High-purity isolation of bullfrog hair bundles and subcellular and topological localization of constituent proteins.

The small number of hair cells in auditory and vestibular organs severely impedes the biochemical characterization of the proteins involved in mechano-electrical transduction. By developing an efficient and clean "twist-off" method of hair bundle isolation, and by devising a sensitive, nonradioactive method to detect minute quantities of protein, we have partially overcome this limitation and have extensively classified the proteins of the bundles. To isolate hair bundles, we glue the saccular macula of the bullfrog to a glass coverslip, expose the tissue to a molten agarose solution, and allow the agarose to solidify to a firm gel. By rotating the gel disk with respect to the fixed macula, we isolate the hair bundles by shearing them at their mechanically weak bases. The plasma membranes of at least 80% of the stereocilia reseal. To visualize the proteins of the hair bundle, we covalently label them with biotin, separate them by SDS-PAGE, and transfer them to a charged nylon membrane. We can detect less than 500 fg of protein by probing the membrane with streptavidin-alkaline phosphatase and detecting the chemiluminescent product from the hydrolysis of the substrate 3-(4-methoxyspiro-(1,2-dioxetane-3,2'-tricyclo-[3.3.1. 1(3.7)]decan)-4-yl) phenyl phosphate (AMPPD). These techniques reveal a distinct constellation of proteins in and associated with hair bundles. Several proteins, such as calmodulin, calbindin, actin, tubulin, and fimbrin, have previously been described. A second class of proteins in the preparation appears to be derived from extracellular sources. Finally, several heretofore undescribed bundle proteins are identified and characterized by their membrane topology, subcellular localization, and glycosidase and protease sensitivities.

Unconventional myosins in inner-ear sensory epithelia.

To understand how cells differentially use the dozens of myosin isozymes present in each genome, we examined the distribution of four unconventional myosin isozymes in the inner ear, a tissue that is particularly reliant on actin-rich structures and unconventional myosin isozymes. Of the four isozymes, each from a different class, three are expressed in the hair cells of amphibia and mammals. In stereocilia, constructed of cross-linked F-actin filaments, myosin-Ibeta is found mostly near stereociliary tips, myosin-VI is largely absent, and myosin-VIIa colocalizes with crosslinks that connect adjacent stereocilia. In the cuticular plate, a meshwork of actin filaments, myosin-Ibeta is excluded, myosin-VI is concentrated, and modest amounts of myosin-VIIa are present. These three myosin isozymes are excluded from other actin-rich domains, including the circumferential actin belt and the cortical actin network. A member of a fourth class, myosin-V, is not expressed in hair cells but is present at high levels in afferent nerve cells that innervate hair cells. Substantial amounts of myosins-Ibeta, -VI, and -VIIa are located in a pericuticular necklace that is largely free of F-actin, squeezed between (but not associated with) actin of the cuticular plate and the circumferential belt. Our localization results suggest specific functions for three hair-cell myosin isozymes. As suggested previously, myosin-Ibeta probably plays a role in adaptation; concentration of myosin-VI in cuticular plates and association with stereociliary rootlets suggest that this isozyme participates in rigidly anchoring stereocilia; and finally, colocalization with cross-links between adjacent stereocilia indicates that myosin-VIIa is required for the structural integrity of hair bundles.

Myosin-I nomenclature.

We suggest that the vertebrate myosin-I field adopt a common nomenclature system based on the names adopted by the Human Genome Organization (HUGO). At present, the myosin-I nomenclature is very confusing; not only are several systems in use, but several different genes have been given the same name. Despite their faults, we believe that the names adopted by the HUGO nomenclature group for genome annotation are the best compromise, and we recommend universal adoption.

Phosphate analogs block adaptation in hair cells by inhibiting adaptation-motor force production.

To ensure optimal sensitivity for mechanoelectrical transduction, hair cells adapt to prolonged stimuli using active motors. Adaptation motors are thought to employ myosin molecules as their force-producing components. We find that beryllium fluoride, vanadate, and sulfate, phosphate analogs that inhibit the ATPase activity of myosin, inhibit adaptation by abolishing motor force production. Phosphate analogs interact with a 120-kDa bundle protein, most likely myosin 1 beta, in a manner that coincides with their effects on adaptation. Features of transduction following inhibition of motor force production suggest that the gating and extent springs of the hair cell orient in parallel at rest and that the negative limit of adaptation arises when force in the stretched extent spring matches the force output of the adaptation motor.

Identification of a 120 kd hair-bundle myosin located near stereociliary tips.

By adapting to sustained stimuli, hair cells of the internal ear maintain their optimal sensitivity to minute displacements. Biophysical experiments have suggested that adaptation is mediated by a molecular motor, most likely a member of the myosin family. To provide direct evidence for the presence of myosin isozymes in hair bundles, we used photoaffinity labeling with vanadate-trapped uridine and adenine nucleotides to identify proteins of 120, 160, and 230 kd in a preparation of hair bundles purified from the bullfrog's sacculus. The photoaffinity labeling properties of these proteins, particularly the 120 kd protein, resembled those of other well-characterized myosins. A 120 kd hair-bundle protein was also recognized by a monoclonal antibody directed against a vertebrate myosin I isozyme. Immunofluorescence microscopy localized this protein near the beveled top edge of the hair bundle, the site of mechanoelectrical transduction and adaptation.

Analysis of the Proteome of Hair-Cell Stereocilia by Mass Spectrometry.

Characterization of proteins that mediate mechanotransduction by hair cells, the sensory cells of the inner ear, is hampered by the scarcity of these cells and their sensory organelle, the hair bundle. Mass spectrometry, with its high sensitivity and identification precision, is the ideal method for determining which proteins are present in bundles and what proteins they interact with. We describe here the isolation of mouse hair bundles, as well as preparation of bundle protein samples for mass spectrometry. We also describe protocols for data-dependent (shotgun) and parallel reaction monitoring (targeted) mass spectrometry that allow us to identify and quantify proteins of the hair bundle. These sensitive methods are particularly useful for comparing proteomes of wild-type mice and mice with deafness mutations affecting hair-bundle proteins.

Molecular machinery of auditory and vestibular transduction.

The hunt for molecules that conduct mechanoelectrical transduction in hair cells has recently intensified. A hair cell's transduction apparatus adapts to sustained stimuli, and myosin I beta and myosin VIIA have been advanced as candidates for the motor that mediates this process. The identity of the transduction channel remains unknown, although a viable suggestion proposes that it belongs to the amiloride-sensitive Na+ channel family.

Plasma membrane Ca2+-ATPase isoform 2a is the PMCA of hair bundles.

Mechanoelectrical transduction channels of hair cells allow for the entry of appreciable amounts of Ca(2+), which regulates adaptation and triggers the mechanical activity of hair bundles. Most Ca(2+) that enters transduction channels is extruded by the plasma membrane Ca(2+)-ATPase (PMCA), a Ca(2+) pump that is highly concentrated in hair bundles and may be essential for normal hair cell function. Because PMCA isozymes and splice forms are regulated differentially and have distinct biochemical properties, we determined the identity of hair bundle PMCA in frog and rat hair cells. By screening a bullfrog saccular cDNA library, we identified abundant PMCA1b and PMCA2a clones as well as rare PMCA2b and PMCA2c clones. Using immunocytochemistry and immunoprecipitation experiments, we showed in bullfrog sacculus that PMCA1b is the major isozyme of hair cell and supporting cell basolateral membranes and that PMCA2a is the only PMCA present in hair bundles. This complete segregation of PMCA1 and PMCA2 isozymes holds for rat auditory and vestibular hair cells; PMCA2a is the only PMCA isoform in hair bundles of outer hair cells and vestibular hair cells and is the predominant PMCA of hair bundles of inner hair cells. Our data suggest that hair cells control plasma membrane Ca(2+)-pumping activity by targeting specific PMCA isozymes to distinct subcellular locations. Because PMCA2a is the only Ca(2+) pump present at appreciable levels in hair bundles, the biochemical properties of this pump must account fully for the physiological features of transmembrane Ca(2+) pumping in bundles.

Feeling force: mechanical transduction by vertebrates and invertebrates.

Detection of mechanical stimuli requires conversion of the signal's inherent information into neuronal electrical signals. Studies of vertebrate hair cells suggest that this is accomplished by elastic links between stereocilia that control the opening of ion channels. Molecular genetics in Caenorhabditis elegans has identified candidate proteins that may be responsible for similar functions in this organism.

Inhibition and stimulation of photoreceptor phosphodiesterases by dipyridamole and M&B 22,948.

Few high affinity inhibitors of the photoreceptor phosphodiesterases have been identified. We show here that dipyridamole and M&B 22,948 (Zaprinast), potent inhibitors of the cGMP-binding, cGMP-specific phosphodiesterase (PDE), also inhibit trypsin- or transducin-activated bovine rod and cone photoreceptor phosphodiesterases at submicromolar concentrations. Dixon plots demonstrated that the inhibition of trypsin-activated rod PDE was competitive, with Ki values of 140 nM for M&B 22,948 and 380 nM for dipyridamole. Both of these drugs were much more potent than other PDE inhibitors, including isobutylmethylxanthine (IBMX). These results reinforce the suggestion that the photoreceptor and the cGMP-binding, cGMP-specific PDE are closely related. In addition, the high affinity and selectivity of these agents should make them useful for probing the regulation and function of PDE in the photoreceptor. At low substrate concentrations, the effects of these drugs on basal unactivated PDE activity were similar to those seen with trypsin- or transducin-activated PDE. At millimolar substrate concentrations, however, the effects of the drugs were biphasic; PDE activity was stimulated at drug concentrations from 1 to 10 microM and inhibited at higher concentrations. Stimulation was not observed with IBMX. This stimulation of activity apparently was not an allosteric effect caused by direct binding of the dipyridamole and M&B 22,948 to the high affinity noncatalytic cGMP binding sites on the PDEs; whereas no cooperativity of cGMP binding to this site has been demonstrated, the drugs actually stimulated the binding of low concentrations of cGMP to this site. In addition, whereas preincubation with cGMP and cGMP analogs blocked the stimulation exerted by the drugs, they did so only at much higher concentrations than those necessary for saturation of the high affinity noncatalytic cGMP site. Because the stimulation can only be seen at higher substrate levels than are thought to exist in the photoreceptor, only the inhibitory effects of the drugs are likely to be pharmacologically relevant. However, the stimulation exerted by these drugs may point to a hitherto unknown allosteric interaction between the catalytic and regulatory sites on the PDE or to a previously unrecognized regulatory site.

cGMP is tightly bound to bovine retinal rod phosphodiesterase.

Although the total concentration of cGMP in rod outer segments is thought to be substantially greater than the free concentration, no quantitatively relevant site for the bound cGMP has been described in mammalian photoreceptors. We have found that preparations of purified bovine rod photoreceptor cyclic nucleotide phosphodiesterase (PDE) contain 1.8 +/- 0.3 mol of tightly bound cGMP per mol of PDE. When subunits of the purified PDE were separated by reverse-phase HPLC in 0.1% trifluoroacetic acid and acetonitrile, a peak of material having spectral properties characteristic of a guanine ring was seen. This material was identified as cGMP by comigration with authentic cGMP on HPLC, conversion to 5-GMP by trypsin-activated rod PDE, and conversion to guanosine by a combination of trypsin-activated PDE and 5'-nucleotidase-containing snake venom. When incubated with 1 microM [3H]cGMP, only 0.1 mol of [3H]cGMP bound per mol of purified PDE, presumably because nearly all binding sites were occupied by tightly bound endogenous cGMP carried through the purification. Scatchard plots of [3H]cGMP binding have indicated that two classes of binding sites are present on the rod PDE. The off-rate of cGMP from the slowly dissociating site is extremely slow; it has a t1/2 of approximately 4 hr at 37 degrees C. At lower temperatures, very little cGMP dissociates; the amount of [3H]cGMP bound to rod PDE after 2 hr at 4 degrees C was essentially the same as at the beginning of the incubation. The observation that stoichiometric amounts of cGMP are tightly bound to PDE accounts for the inability to purify the bovine rod PDE on cGMP affinity columns or to demonstrate stoichiometric high-affinity binding sites with [3H]cGMP. More significantly, the tightly bound cGMP may resolve the apparent discrepancy between the free and total cGMP concentrations of photoreceptor outer segments.

Characterization of a bovine cone photoreceptor phosphodiesterase purified by cyclic GMP-sepharose chromatography.

The biochemical bases for the differences in cone and rod photoreceptor physiology have not been thoroughly examined because of the difficulty in obtaining cone photoreceptor components. We report here the purification and preliminary characterization of a bovine cyclic GMP phosphodiesterase (PDE) which is enriched in cone photoreceptors. The cone PDE was purified at least 15,000-fold to apparent homogeneity from bovine retinas by DEAE-cellulose and cGMP-Sepharose affinity chromatography. The trypsin-activated cone PDE hydrolyzed cGMP with efficiency similar to that of the rod PDE. However, a number of characteristics distinguished the cone PDE from the rod isozyme including the subunit structure. As previously reported, the apparent molecular weight of the cone PDE large subunit (alpha') was slightly larger than either of the large subunits of the rod PDE (93,500 versus 88,000 and 84,000). Three other smaller polypeptides were associated with the alpha' subunit (Mr = 11,000, 13,000, and 15,000), one of which (11,000) may be identical to the rod PDE gamma subunit. Cone phosphodiesterase binds at least 10-fold more cyclic GMP/mol of PDE than the rod photoreceptor isozyme. Cyclic GMP binds to this noncatalytic site with high affinity (Kd = 11 nM) and dissociates very slowly (t1/2 = 10-20 min at 37 degrees C). Purified rod transducin activated the cone PDE in solution to at least 90% of the trypsin-activated level. The concentration of rod transducin required for half-maximal activation of cone PDE (15 nM) was 50-fold lower than that necessary for half-maximal activation of rod PDE. Thus several properties of the cone phosphodiesterase clearly distinguish it from the rod isozyme and could account for some differences in cone and rod physiology.

Improved electrophoresis and transfer of picogram amounts of protein with hemoglobin.

During electrophoresis and electroblotting to transfer membranes, picogram amounts of protein can react irreversibly with the polyacrylamide matrix, preventing complete electrophoresis and efficient electroblotting. Bovine hemoglobin, but not other potential carrier proteins, mitigates this protein loss by migrating with or ahead of other proteins and scavenging reactive groups. Inclusion of 5 micrograms of hemoglobin in sample wells increases by 4-fold the amount of a radiolabeled test protein, myosin I beta, found at its appropriate 120-kDa position in sodium dodecyl sulfate-polyacrylamide gels. For electroblotting, incubating the gel with 0.25 mg/ml hemoglobin prior to transfer improves mobilization of picogram amounts of radiolabeled myosin I beta out of the gel by about 6-fold. For picogram amounts of proteins, therefore, approximately 20-fold more protein transfers to a blotting membrane when hemoglobin is used during both electrophoresis and transfer. This effect is general: transfer of radiolabeled Drosophila embryo proteins is improved dramatically by including hemoglobin in the pretransfer incubation solution. We suggest that electroblot-based detection of small amounts of protein, particularly when in the absence of other potential carrier proteins, can be improved substantially by using hemoglobin.

Chemiluminescence detection of proteins from single cells.

The analysis of proteins from single cells requires techniques of supreme sensitivity. Although radiochemical procedures are capable of detecting small amounts of electrophoretically separated proteins, their sensitivity falls short of that required for routine detection of minor components of single cells. Utilizing the avidin-biotin interaction and the alkaline phosphatase substrate 3-(4-methoxyspiro[1,2-dioxetane-3,2'- tricyclo-[3.3.1.1(3,7)]decan]-4-yl)phenyl phosphate (AMPPD), we have developed an alternative, chemiluminescence-based method for protein detection whose sensitivity exceeds that of other methods. Applying this method to a purified protein, we could detect as little as 63 fg (0.9 amol) of biotinylated bovine serum albumin. The sensitivity of the method was demonstrated by the detection of proteins from individual photoreceptor outer segments, including proteins constituting approximately 1% of the total. Chemiluminescence detection also proved extremely sensitive for immunoblotting: a comparison of five methods for detection of antibody-antigen interactions showed that the AMPPD technique was more sensitive than detection with a colorimetric alkaline phosphatase substrate, 125I-labeled protein A, 125I-labeled anti-mouse IgG, or colloidal gold-conjugated anti-mouse IgG.

Adenine nucleoside diphosphates block adaptation of mechanoelectrical transduction in hair cells.

By adapting to sustained stimuli, hair cells in the internal ear retain their sensitivity to minute transient displacements. Because one model for adaptation asserts that this process is mediated by a myosin isozyme, we reasoned that we should be able to arrest adaptation by interfering with myosin's ATPase cycle though introduction of ADP into hair cells. During tight-seal, whole-cell recordings of transduction currents in cells isolated from bullfrog (Rana catesbeiana) sacculus, dialysis with 5-25 mM ADP gave variable results. In half of the cells examined, the rate of adaptation remained unchanged or even increased; adaptation was blocked in the remaining cells. Because we suspected that the variable effect of ADP resulted from the conversion of ADP to ATP by adenylate kinase, we employed the ADP analog adenosine 5'-[beta-thio]diphosphate (ADP[beta S]), which is not a substrate for adenylate kinase. Adaptation consistently disappeared in the presence of 1-10 mM ADP[beta S]; in addition, the transduction channels' open probability at rest grew from approximately 0.1 to 0.8 or more. Both effects could be reversed by 2 mM ATP. When used in conjunction with the adenylate kinase inhibitor P1,P5-bis(5'-adenosyl) pentaphosphate (Ap5A), ADP had effects similar to those of ADP[beta S]. These results suggest that adaptation by hair cells involves adenine nucleotides, and they lend support to the hypothesis that the adaptation process is powered by a myosin motor.

Molecular basis of mechanosensory transduction.

Mechanotransduction - a cell's conversion of a mechanical stimulus into an electrical signal - reveals vital features of an organism's environment. From hair cells and skin mechanoreceptors in vertebrates, to bristle receptors in flies and touch receptors in worms, mechanically sensitive cells are essential in the life of an organism. The scarcity of these cells and the uniqueness of their transduction mechanisms have conspired to slow molecular characterization of the ensembles that carry out mechanotransduction. But recent progress in both invertebrates and vertebrates is beginning to reveal the identities of proteins essential for transduction.