Analytical Chemistry最新文献

Under the guidance of genome mining combined with bioassay-coupled metabolomic analyses, an unprecedented macrodiolide sanyensin (1), with two flexible macrolides fused by the rigid oxabicyclo[3.3.1]nonane core, was isolated from the deep-sea-derived Streptomyces sp. OUCT16-30. The stereochemistry of 1 was established by NOEs (nuclear Overhauser effects), J-based configuration analysis, Marfey's analysis, and together with a newly introduced stereochemical study workflow, which greatly shortens the time to obtain correct conformations of flexible structures based on the NMR constraints, thus leading to reliable quantum chemical calculations to establish the absolute configurations. This workflow is expected to have broad applicability for elucidating the stereochemistry of flexible natural products. The macrodiolide framework of 1 is proposed to be formed through a biocatalytic cyclodimerization, followed by a series of nonenzymatic reactions. This work leads to new insights into the unexplored biosynthetic potential of deep-sea microbes and also provides a practical streamline for efficient mining of novel natural products, from discovery to stereochemical finalization.

Nondestructive detection methods based on vibrational spectroscopy have been widely used in many critical applications in a variety of fields such as the chemical industry, pharmacy, national defense, security, and so on. As these methods/applications rely on machine learning models for data analysis, studying the threats associated with adversarial examples in vibrational spectroscopy and defenses against them is of great importance. In this paper, we propose a novel adversarial method to attack vibrational spectroscopy, named SynPat, where synthetic peaks produced by a physical model are placed at key locations to form adversarial perturbations. Our new attack generates perturbations that successfully deceive machine learning models for Raman and infrared spectrum analysis while they blend much better into the spectra and hence are unnoticeable to human operators, unlike the existing state-of-the-art adversarial attacking methods, e.g., images and audio. We verified the superiority of the proposed SynPat by an imperceptibility test conducted by human experts and of defense experiments by an AI detector. To the best of our knowledge, this is a first thorough study on the robustness of vibrational spectroscopic techniques against adversarial samples and defense mechanisms. Our extensive experiments show that machine learning models for vibrational spectroscopy, including conventional and deep models for Raman or infrared classification and regression, are all vulnerable to adversarial perturbations and thus may pose serious security threats to our society.

Circulating neutral glycosphingolipids (neutral GSLs (nGSLs)) are a unique subset of nGSLs that detach from organs or cell membranes and enter the bloodstream. Altered molecular distribution of circulating nGSL is increasingly associated with diseases. However, profiling of circulating nGSLs presents a lasting challenge due to their low abundances and structural complexity. Although TiO₂ magnetic nanoparticles (TiO₂ MNPs) were effective for the enrichment of nGSLs in brain tissue, the protocol showed limited selectivity for circulating nGSLs because their abundances were 100-times lower in human plasma than in brain tissue. In this work, we optimized the key parameters of selective enrichment by TiO₂ MNPs and achieved 1:10,000 selectivity for nGSLs over interfering phospholipids, while maintaining ∼70% recovery for different subclasses of nGSLs. By integrating TiO₂ MNP-based selective enrichment with reversed-phase liquid chromatography mass spectrometry and charge-tagging Paternò-Büchi derivatization, we achieved deep profiling of over 300 structures of nGSLs and sulfatides across 5 orders of magnitude in relative abundances, a significant leap regarding lipid coverage. We also depicted the structural atlas of nGSLs with defined headgroup, long-chain base, N-acyl chain, the location of desaturation, and 2-hydroxylation. Such information provides a valuable resource for lipidomic studies concerning the roles of circulating nGSLs in health and diseases.

Based on tunable diode laser absorption spectroscopy and off-axis integrated cavity output spectroscopy, a dual-range methane hybrid sensor was constructed utilizing the overtone absorption band of CH₄ gas molecules at 1653.7 nm. By simultaneously utilizing an off-axis integrated cavity and Herriott cell with an effective absorption path of 11 and 405 m, respectively, the two received photoelectric signals are decomposed into different frequency components by VMD and then reconstructed after SG filtering. Applying the global optimization algorithm DA-ELM to CH₄ concentration inversion, the correlation coefficient R² is as high as 0.9995. Through long-term stability verification, the instrument's standard deviation at 1 ppm is 27 ppb. After Allan-Werle deviation analysis, the sensor's limit of detection is 2.298 ppb at an integration time of 138 s. Using the self-developed sensor, the detection of dual-range trace CH₄ gas is achieved, enabling a dynamic detection range of trace CH₄ gas ranging from 400 ppb to 1000 ppm. The sensor realizes dual-range methane trace detection and actively controls methane emissions to improve environmental quality while taking into account the safety benefits of reducing production accidents.

Rapid detection of amines in complex mixtures presents a significant challenge. Here, we introduce a novel nuclear magnetic resonance (NMR) method for amine detection utilizing a probe with two fluorine atoms in distinct chemical environments. Upon interaction with an amine, the probe generates two atomic resonance peaks, which are used to create coordinates, revealing fluorine chemical shifts on the ¹⁹F NMR spectroscopy. This innovative approach allows for the clear distinction of amine signals in a two-dimensional plane. This method has been effectively employed in analyzing amines in pharmaceuticals and amino acids in Ophiopogon japonicus and dry white wine, providing a robust and general approach for amine analysis.

Chemically modified nucleic acid molecules have been developed as oligonucleotide therapeutics, and its assay is critical in quality assurance. The common DNA/RNA quantification method using UV-260 nm can lack accuracy because of structure modifications and the possible formation of higher-order structure (HOS). Additionally, process-associated water and counterions affect the accuracy in gravimetric analysis. Thus, to improve accuracy, efficiency, and flexibility, in this work a fast (<1 h) externally referenced ³¹P quantitative-NMR (qNMR) method was developed. The qNMR assay results agreed within 1-5% of the UV-260 nm results for the single-stranded DNA standards, confirming the method accuracy. Next, an NMR and UV comparison study was performed on intact oligonucleotide drug products. The ³¹P qNMR method showed 7 ± 2%, 8 ± 1%, and 12 ± 1% lower concentration values compared with drug product labels for eteplirsen, inotersen, and inclisiran, respectively. Meanwhile the UV-260 nm results showed 28 ± 3%, 10 ± 3%, and 10 ± 1% lower concentrations than the label for the same three drugs. The agreement between NMR and UV for phosphorothioate (PS)-based inotersen and mostly phosphodiester (PO)-based inclisiran suggest that the labeled concentration may have been obtained using different extinction coefficients. The underestimate of UV results for eteplirsen, which has a phosphorodiamidate morpholino oligomer (PMO) structure, suggests that the UV-260 nm extinction coefficient may need to be re-established for the PMO based oligonucleotide. Therefore, the ³¹P qNMR method could be a primary assay method for the oligonucleotide drug and reference standard.

Herein, a novel functional DNA structure, acid-stimulated self-assembly DNA nanonetwork (ASDN), was proposed for miRNA-221 sensitive detection and high-resolution living cancer cell imaging. Significantly, the self-assembly of ASDN only occurred in the acidic extracellular environment of cancer cells, which could be endocytosed by cancer cells to eliminate the interference of noncancer cells and deliver the ASDN into cancer cells. Subsequently, endogenous miRNA-221 could trigger the catalytic hairpin assembly within ASDN, resulting in the separation of the fluorophore Cy5 and the quencher BHQ2 to recover the substantial Cy5 fluorescence signals, thus achieving signal amplification for sensitive detection of miRNA-221 with a detection limit of 5.5 pM, as well as facilitating high-resolution and low-background imaging of miRNA-221 in cancer cells. In consequence, this strategy provides an innovative DNA nanonetwork to distinguish cancer cells from other cells for sensitive detection of biomarkers, offering a meaningful reference for the application of DNA nanostructure self-assembly technology in relevant fundamental research and disease diagnosis.

Messenger RNAs (mRNAs) have rapidly emerged as a pivotal class of biotherapeutics with great promise in the prevention and treatment of various diseases. As with other biotherapeutics, the sequence accuracy and integrity of mRNAs are critical quality attributes (CQAs), influencing the translation efficiency and protein expression fidelity of mRNAs. Due to the generation of short and repetitive oligonucleotides, traditional sequence mapping methods, which utilize in-solution RNase T1 digestion followed by LC-MS/MS analysis, face challenges in achieving high sequence coverage. In this study, we developed a novel flow through (FT)-based strategy to achieve the limited RNase T1 digestion of therapeutic mRNAs, leading to improved mRNA sequence mapping by LC-MS/MS analysis. Compared with the traditional in-solution digestion methods, the FT-based digestion method could consistently generate an increased number of unique oligonucleotides with miscleavages, which significantly boosted the overall sequence coverage (over 93%) of therapeutic mRNAs of varying sizes. Moreover, the automated digestion workflow using the AssayMAP platform offers significant advantages in method reproducibility and throughput. The high throughput and high sequence coverage features of this method could facilitate its wide application in the development of mRNA-based therapeutics.

Thermal denaturation (TD), known as antigen retrieval, heats tissue samples in a buffered solution to expose protein epitopes. Thermal denaturation of formalin-fixed paraffin-embedded samples enhances on-tissue tryptic digestion, increasing peptide detection using matrix-assisted laser desorption ionization imaging mass spectrometry (MALDI IMS). We investigated the tissue-dependent effects of TD on peptide MALDI IMS and liquid chromatography-tandem mass spectrometry signal in unfixed, frozen human colon, ovary, and pancreas tissue. In a triplicate experiment using time-of-flight, orbitrap, and Fourier-transform ion cyclotron resonance mass spectrometry platforms, we found that TD had a tissue-dependent effect on peptide signal, resulting in a (22.5%) improvement in peptide detection from the colon, a (73.3%) improvement in ovary tissue, and a (96.6%) improvement in pancreas tissue. Biochemical analysis of identified peptides shows that TD facilitates identification of hydrophobic peptides.