A CAG repeat threshold for therapeutics targeting somatic instability in Huntington's disease

The Huntington's disease mutation is a CAG repeat expansion in the huntingtin gene that results in an expanded polyglutamine tract in the huntingtin protein. The CAG repeat is unstable and expansions of hundreds of CAGs have been detected in Huntington's disease post-mortem brains. The age of disease onset can be predicted partially from the length of the CAG repeat as measured in blood. Onset age is also determined by genetic modifiers, which in six cases involve variation in DNA mismatch repair pathways genes. Knocking-out specific mismatch repair genes in mouse models of Huntington's disease prevents somatic CAG repeat expansion. Taken together, these results have led to the hypothesis that somatic CAG repeat expansion in Huntington's disease brains is required for pathogenesis. Therefore, the pathogenic repeat threshold in brain is longer than (CAG)40, as measured in blood, and is currently unknown. The mismatch repair gene MSH3 has become a major focus for therapeutic development, as unlike other mismatch repair genes, nullizygosity for MSH3 does not cause malignancies associated with mismatch repair deficiency. Potential treatments targeting MSH3 currently under development include gene therapy, biologics and small molecules, which will be assessed for efficacy in mouse models of Huntington's disease. The zQ175 knock-in model carries a mutation of approximately (CAG)185 and develops early molecular and pathological phenotypes that have been extensively characterized. Therefore, we crossed the mutant huntingtin allele onto heterozygous and homozygous Msh3 knockout backgrounds to determine the maximum benefit of targeting Msh3 in this model. Ablation of Msh3 prevented somatic expansion throughout the brain and periphery, and reduction of Msh3 by 50% decreased the rate of expansion. This had no effect on the deposition of huntingtin aggregation in the nuclei of striatal neurons, nor on the dysregulated striatal transcriptional profile. This contrasts with ablating Msh3 in knock-in models with shorter CAG repeat expansions. Therefore, further expansion of a (CAG)185 repeat in striatal neurons does not accelerate the onset of molecular and neuropathological phenotypes. It is striking that highly expanded CAG repeats of a similar size in humans cause disease onset before 2 years of age, indicating that somatic CAG repeat expansion in the brain is not required for pathogenesis. Given that the trajectory for somatic CAG expansion in the brains of Huntington's disease mutation carriers is unknown, our study underlines the importance of administering treatments targeting somatic instability as early as possible.

Benefits of global mutant huntingtin lowering diminish over time in a Huntington's disease mouse model

We have developed an inducible Huntington's disease (HD) mouse model that allows temporal control of whole-body allele-specific mutant huntingtin (mHtt) expression. We asked whether moderate global lowering of mHtt (~50%) was sufficient for long-term amelioration of HD-related deficits and, if so, whether early mHtt lowering (before measurable deficits) was required. Both early and late mHtt lowering delayed behavioral dysfunction and mHTT protein aggregation, as measured biochemically. However, long-term follow-up revealed that the benefits, in all mHtt-lowering groups, attenuated by 12 months of age. While early mHtt lowering attenuated cortical and striatal transcriptional dysregulation evaluated at 6 months of age, the benefits diminished by 12 months of age, and late mHtt lowering did not ameliorate striatal transcriptional dysregulation at 12 months of age. Only early mHtt lowering delayed the elevation in cerebrospinal fluid neurofilament light chain that we observed in our model starting at 9 months of age. As small-molecule HTT-lowering therapeutics progress to the clinic, our findings suggest that moderate mHtt lowering allows disease progression to continue, albeit at a slower rate, and could be relevant to the degree of mHTT lowering required to sustain long-term benefits in humans.

Dynamic single-cell RNA sequencing identifies immunotherapy persister cells following PD-1 blockade

Resistance to oncogene-targeted therapies involves discrete drug-tolerant persister cells, originally discovered through in vitro assays. Whether a similar phenomenon limits efficacy of programmed cell death 1 (PD-1) blockade is poorly understood. Here, we performed dynamic single-cell RNA-Seq of murine organotypic tumor spheroids undergoing PD-1 blockade, identifying a discrete subpopulation of immunotherapy persister cells (IPCs) that resisted CD8+ T cell-mediated killing. These cells expressed Snai1 and stem cell antigen 1 (Sca-1) and exhibited hybrid epithelial-mesenchymal features characteristic of a stem cell-like state. IPCs were expanded by IL-6 but were vulnerable to TNF-α-induced cytotoxicity, relying on baculoviral IAP repeat-containing protein 2 (Birc2) and Birc3 as survival factors. Combining PD-1 blockade with Birc2/3 antagonism in mice reduced IPCs and enhanced tumor cell killing in vivo, resulting in durable responsiveness that matched TNF cytotoxicity thresholds in vitro. Together, these data demonstrate the power of high-resolution functional ex vivo profiling to uncover fundamental mechanisms of immune escape from durable anti-PD-1 responses, while identifying IPCs as a cancer cell subpopulation targetable by specific therapeutic combinations.

Proprotein convertase subtilisin/kexin type 9 (PCSK9) affects gene expression pathways beyond cholesterol metabolism in liver cells

Proprotein convertase subtilisin/kexin type 9 (PCSK9) induces degradation of low-density lipoprotein receptor (LDLR) in the liver. It is being pursued as a therapeutic target for LDL-cholesterol reduction. Earlier genome-wide gene expression studies showed that PCSK9 over-expression in HepG2 cells resulted in up-regulation of genes in cholesterol biosynthesis and down-regulation of genes in stress response pathways; however, it was not known whether these changes were directly regulated by PCSK9 or were secondary to PCSK9-induced changes to the intracellular environment. In order to further understand the biological function of PCSK9 we treated HepG2 cells with purified recombinant wild type (WT) and D374Y gain-of-function PCSK9 proteins for 8, 24, and 48 h, and used microarray analysis to identify genome-wide expression changes and pathways. These results were compared to the changes induced by culturing HepG2 cells in cholesterol-free medium, mimicking the intracellular environment of cholesterol starvation. We determined that PCSK9-induced up-regulation of cholesterol biosynthesis genes resulted from intracellular cholesterol starvation. In addition, we identified novel pathways that are presumably regulated by PCSK9 and are independent of its effects on cholesterol uptake. These pathways included "protein ubiquitination," "xenobiotic metabolism," "cell cycle," and "inflammation and stress response." Our results indicate that PCSK9 affects metabolic pathways beyond cholesterol metabolism in HepG2 cells.

HCV genotyping using statistical classification approach

The genotype of Hepatitis C Virus (HCV) strains is an important determinant of the severity and aggressiveness of liver infection as well as patient response to antiviral therapy. Fast and accurate determination of viral genotype could provide direction in the clinical management of patients with chronic HCV infections. Using publicly available HCV nucleotide sequences, we built a global Position Weight Matrix (PWM) for the HCV genome. Based on the PWM, a set of genotype specific nucleotide sequence "signatures" were selected from the 5' NCR, CORE, E1, and NS5B regions of the HCV genome. We evaluated the predictive power of these signatures for predicting the most common HCV genotypes and subtypes. We observed that nucleotide sequence signatures selected from NS5B and E1 regions generally demonstrated stronger discriminant power in differentiating major HCV genotypes and subtypes than that from 5' NCR and CORE regions. Two discriminant methods were used to build predictive models. Through 10 fold cross validation, over 99% prediction accuracy was achieved using both support vector machine (SVM) and random forest based classification methods in a dataset of 1134 sequences for NS5B and 947 sequences for E1. Prediction accuracy for each genotype is also reported.

P518/Qrfp sequence polymorphisms in SAMP6 osteopenic mouse

Mice lacking GPR103A expression display osteopenia. Analysis of mouse quantitative trait loci literature associated with bone mineral density suggested GPR103A ligand P518/Qrfp (chromosome 2qB) as a candidate osteoporosis gene. Promoter and coding regions of mouse P518/Qrfp were sequenced from genomic DNA obtained from the osteoporosis-prone strain SAMP6 and control strains SAMR1, A/J, AKR/J, BALB/c, C3H/HeJ, C57BL/6J, and DBA/2J. Four single-nucleotide polymorphisms (SNPs) were identified in only SAMP6 genomic DNA, g.-1773 T-->C, g.110 A-->G (N37S), g.188 G-->A (R63K), and g.135 T-->C (H45H). The promoter SNP generated a novel neuron-restrictive silencing factor binding site, a repressor that decreases gene expression in nonneuronal tissues. TaqMan analysis demonstrated fivefold lower P518/Qrfp liver expression in SAMP6 versus SAMR1 or C57BL/6J control strains. Tissue distribution of human, mouse, and rat P518/Qrfp and its receptors showed expression in bone and spinal cord. A direct role for P518/Qrfp function in maintaining bone mineral density is suggested.

Posaconazole is a potent inhibitor of sterol 14alpha-demethylation in yeasts and molds

Posaconazole (POS; SCH 56592) is a novel triazole that is active against a wide variety of fungi, including fluconazole-resistant Candida albicans isolates and fungi that are inherently less susceptible to approved azoles, such as Candida glabrata. In this study, we compared the effects of POS, itraconazole (ITZ), fluconazole (FLZ), and voriconazole (VOR) on sterol biosynthesis in strains of C. albicans (both azole-sensitive and azole-resistant strains), C. glabrata, Aspergillus fumigatus, and Aspergillus flavus. Following exposure to azoles, nonsaponifiable sterols were extracted and resolved by liquid chromatography and sterol identity was confirmed by mass spectroscopy. Ergosterol was the major sterol in all but one of the strains; C. glabrata strain C110 synthesized an unusual sterol in place of ergosterol. Exposure to POS led to a decrease in the total sterol content of all the strains tested. The decrease was accompanied by the accumulation of 14alpha-methylated sterols, supporting the contention that POS inhibits the cytochrome P450 14alpha-demethylase enzyme. The degree of sterol inhibition was dependent on both dose and the susceptibility of the strain tested. POS retained activity against C. albicans isolates with mutated forms of the 14alpha-demethylase that rendered these strains resistant to FLZ, ITZ, and VOR. In addition, POS was a more potent inhibitor of sterol synthesis in A. fumigatus and A. flavus than either ITZ or VOR.

Genome wide in silico SNP-tumor association analysis

Background

Carcinogenesis occurs, at least in part, due to the accumulation of mutations in critical genes that control the mechanisms of cell proliferation, differentiation and death. Publicly accessible databases contain millions of expressed sequence tag (EST) and single nucleotide polymorphism (SNP) records, which have the potential to assist in the identification of SNPs overrepresented in tumor tissue.

Methods

An in silico SNP-tumor association study was performed utilizing tissue library and SNP information available in NCBI's dbEST (release 092002) and dbSNP (build 106).

Results

A total of 4865 SNPs were identified which were present at higher allele frequencies in tumor compared to normal tissues. A subset of 327 (6.7%) SNPs induce amino acid changes to the protein coding sequences. This approach identified several SNPs which have been previously associated with carcinogenesis, as well as a number of SNPs that now warrant further investigation

Conclusions

This novel in silico approach can assist in prioritization of genes and SNPs in the effort to elucidate the genetic mechanisms underlying the development of cancer.

Quantification of single nucleotide polymorphisms by automated DNA sequencing

Single nucleotide polymorphisms (SNPs) are linked to phenotypes associated with diseases and drug responses. Many techniques are now available to identify and quantify such SNPs in DNA or RNA pools, although the information on the latter is limited. The majority of these methodologies require prior knowledge of target sequences, normally obtained through DNA sequencing. Direct quantitation of SNPs from DNA sequencing raw data will save time and money for large amount sample analysis. A high throughput DNA sequencing assay, in combination with a SNP quantitative algorithm, was developed for the quantitation of a SNP present in HCV RNA sequences. For a side-by-side comparison, a Pyrosequencing assay was also developed. Quantitation performance was evaluated for both methods. The direct DNA sequencing quantitation method was shown to be more linear, accurate, sensitive, and reproducible than the Pyrosequencing method for the quantitation of the SNP present in HCV RNA molecules.

Global transcriptional program of p53 target genes during the process of apoptosis and cell cycle progression

The temporal gene expression profile during the entire process of apoptosis and cell cycle progression in response to p53 in human ovarian cancer cells was explored with cDNA microarrays representing 33 615 individual human genes. A total of 1501 genes (4.4%) were found to respond to p53 (approximately 80% of these were repressed by p53) using 2.5-fold change as a cutoff. It was anticipated that most of p53 responsive genes resulted from the secondary effect of p53 expression at late stage of apoptosis. To delineate potential p53 direct and indirect target genes during the process of apoptosis and cell cycle progression, microarray data were combined with global p53 DNA-binding site analysis. Here we showed that 361 out of 1501 p53 responsive genes contained p53 consensus DNA-binding sequence(s) in their regulatory region, approximately 80% of which were repressed by p53. This is the first time that a large number of p53-repressed genes have been identified to contain p53 consensus DNA-binding sequence(s) in their regulatory region. Hierarchical cluster analysis of these genes revealed distinct temporal expression patterns of transcriptional activation and repression by p53. More genes were activated at early time points, while more repressed genes were found after the onset of apoptosis. A small-scale quantitative chromatin immunoprecipitation analysis indicated that in vivo p53-DNA interaction was detected in eight out of 10 genes, most of which were repressed by p53 at the early onset of apoptosis, suggesting that a portion of p53 target genes in the human genome could be negatively regulated by p53 via sequence-specific DNA binding. The approaches and genes described here should aid the understanding of global gene regulatory network of p53.

PepPat, a pattern-based oligopeptide homology search method and the identification of a novel tachykinin-like peptide

PepPat, a hybrid method that combines pattern matching with similarity scoring, is described. We also report PepPat's application in the identification of a novel tachykinin-like peptide. PepPat takes as input a query peptide and a user-specified regular expression pattern within the peptide. It first performs a database pattern match and then ranks candidates on the basis of their similarity to the query peptide. PepPat calculates similarity over the pattern spanning region, enhancing PepPat's sensitivity for short query peptides. PepPat can also search for a user-specified number of occurrences of a repeated pattern within the target sequence. We illustrate PepPat's application in short peptide ligand mining. As a validation example, we report the identification of a novel tachykinin-like peptide, C14TKL-1, and show it is an NK1 (neuokinin receptor 1) agonist whose message is widely expressed in human periphery.

AVAILABILITY

PepPat is offered online at: http://peppat.cbi.pku.edu.cn.

Comparative promoter analysis and its application in analysis of PTH-regulated gene expression

Taking advantage of the "working draft" of the human genome and the MIT shotgun assembly of the mouse genome, we performed a comparative promoter analysis of human RefSeq mRNA (sequences from GenBank's RefSeq database). By combining this analysis with a transcription factor (TF) binding site analysis using a TRANSFAC position weight matrix (PWM) search, 86% of non-specific TF sites were removed. Using a set of genes that are regulated by parathyroid hormone (PTH), a statistical analysis was performed on the conserved TF binding sites among a set of eight human and mouse genes. From among the eight genes tested, we obtained a set of 31 TFs, suggesting possible roles for associated genes in PTH-mediated pathways. All three known PTH-responsive TFs (AP1, RUNX2, CREB) were correctly predicted by this analysis as well as two other potential TFs (VDR and CEBP Delta). Additionally, a model was made to describe the TF site characteristic module of PTH-regulated genes. This model was then used to search all human RefSeq gene promoters with established human-mouse ortholog relationships to identify other PTH-regulated genes. This comparative approach combined with statistical analysis proved to be sufficiently specific to decipher critical TFs involved in PTH-regulated pathways.

[IRE_FINDER-computational search of iron response element in human and mouse UTRs]

The iron response element (IRE) is a highly conserved RNA stem loop structure. It is the binding site of iron regulatory protein (IRP). IRP binding to IRE is regulated by cellular iron. When cells are derived of iron, IRP binds IRE. If IRE is located at 5'UTR, IRP binding will inhibit translation initiation, else if IRE is at 3'UTR, IRP binding will stabilize mRNA and prevent it from degradation. So far all known IREs have C at the 1 position and G at the 5 position of the loop (C1G5 type). In vitro studies suggest that the U1A5 type IRE, which has U and A at the 1 and 5 loop position respectively, binds well to IRP. However, U1A5 type's in vivo existence is still elusive. IRE-IRP binding is involved in the regulation of iron metabolism, oxidative stress and possibly aging. Here we use an improved computation method performing a comprehensive search of IRE in human and mouse genes. We try to catalog potential human and mouse IRE containing genes, at the same time identify potential U1A5 IREs.

A reference database for tumor-related genes co-expressed with interleukin-8 using genome-scale in silico analysis

BACKGROUND

The EST database provides a rich resource for gene discovery and in silico expression analysis. We report a novel computational approach to identify co-expressed genes using EST database, and its application to IL-8.

RESULTS

IL-8 is represented in 53 dbEST cDNA libraries. We calculated the frequency of occurrence of all the genes represented in these cDNA libraries, and ranked the candidates based on a Z-score. Additional analysis suggests that most IL-8 related genes are differentially expressed between non-tumor and tumor tissues. To focus on IL-8's function in tumor tissues, we further analyzed and ranked the genes in 16 IL-8 related tumor libraries.

CONCLUSIONS

This method generated a reference database for genes co-expressed with IL-8 and could facilitate further characterization of functional association among genes.

Hepatitis C virus whole genome position weight matrix and robust primer design

BACKGROUND

The high degree of sequence heterogeneity found in Hepatitis C virus (HCV) isolates, makes robust nucleic acid-based assays difficult to generate. Polymerase chain reaction based techniques, require efficient and specific sequence recognition. Generation of robust primers capable of recognizing a wide range of isolates is a difficult task.

RESULTS

A position weight matrix (PWM) and a consensus sequence were built for each region of HCV and subsequently assembled into a whole genome consensus sequence and PWM. For each of the 10 regions, the number of occurrences of each base at a given position was compiled. These counts were converted to frequencies that were used to calculate log odds scores. Using over 100 complete and 14,000 partial HCV genomes from GenBank, a consensus HCV genome sequence was generated along with a PWM reflecting heterogeneity at each position. The PWM was used to identify the most conserved regions for primer design.

CONCLUSIONS

This approach allows rapid identification of conserved regions for robust primer design and is broadly applicable to sets of genomes with all levels of genetic heterogeneity.

Computational analysis of composite regulatory elements

Combinatorial regulation is a powerful mechanism for generating specificity in gene expression, and it is thought to play a pivotal role in the formation of the complex gene regulatory networks found in higher eukaryotes. The term "Composite Element" (CE) refers to a minimal functional unit where protein-DNA and protein-protein interactions contribute to a highly specific pattern of gene transcriptional regulation. Identification of composite elements will help to better understand gene regulation networks. Experimentally identified CEs are limited in number, and the currently available CE database COMPEL is based on such published information. Here, based on the statistical analysis of over-represented adjacent transcription factor binding sites, we describe a computational method to predict composite regulatory elements in genomic sequences. The algorithm proved to be efficient for extracting composite elements that had been experimentally confirmed and documented in the COMPEL database. Furthermore, putative new composite elements are predicted based on this method, and we have been able to confirm some of our predictions which are not included in the COMPEL database by searching published information.

Analysis of a human brain transcriptome map

BACKGROUND

Genome wide transcriptome maps can provide tools to identify candidate genes that are over-expressed or silenced in certain disease tissue and increase our understanding of the structure and organization of the genome. Expressed Sequence Tags (ESTs) from the public dbEST and proprietary Incyte LifeSeq databases were used to derive a transcript map in conjunction with the working draft assembly of the human genome sequence.

RESULTS

Examination of ESTs derived from brain tissues (excluding brain tumor tissues) suggests that these genes are distributed on chromosomes in a non-random fashion. Some regions on the genome are dense with brain-enriched genes while some regions lack brain-enriched genes, suggesting a significant correlation between distribution of genes along the chromosome and tissue type. ESTs from brain tumor tissues have also been mapped to the human genome working draft. We reveal that some regions enriched in brain genes show a significant decrease in gene expression in brain tumors, and, conversely that some regions lacking in brain genes show an increased level of gene expression in brain tumors.

CONCLUSIONS

This report demonstrates a novel approach for tissue specific transcriptome mapping using EST-based quantitative assessment.

Analyses of p53 target genes in the human genome by bioinformatic and microarray approaches

The completion of the human genome sequence (International Human Genome Sequence Consortium (2001) Nature 409, 860-921; Venter, J. C., et al. (2001) Science 291, 1304-1351) allows for new ways to analyze global cellular regulatory mechanisms. Here we present a strategy to identify genes regulated by specific transcription factors in the human genome, and apply it to p53. We first collected promoters or introns of all genes available using two methods: GenBank(TM) annotation and a computationally derived transcript map. 4,852 genes analyzed in this way contained at least one p53 consensus binding sequence. Of 13 genes randomly selected for mRNA analysis, 11 were shown to respond to p53 expression. Five promoters were analyzed by chromatin immunoprecipitation, which revealed that all were bound by p53 in vivo. We then analyzed 33,615 unique human genes on cDNA microarrays, identifying 1,501 genes that respond to p53 expression. A parameter was derived that demonstrates that in silico prediction greatly enriches for genes that are activated and repressed by p53 and assists us to suggest other signaling pathways that may be connected to p53. The methods shown here illustrate a novel approach to analysis of global gene regulatory network through the integration of human genomic sequence information and genome-wide gene expression analysis.

Accumulation of calbindin in cortical pyramidal cells with ageing; a putative protective mechanism which fails in Alzheimer's disease

There is considerable interest in the status of calbindin immunoreactive neurones in Alzheimer's disease (AD) but previous studies have produced widely differing results. Here we describe calbindin neurones in temporal neocortex from 18 severely demented patients with neuropathologically confirmed AD and 13 age and post-mortem delay matched, neurologically normal controls. Calbindin immunoreactive neurones were small and round in layers II-IV, and pyramidal in layers IIIc and V. There were significantly more calbindin positive neurones in controls than in AD (mean+/- SD, for each comparison P < 0.01): superior temporal lobe, AD = 3.32 +/- 2.24, Control (C) = 24.83 +/- 10.8; middle temporal lobe, AD = 3.6 +/- 4.94, C = 26.09 +/- 15.7; inferior temporal lobe, AD 3.69 +/- 3.6, C = 25.25 +/- 16.9. Furthermore, there was an age-related increase in immunopositive neurones in the superior (r2 = 0.37, P = 0.046) and inferior (r2 = 0.75, P = 0.01) temporal gyri in controls. In AD the number of calbindin positive neurones did not change with age. This is the first report of such an age-related increase in controls, and it suggests that this, rather than a decrease in AD, accounts for the overall difference between AD and controls. It is possible that an increase in intraneuronal calbindin protects these cells from degeneration and that failure of such a neuroprotective mechanism is a significant contributory factor in the pathogenesis of AD.

Structural and functional abnormalities of the hippocampal formation in rats with environmentally induced reductions in prepulse inhibition of acoustic startle

The effects of social isolation on prepulse inhibition of acoustic startle (PPI), electrophysiology and morphology of subicular pyramidal neurons and the densities of interneuronal sub-types in the hippocampal formation were examined. Wistar rats (male weanlings) were housed socially (socials, n=8) or individually (isolates, n=7). When tested eight weeks later, PPI was lower in isolates. Rats then received terminal anaesthesia before slices of hippocampal formation were made in which the electrophysiological properties of a total of 108 subicular neurons were characterized. There were no differences in neuronal sub-types recorded in socials compared with isolates. Intrinsically burst-firing and regular spiking pyramidal neurons were examined in detail. There were no differences in resting membrane potential or input resistance in isolates compared with socials but action potential height was reduced and action potential threshold raised in isolates. A limited morphological examination of Neurobiotin-filled intrinsically burst-firing neurons did not reveal differences in cell-body area or in number of primary dendrites. Sections from the contralateral hemispheres of the same rats were stained with antibodies to calretinin, parvalbumin and the neuronal isoform of nitric oxide synthase (nNOS). In isolates, the density of calretinin positive neurons was increased in the dentate gyrus but unchanged in areas CA3, CA1 and subiculum. Parvalbumin and nNOS positive neuronal densities were unchanged. Hence in rats with environmentally induced reductions in PPI there are structural and functional abnormalities in the hippocampal formation. If the reduction in PPI stems from these abnormalities, and reduced PPI in rats is relevant to schizophrenia, then drugs that correct the reported electrophysiological changes might have antipsychotic effects.

Post-weaning social isolation of rats leads to an abnormal gait

The purpose of this study was to further characterize the phenotype of rats that have experienced prolonged postweaning social isolation, a paradigm that produces changes relevant to neuropsychiatric disorders. At weaning, male Wistar rats from three litters were housed socially (n = 12) or in isolation (n = 13) for 10 weeks. Isolated rats could see, hear and smell other rats. A sophisticated analysis of gait revealed that the stride duration was increased in isolates (12%, P = 0.0024) as a result of increased stance duration (18%, P = 0.0005), but there was no difference in vertical reaction force or velocity. Adrenal glands were heavier in isolates (19%, P = 0.0047). There was no difference in cross-sectional area of the brain or lateral ventricles anywhere along the anterior-posterior axis. All experiments and analysis were performed blind to housing condition. This is the first study to demonstrate that socially isolated rats have an abnormal gait. Further analysis, including pharmacological manipulation, is needed in order to understand the nature of the abnormality.

Genetic footprinting in bacteria

In vivo genetic footprinting was developed in the yeast Saccharomyces cerevisiae to simultaneously assess the importance of thousands of genes for the fitness of the cell under any growth condition. We have developed in vivo genetic footprinting for Escherichia coli, a model bacterium and pathogen. We further demonstrate the utility of this technology for rapidly discovering genes that affect the fitness of E. coli under a variety of growth conditions. The definitive features of this system include a conditionally regulated Tn10 transposase with relaxed sequence specificity and a conditionally regulated replicon for the vector containing the transposase and mini-Tn10 transposon with an outwardly oriented promoter. This system results in a high frequency of randomly distributed transposon insertions, eliminating the need for the selection of a population containing transposon insertions, stringent suppression of transposon mutagenesis, and few polar effects. Successful footprints have been achieved for most genes longer than 400 bp, including genes located in operons. In addition, the ability of recombinant proteins to complement mutagenized hosts has been evaluated by genetic footprinting using a bacteriophage lambda transposon delivery system.

Induction of hyperphosphorylated tau in living slices of rat hippocampal formation and subsequent detection using an ELISA

Although hyperphosphorylated tau is an established feature of Alzheimer's Disease, its role in the disease process is poorly understood, partly because of lack of suitable animal models. We describe the use of living slices of rat hippocampal formation to study tau phosphorylation. Using the AT8 antibody in an ELISA, phosphorylated tau was detected in freshly frozen slices and it increased significantly in slices that were incubated in an electrophysiological recording chamber; the amount detected was greatest when the homogenisation buffer contained phosphatase and kinase inhibitors. The phosphorylated tau content of the slices increased significantly after exposure to the phosphatase 1 and 2A inhibitor okadaic acid (OA) - 1.5 microM. Electrophysiological recordings confirmed that slices were alive and that OA had no acute toxic effect. In control slices phosphorylated tau, detected immunohistochemically, was mainly in the somatodendritic compartment of neurones; in OA treated slices, there was an apparent decrease in somatodendritic AT8 staining and an increase in neuropil staining. Our system enables the induction of hyperphosphorylated tau within living slices, in an experimental environment that can be used to study the biological consequences of such a change, and may therefore help further our understanding of the significance of hyperphosphorylated tau in Alzheimer's Disease.

Heparin injection into the adult rat hippocampus induces seizures in the absence of macroscopic abnormalities

The pathological hallmarks of Alzheimer's disease include neurofibrillary tangles, neuropil threads and neuritic plaques. Neurofibrillary tangles and neuropil threads are comprised of paired helical filaments which are themselves composed of a hyperphosphorylated form of the microtubule-associated protein tau. Neuritic plaques are extracellular deposits of aggregated beta amyloid associated with neurites containing hyperphosphorylated tau. The mechanisms by which the neurofibrillary tangles and neuritic plaques develop in Alzhemier's disease are not clear but it is hypothesized that sulphated glycosaminoglycans are important in their formation. This impression is based on the finding that the glycosaminoglycan, heparan sulphate, is found associated with neurofibrillary tangles, neuritic plaques and neuropil threads while dermatan sulphate, chondroitin sulphate and keratan sulphate immunoreactivity is found around neuritic plaques in brains of Alzheimer's disease patients. Furthermore, in vitro studies demonstrate that sulphated glycosaminoglycans such as heparan sulphate and the closely related molecule heparin interact with tau and potentiate its phosphorylation by a number of serine/threonine kinases, reduce its ability to bind to microtubules and induce paired helical filament formation, all properties associated with tau isolated from Alzheimer's disease brain. Thus, we were interested to learn whether intracerebral injection of the sulphated glycosaminoglycan heparin would give rise to alterations in the cytoskeletal protein tau in the rat brain. Although no cytoskeletal changes were observed, to our considerable surprise we found that the intrahippocampal injection of heparin gave rise to seizures. We have investigated this unexpected effect further in vivo and by using in vitro electrophysiological techniques.

Differential expression of NADPH-diaphorase between electrophysiologically-defined classes of pyramidal neurons in rat ventral subiculum, in vitro

The subiculum is the major output region of the hippocampal formation. We have studied pyramidal neurons in slices of rat ventral subiculum to determine if there is a correlation between nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-d) activity and electrophysiological phenotype. The majority of NADPH-d-positive pyramidal neurons were found in the superficial cell layer (i.e. nearest to the hippocampal fissure) of the subiculum and appreciable NADPH-d activity was absent from pyramidal neurons in area CA1. This distribution of NADPH-d activity was mimicked by that of immunoreactivity for the neuronal isoform of nitric oxide synthase. Subicular pyramidal neurons were classified, electrophysiologically, as intrinsically burst-firing or regular spiking. After electrophysiological characterization, neurons were filled with Neurobiotin and revealed using fluorescence immunocytochemistry. The slices containing these neurons were also processed for NADPH-d. NADPH-d activity was found in six out of eight regular spiking neurons but was not found in any of 13 intrinsically burst-firing neurons (P=0.0008, Fisher's Exact Test). We conclude that in rat ventral subiculum, NADPH-d activity is present in a proportion of pyramidal neurons and indicates the presence of the neuronal isoform of nitric oxide synthase. Furthermore, amongst pyramidal neurons, NADPH-d activity is distributed preferentially to those with the regular spiking phenotype. The distribution of regular spiking neurons suggests that they may not be present to the same extent in all subicular output pathways. Thus, the actions of nitric oxide may be relatively specific to particular hippocampal connections.

Morphology and distribution of electrophysiologically defined classes of pyramidal and nonpyramidal neurons in rat ventral subiculum in vitro

Intracellular electrophysiological recordings were made from 210 ventral subicular neurons in rat brain slices. Recordings were classified as burst-firing or nonburst-firing. Eighteen burst-firing neurons were filled with Neurobiotin, and all had pyramidal morphology. Nine of these recordings were made from intrinsically burst-firing (IB) cell bodies, and nine were made from burst-firing dendrites (BD). Twelve nonburst-firing neurons were also filled with Neurobiotin. Eight were regular spiking (RS) and had pyramidal morphology, four were fast spiking (FS) and nonpyramidal. Additional electrophysiological parameters distinguished IB from BD, RS from FS, and RS from IB recordings. The distribution of IB and RS neurons was examined by using 180 recordings. Information from the first series of experiments was used to distinguish between somatic and dendritic recordings. The deep-superficial axis (alveus-hippocampal fissure) was divided into four equal rows. RS neurons accounted for 12%, 28%, 58%, and 50% of presumed somatic recordings in successively more superficial rows. The proximal-distal (CA1-perforant path) axis was divided into five equal columns. RS cells accounted for 52% of presumed somatic impalements in the central column compared with 16% in the most proximal and 10% in the most distal columns. Thus, two electrophysiological classes of pyramidal neuron were localized to particular regions of the ventral subiculum. In the light of existing knowledge of the topography of subicular inputs and outputs, our results are consistent with the hypothesis that the ratio of RS to IB pyramidal neurons will be different in different transhippocampal circuits.

The subiculum: a potential site of action for novel antipsychotic drugs?

In this article, recent information from neuropathological studies in humans and behavioural and electrophysiological/pharmacological studies in rats is used to examine the hypothesis that the subiculum, the major output region of the hippocampal formation, may contain sites of action for novel antipsychotic drugs. In the first section, the possible interactions between subicular neurons and the sites at which existing antipsychotic drugs act are discussed. These include the interaction implied by convergence of subicular and dopaminergic inputs to the nucleus accumbens. The second section concentrates upon subicular involvement in animal behaviours that are thought to be relevant to schizophrenia, which, in rats, include Latent inhibition and Pre-pulse inhibition of Acoustic Startle. Involvement of the subiculum in the neuropathology of schizophrenia is discussed in the third section. However, few neuropathological studies comment specifically on the subiculum. Those which suggest involvement tend to be recent and, as yet, have not all been replicated. Finally, there is discussion of the possibility that the subiculum contains chemical sites at which drugs could act specifically to produce appropriate physiological effects. Potential sites include, but are not necessarily restricted to, particular ion channels in electrophysiologically defined subclasses of subicular pyramidal neurons, the receptors for neuromodulatory peptides such as somatostatin and cholecystokinin, and the enzyme nitric oxide synthase. It is concluded that a wide range of clinical and basic neuroscience disciplines has provided evidence which, especially when viewed as a whole, is consistent with the hypothesis that the subiculum is a potential site of action for novel antipsychotic drugs.

Neuronal diversity in the subiculum: correlations with the effects of somatostatin on intrinsic properties and on GABA-mediated IPSPs in vitro

1. We used intracellular current-clamp techniques to record from 33 ventral subicular neurons in slices or rat hippocampal formation. Presumed pyramidal neurons were characterized by their responses to depolarizing current pulses as either intrinsically burst firing (IB) or regular spiking (RS). Within the subiculum, IB cells were encountered most frequently in the deep cell layer, whereas RS cells were encountered most frequently in the superficial cell layer. IB cells had more depolarized resting potentials, lower input resistances, and more sag in their voltage responses to hyperpolarizing current pulses. 2. Somatostatin (5 microM) applied in the bathing medium caused a hyperpolarization and reduction in input resistance. These effects were of greater magnitude in IB cells. Somatostatin had no effect on sag in either cell type. These effects of somatostatin were unchanged in the presence of gamma-aminobutyric acid (GABA) receptor antagonists. 3. In a series of experiments conducted in RS cells only, somatostatin reduced the amplitude of the late but not the early component of evoked biphasic inhibitory postsynaptic potentials (IPSPs). 4. A second series of experiments was conducted in RS and IB cells. Somatostatin reduced the amplitude of pharmacologically isolated GABAA IPSPS in both cell types. In IB cells but not RS cells there was a correlation between this effect and the somatostatin-induced hyperpolarization. Somatostatin also reduced the amplitude of isolated GABAB IPSPS in both cell types, but more so in IB cells. 5. Somatostatin had no effect on the reversal potential of either IPSP in either cell type and no effect on the GABAA-mediated conductance in either cell type. In contrast, the GABAB-mediated conductance was reduced, especially in IB cells. 6. The effects of somatostatin on GABAA IPSPS are principally a result of membrane shunting and reductions in ionic driving force, but these mechanisms do not account for the reduction in GABAB IPSPS. 7. We suggest that the combined effects of somatostatin are likely to alter the balance between fast and slow inhibition and to do so more in IB cells than in RS cells.

Effects of somatostatin and related peptides on the membrane potential and input resistance of rat ventral subicular neurons, in vitro

The effects of bath-applied somatostatin (SS), and related peptides on the membrane potential and input resistance of 117 ventral subicular neurons were investigated by intracellular recording in rat brain slices. Electrophysiological properties, which included burst-firing in response to depolarizing current pulses, indicated that the neurons studied were of the pyramidal type. For the 89 cells analyzed quantitatively, membrane potential was -69.1 +/- 0.3 mV (mean +/- S.E.) and input resistance was 23.9 +/- 0.5 megohms. SS (5 microM) caused a hyperpolarization of 3.4 +/- 0.3 mV (n = 9) and reduced input resistance by 16 +/- 3.1% (n = 6). SS D-Trp8, somatostatin, octreotide, CGP 23996 and MK 678 shared these effects, but somatostatin was inactive. SS effects persisted when bathing solutions contained tetrodotoxin, reduced calcium and elevated magnesium concentrations and when both of these treatments were combined. They were unaltered by antagonists at gamma-aminobutyric acid receptors or at ionotropic glutamate receptors. The effects of MK 678, SS, SS D-Trp8 and somatostatin were concentration-dependent, and these peptides were equipotent at 500 nM and at 5 microM. For MK 678, the EC50 was 316 nM for the hyperpolarization and 90 nM for the reduction in input resistance. We conclude that SS acts directly on pyramidal neurons of the rat subiculum to cause a hyperpolarization and a decrease in input resistance. We suggest that these effects are mediated by the SSTR2 receptor subtype.

Homologs of the yeast neck filament associated genes: isolation and sequence analysis of Candida albicans CDC3 and CDC10

Morphogenesis in the yeast Saccharomyces cerevisiae consists primarily of bud formation. Certain cell division cycle (CDC) genes, CDC3, CDC10, CDC11, CDC12, are known to be involved in events critical to the pattern of bud growth and the completion of cytokinesis. Their products are associated with the formation of a ring of neck filaments that forms at the region of the mother cell-bud junction during mitosis. Morphogenesis in Candida albicans, a major fungal pathogen of humans, consists of both budding and the formation of hyphae. The latter is thought to be related to the pathogenesis and invasiveness of C. albicans. We have isolated and characterized C. albicans homologs of the S. cerevisiae CDC3 and CDC10 genes. Both C. albicans genes are capable of complementing defects in the respective S. cerevisiae genes. RNA analysis of one of the genes suggests that it is a regulated gene, with higher overall expression levels during the hyphal phase than in the yeast phase. Not surprisingly, DNA sequence analysis reveals that the proteins share extensive homology at the amino acid level with their respective S. cerevisiae counterparts. Related genes are also found in other species of Candida and, more importantly, in filamentous fungi such as Aspergillus nidulans and Neurospora crassa. A database search revealed significant sequence similarity with two peptides, one from Drosophila and one from mouse, suggesting strong evolutionary conservation of function.

The GEF1 gene of Saccharomyces cerevisiae encodes an integral membrane protein; mutations in which have effects on respiration and iron-limited growth

We have isolated a new class of respiration-defective, i.e petite, mutants of the yeast Saccharomyces cerevisiae. Mutations in the GEF1 gene cause cells to grow slowly on rich media containing carbon sources utilized by respiration. This phenotype is suppressed by adding high concentrations of iron to the growth medium. Gef1- mutants also fail to grow on a fermentable carbon source, glucose, when iron is reduced to low concentrations in the medium, suggesting that the GEF1 gene is required for efficient metabolism of iron during growth on fermentable as well as respired carbon sources. However, activity of the iron uptake system appears to be unaffected in gef1- mutants. Fe(II) transporter activity and regulation is normal in gef1- mutants. Fe(III) reductase induction during iron-limited growth is disrupted, but this appears to be a secondary effect of growth rate alterations. The wild-type GEF1 gene was cloned and sequenced; it encodes a protein of 779 amino acids, 13 possible transmembrane domains, and significant similarity to chloride channel proteins from fish and mammals, suggesting that GEF1 encodes an integral membrane protein. A gef1- deletion mutation generated in vitro and introduced into wild-type haploid strains by gene transplacement was not lethal. Oxygen consumption by intact gef1- cells and by mitochondrial fractions isolated from gef1- mutants was reduced 25-50% relative to wild type, indicating that mitochondrial function is defective in these mutants. We suggest that GEF1 encodes a transport protein that is involved in intracellular iron metabolism.

Parallel pathways of gene regulation: homologous regulators SWI5 and ACE2 differentially control transcription of HO and chitinase

Two independent pathways of transcriptional regulation that show functional homology have been identified in yeast. It has been demonstrated previously that SWI5 encodes a zinc finger DNA-binding protein whose transcription and cellular localization both are cell cycle regulated. We show that ACE2, whose zinc finger region is nearly identical to that of SWI5, shows patterns of cell cycle-regulated transcription and nuclear localization similar to those seen previously for SWI5. Despite their similarities, SWI5 and ACE2 function in separate pathways of transcriptional regulation. SWI5 is a transcriptional activator of the HO endonuclease gene, whereas ACE2 is not. In contrast, ACE2 is a transcriptional activator of the CTS1 gene (which encodes chitinase), whereas SWI5 is not. An additional parallel between the SWI5/HO pathway and the ACE2/CTS1 pathway is that HO and CTS1 both are cell cycle regulated in the same way, and HO and CTS1 both require the SWI4 and SWI6 transcriptional activators. Overproduction of either SWI5 or ACE2 permits transcriptional activation of the target gene from the other pathway, suggesting that the DNA-binding proteins are capable of binding in vivo to promoters that they do not usually activate. Chimeric SWI5/ACE2 protein fusion experiments suggest that promoter specificity resides in a domain distinct from the zinc finger domain.

Bacteriophage SPO1 middle transcripts

Phage SPO1 middle transcripts are known to fall into two classes, m and m1l. Class m1l transcripts continue to be made late in the viral infection, while the synthesis of class m transcripts ceases soon after the onset of replication and late transcription. The experiments that are reported here deal with the regulatory nature of this diversity. The accumulation of transcripts associated with eight middle promoters was analyzed by S1 nuclease mapping. DNA sequence surrounding these middle promoters was determined or redetermined, and the stability of RNA associated with most of these promoters was also analyzed. Class m1l transcription was shown to be associated with SPO1 middle promoters that remain active at late stages of viral development, when middle promoters of class m are repressed. The consensus sequences of class m and m1l middle promoters were found to be indistinguishable and the search for sequences consensual with late promoters yielded only divergent candidates. No other consensus sequence that is specific and exclusive to either class of middle promoters was detected within a hundred base pairs upstream or downstream of these promoters. Considerable variations in the stabilities of SPO1 middle transcripts were found. Two promoters that are only 71 base pairs apart yielded transcripts that had substantially different stabilities. The 5'-flanking segment of the transcript associated with the upstream promoter apparently conferred a high degree of stability on this RNA.

DNA binding by the bacteriophage SPO1-encoded type II DNA-binding protein, transcription factor 1. Formation of nested complexes at a selective binding site

The interactions of the phage SPO1-encoded Type II DNA-binding protein, transcription factor 1 (TF1), with one of its preferred binding sites in SPO1 DNA have been analyzed in detail. The results suggest that TF1 recognizes a high-affinity "core" binding site and that additional protein moieties can accrue to either side of the occupied core site to form higher order complexes. Close contacts between TF1 and the core binding site as well as some of the steric requirements for recognition of the core site were determined. Comparison of the nucleotide sequences of several preferred binding sites for TF1 reveals a striking lack of precise homology but does show common features.

DNA binding by the bacteriophage SPO1-encoded type II DNA-binding protein, transcription factor 1. Site-specific binding requires 5-hydroxymethyluracil-containing DNA

The bacteriophage SPO1-encoded Type II DNA-binding protein, transcription factor 1 (TF1), forms complexes with specific sites in SPO1 DNA. We have investigated the binding of TF1 to one of its preferred sites in which the normal 5-hydroxymethyluracil (hmUra) of SPO1 DNA has been replaced by thymine and have also investigated the binding of a bacterial Type II DNA-binding protein (from Bacillus stearothermophilus) to the hmUra- and thymine-containing forms of the same DNA segment. Our results show that TF1 binds selectively to this high affinity binding site only in hmUra-containing DNA and that the bacterial Type II DNA-binding protein interacts nonspecifically with both forms of DNA.

Synergistic toxicity of endotoxin and hemoglobin

Stroma-free hemoglobin (SFH) is advocated as an oxygen-transporting resuscitation solution. Hemoglobin has been shown to enhance endotoxin lethality when given intraperitoneally. It is possible that SFH could interact with endotoxin when used as an oxygen-transporting resuscitation system for trauma victims with contaminating wounds. To assess the effects of these two agents when given intravascularly, rabbits were infused with SFH (1.75 gm/kg) or albumin (controls; 1.75 gm/kg) with and without endotoxin. Two doses of endotoxin were used. At 14.5 ng/kg of Salmonella enteritidis endotoxin, no effect was seen in the albumin group. However, 50% of the hemoglobin group died. At 14.5 micrograms/kg, the albumin group showed hematologic alterations, but all animals lived. All SFH-treated animals died at the higher endotoxin dose. SFH alone caused cardiac abnormalities (bradycardia in 100%, sinus arrhythmias in 30%, and ventricular arrhythmias in 20%), liver abnormalities (necrosis in 40% and 240% increase in alanine aminotransferase activity by 6 hours), and intravascular thrombi (30%). The only hemoglobin-induced abnormality that was more frequent in the presence of endotoxin was ventricular arrhythmias (up to 75% of animals). Thrombin times were approximately 20% larger in all SFH groups compared with the albumin groups. By 6 hours after infusion, endotoxin prolonged the thrombin time even further, despite the lack of fibrinogen consumption. This study shows that endotoxin and SFH exert synergistic toxicity when SFH is given in a clinically relevant dose for an oxygen-transporting resuscitation system. Only minute quantities of endotoxin are needed to produce this phenomenon. We hypothesize that this synergism is endotoxin enhancement of hemoglobin toxicity.

Toxicity of human hemoglobin solution infused into rabbits

Aqueous human hemoglobin is being considered as an oxygen-transporting resuscitation solution. However, there are some indications that such solutions may be toxic. We report here that at a clinically significant dose (20% to 30% of the estimated blood volume) of hemoglobin solution (10 gm/dl), toxicity similar to that reported in humans is produced in rabbits. This toxicity is characterized by cardiac rhythm disturbances (25%) and coagulation abnormalities, manifested as intravascular thrombi. Hypoxia, tissue necrosis (brain 29%, liver 25%), and death (22%) were seen. Hypoxia appears to be the cause of death, because the arterial PO2 of rabbits that died was significantly lower (43 +/- 8 mm Hg) than that of rabbits that lived (80 +/- 12 mm Hg). We also report the effect of the method of preparation or degree of purity on the observed toxicity. Stroma-free red blood cell hemolysate, stroma-free hemoglobin, zinc-precipitated hemoglobin, and chromatographically purified hemoglobin were infused. Purification of these solutions ranged from removal of stroma to removal of all detectable nonhemoglobin protein components. In addition, the nonhemoglobin proteins were concentrated and infused into rabbits to compare their effects. All of these hemoglobin solutions induced similar toxicity for rabbits infused with stroma-free hemoglobin. The nonhemoglobin proteins, as well as human serum albumin, exhibited no toxicity. These results demonstrate that, at the clinically relevant dose, hemoglobin or something inherently associated with hemoglobin is toxic.

The phage SPO1-specific RNA polymerase, E.gp28, recognizes its cognate promoters in thymine-containing DNA

The bacteriophage SPO1 gene 28-encoded protein, gp28, directs specific recognition of viral middle promoters in hydroxymethyluracil-containing DNA by the Bacillus subtilis host's RNA polymerase core. Using appropriately sensitive methods of detection, we have shown that discrimination against thymine-containing DNA is not absolute and that the gp28-containing RNA polymerase precisely initiates transcription at two thymine-containing SPO1 middle promoters.

Site-specific DNA binding by the bacteriophage SP01-encoded type II DNA-binding protein

The bacteriophage SP01 genome encodes a virus-specific type II DNA-binding protein, TF1. The bacterial proteins of this ubiquitous and evolutionarily conserved class are thought to bind non-specifically to DNA. In contrast, the experiments described here demonstrate that TF1 binds to specific sites in SP01 DNA. Several of these sites have been characterized by DNase I 'footprinting' and four of them have been shown to overlap strong phage promoters for Bacillus subtilis RNA polymerase holoenzyme. We speculate on the possible structural basis of site-selective DNA binding by a protein of this class.

Sequence of the bacteriophage SP01 gene coding for transcription factor 1, a viral homologue of the bacterial type II DNA-binding proteins

The Bacillus subtilis phage SP01, whose DNA contains 5-hydroxymethyluracil (hmUra) in place of thymine, codes for an abundant, small, basic protein called TF1. TF1 binds preferentially to hydroxymethyluracil-containing DNA and thereby selectively inhibits transcription of such DNA in vitro. The gene for TF1 has been sequenced. We find that this viral protein is a homologue of the ubiquitous bacterial type II DNA-binding proteins. The three-dimensional structure of one of these bacterial proteins has recently been determined. We are able to discern common as well as distinctive features in the amino acid sequence and the three-dimensional structure of the homologous viral protein.

SP01 gene 27 is required for viral late transcription

The SP01 mutant sus HA20 (gene 27) was found to be defective for synthesis of viral late RNA. It is known that gene 27 is also required for viral DNA replication. The SP01 gene 27 product resembles the T4 gene 45 product, which also has a dual role in viral DNA replication and late transcription.

Nucleotide sequences of the mRNA's encoding the vesicular stomatitis virus N and NS proteins

The complete nucleotide sequences of the vesicular stomatitis virus (VSV) mRNA's encoding the N and NS proteins have been determined from the sequences of cDNA clones. The mRNA encoding the N protein is 1,326 nucleotides long, excluding polyadenylic acid. It contains an open reading frame for translation which extends from the 5'-proximal AUG codon to encode a protein of 422 amino acids. The N and mRNA is known to contain a major ribosome binding site at the 5'-proximal AUG codon and two other minor ribosome binding sites. These secondary sites have been located unambiguously at the second and third AUG codons in the N mRNA sequence. Translational initiation at these sites, if it in fact occurs, would result in synthesis of two small proteins in a second reading frame. The VSV and mrna encoding the NS protein is 815 nucleotides long, excluding polyadenylic acid, and encodes a protein of 222 amino acids. The predicted molecular weight of the NS protein (25,110) is approximately one-half of that predicted from the mobility of NS protein on sodium dodecyl sulfate-polyacrylamide gels. Deficiency of sodium dodecyl sulfate binding to a large negatively charged domain in the NS protein could explain this anomalous electrophoretic mobility.