A Bioinformatics-Based Alternative mRNA Splicing Code that May Explain Some Disease Mutations Is Conserved in Animals

Deep sequencing of cDNAs made from spliced mRNAs indicates that most coding genes in many animals and plants have pre-mRNA transcripts that are alternatively spliced. In pre-mRNAs, in addition to invariant exons that are present in almost all mature mRNA products, there are at least 6 additional types of exons, such as exons from alternative promoters or with alternative polyA sites, mutually exclusive exons, skipped exons, or exons with alternative 5′ or 3′ splice sites. Our bioinformatics-based hypothesis is that, in analogy to the genetic code, there is an “alternative-splicing code” in introns and flanking exon sequences, analogous to the genetic code, that directs alternative splicing of many of the 36 types of introns. In humans, we identified 42 different consensus sequences that are each present in at least 100 human introns. 37 of the 42 top consensus sequences are significantly enriched or depleted in at least one of the 36 types of introns. We further supported our hypothesis by showing that 96 out of 96 analyzed human disease mutations that affect RNA splicing, and change alternative splicing from one class to another, can be partially explained by a mutation altering a consensus sequence from one type of intron to that of another type of intron. Some of the alternative splicing consensus sequences, and presumably their small-RNA or protein targets, are evolutionarily conserved from 50 plant to animal species. We also noticed the set of introns within a gene usually share the same splicing codes, thus arguing that one sub-type of splicesosome might process all (or most) of the introns in a given gene. Our work sheds new light on a possible mechanism for generating the tremendous diversity in protein structure by alternative splicing of pre-mRNAs.

Transcription poisoning by Topoisomerase I is controlled by gene length, splice sites, and miR-142-3p

Topoisomerase I (Top1) relaxes DNA supercoiling by forming transient cleavage complexes (Top1cc) up- and downstream of transcription complexes. Top1cc can be trapped by carcinogenic and endogenous DNA lesions and by camptothecin, resulting in transcription blocks. Here, we undertook genome-wide analysis of camptothecin-treated cells at exon resolution. RNA samples from HCT116 and MCF7 cells were analyzed with the Affy Exon Array platform, allowing high-resolution mapping along 18,537 genes. Long genes that are highly expressed were the most susceptible to downregulation, whereas short genes were preferentially upregulated. Along the body of genes, downregulation was most important toward the 3'-end and increased with the number of exon-intron junctions. Ubiquitin and RNA degradation-related pathway genes were selectively downregulated. Parallel analysis of microRNA with the Agilent miRNA microarray platform revealed that miR-142-3p was highly induced by camptothecin. More than 10% of the downregulated genes were targets of this p53-dependent microRNA. Our study shows the profound impact of Top1cc on transcription elongation, especially at intron-exon junctions and on transcript stability by microRNA miR-142-3p upregulation.

Abstract 5219: Molecular phenotype of an epithelial-like subset of the NCI-60 human tumor cell lines and relevance to gene expression patterns in TCGA normal breast and breast cancer tissue samples

Better cancer therapy is likely to come from molecular characterization of individual patients’ tumors that will permit selection and development of tailored therapies. To that end, many bioinformatic studies have been carried out on human tumor cell lines and tissue samples. Large gene expression databases have become available for the National Cancer institute's 60 human tumor cell lines (NCI-60) and for normal breast and breast cancer tissue samples from the Cancer Genome Atlas (TCGA) Research Network. We aimed to infer functional relationships among expression-correlated genes based on these data and on published information about molecular interactions. In addition, we asked how or to what extent inferences based on cell line data may apply to human tissue samples.

Among the best-defined cell phenotypes are those of epithelia. Epithelial cells have polarity that distinguishes an apical region, directed towards a surface or lumen, from a basolateral region resting on a connective tissue basement membrane. Polarity is created by transport of appropriate molecular components as cargo in vesicles that are moved by motor proteins along cytoskeletal tracks to the relevant cell regions. Epithelial polarity is maintained by molecular structures, such as tight junctions, that prevent back-flow between the apical and basolateral surfaces of the cell.

We initially defined an epithelial-like phenotype on the basis of expression of genes coding for tight junction and adherence junction proteins and their family members. Among the NCI-60, we found 11 cell lines that express a mutually correlated subset of those genes. The relevant cell lines were breast cancer MCF7 and T47D; colon cancer COLO205, HCC_2998, HCT_116, HCT_15, HT29, and KM12; lung cancer NCI_H322M; and ovary cancer OVCAR_3 and OVCAR_4; we call these the NCI-60 epithelial consensus (NEC) cell lines. The following tight junction genes were selectively expressed in the NEC cell lines: TJP3, CLDN3, CLDN4, CLDN7, OCLN, MARVELD2, and MARVELD3. Highly correlated with these genes was the adherence junction gene CDH1/E-cadherin (as expected).

We then assembled an expanded list of 75 genes that were selectively expressed in the NEC cell lines, and found that many of them function in the epithelia-specific processes summarized above.

With some notable exceptions, the tight junction and adherence junction genes co-expressed in the NEC cell lines were also co-expressed in the normal breast and breast cancer tissue samples. An unexpected finding was that some of the normal breast tissue samples expressed a different set of tight junction family genes.

This work delineates similarities and differences between epithelial-like NCI-60 cell lines and TCGA breast tissue samples in regard to the expression of functionally defined genes.

Citation Format: Kurt W. Kohn, Barry R. Zeeberg, William C. Reinhold, Ari Kahn, Yves Pommier. Molecular phenotype of an epithelial-like subset of the NCI-60 human tumor cell lines and relevance to gene expression patterns in TCGA normal breast and breast cancer tissue samples. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5219. doi:10.1158/1538-7445.AM2013-5219

Gene expression profiles of the NCI-60 human tumor cell lines define molecular interaction networks governing cell migration processes

Although there is extensive information on gene expression and molecular interactions in various cell types, integrating those data in a functionally coherent manner remains challenging. This study explores the premise that genes whose expression at the mRNA level is correlated over diverse cell lines are likely to function together in a network of molecular interactions. We previously derived expression-correlated gene clusters from the database of the NCI-60 human tumor cell lines and associated each cluster with function categories of the Gene Ontology (GO) database. From a cluster rich in genes associated with GO categories related to cell migration, we extracted 15 genes that were highly cross-correlated; prominent among them were RRAS, AXL, ADAM9, FN14, and integrin-beta1. We then used those 15 genes as bait to identify other correlated genes in the NCI-60 database. A survey of current literature disclosed, not only that many of the expression-correlated genes engaged in molecular interactions related to migration, invasion, and metastasis, but that highly cross-correlated subsets of those genes engaged in specific cell migration processes. We assembled this information in molecular interaction maps (MIMs) that depict networks governing 3 cell migration processes: degradation of extracellular matrix, production of transient focal complexes at the leading edge of the cell, and retraction of the rear part of the cell. Also depicted are interactions controlling the release and effects of calcium ions, which may regulate migration in a spaciotemporal manner in the cell. The MIMs and associated text comprise a detailed and integrated summary of what is currently known or surmised about the role of the expression cross-correlated genes in molecular networks governing those processes.

Functional categories associated with clusters of genes that are co-expressed across the NCI-60 cancer cell lines

Background

The NCI-60 is a panel of 60 diverse human cancer cell lines used by the U.S. National Cancer Institute to screen compounds for anticancer activity. In the current study, gene expression levels from five platforms were integrated to yield a single composite transcriptome profile. The comprehensive and reliable nature of that dataset allows us to study gene co-expression across cancer cell lines.

Methodology/Principal Findings

Hierarchical clustering revealed numerous clusters of genes in which the genes co-vary across the NCI-60. To determine functional categorization associated with each cluster, we used the Gene Ontology (GO) Consortium database and the GoMiner tool. GO maps genes to hierarchically-organized biological process categories. GoMiner can leverage GO to perform ontological analyses of gene expression studies, generating a list of significant functional categories.

Conclusions/Significance

GoMiner analysis revealed many clusters of coregulated genes that are associated with functional groupings of GO biological process categories. Notably, those categories arising from coherent co-expression groupings reflect cancer-related themes such as adhesion, cell migration, RNA splicing, immune response and signal transduction. Thus, these clusters demonstrate transcriptional coregulation of functionally-related genes.

RedundancyMiner: De-replication of redundant GO categories in microarray and proteomics analysis

BACKGROUND

The Gene Ontology (GO) Consortium organizes genes into hierarchical categories based on biological process, molecular function and subcellular localization. Tools such as GoMiner can leverage GO to perform ontological analysis of microarray and proteomics studies, typically generating a list of significant functional categories. Two or more of the categories are often redundant, in the sense that identical or nearly-identical sets of genes map to the categories. The redundancy might typically inflate the report of significant categories by a factor of three-fold, create an illusion of an overly long list of significant categories, and obscure the relevant biological interpretation.

RESULTS

We now introduce a new resource, RedundancyMiner, that de-replicates the redundant and nearly-redundant GO categories that had been determined by first running GoMiner. The main algorithm of RedundancyMiner, MultiClust, performs a novel form of cluster analysis in which a GO category might belong to several category clusters. Each category cluster follows a "complete linkage" paradigm. The metric is a similarity measure that captures the overlap in gene mapping between pairs of categories.

CONCLUSIONS

RedundancyMiner effectively eliminated redundancies from a set of GO categories. For illustration, we have applied it to the clarification of the results arising from two current studies: (1) assessment of the gene expression profiles obtained by laser capture microdissection (LCM) of serial cryosections of the retina at the site of final optic fissure closure in the mouse embryos at specific embryonic stages, and (2) analysis of a conceptual data set obtained by examining a list of genes deemed to be "kinetochore" genes.

Genome-wide analysis of novel splice variants induced by topoisomerase I poisoning shows preferential occurrence in genes encoding splicing factors

RNA splicing is required to remove introns from pre-mRNA, and alternative splicing generates protein diversity. Topoisomerase I (Top1) has been shown to be coupled with splicing by regulating serine/arginine-rich splicing proteins. Prior studies on isolated genes also showed that Top1 poisoning by camptothecin (CPT), which traps Top1 cleavage complexes (Top1cc), can alter RNA splicing. Here, we tested the effect of Top1 inhibition on splicing at the genome-wide level in human colon carcinoma HCT116 and breast carcinoma MCF7 cells. The RNA of HCT116 cells treated with CPT for various times was analyzed with ExonHit Human Splice Array. Unlike other exon array platforms, the ExonHit arrays include junction probes that allow the detection of splice variants with high sensitivity and specificity. We report that CPT treatment preferentially affects the splicing of splicing-related factors, such as RBM8A, and generates transcripts coding for inactive proteins lacking key functional domains. The splicing alterations induced by CPT are not observed with cisplatin or vinblastine and are not simply due to reduced Top1 activity, as Top1 downregulation by short interfering RNA did not alter splicing like CPT treatment. Inhibition of RNA polymerase II (Pol II) hyperphosphorylation by 5,6-dichloro-1-β-d-ribofuranosylbenzimidazole (DRB) blocked the splicing alteration induced by CPT, which suggests that the rapid Pol II hyperphosphorylation induced by CPT interferes with normal splicing. The preferential effect of CPT on genes encoding splicing factors may explain the abnormal splicing of a large number of genes in response to Top1cc.

Ontogenomic study of the relationship between number of gene splice variants and GO categorization

MOTIVATION

Splice variation plays important roles in evolution and cancer. Different splice variants of a gene may be characteristic of particular cellular processes, subcellular locations or organs. Although several genomic projects have identified splice variants, there have been no large-scale computational studies of the relationship between number of splice variants and biological function. The Gene Ontology (GO) and tools for leveraging GO, such as GoMiner, now make such a study feasible.

RESULTS

We partitioned genes into two groups: those with numbers of splice variants <or=b and >b (b=1,..., 10). Then we used GoMiner to determine whether any GO categories are enriched in genes with particular numbers of splice variants. Since there was no a priori 'appropriate' partition boundary, we studied those 'robust' categories whose enrichment did not depend on the selection of a particular partition boundary. Furthermore, because the distribution of splice variant number was a snapshot taken at a particular point in time, we confirmed that those observations were stable across successive builds of GenBank. A small number of categories were found for genes in the lower partitions. A larger number of categories were found for genes in the higher partitions. Those categories were largely associated with cell death and signal transduction. Apoptotic genes tended to have a large repertoire of splice variants, and genes with splice variants exhibited a distinctive 'apoptotic island' in clustered image maps (CIMs).

AVAILABILITY

Supplementary tables and figures are available at URL http://discover.nci.nih.gov/OG/supplementaryMaterials.html. The Safari browser appears to perform better than Firefox for these particular items.

SpliceCenter: a suite of web-based bioinformatic applications for evaluating the impact of alternative splicing on RT-PCR, RNAi, microarray, and peptide-based studies

Background

Over 60% of protein-coding genes in vertebrates express mRNAs that undergo alternative splicing. The resulting collection of transcript isoforms poses significant challenges for contemporary biological assays. For example, RT-PCR validation of gene expression microarray results may be unsuccessful if the two technologies target different splice variants. Effective use of sequence-based technologies requires knowledge of the specific splice variant(s) that are targeted. In addition, the critical roles of alternative splice forms in biological function and in disease suggest that assay results may be more informative if analyzed in the context of the targeted splice variant.

Results

A number of contemporary technologies are used for analyzing transcripts or proteins. To enable investigation of the impact of splice variation on the interpretation of data derived from those technologies, we have developed SpliceCenter. SpliceCenter is a suite of user-friendly, web-based applications that includes programs for analysis of RT-PCR primer/probe sets, effectors of RNAi, microarrays, and protein-targeting technologies. Both interactive and high-throughput implementations of the tools are provided. The interactive versions of SpliceCenter tools provide visualizations of a gene's alternative transcripts and probe target positions, enabling the user to identify which splice variants are or are not targeted. The high-throughput batch versions accept user query files and provide results in tabular form. When, for example, we used SpliceCenter's batch siRNA-Check to process the Cancer Genome Anatomy Project's large-scale shRNA library, we found that only 59% of the 50,766 shRNAs in the library target all known splice variants of the target gene, 32% target some but not all, and 9% do not target any currently annotated transcript.

Conclusion

SpliceCenter provides unique, user-friendly applications for assessing the impact of transcript variation on the design and interpretation of RT-PCR, RNAi, gene expression microarrays, antibody-based detection, and mass spectrometry proteomics. The tools are intended for use by bench biologists as well as bioinformaticists.

The EDGE hypothesis: epigenetically directed genetic errors in repeat-containing proteins (RCPs) involved in evolution, neuroendocrine signaling, and cancer

Trans-generational epigenetic phenomena, such as contamination with endocrine-disrupting chemicals (EDCs) that decrease fertility and the global methylation status of DNA in the offspring, are of great concern because they may affect health, particularly the health of children. However, of even greater concern is the possibility that trans-generational changes in the methylation status of the DNA might lead to permanent changes in the DNA sequence itself. By contaminating the environment with EDCs, mankind might be permanently affecting the health of future generations. In this section, we present evidence from our laboratory and others that trans-generational epigenetic changes in DNA might lead to mutations directed to genes encoding amino acid repeat-containing proteins (RCPs) that are important for adaptive evolution or cancer progression. Such epigenetic changes can be induced "naturally" by hormones or "unnaturally" by EDCs or environmental stress. To illustrate the phenomenon, we present new bioinformatic evidence that the only RCP ontological categories conserved from Drosophila to humans are "regulation of splicing," "regulation of transcription," and "regulation of synaptogenesis," which are classes of genes likely to be important for evolutionary processes. Based on that and other evidence, we propose a model for evolution that we call the EDGE (Epigenetically Directed Genetic Errors) hypothesis for the mechanism by which mutations are targeted at epigenetically modified "contingency genes" encoding RCPs. In the model, "epigenetic assimilation" of metastable epialleles of RCPs over many generations can lead to mutations directed to those genes, thereby permanently stabilizing the adaptive phenotype.

VennMaster: area-proportional Euler diagrams for functional GO analysis of microarrays

Background

Microarray experiments generate vast amounts of data. The functional context of differentially expressed genes can be assessed by querying the Gene Ontology (GO) database via GoMiner. Directed acyclic graph representations, which are used to depict GO categories enriched with differentially expressed genes, are difficult to interpret and, depending on the particular analysis, may not be well suited for formulating new hypotheses. Additional graphical methods are therefore needed to augment the GO graphical representation.

Results

We present an alternative visualization approach, area-proportional Euler diagrams, showing set relationships with semi-quantitative size information in a single diagram to support biological hypothesis formulation. The cardinalities of sets and intersection sets are represented by area-proportional Euler diagrams and their corresponding graphical (circular or polygonal) intersection areas. Optimally proportional representations are obtained using swarm and evolutionary optimization algorithms.

Conclusion

VennMaster's area-proportional Euler diagrams effectively structure and visualize the results of a GO analysis by indicating to what extent flagged genes are shared by different categories. In addition to reducing the complexity of the output, the visualizations facilitate generation of novel hypotheses from the analysis of seemingly unrelated categories that share differentially expressed genes.

AffyProbeMiner: a web resource for computing or retrieving accurately redefined Affymetrix probe sets

MOTIVATION

Affymetrix microarrays are widely used to measure global expression of mRNA transcripts. That technology is based on the concept of a probe set. Individual probes within a probe set were originally designated by Affymetrix to hybridize with the same unique mRNA transcript. Because of increasing accuracy in knowledge of genomic sequences, however, a substantial number of the manufacturer's original probe groupings and mappings are now known to be inaccurate and must be corrected. Otherwise, analysis and interpretation of an Affymetrix microarray experiment will be in error.

RESULTS

AffyProbeMiner is a computationally efficient platform-independent tool that uses all RefSeq mature RNA protein coding transcripts and validated complete coding sequences in GenBank to (1) regroup the individual probes into consistent probe sets and (2) remap the probe sets to the correct sets of mRNA transcripts. The individual probes are grouped into probe sets that are 'transcript-consistent' in that they hybridize to the same mRNA transcript (or transcripts) and, therefore, measure the same entity (or entities). About 65.6% of the probe sets on the HG-U133A chip were affected by the remapping. Pre-computed regrouped and remapped probe sets for many Affymetrix microarrays are made freely available at the AffyProbeMiner web site. Alternatively, we provide a web service that enables the user to perform the remapping for any type of short-oligo commercial or custom array that has an Affymetrix-format Chip Definition File (CDF). Important features that differentiate AffyProbeMiner from other approaches are flexibility in the handling of splice variants, computational efficiency, extensibility, customizability and user-friendliness of the interface.

AVAILABILITY

The web interface and software (GPL open source license), are publicly-accessible at http://discover.nci.nih.gov/affyprobeminer.

Sequencing and analysis of 10,967 full-length cDNA clones from Xenopus laevis and Xenopus tropicalis reveals post-tetraploidization transcriptome remodeling

Sequencing of full-insert clones from full-length cDNA libraries from both Xenopus laevis and Xenopus tropicalis has been ongoing as part of the Xenopus Gene Collection Initiative. Here we present 10,967 full ORF verified cDNA clones (8049 from X. laevis and 2918 from X. tropicalis) as a community resource. Because the genome of X. laevis, but not X. tropicalis, has undergone allotetraploidization, comparison of coding sequences from these two clawed (pipid) frogs provides a unique angle for exploring the molecular evolution of duplicate genes. Within our clone set, we have identified 445 gene trios, each comprised of an allotetraploidization-derived X. laevis gene pair and their shared X. tropicalis ortholog. Pairwise dN/dS, comparisons within trios show strong evidence for purifying selection acting on all three members. However, dN/dS ratios between X. laevis gene pairs are elevated relative to their X. tropicalis ortholog. This difference is highly significant and indicates an overall relaxation of selective pressures on duplicated gene pairs. We have found that the paralogs that have been lost since the tetraploidization event are enriched for several molecular functions, but have found no such enrichment in the extant paralogs. Approximately 14% of the paralogous pairs analyzed here also show differential expression indicative of subfunctionalization.

The p53 tumor suppressor network is a key responder to microenvironmental components of chronic inflammatory stress

Activation of the p53 network plays a central role in the inflammatory stress response associated with ulcerative colitis and may modulate cancer risk in patients afflicted with this chronic disease. Here, we describe the gene expression profiles associated with four microenvironmental components of the inflammatory response (NO*, H2O2, DNA replication arrest, and hypoxia) that result in p53 stabilization and activation. Isogenic HCT116 and HCT116 TP53-/- colon cancer cells were exposed to the NO* donor Sper/NO, H2O2, hypoxia, or hydroxyurea, and their mRNA was analyzed using oligonucleotide microarrays. Overall, 1,396 genes changed in a p53-dependent manner (P < 0.001), with the majority representing a "unique" profile for each condition. Only 14 genes were common to all four conditions. Included were eight known p53 target genes. Hierarchical sample clustering distinguished early (1 and 4 hours) from late responses (8, 12, and 24 hours), and each treatment was differentiated from the others. Overall, NO* and hypoxia stimulated similar transcriptional responses. Gene ontology analysis revealed cell cycle as a key feature of stress responses and confirmed the similarity between NO* and hypoxia. Cell cycle profiles analyzed by flow cytometry showed that NO* and hypoxia induced quiescent S-phase and G2-M arrest. Using a novel bioinformatic algorithm, we identified several putative p53-responsive elements among the genes induced in a p53-dependent manner, including four [KIAA0247, FLJ12484, p53CSV (HSPC132), and CNK (PLK3)] common to all exposures. In summary, the inflammatory stress response is a complex, integrated biological network in which p53 is a key molecular node regulating gene expression.

Nova regulates brain-specific splicing to shape the synapse

Alternative RNA splicing greatly increases proteome diversity and may thereby contribute to tissue-specific functions. We carried out genome-wide quantitative analysis of alternative splicing using a custom Affymetrix microarray to assess the role of the neuronal splicing factor Nova in the brain. We used a stringent algorithm to identify 591 exons that were differentially spliced in the brain relative to immune tissues, and 6.6% of these showed major splicing defects in the neocortex of Nova2-/- mice. We tested 49 exons with the largest predicted Nova-dependent splicing changes and validated all 49 by RT-PCR. We analyzed the encoded proteins and found that all those with defined brain functions acted in the synapse (34 of 40, including neurotransmitter receptors, cation channels, adhesion and scaffold proteins) or in axon guidance (8 of 40). Moreover, of the 35 proteins with known interaction partners, 74% (26) interact with each other. Validating a large set of Nova RNA targets has led us to identify a multi-tiered network in which Nova regulates the exon content of RNAs encoding proteins that interact in the synapse.

High-Throughput GoMiner, an 'industrial-strength' integrative gene ontology tool for interpretation of multiple-microarray experiments, with application to studies of Common Variable Immune Deficiency (CVID)

BACKGROUND

We previously developed GoMiner, an application that organizes lists of 'interesting' genes (for example, under-and overexpressed genes from a microarray experiment) for biological interpretation in the context of the Gene Ontology. The original version of GoMiner was oriented toward visualization and interpretation of the results from a single microarray (or other high-throughput experimental platform), using a graphical user interface. Although that version can be used to examine the results from a number of microarrays one at a time, that is a rather tedious task, and original GoMiner includes no apparatus for obtaining a global picture of results from an experiment that consists of multiple microarrays. We wanted to provide a computational resource that automates the analysis of multiple microarrays and then integrates the results across all of them in useful exportable output files and visualizations.

RESULTS

We now introduce a new tool, High-Throughput GoMiner, that has those capabilities and a number of others: It (i) efficiently performs the computationally-intensive task of automated batch processing of an arbitrary number of microarrays, (ii) produces a human-or computer-readable report that rank-orders the multiple microarray results according to the number of significant GO categories, (iii) integrates the multiple microarray results by providing organized, global clustered image map visualizations of the relationships of significant GO categories, (iv) provides a fast form of 'false discovery rate' multiple comparisons calculation, and (v) provides annotations and visualizations for relating transcription factor binding sites to genes and GO categories.

CONCLUSION

High-Throughput GoMiner achieves the desired goal of providing a computational resource that automates the analysis of multiple microarrays and integrates results across all of the microarrays. For illustration, we show an application of this new tool to the interpretation of altered gene expression patterns in Common Variable Immune Deficiency (CVID). High-Throughput GoMiner will be useful in a wide range of applications, including the study of time-courses, evaluation of multiple drug treatments, comparison of multiple gene knock-outs or knock-downs, and screening of large numbers of chemical derivatives generated from a promising lead compound.

Analysis of ATP-Binding Cassette Transporter Expression in Drug-Selected Cell Lines by a Microarray Dedicated to Multidrug Resistance

Discovery of the multidrug resistance protein 1 (MDR1), an ATP-binding cassette (ABC) transporter able to transport many anticancer drugs, was a clinically relevant breakthrough in multidrug resistance research. Although the overexpression of ABC transporters such as P-glycoprotein/ABCB1, MRP1/ABCC1, and MXR/ABCG2 seems to be a major cause of failure in the treatment of cancer, acquired resistance to multiple anticancer drugs may also be multifactorial, involving alteration of detoxification processes, apoptosis, DNA repair, drug uptake, and overexpression of other ABC transporters. As a tool for the study of such phenomena, we designed and created a microarray platform, the ABC-ToxChip, to evaluate relative levels of transcriptional activation among genes involved in the various mechanisms of resistance. In the ABC-ToxChip, a comprehensive set of genes important in toxicological responses (represented by 2200 cDNA probes) is complemented with probes specifically matching ABC transporters as well as oligonucleotides representing 18,000 unique human genes. By comparing the transcriptional profiles of KB-3-1 and DU-145 parental cells with resistant derivatives selected in colchicine (KB-8-5), and 9-nitro-camptothecin (RCO.1), respectively, we demonstrate that ABC transporters (ABCB1/MDR1 and ABCC2/MRP2, respectively) show dramatic overexpression, whereas the glutathione S-transferase gene GST-Pi shows the strongest decrease in expression among the 20,000 genes studied. The results were confirmed by quantitative reverse transcription-polymerase chain reaction and immunohistochemistry. The custom-designed ABC-Tox microarray presented here will be helpful to elucidate mechanisms leading to anticancer drug resistance.

Mistaken identifiers: gene name errors can be introduced inadvertently when using Excel in bioinformatics

Background

When processing microarray data sets, we recently noticed that some gene names were being changed inadvertently to non-gene names.

Results

A little detective work traced the problem to default date format conversions and floating-point format conversions in the very useful Excel program package. The date conversions affect at least 30 gene names; the floating-point conversions affect at least 2,000 if Riken identifiers are included. These conversions are irreversible; the original gene names cannot be recovered.

Conclusions

Users of Excel for analyses involving gene names should be aware of this problem, which can cause genes, including medically important ones, to be lost from view and which has contaminated even carefully curated public databases. We provide work-arounds and scripts for circumventing the problem.

Development of gene ontology tool for biological interpretation of genomic and proteomic data

We have designed and developed a Gene Ontology based navigation tool, GoMiner, which organizes lists of interesting genes from a microarray or a protein array experiment for biological interpretation. It provides quantitative and statistical output files and useful visualization (e.g., a tree-like structure) to map the list of genes to its biological functional categories. It also provides links to other resources such as pubmed, locuslink, and biological molecular interaction map and signaling pathway packages.

GoMiner: a resource for biological interpretation of genomic and proteomic data

We have developed GoMiner, a program package that organizes lists of 'interesting' genes (for example, under- and overexpressed genes from a microarray experiment) for biological interpretation in the context of the Gene Ontology. GoMiner provides quantitative and statistical output files and two useful visualizations. The first is a tree-like structure analogous to that in the AmiGO browser and the second is a compact, dynamically interactive 'directed acyclic graph'. Genes displayed in GoMiner are linked to major public bioinformatics resources.

In vivo autoradiographic competition studies of isomers of [125I]IQNP against QNB demonstrating in vivo m2 muscarinic subtype selectivity for QNB

(R,S)-[125I]IQNB has been used extensively in in vivo studies in rats, and has been of utility in demonstrating the in vivo subtype selectivity of nonradioactive ligands in competition studies. Because of the implications for the study of Alzheimer's disease (AD), those ligands that demonstrate m2 selectivity are of particular interest. Radiolabelled Z- and E-(-,-)-1-azabicyclo[2.2.2]oct-3-yl alpha-hydroxy-alpha-(1-iodo-1-propen-3-yl)-alpha-phenylacetate (Z- and E-(-,-)-[125I]IQNP) are analogs of (R,S)-[125I]IQNB. Rat brain regional dissection studies and in vivo autoradiographic comparison of the time-courses of (R,S)-[125I]IQNB, Z-(-,-)-[125I]IQNP, and E-(-,-)-[125I]IQNP have indicated that Z- and E-(-,-)-[125I]IQNP, in general, are distributed similarly to (R,S)-[125I]IQNB. Z-(-,-)-[125I]IQNP binds to the muscarinic receptors in those brain regions enriched in the m2 subtype with approximately a two- to fivefold higher % dose/g compared with (R,S)-[125I]IQNB. Thus, as we show here autoradiographically, using QNB as the competing nonradioactive ligand in in vivo competition studies against Z-(-,-)-[125I]IQNP provides a sensitive and accurate probe for demonstrating the in vivo m2 selectivity of nonradioactive ligands.

In vitro and in vivo m2 muscarinic subtype selectivity of some dibenzodiazepinones and pyridobenzodiazepinones

Alzheimer's disease (AD) involves selective loss of muscarinic m2, but not m1, subtype receptors in cortical and hippocampal regions of the human brain. Emission tomographic study of the loss of m2 receptors in AD has been limited by the absence of available m2-selective radioligands, which can penetrate the blood-brain barrier. We now report on the in vitro and in vivo m2 muscarinic subtype selectivity of a series of dibenzodiazepinones and pyridobenzodiazepinones determined by competition studies against (R)-3-quinuclidinyl (S)-4-iodobenzilate ((R,S)-[125I]IQNB) or [3H]QNB. Of the compounds examined, three of the 5-[[4-[(4-dialkylamino)butyl]-1-piperidinyl]acetyl]-10, 11-dihydro-5-H-dibenzo[b,e][1,4]diazepin-11-ones (including DIBA) and three of the 11-[[4-[4-(dialkylamino)butyl]-1-phenyl]acetyl]-5, 11-dihydro-6H-pyrido [2,3-b][1,4]benzodiazepin-6-ones (including PBID) exhibited both high binding affinity for the m2 subtype (</=5 nM) and high m2/m1 selectivity (>/=10). In vivo rat brain dissection studies of the competition of PBID or DIBD against (R,S)[125I]IQNB or [3H]QNB exhibited a dose-dependent preferential decrease in the binding of the radiotracer in brain regions that are enriched in the m2 muscarinic subtype. In vivo rat brain autoradiographic studies of the competition of PBID, BIBN 99, or DIBD against (R,S)[125I]IQNB exhibited an insignificant effect of BIBN 99 and confirmed the effect of PBID and DIBD in decreasing the binding of (R,S)[125I]IQNB in brain regions that are enriched in the m2 muscarinic subtype. We conclude that PBID and DIBD are potentially useful parent compounds from which in vivo m2 selective derivatives may be prepared for potential use in positron emission tomographic (PET) study of the loss of m2 receptors in AD.

Pharmacokinetic computer simulations of the relationship between in vivo and in vitro neuroreceptor subtype selectivity of radioligands

Pharmacokinetic computer simulations reveal a discrepancy between the in vivo and in vitro neuroreceptor subtype selectivity of radioligands. For radioligands with an in vitro neuroreceptor subtype selectivity between 0.1 and 10.0, the in vivo neuroreceptor subtype selectivity appears to be constrained to be between 0.1 and 10.0, but, in general, is not equal to the in vitro selectivity. For example, if the in vitro selectivity is 1.0 (that is, the radioligand is nonselective in vitro) the in vivo selectivity may be thought of as a random variable having a significant nonzero probability for values as low as 0.1 or as high as 10.0, with a moderate peak at a value of 1.0. For a radioligand whose in vitro subtype selectivity is greater than 10.0, the in vivo selectivity is bounded above by the in vitro subtype selectivity, but may be several orders of magnitude lower than the in vitro subtype selectivity. Thus, in spite of the discrepancy between the in vivo and in vitro neuroreceptor subtype selectivity of radioligands, there are two useful inferences about the in vivo selectivity that might be drawn from knowledge of the in vitro selectivity: (1) If the in vitro selectivity is between 0.1 and 10.0, then, at best, the in vivo selectivity might be as high as 10.0. (2) If the in vitro selectivity is greater than 10.0, then, at best, the in vivo selectivity might be as high as the in vitro selectivity.

In vivo competition studies of Z-(-,-)-[125I]IQNP against 3-quinuclidinyl 2-(5-bromothienyl)-2-thienylglycolate (BrQNT) demonstrating in vivo m2 muscarinic subtype selectivity for BrQNT

Alzheimer's disease (AD) involves selective loss of muscarinic m2, but not m1, subtype neuroreceptors in cortical and hippocampal regions of the human brain. Until recently, emission tomographic study of the loss of m2 receptors in AD has been limited by the absence of available m2-selective radioligands that can penetrate the blood-brain barrier. We now demonstrate the in vivo m2 selectivity of an analog of (R)-QNB, 3-quinuclidinyl 2-(5-bromothienyl)-2-thienylglycolate (BrQNT), by dissection and autoradiographic studies of the in vivo inhibition of radioiodinated Z-1-azabicyclo[2.2.2]oct-3-yl alpha-hydroxy-alpha-(1-iodo-1-propen-3-yl)-alpha-phenyl-acetate (Z-(-,-)-[125I]IQNP) binding by unlabeled BrQNT in rat brain. In the absence of BrQNT, Z-(-,-)-[125I]IQNP labels brain regions containing muscarinic receptors, with an enhanced selectivity for the m2 subtype. In the presence of 60-180 nmol of co-injected racemic BrQNT, Z-(-,-)-[125I]IQNP labeling in those brain regions containing predominantly m2 subtype is reduced to background levels, while levels of radioactivity in areas not enriched in the m2 subtype do not significantly decrease. We conclude that BrQNT is m2-selective in vivo, and that [76Br]BrQNT, or a radiofluorinated analog, may be of potential use in positron emission tomographic (PET) study of the loss of m2 receptors in AD. In addition, a radioiodinated analog may be of potential use in single photon emission tomographic (SPECT) studies.

Correction of the stereochemical assignment of the benzilic acid center in (R)-(-)-3-quinuclidinyl (S)-(+)-4-iodobenzilate [(R,S)-4-IQNB]

Radioiodinated (R)-quinuclidinyl-4-iodobenzilate (4IQNB) is a high affinity muscarinic antagonist which has been utilized for in vitro and in vivo assays, and for SPECT imaging in humans. 4IQNB exists in four different diastereomeric forms, since there are two asymmetric centers at the quinuclidinyl and benzilic acid centers. Based upon our in vivo studies, we have determined that the absolute stereochemistry previously assigned to the benzilic center was incorrect for the diastereomer that had been previously referred to as '(R)-quinuclidinyl-(R)-4-iodobenzilate' [(R,R)-4IQNB]. The correct designation for this diastereomer is '(R)-quinuclidinyl-(S)-4-iodobenzilate' [(R,S)-4IQNB].

In vivo autoradiography of radioiodinated (R)-3-quinuclidinyl (S)-4-iodobenzilate [(R, S)-IQNB] and (R)-3-quinuclidinyl (R)-4-iodobenzilate [(R,R)-IQNB]. Comparison of the radiolabelled products of a novel tributylstannyl precursor with those of the established triazene and exchange methods

Radioiodinated (R,S)-IQNB and (R,R)-IQNB are prepared either from a triazene precursor or using an exchange reaction. In both cases the radiochemical yield is low. The product of the exchange reaction also suffers from having a fairly low specific activity. A new method for preparing radioiodinated (R,S)-IQNB and (R,R)-IQNB from a tributylstannyl precursor has recently been developed. This method is more convenient and much faster than the triazene and exchange methods, and it reliably results in a high radiochemical yield of a high specific activity product. In rat brain, the in vivo properties of the radioiodinated products of the tributylstannyl method are identical to those of the corresponding radioiodinated (R,S)-IQNB and (R,R)-IQNB prepared using the triazene and exchange methods. Dissection studies of selected brain regions show that at 3 h post injection (R,S)-[125I]IQNB prepared by all three methods have indistinguishable % dose g-1 values in all brain regions studied. Autoradiographic comparison of coronal slices through the anteroventral nucleus of the thalamus, through the hippocampus and through the pons at 2 h post injection shows that (R,S)-[125I]IQNB prepared by the triazene and tributylstannyl methods have indistinguishable patterns of binding.

Autoradiographic evidence that (R)-3-quinuclidinyl (S)-4-fluoromethylbenzilate ((R,S)-FMeQNB) displays in vivo selectivity for the muscarinic m2 subtype

Alzheimer's disease (AD) involves selective loss of muscarinic m2, but not m1, subtype neuroreceptors in cortical and hippocampal regions of the human brain. Until recently, emission tomographic study of the loss of m2 receptors in AD has been limited by the absence of available m2-selective radioligands that can penetrate the blood-brain barrier. We now demonstrate the in vivo m2 selectivity of a fluorinated derivative of QNB, (R)-3-quinuclidinyl (S)-4-fluoromethylbenzilate ((R,S)-FMeQNB), by studying autoradiographically the in vivo inhibition of radioiodinated (R)-3-quinuclidinyl (S)-4-iodobenzilate ((R,S)-[125I]IQNB) binding by unlabelled (R,S)-FMeQNB. In the absence of (R,S)-FMeQNB, (R,S)-[125I]IQNB labels brain regions in proportion to the total muscarinic receptor concentration; in the presence of 75 nmol of (R,S)-FMeQNB, (R,S)-[125I]IQNB labelling in those brain regions containing predominantly m2 subtype is reduced to background levels. We conclude that (R,S)-FMeQNB is m2-selective in vivo, and that (R,S)-[18F]FMeQNB may be of potential use in positron emission tomographic (PET) study of the loss of m2 receptors in AD.

Specific binding component of the "inactive" stereoisomer (S,S)-[125I] IQNB to rat brain muscarinic receptors in vivo

In vivo nonspecific binding can be estimated using the inactive stereoisomer of a receptor radioligand. However, the binding of the inactive stereoisomer may be partially specific. Specific binding of the inactive (S,S)-[125I]IQNB was estimated from the inhibition induced by a competing nonradioactive ligand. This technique differed from the usual approach, since it was used to study the inactive rather than the active stereoisomer. The results indicate that there is substantial specific binding for (S,S)-[125I]IQNB.

Autoradiographic evidence that 3-quinuclidinyl-4-fluorobenzilate (FQNB) displays in vivo selectivity for the m2 subtype

Alzheimer's disease (AD) involves selective loss of muscarinic m2, but not m1, subtype neuroreceptors in cortical and hippocampal regions of the human brain. Emission tomographic study of the loss of m2 receptors in AD is limited by the fact that there is currently no available m2-selective radioligand which can penetrate the blood-brain barrier. We now demonstrate the in vivo m2 selectivity of a fluorine derivative of QNB (FQNB), by studying autoradiographically the in vivo inhibition of radioiodinated (R)-3-quinuclidinyl (S)-4-iodobenzilate ((R,S)-[125I]IQNB) binding by unlabeled FQNB. In the absence of FQNB, (R,S)-[125I]IQNB labels brain regions in proportion to the total muscarinic receptor concentration; in the presence of 30.0 nmol of racemic FQNB, (R,S)-[125I]IQNB labeling in those brain regions containing predominantly the m2 subtype is reduced to background levels. We conclude that FQNB is m2-selective in vivo and that [18F]FQNB or a closely related analogue may be of potential use in positron emission tomographic study of the loss of m2 receptors in AD.

Autoradiographic evidence that quinuclidinyl 4-(bromophenyl)-2-thienylglycolate (QBPTG) displays in vivo selectivity for the muscarinic m2 subtype

Alzheimer's disease (AD) involves selective loss of muscarinic m2, but not m1, subtype neuroreceptors in cortical and hippocampal regions of the human brain. Emission tomographic study of the loss of m2 receptors in AD is limited by the fact that there is currently no available m2-selective radioligand which can penetrate the blood-brain barrier. We now demonstrate the in vivo m2 selectivity of an analogue of QNB, 4-(bromophenyl)-2-thienylglycolate (QBPTG), by studying autoradiographically the in vivo inhibition of radioiodinated (R)-3-quinuclidinyl (S)-4-iodobenzilate ((R,S)-[125I]IQNB) binding by unlabeled QBPTG in rat brain. In the absence of QBPTG, (R,S)-[125I]IQNB labels brain regions in proportion to the total muscarinic receptor concentration; in the presence of 37.5 nmol of racemic QBPTG, (R,S)-[125I]IQNB labeling in those brain regions containing predominantly the m2 subtype is reduced to background levels. We conclude that QBPTG is m2-selective in vivo and that [76Br]QBPTG, or a radiofluorinated analogue, may be of potential use in positron emission tomographic study of the loss of m2 receptors in AD. In addition, a radioiodinated analogue may be of potential use in single photon emission tomographic studies.

Characterization of in vivo brain muscarinic acetylcholine receptor subtype selectivity by competition studies against (R,S)-[125I]IQNB

We have studied the in vivo rat brain muscarinic acetylcholine receptor (mAChR) m2 subtype selectivities of three quinuclidine derivatives: (R)-3-quinuclidinyl benzilate (QNB), E-(+,+)-1-azabicyclo[2.2.2]oct-3-yl alpha-hydroxy-alpha-(1-iodo-1-propen-3-yl)-alpha-phenylacetate (E-(+,+)-IQNP), and E-(+,-)-1-azabicyclo[2.2.2]oct-3-yl alpha-hydroxy-alpha-(1-iodo-1-propen-3-yl)-alpha-phenylacetate (E-(+,-)-IQNP), and two tricyclic ring compounds: 5-[[4-[4-(diisobutylamino)butyl]-1-phenyl]-10,11-dihydro-5H-dibenz o [b,e][1,4]diazepin-11-one [sequence: see text] (DIBD) and 11-[[4-[4-(diisobutylamino)butyl-1-phenyl]acetyl]-5,11-dihydro-6H- pyrido [2,3-b][1,4]benzodiazepin-6-one [sequence: see text] (PBID), by correlating the regional inhibition of (R,S)-[125I]IQNB with the regional composition of the m1-m4 subtypes. Subtle effects are demonstrated after reduction of the between-animal variability by normalization to corpus striatum. Substantial in vivo m2-selectivity is exhibited by QNB and DIBD, modest in vivo m2-selectivity is exhibited by E-(+,+)-IQNP, and little or no in vivo m2-selectivity is exhibited by PBID and E-(+,-)-IQNP. Surprisingly, the in vivo m2-selectivity is not correlated with the in vitro m2-selectivity. For example, QNB, which appears to be the most strongly in vivo m2-selective compound, exhibits negligible in vitro m2-selectivity. These examples indicate that a strategy which includes only preliminary in vitro screening may very well preclude the discovery of a novel compound which would prove useful in vivo.

Autoradiographic evidence that QNB displays in vivo selectivity for the m2 subtype

Alzheimer's disease (AD) involves selective loss of muscarinic m2, but not m1, subtype neuroreceptors in cortical and hippocampal regions of the human brain. Emission tomographic study of the loss of m2 receptors in AD is limited by the fact that there is currently no available m2-selective radioligand which can penetrate the blood-brain barrier. We have previously reported the results of in vivo dissection studies, using both carrier-free and low specific activity [3H]QNB, which show that [3H]QNB exhibits a substantial in vivo m2 selectivity. Because of the expense of the radioligand and the long exposure time required for the X-ray film, performing a large number of direct in vivo autoradiographic studies using [3H]QNB is precluded. Therefore, we now confirm these results autoradiographically by studying the in vivo inhibition of radio-iodinated (R)-3-quinuclidinyl (S)-4-iodobenzilate ((R,S)-[125I]IQNB) binding by unlabeled QNB. In the absence of QNB, (R,S)-[125I]IQNB labels brain regions in proportion to the total muscarinic receptor concentration; in the presence of 15 nmol QNB, (R,S,)-[125I]IQNB labeling in those brain regions containing predominantly m2 subtype is reduced to background levels. We conclude that QNB is m2-selective in vivo and that a suitably radiolabeled derivative of QNB, possibly labeled with 18F, may be of potential use in positron emission tomographic study of the loss of m2 receptors in AD.

Theoretical relationships of receptor and delivery sensitivities and measurable parameters in in vivo neuroreceptor-radioligand interactions

In vivo quantification of neuroreceptors in human brains by PET or SPECT is complicated by the fact that a number of variables other than receptor concentration may influence the observed radioactivity in a brain region. This consideration has led the authors to formulate rigorous mathematical definitions of the concepts of receptor and delivery sensitivities. It has been speculated that a neuroreceptor-radioligand system having a high (low) receptor sensitivity would have a low (high) delivery sensitivity, and that the receptor sensitivity of a neuroreceptor-radioligand system can be determined by observing the time-course of the brain radioligand concentration following injection of no carrier added (nca) radioligand. Computer simulation studies of the characteristics of a simple model for in vivo neuroreceptor-radioligand interaction show that, under a set of realistic restrictions, there is a unique and intuitively satisfying relationship between receptor and delivery sensitivities: receptor sensitivity+delivery sensitivity approximately 1. In addition, the receptor sensitivity can be computed as a function of the observable parameters of the nca radioligand time course. These straightforward relationships are surprising in light of the complexity of the analytical solutions.

[3H]QNB displays in vivo selectivity for the m2 subtype

Alzheimer's disease (AD) involves selective loss of muscarinic m2, but not m1, subtype neuroreceptors in the posterior parietal cortex of the human brain. Emission tomographic study of the loss of m2 receptors in AD is limited by the fact that there is currently no available m2-selective radioligand which can penetrate the blood-brain barrier. [3H](R)-3-quinuclidinylbenzilate ([3H]QNB) is commonly used for performing in vitro studies of the muscarinic acetylcholine receptor (mAChR), either with membrane homogenates or with autoradiographic slices, in which [3H]QNB is nonsubtype-selective. We report here the results of in vivo studies, using both carrier-free and low specific activity [3H]QNB, which show that [3H]QNB exhibits a substantial in vivo m2-selectivity. Previously reported in vivo (R)-3-quinuclidinyl (R)-4-iodobenzilate ((R,R)-[125I]IQNB) binding appears to be nonsubtype-selective. Apparently the bulky iodine substitution in the 4 position reduces the subtype selectivity of QNB. It is possible that a less bulky fluorine substitution might permit retention of the selectivity exhibited by QNB itself. We conclude that a suitably radiolabeled derivative of QNB, possibly labeled with 18F, may be of potential use in positron emission tomographic (PET) study of the loss of m2 receptors in AD.

Evaluation of reconstruction algorithms in SPECT neuroimaging: II. Computation of deterministic and statistical error components

For the reconstruction of a series of computer simulations of statistically-independent noisy realizations of projection data, the total error of the ith reconstructed voxel in the rth realization, Er,i, is composed of the statistical error, Sr,i, and the (deterministic) inaccuracy in the presence of noise, Di+. Di+ is composed of the (deterministic) inaccuracy in the absence of noise, Di-, and the (deterministic) additional inaccuracy in the presence of noise, Di delta. E(Er,i), the theoretical expected value of Er,i, is given by E(Er,i) = E(Di+) + E(Sr,i). Similarly, E(Di+) = E(Di-) + E(Di delta). The corresponding theoretical variances are given by sigma 2(Er,i) = sigma 2(Di+)+2C(Di+, Sr,i)+ sigma 2(Sr,i) and sigma 2(Di+) = sigma 2(Di-)+2C(Di-, Di delta)+ sigma 2(Di delta), where C(.,.) is the covariance. We have utilized these relationships to evaluate three reconstruction algorithms: standard filtered back projection (FBP), an iterative reconstruction algorithm (IRA), and a version of the IRA which incorporates a linear transformation (TIRA). For simulated brain images in which the projection data (500,000 events detected) were degraded as the result of convolution of the true radioactivity distribution with a realistic distance-dependent detector response function, for FBP the major contribution to both E(Er,i) and sigma 2(Er,i) was Di-. For the IRA and TIRA, the major contributions to E(Er,i) were Di- and Di delta, and the major contribution to sigma 2(Er,i) was Sr,i, although in some cases Di delta was also a contributing factor. Furthermore, the errors due to sigma 2(Er,i) (that is, [sigma 2(Er,i)]0.5) were more severe than those due to E(Er,i). We conclude that, in contrast to FBP, the effects of statistical noise are an important limiting factor for the IRA and TIRA, and that the future development of tomographic devices with higher sensitivity would expand the quantitative potential of the IRA and TIRA.

Evaluation of reconstruction algorithms in SPECT neuroimaging: I. Comparison of statistical noise in SPECT neuroimages with 'naive' and 'realistic' predictions

In the presence of statistical noise, an iterative reconstruction algorithm (IRA) for the quantitative reconstruction of single-photon-emission computed tomographic (SPECT) brain images overcomes major limitations of applying the standard filtered back projection (FBP) reconstruction algorithm to projection data which have been degraded by convolution of the true radioactivity distribution with a finite-resolution distance-dependent detector response: (a) the non-uniformity within the grey (or white) matter voxels which results even though the true model is uniform within these voxels; (b) a significantly lower ratio of grey/white matter voxel values than in the true model; and (c) an inability to detect an altered radioactivity value within the grey (or white) matter voxels. It is normally expected that an algorithm which improves spatial resolution and quantitative accuracy might also increase the magnitude of the statistical noise in the reconstructed image. However, the noise properties in the IRA images are very similar to those in the FBP images. In fact, the noise magnitude in both the FBP and IRA images is only slightly greater than that computed by the 'naive prediction', which presumably is a lower limit to the amount of statistical noise in a reconstructed image. Thus, the IRA should provide the potential for quantitative SPECT imaging of normal physiological responses or diseases involving both the brain grey and white matter.

Estimation of relative regional neuroreceptor concentration by PET or SPECT: Theoretical comparisons of using a single late image or a late plus early image

The potential for using a single (SPECT) single-photon-emission computed tomography or (PET) positron emission tomography image to estimate quantitatively the relative regional neuroreceptor concentration depends on acquiring the image at a time when changes in the regional radioactivity localization are much more sensitive to changes in receptor concentration than to changes in delivery. Using the binding of [(11)C]carfentanil to the opiate receptor as a model, the authors have applied a computer simulation approach to determine the joint and marginal probability distributions for the ipsilateral/contralateral ratio of receptor concentrations and delivery. They have found that the probability distributions depend on the sensitivities for both delivery and receptor. Incorporation of data at an early time point results in a significant sharpening of the probability distributions. There is an insignificant effect of subtraction of the radioactivity localization in a control region.

A novel muscarinic receptor ligand which penetrates the blood brain barrier and displays in vivo selectivity for the m2 subtype

Alzheimer's disease (AD) involves selective loss of muscarinic m2, but not m1, subtype neuroreceptors in the posterior parietal cortex of the human brain. Emission tomographic study of the loss of m2 receptors in AD is limited by the fact that there is currently no available m2-selective radioligand which can penetrate the blood-brain barrier. In our efforts to prepare such a radioligand, we have used competition studies against currently existing muscarinic receptor radioligands to infer the in vitro and in vivo properties of a novel muscarinic receptor ligand, 5-[[4-[4-(diisobutylamino)butyl]-1-phenyl]acetyl]-10,11-dihydro-5H - -dibenzo [b,e][1,4]diazepin-11-one (DIBD). In vitro competition studies against [3H](R)-3-quinuclidinylbenzilate ([3H]QNB) and [3H]N-methylscopolamine ([3H]NMS), using membranes derived from transfected cells expressing only m1, m2, m3, or m4 receptor subtypes, indicate that DIBD is selective for m2/m4 over m1/m3. In vivo competition studies against (R,R)-[125I]IQNB indicate that DIBD crosses the blood brain barrier (BBB). The relationship of the regional percentage decrease in (R,R)-[125I]IQNB versus the percentage of each of the receptor subtypes indicates that DIBD competes more effectively in those brain regions which are known to be enriched in the m2, relative to the m1, m3, and m4, receptor subtype; however, analysis of the data using a mathematical model shows that caution is required when interpreting the in vivo results. We conclude that a suitably radiolabeled derivative of DIBD may be of potential use in emission tomographic study of changes in m2 receptors in the central nervous system.

Compensation for three-dimensional detector response, attenuation and scatter in SPECT grey matter imaging using an iterative reconstruction algorithm which incorporates a high-resolution anatomical image

The use of SPECT to diagnose physiological alterations in disease states depends on the potential of SPECT to provide a quantitatively accurate reconstructed image. However, the reconstructed values depend upon the shape and size of the brain region as strongly as they depend upon true radioactivity concentration. We report here the results of applying an iterative reconstruction algorithm (IRA) to compensate for shape- and size-dependence, as well as for attenuation and scatter. The IRA is designed only for the reconstruction of images for which the true radioactivity in the white matter within the actual brain is negligible compared with the true radioactivity in the grey matter within the actual brain. The IRA incorporates an accurate three-dimensional model of detector response and utilizes an MRI image which defines the anatomical features of the brain being imaged by segmenting the grey, white and ventricular regions. It is the assumption of radioactivity localization exclusively in the grey matter which permits the efficient incorporation of the MRI image. The IRA was validated by simulation studies that utilized a slice through the basal ganglia in the realistic Hoffman three-dimensional mathematical brain model. FBP images deviate significantly from true radioactivity distribution, whereas IRA images are nearly identical to true radioactivity distribution, except for random fluctuations due to the presence of statistical noise. These results indicate that the application of the IRA will permit SPECT to distinguish deficits due to true physiological changes from apparent deficits due to imaging/reconstruction artifacts.

Pharmacokinetic simulations of SPECT quantitation of the M2 muscarinic neuroreceptor subtype in disease states using radioiodinated (R,R)-4IQNB

Alzheimer's disease (AD) involves selective loss of muscarinic M2, but not M1, subtype neuroreceptors in the posterior parietal cortex of the human brain. Emission tomographic study of the loss of M2 receptors in AD is limited by the fact that there is currently no available M2-selective radioligand which can penetrate the blood-brain barrier. However, by taking advantage of the different pharmacokinetic properties of (R,R)-[123I]IQNB for the M1 and M2 subtypes, it may be possible to estimate losses in M2. It has previously been hypothesized that the difference between an early study and a late study should provide information on the M2 receptor population. In order to test this hypothesis, we present here the results of pharmacokinetic simulations of the in vivo localization of (R,R)-[123I]IQNB in brain regions containing various proportions of M1 and M2 subtypes. These results permit us to conclude that SPECT imaging of (R,R)-[123I]IQNB localization can potentially be used to quantitate changes in the M2 subtype in a disease state within a brain region for which the ratio M2/M1 is sufficiently high in normal individuals.

3-D SPECT simulations of a complex 3-D mathematical brain model: effects of 3-D geometric detector response, attenuation, scatter, and statistical noise

The quantitative imaging characteristics of ultrahigh-resolution parallel-hole SPECT, including 3-D geometric detector response, attenuation, scatter, and statistical noise, were investigated by simulations based on a complex digitized 3-D brain model of the gray and white matter distributions. The projection data resulting from a uniform distribution of gray and white matter radioactivity, in a ratio of 5:1, were simulated. The results demonstrate significant qualitative and quantitative artifacts in reconstructed human brain images. In the absence of attenuation, scatter, and noise, artifactual variation caused inaccuracies in regional radioactivity quantification. Inclusion of attenuation scatter, and noise in the simulation caused additional artifacts, and resulted in reconstructed images which qualitatively and quantitatively corresponded very closely to reconstructed images of the actual 3-D brain phantom which was constructed from the same set of data as the mathematical 3-D brain model. It is concluded that the major degrading factor in SPECT neuroimaging is the 3-D geometric detector response function.

Three-dimensional SPECT simulations of a complex three-dimensional mathematical brain model and measurements of the three-dimensional physical brain phantom

We have developed a three-dimensional computer simulation of SPECT imaging. We have applied the simulation procedure to the realistic mathematical Hoffman three-dimensional brain model to generate the projection data (in the absence of attenuation, scatter, or noise) of both a parallel-hole and a multidetector SPECT system with point-focusing collimators. The simulated projection data were then reconstructed using standard software. The projection data resulting from the distribution of grey matter alone, or grey and white matter, were simulated. The results of these simulations indicate the existence of significant qualitative and quantitative artifacts in reconstructed human brain images. For example, the reconstructed values for grey matter along a cortical circumferential profile in a transverse slice through the basal ganglia varied by a factor of 2.40 (parallel-hole) and 2.99 (point-focusing), although the original grey matter values were identical in all cortical regions in the model. We have compared the simulated reconstructed images with those obtained by imaging the physical three-dimensional Hoffman brain phantom, which was constructed based upon the same set of data from which the mathematical three-dimensional Hoffman brain model was derived. Although the simulation did not include all of the degrading factors present in the physical imaging, the two images were in good agreement, indicating the applicability of the simulation to a realistic situation and the importance of the detector resolution effect.

Binding of radioiodinated SPECT ligands to transfected cell membranes expressing single muscarinic receptor subtypes

The equilibrium dissociation constant and the kinetic rate constants were determined for the binding of (R)-[3H]3-quinuclidinyl benzilate ([3H]QNB) and [125I]3-quinuclidinyl-4-iodobenzilate ((R,R)- and (R,S)-[125I]IQNB) to transfected cell membranes expressing one single muscarinic acetylcholine receptor (mAChR) subtype. The association and dissociation kinetics for the m2 subtype were more rapid than for the m1 and m3 subtypes. The differential kinetic properties may be useful for the single photon emission computed tomographic (SPECT) evaluation of regional mAChR subtype alterations in disease states.

In vivo dissociation kinetics of [3H]quinuclidinyl benzilate: relationship to muscarinic receptor concentration and in vitro kinetics

The in vivo washout kinetics of [3H]quinuclidinyl benzilate ([3H]QNB) varies significantly in various structures in the rat brain. The slowest washout rates are from the hippocampus, corpus striatum, and cortex, intermediate rates are exhibited from the thalamus and colliculi, while the fastest washout rate is from the cerebellum. We have also demonstrated a difference in the in vitro dissociation rates (k-1) of [3H]QNB from various structures. The k-1 for the hippocampus, corpus striatum and cortex, is two-fold slower than that observed in the thalamus, colliculi, and cerebellum. The differences in the in vitro dissociation kinetics are not, however, sufficient to explain the differences in the in vivo washout kinetics. We have developed a theoretical formulation which describes conditions under which the washout kinetics are a function of the concentration of receptor in a structure. Furthermore, we present a graphical method in which a plot of the reciprocal of the observed washout rate constant, 1/k(obs), vs receptor concentration is linear. Analysis of the washout kinetics of [3H]QNB from various structures of the CNS of rat were well described by this theory when the differences in in vitro k-1 are included.

Three-dimensional simulations of multidetector point-focusing SPECT imaging

We have applied an efficient algorithm for mathematically simulating the three-dimensional (3-D) response of a SPECT imaging system with a depth-dependent 3-D point spread function (3-DPSF). The input object whose reconstructed image is to be simulated is restricted to a binary map; more complex objects may be treated as linear combinations of binary maps. The 3-D convolution reduces to a sequence of additions of a 3-D line spread function (3-DLSF), appropriately translated, to the 3-D response. We have simulated the projection data from a multidetector SPECT system with point-focusing collimators. The simulated projection data were then reconstructed using the manufacturer's software. The objects simulated included simple geometrical solids such as spheres and cylinders, as well as the distribution of muscarinic cholinergic receptors in a realistic brain slice. The results of these simulations indicate the existence of significant qualitative and quantitative artifacts in reconstructed human brain images.