be manipulated with size-selectivity.

Temperature- and flow-enhanced detection specificity of mutated DNA against the wild type with reporter microspheres

Detection of mutated (MT) deoxyribonucleic acid (DNA) amongst the wild type (WT) requires the probe DNA (pDNA) that is complementary to the MT to discriminate the WT by one or two nucleotide mismatches. Traditionally this is achieved by raising the temperature to above the melting temperature (Tm) of the WT (TWT) but below that of the MT (TMT). However, a raised temperature is also accompanied by a weakened binding of the MT to the pDNA which can reduce the detection sensitivity. In this study, we investigated flow as a way to enhance MT detection specificity at a lower temperature. Gold-coated glass (GCG) slides immobilized with pDNA complementary to the target MT were placed at the center of the flow cell. The detection was done by flowing MT or WT at various concentrations followed by flowing 10(5) ml(-1) fluorescent reporter microspheres (FRMs) that were 6 μm in size and coated with reporter DNA complementary to the MT or WT but different from the pDNA at various flow rates and temperatures. The detection of MT or WT was characterized by counting the FRMs captured on the GCG. Hepatitis B virus 1762/1764 double mutation (HBV DM) was the model MT and the TMT and TWT were 47 °C and 22 °C, respectively. It was shown that at room temperature, flow initially increased the binding of both the MT and WT at lower flow rates but decreased the binding at flow rates ≥4 ml min(-1) due to the increase in the flow-induced impingement force on the FRMs to overcome the binding of the MT and the WT to the GCG at higher flow rates. At ≥30 °C the decrease in binding of the WT with an increasing flow rate was more than that of the MT because 30 °C was above the TWT but still well below the TMT. As a result, the detection of MT at 30 °C with a flow rate of 4 ml min(-1) was more specific than at 35 °C without flow. These results indicate that flow can diminish WT binding at a lower temperature than without flow and allow MT detection to occur at a lower temperature with high specificity.

Was the Scanner Calibration Slide used for its intended purpose?

In the article, Scanner calibration revisited, BMC Bioinformatics 2010, 11:361, Dr. Pozhitkov used the Scanner Calibration Slide, a key product of Full Moon BioSystems to generate data in his study of microarray scanner PMT response and proposed a mathematic model for PMT response [1]. In the end, the author concluded that "Full Moon BioSystems calibration slides are inadequate for performing calibration," and recommended "against using these slides." We found these conclusions are seriously flawed and misleading, and his recommendation against using the Scanner Calibration Slide was not properly supported.

Shear unzipping of double-stranded DNA

We use a simple nonlinear scaler displacement model to calculate the distribution of effects created by a shear stress on a double-stranded DNA (dsDNA) molecule and the value of shear force F(c) that is required to separate the two strands of a molecule at a given temperature. It is shown that for molecules of base pairs fewer than than 21, the entire single strand moves in the direction of applied force, whereas for molecules having base pairs more than 21, part of the strand moves in the opposite direction under the influence of force acting on the other strand. This result as well as the calculated values of F(c) as a function of length of dsDNA molecules are in very good agreement with the experimental values of Hatch et al. [Phys. Rev. E 78, 011920 (2008)].

Manipulating DNA probe presentation via enzymatic cleavage of diluent strands

We previously reported a system for the controlled redispersion of DNA-linked aggregates using secondary, competitive hybridization events and found that complete redispersion is contingent upon dilution of the active 20 base-long probe strands with 20 base-long nonhybridizing strands. Here, to reduce the steric interference of nonhybridizing or diluent strands on probe activity, we investigate the effect of shorter diluent strands on the hybridization activity of immobilized probes using the following two approaches: (1) simultaneously coupling shorter diluent strands and longer probe strands to microspheres and (2) simultaneously coupling diluent and probe strands of the same base length to microspheres and then clipping diluent strands with the restriction endonuclease AluI. Results indicate that one can reduce the duplex density down by 50-70% of its initial value, depending on the location of the recognition motif along the hybridization segment. In addition, tighter control over the number of probe-target duplexes is achieved with the enzyme-based approach.

Demonstration that the shear force required to separate short double-stranded DNA does not increase significantly with sequence length for sequences longer than 25 base pairs

We have measured the shear force for short double-stranded DNA sequences pulled by either the 3'3' or 5'5' ends and find that the shear force is independent of the pulling technique. For the 50% GC sequences examined, the force is a linear function of DNA length up to 20 base pairs (bp); however, we show that, as predicted by deGennes, the shear force approaches an asymptotic value in the limit where the number of base pairs approaches infinity, where the shear force for a 32 bp sequence is within 5% of the asymptotic value of 61.4 pN . Fits to deGennes' theory suggest that the shear force is distributed over fewer than 10 bp at the end of the sequence, with the rest of the sequence experiencing negligible shear force. The single base pair rupture force and the ratio of the backbone spring constant to the base pair spring constant determined from fits of the data to deGennes' theory are consistent with ab initio predictions.

Diverse approaches to the controlled generation of nanotextured surfaces

Smooth, chemically uniform surfaces are seldom found in nature. Mimicry of natural variegation is a powerful approach for controlling chemical affinity at the nanoscale. Molecular recognition is one of the fundamental concepts underlying the functioning of living cells, and it depends on a particular relationship between the nanoscale, i.e., molecular, variegations of two potentially interacting molecular partners. The primary subject matter of this paper is how to articially generate appropriate nanoscale texture at the surfaces of materials. Excluding "pick and place" chemistry, in which essentially a Maxwellian demon intervenes to place objects with atomic precision, and nowadays achievable through an adaptation of atomic force microscopy, on the grounds that it is too slow to be practicable for fabricating useful quantities of material, three approaches are explored in some detail: (i) "powder", i.e., mixing at least two individually monofunctional (with respect to the ultimate molecular recognition task) precursor components (possibly with secondary functionality enabling them to appropriately self-assemble on a substratum); (ii) mixing polymers with the possibility of phase separation and frustrated phase separation with block copolymers; and (iii) felting. The emphasis is on processes that create more or less irregular structures, rather than regular arrays. The final section deals with the metrology of nanotexture.

Flow of microgel capsules through topographically patterned microchannels

We investigated the flow dynamics of microgel capsules in topographically patterned microfluidic devices. For microgels flowing through channel constrictions, or orifices, we observed three phenomena: (i) the effect of confinement, (ii) the role of interactions between the microgels and the channel surface, and (iii) the effect of the velocities of microgels prior to their passage through an orifice. We studied negatively charged alginate microgels and positively charged alginate microgels coated with N-(2-hydroxy)propyl-3-trimethylammonium chitosan chloride (HTCC). Aqueous dispersions of microgels were driven through poly(dimethyl siloxane) microchannels carrying a weak negative surface charge. The velocity of the continuous phase, and hence, the velocity of the microgels increased as they passed through topographically patterned orifices. Alginate microgels were observed to have a larger increase in velocity relative to HTCC-coated alginate microgels. This effect, which was attributed to electrostatic attraction or repulsion, was found to be strongest for orifices with dimensions close to the microgel diameter. For example, when 75 microm-diameter microgels flowed through a 76 microm orifice, alginate gels (negatively charged) experienced a 2x greater increase in velocity than HTCC-coated (positively charged) microgels. This effect was exaggerated at lower initial flow rates. For example, when 75 microm-diameter microgels flowed through an 80 microm orifice, a two-fold difference in the velocity changes of the two microgel types was observed when the initial flow rate was 275 microm s(-1), while a three-fold difference in velocity changes was observed when the initial flow rate was 130 microm s(-1). We speculate that these studies will be useful for modeling the flow of suspensions of cells or other biologically relevant particles for a wide range of applications.