UltraPrep is a scalable, cost-effective, bead-based method for purifying cell-free DNA

UltraPrep is an open-source, two-step method for purification of cell-free DNA that entails extraction of total DNA followed by size-selective enrichment of the smaller fragments that are characteristic of DNA originating from fragmentation between nucleosome. The advantages of the two related protocols that are described are that they can easily accommodate a wide range of sample input volumes, they rely on simple, magnetic bead-based technology, the yields of cfDNA are directly comparable to the most popular methods for cfDNA purification, and they dramatically reduce the cost of cfDNA isolation relative to currently available commercial methods. We provide a framework for physical and molecular quality analysis of purified cfDNA and demonstrate that the cfDNA generated by UltraPrep meets or exceeds the quality metrics of the most commonly used procedure. In addition, our method removes high molecular weight genomic DNA (hmwgDNA) that can interfere with downstream assay results, thereby addressing one of the primary concerns for preanalytical collection of blood samples.

A Prospective Study of Circulating Tumor DNA to Guide Matched Targeted Therapy in Lung Cancers

BACKGROUND

Liquid biopsy for plasma circulating tumor DNA (ctDNA) next-generation sequencing (NGS) is commercially available and increasingly adopted in clinical practice despite a paucity of prospective data to support its use.

METHODS

Patients with advanced lung cancers who had no known oncogenic driver or developed resistance to current targeted therapy (n = 210) underwent plasma NGS, targeting 21 genes. A subset of patients had concurrent tissue NGS testing using a 468-gene panel (n = 106). Oncogenic driver detection, test turnaround time (TAT), concordance, and treatment response guided by plasma NGS were measured. All statistical tests were two-sided.

RESULTS

Somatic mutations were detected in 64.3% (135/210) of patients. ctDNA detection was lower in patients who were on systemic therapy at the time of plasma collection compared with those who were not (30/70, 42.9% vs 105/140, 75.0%; OR = 0.26, 95% CI = 0.1 to 0.5, P < .001). The median TAT of plasma NGS was shorter than tissue NGS (9 vs 20 days; P < .001). Overall concordance, defined as the proportion of patients for whom at least one identical genomic alteration was identified in both tissue and plasma, was 56.6% (60/106, 95% CI = 46.6% to 66.2%). Among patients who tested plasma NGS positive, 89.6% (60/67; 95% CI = 79.7% to 95.7%) were also concordant on tissue NGS and 60.6% (60/99; 95% CI = 50.3% to 70.3%) vice versa. Patients who tested plasma NGS positive for oncogenic drivers had tissue NGS concordance of 96.1% (49/51, 95% CI = 86.5% to 99.5%), and directly led to matched targeted therapy in 21.9% (46/210) with clinical response.

CONCLUSIONS

Plasma ctDNA NGS detected a variety of oncogenic drivers with a shorter TAT compared with tissue NGS and matched patients to targeted therapy with clinical response. Positive findings on plasma NGS were highly concordant with tissue NGS and can guide immediate therapy; however, a negative finding in plasma requires further testing. Our findings support the potential incorporation of plasma NGS into practice guidelines.

Focused size selection of cell-free DNA samples for liquid biopsy applications that rely on next-generation sequencing

Genomic analysis of circulating, cell-free DNA (cfDNA) is being used extensively for molecular diagnostics. Many approaches rely on the construction of cfDNA genomic libraries, targeted retrieval of specific genomic regions and analysis by next-generation DNA sequencing. Several steps during sample preparation require isolation of DNA fragments within a particular size range. In this Benchmark article, two related methods for size-selective DNA fragment enrichment are described.

Longitudinal Cell-Free DNA Analysis in Patients with Small Cell Lung Cancer Reveals Dynamic Insights into Treatment Efficacy and Disease Relapse

INTRODUCTION

Patients with SCLC have a poor prognosis and limited treatment options. Because access to longitudinal tumor samples is very limited in patients with this disease, we chose to focus our studies on the characterization of plasma cell-free DNA (cfDNA) for rapid, noninvasive monitoring of disease burden.

METHODS

We developed a liquid biopsy assay that quantifies somatic variants in cfDNA. The assay detects single nucleotide variants, copy number alterations, and insertions or deletions in 14 genes that are frequently mutated in SCLC, including tumor protein p53 gene (TP53), retinoblastoma 1 gene (RB1), BRAF, KIT proto-oncogene receptor tyrosine kinase gene (KIT), notch 1 gene (NOTCH1), notch 2 gene (NOTCH2), notch 3 gene (NOTCH3), notch 4 gene (NOTCH4), phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha gene (PIK3CA), phosphatase and tensin homolog gene (PTEN), fibroblast growth factor receptor 1 gene (FGFR1), v-myc avian myelocytomatosis viral oncogene homolog gene (MYC), v-myc avian myelocytomatosis viral oncogene lung carcinoma derived homolog gene (MYCL1), and v-myc avian myelocytomatosis viral oncogene neuroblastoma derived homolog gene (MYCN).

RESULTS

Over the course of 26 months of peripheral blood collection, we examined 140 plasma samples from 27 patients. We detected disease-associated mutations in 85% of patient samples with mutant allele frequencies ranging from 0.1% to 87%. In our cohort, 59% of the patients had extensive-stage disease, and the most common mutations occurred in TP53 (70%) and RB1 (52%). In addition to mutations in TP53 and RB1, we detected alterations in 10 additional genes in our patient population (PTEN, NOTCH1, NOTCH2, NOTCH3, NOTCH4, MYC, MYCL1, PIK3CA, KIT, and BRAF). The observed allele frequencies and copy number alterations tracked closely with treatment responses. Notably, in several cases analysis of cfDNA provided evidence of disease relapse before conventional imaging.

CONCLUSIONS

These results suggest that liquid biopsies are readily applicable in patients with SCLC and can potentially provide improved monitoring of disease burden, depth of response to treatment, and timely warning of disease relapse in patients with this disease.

Abstract 4949: Longitudinal monitoring of cell-free DNA in patients with small cell lung cancer reveals dynamic insights into treatment efficacy and disease relapse

Introduction: Small cell lung cancer (SCLC) is a highly lethal neuroendocrine malignancy that accounts for approximately 10-15% of all lung cancers and is responsible for approximately 30,000 deaths annually in the United States and 200,000 deaths worldwide every year1. There is an urgent need to develop novel treatment strategies for patients with this disease. We sought to improve the quality of patient care by establishing a liquid biopsy assay for rapid, noninvasive monitoring of disease burden.

Design: The SCLC assay relies on targeted next-generation DNA sequencing of cell-free DNA (cfDNA) collected from patient plasma. The assay targets a panel of 14 genes that are frequently mutated in SCLC2. We examined a total of 141 plasma samples from a cohort of 27 patients. 11 patients had limited stage SCLC and 16 patients had extensive stage SCLC. The analyzed plasma samples were collected during the course of patient treatment and included time points before and after chemotherapy or immunotherapy.

Results: We detected somatic, disease-associated mutations in the cfDNA of 78% of patient samples (21/27). The allele frequency of cfDNA ranged from ≤0.5% to ≥85%. The most commonly mutated genes were TP53 and RB1, which were found in 17/27 and 10/27 samples, respectively. We also detected single nucleotide variants in PIK3CA (3/27) and PTEN (1/27) as well as copy number variants in MYC and MYCL1 (2/27). The observed mutant allele frequencies in longitudinal samples tracked closely with treatment responses. Strikingly, we found instances where the assay detected the reappearance of tumor-associated markers several weeks before clinical evidence of relapse was detected.

Conclusions: cfDNA sequencing allows for improved monitoring of disease burden, depth of responses to treatment, and timely warning of disease relapse in patients with SCLC.

References:

1. Society, A.C., Cancer Facts and Figures 2015. American Cancer Society, Atlanta, Ga, 2015.

2. George, J. et al., Comprehensive genomic profiles of small cell lung cancer, Nature 524:47

Citation Format: Christine M. Lovly, Karinna Almodovar, Wade T. Iams, Catherine B. Meador, Sally York, Leora Horn, Christopher K. Raymond, Jennifer Hernandez, Lee P. Lim. Longitudinal monitoring of cell-free DNA in patients with small cell lung cancer reveals dynamic insights into treatment efficacy and disease relapse [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4949. doi:10.1158/1538-7445.AM2017-4949

Collection of cell-free DNA for genomic analysis of solid tumors in a clinical laboratory setting

The breadth of diagnostic procedures that utilize cell free DNA (cfDNA) from human plasma has increased dramatically in recent years. Here, we confirm that tumor-derived cfDNA fragments are similar in size distribution to cfDNA derived from normal tissues. Therefore, collection procedures optimized with healthy donor specimens are likely to be applicable to the diagnosis and monitoring of many different cancer types. We verify that the distribution and DNA sequences of fragmentation sites in cfDNA from both normal-germline and tumor-derived cfDNA are non-random. A broad survey of cfDNA from healthy donors suggests that average individuals possess ~6 ng of cfDNA per mL of plasma. Importantly, the cfDNA present in plasma samples that were initially collected as whole blood in K2-EDTA tubes and subsequently processed by centrifugation is stable for several days at ambient temperatures. This observation has the potential to significantly reduce the cost and logistical complexity of shipping clinical samples from the site of collection to centers proficient in diagnostic analysis. Finally, plasma samples collected with high-volume plasma collection devices possess abundant quantities of cfDNA. Since the quantity of analyzed cfDNA is directly proportional to sensitivity of diagnostic assays, this method of plasma collection, where available, could enable highly sensitive post-treatment disease monitoring and early detection of cancer in at-risk individuals.

NSR-seq transcriptional profiling enables identification of a gene signature of Plasmodium falciparum parasites infecting children

Malaria caused by Plasmodium falciparum results in approximately 1 million annual deaths worldwide, with young children and pregnant mothers at highest risk. Disease severity might be related to parasite virulence factors, but expression profiling studies of parasites to test this hypothesis have been hindered by extensive sequence variation in putative virulence genes and a preponderance of host RNA in clinical samples. We report here the application of RNA sequencing to clinical isolates of P. falciparum, using not-so-random (NSR) primers to successfully exclude human ribosomal RNA and globin transcripts and enrich for parasite transcripts. Using NSR-seq, we confirmed earlier microarray studies showing upregulation of a distinct subset of genes in parasites infecting pregnant women, including that encoding the well-established pregnancy malaria vaccine candidate var2csa. We also describe a subset of parasite transcripts that distinguished parasites infecting children from those infecting pregnant women and confirmed this observation using quantitative real-time PCR and mass spectrometry proteomic analyses. Based on their putative functional properties, we propose that these proteins could have a role in childhood malaria pathogenesis. Our study provides proof of principle that NSR-seq represents an approach that can be used to study clinical isolates of parasites causing severe malaria syndromes as well other blood-borne pathogens and blood-related diseases.

Genetic validation of whole-transcriptome sequencing for mapping expression affected by cis-regulatory variation

Background

Identifying associations between genotypes and gene expression levels using microarrays has enabled systematic interrogation of regulatory variation underlying complex phenotypes. This approach has vast potential for functional characterization of disease states, but its prohibitive cost, given hundreds to thousands of individual samples from populations have to be genotyped and expression profiled, has limited its widespread application.

Results

Here we demonstrate that genomic regions with allele-specific expression (ASE) detected by sequencing cDNA are highly enriched for cis-acting expression quantitative trait loci (cis-eQTL) identified by profiling of 500 animals in parallel, with up to 90% agreement on the allele that is preferentially expressed. We also observed widespread noncoding and antisense ASE and identified several allele-specific alternative splicing variants.

Conclusion

Monitoring ASE by sequencing cDNA from as little as one sample is a practical alternative to expression genetics for mapping cis-acting variation that regulates RNA transcription and processing.

Digital genome-wide ncRNA expression, including SnoRNAs, across 11 human tissues using polyA-neutral amplification

Non-coding RNAs (ncRNAs) are an essential class of molecular species that have been difficult to monitor on high throughput platforms due to frequent lack of polyadenylation. Using a polyadenylation-neutral amplification protocol and next-generation sequencing, we explore ncRNA expression in eleven human tissues. ncRNAs 7SL, U2, 7SK, and HBII-52 are expressed at levels far exceeding mRNAs. C/D and H/ACA box snoRNAs are associated with rRNA methylation and pseudouridylation, respectively: spleen expresses both, hypothalamus expresses mainly C/D box snoRNAs, and testes show enriched expression of both H/ACA box snoRNAs and RNA telomerase TERC. Within the snoRNA 14q cluster, 14q(I-6) is expressed at much higher levels than other cluster members. More reads align to mitochondrial than nuclear tRNAs. Many lincRNAs are actively transcribed, particularly those overlapping known ncRNAs. Within the Prader-Willi syndrome loci, the snoRNA HBII-85 (group I) cluster is highly expressed in hypothalamus, greater than in other tissues and greater than group II or III. Additionally, within the disease locus we find novel transcription across a 400,000 nt span in ovaries. This genome-wide polyA-neutral expression compendium demonstrates the richness of ncRNA expression, their high expression patterns, their function-specific expression patterns, and is publicly available.

ACC2 is expressed at high levels in human white adipose and has an isoform with a novel N-terminus [corrected]

Acetyl-CoA carboxylases ACC1 and ACC2 catalyze the carboxylation of acetyl-CoA to malonyl-CoA, regulating fatty-acid synthesis and oxidation, and are potential targets for treatment of metabolic syndrome. Expression of ACC1 in rodent lipogenic tissues and ACC2 in rodent oxidative tissues, coupled with the predicted localization of ACC2 to the mitochondrial membrane, have suggested separate functional roles for ACC1 in lipogenesis and ACC2 in fatty acid oxidation. We find, however, that human adipose tissue, unlike rodent adipose, expresses more ACC2 mRNA relative to the oxidative tissues muscle and heart. Human adipose, along with human liver, expresses more ACC2 than ACC1. Using RT-PCR, real-time PCR, and immunoprecipitation we report a novel isoform of ACC2 (ACC2.v2) that is expressed at significant levels in human adipose. The protein generated by this isoform has enzymatic activity, is endogenously expressed in adipose, and lacks the N-terminal sequence. Both ACC2 isoforms are capable of de novo lipogenesis, suggesting that ACC2, in addition to ACC1, may play a role in lipogenesis. The results demonstrate a significant difference in ACC expression between human and rodents, which may introduce difficulties for the use of rodent models for development of ACC inhibitors.

The VPS1 protein is a dynamin-like GTPase required for sorting proteins to the yeast vacuole

VPS1 encodes a 79 kDa protein required for the proper sorting of soluble vacuolar proteins in Saccharomyces cerevisiae. The N-terminal half of Vps1p, which contains a consensus GTP-binding motif, shares extensive homology with a growing family of high molecular mass GTP-binding proteins. Members of this family have been implicated in a number of cellular processes. Vps1p most closely resembles the microtubule-associated protein dynamin. As predicted from the sequence, Vps1p binds and hydrolyses GTP. However, no requirement for microtubules was found for Vps1p function in protein sorting. In subcellular fractionation experiments Vps1p associates with the membrane fraction; the C-terminal half of Vps1p is important for this association. Mutational analysis of VPS1 generated two classes of mutations, dominant negative and recessive. The dominant mutations all mapped to the N-terminal half of the protein. Recessive mutations gave rise to either truncated or unstable proteins. A potential Vps1p-interacting protein (Mvp1p) has been isolated by screening for suppressors of the dominant alleles of VPS1. Taken together these results suggest that Vps1p is a two-domain protein that is part of a multi-subunit protein complex involved in vacuolar protein sorting.

Transcripts targeted by the microRNA-16 family cooperatively regulate cell cycle progression

microRNAs (miRNAs) are abundant, approximately 21-nucleotide, noncoding regulatory RNAs. Each miRNA may regulate hundreds of mRNA targets, but the identities of these targets and the processes they regulate are poorly understood. Here we have explored the use of microarray profiling and functional screening to identify targets and biological processes triggered by the transfection of human cells with miRNAs. We demonstrate that a family of miRNAs sharing sequence identity with miRNA-16 (miR-16) negatively regulates cellular growth and cell cycle progression. miR-16-down-regulated transcripts were enriched with genes whose silencing by small interfering RNAs causes an accumulation of cells in G(0)/G(1). Simultaneous silencing of these genes was more effective at blocking cell cycle progression than disruption of the individual genes. Thus, miR-16 coordinately regulates targets that may act in concert to control cell cycle progression.

The colorectal microRNAome

MicroRNAs (miRNAs) are a class of small noncoding RNAs that have important regulatory roles in multicellular organisms. The public miRNA database contains 321 human miRNA sequences, 234 of which have been experimentally verified. To explore the possibility that additional miRNAs are present in the human genome, we have developed an experimental approach called miRNA serial analysis of gene expression (miRAGE) and used it to perform the largest experimental analysis of human miRNAs to date. Sequence analysis of 273,966 small RNA tags from human colorectal cells allowed us to identify 200 known mature miRNAs, 133 novel miRNA candidates, and 112 previously uncharacterized miRNA* forms. To aid in the evaluation of candidate miRNAs, we disrupted the Dicer locus in three human colorectal cancer cell lines and examined known and novel miRNAs in these cells. These studies suggest that the human genome contains many more miRNAs than currently identified and provide an approach for the large-scale experimental cloning of novel human miRNAs in human tissues.

Simple, quantitative primer-extension PCR assay for direct monitoring of microRNAs and short-interfering RNAs

There has been a surge of interest in the biology of microRNAs and the technology of RNA interference. We describe a simple, robust, inexpensive assay for quantitative analysis of microRNAs and short-interfering RNAs. The method relies on primer extension conversion of RNA to cDNA by reverse transcription followed by quantitative, real-time PCR. Technical parameters critical to the success of the assay are presented. Measurements of microRNA levels are sensitive, with most assays allowing measurements in the femtomolar range, which corresponds to tens of copies per cell or less. The assay has a high dynamic range and provides linear readout over differences in microRNA concentrations that span 6-7 orders of magnitude. The assay is capable of discriminating between related microRNA family members that differ by subtle sequence differences. We used the method for quantitative analysis of six microRNAs across 12 tissue samples. The data confirm striking variation in the patterns of expression of these noncoding regulatory RNAs.

Targeted, haplotype-resolved resequencing of long segments of the human genome

Currently, challenges exist to acquire long-range (hundreds of kilobase pairs) phase-discriminated sequence across substantial numbers of individuals. We have developed a straightforward method for isolating and characterizing specific genomic regions in a haplospecific manner. Real-time PCR is carried out to STS content map and genotype pools of fosmid clones arrayed in 384-well microtiter plates. Single-nucleotide polymorphisms, microsatellite markers, and insertion-deletion polymorphisms are used to differentiate the target region into haplotype-specific tiling paths. DNA of clones from these tiling paths is retrieved from the library and either sequenced by standard shotgun methods or amplified in vitro and sequenced by a primer-based, directed method. This approach provides convenient access to complete, haplotype-resolved resequencing data from multiple individuals across tens to hundreds of thousands of basepairs. We illustrate its implementation with a detailed example of more than 400 kbp from the human CFTR region, across 15 individuals, and summarize our experience applying it to many other human loci.

Ancient haplotypes of the HLA Class II region

Allelic variation in codons that specify amino acids that line the peptide-binding pockets of HLA's Class II antigen-presenting proteins is superimposed on strikingly few deeply diverged haplotypes. These haplotypes appear to have been evolving almost independently for tens of millions of years. By complete resequencing of 20 haplotypes across the approximately 100-kbp region that spans the HLA-DQA1, -DQB1, and -DRB1 genes, we provide a detailed view of the way in which the genome structure at this locus has been shaped by the interplay of selection, gene-gene interaction, and recombination.

Expression of alternatively spliced sodium channel alpha-subunit genes. Unique splicing patterns are observed in dorsal root ganglia

Molecular medicine requires the precise definition of drug targets, and tools are now in place to provide genome-wide information on the expression and alternative splicing patterns of any known gene. DNA microarrays were used to monitor transcript levels of the nine well-characterized alpha-subunit sodium channel genes across a broad range of tissues from cynomolgus monkey, a non-human primate model. Alternative splicing of human transcripts for a subset of the genes that are expressed in dorsal root ganglia, SCN8A (Na(v)1.6), SCN9A (Na(v)1.7), and SCN11A (Na(v)1.9) was characterized in detail. Genomic sequence analysis among gene family paralogs and between cross-species orthologs suggested specific alternative splicing events within transcripts of these genes, all of which were experimentally confirmed in human tissues. Quantitative PCR revealed that certain alternative splice events are uniquely expressed in dorsal root ganglia. In addition to characterization of human transcripts, alternatively spliced sodium channel transcripts were monitored in a rat model for neuropathic pain. Consistent down-regulation of all transcripts was observed, as well as significant changes in the splicing patterns of SCN8A and SCN9A.

Hypervariability, suppressed recombination and the genetics of individuality

We define 'genetic individuality' as intraspecies variation that has substantial heritability and involves traits that are sufficiently common that they can be observed in any modest-sized sampling of individuals. We propose that genetic individuality is largely shaped by the combinatory shuffling of a modest number of genes, each of which exists as a family of functionally and structurally diverged alleles. Unequivocal examples of such allele families are found at the O-antigen-biosynthetic locus in Pseudomonas aeruginosa and the human leucocyte antigen locus in humans. We examine characteristic features of these allele families and explore the possibility that genetic loci with similar characteristics can be recognized in a whole-genome scan of human genetic variation.

Effects of methanol concentration on expression levels of recombinant protein in fed-batch cultures of Pichia methanolica

The methylotrophic yeast Pichia methanolica can be used to express recombinant genes at high levels under the control of the methanol-inducible alcohol oxidase (AUG1) promoter. Methanol concentrations during the induction phase directly affect cellular growth and protein yield. Various methanol concentrations controlled by an on-line monitoring and control system were investigated in mixed glucose/methanol fed-batch cultures of P. methanolica expressing the human transferrin N-lobe protein. The PMAD18 P. methanolica strain utilized is a knock-out for the chromosomal AUG1 gene locus, resulting in a slow methanol utilization phenotype. Maximum growth of 100 g of dry cell weight per liter of culture was observed in cultures grown at 1.0% (v/v) methanol concentration. Maximum recombinant gene expression was observed for cultures controlled at 0.7% (v/v) methanol concentration, resulting in maximum volumetric production of 450 mg of transferrin per liter after 72 h of elapsed fermentation time.

Whole-genome sequence variation among multiple isolates of Pseudomonas aeruginosa

Whole-genome shotgun sequencing was used to study the sequence variation of three Pseudomonas aeruginosa isolates, two from clonal infections of cystic fibrosis patients and one from an aquatic environment, relative to the genomic sequence of reference strain PAO1. The majority of the PAO1 genome is represented in these strains; however, at least three prominent islands of PAO1-specific sequence are apparent. Conversely, approximately 10% of the sequencing reads derived from each isolate fail to align with the PAO1 backbone. While average sequence variation among all strains is roughly 0.5%, regions of pronounced differences were evident in whole-genome scans of nucleotide diversity. We analyzed two such divergent loci, the pyoverdine and O-antigen biosynthesis regions, by complete resequencing. A thorough analysis of isolates collected over time from one of the cystic fibrosis patients revealed independent mutations resulting in the loss of O-antigen synthesis alternating with a mucoid phenotype. Overall, we conclude that most of the PAO1 genome represents a core P. aeruginosa backbone sequence while the strains addressed in this study possess additional genetic material that accounts for at least 10% of their genomes. Approximately half of these additional sequences are novel.

Genetic variation at the O-antigen biosynthetic locus in Pseudomonas aeruginosa

The outer carbohydrate layer, or O antigen, of Pseudomonas aeruginosa varies markedly in different isolates of these bacteria, and at least 20 distinct O-antigen serotypes have been described. Previous studies have indicated that the major enzymes responsible for O-antigen synthesis are encoded in a cluster of genes that occupy a common genetic locus. We used targeted yeast recombinational cloning to isolate this locus from the 20 internationally recognized serotype strains. DNA sequencing of these isolated segments revealed that at least 11 highly divergent gene clusters occupy this region. Homology searches of the encoded protein products indicated that these gene clusters are likely to direct O-antigen biosynthesis. The O15 serotype strains lack functional gene clusters in the region analyzed, suggesting that O-antigen biosynthesis genes for this serotype are harbored in a different portion of the genome. The overall pattern underscores the plasticity of the P. aeruginosa genome, in which a specific site in a well-conserved genomic region can be occupied by any of numerous islands of functionally related DNA with diverse sequences.

Linker-mediated recombinational subcloning of large DNA fragments using yeast

The homologous recombination pathway in yeast is an ideal tool for the sequence-specific assembly of plasmids. Complementary 80-nucleotide oligonucleotides that overlap a vector and a target fragment were found to serve as efficient recombination linkers for fragment subcloning. Using electroporation, single-stranded 80-mers were adequate for routine plasmid construction. A cycloheximide-based counterselection was introduced to increase the specificity of cloning by homologous recombination relative to nonspecific vector background. Reconstruction experiments suggest this counterselection increased cloning specificity by 100-fold. Cycloheximide counterselection was used in conjunction with 80-bp linkers to subclone targeted regions from bacterial artificial chromosomes. This technology may find broad application in the final stages of completing the Human Genome Sequencing Project and in applications of BAC clones to the functional analysis of complex genomes.

General method for plasmid construction using homologous recombination

We describe a general method for plasmid assembly that uses yeast and extends beyond yeast-specific research applications. This technology exploits the homologous recombination, double-stranded break repair pathway in Saccharomyces cerevisiae to join DNA fragments. Synthetic, double-stranded "recombination linkers" were used to "subclone" a DNA fragment into a plasmid with > 80% efficiency. Quantitative data on the influence of DNA concentration and overlap length on the efficiency of recombination are presented. Using a simple procedure, plasmids were shuttled from yeast into E. coli for subsequent screening and large-scale plasmid preps. This simple method for plasmid construction has several advantages. (i) It bypasses the need for extensive PCR amplification and for purification, modification and/or ligation techniques routinely used for plasmid constructions. (ii) The method does not rely on available restriction sites, thus fragment and vector DNA can be joined within any DNA sequence. This enables the use of multifunctional cloning vectors for protein expression in mammalian cells, other yeast species, E. coli and other expression systems as discussed. (iii) Finally, the technology exploits yeast strains, plasmids and microbial techniques that are inexpensive and readily available.

Development of the methylotrophic yeast Pichia methanolica for the expression of the 65 kilodalton isoform of human glutamate decarboxylase

We describe a protein expression system in the methylotrophic yeast, Pichia methanolica. Methods for transformation and genetic manipulation of the organism were developed using an ade2 strain and the wild-type ADE2 gene. A vacuolar protease-deficient strain was constructed. Two genes encoding alcohol oxidases were found, yet a single isoform of alcohol oxidase was produced during methanol-fed fermentations. The promoter from this gene was used to drive expression. An integrating plasmid for the cytoplasmic expression of the 65 kDa isoform of human glutamate decarboxylase (human GAD65) was assembled. A strain harboring eight copies of this plasmid expressed enzymatically active human GAD65 at levels approaching 0.5 g/l. Identical amounts were made in Pichia pastoris. The recombinant GAD65 was purified to greater than 90% purity.

Morphological classification of the yeast vacuolar protein sorting mutants: evidence for a prevacuolar compartment in class E vps mutants

The collection of vacuolar protein sorting mutants (vps mutants) in Saccharomyces cerevisiae comprises of 41 complementation groups. The vacuoles in these mutant strains were examined using immunofluorescence microscopy. Most of the vps mutants were found to possess vacuolar morphologies that differed significantly from wild-type vacuoles. Furthermore, mutants representing independent vps complementation groups were found to share aberrant morphological features. Six distinct classes of vacuolar morphology were observed. Mutants from eight vps complementation groups were defective both for vacuolar segregation from mother cells into developing buds and for acidification of the vacuole. Another group of mutants, represented by 13 complementation groups, accumulated a novel organelle distinct from the vacuole that contained a late-Golgi protein, active vacuolar H(+)-ATPase complex, and soluble vacuolar hydrolases. We suggest that this organelle may represent an exaggerated endosome-like compartment. None of the vps mutants appeared to mislocalize significant amounts of the vacuolar membrane protein alkaline phosphatase. Quantitative immunoprecipitations of the soluble vacuolar hydrolase carboxypeptidase Y (CPY) were performed to determine the extent of the sorting defect in each vps mutant. A good correlation between morphological phenotype and the extent of the CPY sorting defect was observed.

The VPS1 protein, a homolog of dynamin required for vacuolar protein sorting in Saccharomyces cerevisiae, is a GTPase with two functionally separable domains

The product of the VPS1 gene, Vps1p, is required for the sorting of soluble vacuolar proteins in the yeast Saccharomyces cerevisiae. We demonstrate here that Vps1p, which contains a consensus tripartite motif for guanine nucleotide binding, is capable of binding and hydrolyzing GTP. Vps1p is a member of a subfamily of large GTP-binding proteins whose members include the vertebrate Mx proteins, the yeast MGM1 protein, the Drosophila melanogaster shibire protein, and dynamin, a bovine brain protein that bundles microtubules in vitro. Disruption of microtubules did not affect the fidelity or kinetics of vacuolar protein sorting, indicating that Vps1p function is not dependent on microtubules. Based on mutational analyses, we propose a two-domain model for Vps1p function. When VPS1 was treated with hydroxylamine, half of all mutations isolated were found to be dominant negative with respect to vacuolar protein sorting. All of the dominant-negative mutations analyzed further mapped to the amino-terminal half of Vps1p and gave rise to full-length protein products. In contrast, recessive mutations gave rise to truncated or unstable protein products. Two large deletion mutations in VPS1 were created to further investigate Vps1p function. A mutant form of Vps1p lacking the carboxy-terminal half of the protein retained the capacity to bind GTP and did not interfere with sorting in a wild-type background. A mutant form of Vps1p lacking the entire GTP-binding domain interfered with vacuolar protein sorting in wild-type cells. We suggest that the amino-terminal domain of Vps1p provides a GTP-binding and hydrolyzing activity required for vacuolar protein sorting, and the carboxy-terminal domain mediates Vps1p association with an as yet unidentified component of the sorting apparatus.

Molecular analysis of the yeast VPS3 gene and the role of its product in vacuolar protein sorting and vacuolar segregation during the cell cycle

vps3 mutants of the yeast Saccharomyces cerevisiae are impaired in the sorting of newly synthesized soluble vacuolar proteins and in the acidification of the vacuole (Rothman, J. H., and T. H. Stevens. Cell. 47:1041-1051; Rothman, J. H., C. T. Yamashiro, C. K. Raymond, P. M. Kane, and T. H. Stevens. 1989. J. Cell Biol. 109:93-100). The VPS3 gene, which was cloned using a novel selection procedure, encodes a low abundance, hydrophilic protein of 117 kD that most likely resides in the cytoplasm. Yeast strains bearing a deletion of the VPS3 gene (vps3-delta 1) are viable, yet their growth rate is significantly reduced relative to wild-type cells. Temperature shift experiments with strains carrying a temperature conditional vps3 allele demonstrate that cells rapidly lose the capacity to sort the vacuolar protein carboxypeptidase Y upon loss of VPS3 function. Vacuolar morphology was examined in wild-type and vps3-delta 1 yeast strains by fluorescence microscopy. The vacuoles in wild-type yeast cells are morphologically complex, and they appear to be actively partitioned between mother cells and buds during an early phase of bud growth. Vacuolar morphology in vps3-delta 1 mutants is significantly altered from the wild-type pattern, and the vacuolar segregation process seen in wild-type strains is defective in these mutants. With the exception of a vacuolar acidification defect, the phenotypes of vps3-delta 1 strains are significantly different from those of mutants lacking the vacuolar proton-translocating ATPase. These data demonstrate that the acidification defect in vps3-delta 1 cells is not the primary cause of the pleiotropic defects in vacuolar function observed in these mutants.

A putative GTP binding protein homologous to interferon-inducible Mx proteins performs an essential function in yeast protein sorting

Members of the Mx protein family promote interferon-inducible resistance to viral infection in mammals and act by unknown mechanisms. We identified an Mx-like protein in yeast and present genetic evidence for its cellular function. This protein, the VPS1 product, is essential for vacuolar protein sorting, normal organization of intracellular membranes, and growth at high temperature, implying that Mx-like proteins are engaged in fundamental cellular processes in eukaryotes. Vps1p contains a tripartite GTP binding motif, which suggests that binding to GTP is essential to its role in protein sorting. Vps1p-specific antibody labels punctate cytoplasmic structures that condense to larger structures in a Golgi-accumulating sec7 mutant; thus, Vps1p may associate with an intermediate organelle of the secretory pathway.

Acidification of the lysosome-like vacuole and the vacuolar H+-ATPase are deficient in two yeast mutants that fail to sort vacuolar proteins

Organelle acidification plays a demonstrable role in intracellular protein processing, transport, and sorting in animal cells. We investigated the relationship between acidification and protein sorting in yeast by treating yeast cells with ammonium chloride and found that this lysosomotropic agent caused the mislocalization of a substantial fraction of the newly synthesized vacuolar (lysosomal) enzyme proteinase A (PrA) to the cell surface. We have also determined that a subset of the vpl mutants, which are deficient in sorting of vacuolar proteins (Rothman, J. H., and T. H. Stevens. 1986. Cell. 47:1041-1051; Rothman, J. H., I. Howald, and T. H. Stevens. EMBO [Eur. Mol. Biol. Organ.] J. In press), failed to accumulate the lysosomotropic fluorescent dye quinacrine within their vacuoles, mimicking the phenotype of wild-type cells treated with ammonium. The acidification defect of vpl3 and vpl6 mutants correlated with a marked deficiency in vacuolar ATPase activity, diminished levels of two immunoreactive subunits of the protontranslocating ATPase (H+-ATPase) in purified vacuolar membranes, and accumulation of the intracellular portion of PrA as the precursor species. Therefore, some of the VPL genes are required for the normal function of the yeast vacuolar H+-ATPase complex and may encode either subunits of the enzyme or components required for its assembly and targeting. Collectively, these findings implicate a critical role for acidification in vacuolar protein sorting and zymogen activation in yeast, and suggest that components of the yeast vacuolar acidification system may be identified by examining mutants defective in sorting of vacuolar proteins.

Characterization of three genomic loci encoding Rhizobium sp. strain ORS571 N2 fixation genes

Sixty-five independent, N2 fixation-defective (Nif-) vector insertion (Vi) mutants were selected, cloned, and mapped to the ORS571 genome. The recombinant Nif::Vi plasmids obtained in this way were used as DNA hybridization probes to isolate homologous phages from a genomic library of ORS571 constructed in lambda EMBL3. Genomic maps were drawn for three ORS571 Nif gene loci. Forty-five Nif::Vi mutants in genomic Nif locus 1 defined two gene clusters separated by 8 kilobase pairs (kb) of DNA. In the first cluster, 36 Nif::Vi mutants mapped to a 7-kb DNA segment that showed DNA homology with Klebsiella pneumoniae nifHDKE and encoded at least two Nif operons. In the other cluster, nine Nif::Vi mutants mapped to a 1.5-kb DNA segment that showed homology with K. pneumoniae and Rhizobium meliloti nifA; this DNA segment encoded a separate Nif operon. Fifteen Nif::Vi mutants mapped to a 3.5-kb DNA segment defined as Nif locus 2 and showed DNA homology with the R. meliloti P2 fixABC operon. Nif locus 2 carries a second nifH (nifH2) gene. Four Nif::Vi mutants mapped to a 2-kb DNA segment defined as Nif locus 3 and showed DNA homology with K. pneumoniae nifB. DNA from lambda Nif phages comprising all three genomic Nif loci was subcloned in plasmid vectors able to stably replicate in ORS571. These plasmid subclones were introduced into ORS571 strains carrying physically mapped Nif::Vi insertions, and genetic complementations were conducted. With the exception of certain mutants mapping to the nifDK genes, all mutants could be complemented to Nif+ when they carried plasmid subclones of defined genomic DNA regions. Conversely, most nifDK mutants behaved as pseudodominant alleles.

Extension of bacteriophage lambda host range: selection, cloning, and characterization of a constitutive lambda receptor gene

A set of plasmids has been constructed that carry a constitutive lamB gene (LamBc phenotype) from Escherichia coli and that confer functional phage lambda receptors to bacteria other than E. coli. This E. coli LamBc strain has been selected to escape both maltose-inducible and glucose-repressible control. Constitutivity results from an IS-3 insertion, carrying a mobile promoter, proximal to lamB. The LamBc DNA has been cloned into both broad and narrow host-range plasmids, and the resulting pTROY plasmids have been transferred to diverse bacteria. Both Salmonella typhimurium/pTROY and Klebsiella pneumoniae/pTROY strains efficiently adsorb phage lambda; Pseudomonas aeruginosa/pTROY strains do not. Introduction of a functional E. coli LamB protein into foreign bacterial will allow these bacteria carrying pTROY plasmids to be infected by phage lambda recombinant DNA libraries, phage lambda::Tn insertion mutagenesis vectors, and in vivo lambda-packaged cosmids.