Microfluidic Mixers for Studying Protein Folding

The process by which a protein folds into its native conformation is highly relevant to biology and human health yet still poorly understood. One reason for this is that folding takes place over a wide range of timescales, from nanoseconds to seconds or longer, depending on the protein¹. Conventional stopped-flow mixers have allowed measurement of folding kinetics starting at about 1 ms. We have recently developed a microfluidic mixer that dilutes denaturant ~100-fold in ~8 μs². Unlike a stopped-flow mixer, this mixer operates in the laminar flow regime in which turbulence does not occur. The absence of turbulence allows precise numeric simulation of all flows within the mixer with excellent agreement to experiment^3-4.Laminar flow is achieved for Reynolds numbers Re ≤100. For aqueous solutions, this requires micron scale geometries. We use a hard substrate, such as silicon or fused silica, to make channels 5-10 μm wide and 10 μm deep (See Figure 1). The smallest dimensions, at the entrance to the mixing region, are on the order of 1 μm in size. The chip is sealed with a thin glass or fused silica coverslip for optical access. Typical total linear flow rates are ~1 m/s, yielding Re~10, but the protein consumption is only ~0.5 nL/s or 1.8 μL/hr. Protein concentration depends on the detection method: For tryptophan fluorescence the typical concentration is 100 μM (for 1 Trp/protein) and for FRET the typical concentration is ~100 nM.The folding process is initiated by rapid dilution of denaturant from 6 M to 0.06 M guanidine hydrochloride. The protein in high denaturant flows down a central channel and is met on either side at the mixing region by buffer without denaturant moving ~100 times faster (see Figure 2). This geometry causes rapid constriction of the protein flow into a narrow jet ~100 nm wide. Diffusion of the light denaturant molecules is very rapid, while diffusion of the heavy protein molecules is much slower, diffusing less than 1 μm in 1 ms. The difference in diffusion constant of the denaturant and the protein results in rapid dilution of the denaturant from the protein stream, reducing the effective concentration of the denaturant around the protein. The protein jet flows at a constant rate down the observation channel and fluorescence of the protein during folding can be observed using a scanning confocal microscope⁵.     

Replicon size and rate of fork movement in early S of higher plant cells (Pisum sativum)

Measurements of chromosomal DNA fiber replication of cells of cultured pea root meristems in early S via autoradiography showed a 3-fold increase in rate of fork movement in the first 2 h. The initial rate was 4.5–6 μm h⁻¹ but forks active after 90 min moved at nearly 18 μm h⁻¹. The faster movement was not characteristic of all replicons. Certain fibers consisted of replicons of a smaller mean size (38–42 μm) with slowly moving forks (4.5–6 μm h⁻¹ fork⁻¹) and others had replicons almost 50 μm long with forks that moved more rapidly.     

Quantitative micro-assay for RNA/DNA hybrids in the study of nucleolar RNA from Chironomus tentans salivary gland cells

A microassay for RNA/DNA hybrids has been designed for the study of RNA from different nuclear components of Chironomus tentans salivary gland cells. The procedure comprises a scale reduction of the conventional filter method for hybridization, using ultraviolet microphotometry for quantitation of RNA and DNA. Hybridization is performed in 0.3 μl of 2 × SSC containing 1–2 × 10^-2 μg DNA, immobilized on a 0.2 mm² ‘micro-filter’, and 0.5–5 × 10⁻² μg RNA, with a specific activity of more than 10⁶ cpm/μg. Results obtained by the microtechnique are found to agree with results obtained by a large-scale, standard procedure. The applicability of the microtechnique is demonstrated in saturation and presaturation-competition experiments. RNA from micro-isolated nucleoli hybridizes a maximum of 0.22% of Chironomus tentans DNA, which corresponds to about 100 cistrons for the 38S ribosomal precursor in the haploid genome. The hybrids show a steep thermal dissociation profile with a T_m of 79 °C, close to the value expected for hybrids with a G + C content of 42%. Presaturation of filter-bound DNA by total unlabelled nucleolar RNA prevents 80% of the subsequent hybridization by labeled nucleolar Presaturation by RNA from one of the two nucleolar organizers prevents to a similar degree the subsequent hybridization by RNA from the other nucleolar organizer. This result indicates a sequence similarity of RNA transcribed in different nucleolar organizers. Further applications of the microtechnique are presented in the accompanying paper where the hybridization properties of chromosomal and nuclear sap RNA are investigated.     

Large circular mitochondrial DNA in Crithidia luciliae

A novel circular DNA, 11.3 μm in contour length, has been found in a pure kinetoplast DNA fraction of Crithidia luciliae. The mitochondrial nature of the kinetoplast and the absence of these large circular molecules in the nuclear fraction of DNA suggest that they constitute the mitochondrial genome of this species.     

A simplified mass-transfer model for visual pigments in amphibian retinal-cone outer segments

When radiolabeled precursors and autoradiography are used to investigate turnover of protein components in photoreceptive cone outer segments (COSs), the labeled components—primarily visual pigment molecules (opsins)—are diffusely distributed along the COS. To further assess this COS labeling pattern, we derive a simplified mass-transfer model for quantifying the contributions of advective and diffusive mechanisms to the distribution of opsins within COSs of the frog retina. Two opsin-containing regions of the COS are evaluated: the core axial array of disks and the plasmalemma. Numerical solutions of the mass-transfer model indicate three distinct stages of system evolution. In the first stage, plasmalemma diffusion is dominant. In the second stage, the plasmalemma density reaches a metastable state and transfer between the plasmalemma and disk region occurs, which is followed by an increase in density that is qualitatively similar for both regions. The final stage consists of both regions slowly evolving to the steady-state solution. Our results indicate that autoradiographic and cognate approaches for tracking labeled opsins in the COS cannot be effective methodologies for assessing new disk formation at the base of the COS.Abbreviations used: A, area (μm²), COS, cone outer segment, D, mass diffusion coefficient (μm²/s), h_m, mass transfer coefficient (μm/s), L, cone outer segment length (μm), PDE, partial differential equation, r, radius (μm), t, time (s), T, plasmalemma thickness (μm), u, plasmalemma or disk region (axial) velocity (μm/s), V, volume (μm³), W, plasmalemma width (μm), x, axial direction, v, disk to plasmalemma velocity (μm/s), ρ₁, disk label density, ρ₂, plasmalemma label density, ϕ, nonvoid fraction     

A sensitive fluorometric assay for the determination of DNA

A sensitive fluorometric assay for DNA determination using m-diaminobenzoic acid dihydrochloride for the reaction with deoxyribose liberated by perchloric acid is described. Fluorescence is proportional to the amount of DNA over the range 0.05–5 μg of DNA when the reaction is conducted in a volume of 400 μl and over the range 0.01–0.5 μg when the reaction is conducted in a volume of 40 μl. The assay is suitable for estimating DNA both in perchloric acid extracts and in complex mixtures with protein and RNA. The determination of DNA in cells grown as a monolayer by the described method is simple and rapid.     

Physicochemical characterization of mitochondrial DNA from soybean

Mitochondrial DNA (mtDNA) of soybean (Glycine max L.) was isolated and its buoyant density was contrasted with that of nuclear (nDNA) and chloroplast (ctDNA) DNA. Each of the three DNAs banded at a single, characteristic buoyant density when centrifuged to equilibrium in a CsCl gradient. Buoyant densities were 1.694 g/cm³ for nDNA and 1.706 g/cm³ for mtDNA. These values correspond to G-C contents of 34.7 and 46.9%, respectively. Covalently closed, circular mtDNA molecules were isolated from soybean hypocotyls by ethidium bromide-cesium chloride density gradient centrifugation. Considerable variation in mtDNA circle size was observed by electron microscopy. There were seven apparent size classes with mean lengths of 5.9 μm (class 1), 10 μm (class 2), 12.9 μm (class 3), 16.6 μm (class 4), 20.4 μm (class 5), 24.5 μm (class 6), and 29.9 μm (class 7). In addition, minicircles were observed in all preparations. Partially denatured, circular mtDNA molecules with at least one representative from six of the seven observed size classes were mapped. In class 4, there appear to be at least three distinct denaturation patterns, indicating heterogeneity within this class. It is proposed that the mitochondrial genome of soybean is distributed among the different size circular molecules, several copies of the genome are contained within these classes and that the majority of the various size molecules may be a result of recombination events between circular molecules.     

Surface area change at fertilization: Resorption of the mosaic membrane

Eggs of Strongylocentrotus purpuratus (sea urchin) have a surface area of 41,000 μm² before fertilization as determined by quantitative transmission and scanning electron microscopy. Within a minute after fertilization 18,000 cortical vesicles contribute an additional 57,000 μm² to form a mosaic membrane with the original plasma membrane. However, by 16 min after fertilization the total area of the egg is only 45,000 μm², indicating a rapid resorption of surface. Calculations of surface area depend in large part upon the numbers and dimensions of microvilli, after careful compensations are made for specimen shrinkage. The 134,000 microvilli per egg are 0.35 μm long before fertilization. They elongate to 1.0 μm in the first few minutes and then soon shorten to 0.5 μm. Even at their longest, microvilli do not accommodate all of the surface area of cortical vesicle membrane. The merger of cortical vesicle membranes and the plasma membrane was demonstrated many years ago and is not in doubt; however, this study indicates that the resulting mosaic membrane is not a long-lived, simple arithmetic combination of its components. Rather, the mosaic membrane undergoes a rapid and dynamic shrinkage by a mechanism which is not apparent on the basis of egg topography alone. The absolute values of egg surface area and dynamic changes in the surface are discussed in relation to physiological events accompanying fertilization.     

Neither loss of Bik alone, nor combined loss of Bik and Noxa, accelerate murine lymphoma development or render lymphoma cells resistant to DNA damaging drugs

The pro-apoptotic BH3-only protein, BIK, is widely expressed and although many critical functions in developmental or stress-induced death have been ascribed to this protein, mice lacking Bik display no overt abnormalities. It has been postulated that Bik can serve as a tumour suppressor, on the basis that its deficiency and loss of apoptotic function have been reported in many human cancers, including lymphoid malignancies. Evasion of apoptosis is a major factor contributing to c-Myc-induced tumour development, but despite this, we found that Bik deficiency did not accelerate Eμ-Myc-induced lymphomagenesis. Co-operation between BIK and NOXA, another BH3-only protein, has been previously described, and was attributed to their complementary binding specificities to distinct subsets of pro-survival BCL-2 family proteins. Nevertheless, combined deficiency of Bik and Noxa did not alter the onset of Eμ-Myc transgene induced lymphoma development. Moreover, although p53-mediated induction of Bik has been reported, neither Eμ-Myc/Bik^−/− nor Eμ-Myc/Bik^−/−Noxa^−/− lymphomas were more resistant than control Eμ-Myc lymphomas to killing by DNA damaging drugs, either in vitro or in vivo. These results suggest that Bik, even in combination with Noxa, is not a potent suppressor of c-Myc-driven tumourigenesis or critical for chemotherapeutic drug-induced killing of Myc-driven tumours.     

The comparative effects of short-term DNA inhibition on replicon synthesis in mammalian cells

The inter-replicon distance (ID) and rate (R) of DNA chain growth along the replicon were investigated with a [³H]TdR pulse-chase protocol in DNA autoradiographs of cells from seven different cultures of mammalian cells from various species. Asynchronous cultures were labelled with or without a 4 h pretreatment with the DNA inhibitor 5-fluorodeoxyuridine (FUdR). DNA inhibition was found to reduce both the mean ID and R by different amounts in the different cultures. This reduction appeared to correlate with the effectiveness of the inhibition in reducing cell viability. These findings generally could account for the considerable variability found in published data where FUdR pretreatment has been used. When individual values of ID and R in units of μm are plotted against each other, their relationship is given by the mean linear regressions: R = 0.26 ± 0.04 + (0.88 ± 0.05) 10⁻²ID for control, and R = 0.16 ± 0.04 + (1.04 ± 0.06) 10⁻²ID for FUdR-pretreated cultures.The relationship between ID and R in both sets of cultures suggests the presence of a regulating mechanism within a cell which maintains a relatively constant overall rate of chain growth over long stretches of DNA. A mechanism involving changes in the levels of various DNA replication complexes is suggested as one explanation for this relationship.     

Rearrangement of concanavalin A receptor sites on cells tagged with dinitrofluorobenzene: Evidence for the chemical induction of a change usually associated with malignant transformation

Normal human peripheral blood granulocytes which are tagged with 1-fluoro-2,4-dinitrobenzene (DNFB) are agglutinated by concanavalin A (ConA) in a way which resembles the pattern of reactivity displayed by leukemic cells. The present study further defines this reaction. The binding of ConA to untagged and DNP-tagged granulocytes, treated with DNFB at a ratio of 10¹¹ molecules/cell, was quantified by isotopic dilution experiments employing [³H]ConA. Similar amounts of the lectin were bound to untagged and DNP-tagged cells following incubation for 5 min at 4 °C or 30 min at 24 °C: 1.1 × 10⁵ molecules/cell, 4.6 × 10²/μ² of surface area, and 1.6 × 10³/μg of protein. The binding of [³H]ConA to both untagged and DNP-tagged cells was inhibited to the same degree by α-methylglucopyranoside (α-MG). Fixation with either glutaraldehyde or formaldehyde, which immobilizes ConA receptor sites, completely inhibited the agglutination of both untagged and DNP-tagged cells although lectin binding was unchanged. This suggests that the inhibition of agglutination was not due to the blocking of ConA-binding sites by aldehyde groups but rather to the immobilization of lectin receptors. We conclude that dinitrophenylation of normal granulocytes facilitates the rearrangement of lectin receptors in a way which resembles the ConA-induced clustering of sites which have been observed with malignant and transformed cells.     

Dictyostelium discoideum chemotaxis: Threshold for directed motion

The chemotactic response of Dictyostelium discoideum cells to stationary, linear gradients of cyclic adenosine 3′,5′-monophosphate (cAMP) was studied using microfluidic devices. In shallow gradients of less than 10⁻³ nM/μm, the cells showed no directional response and exhibited a constant basal motility. In steeper gradients, cells moved up the gradient on average. The chemotactic speed and the motility increased with increasing steepness up to a plateau at around 10⁻¹ nM/μm. In very steep gradients, above 10 nM/μm, the cells lost directionality and the motility returned to the sub-threshold level. In the regime of optimal response the difference in receptor occupancy at the front and back of the cell is estimated to be only about 100 molecules.     

The effects of actinomycin D and cordycepin on neurite formation and acetylcholinesterase activity in mouse neuroblastoma cells

Actinomycin D (actD) (0.003–0.10 μg/ml) and cordycepin (3–30 μg/ml) were used to examine the requirement of de novo RNA synthesis in the pH 6.6-induced expression of neurites and acetylcholinesterase activity in C-1300 mouse neuroblastoma cells. ActD at 0.03 and 0.10 μg/ml caused a pronounced stimulation in neurite formation following 20 h of treatment, although by 30 h exposure to actD (0.01–0.10 μg/ml), neurite formation had rapidly declined. Cordycepin (3–30 μg/ml) also inhibited neurite formation in a concentration- and time-dependent manner, although it did not produce an initial stimulation in neurite formation. The pH 6.6-induced increase in acetylcholinesterase activity was inhibited by both actD and cordycepin in a concentration- and time-dependent manner. Cell viabilities in the presence of actD and cordycepin were 90% or greater throughout the course of these studies.The effects of actD on [³H]uridine and [³H]leucine transport into cells and on incorporation into acid-insoluble material showed that actD inhibited RNA synthesis to a greater extent than it inhibited protein synthesis. Cordycepin caused only minor effects on [³H]uridine and [³H]leucine transport into cells and incorporation into acid-insoluble material; these effects were variable and neither concentration- nor time-dependent. The results of this study show that actD can inhibit the pH 6.6-induced expression of neurites and acetylcholinesterase activity in mouse neuroblastoma cells at concentrations which were relatively non-toxic and which caused a greater inhibition of RNA synthesis than of protein synthesis. This suggests that de novo RNA synthesis is required for the expression and maintenance of neurites and acetylcholinesterase activity in mouse neuroblastoma cells. Experiments with cordycepin were consistent with this conclusion.     

Polyamine content of a non-aqueously isolated chromosome preparation

Chromosomes were isolated from synchronized metaphase HeLa cells by a non-aqueous technique. Labelling of the cells with [¹⁴C]polyamine demonstrated that there is very little redistribution of polyamine from the cytoplasm using this method. The chromosomes were not obtained singly but were released from the cells in clusters. The polyamine content of these clusters was estimated as 135.9 pmoles spermine/μg DNA and 116.1 pmoles spermidine/μg DNA, producing a spermidine: spermine ratio of 1:1.17 as compared with a ratio of 1:0.67 for the whole cell.     

Cytochrome c2 and reaction center of Rhodospeudomonas spheroides Ga. membranes. Extinction coefficients, content, half-reduction potentials, kinetics and electric field alterations

1. The reduced minus oxidized extinction coefficients (Δ^red-ox) of reaction center P605 when in the chromatophore is about 20% smaller than in the detergent-isolated state. Presumably the coupling of the reaction center protein to the antenna bacteriochlorophylls and carotenoids causes this hypochromism. The chromatophore values for P605 are 19.5 mM⁻¹ · cm⁻¹ with the spectrophotometer on single beam mode at 605 nm, and 29.8 mM⁻¹ · cm⁻¹ on dual wavelength mode set at 605 – 540 nm. Cytochrome c₂, which is not affected by detergent, has a Δ^red-ox value at 550-540 nm of 19.0 mM⁻¹ · cm⁻¹.2. The total bacteriochlorophyll to reaction center bacteriochlorophyll protein (P) ratio is about 100 : 1. The cytochrome c₂: reaction center protein ratio approaches 2. In current French press chromatophore preparations, about 70% of the reaction centers are each associated on a rapid kinetic basis with two cytochrome c₂ molecules (intact P-c₂ units). The remaining reaction center proteins are not associated with cytochrome c₂ on a kinetically viable basis and may be the result of damage incurred during mechanical rupture of the cells.3. The half-reduction potential of cytochrome c₂ in the isolated state is 345 mV. In the chromatophore, two electrochemical species of cytochrome c₂ are recognized. The majority has a value of approx. 295 mV and is identifiable with cytochrome c₂ in a reaction center protein-associated state (kinetically active, intact P-c₂ units); the remainder has an approx. 350 mV half-reduction potential and is probably cytochrome c₂ in the “free” or reaction center-dissociated state (possibly from damaged P-c₂ units). It appears that there is no exchange of cytochrome c₂ between the reaction center-associated and the reaction center-dissociated state.4. The half-reduction potential of cytochrome c₂ is pH independent (from pH 5 to 9) whether measured in the free state or when associated with the chromatophore membrane. This shows that a proton is not involved in the oxidation and reduction of cytochrome c₂ in the physiological pH range.5. The kinetics of the intact reaction center, P, and cytochrome c₂ units in chromatophores and whole cells of Rhodopseudomonas spheroides are described. The two cytochrome c₂ molecules which are associated with one P exhibit similar oxidation kinetics; both are biphasic. The fast phase is estimated to be 20–40 μs in half time. The second slower phase is variable depending on the ionic strength of the medium used for the preparation of the chromatophores; it varies from 0.3 to 8 ms.6. An equilibrium for cytochrome c₂ and the reaction center and/or the membrane is suggested. The two states of the equilibrium are described by a population of cytochrome c₂ functionally “close” to the P⁺, and a population functionally distant from the P⁺, which might be physically off the binding site, or orientated unfavorably to the P⁺. The former population is identified by the 20–40 μs oxidation rate; the latter variable and somewhat slower oxidation (0.3–8 ms) is that whose rate is governed by the diffusional processes of the equilibrium which brings the cytochrome to the close position.7. Carotenoid bandshifts are kinetically compatible (a) with the P oxidation which is too fast to measure, and (b) with the two phases of cytochrome c₂ oxidation. These are interpreted as arising from local electric field alterations occurring during the electron transfer events in the reaction center and cytochrome c₂.     

A chemical and spectral characterization of the extrapallial fluid of Mytilus edulis

The extrapallial fluid of Mytilus edulis is analyzed for inorganic and organic constituents. Disc gel electrophoresis reveals the presence of at least five protein components. The insoluble fraction is sulfated carbohydrate material while the ultrafiltrable portion contains free amino acids. The concentration of calclium is 9.8 ± 0.4 m , 84.7% of which is complexed; 74.3% of the calcium is bound to small molecules, 9.2% to insoluble carbohydrate, 0.88% to soluble macromolecular components, and 15.3% is free. The data suggest that CaCO₃ deposition may be controlled by complexation of Ca²⁺ by small chelates produced by the animal. EPR measurements indicate that the majority of manganese present in the fluid is also bound to small molecules. Titrations of the native fluid with Mn²⁺ and H⁺ establish that the chelating capacity of the fluid for Mn²⁺ is nominally 10⁻⁴ . Mn²⁺ forms 1:1 complexes in the fluid with ligands having an average pK_a′ value of 5.2     

Similar replicon properties of higher plant cells with different S periods and genome sizes

Root meristematic cells of nine unrelated diploid higher plants with genome sizes that differ 82-fold and with S periods that differ 4-fold have similar replicon sizes and single replication fork rates that average 22 μm and 8 μm/h respectively. The average replicon size of 22 μm is near the 18 μm obtained by extrapolation of measurements, taken from DNA fiber autoradiograms, to zero pulse time with [³H]thymidine. The data suggest that the duration of S is determined by the minimal number of replicon families that function sequentially during DNA replication.     

Measurement of 5′-adenylic acid by stimulation of phosphorylase a

A method for measuring 5′-adenylic acid is described which is based on its capacity to stimulate the activity of phosphorylase a in the presence of low levels of glycogen and inorganic P. The method is useful with very low concentrations of AMP (0.02 to 0.5 μm) and is unaffected by much higher levels of ADP and ATP. Other common mononucleotides do not interfere at concentrations likely to be encountered. Yields of final product (NADPH) as high as 500 mol/mol of AMP can be obtained. Protocols are given for measuring as much as 5 × 10⁻¹⁰ mol at one extreme and as little as 10⁻¹⁵ mol at the other.     

Lactose-H(OH) transport system of Escherichia coli Multistate gated pore model based on half-sites stoichiometry for high-affinity substrate binding in a symmetrical dimer

A model is proposed for the d-galactoside-H⁺(⁻OH) transporter of Escherichia coli that accounts for essentially all the experimental observations established for this system to date. In this model, the functional unit is postulated to be a dimer (consisting of two copies of lacY-specified polypeptide) which spans the membrane with a 2-fold symmetry axis in the membrane plane (Lancaster, J.R. (1978) J. Theor. Biol. 75, 35–50). The functional dimer is assumed to possess a single pore flanked by an inner gate (g_i) and an outer gate (g_o) and encompassing two oppositely oriented galactoside binding sites, designated m and μ. When g_o is open and g_i is closed under non-energized conditions, binding site m adopts a configuration defined as State A (i.e., m_o^A) exhibiting high affinity toward Class G^a galactosides (thiodigalactoside, melibiose, α-p-nitrophenylgalactoside) but low affinity for Class G^b galactosides (lactose, β-o-nitrophenylgalactoside, β-isopropylthiogalactoside), whereas binding site μ adopts State B (i.e., μ_o^B) displaying relatively high affinity toward Class G^b galactosides but comparatively low affinity for Class G^a galactosides; further, each m_o^A : μ_o^B dimer contains one thiol group whose reaction with N-ethylmaleimide inactivates the transporter unless blocked by galactoside binding at site m_o^A, while the second homologous thiol of the dimer is unreactive toward thiol reagents. Translocation of the m_o^A : μ_o^B dimer involves closing of g_o followed by opening of g_i, and causes the two thiols (as well as sites m and μ) to interchange roles in a symmetrical fashion: m_o^A : μ_o^B ↔ m_i^B : μ_i^A. In the presence of a substantial (negative) transmembrane Δμ~_H⁺, the m : μ dimer is postulated to undergo an electrogenic protein conformational change to a second form, ^*(m : μ), in which both sites m and μ possess low affinity toward internal Class G^b substrates; galactoside transport in both m : μ and ^*(m : μ) is assumed to be coupled to H⁺-symport (⁻OH-antiport) with a stoichiometry of approximately 1 : 1. Finally, five characteristic predictions of the half-sites model are outlined for further tests of its validity.