Lower vertebrates have an intrinsically-photosensitive iris and thus a local pupillary light reflex (PLR). In contrast, it has been a dogma that the PLR in mammals generally requires neuronal circuitry connecting the eye and the brain. We report here that an intrinsic component of the PLR is actually widespread in nocturnal and crepuscular mammals. In mouse, this intrinsic PLR requires the visual pigment, melanopsin. It also requires PLCβ4, the vertebrate homolog of the Drosophila NorpA phospholipase C mediating rhabdomeric phototransduction. The Plcβ4−/− genotype, besides removing the intrinsic PLR, also essentially eliminates the intrinsic light response of the M1-subtype of melanopsin-expressing, intrinsically-photosensitive retinal ganglion cells (M1-ipRGCs), by far the most photosensitive ipRGCs and with the largest responses. Ablating in mouse the expression of both TRPC6 and TRPC7, members of the TRP channel superfamily, likewise essentially eliminated the M1-ipRGC light response, but spared the intrinsic PLR. Thus, melanopsin signaling exists in both iris and retina, involving a PLCβ4-mediated pathway that nonetheless diverges in the two locations.
The Ca2+-binding protein recoverin may regulate visual transduction in retinal rods and cones, but its functional role and mechanism of action remain controversial. We compared the photoresponses of rods from control mice and from mice in which the recoverin gene was knocked out. Our analysis indicates that Ca2+-recoverin prolongs the dark-adapted flash response and increases the rod's sensitivity to dim steady light. Knockout rods had faster Ca2+ dynamics, indicating that recoverin is a significant Ca2+ buffer in the outer segment, but incorporation of exogenous buffer did not restore wild-type behavior. We infer that Ca2+-recoverin potentiates light-triggered phosphodiesterase activity, probably by effectively prolonging the catalytic activity of photoexcited rhodopsin.
calcium-binding proteins; knockout mice; light adaptation; photoreceptors; phototransduction
Bacterial chemotaxis results from the ability of flagellated bacteria to control the frequency of switching between smooth-swimming and tumbling episodes in response to changes in concentration of extracellular substances. High levels of phosphorylated CheY protein are the intracellular signal for inducing the tumbling mode of swimming. The CheZ protein has been shown to control the level of phosphorylated CheY by regulating its rate of dephosphorylation. To identify functional domains in the CheZ protein, we made mutants by random mutagenesis of the cheZ gene and constructed a series of deletions. The map position and the in vivo and in vitro activity of the resulting gain- or loss-of-function mutant proteins define separate functional domains of the CheZ protein.
An expressed sequence tag homologous to cheA was previously isolated by random sequencing of Thermotoga maritima cDNA clones (C. W. Kim, P. Markiewicz, J. J. Lee, C. F. Schierle, and J. H. Miller, J. Mol. Biol. 231: 960-981, 1993). Oligonucleotides complementary to this sequence tag were synthesized and used to identify a clone from a T. maritima lambda library by using PCR. Two partially overlapping restriction fragments were subcloned from the lambda clone and sequenced. The resulting 5,251-bp sequence contained five open reading frames, including cheA, cheW, and cheY. In addition to the chemotaxis genes, the fragment also encodes a putative protein isoaspartyl methyltransferase and an open reading frame of unknown function. Both the cheW and cheY genes were individually cloned into inducible Escherichia coli expression vectors. Upon induction, both proteins were synthesized at high levels. T. maritima CheW and CheY were both soluble and were easily purified from the bulk of the endogenous E. coli protein by heat treatment at 80 degrees C for 10 min. CheY prepared in this way was shown to be active by the demonstration of Mg(2+)-dependent autophosphorylation with [32P]acetyl phosphate. In E. coli, CheW mediates the physical coupling of the receptors to the kinase CheA. The availability of a thermostable homolog of CheW opens the possibility of structural characterization of this small coupling protein, which is among the least well characterized proteins in the bacterial chemotaxis signal transduction pathway.
A gene encoding only the C-terminal portion of the receptor-transducer protein Tar of Escherichia coli was constructed. The gene product was detected and localized in the cytoplasmic fraction of the cell by immunoblotting with anti-Tar antibodies. The C-terminal fragments from wild-type and mutant tar genes were characterized in vivo. The C-terminal fragment generated from tar-526, a mutation that results in a dominant "tumble" phenotype, was found to be deamidated and methylated by the CheB and CheR proteins, respectively. The C-terminal fragment derived from a wild-type gene was poorly deamidated, and the C-terminal fragment derived from tar-529, a dominant mutant with a "smooth swimming" phenotype, was not apparently modified. Cells carrying the C-terminal fragment with the tar-526 mutation as the sole receptor-transducer protein showed a high frequency of tumbling and chemotaxis responses to changes in intracellular pH. These results suggest that the cytoplasmic C-terminal fragment of Tar retains some of the functions of the whole protein in vivo.
The nucleotide sequence of DNA which contains five chemotaxis-related genes of Escherichia coli, cheW, cheR, cheB, cheY, and cheZ, and part of the cheA gene was determined. Molecular weights of the polypeptides encoded by these genes were calculated from translated amino acid sequences, and they were 18,100 for cheW, 32,700 for cheR, 37,500 for cheB, 14,100 for cheY, and 24,000 for cheZ. Nucleotide sequences which could act as ribosome-binding sites were found in the upstream region of each gene. After the termination codon of the cheW gene, a typical rho-independent transcription termination signal was observed. There are no other open reading frames long enough to encode polypeptides in this region except those which code for the two previously reported genes tar and tap.
Flagellar proteins controlling motility and chemotaxis in Escherichia coli were selectively labeled in vivo with [35S]methionine. This distribution of these proteins in subcellular fractions was examined by sodium dodecyl sulfatepolyacrylamide gel electrophoresis and autoradiography. The motA, motB, cheM, and cheD gene products were found to be confined exclusively to the inner cytoplasmic membrane fraction, whereas the cheY, cheW, and cheA (66,000 daltons) polypeptides appeared only in the soluble cytoplasmic fraction. The cheB, cheX, cheZ, and cheA (76,000 daltons) proteins, however, were distributed in both the cytoplasm and the inner membrane fractions. The hag gene product (flagellin) was the only flagellar protein examined that copurified with the outer lipopolysaccharide membrane. Differences in the intracellular locations of the che and mot gene prodcuts presumably reflect the functional attributes of these components.
A procedure for the isolation of Bacillus subtilis mutants that appear to be defective in septum synthesis has been described. Fourteen mutants isolated with this technique were found to be located at four distinct loci on the B. subtilis chromosome. These have been designated divA, divB, divC, and divD. The four mutants in the divA group synthesize septa; however, they do so with a high frequency of error resulting in minicell production. Mapping data were obtained by scoring cotransduction frequencies using PBS1-transducing lysates. Thus, some of the genes apparently involved in septum synthesis and their order on the genome have been established. The results of a study by electron microscope of some of the mutants is also presented.
A method for preparing bacterial flagellar hook structures is described. The method involves isolating intact flagella from a mutant which makes thermally labile flagellar filaments and heat-treating them to disaggregate the filament preferentially. The resulting hook preparation can be separated and purified by velocity and isopycnic centrifugation. The purified hooks sediment at a relative S value of 77. On acrylamide gel electrophoresis in sodium dodecyl sulfate, they show one major and a number of minor protein bands. The purified hooks can be used to immunize rabbits, and the resulting antiserum is hook-specific. These results support the notion that hooks are composed of a protein that differs from flagellin.
The CheA kinase is a central protein in the signal transduction network that controls chemotaxis in Escherichia coli. CheA receives information from a transmembrane receptor (e.g., Tar) and CheW proteins and relays it to the CheB and CheY proteins. The biochemical activities of CheA proteins truncated at various distances from the carboxy terminus were examined. The carboxy-terminal portion of CheA regulates autophosphorylation in response to environmental signals transmitted through Tar and CheW. The central portion of CheA is required for autophosphorylation and is also presumably involved in dimer formation. The amino-terminal portion of CheA was previously shown to contain the site of autophosphorylation and to be able to transfer the phosphoryl group to CheB and CheY. These studies further delineate three functional domains of the CheA protein.
A cDNA corresponding to a known G protein alpha subunit, the alpha subunit of Go (Go alpha), was isolated and sequenced. The predicted amino acid sequence of C. elegans Go alpha is 80-87% identical to other Go alpha sequences. An mRNA that hybridizes to the C. elegans Go alpha cDNA can be detected on Northern blots. A C. elegans protein that crossreacts with antibovine Go alpha antibody can be detected on immunoblots. A cosmid clone containing the C. elegans Go alpha gene (goa-1) was isolated and mapped to chromosome I. The genomic fragments of three other C. elegans G protein alpha subunit genes (gpa-1, gpa-2, and gpa-3) have been isolated using the polymerase chain reaction. The corresponding cosmid clones were isolated and mapped to disperse locations on chromosome V. The sequences of two of the genes, gpa-1 and gpa-3, were determined. The predicted amino acid sequences of gpa-1 and gpa-3 are only 48% identical to each other. Therefore, they are likely to have distinct functions. In addition they are not homologous enough to G protein alpha subunits in other organisms to be classified. Thus C. elegans has G proteins that are identifiable homologues of mammalian G proteins as well as G proteins that appear to be unique to C. elegans. Study of identifiable G proteins in C. elegans may result in a further understanding of their function in other organisms, whereas study of the novel G proteins may provide an understanding of unique aspects of nematode physiology.
The carboxy-terminal half of the Escherichia coli Tar chemoreceptor protein was cloned into an overproducing plasmid with the transcription of the insert under the control of the strong hybrid tac promoter. Two dominant mutations in the tar gene, which result in "tumble-only" (tar-526) or "swim-only" (tar-529) phenotypes and which are postulated to produce proteins locked in specific signalling modes, were introduced separately onto the overproducing plasmid. After induction with isopropyl-beta-D-thiogalactopyranoside, cells containing the plasmids produced about 10% of their soluble cellular protein as the carboxy-terminal fragments. A scheme to purify the overproduced fragments was developed. Typical yields of pure fragment were 5, 30, and 20 mg per liter of induced culture for the wild type, 526 mutant, and 529 mutant, respectively. Fast-protein liquid chromatography-gel filtration analysis of the pure fragments showed that they all existed as oligomers (ca. 103,000 daltons), possibly trimers or tetramers (monomer size is 31,000 daltons). However, the 529 mutant fragment showed an additional oligomeric form (240,000 daltons) corresponding approximately to an octamer. When chromatographed in the presence of 1% octylglucoside, all three fragments showed an identical single oligomeric size of about 135,000 daltons. Further differences between the fragments such as ion-exchange behavior and susceptibility to degradation were found. Taken together, these results suggest that conformational differences between the 529 mutant fragment and the other fragments exist and that these differences may correlate with the phenotypic effects of the tar-529 mutation.
Hydroxylamine mutagenesis was used to alter the tar gene that encodes the transmembrane Tar protein required for chemotaxis. Mutants defective in chemotaxis were selected, and the mutation was characterized by DNA sequencing. Two classes of mutations were found: nonsense and missense. The nonsense mutations were distributed throughout the gene, while the missense mutations were found to cluster in a region that includes 185 amino acids at the C-terminal end of the Tar protein. Partial characterization of mutant phenotypes suggested that some are completely defective in signaling while responding to attractants and repellents by differential methylation. Other mutants are undermethylated and constantly tumble, while yet another class of mutants is overmethylated and biased toward constant swimming with little or no tumbling. These mutants will be useful in experiments designed to understand the mechanism of chemotaxis.
The alternate expression of the Salmonella flagellin genes H1 and H2 is controlled by the orientation of a 995-base-pair invertible segment of DNA located at the 5' end of the H2 gene. The hin gene, which is encoded within the invertible region, is essential for the inversion of this DNA segment. We cloned the hin gene into Escherichia coli and placed it under the control of the PL promoter of bacteriophage lambda. These cells overproduced the Hin protein. In vivo inversion activity was measured by using a recombinant lambda phage which contains the H2 and lacZ genes under the control of the invertible region. Using this phage, we showed that the amount of inversion activity is proportional to the amount of Hin protein in the cell. An inactive form of the protein was purified by using the unusual solubility properties of the overproduced protein. The amino acid composition of the protein agreed with the DNA sequence of the hin gene. Antibodies were made to the isolated protein. These antibodies cross-reacted with two other unidentified E. coli proteins.
The murine homologs of two viral oncogenes associated with tyrosine-specific kinase activity have been assigned to different loci in the mouse genome. The segregation of restriction site polymorphisms, as detected by probes that are specific for endogenous c-fes and c-src sequences, was followed in the DNA of recombinant inbred strains. The c-fes gene was mapped to the proximal portion of chromosome 7, very close to the Gpi-1 locus, whereas c-src was linked to the Psp locus on the distal half of chromosome 2.
The tsr and tar genetic loci of Escherichia coli determine the presence in sodium dodecyl sulfate-polyacrylamide gel electrophoresis of methyl-accepting chemotaxis proteins (MCPs) I and II, respectively, each of which consists of a distinct group of multiple bands. Synthesis of the tsr and tar products was directed in ultraviolet-irradiated bacteria by lambda transducing phages. The addition of appropriate chemotactic stimuli to these cells resulted in the appearance of additional, faster migrating electrophoretic forms of the Tsr and Tar polypeptides which disappeared upon removal of the stimulus. The stimulus-elicited forms comigrated with component bands of the corresponding MCPs. These results indicate that methylation itself caused shifts in electrophoretic mobility and hence led to the observed MCP band patterns. The number of Tsr species suggested that there were at least three methylated sites on the Tsr polypeptide. The conclusion that methylation generates multiplicity was supported by the results of experiments in which the tsr product was synthesized in mutant bacteria defective in specific chemotaxis functions concerned with methylation or demethylation of MCPs. Thus, the presence of a cheX defect blocked the stimulus-elicited appearance of faster migrating forms of the tsr product; conversely, the presence of a cheB defect resulted in a pronounced shift toward these forms in the absence of a chemotactic stimulus.
Purified flagellins derived from 16 strains of Bacillus subtilis were classified into at least five distinct groups on the basis of their reaction with antiflagellar filament antibody and antiflagellin antibody. This classification was in good accord with that derived independently on the basis of amino acid analyses of the flagellins. Flagellar antigenicity appears to provide a useful typological character in classifying B. subtilis strains.
We have developed a computer program that simulates the intracellular reactions mediating the rapid (nonadaptive) chemotactic response of Escherichia coli bacteria to the attractant aspartate and the repellent Ni2+ ions. The model is built from modular units representing the molecular components involved, which are each assigned a known value of intracellular concentration and enzymatic rate constant wherever possible. The components are linked into a network of coupled biochemical reactions based on a compilation of widely accepted mechanisms but incorporating several novel features. The computer motor shows the same pattern of runs, tumbles and pauses seen in actual bacteria and responds in the same way as living bacteria to sudden changes in concentration of aspartate or Ni2+. The simulated network accurately reproduces the phenotype of more than 30 mutants in which components of the chemotactic pathway are deleted and/or expressed in excess amounts and shows a rapidity of response to a step change in aspartate concentration similar to living bacteria. Discrepancies between the simulation and real bacteria in the phenotype of certain mutants and in the gain of the chemotactic response to aspartate suggest the existence of additional as yet unidentified interactions in the in vivo signal processing pathway.
Instability of complex mammalian genomic DNA inserts is commonplace in cosmid libraries constructed in conventional multicopy vectors. To develop a means to construct stable libraries, we have developed a low copy number cosmid vector based on the E. coli F factor replicon (Fosmid). We have tested relative stability of human DNA inserts in Fosmids and in two conventional multicopy vectors (Lawrist 16 and Supercos) by comparing the frequency of changes in restriction patterns of the inserts after propagating randomly picked human genomic clones based on these vectors. We found that the clones based on Fosmid vector undergo detectable changes at a greatly reduced frequency. We also observed that sequences that undergo drastic rearrangements and deletions during propagation in a conventional vector were stably propagated when recloned as Fosmids. The results indicate that Fosmid system may be useful for constructing stable libraries from complex genomes.
We have previously shown that asymmetric-voltage field inversion electrophoresis produces more uniform separation for fragments between 1 and 50 kilobases (kb) than other modes of pulsed field gel electrophoresis. We now report on the basis of this phenomenon. As in conventional electrophoresis, the pulsed field mobility of DNAs between 1 and 50 kb varies with voltage in a size dependent manner. The complex migration pattern obtained with asymmetric-voltage field inversion electrophoresis reflects the difference between the mobilities of each sized fragment under the conditions used for the forward and reverse fields. We have applied this technique to DNA sequencing gels and find improvement in resolution for single-stranded fragments in polyacrylamide gels.
Quantitative measurement of DNA migration in gel electrophoresis requires precisely controlled homogeneous electric fields. A new electrophoresis system has allowed us to explore several parameters governing DNA migration during homogeneous field pulsed field gel (PFG) electrophoresis. Migration was measured at different switch times, temperatures, agarose concentrations, and voltage gradients. Conditions which increase DNA velocities permit separation over a wider size range, but reduce resolution. We have also varied the angle between the alternating electric fields. Reorientation angles between 105 degrees and 165 degrees give equivalent resolution, despite significant differences in DNA velocity. Separation of DNA fragments from 50 to greater than 7000 kilobases (Kb) can easily be optimized for speed and resolution based on conditions we describe.
We describe a technique that uses reverse transcription and the polymerase chain reaction (pcr) to rapidly quantitate numbers of specific mRNA transcripts from nanogram quantities of total cellular RNA. Linearity of input molecules to output signal was maintained by limiting the cycle number and the amount of input RNA and by minimizing the number of manipulations. Absolute levels of specific transcripts were determined by the inclusion of a separate standard curve composed of serially diluted in vitro transcribed RNA run alongside the experimental samples. This allowed rapid quantitation of many samples simultaneously. We applied this technique to measuring the expression of phosphoglycerate kinase 2 (Pgk-2) transgenes in the mouse testis during development. A human PGK-2 transgene, a PGK-2/CAT transgene, and the endogenous mPgk-2 gene all displayed similar patterns and levels of expression, consistent with the conclusion that peak RNA accumulation occurs in pachytene spermatocytes. Mouse protamine 2 (mP2) is expressed at a level approximately tenfold higher than Pgk-2 and displays a different pattern of expression consistent with initiation of transcription occurring in haploid round spermatids.