Here we report that nanopillar array structures have an intrinsic ability to suppress electroosmotic flow (EOF). Currently using glass chips for electrophoresis requires laborious surface coating to control EOF, which works as a counterflow to the electrophoresis mobility of negatively charged samples such as DNA and sodium dodecyl sulfate (SDS) denatured proteins. Due to the intrinsic ability of the nanopillar array to suppress the EOF, we carried out electrophoresis of SDS-protein complexes in nanopillar chips without adding any reagent to suppress protein adsorption and the EOF. We also show that the EOF profile inside a nanopillar region was deformed to an inverse parabolic flow. We used a combination of EOF measurements and fluorescence observations to compare EOF in microchannel, nanochannel, and nanopillar array chips. Our results of EOF measurements in micro- and nanochannel chips were in complete agreement with the conventional equation of the EOF mobility (μ(EOF-channel) = αC(i)(-0.5), where C(i) is the bulk concentration of the i-ions and α differs in micro- and nanochannels), whereas EOF in the nanopillar chips did not follow this equation. Therefore we developed a new modified form of the conventional EOF equation, μ(EOF-nanopillar) ≈ β[C(i) - (C(i)(2)/N(i))], where N(i) is the number of sites available to i-ions and β differs for each nanopillar chip because of different spacings or patterns, etc. The modified equation of the EOF mobility that we proposed here was in good agreement with our experimental results. In this equation, we showed that the charge density of the nanopillar region, that is, the total number of nanopillars inside the microchannel, affected the suppression of EOF, and the arrangement of nanopillars into a tilted or square array had no effect on it.

We report a CE-LIF method for the separation and detection of five synthetic amyloid-β peptides corresponding to an important family of CSF-biomarkers in the context of Alzheimer disease (AD). The presumed most relevant peptides (Aβ1-42, Aβ1-40, and Aβ1-38) that may support the differentiation between AD and healthy patients or other dementias were successfully detected in CSF by incorporating an immunoconcentration step prior to CE analysis of derivatized peptides. We labeled the Aβ peptides with a fluoroprobe dye before CE-LIF analysis. This reagent reacts with the amino groups of lysine residues and produced mostly ditagged Aβ peptides under the proposed experimental conditions. The labeling reaction displayed similar efficiency with each one of the five different synthetic Aβ peptides that were tested. The limit of detection of the CE-LIF method approached 280 attomoles of injected synthetic labeled Aβ peptides. We obtained excellent correlation between peak areas and peptide concentrations from 35 nM to 750 nM. For the detection of Aβ peptides in human CSF samples, we enriched the peptides by immunoprecipitation prior to the CE-LIF analysis. The comparison of the CE-LIF profiles obtained from CSF samples from 3 AD patients and 4 non-demented control subjects indicated noticeable differences, suggesting that this method, which relies on a multibiomarker approach, may have potential as a clinical diagnostic test for AD.

Online preconcentration of human serum albumin (HSA) and its immunocomplex with a monoclonal antibody by on-chip transient isotachophoresis is reported. An 800-fold signal enhancement was achieved following the preconcentration on standard cross-channel microchips made of poly (methyl methacrylate). Sample injection, preconcentration, and separation were done continuously and controlled solely by a sequential voltage switching program. The preconcentration was followed by on-chip nondenaturing gel electrophoresis in methylcellulose solution. The method was applied to microchip electrophoresis immunoassay of HSA. Baseline separation of HSA and its immunocomplex was achieved in 25 s in the first 1 cm of the microchannel. In a direct immunoassay, the minimum detectable concentration of fluorescent labeled HSA by laser-induced fluorescence detection was 7.5 pM.