Statistical analysis of MPSS measurements: application to the study of LPS-activated macrophage gene expression.

Massively Parallel Signature Sequencing (MPSS), a recently developed high-throughput transcription profiling technology, has the ability to profile almost every transcript in a sample without requiring prior knowledge of the sequence of the transcribed genes. As is the case with DNA microarrays, effective data analysis depends crucially on understanding how noise affects measurements. We analyze the sources of noise in MPSS and present a quantitative model describing the variability between replicate MPSS assays. We use this model to construct statistical hypotheses that test whether an observed change in gene expression in a pair-wise comparison is significant. This analysis is then extended to the determination of the significance of changes in expression levels measured over the course of a time series of measurements. We apply these analytic techniques to the study of a time series of MPSS gene expression measurements on LPS-stimulated macrophages. To evaluate our statistical significance metrics, we compare our results with published data on macrophage activation measured by using Affymetrix GeneChips.

Two dominant mutations in the mouse fused gene are the result of transposon insertions.

The mouse Fused locus encodes a protein that has been implicated in the regulation of embryonic axis formation. The protein, which has been named Axin to distinguish it from the product of the unrelated Drosophila melanogaster gene fused, contains regions of similarity to the RGS (regulators of G-protein signaling) family of proteins as well as to dishevelled, a protein that acts downstream of Wingless in D. melanogaster. Loss-of-function mutations at Fused lead to lethality between days 8 and 10 of gestation. Three dominant mutations result in a kinked tail in heterozygotes. Two of the dominant mutations, Fused and Knobbly, result from insertion of intracisternal A particle retrotransposons into the gene. The insertion in Fused, within the sixth intron, creates a gene that produces wild-type transcripts as well as mutant transcripts that initiate at both the authentic promoter and the 3'-most long terminal repeat of the insertion. Knobbly, an insertion of the retrotransposon into exon 7, precludes the production of wild-type protein. Thus the Fused homozygote is viable whereas Knobbly is a recessive embryonic lethal. In both mutants the dominant kink-tailed phenotype is likely to result from the synthesis of similar amino-terminal fragments of Axin protein that would contain the RGS domain, but lack the dishevelled domain.

The mouse Fused locus encodes Axin, an inhibitor of the Wnt signaling pathway that regulates embryonic axis formation.

Mutations at the mouse Fused locus have pleiotropic developmental effects, including the formation of axial duplications in homozygous embryos. The product of the Fused locus, Axin, displays similarities to RGS (Regulators of G-Protein Signaling) and Dishevelled proteins. Mutant Fused alleles that cause axial duplications disrupt the major mRNA, suggesting that Axin negatively regulates the response to an axis-inducing signal. Injection of Axin mRNA into Xenopus embryos inhibits dorsal axis formation by interfering with signaling through the Wnt pathway. Furthermore, ventral injection of an Axin mRNA lacking the RGS domain induces an ectopic axis, apparently through a dominant-negative mechanism. Thus, Axin is a novel inhibitor of Wnt signaling and regulates an early step in embryonic axis formation in mammals and amphibians.

Phenotypic and molecular analysis of a transgenic insertional allele of the mouse Fused locus.

Spontaneous mutations at the mouse Fused (Fu) locus cause dominant skeletal and neurological defects and recessive lethal embryonic defects including neuroectodermal abnormalities and axial duplications. Here, we describe a new allele at the Fu locus caused by a transgenic insertional mutation, H epsilon 46. Embryos homozygous for the H epsilon 46 insertion die at day 9-10 post coitum and display phenotypic defects similar to those associated with Fu alleles. The H epsilon 46 locus was cloned and shown to contain a 20-kb deletion at the site of transgene insertion with no other detectable rearrangements. Genomic probes from the H epsilon 46 locus were mapped to a genetic locus closely linked to Fu on chromosome 17 and were hybridized to a YAC contig covering the FuKi critical region. Compound heterozygotes between H epsilon 46 and FuKi were inviable and displayed abnormalities at the same stage of embryogenesis as do homozygotes for either of the two mutations, demonstrating that these two recessive lethal mutations belong to the same complementation group. A genomic probe from the wild-type H epsilon 46 locus detected a transcript that is disrupted by the transgenic insertion, representing a candidate for the wild-type allele of Fused.

B-less: a strain of profoundly B cell-deficient mice expressing a human lambda transgene.

We have created several transgenic mouse strains that bear the human lambda light chain gene driven by its own promoter and a mouse immunoglobulin heavy chain enhancer. The transgene is expressed in many tissues, with particularly high levels of expression in the bone marrow, thymus, spleen, and lymph nodes. One of these transgenic lines, B-less, displays a dramatic phenotype characterized by an acute susceptibility to bacterial and viral infections. Analysis of this strain shows it to be profoundly deficient in both immature (pre-B) and mature B cells, as well as in circulating immunoglobulin. The pre-B and B cell defects are cell autonomous, as judged by cell culture and bone marrow graft chimeras. Despite this B cell deficiency, the T cell lineage appears grossly normal as assessed by flow cytometric analysis and by its response to mitogen stimulation. Since an independently derived transgenic strain bearing the same human lambda construct displays a partial B-less phenotype, it is likely that the B lineage deficiency is due to a dominant effect of transgene expression rather than to the insertional perturbation of an endogenous mouse gene. It is interesting that the deficiency phenotype is fully expressed in the FVB/N genetic background, but is suppressed in F1 hybrids formed between the FVB/N and C57BL/6 inbred strains. Evidently, there are one or more dominant genetic suppressors of B-less in the C57BL/6 genome.

Structure and expression of the human immunoglobulin lambda genes.

We determined the DNA sequence of two large regions of chromosome 22: 33.7 kb containing the C lambda complex; and 5.2 kb 5' of the functionally rearranged lambda gene from the human myeloma, U266. Analysis of these sequences reveals the complete structure of the human C lambda complex and a previously undescribed seventh C lambda region that may encode the Ke+Oz- lambda protein. The seven constant regions are organized in a tandem array, and each is preceded by a single J lambda region. lambda 1, lambda 2, lambda 3, and lambda 7 are apparently active genes, while lambda 4, lambda 5, and lambda 6 are pseudogenes. There are no other J lambda or C lambda regions within a 60-kb region surrounding the C lambda complex; however, there are at least four other lambda-like genes and lambda pseudogenes in the human genome. The lambda genes appear to have evolved via a series of gene duplication events resulting from unequal crossing over or gene conversion between the highly conserved C lambda regions on mispaired chromosomes. The lack of Alu sequences in this large segment of DNA suggests that the C lambda complex resulted from a recent amplification of a smaller Alu-free segment of DNA. Illegitimate recombination between repeated sequences containing lambda 2 and lambda 3 may be responsible for variable amplification of the lambda genes. We also found a 1,377-bp open reading frame (ORF) located on the opposite strand in the region containing lambda 7. While this ORF is flanked by potential RNA splicing signals, we have no evidence that it is part of a functional gene. We also discovered a V lambda pseudogene, called psi V lambda 1, 3 kb upstream of the U266 lambda gene. Using primer extension analysis to map the transcription start in the human lambda gene, we have identified its initiation point 41 bp upstream of the initiation codon. Analysis of the lambda promoter reveals that it contains a TATAA box at position -29 relative to the transcription initiation site and an octamer sequence at -67. Computer analysis of 40 kb of DNA sequences surrounding the human lambda locus has revealed no sequences resembling the kappa or IgH transcriptional enhancers, nor have in vitro analyses for function revealed enhancer activity. A comparison of these results with those obtained in separate studies with transgenic mice point to a complex, developmentally linked mechanism of transcriptional activation.

Nucleotide sequence of the human parathyroid hormone gene.

From a lambda phage gene library we have isolated phage containing the gene encoding human preproparathyroid hormone. The phage were isolated by using both the plaque-hybridization technique and the in vivo recombination-selection technique. The human preproparathyroid hormone gene contains two intervening sequences that separate the gene into a 5' noncoding domain, a "prepro" sequence domain, and a domain containing the parathyroid hormone sequence and the 3' noncoding region. The gene is approximately 4,200 base pairs long. Restriction endonuclease analysis of human leukocyte DNA shows that the haploid human genome contains one copy of the preproparathyroid hormone gene. A 14-base-pair sequence of alternating purines and pyrimidines that has the potential of adopting the Z-DNA conformation lies 134 base pairs upstream from the presumed site of initiation of transcription.

Transport and cleavage of bacterial pre-beta-lactamase by mammalian microsomes.

Eucaryotic endoplasmic reticulum and bacterial inner cell membranes use strikingly similar mechanisms to transport proteins from one cellular compartment to another. Using a mammalian linked transcription-translation system, we show here that canine pancreatic microsomes accurately cleave off the signal sequence of beta-lactamase, a secreted bacterial protein. We suggest that the apparent differences between bacterial and eukaryotic protein transport may involve only minor modulations of a profoundly conserved mechanism.