Thursday, June 21 – 4:30 PM
BioMEMS / Instrumentation
Highspeed solution exchange in an artificial bilayer system for high-throughput screening of drugs targeting ion channels
Shiv Acharya, Ivan Pushkasrky, Greg Brown, Jacob Schmidt
UC Los Angeles
Ion channel targeting drugs are relatively unexplored due to the limitations of current drug screening technologies. Patchclamp requires dedicated time and labor to maintain expensive live cell cultures, and is limited to measurements of ion channels in stringent cell environments. Sessile droplet artificial bilayer platforms provide an inexpensive alternative to patch clamp, with no requirement for live cells, flexibility of the membrane components and environment, and the ability to change the membrane environment with soluton exchange. The current high speed solution exchange technology allows solution exchange flowrates of 20 μL/min, limited by bilayer stability during exchange. With an agarose hydrogel supporting the bilayer, we have developed an artificial bilayer system capable of exchanging solutions at 69.5 mL/min flowrates. Measurements of physiologically relevant ion channels demonstrate that the technology is capable of measuring a drug potency metric (IC50) within 2 minutes. With the advances made in this project, the artificial bilayer system has become a high throughput, controlled technology for the screening of ion channel targeting drugs.
Programming Fluid Flow in Microchannels Using Microstructure
H. Amini, M. Masaeli, E. Sollier, Y. Xie, B. Ganapathysubramanian, H. A. Stone, D. Di Carlo
UC Los Angeles
Control of fluid streams is useful in biological processing, chemical reaction engineering, and creating structured materials. However, general strategies to engineer the cross-sectional form and motion of fluid streams have been limited. Strategies to mix fluid and control particles using engineered systems exist, often relying on chaotic fluid transformations to disrupt sustained regions of order in the flow. Rather than apply flow transformations to prevent order, here we develop a hierarchical approach to engineer fluid streams into a broad class of complex configurations.
We use cylindrical pillars to induce significant deformations in laminar flow. Numerical simulations predict that as fluid passes centrally positioned pillars in a straight microchannel, the fluid parcels near the channel centerline move towards the side walls, while fluid parcels near the top and bottom walls move towards the channel center. This phenomenon, validated experimentally, effectively creates a set of net rotational secondary flows within the microchannel. Hence, the flow is irreversibly twisted near the pillar, leading to a significant final flow deformation. The lateral position of the pillar can be used to tune the lateral position of the net recirculating flow (Figure).
Hierarchical flow deformation operations can be integrated to execute sophisticated programs and render complex flow-shapes. We can numerically predict the deformation near a single pillar with high precision and accordingly, predict the total transformation function of any potential program. Consequently, a user can use a library of pre-simulated motions and engineer a flow-shape of interest quickly, at a low cost, and with high accuracy. Systematic discretization of the pillar positions enables each program to be simply communicated using the inlet condition and the sequence of pillar positions. Therefore, analogous to programming of software, a designer can build upon previously demonstrated functions and integrate them in new ways to create more complex and useful outcomes.
Brain Injury Screening Diagnostics for Emergency Medicine: Quantitation of Cerebrospinal Fluid Specific Proteins in Human Nasal Discharge
Akwasi A. Apori, Amy E. Herr
UC Berkeley
To expedite and automate raw sample processing for clinical diagnostics, we introduce an on-chip sample preparation and immunosubtraction diagnostic for emergency room screening of cerebrospinal fluid (CSF) rhinorrhea. CSF rhinorrhea occurs when CSF from the cranial cavity leaks into the nasal cavity owing to head trauma. This can lead to bacterial and viral infection or stroke. State-of-the-art detection includes CT scan, MRI, endoscopic examination, or ELISA. These techniques have drawbacks including limited availability and high cost (imaging), or long assay times and large sample volume requirements (ELISA). Consequently, emergency diagnostics would benefit from a rapid and non-invasive nasal discharge screening assay for CSF- specific proteins [Warnecke et al. 2004]. Here we report technology for portable screening assays through analysis of nasal discharge from patient cohorts. The platform integrates multiple sample preparation steps into an automated and rapid device for multiplexed detection of differentiating CSF biomarkers indicative of head trauma in raw nasal discharge (prolactin inducible protein, PIP & transthyretin, TTR [Burkhard et al. 2001]).
Advances in in-situ photo-polymerization are essential to integrating sample preparation with analytical functions harnessing small transport length scale architectures. First, raw human nasal discharge is processed on-chip via labeling, enrichment, and target antibody binding for on-the-fly multi-target identification.Then, a discontinuous polyacrylamide gel is used to create a step decrease in nanopore size (immuno-filter) at the start of a polyacrylamide gel electrophoresis (PAGE) separation channel. Target analyte bound to the capture antibody is subtracted (via size-based exclusion) from subsequent PAGE analysis (Figure). Our preliminary assays rapidly (15 min) generate dose-response information for putative biomarkers simultaneously. These pilot patient cohort studies support the assertion that TTR and PIP comprise an effective protein biomarker panel for quantitation of nasal mucous or CSF content in unknown nasal discharge samples. Early results detect ~95% endogenous TTR from CSF in <1 min.
Custom Electrokinetic Patterning of Proteins in Microfluidic Gels and Applications in Confirmatory Immunoassays
Muhammet Kursad Araz, Akwasi Apori, Amy E. Herr
UC Berkeley
Enzyme immunoassays (EIA) are widely used for infectious disease diagnostic screening due to their low cost. EIA have high sensitivities, however due to their questionable specificities, a positive result is followed by a confirmatory assay. Confirmatory immunoassays use immunological blotting techniques in which antigens fixed into a membrane or strip capture the target antibodies, if present in the sample solution in which the strip is incubated. Diffusion based antibody capture assays require up to overnight incubation along with trained staff and medium laboratory infrastructure. In the Herr Lab, we have developed a novel antigen patterning method which enables electrokinetic immobilization of antigens in a polyacrylamide gel, polymerized in a glass microchannel with only two openings for input and output. Biotin conjugated antigens are delivered into the channel with an applied electric field and immobilized via streptavidin functionalized sites in the gel matrix. Antigens captured in the gel matrix form a 3D sieving and capturing mechanism leading to a maximized antibody-antigen interactions. Presently, we’ve demonstrated patterning of 3 different antigens in a short (2.5mm) microchannel with excellent purities. The custom multiplexed single- channel patterning enabled the capture of electrophoretically delivered antibodies spiked in complex fluids such as human sera, at their respective antigen band locations. Assay performance demonstrated with human serum samples shows quantitative detection of specific human Hepatitis C antibodies at clinically relevant sensitivities in 30 minutes, compared to 8 hours for typical confirmatory assays. The antigen patterning step uses 1microWatt power at 2-10Volt drive enabling battery operable custom patterning. The two input and short channel infrastructure leads to high-density device fabrication with 3-4 devices per cm2 of glass substrates, with significantly reduced expensive biosample required for patterning. Present device architecture enables patterning of 10,000 individual channels with the use of 1 mg of desired antigen.
Electrostatic immobilization of SDS coated proteins for advanced microfluidic western blotting
Minsub Chung, Dohyun Kim, Amy E. Herr
UC Berkeley
We report a novel protein immobilization matrix for fully integrated microfluidic Western blotting (WB). The electrostatic immobilization gel (EIG) enables immobilization of all proteins sized using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), for subsequent electrophoretic probing with detection affinity reagents (e.g., labeled antibodies). The electrostatic capture strategy uses polyacrylamide gel grafted with concentrated point charges (charge carrying macromolecules), in contrast to existing microfluidic WB strategies that rely on a sandwich immunoassay format for analyte immobilization and detection. Sandwich approaches limit analyte immobilization to capture of only a priori known targets. In contrast to the electrostatic immobilization of positively charged protein- detergent complex reported earlier in our group1, a design for immobilizing more widely used negatively charged SDS-protein complex is developed here by positively charged poly-lyisne. Unlike previously approach using β-galactosidase as a negatively charged immobilizing agent, poly-lysine provides control over size and charge in more defined manner. We demonstrate pan-analyte immobilization of sized SDS-laden model proteins (protein G, bovine serum albumin, phosphorylase b) on the EIG with high capture efficiencies. Target proteins fixed on the EIG (protein G) are detected using antibody probes. The dependence of protein capture efficiency on both the concentration of copolymerized charges and poly-lysine length gives important insight of electrostatic immobilization mechanism. The approach advances microfluidic protein immunoblotting which can be directly correlated to the conventional SDS-PAGE based slab-gel WB.
1. Dohyun Kim, Kelly Karns, Samuel Q. Tia, Mei He, and Amy E. Herr, Analytical Chemistry,2012, 84 (5) 2533-2540
A fully integrated microfluidic device for tumor tissue dissociation into single cells
Janice De Jesus, Elliot Hui, and Jered B. Haun
UC Irvine
Fine needle aspiration (FNA) is a method of tissue biopsy that is rapidly gaining in frequency for the diagnosis of solid tumors because it is far less invasive than traditional cores. However, the small amount of tissue obtained by FNA has limited clinical use to histological characterization. To facilitate the use of FNA for molecular diagnostic platforms such as flow cytometry and advanced miniaturized detection devices, a controlled and reliable method to disrupt the tissue into single cells would be advantageous and improve clinical utility. We have developed and fabricated a microfluidic platform to achieve these goals. This device includes a dissociation chamber designed with a series of flow constrictions and curved features to mechanically dissociate tissue into single cells but avoid cell damage. The device features gradually decrease in size from that of the needle canula (2000 μm) down to 100 μm throughout the three‐layer device (see Figure). The device was fabricated through ALine Inc. whose technology uses lasers to design plastic sheets, adhesive to bond the layers, and the integration of pumps and valves to create a fully integrated chip. After loading the tissue, integrated pneumatic pumps and valves will control the flow of fluid through the dissociation chamber and back again for multiple passes to more effectively disrupt the tissue. Finally, a membrane filter with pores sized to allow the passage of single cells is included at the outlet. We plan to test the device using cancer cell line models before validating performance with FNA from solid tumors. This fully integrated system should significantly improve cell yield and sample quality as well as increase sample throughput by means of parallel processing, and thus help enable the application of FNA to molecular diagnostics in the clinic.
High‐throughput, Ultrafast Solution Exchange Around Cells and Particles via Inertial Microfluidics
Jaideep S. Dudani, Daniel R. Gossett, Henry Tat Kwong Tse, Keisuke Goda, Travis A. Woods, Steven W. Graves, Dino Di Carlo
UC Los Angeles
A passive system that can achieve precise control of fluids around particles would allow automation of various biological and chemical operations for single‐cell analysis, biophysics research, and solid‐phase chemistry. Utilizing passive inertial lift forces, we have developed a rapid inertial solution exchange (RInSE) system that precisely transfers cells and particles across fluid streams and positions them for downstream analysis. Importantly, the dominant force leading to particle transfer acts parallel to the particle rotation direction ensuring minimal disturbance of fluid interfaces. The technique operates at rates larger than 1000 particles or cells per second with sub‐millisecond transfer times. Solution exchange is rapid and efficient with 100% solution purity for a single solution exchange. In cell biology, assays often require solution exchange steps to lower background noise for analysis. We demonstrate the utility of RInSE for preparation of hematological samples, as well as numerous other cell‐based assays, for automating sample preparation. Fluorescently labeled leukocytes isolated from other blood components via RInSE showed a 70% reduction in background noise. Further, we illustrate improvements to inline flow cytometry after RInSE of excess fluorescent dye. Signal to noise ratio (SNR) was improved by a factor of 1.8 after solution exchange. Lastly, we introduce a system with two solution exchange steps, with a pinched center stream. This enables particles to have a very short residence time in the pinched stream (on the order of a few milliseconds). The final solution is greater than 99.9% pure. This short incubation time can enable short time scales for solid‐phase chemistry, extending the dynamic range of kinetic studies for flow cytometry, and identification of low‐affinity cell surface binders. This system is able to simply manipulate cells and particles for broad applications in cell and molecular biology.
From Bench to Bedside: Realizing On-Chip Electrophoretic Immunoassays for Protein Biomarkers Using a Standard 9V Battery
Todd A. Duncombe and Amy E. Herr
UC Berkeley
While significant recent attention has been paid to point-of-care (POC) screening diagnostics, innovation in confirmatory diagnostics has lagged. Critically important to infectious disease diagnosis (HIV, hepatitis C), confirmation of a positive screening result requires robust quantitation of biomarkers. Analytical grade performance – not simply the ‘yes/no’ screening readout (lateral flow assays) – is a hallmark of confirmatory tests. To this end, powerful bench-top electrophoretic immunoassays routinely provide robust, quantitative results that directly inform patient treatment. Nevertheless, a major technological gap exists. Electrophoresis drives species to differentially electromigrate through a sieving matrix, thus yielding protein specific identifying information – but thousands of volts are needed to drive separations over centimeters of separation length. Consequently, the resources (power, time, material, labor) demanded by electrophoresis severely limit POC applicability
To move electrophoretic immunoassays from the bench to the POC, we report the first, to our knowledge, confirmatory grade electrophoretic assay designed for use with a standard 9V battery. Key to our design strategy is an immunoassay that requires only a single, ultra-short (300 um) separation channel with 2 terminal fluid reservoirs to inject a moving boundary (front of material). In contrast, conventional approaches define a ‘plug’ (not a front) of material in the separation channel and, thus, require 2 intersecting channels (mm-to-cm in length) and 4 ports. Owing to the millimeters of separation distance demanded by on- chip electrophoretic assays, applied potentials of ~103 V are routine. The striking reduction in power realized here stems from rational device, separation matrix, and assay design. A standard battery technology is now feasible in lieu of high voltage supplies. By surmounting power and time shortcomings, we make electrophoretic immunoassays feasible at the POC for the first time. This versatile diagnostic format paves the way for analytical quality in near-patient, emergency, and global health settings including compatibility with cellular phone technologies.
Electrochemical Detection of Hydrogen Peroxide Release from Alcohol‐Injured Hepatocytes
James Enomoto, Zimple Matharu, Alexander Revzin
UC Davis
Alcohol metabolism by the liver damages hepatocytes, setting off a complex sequence of inflammatory and fibrogenic responses. There is increasing evidence that hepatocytes participate in pathophysiological signaling through the production of reactive oxygen species (ROS), important indicators of inflammation associated with alcohol consumption. The present work describes the fabrication of an enzyme‐ based biosensor for the electrochemical detection of H2O2, one of the major ROS released from hepatocytes. Our biosensor is comprised of an array of micropatterned Au electrodes prepared through standard photolithography and metal etching methods. A pre‐polymer hydrogel solution consisting of PEG‐diacrylate (DA), horseradish peroxidase (HRP) and glutaraldehyde is coated onto the surface and exposed to UV radiation through an aligned photo mask. Collagen is then adsorbed around the HRP‐PEG covered electrodes to promote hepatocyte attachment. The PEG‐HRP patterned slide is then integrated with a PDMS microfluidic device and connected to a potentiostat for electrochemical measurements.
The sensor showed a detection limit of 1 μM when challenged with varying concentrations of H2O2. In cell experiments, hepatocytes were stimulated by infusing ethanol (100mM) into the microfluidic device and were analyzed through cyclic voltammetry. The reduction current increased after about 60 min of ethanol incubation, indicating the production of H2O2. The increase in reductive current was reduced by the pre‐ treatment of hepatocytes with various anti‐oxidants. When coupled with primary hepatocytes exposed to alcohol, this biosensor allowed for the continuous monitoring of extracellular peroxide that ranged from 4 μM at 60 min to 12 μM after 3 hrs. This study is one of the first to focus on extracellular hepatic oxidative stress generated during alcohol injury. Our findings demonstrate that hepatocytes represent an important source of oxidative stress during liver injury and that these cells may possibly act as instigators of inflammatory signaling in the liver.
Isolation of Rare Circulating Tumor Cells From Blood Using Microscale Fluid Vortices
Derek E. Go, Albert J. Mach, Elodie Sollier, Xu Yi, Ish Talati, James Che, Jonathan Goldman, Rajan Kulkarni, and Dino Di Carlo
UC Los Angeles
Blood-based isolation of potentially malignant cells can be used to obtain information concerning cancer relapse or treatment effectiveness at low-cost, and without radiation-intensive imaging. However, malignant cells can be extremely rare; with only one of these cells is 500 million and 10 million red and white blood cells, respectively. Current tumor cell isolation technologies have shortcomings with either limited throughput, cell purity, or post-processing retrieval of viable cells. Here, we present a microfluidic system, the Centrifuge-on-a-Chip, that makes use of size-based trapping in microscale vortices that occurs due to fluid dynamic shear gradient lift forces. A massively parallel device design and an automated pressure driven system allows high-throughput processing and concentration of cell samples followed by controlled release. Processing patient samples through our device permits retrieval of larger circulating tumor cells that selectively enter and are maintained in vortices while smaller contaminating blood components are filtered out. The larger isolated cells are then released in a small liquid volume that can easily be manipulated to perform fluorescent labeling with intra- and extra-cellular antigens for downstream imaging and cell classification, or because of our high purity, molecular diagnostics. To test clinical applicability, thirteen patient samples to date, across a range of cancer types, were processed and in 4/13 putative circulating tumor cells were successfully isolated and detected. The Centrifuge-on-a-Chip has the potential to be integrated into the clinic as a sample preparation tool aiding cytopathologists in obtaining diagnostically useful cells for relapse detection and treatment monitoring.
A Point-of-care Microfluidic Magnetic Particle Flow Cytometer
Igor Izyumin, Prof. Bernhard E. Boser
UC Berkeley
Flow cytometers are powerful laboratory instruments that allow the rapid measurement of multiple parameters with very high throughput; however, they are also large, expensive, and difficult to operate. Numerous obstacles prevent flow cytometers from being successfully implemented in a low-cost point-of- care format. Optical filters and detectors have proven difficult to miniaturize. More critically, the complex sample-dependent calibration process required to eliminate the effects of autofluorescence and fluorescent label variability prevents use of the technology outside a laboratory setting. Magnetic nanoparticle labels offer a solution to these problems. Unlike fluorescent labels, their properties are not affected by storage conditions or chemical interactions, and there is no sample background; many of the calibration steps can therefore be eliminated. In addition, the required sensors and readout electronics can be incorporated into a low-cost integrated circuit, dramatically reducing size and cost and allowing the flow cytometer to be realized in a single-use point-of-care format.
Unfortunately, existing magnetic label-based systems cannot discriminatemultiple labels and are thus limited to single-color experiments. The proposed work eliminates this limitationby exploiting the magnetization dynamics of superparamagnetic nanoparticles. A time-varying magnetic polarization field allows different types of beads to be distinguished by measuring their Neel relaxation time. Preliminary experimental work has been performed with 4.5 and 12 micron diameter polystyrene beads with embedded magnetic nanoparticles as substitutes for labeled cells. Currently, the system is capable of counting up to 100 beads per second while distinguishing between two different labels.
Kinetic Polyacrylamide Gel Electrophoresis: Microfluidic Binding Assay Enables Measurements of Kinetic Rates
Monica A. Kapil and Amy E. Herr, Ph,D.
UC Berkeley
We present an inexpensive, rapid method for the determination of rate constants of complex formation, Kon, and dissociation, koff directly without the use of invasive chemicals for surface immobilization. This method is termed Kinetic Polyacrylamide Gel Electrophoresis (K-PAGE). In this multistep assay, labeled target species are electrophoretically injected into microfluidic immunofilter with varying electromigration velocities (interaction times). Species having an affinity to the immobilized antibody form complex (Fig1 A). Kon is then determined via a simple exponential fit of complex formation with varying interaction times (Fig1 B). A subsequent buffer wash was then introduced to the complex in order to stop further association and Koff is determined (Fig1 C). In order to optimize operational parameters of this assay and maximize the analytical sensitivity, the interplay between competing transport processes such as diffusion, convection, and reaction, a analytical model captures electromigration through a porous gel, diffusion, and antibody/antigen reaction kinetics was developed. The K-PAGE assay was applied in a study of Prostate-specific antigen (PSA), a 30 kDa protein that is used as a widespread cancer marker, both for initial diagnosis and for monitoring of response. As PSA screening has become more widely used, it has generated increasing controversy, as there has been limited success and no reduction in prostate cancer mortality. Consequently, there is need for progress in protein-‐based diagnostics and methods for scouting out better binders of cancer induced proteins like PSA for drug screening, the selection of antibodies for clinical and research applications such as disease diagnostics(e.g.,immunoassays), novel therapeutics for cancer treatments as well as vaccine development.
Cationic Detergent Microfluidic Western blotting integrated with Electrostatic Protein blotting
Dohyun Kim,_Kelly Karns, Samuel Q. Tia, Mei He, and Amy E. Herr
UC Berkeley
Western blotting (WB) is a powerful and ubiquitous protein analysis technique used in many fields spanning from biology to biomedicine. However, WB is a bench-top wet technique with apparent drawbacks: copious sample consumption (1-40 μg), slow assay (1-2 days), and manual operation. In an effort to address these issues, for the first time we report fully-functional microfluidic WB that seamlessly integrate 1) cationic- detergent polyacrylamide gel electrophoresis (PAGE), 2) novel polyacrylamide protein immobilization matrix, and 3) antibody-based immunodetection of target proteins. In a central microchamber of a glass microfluidic chip, three contiguous polyacrylamide gel regions, loading, separation, and electrostatic immobilization gel (EIG), are photopatterned to realize integration of three consecutive WB assay steps: PAGE separation, protein blotting, and immunoprobing. Electrophoretic sample control and microfluidic electric field shaping strategy allow a compact sample plug to be injected and manipulated with minimal band broadening in the microchamber. Using novel cationic-detergent (cetyltrimethylammonium bromide) PAGE, proteins are separated and molecular mass is accurately determined. After protein sizing, resolved proteins are electrophoretically transferred and strongly immobilized on the EIG region. In contrast to existing microfluidic WB methods, all the separated bands are immobilized. Positively charged CTAB-protein complexes are immobilized owing to electrostatic interaction with charged polyacrylamide gel copolymerized with negative capture moiety β-galactosidase. After immobilization, antibody probes are injected to the EIG to detect protein targets. In this presentation, assay performance is thoroughly characterized and protein capture mechanism is elucidated. Assay performance is tremendously improved over conventional WB: reduced sample consumption (~10 ng), rapid assay completion (~2.5 hours), and automated assay control. Using the microfluidic WB assay, we assessed lactoferrin in human tear fluids with a goal of non-biopsy diagnosis of Sjögren’s Syndrome, an autoimmune disease. The automated, high-speed, multi-stage microfluidic WB will form a versatile platform technology for advancing ‘high throughput’ proteomics.
Low Cost MR Compatible Neural Prosthesis
Sung June Kim*, Sung Eun Lee, Joon Soo Jeong, Kyu Sik Min, Jin Ho Kim
(all of Seoul National University, * on sabbatical leave visit to UC Berkeley Bioengineering in 2012)
UC Berkeley
The price of a cochlear implant device, is around $25,000 while about 80% of world’s 120 million hearing impaired people live in developing countries where the device is not or rarely affordable. Low cost but effective neural prosthesis is needed. A costly component of the device is the hermetic packaging. Typically IC’s are encapsulated using medical grade titanium where sealing is considered hermetic and good protection against sudden impact is expected. However, the Ti package is often bulky, involves complicated manufacturing process, and causes MR image artifact problem, On the other hand, biocompatible polymers can offer compact packaging, RF transparency, and simpler manufacturing process. Still, widely used biocompatible polymers such as polyimide and paralene-C, have shown limited life time in implanted conditions due to high water absorption rate. As an alternative, liquid crystal polymer (LCP) material is used in this study. This thermoplastic material has extremely small water vapor transmission.. In accelerated soak test at 75 C PBS, the mean time to failure is recorded to be 380 days while that of polyimide is 75 days. The retina electrode made using LCP, implanted in rabbit eye and explanted after two years, came out intact showing no sign of degradation. Fabrication steps have been developed for patterning and insulation of microelectrode arrays, and electronics packaging. These methods enabled monolithic integration of the neural interface and the packaging, removing the need for the yield limiting feed-through connection between the two, which is also the major source for long-term failure of implants. Several device prototypes have been developed for a depth type microelectrode array, a retinal implant with eye surface conforming structure, and a cochlear implant (Figure 1). T1-weighted 3.0 T MR image taken using the LCP-based cochlear implant showed very little distortion (un-measurable) while the artifact from a titanium based counterpart was severe (69 mm).
High-density Thermo-compression-bonded flex-cables for Neural Nanoprobes
P. Ledochowitsch, M. M. Maharbiz, T. J. Blanche
UC Berkeley
This paper reports on the design, microfabrication and testing of ultra-high density dual-layer Parylene cables for interfacing with next-generation silicon neural probes (‘nanoprobes’ [1]). This work presents a significant improvement over the fabrication process we reported last year for μECoG arrays [2]. The trace/space width was reduced 3-fold from 15 μm to 5 μm and continuous metal traces with a length-width-ratio of up to 13,000 were successfully patterned by lift-off. We characterized the cables with electrochemical impedance spectroscopy (EIS) between 1 Hz and 5 kHz using a 64- channel impedance analyzer (nanoZ, White Matter LLC) in phosphate-buffered saline (PBS).
Neurophysiologists are struggling to scale up the number of simultaneously recorded neurons while minimizing the impact of probe implantation on brain function. Avoiding tissue damage is of particular importance for viable chronic recordings [3]. Miniaturizing the cable minimizes tethering forces, allows multiple nanoprobes to be implanted in adjacent brain regions, and decreases the damage associated with deep brain implants. Our cables compare favorably to state of the art on Parylene flex cables for neural probes [4] in terms of length, trace density and total number of channels. The cable width is one order of magnitude below that of commercially available devices [5]. It matches the widest dimension of the nanoprobe, the bond pad array, to achieve ultra-compliant, ‘free-floating’ cortical implants and to avoid displacing more than the bare minimum of tissue for sub-cortical implantations.
This work constitutes an important milestone towards the goal of making distributed neuronal recordings in both superficial and deep brain structures.
[1] J. Du and S.C. Masmanidis, PLoS ONE (in press).
[2] P. Ledochowitsch and M.M. Maharbiz, MEMS. IEEE; 2011:1031-1034.
[3] M. P. and P. P. Irazoqui, Brain research. 2009;1282:183-200.
[4] C. Pang and R.A. Andersen, Engineering In Medicine And Biology. 2005:4-7.
[5] Neuronexus H64 hybrid flex-cable (https://neuronexustech.com).
Shrink-Induced Superhydrophobic and Hydrophilic Microfluidic Devices
Jolie McLane, Lauren Freschauf, Sophia Lin, Michelle Khine
UC Irvine
Point-of-care diagnostic (POC) devices should be inexpensive, quick, and simple without using external equipment so diseases can be detected immediately. By leveraging our superhydrophobic plastic, we obviate the need for pressure driven flow and external equipment with a functional POC device.1 Integrating a glass-bottom microchannel with the superhydrophobic surface, fluid can wick the glass channel by capillary action within seconds. Capture antibodies can be selectively deposited on-chip during device fabrication. When the target is introduced, antigens selectively bind to the capture antibodies. Secondary antibodies and enzymes will then yield fluorescence detection. This device is not limited to specific antibodies and has potential to perform immunoassays for infectious disease as well as food-borne pathogens.2
The device uses commercially available hard plastics, adhesive, and glass so that device fabrication is simple and inexpensive. The key aspect of the device is the hydrophilic channel surrounded by a superhydrophobic barrier (Figure 1). A superhydrophobic surface repels water because it is more energetically favorable for the water to bead up and roll off the surface.3-6 When fluid contacts the surface of the device, it rolls off the superhydrophobic region into the channel, so no reagents are lost from nonspecific binding. The fluid then quickly wets the hydrophilic glass channel to perform an immunoassay.7
Overall, the device is robust, inexpensive to manufacture, requires no external pumps, yields fast results, and can have a simple readout for the user. This device is therefore ideal for POC and has potential as a fabrication method for many biological assays.
A Micro-Drive Hearing Aid: A Non-Invasive Hearing Prosthesis
Peyton Paulick, Hossein Mahboubi, Hamid Djalilian, Mark Bachman
UC Irvine
The direct hearing device (DHD) is a new auditory prosthesis that combines conventional hearing aid and middle ear implant technologies into a single device. Both conventional hearing aids and middle ear implants have their shortcomings. For conventional hearing aids, these shortcomings include visibility and sound quality issues, such as feedback and the occlusion effect. Middle ear implants require invasive surgery, are not easily reversible, and the performance of the implant is not known until after implantation. The DHD is located deep in the ear canal and recreates sounds with mechanical movements of the tympanic membrane. A critical component of the DHD is the actuator, which must be capable of moving the tympanic membrane at frequencies and magnitudes appropriate for normal hearing, with little distortion, and with minimal acoustic noise generation. The DHD actuator reported here utilized a voice coil actuator design and is 3.7 mm in diameter. Its measured frequency response is similar to a tympanic membrane during normal hearing, total harmonic distortion between 425 Hz and 10 kHz is below 0.5%, and acoustic noise generation is minimal. The DHD has been validated as a tympanic membrane driver on cadaveric temporal bones where the device was coupled to the umbo of the tympanic membrane. The DHD successfully recreates ossicular chain movements across the frequencies of human hearing while demonstrating controllable magnitude.
High Surface Area Electrodes for Easy Integration into Polymer Microfluidics
Jonathan Pegan, Mark Bachman, Michelle Khine
UC Irvine
Integration of electronics into microfluidics offers a valuable detection mode providing high sensitivity, portability, low-power requirements, and less reliance on external equipment as compared to other modes. Many current devices that utilize electronic components are manufactured using planar techniques to deposit metal onto glass or Si substrates [1-3]. To make a complete Lab-on-a-chip (LOC) device a fluidic network must be carefully aligned with the electronic component. This is increasingly challenging at lower length scales.
This work presents a novel fabrication method for producing high surface area metal structures, patterned on polyolefin (PO) films. These complex 3D electrodes can be more easily integrated into microfluidics utilizing a pattern then shrink process previously reported by our lab [4,5]. While the use of shape memory polymers to create complex 3D structures has been previously reported [6,7], this is the first demonstration of creating complex metallic structures for applications as electrical components.
The Au thinfilm buckles during compression of the PO film support in the thermal shrinking process. This buckling creates complex, 3D wrinkled structures (fig. 1). Leveraging a 20x miniaturization of patterned features by shrinking the PO film, 2μm resolution can be achieved. Kelvin resistance measurements show a reduction of the Au thinfilm resistivity from 7.5×10-8 Ωm to 2.5×10-8 Ωm after shrinking, similar to that of bulk gold (2.4×10-8 Ωm).
This method can be integrated into the fabrication of complete LOC devices previously reported by our lab. By patterning electrodes at a larger length scale it is easier to align them properly with a microfluidic network. The electrical and fluidic components can be thermally shrunk together to bond the two layers. The resulting LOC device is fabricated completely out of robust polymeric materials as opposed to many current integrated devices made from a combination of polymeric materials and fragile glass or Si substrates.
Digital Readout Platform for Water-in-oil Droplet Immunoassays Running on a Cell-Phone for Point of Care Viral Load Sensing
Patrick A. Sandoz, Ahmet F. Coskun, Aram J. Chung, Westbrook M. Weaver, Oladunni Adeyiga, Delaram Khodadadi, Aydogan Ozcan, Dino Di Carlo
UC Los Angeles
Here we report a new technique to improve signal fidelity for implementing on-chip digital immunoassays running on a cell-phone. Unlike sensing by digital PCR in which the two copies of viral nucleic acids are amplified, our platform targets viral capsid protein for a microparticle-based sandwich enzyme-linked immunosorbent assay (ELISA) performed on chip. Digitization is achieved by encapsulating the microparticles in water-in-oil droplets. Target protein capture is reported from each droplet with protein-bound microparticles by the horseradish peroxidase (HRP) enzyme conjugated to the secondary antibody. In the presence of fluorogenic substrate, HRP produces a strong fluorescent signal.
Readout fluorophores are generally amphipathic, resulting in poor signal stability with water-in-oil digital systems. Indeed, this phenomenon is observed with our system in the presence of surfactant, and seems to result from an interaction of the fluorophore with the surfactant at the droplet interface. We have overcome this previously fundamental limitation with the addition of sucrose in the aqueous phase, significantly conserving the signal stability. The fluorophore leakage was quantified on-chip from the droplet signal intensities in chambers with different sucrose dilutions to define a relevant concentration. This technique significantly improves signal fidelity for generating and imaging droplets.
Finally, we increase the dynamic range of concentrations that can be detected by imaging over a large field of view using a variant of cell-phone based fluorescent imaging adapted from previous work. This cost effective and portable imaging device has a resolution of ~10μm, a wide field-of-view of 1cm2, a low cost lens and color plastic filters enables future multiplexing possibilities. The imager is directly mounted on a cell- phone camera or on a webcam and is well adapted for POC testing or telemedicine applications.
Infrared Neural Stimulation via a Highly General Capacitive Mechanism
Shapiro MG, Homma K, Villarreal S, Richter CP, Bezanilla F.
UC Berkeley
Optical stimulation has enabled important advances in the study of brain function and other biological processes, and holds promise for medical applications ranging from hearing restoration to cardiac pace making. In particular, pulsed laser stimulation using infrared (IR) wavelengths > 1.5μm has therapeutic potential based on its ability to directly stimulate nerves and muscles without any genetic or chemical pre-treatment. However, the mechanism of IR stimulation has been a mystery, hindering its path to the clinic.
We recently discovered that IR light excites cells through a novel, highly general electrostatic mechanism (Shapiro et al, 2012). IR pulses are absorbed by water, producing a rapid local increase in temperature. This heating reversibly alters the electrical capacitance of the plasma membrane, depolarizing the target cell. This mechanism is fully reversible and requires only the most basic properties of cell membranes. Here we will review our findings, provide further insights on the mathematical model explaining the capacitive mechanism, and discuss the implications of our work for the design of implantable neural prosthetic devices using IR neural stimulation.
Reference: Shapiro MG, Homma K, Villarreal S, Richter CP, Bezanilla F. Infrared light excites cells by changing their electrical capacitance. Nature Communications 3:736 (2012).
Fluorescence Enhancement with Gold Nanostructures Generated on a Thermoplastic
Himanshu Sharma, Sophia Lin, Avina Gupta, Michelle Khine
UC Irvine
Fluorescence is widely used in biomedical diagnostics and environmental monitoring due to its sensitivity and specificity, but significant amplification of the fluorescence signal relative to the background noise remains a problem. Recent work has demonstrated that localized electromagnetic field enhancements generated at sharp tip gold (Au) structures and nano-scale gaps between Au structures are a source for stronger fluorescence enhancements[1]. We have developed a simple approach to create Au nanostructures onto a thermoplastic polymer for enhancing molecular fluorescence. A thin nickel (Ni) film followed by Au film is first deposited onto the thermoplastic substrate. Upon heating, the thermoplastic substrate retracts while the Au film does not, which makes the Au film buckle into nanostructures with high density nanogaps between the nanostructures (Figure 1A). 5 nm of silica (SiO2) is sputtered on top of the Au structures before an antibody conjugated to a fluorophore, Alexa Fluor 555 goat anti-rabbit IgG (A555), is adsorbed on the Au structures. The silica acts as a thin spacer between the Au structures and the fluorophore which has been found to be important for obtaining maximum fluorescence enhancements. It was determined that the strong surface plasmons generated by these Au nanostructures enhance the fluorescent signal by more than 30 fold, relative to the signal observed without the Au nanostructures (Figure 1B). The emission spectra also indicate that the bright fluorescent spots are specific to A555 (Figure 1C). Relative to a planar surface, these Au structures have an increased surface area to volume ratio which can result in increased proteins binding to the surface. In the future, lower concentrations of A555 will be tested to determine the actual LOD. The tunable Au nanostructures will also be integrated into a portable microfluidic device to serve as a relatively low-cost and powerful technique for fabricating highly sensitive microfluidic sensors.
Rapid Microfluidic Western Blotting of Multiple Native Analytes
Samuel Tia, Mei He, Dohyun Kim & Amy E. Herr
UC Irvine
Microfluidic “lab on a chip” devices hold great promise for the integration of disparate analytical and sample handling processes that can be difficult to integrate at the macroscale. For example, Western blotting of multiple proteins is a frequently employed procedure which can multiply the time and labor requirements of the immunoblot process. One of the most common approaches to multi-analyte blotting involves the application of heat and harsh detergents to strip previously-bound antibodies away from the blotting membrane, allowing for new antibodies to be incubated and detected (“reblotting”). These harsh stripping processes result in analyte losses which render quantitative comparison impossible and so the cycle can only be repeated for a limited number of protein targets. In response to these challenges, we present a microfluidic chip for performing PAGE separation, blotting transfer and immunodetection of multiple native proteins within minutes – a time savings of several orders of magnitude when compared against conventional reblot techniques. The work presented here significantly advances on our previously reported blotting assays (He, Herr, JACS, 2010) by demonstrating separation and immunoblot of multiple targets within a single step, as well as a method for visualization of unlabeled target molecules. Gapless integration of multiple functions is made possible through a novel photolithographic technique which polymerizes gel structures of different physical and functional properties at high spatial resolution within a ~1 x 1.5 mm central chamber. The capture and detection mechanism is analogous to a sandwich assay, enabling high capture efficiencies (>85%). Langmuir model simulation is applied to system binding kinetics and shown to enable rational device design. This multi-immunoblotting device is especially useful for studies of isoforms and post-translational modifications. In particular, studies of acetylation are discussed with a specific emphasis on aging and regeneration through ongoing work with our collaborators (D. Chen, UC Berkeley).
Combining Antibody Arrays with Reconfigurable Microfluidics to Enhance Sensitivity of Cell-Secreted CytokineDetection
Tam Vu, Arnold Chen, Gulnaz Stybayeva, Tingrui Pan, Alexander Revzin
UC Davis
During a viral or bacterial infection, CD4 and CD8 T-cells secrete cytokines such as interferon (IFN) gamma and tumor necrosis factor (TNF) alpha to orchestrate an immune response. Detection of these cytokines can provide valuable diagnostic data but requires a large sample volume, is expensive, and is time-consuming with traditional techniques such as flow cytometry or enzyme-linked immunospot assay. Our lab has previously designed lab-on-a-chip devices capable of capturing CD4 and CD8 T-cells and detecting released cytokines from a small volume of minimally processed blood. This presentation will describe an approach whereby antibody arrays for capturing cells and secreted cytokines are integrated into a reconfigurable microfluidic device that helps further reduce volume of the channel to concentrate the released cytokines, amplifying the signal and reducing the time to detection. A reconfigurable microfluidic device was mounted onto a hydrogel-coated substrate imprinted with cell and cytokine-specific antibody spots. T-cells in PBMC isolated blood were flown through the device, captured, and stimulated with a mitogenic solution. After cell capture, the microfluidic device was reconfigured to lower pL volume (8 mm3 channel to six 0.03 mm3 grids) cups over the cells and adjacent cytokine detection spots. IFN-gamma and TNF-alpha molecules captured on these detection spots were immunostained and then fluorescently scanned to measure the captured cytokines concentration. The collapsed device detected about 5 times more cytokine signal than a device that was not collapsed. In the future, this microfluidic device may have clinical applications in detecting leukocyte subset-specific cytokine signatures in the monitoring of infectious diseases and immune competency.
Imaging
Ultrasensitive Detection of Tuberculosis and Fungi Pathogens in Clinical Samples Using Modified Commercial ELISA with 4 Orders Higher Sensitivity
Abhinav P Acharya, Xuli Feng, Kousik Kundu, Ben Kline, Aaron Whiteley, Dan Portnoy, Niren Murthy
UC Berkeley
Introduction: Infectious diseases cause over 10 million deaths every year with a threat from certain pathogens to evolve into epidemic and pandemics. The areas of pathogen detection in the food and water sources and early detection of these pathogens in biological fluids have been studied extensively. However, there is a lack of technology that can accurately detect pathogens non invasively at an early stage of pathogenesis. In this report we demonstrate that the detection sensitivity of commercial ELISA kits can be increased by four to five orders of magnitude, using a new horse radish peroxidase (HRP) substrate composed of deuterocyanines coupled to a gold nanorod (D-Au). Results and Discussion: D-Au was generated by reducing gold conjugated cyanine dye via a linker. D-Au provided 20 picomolar sensitivity to H2O2 in Fenton’s reagent. Next, reactive oxygen species generated by HRP in the presence of H2O2 oxidized D-Au and provided 56 attomolar detection sensitivity of HRP as measured by a generic plate reader. HRP is the most commonly used reporter enzyme in ELISA, and hence D-Au is an excellent novel substrate for ELISA. We utilized D-Au in conjunction with commercially available ELISA kits to improve the detection sensitivity of cytokines such as IL-10, IL-12p40 and IL-1 by 4-5 orders of magnitude. Further, we demonstrate that D-Au based ELISAs are able to detect bacterial muscle infections in mice composed of a 1000 bacteria, via ELISA of the blood for leaked antigens. Next, infections induced by orally gavaged and intravenous injections of listeria monocytogenes could be detected after 3 days of infection in the stool samples of mice. Further, this is the first instance where an ELISA for detecting lipoarabionomannan an antigen produced by tuberculosis (TB) mycobacterium in saline and serum with high sensitivity was developed. This new ELISA technique is currently being utilized to evaluate antigen concentration in human bronchoalveolar lavage fluid for the presence of TB pathogen.
Detection of Radiation Therapy Induced Cerebral Microbleeds in Gliomas: Does High Field Mean High Yield?
Wei Bian1,2, Christopher P. Hess2, Susan M. Chan3, Sarah J. Nelson1,2,4, Janine M. Lupo2
UC San Francisco
BACKGROUND: Cerebral microbleeds have been detected in the brain of patients with gliomas after receiving radiation therapy. Although studies have shown improved sensitivity to microbleeds on MR magnitude images or via using susceptibility‐weighted imaging (SWI) at 3T and 7T field strengths compared to 1.5 T, it is not clear how much sensitivity is gained with 7T over 3T for microbleed detection.
OBJECTIVE: The purpose of this study was to compare radiation‐induced microbleed detection between 3T and 7T and determine whether 7T SWI detect more radiation‐induced microbleeds than 3T SWI.
SUBJECTS and METHODS: Ten patients with gliomas who had radiation therapy were scanned at both 3T and 7T GE scanners on the same day. Both acquisitions used a 3D SPGR sequence with GRAPPA‐based parallel imaging. Acquisition parameters were as following: GRAPPA R = 2 (3T) or 3 (7T), TR/TE = 28/56ms (3T) or 16ms/50ms (7T), Flip angle 20°, FOV 24cm, 0.5 x 0.5 x 2 mm resolution, total scan time around 6 min. Minimum intensity projection images through 8 mm‐thick slabs were generated from both magnitude and SWI images. Microbleeds were identified as small circular hypointense foci on the projected images. Wilcoxon signed rank test was performed to test whether there was a significant difference in microbleed detection between groups.
RESULTS: No group differences in microbleed detection were found between 7T SWI and 3T SWI when including all 10 patients. However, a significantly more microbleeds were detected on 7T SWI after excluding 3 patients who had microbleeds surrounding air‐tissue interfaces.
CONCLUSION:S 7T SWI is more sensitive to radiation‐induced cerebral microbleeds than SWI at 3T as long as the location of microbleeds is not in areas with heightened susceptibility artifacts. To achieve an optimal microbleed detection rate, tumor location should be considered in conjunction with field strength when managing these patients.
Characterizing anatomical variability in breast images: NPS slope “beta” comparisons between breast CT, Tomosynthesis, and Mammography
Chen L, Abbey CK, Nosrateih A, Lindfors KK, and Boone JM
UC Davis
In breast imaging, tumor detection requires identification of abnormalities against a background of normal tissues. Thus, the anatomical fluctuations of normal tissues in the breast can be treated as “noise”, usually referred as “structured” or “anatomical” noise. Previous researches showed that the anatomical background plays an important role in diagnosis, even outweighing other noise components such as quantum noise.
It has been shown the natural structure background of human breasts represented in breast images has the power spectra could be described by a power-law at low frequencies: formula missing, where the ƒ here is the spatial frequency, α and β here are constant scalars. Also, Burgess et al investigated the anatomical influence in mammographic lesion detection, and pointed out that the power-law exponent β plays an important role to detect tumors as the smaller the β is, the easier the tumor- detection would be.
The study focused on characterization of anatomical noise variability from images obtained by three modalities: Breast CT, Breast Tomosynthesis & Mammography. Patients (BIRADS 4 and 5) underwent IRB-approved imaging by mammography, breast tomosynthesis, and breast CT on the same day. The 2D noise power spectrum (NPS) was computed, and the value of β was determined for all the three modalities. According to recent findings, it appears that β is fundamentally a function of the thickness of the plane which cuts through the breast. What we find interesting is that for tomosynthesis, the value of β for the thin slice tomosynthesis images is very similar to that of mammography (3.0 versus 3.2), as significantly different to that of breast CT (~1.8). This study suggests, based on the metric β, that there might be a fundamental difference between breast CT and mammography, and that breast tomosynthesis has similar anatomical noise as mammography.
Integration of Lensfree On-Chip Fluorescent Imaging with Lens-based Microscopy toward High- throughput Rare-cell Analysis
Ahmet F. Coskun, Danny R. Gossett, Albert J. Mach, David Herman, YeongSeok Suh, Dino Di Carlo, and Aydogan Ozcan
UC Los Angeles
Optical microscopy has been the workhorse for various fields including biomedical sciences [1]. Although optical microscopes revolutionized our probing capabilities at the micro- and nano-scales, they are still relatively low-throughput, requiring tedious mechanical scanning to monitor e.g., large-area microfluidic devices [2]. To provide an alternative solution to such high-throughput imaging needs, here we demonstrate an integrated microscopy platform that can rapidly determine the position of fluorescently labeled cells over a wide field-of-view (FOV) of e.g., 8 cm2 with ~ 4-10 μm spatial resolution using a wide-field lensfree microscopy platform [3,4] and then perform multi-parameter (bright-field and multi-color fluorescence) analysis of the selected region-of-interests (ROIs) employing a high-resolution (e.g., 10X-100X Objective) lens-based microscope (see Fig. 1). In this integrated platform, lensfree fluorescence imaging operates based on an opto-fluidic waveguide (i.e., PDMS, liquid core and glass bottom layers), where the fluorescent samples of interest within a micro-channel are excited through the side-facets of the micro-fluidic chip using butt-coupled LEDs. Upon uniformly guided excitation, fluorescence signal is then sampled through a faceplate [4,5] by a sensor-array (e.g. a CCD or CMOS chip). To demonstrate the feasibility of the integrated screening platform, fluorescently labeled breast cancer model MCF7 cells were imaged using the wide-field lensfree fluorescence imager. After the ROIs of the cells are defined based on the wide-field image, the stage is moved to selected ROI’s to perform multi-parameter microscopic analysis of these selected cells using a conventional lens-based high-resolution microscope. This dual-microscopy platform, merging both lensfree and lens-based imaging on the same chip, could be very useful to significantly reduce the number of frames that are acquired to monitor/probe large area microfluidic chips.
References: [1] S. Nagrath, et al., Nature, vol 450, 2007, p. 1235-1239
[2] S. L. Scott, et al., PNAS, vol 107, 2010, p. 18392–18397
[3] A.F. Coskun, T. Su, and A. Ozcan, Lab on a Chip, vol. 10, 2010, p. 824. [4] A.F. Coskun, I. Sencan, T. Su, and A. Ozcan, Optics Express, vol. 18, May. 2010, p. 10510-10523 [5] A.F. Coskun, I. Sencan, T. Su, and A. Ozcan, Analyst, 2011, vol 136, 2011, p. 3512-3518
Relaxation Effects in Magnetic Particle Imaging
Laura R. Croft, Patrick Goodwill, Steven Conolly
UC Berkeley
Magnetic particle imaging (MPI) is an emerging medical imaging modality capable of high‐sensitivity images with unprecedented contrast and without ionizing radiation [1]. MPI uses a strong magnetic field gradient to spatially localize the induction response of ultra-small superparamagnetic iron oxide nanoparticles (USPIOs), which are currently approved as a contrast agent for MRI. MPI relies on USPIO dipole moments aligning quickly with the applied magnetic field, but relaxation mechanisms can significantly retard this alignment. By causing a lag in magnetization alignment, relaxation ultimately degrades the resolution and accuracy of the MPI method. In this work, we update our laboratory’s one-dimensional x-space formulation [2,3] to allow for non-instantaneous USPIO magnetization. To validate our theoretical predictions, we built an x-space MPI relaxometer, which measures the USPIO diameter, relaxation time constant, and point spread function without an imaging gradient. We show that the inclusion of these relaxation effects is essential for theoretical predications to agree with experimental MPI data, and we demonstrate experimentally how relaxation adversely affects image quality. This knowledge will enable us to understand how to design MPI x-space scanning to mitigate the negative effects of relaxation and to achieve desirable image resolution, accuracy, and signal strength.
1. Gleich B, Weizenecker J (2005) Nature 435:1214-7.
2. Goodwill P, Conolly S (2010) IEEE Trans Med Imag 29:1851-9
3. Goodwill P, Conolly S (2011) IEEE Trans Med Imag 30: 1581-1590
Detection of Seizure Progression in vivo with Optical Coherence Tomography
Melissa M. Eberle, Carissa L. Reynolds, Jenny I. Szu, Yan Wang, Mike S. Hsu, Devin K. Binder, B. Hyle Park
UC Riverside
Seizures have very serious health consequences and impact millions of people around the world. The most common technology for seizure detection and monitoring is with electroencephalography (EEG). However, EEGs have low spatial resolution and minimal depth discrimination, making it difficult to map the progression of an electrical event through the brain. Optical techniques using near-infrared (NIR) light have been implemented in an attempt to improve upon EEG technology by decreasing noise and increasing spatial resolution [1, 2]. Due to the changes in optical properties of the tissue, a decrease in intensity of backscattered light is observed during the progression of seizures [2, 3]. We have demonstrated that with optical coherence tomography (OCT), a high resolution, minimally invasive, imaging technique, we can detect a significant decrease in intensity from the baseline during seizure progression in vivo in mice. We continuously imaged over the motor cortex region of the right hemisphere as we induced a general seizure through the use of pentylenetetrazol (PTZ). The brain tissue intensities were determined by averaging a region of interest (ROI) and were then plotted versus time. A decrease from baseline was observed before the physical manifestations of a full seizure. From these results, we are developing an optical trigger for pre- seizure detection, improving upon the timing and resolution of EEG technology.
[1] Lee H, et al., Jour. of the Korean Phy. Soc., 58:6 (2011). [2] Weber J, et al., Biomed. Opt. OSA CD, (2010). [3] Rajneesh KF, et al., Jour. Of Neuosur. Abs, 2:113 (2010).
Non-destructive evaluation of bioengineered tissue using time resolved fluorescence measurements, high frequency ultrasound and acoustic radiation force impulse imaging.
Hussain Fatakdawala, Donald Responte, Yang Sun, Kyriacos Athanasiou, Laura Marcu.
UC Davis
The success in developing a useful engineered tissue relies heavily on evaluating its mechanical and biochemical properties before and after implantation as well as during development. Traditional evaluation methods are a clear impediment to such a setting due to their destructive nature and inefficiency in time and cost. The overarching goal of this work is to develop a non-destructive evaluation tool for continuous monitoring of engineered tissue using time resolved fluorescence measurements, high frequency ultrasound (US) and acoustic radiation force impulse (ARFI) imaging. Numerous evaluation parameters are routinely monitored in tissue engineering. We aim to explore the possibility of quantifying (1) cell viability in cell culture using fluorescence lifetime imaging microscopy (FLIM) of cellular nicotine-amide adenine dinucleotide (NADH), (2) extracellular matrix (ECM) collagen content and cross link formation using time resolved fluorescence spectroscopy (TRFS) and (3) ECM mechanical property changes (stiffness and time to relaxation) using US radio frequency analysis and ARFI imaging in developing engineered cartilage tissue constructs. Here we present preliminary results from measurements on native and engineered cartilage tissue. US integrated reflection coefficient (IRC) values were found to be statistically significantly different (p<0.05) when comparing (1) control and glutaraldehyde (1%, 3 days) treated and (2) superficial and deep (close to bone) harvested native bovine cartilage. A statistically significant difference (p<0.05) was seen between control and collagenase (0.2%, 30 min) treated engineered cartilage constructs when comparing average lifetime values (440/40nm), and fractional contribution of slow decay component values. A similar statistically significant trend was seen in IRC values as in native tissue. Experimental results were validated by observing statistically significant changes in Young’s Moduli and compressive moduli from standard tensile and compressive testing methods. Results from this work will allow future development of an integrated system that allows continuous monitoring of developing engineered tissue in a non-invasive manner.
Giga-Pixel Lensfree Computational Microscopy On a Chip
Alon Greenbaum, Uzair Sikora and Aydogan Ozcan
UC Los Angeles
Bright-field microscopy is used in various bio-medical applications. Conventional lens-based microscopes, however, have relatively small field-of-views, and therefore microscopists are required to scan the samples to detect rare features of interest such as abnormal cells and parasites, a labor intensive and expensive task. To provide a high-throughput solution to such imaging needs, lensfree on-chip microscopy offers sub-micron resolution over a large field-of-view (e.g., 30mm2) without the need for any mechanical scanning. While especially advantageous for resource limited settings and field use, lensfree on-chip imaging has been limited so far to relatively low spatial density samples due to its transmission geometry.
Here we demonstrate a new approach, which permits imaging of dense and connected samples within a lensfree and compact design (see Fig. 1(a)), without compromising resolution or field-of-view. Rather than recording only one intensity measurement of the sample’s diffraction pattern, a few successive intensity measurements (e.g., 2-5) are recorded using an opto-electronic sensor-array (e.g., a CMOS chip) at different sample-to-sensor vertical distances. As a result, phase and amplitude images of dense samples can be reconstructed using an iterative multi-height phase recovery method, which propagates back-and-forth between different heights until the reconstruction process converges without requiring additional information such as the object support. To demonstrate the relevance of this field-portable lensfree microscope, we imaged a dense Papanicolaou (i.e., Pap) test that is frequently used for screening of cervical cancer. The reconstructed phase and amplitude images of the sample are shown in Fig. 1(b and c) respectively and are in a good agreement with a 40x objective-lens (0.65NA) image of the same specimen (Fig. 1(d)). Using state-of-the-art image sensors, this lensfree super-resolution based imaging device can create images with more than one Giga pixels, reaching an extremely high space-bandwidth product.
A Novel Functional MRI Method for Examining Modulations of Brain Blood Flow and Oxygen Metabolism in Humans
Valerie E. M. Griffeth, Nicholas P. Blockley, Aaron B. Simon, Farshad Moradi, and Richard B. Buxton
UC San Diego
Background: Functional magnetic resonance imaging (fMRI) based on the blood oxygenation level dependent (BOLD) signal is a sensitive tool for detecting cerebral blood flow and oxygen metabolism changes associated with neural activity. Yet modulations of the signal amplitude are difficult to interpret because they depend on both blood flow and metabolism changes, two distinct physiological responses for which the relationship may vary with the stimulus, attention or the brain’s physiological state. While oxygen metabolism is linked to the energy-demanding task of reversing the ion flux from action potentials and post-synaptic potentials, brain blood flow is hypothesized to be driven in a feed-forward way by neural signaling. Here, we present a new method for analyzing combined BOLD and blood flow measurements to test for flow/metabolism coupling modulations. Unlike the calibrated-BOLD methodology, this new method does not depend on measurements involving inhalation of altered gases.
Methodology/Results: A new simpler model that captures the behavior of the BOLD signal was tested with a previously developed detailed model, which includes all physiological factors affecting the BOLD response. The form of this new model suggests a methodology for testing whether flow/metabolism coupling differs between two responses. If additional information is available on the coupling ratio for any single stimulus (e.g., a calibration experiment or a pre-determined ratio for a standard stimulus), the absolute coupling ratios for all responses from the same baseline state can be determined. As an initial test, this method was applied to previous fMRI studies confirming changes in flow/metabolism coupling.
Conclusions/Significance: The new method proposed here provides a straightforward way to test how the coupling of blood flow and oxygen metabolism is modulated under a wide range of conditions. This will improve our understanding of the mutable relationship between neural activity, oxygen metabolism and blood flow in a wide variety of conditions.
Development of a registered optical coherence tomography and fluorescence microscopy system for studying structural and functional changes in nerves during activity.
Md. Rezuanul Haque, M. Shahidul Islam, Christian M. Oh, B. Hyle Park
UC Riverside
Neurons communicate with each other by sending and receiving electrical signals known as action potentials. A detailed investigation of a neural network or the nervous system overall is largely dependent on the monitoring of neural activity at the neuron level. Although a voltage change is the gold standard hallmark for detection of an action potential, there are a number of other changes that accompany neural spike propagation. These transient changes include alterations in scattering, birefringence, fluorescence, and thickness. Our preliminary attempts toward the development of an optical electrode have demonstrated the potential of phase- sensitive optical coherence tomography (OCT) to detect nanometer level thickness change of neurons associated with a functionally-stimulated limulus lateral compound eye. However, while OCT excels at acquisition of structural information, it lacks chemical specificity. In order to better understand the mechanism behind structural changes associated with action potential propagation, we are currently constructing a spatially and temporally registered OCT and fluorescence imaging system. In addition to this, we have developed a cold block for switchable activation and deactivation of action potential propagation through the nerve. The cold block will be used as control method in our experiments.
Fluorescence Lifetime Spectroscopy and Imaging Studies of Xenograftic Glioma Metabolic States
Brad Hartl, Anthony Valenzuela, Yang Sun, Rudolph Schrot, Jeffery Walton, Simon Cherry, Fred Gorin, Laura Marcu
UC Davis
We report on the application of a new bimodal imaging system, combining time-resolved fluorescence spectroscopy (TRFS) and ultrasound backscatter microscopy (UBM), to improve brain tumor margin detection using a xenograftic U-87 glioma rat model. The TRFS system employs a pulsed nitrogen laser (337 nm) to excite naturally occurring fluorophores found in brain tissue; the primary constituent of which is the reduced form of nicotinamide adenine dinucleotide (NADH). We expect to see differences in the fluorescence intensity decay profiles of NADH, which are ultimately caused by the altered metabolic pathways of tumor tissues. The UBM system utilizes a scanning single element transducer to provide high resolution structural imaging. In combining these techniques, concurrent detection of biochemical, functional and structural tissue features can be measured. In order to bridge the gap between conventional pre-operative diagnostic techniques and the proposed bimodal system we utilize magnetic resonance spectroscopy (MRS) and magnetic resonance imaging (MRI). MRS and MRI are employed to validate the hybrid system’s ability to measure tissue metabolic status and delineate tumor margins, respectively. Initial animal studies were performed in vivo; after the xenograftic tumor grew to appropriate size, rats were measured with MRI and MRS, followed by a craniotomy to expose the brain tumor, and finally TRFS and UBM measurements. Preliminary results from TRFS and MRS suggest that fluorescence emission differences can be detected between tumor and healthy tissue types by studying the fluorescence integrated intensities and characteristics of the intensity decays (lifetimes and Laguerre coefficients) at specific wavelengths. When compared with MRI, UBM results also suggest that the xenograftic tumor boundaries are detectible using b-mode images and acoustical parameters computed from radiofrequency data. Findings from this study have the potential to provide improved intraoperative brain tumor diagnosis and delineation, and also establish new paradigms for the research of these tumors.
Optical coherence tomography for non-invasive structural analysis of nerves and recording of neural activity
M. Shahidul Islam, M. Rezuanul Haque, Michael C. Oliveira, Christian Oh, Yan Wang, B. Hyle Park
UC Riverside
The first part of this study is about polarization-sensitive optical coherence tomography (PS-OCT) imaging of rat sciatic nerves. We have recently demonstrated that PS-OCT combined with regular intensity based OCT is a very useful tool to extract important features of peripheral nerve microstructure. These include measurement of epineurium boundary and thickness, estimation of extinction coefficient and birefringence, quantification of bands of Fontana and determination of boundaries between nerve, muscle and adipose tissues. The results from this part of study demonstrate that we can clearly and quantifiably delineate the nerve structure and the second part of this project is aimed to develop an optical tool to image the changes in those nerve structures that correspond with neural activity. A major portion of current technologies for neural recording use either different varieties of electrodes or the introduction of exogenous contrast agents and both of these are invasive to some extent. It has been known for past several decades that nerves undergo small rapid transient thickness change during action potential propagation and this sub-nanometer thickness change (~1-50nm) has good temporal correlation with action potential. Phase-resolved OCT (pr-OCT) has the capability of measuring very small (tens of picometers) displacements using differential phase measurements. Our developed common path pr-OCT system is currently capable of detecting thickness changes as small as 0.5nm with a temporal resolution of 10μs. Recently, electrical recording of “single action potentials” from the optic nerve associated with the lateral eye of horseshoe crab (Limulus polyphemus) has been demonstrated. Preliminary optical measurements from Limulus optic nerve have shown some evidence of activity which matches with concurrent electrical recording. To controllably deactivate and reactivate the propagation of nerve impulses, recently a cold block system has been also developed. Current and future works involve verifying simultaneous optical and electrical recordings of activity using cold block.
Magnetic Resonance of 2-Hydroxyglutarate (2HG) in IDH1-mutant Low-grade Gliomas
Llewellyn Jalbert, Adam Elkhaled, Joanna J. Phillips, Hikari A. Yoshihara, Rupa Parvataneni, Radhika Srinivasan, Gabriela Bourne, Susan M. Chang, Soonmee Cha, Sarah J. Nelson
University of California, San Francisco, San Francisco, CA
Infiltrating gliomas are heterogeneous tumors of the central nervous system that include astrocytomas, oligodendrogliomas, and mixed oligoastrocytomas. Recent studies have indicated that a significant survival advantage and favorable response to temozolamide is conferred to glioma patients whose lesions harbor mutations in the isocitrate dehydrogenase 1 and 2 (IDH1/2) genes (Yan et al., 2009) (Houillier et al., 2010). Interestingly, these mutations result in the aberrant production of a potential onco-metabolite: d-2- hydroxyglutarate (2HG) (Dang et al., 2009). Here we report the ex vivo detection of 2HG in pathologically confirmed IDH1-mutated tissue samples from fifty-two recurrent low-grade glioma patients using the nuclear magnetic resonance (NMR) technique of high-resolution magic-angle-spinning (1H HR- MAS) spectroscopy.
Relative 2HG levels and mutant IDH1 cells demonstrated correlations with various histopathology parameters including mitoses, vascular neoplasia, axonal disruption, relative tumor content, and increased cellularity. Of interest is that ex vivo spectroscopic measures of choline-containing species were also correlated with 2HG levels, as well as in vivo magnetic resonance (MR) measures of the apparent diffusion coefficient (ADC). These data provide novel and extensive characterization of mutant IDH1 lesions, while confirming the potential diagnostic value of 2HG and related imaging parameters as markers of clinically relevant tumor characteristics. Considering the survival benefits associated with IDH1/2 mutations, the prognostic value for patients with low-grade gliomas is significant. This information may augment the ability of clinicians to monitor therapeutic response in patient studies and provide criteria for stratifying patients to specific treatment regimens. If 2HG is confirmed to be a tumor supporting onco-metabolite, additional monitoring of 2HG levels may become important, especially for therapies targeting the IDH1 pathway. The magnetic resonance characteristics provided in this study may aid the development of in vivo sequences for use in the clinic.
Experimental 3D X-Space Magnetic Particle Imaging Using Filtered Backprojection
Justin Konkle, Patrick Goodwill, Steven Conolly
UC Berkeley
Magnetic Particle Imaging (MPI) is an emerging medical imaging technique that directly detects iron oxid nanoparticle tracers. Tomographic MPI using a shifted and rotated field free line (FFL) with filtered backprojection image reconstruction can approach an order of magnitude SNR improvement over standard MPI, which images using a field free point (FFP). We have built an imager to demonstrate this SNR gain of projection reconstruction imaging in MPI. The imager consists of a 2.4 T/m permanent magnet FFL gradient, a Helmholtz pair of off-the-shelf electromagnets, a solenoidal transmit coil and a gradiometer receive coil. A motor driven rotary table rotates the sample and the system acquires multiple projection images at evenly spaced angles between zero degrees and 180 degrees. Filtered back-projection is used to reconstruct a three-dimensional tomographic image volume. Sample rotation, which is sometimes employed in commercial mi-croCT scanners, has been used to test this method. Future systems may potentially rotate the gradient similar to a human-sized CT gantry or may generate an electronically rotated FFL gradient. In previous work, we have sown an MPI capable FFL scanner. Here, we show 3D experimental results of out PR-MPI scanner using acrylic imagining phantoms and post-mortem mice.
Noninvasive In-vivo Assessment of Vascular Remodeling in a Subcutaneous site following Xenogeneic Islet Transplantation
Rahul Krishnan, Rajan Arora, Hannan Dogar, Vatche Satamian, Morgan Lamb, Sean M White, Rick Storrs, Clarence Foster, Elliot Botvinick, Bernard Choi, Jonathan RT Lakey.
UC Irvine
Background: Xenotransplantation would help circumvent the current shortage of cadaveric human islet donors, inconsistent islet yields and poor graft function that currently hamper islet allotransplantation. Islet xenografts require life-long immunosuppression or biomaterial encapsulation to evade immune-mediated destruction. However biomaterial encapsulation results in suboptimal islet survival due to insufficient oxygen delivery via the microcirculation. Existing in-vivo imaging models cannot evaluate and quantify the vascular remodeling that occurs in response to subcutaneous biomaterial implantation and islet transplantation. We address this problem by employing the mouse dorsal window (DW) model and Laser Speckle Imaging (LSI) to quantify changes in subcutaneous functional vascular density (FVD) post-transplantation.
Methods: Dorsal window chamber models were implanted on C57BL/6 albino mice and either transplanted with blank alginate sheets or alginate sheets containing porcine islets. The vascular response to the transplanted sheets was studied over a 7-day period (Figure 1). LSI was performed on days 0, 2, 4 and 7 to map the changes in blood flow within the window. The images obtained were used to generate speckle contrast maps which were then analyzed using an algorithm to evaluate changes in functional vascular density.
Results: Porcine islet sheet implants elicited a qualitative increase in the rate of blood flow and peri-implant neovascularization by day 7 (Figure 1. F, H). An increase in the FVD was observed in chambers implanted with blank sheets (45.03%) and porcine islet sheets (43.73%) during the 7 days of implantation (Figure 1H). The rate of increase in the FVD over time was higher in chambers containing islet implants (1.70±0.04) as compared to blank sheets (1.41±0.02).
Conclusion: We postulate that our model can efficiently demonstrate and quantify biomaterial implant-induced angiogenesis. Future experiments will attempt to correlate these findings with islet viability and function for up to 14 days post transplantation.
Quantitative Biomarkers of Cancer From Metabolic Activity Decomposition using Stimulated—Echoes and Hyperpolarized Carbon-13 MR
Christine M Leon1,2, Peder EZ Larson2, Adam B Kerr, Robert Bok1, Sarah Nelson3, John Pauly3, John Kurhanewicz1,2, and Daniel B Vigneron1,2
1UC Berkeley -UCSF Graduate Group in Bioengineering, 2Department of Radiology and Biomedical Imaging, University of California, San Francisco, 3Department of Electrical Engineering, Stanford University
Introduction: Changes in enzyme expression precede tumorigenesis providing early biomarkers of disease1. Specifically, it is well known that the lactate dehydrogenase (LDH) enzyme is significantly upregulated in cancers2. However, in vivo the acquired signal is complicated by the vasculature4. We applied the stimulated echo acquisition mode (STEAM) technique to suppress vascular effects4. We propose a new, robust method for quantification of real-time cancer metabolism.
Methods and Modeling: STEAM in the presence of metabolic conversion creates a phase shift that depends on the resonance frequency and echo time (TE), Δφ=2πΔfTE/27. By choosing Δφ=π/2, the spins generated during a mixing time (TM), can then be separated during reconstruction. Dynamic spectra were obtained in TRAMP tumor model(n=5) and normal(n=5) mice. Conventional two-site exchange using only the magnitude of the data yields four unknowns but only two equations; an underdetermined system of equations with multiple local minima. Using MAD-STEAM, twice the amount of information can be obtained from a same acquisition providing a well-conditioned system of equations. Conventional modeling can underestimate conversion rates, which will not describe the actual kinetics of the system. However, Metabolic Activity Decomposition reconstruction of the same data yielded more accurate rates of conversion.
Results: MAD-STEAM was able to differentiate generated lactate from high rates of conversion in a tumor model, which is known to have high LDH activity. Simulations showed that Metabolic Activity Decomposition more accurately calculates relaxation and conversion rates. Moreover, fitting in vivo data with MAD-STEAM yielded KPyr>Lac values, which were better able to distinguish tumor versus normal, than conventional modeling.
Conclusion: MAD-STEAM improved real-time conversion and T1 relaxation measurement for better observation of enzyme activity. The ability to probe enzyme activity directly and measure relaxation provides new quantitative biomarkers, which could improve assessments of early tumor formation and regression with increased specificity.
References: [1] Hu, et al. Cell Metab. 2011; 14:131-142. [2] Kroemer and Pouyssegur. Cancer Cell; 2008, 13(6): 472-382. [3] Merritt. P Natl Acad Sci USA 104, 19773-19777 (2007). [4] Larson et al. IEE Trans Med Imaging, 2011; in press. [5] Kurhanewicz, et al. Neoplasia. 2011; 13(2): 81-97. [6] Day. Nature Med 13, 1382-1387 (2007). Acknowledgments: We would like to acknowledge Murat Arcak, Galen Reed, Simon Hu, Cornelius Von Morze, Hikari Yoshihara, Peter Shin and Kristen Scott for technical assistance, as well as, funding from NIH P41-EB01359, NIGMS-IMSD R25-GM56847 and NIH R00-EB012064.
Linearity and Shift-Invariance in Magnetic Particle Imaging, A New Medical Imaging Technique
Kuan Lu, Patrick W. Goodwill, Emine U. Saritas, Steve M. Conolly
UC Berkeley
Magnetic Particle Imaging [1] (MPI) is a new medical imaging modality that uses completely different hardware from existing modalities and holds great promise for high sensitivity, high contrast while avoiding tissue penetration and ionizing radiation. In all medical imaging modalities, linearity and shift-invariance (LSI) are critical characteristics for quantitative imaging. Our lab has developed the x-space MPI theory to analyze the MPI signal and image formation and this theory shows that MPI could be modeled as an LSI system [2] assuming negligible signal loss in the acquisition. However, MPI does suffer from signal loss in data acquisition. Unlike MRI, where the received signal and excitation are temporally decoupled, MPI drives the excitation coil and detects the particle signal simultaneously. The excitation field thus undesirably “feeds through” the receiver and corrupts the particle signal at fundamental drive frequency. To remove the direct- feedthough interference in the acquired MPI signal, we apply significant electronic filtering to suppress the entire fundamental frequency component. However, this filtering operation inevitably also removes particle signal at the same frequency, and compromises the LSI properties of the MPI system. In order to use MPI as a quantitative imaging technique, it is important to recover MPI’s LSI property by recovering the lost signal. In this study, we show that filtering out the fundamental frequency signal only leads to low spatial frequency loss in x-space and is completely recoverable with robust image processing methods [3]. The preliminary recovered image shows a good match to the theoretical simulation using LSI x-space theory. In conclusion, we prove that we can restore LSI properties to the MPI system.
[1] B. Gleich and J. Weizenecker. Nature, 435:1214–17, 2005.
[2] P. Goodwill and S. Conolly. IEEE Trans Med Imaging, 2010.
[3] Kuan Lu, P. Goodwill and S. Conolly. SPIE Conference Proceeding, 2011.
High Resolution Optical Molecular Imaging of Changes in Choline Metabolism in Oral Neoplasia
Zhen Luo1, Melissa Loja2, D. Greg Farwell4, Quang C. Luu4, Paul J. Donald4, Regina Gandour-Edwards5, Nitin Nitin1&3
1. Department of Biological and Agricultural Engineering, University of California – Davis
2. School of Medicine, University of California – Davis
3. Food Science and Technology, University of California – Davis
4. Department of Otolaryngology – Head and Neck Surgery, University of California – Davis
5. Department of Pathology, University of California – Davis
Choline containing phospholipids are important structural components of membranes. Upregulation in choline (Cho) uptake and intracellular retention as compared to normal tissue has been used as a diagnostic molecular imaging marker for cancer. The overall objectives of this study are to develop an optical molecular imaging approach to measure changes in choline uptake and metabolism in oral neoplasia and to evaluate its potential for early stage detection of the disease.
Choline uptake and intracellular retention was measured using a click chemistry analog of choline, propargylcholine. The rapid uptake of propargylcholine and specific detection of internalized propargyl choline in cells were demonstrated in 2-D cell culture. To evaluate the potential of topical delivery of propargyl choline in epithelial tissues, uptake of topically delivered propargyl choline was measured in 3-D tissue phantoms. Measurements using high resolution fluorescence imaging in 3-D tissue phantoms showed a rapid penetration and internalization of propargylcholine in 3-D multicellular structure.
To further evaluate the potential of this approach to detect upregulation in choline uptake and metabolism in oral cancer using topically delivered propargyl choline, 15 pairs of fresh oral biopsies (clinically abnormal and normal biopsies) from consenting patients at UCDMC were evaluated. High-resolution imaging results showed significantly higher fluorescence signal intensity in neoplastic tissues as compared to matched clinically normal paired biopsies. The imaging results were compared with pathological analysis of these paired biopsies and the results show significant agreement of imaging results with pathological diagnosis.
Overall results show that propargylcholine can be rapidly delivered to epithelium in oral tissue using topical application. The fluorescence intensity of propargylcholine uptaken in tissues was able to be non- invasively measured and quantified by optical imaging methods at the signal cell level. This novel optical molecular imaging approach has significant potential to be a useful tool for early detection of oral neoplasia.
Optical Molecular Imaging Approach for Rapid Assessment of Response of Individual Cancer Cells to Chemotherapy
Zhen Luo1, Rohan Vijay Tikekar2, Kiana Michelle Samadzadeh1, Nitin Nitin1&2
1. Department of Biological and Agricultural Engineering, University of California-Davis
2. Food Science and Technology, University of California – Davis
Predicting response of individual patients to cyto-toxic chemotherapy drugs is critical for developing an individualized therapy. With this motivation, an optical molecular imaging approach was developed to noninvasively and rapidly measure changes in choline metabolic activity in individual cells in response to chemotherapy. This imaging approach is based on molecular imaging of changes in uptake and intracellular retention of choline using propargyl choline (a click chemistry analogue of choline). To demonstrate potential of this approach, cancer cells in 2-D cell culture and 3-D spheroid models were treated with cisplatin. Changes in choline uptake in these model systems in response to cisplatin treatment were quantified by fluorescence microscopy and compared with conventional assays. The results of this study demonstrate that the uptake and retention of propargyl choline decreases with an increase in concentration of cisplatin. Changes in uptake of propargyl choline can be detected in approximately 3 hours post treatment of cells with cisplatin. Comparison between propargyl choline and 2- NBDG (a fluorescence analogue of glucose) demonstrate that the imaging approach based on choline analogue ]is more sensitive for detecting early response of cancer cells. Compared with the conventional cell viability assays (MTT and Annexin-V/PI staining), imaging approaches based on propargyl choline and 2-NBDG has higher sensitivity to evaluate the early response of cancer cells to cisplatin treatment. Results of the study also demonstrate the feasibility of imaging drug response in 3-d tumor spheroids. Overall, the results of this study demonstrate the potential of high resolution molecular imaging of changes in choline uptake to assess response of cancer cells and tissues to chemotherapy. The ability of this approach to measure drug response at a single cell resolution within a 3-D tissue is complementary to PET and MR spectroscopy methods that are predominantly limited to whole body/ tissue imaging.
Automated Simultaneous Time- and Wavelength-Resolved Fluorescence Spectroscopy For Diagnosis of Atherosclerosis
Dinglong Ma, Diego Yankelevich, Julien Bec, Yinghua Sun, Daniel Elson, Laura Marcu
UC Davis
Time-resolved fluorescence spectroscopy (TRFS) is currently recognized for its ability to detect biochemical features in atherosclerotic plaques. The application of TRFS in atherosclerosis diagnosis requires fast acquisition and analysis of autofluorescence data from clinically relevant endogenous fluorophores. Here we report the development and validation of a multichannel fast scanning-TRFS system that can be deployed in-vivo in combination with an intravascular ultrasound (IVUS) system for diagnosis of atherosclerosis. Autofluorescence in arterial wall is induced by a 355 nm, 35 ps laser pulse and split into four wavelength sub-bands with different optical delay lengths so that the sub- band fluorescence responses arrive at the detector in different time ranges. A single shot acquisition takes 200 ns, and the instrument repetition rate can be as fast as 10,000 points per second. With this fast system, dual modal intravascular imaging is achieved by combining it with a commercial IVUS system and adopting a multimodal intravascular catheter. The optical fiber and ultrasound transducer are housed in a custom built balloon catheter and can be rotated and pulled-back separately with DC motors and steppers. The motion sequence is controlled by the controller of the digitizer in TRFS system to achieve co-registration of the two modalities. An all-in-one software is programmed to take control of all components in the combined system and to process and display data for immediate accessibility in clinical condition. This system was characterized in terms of optical efficiency, measurement variation, imaging resolution and data throughput. Its performance was validated in artificial phantom and pig heart phantom. Current results demonstrate promising performance for future intravascular studies in vivo humans.
Bio-physical interaction of Mytilus californianus adhesive plaques on thin films
Nadine R. Martinez-Rodriguez, Chris Broomell, Herbert Waite and Jacob Israelachivili
UC Santa Barbara
Nature has surpassed man’s best efforts to create robust adhesives that function well in wet conditions. Marine mussels for example, utilize a proteinaceous glue for anchorage to a variety of hard, wet surfaces into order to avoid being swept away and destroyed by waves. The mussels are tethered to these surfaces by hundreds of byssal threads, often accumulating to scores of fibers, which terminate in an adhesive plaque at the surface interface. Byssal thread formation in M. californianus is thought to initiate by environmental cues that prompt the foot to emerge from the gap between the two valves for exploration of the surface. Once the foot has selected a suitable region for adherence it begins the process of “scrubbing” or surface preparation. The mussel then takes the next “committed” step, during which it becomes motionless, fixed by its tip to the designated attachment location. At this time the protein components for the new plaque and thread are assembled in the ventral groove. The successful design of high performance mussel-inspired wet adhesive polymers requires a mechanistic understanding of structure-function relationships in the plaque proteins and well as an understanding of plaque formation at multiple length and time scales; however, the process of surface preparation has yet to be characterized. When the mussel faces a fouled surface, does it wipe the surface clean or does it deposit a plaque on top of the fouling agent? Studies using three-day biofilm covered glass, gold sputtered and platinum evaporated glass aid in determining if the mussel is actively removing thin films as it forms a new adhesive plaque. Here we present the results of thin films on glass that gives clues as to how mussels interact with thin films as it prepares to form new adhesive plaques.
Quantitative method for standardization of contrast-enhanced head computed tomography protocols
Sarah E. McKenney, John M. Boone
UC Davis
Purpose: Technique factors for clinical contrast-enhanced head CT protocols are typically developed and implemented ad hoc. A quantitative method was implemented to optimize the technique factors by examining the effects of tube potential and current modulation on image contrast and noise.
Method: Five dilutions of iodine contrast were inserted into a 16 cm diameter PMMA cylindrical phantom. The phantom was scanned on two clinical CT scanners, a GE LightSpeed VCT and Siemens SOMATOM Definition. Images were collected from a series of permutations of (1) x-ray tube potential and (2) x-ray tube current modulation metrics. Using regions of interest (ROIs) within each insert, the contrast and signal- difference-to-noise ratio (SDNR) were measured. The image noise was also assessed using a larger ROI in the phantom. A protocol translation scheme, based on matching the x-ray output of each scanner was developed.
Results: It was found that lower x-ray tube potentials improved the image contrast for all dilutions of iodine contrast. Image noise was found to be linearly proportional to the modulation metric of the GE scanner and image noise followed a power relationship with the modulation metric of the Siemens scanner.
Conclusions: For the detection of iodine contrast agents within a head CT image set, a lower tube potential, such as 80 kV, improves image contrast (fig. 1) without increasing dose to the patient even when image noise levels are maintained. While more aggressive modulation metrics increase the dose to a patient, the contrast-to-noise ratio also increases. When the patient dose and image noise have been properly optimized, translating a protocol between different manufacturers of scanners is feasible.
Cell-phone based Rapid-Diagnostic-Test Reader Platform for Spatio-temporal Mapping of Diseases
Onur Mudanyali, Stoyan Dimitrov, Uzair Sikora, Swati Padmanabhan, Isa Navruz, and Aydogan Ozcan
UC Los Angeles
Rapid-diagnostic-tests (e.g., lateral-flow based immuno-chromatographic tests) take the center stage in surveillance of infectious diseases especially in the developing parts of the world. Such rapid-diagnostic-tests (RDTs) in general exhibit important advantages over conventional approaches (e.g., clinical examination) especially in resource poor settings. For providing an integrated digital network for automated reading and distribution of RDT results, we developed a universal smart RDT reader installed on a cell-phone that can work with various lateral-flow based tests to automatically digitize and evaluate test results (Fig. 1). Weighing ~60 grams, this RDT reader attachment utilizes an inexpensive plano-convex lens and three light-emitting-diode (LED) arrays to digitally acquire transmission and reflection mode images of RDTs of interest. A custom-developed cell-phone application (see Figure 1(c)) processes these raw images in real time to generate an extensive digital test report including test validation and semi-quantitative test reading results. The same cell-phone application also connects this RDT reader to a global database/server that stores and organizes RDT results uploaded by users, creating a real-time infectious disease map (running on Google Maps). Powered by two AAA batteries (see Figure 1(a)) or by the cell-phone battery through a USB connection (see Figure 1(b)), this smart RDT reader attachment can be customized to fit on different cell-phone devices (i.e., iPhone or Android based phones) without any major changes to its design. To test the performance of this smart RDT reader, we evaluated malaria, HIV and TB rapid lateral-flow tests that were activated with positive control antigens at various density levels. This integrated digital RDT reader running on a cellphone offers a dynamic and cost-effective telemedicine platform that can assist health-care professionals and policy makers by providing large-scale spatio-temporal statistics for the prevalence of various infectious diseases.
Maltodextrin based imaging probes detect bacteria in vivo with high sensitivity and specificity
Xinghai Ning1, Seungjun Lee1, Won Seo2, Mark Goodman2 and Niren Murthy1
University of California at Berkeley1, and Emory University2
Bacterial infections remain one of leading causes of death worldwide by killing millions of people each year, and affect all areas of medicine ranging from cardiology to oncology. A substantial hindrance to development of new diagnostics and antibiotics is lack of specific probes that can rapidly localize infections and monitor response to treatment. In this report, we present a new family of contrast agents, termed maltodextrin based imaging probes (MDPs), which can for the first time image bacteria in vivo with the specificity and sensitivity needed to detect early stage infections and measure drug resistance in vivo. MDPs are composed of a contrast agent conjugated to maltohexaose, and are rapidly internalized through the bacteria‐specific maltodextrin transport pathway, endowing the MDPs with a unique combination of high sensitivity and specificity for bacteria. Here, we show that MDPs selectively accumulate within bacteria at millimolar concentrations, and are a thousand‐fold more specific for bacteria than mammalian cells. In addition, we demonstrate that MDPs can image as few as 105 colony forming units (CFUs) in vivo and can discriminate between active bacteria and inflammation induced by either lipopolysaccharides (LPS) or metabolically inactive bacteria. Finally, we demonstrate that MDPs can monitor therapeutic efficacy in vivo and identify beta lactam resistance in E.coli, thus providing physicians with a powerful tool for guiding antibiotic selection. We anticipate numerous clinical applications of MDPs given pervasiveness of infections in medicine.
Effect of angular acquisition on the slice sensitivity profile (SSP) in tomographic breast imaging
Anita Nosratieh
UC Davis
It is believed that the addition of acquisition over an angular range will resolve the limitations of mammography by adding volumetric breast-tissue information. A prototype, digital breast tomosynthesis (dBT), by Hologic Inc, system was approved by the Food and Drug Administration for use as a breast cancer screening tool in February 2011. The dBT system acquires tomographic images at an angular range of 15°. An in- house designed and manufactured breast computed tomography (bCT) system acquires images over a 360° angular and is yet to be employed as a screening tool. The slice sensitivity profile (SSP) was used to characterize resolution between tomographic planes. This study experimentally evaluated the SSP and its relationship between acquisition angle, ranging from 15-360°, and object size. Brass disks were placed within adipose tissue equivalent breast phantoms and images were acquired on both the dBT and bCT systems. Angular ranges between 15° and 360° were studied using a subset of the projection images acquired from the bCT scanner. The SSP was determined by measuring a background corrected mean gray scale value as a function of the z-position (axis normal to the plane of the detector). The results show that SSP improves when increasing angular acquisition range, approaching a delta function above 180°. Smaller objects have a narrower SSP and the SSP is not significantly dependent on the cone 1angle. For a 2.5, 5, 10 mm disk, the full width at half maximum (FWHM) of the SSP was 35, 61, 115 mm, respectively, on the tomosynthesis system and was 0.5 mm for all disk diameters on the bCT scanner. The SSP is dependent on object size and angular acquisition range. These dependencies are overcome once the angular acquisition range is increased beyond 180°.
Parallelized algorithms for high-speed multi-functional optical coherence tomography imaging
Michael Olveira
Yan Wang, Christian Oh, Mohammad Shahidul Islam, Arthur Ortega, Hyle Park
UC Riverside
Simultaneous optical coherence tomography (OCT), polarization-sensitive OCT and Doppler OCT imaging is desirable for assessing the structure, physiology and function of biological samples. While advances indetection hardware have enabled line acquisition rates in excess of 300 kHz for spectrometer-based and swept-source optical coherence tomography (OCT) systems, the heavy computational load required to process the acquired data stream creates a bottleneck in the realization of real-time multi-functional OCT imaging. We present the development and implementation of multi-functional OCT data processing algorithms onto graphics processing units (GPUs) for high-speed real-time multi-functional imaging. GPU- accelerated algorithms were developed using NVIDIA’s Compute Unified Device Architecture (CUDA) and implemented onto the Tesla C1060 GPU for processing raw data into structural and functional images. The key improvement in the processing speed comes from both exploiting the inherent parallelism of GPUs and determining how the processing algorithms could be parallelized for GPU implementation. Results show that parallelizing the functional algorithms significantly reduces the processing time required to reconstructstructural and functional images. The GPU-accelerated algorithms boast superior processing times for all image sizes compared with the CPU-based algorithms. Structural OCT images can be acquired and processed at 20 frames per second at data sizes of 1024 pixels by 2048 A-lines. This is demonstrated by OCT imaging of the eye of a horseshoe crab. Additionally, multi-functional OCT imaging is performed at 10 frames per second using hybrid CPU-GPU-based processing compared with 2 frames per second using CPU processing only. This is demonstrated by live in vivo imaging of a mouse brain. High-speed multi‐functional OCT imaging is useful is performing preview scans for quick structural and functional assessment of biological samples.
PET/MRI evaluation of 2-18F-fluoroacetate as a PET tracer of glial metabolism
Yu Ouyang, Jan Marik, Simon R. Cherry
UC Davis
Specific in vivo imaging markers of glial metabolism are potentially valuable for translational research of stroke and related neuropathologies. Recent work in rodent models of brain ischemia demonstrates that 2-18F-fluoroacetate (18F- FACE), a positron emission tomography (PET) radiotracer, has elevated uptake in lesions. In this study, we used PET and magnetic resonance imaging (MRI) to evaluate parameters of 18F-FACE uptake in a mouse model of cerebral ischemia- hypoxia (I-H).
Methods: Brain lesions were induced in 8-12 week old, male, C57BL/6 mice via I-H and followed approximately 24 hours later by MRI and PET. Parametric maps of apparent diffusion coefficient (ADC) and T2 were acquired via MRI (4.7 T, Varian, Inc.). PET data were acquired (Inveon, Siemens AG) by 30 minute dynamic scan immediately after tail vein injection of 18F-FACE (300 μCi in 100 μL). Time activity curves of 18F-FACE were fit to a two-tissue model of fluoroacetate uptake. Rate constants are k1 (influx from plasma to brain), k2 (efflux from brain to plasma), and k3 (metabolism trapping in brain). Image registration, analysis, and kinetic modeling were performed in PMOD (PMOD Technologies Ltd.).
Results: Regions of elevated 18F-FACE uptake corresponded in morphology to lesions determined by the T2 parametric map (Fig. A&B). Kinetic modeling shows that k3 was elevated in lesion compared to contralateral brain. Areas of elevated k1/k2 appeared in the general location of the lesion, but areas of elevated k3 did not and instead appeared to border areas of elevated k1/k2 (Fig. C&D).
Conclusions: Data suggest the necrotic core of the lesion may be indicated by areas of increased 18F-FACE influx-to-efflux (k1/k2) but low metabolism (k3). In addition, increased glial metabolism occurring in the lesion periphery may be indicative of activity by astrocytes in injured but potentially viable tissue. Immunohistochemical staining may support this conclusion.
In Vivo Identification of Brain Structures Using Optical Coherence Tomography
Carissa L. Reynolds, Melissa M. Eberle, Jenny I. Szu, Mike H. Hsu, Yan Wang, Devin K. Binder, B. Hyle Park
UC Riverside
Optical coherence tomography (OCT) is a nondestructive optical imaging technique which has proven to be an invaluable clinical tool in fields such as ophthalmology and cardiology. It is used as a real-time image guiding device during surgery as well as for performing in vivo optical biopsy of tissue1. Neurosurgery could greatly benefit from such a tool, although little work has been done to meet this need. Currently during neurosurgery any probes placed into the brain are inserted blindly by the neurosurgeon, guided only by standard stereotactic procedures. Real-time imaging to identify tissues and blood vessels can greatly improve the accuracy and reduce the risk for bleeding during such procedures2. In this work, we demonstrate the capability of OCT for identifying brain tissue layers in vivo. Using a thinned-skull cortical window mouse model, we collected OCT images in the cerebral cortex region using a spectral-domain OCT setup. Comparison of our OCT images with histology revealed that the cortex and the corpus callosum are visible with OCT. This is further verified with measurement of the tissue layer thicknesses, which are consistent with the stereotactic atlas. The attenuation coefficients of the white and gray matter regions were also extracted from the OCT data to quantitatively discriminate between the tissue layers. This work demonstrates that OCT is capable of imaging and identifying different brain tissue layers in vivo and highlights OCT’s potential as a valuable imaging device for neurosurgery.
1. Fujimoto, J.G. Optical coherence tomography for ultrahigh resolution in vivo imaging. Nature biotechnology 21, 1361-7 (2003).
2. Liang, C.-P. et al. A forward-imaging needle-type OCT probe for image guided stereotactic procedures. Optics express 19, 26283-94 (2011).
High-throughput Color Imaging on a Chip Using Sparse Signal Recovery
Ikbal Sencan, Ahmet F. Coskun, Ting-Wei Su, David G. Herman, YeongSeok Suh, and Aydogan Ozcan
UC Los Angeles
Biomedical applications such as cytometry, microarray imaging and rare cell analysis need cost-effective and compact imaging systems with decent resolution and high throughput. However, conventional lens-based optical microscopes are still far from comprehensively addressing these needs all at once. Lensfree computational imaging techniques, on the other hand, provide opportunities to replace bulky and costly designs of conventional microscopes with much simpler, compact and cost-effective on-chip imaging architectures, which can achieve a decent spatial resolution over a significantly larger imaging area compared to conventional optical microscopes. One example of such an emerging computational microscopy technique involves lensfree fluorescent imaging on a chip using sparse signal recovery (i.e., compressive decoding). In this approach, an opto-electronic sensor-array (e.g., a CCD or CMOS chip) records lensfree diffraction images of incoherent objects located on a sample holder e.g., a bio-chip that is placed at <0.5mm away from the active area of the sensor-chip (Fig. 1). Using the 2D point-spread function of this lensfree imaging platform, which can be experimentally measured, the object distribution on the chip is rapidly reconstructed by utilizing sparse signal recovery algorithms based on compressive sensing/sampling. Here we specifically focus on multi-color imaging performance of this lensfree incoherent microscopy technique achieving a spatial resolution of e.g., ~2-3 μm in each color channel (i.e., red, green and blue) over a wide field-of-view of e.g., ~60 mm2. With its decent multi-color resolution and large field-of-view, this compact and cost-effective imaging architecture could be especially valuable for biomedical applications, e.g., high-throughput cytometry, microarray imaging and rare cell detection methods involving large-area micro-fluidic channels.
Interaction of myoglobin with fatty acid
Lifan Shih
UC Davis
The NMR studies of myoglobin functions have not only raised the questions for the role of myoglobin as the major oxygen carrier in the muscle cells, but also presented evidence consistent with a specific and non specific interaction between myoglobin and fatty acid. The specific interaction occurred in the structural region near the heme 8 methyl group of myoglobin cyanide indicated from the intensity decreasing and line broadening of its corresponding peak at the hyperfine region, and the nonspecific interaction was observed from the increase of the fatty acid methylene peak area and 0.1 ppm change in the chemical shift of the peak in the presence of myoglobin. These interactions have suggested a new role for myoglobin in physiology, which could be further applied in the diagnostic procedures for lipid metabolism.
High-Power Active Interference Suppression in Magnetic Particle Imaging
Bo Zheng, Travis Massey, Patrick Goodwill, Steven Conolly
UC Berkeley
Magnetic particle imaging (MPI) can be used for high contrast, high sensitivity angiography and in vivo stem cell imaging. At present, however, the sensitivity and signal-to-noise ratio (SNR) in MPI can be limited by distortion interference arising from various sources in the imaging hardware. In this work, we investigate the effectiveness of actively canceling the distortion interference in MPI systems using a coupled transformer. We then describe the design and construction of the active cancellation circuit. Moreover, we demonstrate the suppression of 8 interfering frequencies to below the system noise floor, achieving better than 45 dB suppression at the major interfering frequencies. This technique can potentially improve the sensitivity and SNR of magnetic particle imaging by over two orders of magnitude.
Synthetic Biology / Systems Biology / Computational Biology / Genomics
Design of a Novel Parallel Flow Device to Decouple Mass Transfer and Mechanotransduction at the Endothelium
Heran C Bhakta, Ryan Kozakaa, Chris Halea, Prashanthi Vandrangia, Victor G. J. Rodgersa
UC Riverside
The vascular endothelium triggers various signaling pathways to maintain its homeostasis. The regulation of these signaling pathways has been attributed to mechanical transducers contrast to chemical transfer. In our research, we will test the role of endothelial mass transfer independent of mechanotransduction. To investigate this, we designed in-vitro experiments utilizing a modified parallel flow device to decouple mass transfer and mechanotransduction. A constant shear stress will be applied while modifying the concentration gradient across the membrane by varying the transmembrane pressure. A parallel flow device (8.5 cm x 3 cm) using polymethyl methacrylate (PMMA) was designed. The device encompasses a top and bottom chamber, each containing slits for an inlet and outlet. A cell-seeded polycarbonate membrane with 5.0 μm pore size (Whatman (New Jersey)) separated both chambers. The inlet for the top chamber is used to facilitate the transmembrane pressure. To test the optimal working of our device, a simulated model was created using COMSOL Multiphysics (Version 3.5) to analyze the mass transfer of Ca2+ from the culture media into the endothelial bed. The fluid phase (μ=0.0098 g/(cm-s)) was governed by the continuity equation and the incompressible Navier-Stokes equation. The chemical species (D=3.2 x 10-5 cm2/s) flowing through the system follows the equation of conservation of mass flowing through an incompressible medium and Fick’s Law of diffusion. Our preliminary computational results show substantial change in concentration gradients while maintaining constant shear profiles. We will further test the designed flow device to delineate mass transfer from mechanotransduction by seeding endothelial cells on the membranes and studying the impact of varying transmembrane pressures on them.
Analysis and Validation of a Multi-Scale Cerebral Blood Flow Model
Mark J Connolly, Xing He Ph.D., Xiao Hu Ph.D.
UC Los Angeles
Due to the rigidity of the cranial vault, it is difficult to accurately study cerebral blood flow dynamics. One approach is to utilize a mathematical model. The model presented here is a novel combination of a one dimensional fluid flow model representing the basal vessels of the circle of Willis (CoW), where each of the six major vessels of the CoW outlets into an individually parameterized, lumped parameter, auto-regulation model of the distal vasculature. The signal studied is cerebral blood velocity (CBV) of the middle cerebra artery (MCA), which exhibits a stereotypical pulse in response to a heartbeat. The simulated pulse was analyzed using the Morphological Clustering and Analysis of Intracranial Pressure (MOCAIP) algorithm, which calculates features describing the shape of the pulse. This first analysis focuses only on the capacity of the model to reproduce pulse shape changes associated with vasodilation, as seen clinically during CO2 rebreathing tests.
A fractional factorial design was used to obtain the sensitivity of each feature to each parameter variation. Using this sensitivity matrix, the correlation matrix was obtained to identify correlating parameters. Subsequently, a linear leasts squares approach was utilized to determine the parameter values necessary to replicate clinically obtained data.
Clinical observations have identified twelve features that specifically increase or decrease during vasodilation. Using this technique, it was possible to identify physiologically plausible parameter combinations that were able to correctly reproduce changes in nine out of the twelve CBV features associated with vasodilation. These parameters were related to the active and passive tension of the arterial wall and cerebrospinal fluid circulation. Subsequent analyses including the parameters controlling craniospinal compliance have shown promise in more accurately replicating vasodilation data. These results demonstrate the potential for this model to help study cerebrovascular dynamics.
Hot-Wired E. coli: Expression of a Complete Electron Transfer Pathway
Matthew Hepler, Heather Jensen, Dr. Jay Groves, Dr. Caroline Ajo-Franklin
UC Berkeley
Cellular-electrical connections have the potential to combine the distinct capabilities of inorganic technologies and living systems. However, cell membranes are natural insulators, creating a barrier between intracellular electron carriers and the outside world. To bypass this barrier, we have ‘grown’ electrical connections in living cells to provide a well-defined electron conduit out of the cell. The dissimilatory metal- reducing microbe Shewanella oneidensis MR-1 inspired our approach: it has the unusual ability to transport electrons to extracellular minerals through a trans-membrane electron transfer (ET) pathway using c-type cytochromes. We seek to generalize this ability to grow electrical contacts between microbes and inorganic materials, and thus have genetically re-engineered a full ET pathway into a model microbe, Escherichia coli (Figure 1). The re-built pathway consists of four Shewanella proteins: three cytochromes (CymA, MtrA and MtrC) and one outer membrane β-barrel protein (MtrB). The proper localization of the recombinant proteins in E. coli was verified through cellular fractionation and western blotting using antibodies of native E. coli proteins to confirm proper separation of the cellular components. The concentration of each cytochrome was determined using densitometry on a heme-stained SDS-page gel and UV-Vis spectroscopy of chemically reduced whole cells. Electrochemical measurements showed that these recombinant strains have an electron transfer rate ~70% of S. oneidensis. Our next objective is to increase the Coulombic efficiency of the strains and use genetic engineering to build additional electron pathways from other metal reducing organisms to find the most efficient ET pathway. In the future, this work may be applied toward electrosynthesis of fuels and other materials.
A Thyroid Hormone Dynamics Simulator for Optimizing Treatment of Congenital Hypothyroidism (CH) in Newborn Babies
King Chung Ho, Marisa Eisenberg and Joe DiStefano III
UC Los Angeles
Congenital Hypothyroidism (CH) is a condition of thyroid hormone (TH) deficiency present at birth, a serious problem with no single cause. One in ~ 3500 newborns have a severe deficiency which, if untreated, leads to growth failure and permanent mental retardation. Milder deficiencies typically lead to milder but similarly deleterious effects on structural and cognitive growth. Regardless of the cause, treatment is clear, necessary and straightforward – a daily oral dose of thyroxine (T4). Nearly all the developed world practices newborn screening to detect and treat CH in the first weeks of life. With incomplete knowledge of specific causes, possibly many, a major treatment issue is knowing how much T4 to give and how to adjust dosages over time. While enough is needed to be effective, overdosing can have harmful side effects. Blood hormone levels are normally assessed periodically before and after onset of treatment, followed by dosage adjustments, but dosage control remains essentially “open-loop”, with “variable delayed feedback” corrections. Our goal is to model the disease and quantitatively rationalize patient-specific treatments, using patient data and clinician-informed, optimized dosing regimens. Our new CH neonate model (below) is an adaptation of a published adult-children, hypo- to hyperthyroid condition simulator of feedback control of human TH levels*. Neonate model dynamics are being quantified and validated from exceptionally available literature data and close cooperation with authors of this early data. We anticipate preliminary results by symposium time.
*R Ben-Shachar, M Eisenberg, SA Huang, JJ. DiStefano III. Simulation of Post-Thyroidectomy Treatment Alternatives for Triiodothyronine or Thyroxine Replacementin Pediatric Thyroid Cancer Patients. THYROID, 22 (2012)
A New Architecture for Patterning Gene Expression Using Zinc Finger Proteins and Small RNAs
Justin Hsia, William J. Holtz, Michel M. Maharbiz, and Murat Arcak
UC Berkeley
Biologists have long searched for reaction networks that can spontaneously generate spatial patterns in the presence of diffusion. In our search for a viable synthetic gene network that can spontaneously produce patterning, we previously identified a new class of networks that we call quenched oscillator systems. These systems consist of a primary feedback loop that serves as an oscillator and a secondary feedback loop that quenches the oscillations and incorporates a diffusible molecule. We demonstrated that quenched oscillator systems are capable of generating spatio-temporal patterning with appropriate parameter values. Here we take advantage of recent work with zinc finger proteins (ZFPs) and small RNAs (sRNAs) to propose a new implementation using a new oscillator subsystem and a modified quenching loop. The new oscillator subsystem takes advantage of orthogonal sets of ZFP-promoter pairs and increased cooperativity via mRNA sequestration using sRNAs.
Chemogenomic profiling of Zymomonas mobilis in plant hydrolysates leads to identification of tolerance genes and an engineered strain with improved ethanol production
Jeffrey M. Skerker 1,2,3, Dacia Leon 1,4, Morgan Price 3, Jordan Mar 1,4, Dan Tarjan 1,4, Jason Baumohl 3, Stefan Bauer 1, Ana Ibanez 1, Valerie Mitchell 1, Cindy Wu 4, Ping Hu 4, Terry Hazen 4, Adam P. Arkin 1,2,3
1Energy Biosciences Institute, University of California, Berkeley
2Department of Bioengineering, University of California, Berkeley
3Physical Biosciences Division, LBNL 4Earth Sciences Division, LBNL
Plant hydrolysates derived from lignocellulosic biomass are a rich source of sugar and therefore, can serve as promising substrates for an industrial scale fermentation process. A major hurdle in this method is the recalcitrance of lignocellulosic biomass that renders the pentose and hexose sugars inaccessible to the fermenting microbes. Several pretreatment strategies exist to release these sugars, but the process causes formation of many toxic inhibitors such as organic acids, phenolics, and sugar-dehydration products. These compounds inhibit microbial growth and fermentation, which makes engineering tolerance to this mixture an essential step in developing commercial cellulosic biofuel. Zymomonas mobilis is a favorable candidate for fuel production, mainly due to its ability to produce ethanol at near theoretical yields. In our work, we aimed to understand the genetic basis of hydrolysate stress in Z. mobilis and to rationally engineer a tolerant strain that can efficiently ferment sugars to ethanol. Using a chemogenomic approach, pooled fitness assays were performed in plant hydrolysates and 37 chemical components. We identified 49 genes that are important for hydrolysate tolerance. Interestingly, these genes spanned a wide variety of functional classes, indicating that hydrolysate tolerance is a complex trait. A subset of these genes was selected for follow-up studies where individual mutants were phenotypically characterized and complemented. To rationally engineer Z. mobilis, we systematically overexpressed each tolerance gene and discovered that overexpression of ZMO1875, a gene of unknown function, improved specific ethanol productivity 2.4-fold in the presence of plant hydrolysate. Our studies demonstrate the first example of successfully engineering hydrolysate tolerance in Z. mobilis.
Lignocellulosic biomass to polyester conversion in Haloarchaea
Michael O. Starr, Elizabeth G. Wilbanks, Andrew I. Yao, Marc T. Facciotti
UC Davis
Polyhydroxyalkanoates (PHAs) are a class of polyesters produced by a wide variety of organisms as a means of carbon and energy storage. PHAs are biodegradable under a variety of conditions and have variable chemical and mechanical properties that make them ideal for use in industries from consumer packaging to medical implants.
I am interested in using systems tools to gain a more thorough understanding of the mechanisms underlying the regulation of PHA metabolism in Haloarchaea. Haloarchaea are a group of organisms that require a high salinity environment and have mechanisms to protect against environmental stresses such as desiccation and UV exposure. They have been shown to produce PHA granules under conditions of carbon limitation and nutrient excess. Furthermore, I have performed a preliminary growth and PHA production screen of ~60 of the ~80 sequenced Haloarchaea on preprocessed ligninocellulosic biomass. The majority of these Haloarchaea are able to grow, and several show significant levels of PHA production. The ability to produce an industrially useful polyester from a waste feedstock under conditions that would inhibit the growth of most other organisms makes Haloarchaea a promising organism for contamination-resistant industrial PHA production.
In addition to growth and production screens, I am finalizing a proteomics pipeline and a ChIP-seq data analysis pipeline as tools for monitoring protein levels and transcriptional regulation of relevant pathways in Haloarchaea. This, combined with transcriptomic data, will elucidate the state of the cell and the control points for substrate to PHA conversion. Understanding the regulation of these pathways will allow fine-tuning of both substrate utilization and the chemical and mechanical properties of the PHA produced.