ICU Cockpit: a platform for collecting multimodal waveform data, AI-based computational disease modeling and real-time decision support in the intensive care unit.

ICU Cockpit: a secure, fast, and scalable platform for collecting multimodal waveform data, online and historical data visualization, and online validation of algorithms in the intensive care unit. We present a network of software services that continuously stream waveforms from ICU beds to databases and a web-based user interface. Machine learning algorithms process the data streams and send outputs to the user interface. The architecture and capabilities of the platform are described. Since 2016, the platform has processed over 89 billion data points (N = 979 patients) from 200 signals (0.5-500 Hz) and laboratory analyses (once a day). We present an infrastructure-based framework for deploying and validating algorithms for critical care. The ICU Cockpit is a Big Data platform for critical care medicine, especially for multimodal waveform data. Uniquely, it allows algorithms to seamlessly integrate into the live data stream to produce clinical decision support and predictions in clinical practice.

Novel 1:1 labeling and purification process for C-terminal thioester and single cysteine recombinant proteins using generic peptidic toolbox reagents.

We developed a versatile set of chemical labeling reagents which allow dye ligation to the C-terminus of a protein or a single internal cysteine and target purification in a simple two-step process. This simple process results in a fully 1:1 labeled conjugate suitable for all quantitative fluorescence spectroscopy and imaging experiments. We refer to a "generic labeling toolbox" because of the flexibility to choose one of many available dyes, spacers of different lengths and compositions which increase the target solubility, a variety of affinity purification tags, and different cleavage chemistries to release the 1:1 labeled proteins. Studying protein function in vitro or in the context of live cells and organisms is of vital importance in biological research. Although label free detection technologies gain increasing interest in molecular recognition science, fluorescence spectroscopy is still the most often used detection technique for assays and screens both in academic as well as in industrial groups. For generations, fluorescence spectroscopists have labeled their proteins of interest with small fluorescent dyes by random chemical linking on the proteins' exposed lysines and cysteines. Chemical reactions with a certain excess of activated esters or maleimides of longer wavelength dyes hardly ever result in quantitative labeling of the target protein. Most of the time, more than one exposed amino acid side chain reacts. This results in a mixture of dye-protein complexes of different labeling stoichiometries and labeling sites. Only mass spectrometry allows resolving the precise chemical composition of the conjugates. In "classical" ensemble averaging fluorescent experiments, these labeled proteins are still useful, and quantification of, e.g., ligand binding experiments, is achieved via knowledge of the overall protein concentration and a fluorescent signal change which is proportional to the amount of complex formed. With the development of fluorescence fluctuation analysis techniques working at single molecule resolution, like fluorescence correlation spectroscopy (FCS), fluorescence cross correlation spectroscopy (FCCS), fluorescence intensity diffusion analysis (FIDA), etc., it became important to work with homogeneously labeled target proteins. Each molecule participating in a binding equilibrium should be detectable when it freely fluctuates through the confocal focus of a microscope. The measured photon burst for each transition contains information about the size and the stoichiometry of a protein complex. Therefore, it is important to work with reagents that contain an exact number of tracers per protein at identical positions. The ideal fluorescent tracer-protein complex stoichiometry is 1:1. While genetic tags such as fluorescent proteins (FPs) are widely used to detect proteins, FPs have several limitations compared to chemical tags. For example, FPs cannot easily compete with organic dyes in the flexibility of modification and spectral range; moreover, FPs have disadvantages in brightness and photostability and are therefore not ideal for most biochemical single molecule studies. We present the synthesis of a series of exemplaric toolbox reagents and labeling results on three target proteins which were needed for high throughput screening experiments using fluorescence fluctuation analysis at single molecule resolution. On one target, Hu-antigen R (HuR), we demonstrated the activity of the 1:1 labeled protein in ribonucleic acid (RNA) binding, and the ease of resolving the stoichiometry of an RNA-HuR complex using the same dye on protein and RNA by Fluorescence Intensity Multiple Distribution Analysis (FIMDA) detection.

On-bead screens sample narrower affinity ranges of protein-ligand interactions compared to equivalent solution assays.

Conceptually, on-bead screening is one of the most efficient high-throughput screening (HTS) methods. One of its inherent advantages is that the solid support has a dual function: it serves as a synthesis platform and as a screening compartment. Compound purification, cleavage and storage and extensive liquid handling are not necessary in bead-based HTS. Since the establishment of one-bead one-compound library synthesis, the properties of polymer beads in chemical reactions have been thoroughly investigated. However, the characterization of the kinetics and thermodynamics of protein-ligand interactions on the beads used for screening has received much less attention. Consequently, the majority of reported on-bead screens are based on empirically derived procedures, independent of measured equilibrium constants and rate constants of protein binding to ligands on beads. More often than not, on-bead screens reveal apparent high affinity binders through strong protein complexation on the matrix of the solid support. After decoding, resynthesis, and solution testing the primary hits turn out to be unexpectedly weak binders, or may even fall out of the detection limit of the solution assay. Only a quantitative comparison of on-bead binding and solution binding events will allow systematically investigating affinity differences as function of protein and small molecule properties. This will open up routes for optimized bead materials, blocking conditions and other improved assay procedures. By making use of the unique features of our previously introduced confocal nanoscanning (CONA) method, we investigated the kinetic and thermodynamic properties of protein-ligand interactions on TentaGel beads, a popular solid support for on-bead screening. The data obtained from these experiments allowed us to determine dissociation constants for the interaction of bead-immobilized ligands with soluble proteins. Our results therefore provide, for the first time, a comparison of on-bead versus solution binding thermodynamics. Our data indicate that affinity ranges found in on-bead screening are indeed narrower compared to equivalent interactions in homogeneous solution. A thorough physico-chemical understanding of the molecular recognition between proteins and surface bound ligands will further strengthen the role of on-bead screening as an ultimately cost-effective method in hit and lead finding.

Confocal nanoscanning, bead picking (CONA): PickoScreen microscopes for automated and quantitative screening of one-bead one-compound libraries.

Solid phase combinatorial chemistry provides fast and cost-effective access to large bead based libraries with compound numbers easily exceeding tens of thousands of compounds. Incubating one-bead one-compound library beads with fluorescently labeled target proteins and identifying and isolating the beads which contain a bound target protein, potentially represents one of the most powerful generic primary high throughput screening formats. On-bead screening (OBS) based on this detection principle can be carried out with limited automation. Often hit bead detection, i.e. recognizing beads with a fluorescently labeled protein bound to the compound on the bead, relies on eye-inspection under a wide-field microscope. Using low resolution detection techniques, the identification of hit beads and their ranking is limited by a low fluorescence signal intensity and varying levels of the library beads' autofluorescence. To exploit the full potential of an OBS process, reliable methods for both automated quantitative detection of hit beads and their subsequent isolation are needed. In a joint collaborative effort with Evotec Technologies (now Perkin-Elmer Cellular Technologies Germany GmbH), we have built two confocal bead scanner and picker platforms PS02 and a high-speed variant PS04 dedicated to automated high resolution OBS. The PS0X instruments combine fully automated confocal large area scanning of a bead monolayer at the bottom of standard MTP plates with semiautomated isolation of individual hit beads via hydraulic-driven picker capillaries. The quantification of fluorescence intensities with high spatial resolution in the equatorial plane of each bead allows for a reliable discrimination between entirely bright autofluorescent beads and real hit beads which exhibit an increased fluorescence signal at the outer few micrometers of the bead. The achieved screening speed of up to 200,000 bead assayed in less than 7 h and the picking time of approximately 1 bead/min allow exploitation of one-bead one-compound libraries with high sensitivity, accuracy, and speed.

Single bead labeling method for combining confocal fluorescence on-bead screening and solution validation of tagged one-bead one-compound libraries.

Screening of one-bead one-compound libraries by incubating beads with fluorescently labeled target protein requires isolation and structure elucidation of a large number of primary hit beads. However, the potency of the identified ligands is only revealed after time consuming and expensive larger scale resynthesis and testing in solution. Often, many of the resynthesized compounds turn out to be weak target binders in solution due to large differences between surface and solution binding affinities. For an industry style high-throughput screening (HTS) process a high false positive rate is detrimental. We have therefore combined single bead and single molecule/single cell techniques into an integrated HTS process in which the picomole amount of substance contained on one isolated hit bead is sufficient for quality control, structure determination, and precise affinity determination to the target protein in solution.

Identification and mechanistic characterization of low-molecular-weight inhibitors for HuR.

Careful regulation of mRNA half-lives is a fundamental mechanism allowing cells to quickly respond to changing environmental conditions. The mRNA-binding Hu proteins are important for stabilization of short-lived mRNAs. Here we describe the identification and mechanistic characterization of the first low-molecular-weight inhibitors for Hu protein R (HuR) from microbial broths (Actinomyces sp.): dehydromutactin (1), MS-444 (2) and okicenone (3). These compounds interfere with HuR RNA binding, HuR trafficking, cytokine expression and T-cell activation. A mathematical and experimental analysis of the compounds' mode of action suggests that HuR homodimerizes before RNA binding and that the compounds interfere with the formation of HuR dimers. Our results demonstrate the chemical drugability of HuR; to our knowledge HuR is the first example of a drugable protein within the Hu family. MS-444, dehydromutactin and okicenone may become valuable tools for studying HuR function. An assessment of HuR inhibition as a central node in malignant processes might open up new conceptual routes toward combatting cancer.

Multifocal multiphoton microscopy based on multianode photomultiplier tubes.

Multifocal multiphoton microscopy (MMM) enhances imaging speed by parallelization. It is not well understood why the imaging depth of MMM is significantly shorter than conventional single-focus multiphoton microscopy (SMM). In this report, we show that the need for spatially resolved detectors in MMM results in a system that is more sensitive to the scattering of emission photons with reduced imaging depth. For imaging depths down to twice the scattering mean free path length of emission photons (2xl (s) (em)), the emission point spread function (PSF(em)) is found to consist of a narrow, diffraction limited distribution from ballistic emission photons and a broad, relatively low amplitude distribution from scattered photons. Since the scattered photon distribution is approximately 100 times wider than that of the unscattered photons at 2xl (s) (em), image contrast and depth are degraded without compromising resolution. To overcome the imaging depth limitation of MMM, we present a new design that replaces CCD cameras with multi-anode photomultiplier tubes (MAPMTs) allowing more efficient collection of scattered emission photons. We demonstrate that MAPMT-based MMM has imaging depth comparable to SMM with equivalent sensitivity by imaging tissue phantoms, ex vivo human skin specimens based on endogenous fluorophores, and green fluorescent protein (GFP) expressing neurons in mouse brain slices.

Multi-photon excitation of intrinsic protein fluorescence and its application to pharmaceutical drug screening.

The majority of proteins contain intrinsic fluorophores as natural sensors of molecular structures, dynamics, and interactions. The intrinsic protein fluorescence signal allows for the label-free and, hence, undisturbed and rapid study of protein-ligand interactions. Ultraviolet-based drug screening is hampered by the background, photobleaching, light scattering, inner filter effects, and interfering assay compounds. Such problems can be overcome by means of molecular three-photon excitation (3PE) with infrared femtosecond light pulses since longer excitation wavelengths result in less Raleigh scattering, and the subfemtoliter (confocal-like) 3PE volume minimizes out-of-focus photobleaching, background generation, and inner filter effects. We demonstrate the general feasibility of 3PE for protein spectroscopy and illustrate the technique's excellent potential for high-throughput screening. By using the intrinsic fluorescence intensity of a protein-substrate, we were able to discriminate between ligands of different affinities in binding assays.

Single-photon counting multicolor multiphoton fluorescence microscope.

We present a multicolor multiphoton fluorescence microscope with single-photon counting sensitivity. The system integrates a standard multiphoton fluorescence microscope, an optical grating spectrograph operating in the UV-Vis wavelength region, and a 16-anode photomultiplier tube (PMT). The major technical innovation is in the development of a multichannel photon counting card (mC-PhCC) for direct signal collection from multi-anode PMTs. The electronic design of the mC-PhCC employs a high-throughput, fully-parallel, single-photon counting scheme along with a high-speed electrical or fiber-optical link interface to the data acquisition computer. There is no electronic crosstalk among the detection channels of the mC-PhCC. The collected signal remains linear up to an incident photon rate of 10(8) counts per second. The high-speed data interface offers ample bandwidth for real-time readout: 2 MByte lambda-stacks composed of 16 spectral channels, 256 x 256 pixel image with 12-bit dynamic range can be transferred at 30 frames per second. The modular design of the mC-PhCC can be readily extended to accommodate PMTs of more anodes. Data acquisition from a 64-anode PMT has been verified. As a demonstration of system performance, spectrally resolved images of fluorescent latex spheres and ex-vivo human skin are reported. The multicolor multiphoton microscope is suitable for highly sensitive, real-time, spectrally-resolved three-dimensional imaging in biomedical applications.

Mitigating thermal mechanical damage potential during two-photon dermal imaging.

Two-photon excitation fluorescence microscopy allows in vivo high-resolution imaging of human skin structure and biochemistry with a penetration depth over 100 microm. The major damage mechanism during two-photon skin imaging is associated with the formation of cavitation at the epidermal-dermal junction, which results in thermal mechanical damage of the tissue. In this report, we verify that this damage mechanism is of thermal origin and is associated with one-photon absorption of infrared excitation light by melanin granules present in the epidermal-dermal junction. The thermal mechanical damage threshold for selected Caucasian skin specimens from a skin bank as a function of laser pulse energy and repetition rate has been determined. The experimentally established thermal mechanical damage threshold is consistent with a simple heat diffusion model for skin under femtosecond pulse laser illumination. Minimizing thermal mechanical damage is vital for the potential use of two-photon imaging in noninvasive optical biopsy of human skin in vivo. We describe a technique to mitigate specimen thermal mechanical damage based on the use of a laser pulse picker that reduces the laser repetition rate by selecting a fraction of pulses from a laser pulse train. Since the laser pulse picker decreases laser average power while maintaining laser pulse peak power, thermal mechanical damage can be minimized while two-photon fluorescence excitation efficiency is maximized.