Progress 10/01/07 to 09/30/08

Outputs
OUTPUTS: The portable fully-automated Microfluidics Desktop platform consists of five modules: (a) electro-hydraulic pumps (b) cell concentrator (c) sample preparation (d) DNA purification (e) real-time PCR amplification and detection of bacterial DNA. Device fabrication: PDMS (polydimethylsiloxane) pre-polymer was mixed with a curing agent at a 10:1 mass ratio and degassed, poured into an SU-8 mold, and baked at 60C. 250nm-thick gold electrodes were e-beam evaporated onto a bare silicon wafer and annealed at 400C for 120 mins. Small gauge tubing was inserted into the final device assembly inlets/outlets and fixed with silicone sealant. Copper wires were bonded to the electrodes. Electro-hydraulic pump testing: The electrolyte chamber was filled with TEA buffer, the hydraulic chambers were filled with DI water, and the reagent chamber was filled with DI water for testing. AC and DC voltages of 5V, 7.5V, and 10V were applied to the electrodes in the 3mm-diameter (7uL volume) electrolyte chamber to produce electrolytic gas bubbles, with average pumping rates of 1.30, 1.44, and 2.57uL/min., respectively, with AC voltages increasing electrode longevity from 5 min. to over 30 min. Cell concentrator & sample prep testing: 100uL of target cells were mixed with 10mL of PBS-Tween buffer and 75uL of anti-Salmonella Dynabeads for 10 min. incubation and 5 min. magnetic settling. The excess solution was pushed out of the cell concentrator, and 150mL of cell-bead solution was pumped into sample preparation module, magnetically-captured and lysed with 4.0M GuSCN for 5 minutes. The percentage of uncaptured beads was studied as a function of magnetic strength at flow rates of 5, 10, and 15 uL/min and found to be 0.26%, 0.84%, and 2.71%, respectively. The DNA in the collected lysate was analyzed by ABI 7000 real-time thermocycler, and cycled at 15s at 95C, 30s at 60C for a total of 55 cycles. Increasing concentrations of S. typhi (0, 10, 1E2, 1E3, 1E4, 1E5, 1E6, 1E7, 1E8 CFU/mL) is correlated with decreasing threshold thermal cycle times (38, 34.5, 33, 33.5, 33, 31, 27, 25, 23). System testing: This testing included platform sections (b) through (e). Continuing on from the testing described previously, the lysate was purified the DNA capture region and mixed with PCR master mix before pumping into the PCR amplification region. The device was pressurized and cycled at 15s at 95C, 30s at 60C for 55 cycles, and TaqMan probe fluorescence was detected using a photodetector. Between 100 and 1000 S. typhi cells could be detected using this system, with the total testing time from raw sample to result of 1.5 hrs, but reducible to 1 hr with optimized pumping speeds.Flow cytometer testing: A compact fluorescence-based particle counter capable of discriminating between several DNA by hybridization to fluorescent particles containing independent optical signatures was developed for multiplexing purposes. Characterization of the fluid focusing manifold was conducted to ensure passage of the particles through the beam of the excitation laser. In addition, we have succeeded in hybridizing fluorescent DNA to microspheres surface-modified with specific-sequence probe DNA. PARTICIPANTS: Carl A. Batt: Principal Investigator; Clarissa Lui: Ph.D. Student; Matthew Kennedy: Ph.D. Student; Scott Stelick: Engineer; Ines Calleja-Lopez: Post-doctoral from Spain; Nathaniel A. Cady: Collaborator, University at Albany TARGET AUDIENCES: The Microfluidics Desktop platform will find use in areas such as environmental microbiology, food safety, and rapid forensic evidence evaluation. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
With recent outbreaks of food pathogens in the U.S., there is an increasing need for a field-deployable instrument that can detect specific DNA sequences from a heterogeneous mixture of DNA in order to quickly assess the type of pathogen and determine the source of outbreak. A field deployable instrument should be compact, portable, and have a short total run-time with minimal operator intervention. The goal of this work is to help develop such an instrument for uses in industries such as food safety and forensics.

Publications

• No publications reported this period

Progress 01/01/07 to 09/30/07

Outputs
QUANTUM DOTS During this reporting period, we have developed a simple synthetic approach to produce water-soluble, near-infrared (NIR) emitting PbS and PbSe quantum dots (QDs) based on commercial QDs in organic solvents. The QDs are amenable to biomolecule functionalization via carboxylic acid groups on the QD surface. Furthermore, the quantum dots have been shown to associate with human colon carcinoma cells thus moving towards a proof-of principle means for visualizing cancer in vivo at the cellular level in the NIR. Water suspended QDs. Commercially available PbS and PbSe QDs (average wavelength = 1040 nm, diameter = 3.2 nm) capped with oleic acid and in organic solvents were transferred into water using 16-mercaptohexadecanoic acid, 11-mercaptoundecanoic acid, and amino-ethanethiol-HCl. Spectroscopic Measurements and Near-Infrared Fluorescent Imaging. Absorption spectra were measured at room temperature using a Shimadzu UV-3101 spectrophotometer. Emission spectra were recorded at room temperature on an infrared fluorometer equipped with a monochromator and a 25 mW HeNe laser as the excitation source. Imaging was done using an Olympus IX 71 inverted microscope was modified to allow NIR fluorescence imaging. A thermoelectrically cooled InGaAs camera was used instead of the normal silicon camera, along with an appropriate filter set. The 546 nm line from a mercury lamp was used for the excitation. Cell Preparation Procedure. HT29 cells were subcultured during log-phase growth: 5000 cells were transferred onto chambered cell culture microscope slides. Quantum dots were added to the cells and allowed to incubate then where subsequently washed to remove excess QDs. The cells were then fixed and mounted for imaging. These findings have been disseminated to communities of interest through peer-reviewed publication. NANOBIOTECHNOLOGICAL DEVICES We have initiated efforts in developing three dimensional carbon nanotubes architectures, to control, guide and eventually monitor cellular interactions. Carbon nanotubes micro-architectures are fabricated by lithographic pattern of the nickel (Ni 3nm) catalyst and consequent chemical vapor deposition for the growth of vertically aligned CNTs. We have developed a chemical functionalization protocol that allows for the specific functionalization of the carbon nanotubes platforms with biomacromolecules. The CNT architectures were first used to specifically target endothelial mammalian colon carcinoma cells. In order to target these cells, the humanized single-chain variable domain fragment antibody (A33scFv) that recognizes the A33 cell surface glycoprotein expressed in the SW1222 colon cancer cell line, was conjugated on the CNTs. Indeed the CNT-A33 architectures were successful in promoting specific cancer cell adhesion when compared to pristine tubes, that did not promote cell adhesion overall or even when compared with CNTs functionalized with a non-specific antibody. These findings have been disseminated to communities of interest through conference proceedings.

Impacts
QUANTUM DOTS Our findings show the first images of cells labeled with QDs and recorded at a wavelength above 1000 nm. Near-infrared emitting optical probes are of interest for in vivo imaging since water and many physiological tissues are known to absorb and scatter less light in NIR range between 700-1400nm. In wavelength region, light can then penetrate up to 10-15cm of tissue thickness and there is much less background signal generated in tissue compared to using visible light. Specifically labeling cancerous tissue with NIR probes opens the possibility to optically diagnose cancer in vivo. Furthermore, QDs are known to associate with individual cells attaching to the cell surface and becoming internalized into sub-cellular compartments. In addition, QDs can be functionalized with biomolecules such as antibodies allowing the association of QDs only to specific cells of interest. In this manner, QD-protein constructs can be developed in order to optically detect cancerous tissue at the cellular level in vivo, perhaps resulting in very early-stage detection of cancer. NANOBIOTECHNOLOGICAL DEVICES Our work in surface bio-functionalized CNTs is crucial for the development of bioelectronic devices as well as implantable devices for biomedical applications. This work highlights the potential of CNT architectures in controlling cell-surface interactions with implications in understanding basic cellular functions, with the ability to probe sub-cellular components and functions, crucial also for our understanding of cancer cellular mechanisms, but also in targeting and sensing technologies.

Publications

• Hyun, B.R., Chen, H.Y., Rey, D.A., Wise, F.W. and Batt, C.A. 2007. Near-infrared fluorescence imaging with water-soluble Pb salt quantum dots. J. of Phys. Chem. B. 111(20):5726-5730.

Progress 01/01/06 to 12/31/06

Outputs
We have continued our previous results of S-layer templated nanoparticle arrays, by expanding our efforts in a variety of nanoparticles both metallic and semiconducting and in evaluating two S-layers of different crystal lattices as biotemplates. The honeycomb lattice of the Sulfolobus acidocaldarius S-layer and the hexagonal lattice of Deinococcus radiodurans S-layer, were used as a biotemplate to generate arrays of dendrimer-encapsulated platinum nanoparticles (Pt-DENs) under a range of different pH conditions via non-covalent nanoparticle-protein interactions. We also evaluated the S-layer crystal lattice for its sensing properties, by developing an S-layer bioelectronic device and evaluated its ion gating characteristics. We recorded the Electrochemical Impedance Spectra of the D. radiodurans protein fragments chemisorbed on a modified silicon surface, in the presence of four chloride cations (M+Cl-, M = Na+, K+, Ca2+, Mg2+) and three sodium anions (Na+X-, X= NO3-, SO42-, citrate-3 ) of varying charge and lipophilic properties. Capacitance values revealed selective response to cationic species with the highest one observed for calcium. It was determined that calcium ions were transported mainly through the vertex regions of the protein membrane giving rise to an ionic current flux. These results introduce the S-layer as a new class of nanopore sensing devices. We have also intensified our efforts in obtaining spatial control and oriented immobilization of the S-layers on solid supports to allow growth of 2D crystals over larger surface areas. We have focused on the study of the B. sphaericus S-layer, that offers the advantage of being dissasembled and recrystallized on surfaces from individual protein subunits. We have taken a dual approach on this: On one hand we have designed biomimetic interfaces, based on self assembled monolayers on gold and silicon, to exploit the natural binding mechanisms of S-layers that involve carbohydrate and electrostatic interactions. Hence, we have designed surfaces that include sugar moieties or variable surface charges. These surfaces have been fully characterized using ellipsometry and grazing angle FT-IR spectroscopy. Initial binding studies have also been performed using Surface plasmon resonance (SPR) spectroscopy and the model lectin protein (Bandeiraea simplicifolia lectin II) showing specificity to the surfaces presenting carbohydrate functionalities. We have been successful in incorporating the amino acid in the C-terminus of a truncated S-layer from Bacillus sphaericus (SbpA31-1068). This was performed by introducing the orthogonal pair tRNA/tRNA synthetase from an Archea (Methanococcus janaschii) in E. coli. This orthogonal pair recognizes an amber codon (ATG) and incorporates the unnatural amino acid in this specific point mutation when expressed in E. coli. The mutation and the incorporation of the unnatural amino acid was already done and tested. Strategies for the separation of proteins containing the unnatural amino acid, from the ones without unnatural amino acid are being optimized.

Impacts
We have achieved significance progress in realizing the full technological significance of the S-layer templated bionanofabrication, by exploring a series of applications and by controlling the long range order of the protein crystal. We have shown the great versatility of the S-layer scaffold by providing arrays of various nanoparticles (both metallic and semiconducting) with different lattice geometries. We have also shown that the S-layer membrane employed the S-layer crystal lattice as a sensing element of an ion gate. The device was sensitive and practically mono-selective for calcium with performance characteristics analogous to its solid-state counterparts. This is expected to have a great impact in the nanoporous sensing technologies, since it is introducing a completely new class of nanopores. This work is also anticipated to have implications in the general understanding of the mechanisms of ion-channel functions and in providing new insights on the possible physiological role of S-layers. Our work in controlling protein-interface interactions will have implications in understanding protein-surface interactions that are of great importance in the development of biosensor technologies (immunoassays, biochips, high throughput techniques) in biomedicine (implants) and in drug delivery systems. The carbohydrate -protein interfaces are also off great importance for cell biology since they mainly occur on the cell membrane and can elucidate the mechanism of cell adhesion.

Publications

• Mark, S., M. Bergkvist, X. Yang, L.M. Teixeira, P. Bhatnagar, E. Angert, C.A. Batt. 2006. Bionanofabrication of metallic and semiconductor nanoparticle arrays using S-layer protein lattices with different lateral spacings & geometries. Langmuir 22:3763.
• Mark, S., M. Bergkvist, X. Yang, E.R. Angert and C.A. Batt. 2006. Self-Assembly of Dendrimer-Encapsulated Nanoparticle Arrays Using 2-D Microbial S-Layer Protein Biotemplates. Biomacromolecules 7:1884.

Progress 01/01/05 to 12/31/05

Outputs
Arrays of Au nanoparticles and semiconducting quantum dots (QDs) were created through biomolecular templating methods using 2-D surface layer (S-layer) proteins isolated from the Gram positive bacterium Deinococcus radiodurans and acidothermophilic archaeon Sulfolobus acidocaldarius. Self-assembling S-layer protein lattices display a highly repetitive surface structure that makes them particularly suitable as biotemplates to fabricate ordered nanostructures and arrays. Transmission electron microscopy (TEM), Fourier transform analyses, and pair correlation function (PCF) calculations revealed that ordered nanostructured arrays with a range of spacings (7-22 nm) and different geometrical arrangements could be fabricated through the use of the two types of S-layers. These results demonstrate that it is possible to exploit the physico-chemical/structural diversity of prokaryotic S-layer scaffolds to vary the morphological patterning of nanoscale metallic and semiconductor nanoparticle arrays. In addition, we have demonstrated a novel electron-beam lithography (EBL)-based approach for patterning biological macromolecules that does not involve the use of resist, hence eliminating the exposure of these biomolecules to harsh resist stripping processes that are normally employed in conventional thin film processing routines. A non-biofouling poly(ethylene glycol) self-assembled monolayer (PEG-SAM) was selectively removed by e-beam and patterned with aldehyde terminated polyamidoamine dendrimers (ald-PAMAM-SAM) in layer-by-layer (LbL) manner. Through fluorescence microscopy techniques, we found that amino-modified ssDNA probe molecules covalently immobilized to this 2-D ultrathin-film matrix hybridized specifically to their complementary target sequences and could be stripped and then re-probed. Generation-6 (G-6) PAMAM molecules (6.7 nm diameter) terminating with 256 surface primary amine groups were used to increase the areal surface density of aldehyde functional groups, so that an enhancement in the oligonucleotide immobilization efficiency could be achieved.

Impacts
We have investigated a number of novel processes which we term bionanofabrication that harnesses biological tools to create nanostructured array with compositions that can potentially be tailored to specific functional needs. We have also developed a novel electron beam lithography-based method for patterning sub-micron arrays of biological molecules in a manner compatible with existing semiconductor manufacturing processes. The expectation is that these methods will lead to the fabrication of ordered arrays that will be useful for fundamental investigations of, for example, enhanced optical nonlinear effects. Furthermore, it is anticipated that such types of arrays may also exploited for the creation of highly efficient, portable sensor devices.

Publications

• No publications reported this period

Progress 01/01/04 to 12/31/04

Outputs
Arrays of Au nanoparticles were created using the inherent repeating patterns of bacterial surface layer (S-layer) proteins. Bacterial self-assembling S-layer protein lattices display a highly repetitive surface structure that makes them particularly suitable as biotemplates to fabricate metallic/semiconducting nanostructures and arrays. Our results demonstrate that the unique surface topography, chemical properties, and highly repetitive structure makes S-layers particularly suitable as biotemplates to fabricate nanostructures and arrays in a parallel fashion. In addition, we have developed a multistep route for the immobilization of DNA and proteins on chemically modified gold substrates using fourth-generation NH(2)-terminated poly(amidoamine) dendrimers supported by an underlying amino undecanethiol (AUT) self-assembled monolayer (SAM). Bioactive ultrathin organic films were prepared via layer-by-layer self-assembly methods and characterized by fluorescence microscopy, variable angle spectroscopic ellipsometry, atomic force microscopy (AFM), X-ray photoelectron spectroscopy (XPS), and attenuated total internal reflection Fourier transform infrared spectroscopy (ATR-FTIR). Surface plasmon resonance (SPR) studies demonstrated that sensor surfaces containing a dendrimer layer display an increased protein immobilization capacity, compared to sensor surfaces without dendrimer molecules.

Impacts
We have investigated a number of novel processes which we term bionanofabrication that harnesses biological tools to create submicron structures with compositions tailored to specific functional needs. Using both 'natural' and 'unnatural' substrates fed to an immobilized enzyme system, we are able to grow unique structures that are confined to a specified region. The expectation is that these ordered arrays will show enhanced nonlinear optical effects that can be exploited for the creation of highly efficient, portable sensor devices.

Publications

• Bergkvist, M., Mark, S.S., Yang, X., Angert, E.R., and Batt, C.A. 2004. Bionanofabrication of ordered nanoparticle arrays: Effect of particle properties and adsorption conditions. J. Phys. Chem., B. 108(24):8241-8248.
• Campagnolo C., Meyers K.J., Ryan T., Atkinson R.C., Chen Y.T., Scanlan M.J., Ritter G., Old L.J., and Batt, C.A. 2004. Real-Time, label-free monitoring of tumor antigen and serum antibody interactions. J. Biochem. Biophys. Methods. 61(3):283-98.
• Mark, S.S., Sandhyarani, N., Zhu, C., Campagnolo, C., and Batt, C.A. 2004. Dendrimer-functionalized self-assembled monolayers as a surface plasmon resonance sensor surface. Langmuir. 20:6808-6817.
• Niamsiri N., Delamarre S.C., Kim Y.R., and Batt, C.A. 2004. Engineering of chimeric class II polyhydroxyalkanoate synthases. Appl. Environ. Microbiol. 70(11):6789-99.

Progress 01/01/03 to 12/31/03

Outputs
The Nanobiotechnology Center (NBTC) was established in 1999 upon its designation as a Science and Technology Center by the National Science Foundation. Support by NSF is for an initial five years, totalling almost $20M with a second continuing five year period of support. In addition to NSF, New York State is providing $300,000 in annual support through the New York State Technology and Advance Research Agency (NYSTAR). In addition, the W.M. Keck Foundation has recently awarded seven faculty at Cornell University, $1.2M to support graduate research assistants and equipment. Faculty from six institutions, Cornell University, Princeton University, Wadsworth Center (NYS Department of Health), Oregon Health Sciences University, Clark Atlanta University and Howard University are members of the NBTC. Its adminstrative home is located in the Biotechnology Building on the Cornell University, Ithaca, New York campus. Core facilities consisting of laboratories for microfabrication, characterization of microscale structures and cellular/molecular biology are being established to support the research and training activities. The laboratories will be equipped with instrumentation including an excimer laser, a fluorescence microscope, a thermal embosser and other unique instruments necessary for fabrication and characterization of biological materials. These core facilities are being assembled with the support of the NSF and also a grant from the W.M Keck Foundation. In addition, close linkages with the Cornell Nanofabrication Facility and the Cornell Center for Materials Research create a highly complementary and robust infrastructure to support the research activities of the NBTC.

Impacts
We have established a new process which we term bionanofabricaion that harnesses biological tools to create submicron polymeric structures with compositions tailored to specific functional needs. Bionanofabrication can be used to create 3-D structures in existing microfluidic networks that can serve as both valves and filters as well as providing a robust scaffold for post synthetic modifications that could be used for a variety of applications. Using both 'natural' and 'unnatural' substrates fed to an immobilized enzyme system we are able to grow unique structures that are fitted to a specified region. Nature displays a vast diversity of nano-scale structures that are reproduced using a relatively simple set of components and which can be directly (albeit not always predictively) modified using molecular biology approaches. Fabrication of micron and submicron structures can be achieved using a variety of tools and processes with extreme precision. Progress in understanding how biology controls structure and how then those mechanisms can be exploited is the foundation for nanobiotechnology. To understand even the most simple events, such as the folding of a peptide, but also to then be able to manipulate it in a predictable fashion would unlock the vast potential of biological synthesis of micro and macro structures.

Publications

• Pollac, L., Tate, M.W., Finnefrock, A.C., Kalidas, C., Trotter, S., Darnton, N.C., Lurio, L., Austin, R.H., Batt, C.A., Gruner, S.M., and Mochrie, S.G. 2001. Time resolved collapse of a folding protein observed with small angle x-ray scattering. Phys Rev Lett 86: 4962-4965.
• Kauffman, E., Darnton, N.C., Austin, R.H., Batt, C.A. and Gerwert, K. 2001. The B-sheet to a-helix transition of B-lactoglobulin monitored in real-time with a microfabricated IR mixer. Proc. Natl. Acad. Sci. USA 98:6646-6649.
• Kuwata, K., Shatry, R., Cheng, H., Hoshino, M., Batt, C.A., Goto, Y., Roder, H. 2001. Structural and kinetic characterization of early folding events in beta-lactoglobulin. Nat Struct Biol. 8:151-155.

Progress 01/01/02 to 12/31/02

Outputs
The Nanobiotechnology Center (NBTC) was established in 1999 upon its designation as a Science and Technology Center by the National Science Foundation. Support by NSF is for an initial five years, totally almost $20M with a second continuing five year period of support. In addition to NSF, New York State is providing $300,000 in annual support through the New York State Technology and Advance Research Agency (NYSTAR). In addition, the W.M. Keck Foundation has recently awarded seven faculty at Cornell University, $1.2 M to support graduate research assistants and equipment. Faculty from six institutions, Cornell University, Princeton University, Wadsworth Center (NYS Department of Health), Oregon Health Sciences University, Clark Atlanta University and Howard University are members of the NBTC. Its administrative home is located in the Biotechnology Building on the Cornell University, Ithaca New York campus. Core facilities consisting of laboratories for microfabrication, characterization of microscale structures and cellular/molecular biology are being established to support the research and training activities. The laboratories will be equipped with instrumentation including an excimer laser, a fluorescence microscope, a thermal embosser and other unique instruments necessary for fabrication and characterization of biological materials. These core facilities are being assembled with the support of the NSF and also a grant from the W.M. Keck Foundation. In addition, close linkages with the Cornell Nanofabrication Facility and the Cornell Center for Materials Research create a highly complementary and robust infrastructure to support the research activities of the NBTC.

Impacts
Nanobiotechnology is the application of nano- and micro-fabrication methods to build tools for exploring the mysteries of biological systems. The NBTC seeks to learn from and be inspired by observing biology as well as apply our current ability to create tools to probe biological systems. The challenge is to work with novel biocompatible materials and to design new devices that can manipulate, interface and probe even single cells.

Publications

• No publications reported this period

Progress 01/01/01 to 12/31/01

Outputs
The Nanobiotechnology Center (NBTC) was established in 1999 upon its designation as a Science and Technology Center by the National Science Foundation. Support by NSF is for an initial five years, totally almost $20M with a second continuing five year period of support. In addition to NSF, New York State is providing $300,000 in annual support through the New York State Technology and Advance Research Agency (NYSTAR). In addition, the W.M. Keck Foundation has recently awarded seven faculty at Cornell University, $1.2 M to support graduate research assistants and equipment. Faculty from six institutions, Cornell University, Princeton University, Wadsworth Center (NYS Department of Health), Oregon Health Sciences University, Clark Atlanta University and Howard University are members of the NBTC. Its administrative home is located in the Biotechnology Building on the Cornell University, Ithaca New York campus. Core facilities consisting of laboratories for microfabrication, characterization of microscale structures and cellular/molecular biology are being established to support the research and training activities. The laboratories will be equipped with instrumentation including an excimer laser, a fluorescence microscope, a thermal embosser and other unique instruments necessary for fabrication and characterization of biological materials. These core facilities are being assembled with the support of the NSF and also a grant from the W.M. Keck Foundation. In addition, close linkages with the Cornell Nanofabrication Facility and the Cornell Center for Materials Research create a highly complementary and robust infrastructure to support the research activities of the NBTC.

Impacts
Nanobiotechnology is the application of nano- and micro-fabrication methods to build tools for exploring the mysteries of biological systems. The NBTC seeks to learn from and be inspired by observing biology as well as apply our current ability to create tools to probe biological systems. The challenge is to work with novel biocompatible materials and to design new devices that can manipulate, interface and probe even single cells.

Publications

• Pollack, L., Tate, M.W., Finnefrock, A.C, Kalidas, C., Trotter, S., Darnton, N.C., Lurio, L., Austin. R.H., Batt, C.A., Gruner, S.M., and Mochrie, S.G. (2001) Time resolved collapse of a folding protein observed with small angle x-ray scattering. Phys Rev Lett 86: 4962-4965
• Kauffmann, E., Darnton, N.C., Austin, R.H., Batt, C.A. and Gerwert, K. (2001) The b-sheet to a-helix transition of b-lactoglobulin monitored in real-time with a microfabricated IR mixer. Proc. Natl. Acad. Sci. USA 98:6646-6649
• Kuwata K, Shastry R, Cheng H, Hoshino M, Batt CA, Goto Y, Roder H. (2001) Structural and kinetic characterization of early folding events in beta-lactoglobulin. Nat Struct Biol. 8:151-155.