Lymphoid V(D)J recombination. Functional analysis of the spacer sequence within the recombination signal

The V(D)J recombination reaction is directed by a pair of signal sequences, each consisting of a palindromic heptamer, an A/T-rich nonamer, and an intervening spacer sequence of 12 or 23 base pairs. The spacer sequence previously has not been analyzed for a functional role. In this study, numerous simultaneous sequence changes have been made in the spacer of each signal to test their functional importance. All of the AT base pairs in each signal were changed to GC base pairs. This particular change is of interest because it markedly increases the energy that would be required to melt out the two strands of each signal to permit the intersignal base pairing proposed in a commonly invoked model for signal-signal interaction in V(D)J recombination. We find that changing 6 to 12 AT base pairs in the 12-signal to GC does not affect V(D)J recombination, nor does changing 11 of 23 AT base pairs in the 23-signal. Substrates with all-GC spacer sequences in both the 12- and the 23-signal also recombine at efficiencies that are not significantly reduced. These studies demonstrate that the sequences at these particular positions are not recognized by the recombinase. In addition, the data do not support models invoking signal-signal base pairing.

Construction of high coverage whole-genome sequencing libraries from single colon crypts without DNA extraction or whole-genome amplification

OBJECTIVE

Comprehensive and reliable genome-wide variant analysis of a small number of cells has been challenging due to genome coverage bias, PCR over-cycling, and the requirement of expensive technologies. To comprehensively identify genome alterations in single colon crypts that reflect genome heterogeneity of stem cells, we developed a method to construct whole-genome sequencing libraries from single colon crypts without DNA extraction, whole-genome amplification, or increased PCR enrichment cycles.

RESULTS

We present post-alignment statistics of 81 single-crypts (each contains four- to eight-fold less DNA than the requirement of conventional methods) and 16 bulk-tissue libraries to demonstrate the consistent success in obtaining reliable coverage, both in depth (≥ 30X) and breadth (≥ 92% of the genome covered at ≥ 10X depth), of the human genome. These single-crypt libraries are of comparable quality as libraries generated with the conventional method using high quality and quantities of purified DNA. Conceivably, our method can be applied to small biopsy samples from many tissues and can be combined with single cell targeted sequencing to comprehensively profile cancer genomes and their evolution. The broad potential application of this method offers expanded possibilities in cost-effectively examining genome heterogeneity in small numbers of cells at high resolution.

Abstract 2634: Polymorphisms in DNA repair genes and risk of multiple myeloma

Background and Objective: DNA repair is an essential process for development and differentiation of all cells, including plasma cells. Associations between single nucleotide polymorphisms (SNPs) in DNA repair genes and multiple myeloma risk have been reported but generally not replicated. In a pilot study, we observed suggestive evidence of an association between multiple myeloma risk and five nonsynonymous SNPs in DNA repair genes (rs25489 in XRCC1, rs1801516 in ATM, rs2227999 in XPC, rs2228528 in CSB and rs2228615 in ICAM5). The objective of this study was to examine the effect of these five SNPs on multiple myeloma risk in a large multi-center analysis. Methods: We conducted a pooled analysis in a multi-ethnic sample of 489 multiple myeloma cases and 939 controls. Studies contributing samples included two population-based case-control studies conducted in Los Angeles and Seattle/Detroit that used the Surveillance, Epidemiology and End Results (SEER) cancer registries to identify cases, and three nested case-control studies using subjects from the Multiethnic Cohort (MEC), the Health Professionals Follow Up Study (HPFS), and the Nurses’ Health Study (NHS). The SNPs coded for an amino acid substitution, and were assumed to have the same effect across ethnic groups, and thus subjects from all racial/ethnic groups were included in the analysis. Assuming a log-additive inheritance model, odds ratios (OR) and 95% confidence intervals (95% CIs) were calculated for allele-specific risk comparing cases to controls using unconditional logistic regression, adjusting for age, race/ethnicity, sex, and study site. Results: We observed a 40% decreased risk per variant T allele in the XRCC1 SNP rs25489 (OR=0.60, 95% CI=0.40, 0.91, p-value=0.02). The minor allele frequency in cases and controls was 3.4% and 5.6%, respectively. Non-significant OR's < 1.0 were observed when whites (OR=0.67, 95% CI=0.38, 1.17) and African Americans (OR=0.33, 95% CI= 0.09, 1.20) were considered separately. The SNPs in other DNA repair genes, ATM (rs1801516), XPC (rs2227999), CSB (rs2228528) and ICAM5 (rs2228615), showed no association with multiple myeloma risk. Conclusion: We observed an association between multiple myeloma and XRCC1, a gene that encodes a protein involved in the repair of single-stranded DNA breaks caused by ionizing radiation and alkylating agents. The variant allele of rs25489 results in a missense mutation substituting an amino acid (Arg > His). Several studies have shown diminished DNA repair capacity associated with the variant allele of this SNP, an effect which may be relevant to multiple myeloma etiology. Replication in additional studies is needed to validate this finding.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2634. doi:1538-7445.AM2012-2634

Interaction of chlorpromazine with the human erythrocyte membrane

The interaction of the amphipath chlorpromazine (CPZ) with the human erythrocyte membrane was evaluated. The partition coefficient of CPZ between the membrane bilayer and the aqueous compartment, measured spectrophotometrically, ranged between 1 and 3 X 10(3). An independent estimate, 4.6 X 10(3), was obtained by a novel method which avoided the measurement of binding and determined instead the variation of the hemolytic potency of the amphipath with the ratio of buffer volume to membrane volume. The maximal uptake of CPZ exceeded 2 X 10(9) molecules/red cell, corresponding to a volume greater than that of the bilayer itself. Such heavily loaded membranes were increased in thickness more than 2-fold, suggesting the formation of a CPZ-rich zone at the center of the bilayer. Ghosts loaded with massive levels of CPZ condensed approximately 20-fold in surface area and increased proportionately in thickness, suggesting the formation of a novel CPZ-lipid solution. CPZ caused hemolysis by a colloid-osmotic mechanism. By measuring the simultaneous uptake of mannitol and sucrose, we determined that CPZ induced holes of constant size but variable number. If circular, the holes would have had a diameter of approximately 14 A. The time-averaged number of holes ranged from 0.09 per cell (signifying intermittency) to 16. Freeze-fracture electron microscopy of CPZ-treated red cells revealed multiple round patches of nearly particle-free bilayer up to 0.3 micron in diameter with crowding of the intramembrane particles into the surrounding membrane. We interpret these images to signify lateral phase separation within the CPZ-treated bilayer. Hemolysis could, therefore, result from the intermittent opening of weak seams at phase boundaries; these could then be fluctuating slits approximately 14 A in width and of variable length, rather than simple circular holes.

Chromosomal translocations and non-B DNA structures in the human genome

The mechanisms of chromosomal translocations in mammalian cells have been largely undefined. Recent progress on the most common translocation in human cancer, t(14;18), highlights interesting issues in DNA structure and in the enzymes involved in the cutting and joining phases of the process.

Human DNA-activated protein kinase (DNA-PK) is homologous to phosphatidylinositol kinases

DNA-activated protein kinase (DNA-PK) is a serine/threonine protein kinase that interacts with a DNA end-binding heterodimeric protein, Ku, and is activated by double-stranded DNA. Genomic clones that contain the DNA-PK gene complement the murine scid defect, indicating that DNA-PK affects double-strand break repair and V(D)J recombination. Here we describe the cDNA sequence of the region that corresponds to about 100 kDa of C-terminal sequence of this large (> p350 kDa) protein. This region contains a kinase domain that has strong homology to phosphatidylinositol kinases.

Distinct roles for RAG-1 in the initiation of V(D)J recombination and in the resolution of coding ends

Although RAG-1 and RAG-2 have been shown to be indispensible for V(D)J recombination, their exact role in this reaction remains unclear. Co-transfecting RAG-1 and RAG-2 expression vectors into NIH3T3 fibroblasts confers V(D)J recombination activity to these otherwise recombinationally inactive cells. In this report we have found that in transient transfections of mouse NIH3T3 fibroblasts with RAG-1 and RAG-2 and the appropriate recombination substrates, one RAG-1 expression vector, pRAG-1A, is capable of yielding both signal joints and coding joints, while another RAG-1 expression vector, pRAG-1B, yields only signal joints. The RAG-1 open reading frame for these two expression vectors is interchangeable, indicating that the inability to resolve coding joints is due to the 45-base pair difference found in the 5'-untranslated regions of these constructs. Differences in this region result in a 15-fold difference in gene expression when the luciferase coding region is substituted for the RAG-1 cDNA. This report provides evidence that RAG-1 may have a role in both the initiation of V(D)J recombination as well as the resolution of coding ends. The data also suggest that these RAG-1 activities may be dependent on different levels of RAG-1 expression.

V(D)J recombination on minichromosomes is not affected by transcription

It has been shown previously by others that transcription is temporally correlated with the onset of V(D)J recombination at the endogenous antigen receptor loci. We have been interested in determining whether this temporal correlation indicates a causal connection between these two processes. We have compared V(D)J recombination minichromosome substrates that have transcripts running through the recombination zone with substrates that do not in a transient transfection assay. In this system, the substrates acquire a minichromosome conformation within the first several hours after transfection. We find that the substrates recombine equally well over a 100-fold range in transcriptional variation. In additional studies, we have taken substrates that have low levels of transcription and inhibited transcription further by methylating the substrate DNA or by treating the cells with a general transcription inhibitor (alpha-amanitin). Although these treatments decrease the level of expression an additional 10-100-fold, there is still no observable effect on V(D)J recombination. Based on these results, we conclude that transcription is not necessary for the V(D)J reaction mechanism and does not alter substrate structure at the DNA level or at the simplest levels of chromatin structure in a way that affects the reaction.

The flexible and iterative steps within the NHEJ pathway

Cellular and biochemical studies of nonhomologous DNA end joining (NHEJ) have long established that nuclease and polymerase action are necessary for the repair of a very large fraction of naturally-arising double-strand breaks (DSBs). This conclusion is derived from NHEJ studies ranging from yeast to humans and all genetically-tractable model organisms. Biochemical models derived from recent real-time and structural studies have yet to incorporate physical space or timing for DNA end processing. In real-time single molecule FRET (smFRET) studies, we analyzed NHEJ synapsis of DNA ends in a defined biochemical system. We described a Flexible Synapsis (FS) state in which the DNA ends were in proximity via only Ku and XRCC4:DNA ligase 4 (X4L4), and in an orientation that would not yet permit ligation until base pairing between one or more nucleotides of microhomology (MH) occurred, thereby allowing an in-line Close Synapsis (CS) state. If no MH was achievable, then XLF was critical for ligation. Neither FS or CS required DNA-PKcs, unless Artemis activation was necessary to permit local resection and subsequent base pairing between the two DNA ends being joined. Here we conjecture on possible 3D configurations for this FS state, which would spatially accommodate the nuclease and polymerase processing steps in an iterative manner. The FS model permits repeated attempts at ligation of at least one strand at the DSB after each round of nuclease or polymerase action. In addition to activation of Artemis, other possible roles for DNA-PKcs are discussed.

The scid factor on human chromosome 8 restores V(D)J recombination in addition to double-strand break repair

The murine severe combined immune deficiency mutation (scid) is characterized by a lack of B- and T-lymphoid cells due to a defect in lymphoid V(D)J recombination. Moreover, defective rejoining of DNA double-strand breaks (dsb) in scid cells also results in a marked increase in sensitivity to ionizing radiation. Recently, the putative human homologue of the murine scid gene locus, HYRC1, was assigned to human chromosome 8q11, based on the radiation sensitivity of scid cells as compared to scid:human cell hybrids carrying portions of human chromosome 8. Given the precedent (e.g., ataxia-telangiectasia) for genes other than the affected one being able to complement radiation defects, we were interested in determining if the V(D)J recombination defect was also corrected by the HYRC1 locus. The V(D)J recombination analysis using extrachromosomal DNA substrates in control scid cells (SC3VA2) versus complemented cells (RD13B2) indicates that the radiation sensitivity-complemented cells (RD13B2) are also fully complemented for the V(D)J recombination reaction, whereas the control (uncomplemented) cells (SC3VA2) fail to carry out V(D)J recombination normally. Slightly over 60% of the radiation-induced dsb are rejoined even in scid cells, and this alternative pathway is temperature sensitive. Only the remaining 30-35% of dsb require the introduction of the HYRC1 locus, and this pathway is not temperature sensitive. This merely partial contribution of the scid factor to the repair process suggests the presence of another pathway of dsb repair. Our results indicate that the HYRC1 locus, assigned to human chromosome 8q11, encodes the scid factor, which is involved in all V(D)J recombination coding joint formation and in 30-35% of dsb repair by the temperature-resistant pathway.

V(D)J recombination in ataxia telangiectasia, Bloom's syndrome, and a DNA ligase I-associated immunodeficiency disorder

Ataxia telangiectasia (AT) and Bloom's syndrome (BS) patients are characterized by sensitivity to radiation, increased lymphoid malignancy, and frequent translocations to the antigen receptor loci. Because of these features, there has been a persistent question as to whether the V(D)J recombinase might be abnormal in cells from these patients. Such abnormalities might be due to inappropriate to inaccurate expression of components of the V(D)J recombinase or due to mutation in a component shared between V(D)J recombination and other cellular processes, such as DNA repair. Bloom's syndrome is associated with a ligation deficiency, and this activity may contribute in the end resolution steps of both site-specific and general DNA-processing reactions. In the current study, we have activated V(D)J recombination in normal, AT, and BS fibroblasts and in fibroblasts from a patient with mutations that largely abolish DNA ligase I activity. We find that the signal and coding joint formation of the V(D)J recombination reaction are entirely normal in AT, BS, and DNA ligase I mutant cells. In addition to ruling out abnormalities of the V(D)J recombinase in AT, BS, and DNA ligase I mutant cells, these studies suggest that DNA ligase I is unlikely to be required for signal or coding end joining in the V(D)J recombination reaction.

Pol X DNA polymerases contribute to NHEJ flexibility

New work on DNA polymerase λ highlights its remarkable flexibility. This fits with the generally adaptable nature of the DNA-repair process in which this enzyme is involved — nonhomologous end-joining — which allows this mechanism to handle diverse types of broken DNA ends in order to restore the duplex structure, albeit with a loss of information at the join.