Anatomical pulmonary magnetic resonance imaging segmentation for regional structure-function measurements of asthma.

PURPOSE: Pulmonary magnetic-resonance-imaging (MRI) and x-ray computed-tomography have provided strong evidence of spatially and temporally persistent lung structure-function abnormalities in asthmatics. This has generated a shift in their understanding of lung disease and supports the use of imaging biomarkers as intermediate endpoints of asthma severity and control. In particular, pulmonary (1)H MRI can be used to provide quantitative lung structure-function measurements longitudinally and in response to treatment. However, to translate such biomarkers of asthma, robust methods are required to segment the lung from pulmonary (1)H MRI. Therefore, their objective was to develop a pulmonary (1)H MRI segmentation algorithm to provide regional measurements with the precision and speed required to support clinical studies. METHODS: The authors developed a method to segment the left and right lung from (1)H MRI acquired in 20 asthmatics including five well-controlled and 15 severe poorly controlled participants who provided written informed consent to a study protocol approved by Health Canada. Same-day spirometry and plethysmography measurements of lung function and volume were acquired as well as (1)H MRI using a whole-body radiofrequency coil and fast spoiled gradient-recalled echo sequence at a fixed lung volume (functional residual capacity + 1 l). We incorporated the left-to-right lung volume proportion prior based on the Potts model and derived a volume-proportion preserved Potts model, which was approximated through convex relaxation and further represented by a dual volume-proportion preserved max-flow model. The max-flow model led to a linear problem with convex and linear equality constraints that implicitly encoded the proportion prior. To implement the algorithm, (1)H MRI was resampled into ∼3 × 3 × 3 mm(3) isotropic voxel space. Two observers placed seeds on each lung and on the background of 20 pulmonary (1)H MR images in a randomized dataset, on five occasions, five consecutive days in a row. Segmentation accuracy was evaluated using the Dice-similarity-coefficient (DSC) of the segmented thoracic cavity with comparison to five-rounds of manual segmentation by an expert observer. The authors also evaluated the root-mean-squared-error (RMSE) of the Euclidean distance between lung surfaces, the absolute, and percent volume error. Reproducibility was measured using the coefficient of variation (CoV) and intraclass correlation coefficient (ICC) for two observers who repeated segmentation measurements five-times. RESULTS: For five well-controlled asthmatics, forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) was 83% ± 7% and FEV1 was 86 ± 9%pred. For 15 severe, poorly controlled asthmatics, FEV1/FV C = 66% ± 17% and FEV1 = 72 ± 27%pred. The DSC for algorithm and manual segmentation was 91% ± 3%, 92% ± 2% and 91% ± 2% for the left, right, and whole lung, respectively. RMSE was 4.0 ± 1.0 mm for each of the left, right, and whole lung. The absolute (percent) volume errors were 0.1 l (∼6%) for each of right and left lung and ∼0.2 l (∼6%) for whole lung. Intra- and inter-CoV (ICC) were <0.5% (>0.91%) for DSC and <4.5% (>0.93%) for RMSE. While segmentation required 10 s including ∼6 s for user interaction, the smallest detectable difference was 0.24 l for algorithm measurements which was similar to manual measurements. CONCLUSIONS: This lung segmentation approach provided the necessary and sufficient precision and accuracy required for research and clinical studies.

Mapping of genes and transcribed sequences in a gene rich 400-kb region on human chromosome 11p15.1-->p14.

We have identified a number of transcribed sequences within a 400-kb interval on chromosome 11p15.1--> p14. Six genes and 13 novel transcripts including ESTs, cDNAs and exons have been identified and assigned to this region. Comparison of mRNA sequence with genomic sequence has enabled us to determine the exon/intron structure of four of the genes (NUCB2, PIK3C2A, RPS13 and OR7E14P).

Comparative structure, proximal promoter elements, and chromosome location of the human eosinophil major basic protein genes.

Human eosinophil major basic protein (MBP) is strongly implicated as a mediator of disease, especially bronchial asthma. We recently isolated a highly divergent human homologue of MBP (MBPH). Given human MBP's importance in disease and the restricted expression of it and human MBPH, we isolated the 4.6-kb human MBPH gene (HGMW-approved symbol PRG3). Comparisons among the human MBP (PRG2), human MBPH, and murine MBP-1 (mMBP-1; Prg2) genes suggest that the human MBP and mMBP-1 genes are more closely related than either is to the human MBPH gene. Proximal promoters of these three genes show conservation of potential binding sites for IK2 and STAT and of a known GATA site. However, a known C/EBP site is altered in the human MBPH gene's proximal promoter. The human MBP and MBPH genes localized to chromosome 11 in the centromere to 11q12 region. Thus, the human MBP and MBPH genes have diverged considerably, probably following a gene duplication event. Furthermore, the identified conserved and distinct proximal promoter elements likely contribute to the eosinophil-restricted and relatively reduced transcription of the human MBPH gene.

Spermatid-specific expression of the novel X-linked gene product SPAN-X localized to the nucleus of human spermatozoa.

Formation of mature spermatozoa involves a series of dramatic molecular and morphological changes in the male germ cell lineage. These changes result from the temporally regulated transcription and translation of several testis-specific gene products. Here, we describe a novel, testis-specific protein designated SPAN-X for sperm protein associated with the nucleus on the X chromosome. SPAN-X sequences showed no significant similarity with known cDNA or peptide sequences. The SPAN-X peptide sequences contained three overlapping consensus nuclear localization signals, a high percentage (33%-37%) of charged amino acid residues, and a relatively acidic isoelectric point (pI; 4.88-6.05). Northern analysis of mRNA from multiple human tissues identified a SPAN-X transcript exclusively in the testis. In situ hybridization of human testes sections showed SPAN-X mRNA expression in haploid, round, and elongating spermatids. The SPANX gene was mapped to chromosome Xq27. 1 by fluorescence in situ hybridization and by Southern blot analysis of human/mouse somatic cell hybrids. On Western blots of human sperm proteins, antirecombinant SPAN-X antibodies reacted with broad bands migrating between 15-20 kDa. Immunofluorescent labeling of human spermatozoa demonstrated SPAN-X localization to nuclear craters and cytoplasmic droplets. Expression of SPAN-X, an X-linked gene product, exclusively in haploid spermatids leads to interesting questions regarding the transcription of sex-linked genes during spermiogenesis.

Coding sequence and expression patterns of mouse chordin and mapping of the cognate mouse chrd and human CHRD genes.

Chordin is a key developmental protein that dorsalizes early vertebrate embryonic tissues by binding to ventralizing TGF-beta-like bone morphogenetic proteins and sequestering them in latent complexes. Here we report the first characterization of mammalian chordin. The full-length cDNA sequence for mouse chordin is given, and RNA blot analysis shows the murine chordin gene Chrd to be expressed at relatively high levels in 7-day postcoitum mouse embryos and at much decreased levels at later developmental times and in adult tissues. These results imply a major role for chordin during gastrulation of the mammalian embryo. Nevertheless, both murine and human chordin genes are shown to be expressed at readily detectable levels in several fetal and adult tissues, most notably liver and cerebellum, suggesting additional roles in organogenesis and homeostasis. Chrd was mapped to mouse chromosome 16 using interspecific crosses, and the cognate human gene CHRD was localized to human chromosome 3q27 by radiation hybrid mapping.

Promoter sequence, expression, and fine chromosomal mapping of the human gene (MLP) encoding the MARCKS-like protein: identification of neighboring and linked polymorphic loci for MLP and MACS and use in the evaluation of human neural tube defects.

The MARCKS-like protein (MLP), also known as F52, MacMARCKS, or MARCKS-related protein, is a widely distributed substrate for protein kinase C (PKC). Recent studies using gene disruption in vivo have demonstrated the importance of both MARCKS and MLP to the development of the central nervous system; specifically, mice lacking either protein exhibit a high frequency of neural tube defects. We isolated a genomic clone for human MLP and discovered a directly linked polymorphism (MLP1) useful for genetic linkage analysis. The MLP promoter was 71% identical over 433 bp to that of the corresponding mouse gene, Mlp, with conservation of many putative transcription factor-binding sites; it was only 36% identical over 433 bp to the promoter of the human gene, MACS, which encodes the MLP homologue MARCKS. This 433-bp fragment drove expression of an MLP-beta-galactosidase transgene in a tissue-specific and developmental expression pattern that was similar to that observed for the endogenous gene, as shown by in situ hybridization histochemistry. In contrast to MACS, the MLP and Mlp promoters contain a TATA box approximately 40 bp 5' of the presumed transcription initiation site. MLP was localized to chromosome 1p34-->1pter by analysis of human-mouse somatic cell hybrid DNA and to 1p34 by fluorescence in situ hybridization. Radiation hybrid mapping of MLP placed it between genetic markers D1S511 (LOD > 3.0) and WI9232. MACS was localized to 6q21 between D6S266 (LOD > 3.0) and AFM268uh5 by the same technique. We tested the novel MLP1 polymorphism and the MACS flanking markers in a series of 43 Caucasian simplex families in which the affected child had a lumbosacral myelomeningocele. We found no evidence of linkage disequilibrium, suggesting that these loci were not major genes for spina bifida in these families. Nonetheless, the identification of linked and neighboring polymorphisms for MACS and MLP should permit similar genetic studies in other groups of patients with neural tube defects and other neurodevelopmental abnormalities.

The human transaldolase gene (TALDO1) is located on chromosome 11 at p15.4-p15.5.

Transaldolase (TAL) is a key enzyme of the pentose phosphate pathway, which is responsible for generation of reducing equivalents to protect cellular integrity from reactive oxygen intermediates. While exons 2 and 3 are highly repetitive, the complete TAL-H gene is mapped to a single genomic locus (TALDO1(2)) by several independent approaches. Southern blot hybridization of a 827-bp 3' EcoRI fragment of the TAL-H cDNA to human-mouse somatic cell hybrid DNA localized TALDO1 to the p13-->pter region of chromosome 11. Fluorescence in situ hybridization with a 15-kb genomic fragment harboring exons 1 and 2 mapped TALDO1 to 11p15.4-p15.5. A truncated and mutated segment of TAL-H exon 5 terminating with a poly(A) tail was identified in a pseudogene locus (TALDOP1) on chromosome 1. Reverse transcriptase-PCR studies of human-mouse somatic cell hybrids revealed the presence of the functional TAL-H gene on chromosome 11 and its absence on human chromosome 1. Mapping of radiation hybrids placed TALDO1 between markers WI-1421 and D11S922 on 11p15.

Human fertilin beta: identification, characterization, and chromosomal mapping of an ADAM gene family member.

Fertilin alpha/beta (PH30 alpha/beta) is a heterodimeric sperm surface protein containing binding and fusion domains with potential for interaction with integrin receptors on the oocyte. We report the cDNA cloning, deduced amino acid sequence, tissue specificity, and chromosomal mapping of human fertilin beta. Encoded by a 2205 nucleotide open reading frame, the deduced amino acid sequence of human fertilin beta contains pro-, metalloprotease-like, disintegrin-like, cysteine-rich, epidermal growth factor-like (EGF) repeat, transmembrane, and cytoplasmic domains. Due to this domain organization, human fertilin beta has been identified as a member of the ADAM family, which is composed of membrane-anchored proteins having A Disintegrin And Metalloprotease domain. The amino acid sequence of human fertilin beta shares 90%, 56%, and 55% identity, respectively, to monkey, guinea pig, and mouse fertilin beta homologs. A phenylalanine-glutamate-glutamate (FEE) binding tripeptide within the disintegrin-like domain of human fertilin beta, homologous to other fertilin beta RGD-like (arginine-glycine-aspartic acid) tripeptides, could compete for recognition by integrins and other receptors. Northern analysis from 16 human tissues revealed human fertilin beta's 2.9 kb message only in testis, which raises interest in possible clinical applications of this molecule as a contraceptive vaccinogen. Human fertilin beta maps to chromosome 8, band p11.2, by fluorescence in situ hybridization and mouse/human somatic cell hybrid Southern hybridization.

Chromosomal localization of a human mucin gene (MUC8) and cloning of the cDNA corresponding to the carboxy terminus.

A partial cDNA (pAM1) encoding a major airway mucin glycoprotein with novel tandem repetitive sequence has recently been cloned (Shankar, V., M. S. Gilmore, R. C. Elkins, and G. P. Sachdev. 1994. Biochem. J. 300:295-298). In this article, we report additional new sequence derived by 3'-rapid amplification of cDNA ends technique. The sequence corresponds to a stop codon, 3'-untranslated region of 458 bp, a polyadenylation signal, and poly A+ tail, and represents the extreme carboxy terminus of MUC8. A plasmid construct (pAM3) in pBluescript was generated by in-frame ligation of pAM1 to the 479-bp 3'UTR of MUC8. A 5'-end 325-bp fragment of this cDNA subcloned into the protein fusion and expression vector pET28b(+) was used to generate fusion protein under the control of T7 promoter. The purified fusion protein as well as synthetic peptide corresponding to the MUC8 repeat sequence (TSCPRPLQEGTPGS) were used to raise polyclonal antibodies in rabbits. The antiserum to the fusion protein and to the synthetic peptide reacted with the deglycosylated major tracheobronchial mucin. Immunohistochemical studies using the above antibodies localized the MUC8 protein product to submucosal glands in human tracheal epithelium. Furthermore, the gene from which this cDNA is derived, was mapped to chromosome 12 using DNA from a panel of human-mouse somatic cell hybrids. Fluorescence in situ hybridization was used to assign the regional localization to 12q24.3. Since the eight known human mucin genes map to other chromosomes, we have named this gene MUC8, in accordance with mucin gene nomenclature.

A human gene (DDX10) encoding a putative DEAD-box RNA helicase at 11q22-q23.

A human gene encoding a putative RNA helicase, designated DDX10, was identified 400 kb telomeric to the ataxia-telangiectasia gene at chromosome 11q22-q23. The predicted amino acid sequence shows very high similarity to a subgroup of DEAD-box RNA helicases involved in ribosome biogenesis. This novel gene encodes a 3.2-kb transcript in a variety of human tissues. A processed pseudogene of DDX10 was detected at chromosome 9q21-q22. We observed a rare trinucleotide repeat length polymorphism within the coding sequence of DDX10.

New members of the chemokine receptor gene family.

Chemokines are relatively small peptides with potent chemoattractant and activation activities for leukocytes. Several chemokine receptors have been cloned and characterized and all are members of the G protein-coupled receptor superfamily. Using degenerate oligonucleotides and polymerase chain reaction, we have identified seven novel receptors. Two of these sequences are presented here for the first time. We have shown, with gene mapping studies, that receptors with the highest sequence similarity are closely linked on human chromosomes. This close genetic association suggests a functional relationship as well.

Mouse and human homologues of the yeast origin of replication recognition complex subunit ORC2 and chromosomal localization of the cognate human gene ORC2L.

ORC2 is a subunit of the origin recognition complex in yeast and has been implicated in the initiation of DNA replication and transcriptional silencing. We have isolated mouse and human cDNA clones encoding proteins with 47.9 and 46.3% similarity, respectively, to yeast ORC2. This degree of similarity and the alignment of sequences suggest that these clones may represent a mammalian ORC2 homologue. The existence of such a homologue would, in turn, suggest the existence of a mammalian origin recognition complex, similar to that found in yeast. Although Northern blot analysis of various adult mouse tissues found the highest levels of expression of ORC2-like (ORC2L) RNA in testes, strong signals did not always correspond to tissues in which high levels of DNA replication would be expected. This finding may reflect functional roles of ORC2L distinct from those that it may play in DNA replication. Analyses of somatic cell hybrid DNA and fluorescence in situ hybridization were employed to map the human ORC2L gene to chromosome 2q33.

The orphan G-protein-coupled receptor-encoding gene V28 is closely related to genes for chemokine receptors and is expressed in lymphoid and neural tissues.

A polymerase chain reaction (PCR) strategy with degenerate primers was used to identify novel G-protein-coupled receptor-encoding genes from human genomic DNA. One of the isolated clones, termed V28, showed high sequence similarity to the genes encoding human chemokine receptors for monocyte chemoattractant protein 1 (MCP-1) and macrophage inflammatory protein 1 alpha (MIP-1 alpha)/RANTES, and to the rat orphan receptor-encoding gene RBS11. When RNA was analyzed by Northern blot, V28 was found to be most highly expressed in neural and lymphoid tissues. Myeloid cell lines, particularly THP.1 cells, showed especially high expression of V28. We have mapped V28 to human chromosome 3p21-3pter, near the MIP-1 alpha/RANTES receptor-encoding gene.

The Brn-3 family of POU-domain factors: primary structure, binding specificity, and expression in subsets of retinal ganglion cells and somatosensory neurons.

A search for POU domain sequences expressed in the human retina has led to the identification of three closely related genes: Brn-3a, Brn-3b, and Brn-3c. The structure and expression pattern of Brn-3b was reported earlier (Xiang et al., 1993); we report here the structures and expression patterns of Brn-3a and Brn-3c. Antibodies specific for each Brn-3 protein were generated and shown to label only ganglion cells in a variety of vertebrate retinas. A complex pattern of strongly and weakly immunolabeled ganglion cells was observed in mouse, cat, and monkey retinae. In mouse and cat retinae, Brn-3a and Brn-3b proteins are found in a large fraction of ganglion cells, whereas Brn-3c is present in fewer ganglion cells. In the cat retina, anti-Brn-3a immunoreactivity was strong in the small ganglion cells (gamma cells) and weak in the remaining ganglion cells (alpha and beta cells); anti-Brn-3b immunoreactivity was present in all ganglion cells; and anti-Brn3c immunoreactivity was confined to the small ganglion cells. Immunolabeling of macaque retinae following retrograde labeling from the lateral geniculate nucleus revealed strong anti-Brn-3a immunoreactivity in a minority of retrogradely labeled P-type ganglion cells, and weak Brn-3a immunoreactivity in all of the remaining P- and M-type ganglion cells. In the same retinae, strong anti-Brn-3b immunoreactivity was seen in nearly all P-type ganglion cells and weak immunoreactivity in nearly all M-type ganglion cells. Each of the Brn-3-specific antibodies also labeled subsets of neurons in the dorsal root and trigeminal ganglia, suggesting that primary somatosensory neurons and retinal ganglion cells share genetic regulatory hierarchies. In vitro selection of an optimal DNA binding site using the Brn-3b POU domain has revealed a consensus [(A/G)CTCATTAA(T/C)] that is recognized by each of the Brn-3 POU domains and is distinct from binding sites previously described for other POU domain proteins.

Genomic organization and chromosomal localization of the gene encoding human P-selectin glycoprotein ligand.

The gene for P-selectin glycoprotein ligand (PSGL-1) has been cloned from a human placenta genomic DNA library. A single intron of approximately 9 kilobases was found in the 5'-untranslated region and the complete coding region resides in exon 2. The genomic clone differs from the cDNA clone isolated from HL-60 cells in that it encodes an extra copy of the decameric repeat located in the extracellular domain of PSGL-1. Further analysis indicated that the PSGL-1 genes of HL-60 and U-937 cells contain 15 repeats, whereas the PSGL-1 genes of polymorphonuclear leukocytes, monocytes, and several other cell lines contain 16 repeats. Transfection experiments did not indicate a functional difference between these two variants of PSGL-1. The two previously observed PSGL-1 mRNA species of 2.5 and 4 kilobases most likely arise from differential utilization of polyadenylation signal sequences. The organization of the PSGL-1 gene closely resembles those of CD43 and human platelet glycoprotein GPIb alpha, both of which have an intron in the 5'-noncoding region, a long second exon containing the complete coding region, and TATA-less promoters. The gene for human PSGL-1, which has been designated SELPLG by the Human Gene Nomenclature Committee, was mapped to chromosome 12q24 using Southern blot analysis of DNA from a set of human-mouse cell hybrids, and fluorescent in situ hybridization on metaphase chromosome spreads.

Three genes that encode human beta-galactoside alpha 2,3-sialyltransferases. Structural analysis and chromosomal mapping studies.

The synthesis of alpha 2,3-linked sialic acid to Gal(beta 1,3)GalNAc is mediated by at least three beta-galactoside alpha 2,3-sialyltransferases (EC 2.4.99.4, SiaT-4) that are encoded by three distinct genes. In contrast, only a single gene encodes the beta-galactoside alpha 2,6-sialyltransferase (EC 2.4.99.1, SiaT-1). This report assesses the relationship and nature of the SiaT-4 genes. Analysis of human-mouse somatic cell hybrids demonstrates that the sialyltransferase genes are dispersed in the human genome. The gene for SiaT-4 resides in chromosome 8, that for SiaT-4b resides in p21-p34 of chromosome 1 and that for SiaT-4c in q23.3-qter of chromosome 11. The gene symbols for these genes have been designated SIAT4A, SIAT4B and SIAT4C, respectively. To assess the structural organization of one of the SiaT-4 genes, a human SiaT-4a cDNA from submaxillary glands was isolated and characterized. Rapid amplification of cDNA 5' ends (5'-RACE) analysis indicates an unusually long 1 kb 5'-untranslated leader. The catalytic domain of the cloned sequence was expressed in transfected cells and was shown to be competent in mediating the specific synthesis of sialic acid alpha 2,3 to Gal(beta 1,3)GalNAc-R. Genomic sequences for SiaT-4a were also isolated and examined. The data demonstrate that coding information for SiaT-4a protein is dispersed into seven discrete exon segments in a manner reminiscent of the SiaT-1 gene. Furthermore, as in the SiaT-1 gene, intervening sequences interrupt both sialylmotif domains, regions that are conserved among all known sialyltransferases.

Human cDNA clones that modify radiomimetic sensitivity of ataxia-telangiectasia (group A) cells.

Genes responsible for genetic diseases with increased sensitivity to DNA-damaging agents can be identified using complementation cloning. This strategy is based on in vitro complementation of the cellular sensitivity by gene transfer. Ataxia-telangiectasia (A-T) is a multisystem autosomal recessive disorder involving cellular sensitivity to ionizing radiation and radiomimetic drugs. A-T is genetically heterogeneous, with four complementation groups. We attempted to identify cDNA clones that modify the radiomimetic sensitivity of A-T cells assigned to complementation group [A-T(A)]. The cells were transfected with human cDNA libraries cloned in episomal vectors, and various protocols of radiomimetic selection were applied. Thirteen cDNAs rescued from survivor cells were found to confer various degrees of radiomimetic resistance to A-T(A) cells upon repeated introduction, and one of them also partially influenced another feature of the A-T phenotype, radioresistant DNA synthesis. None of the clones mapped to the A-T locus on chromosome 11q22-23. Nine of the clones were derived from known genes, some of which are involved in cellular stress responses. We concluded that a number of different genes, not necessarily associated with A-T, can influence the response of A-T cells to radiomimetic drugs, and hence the complementation cloning approach may be less applicable to A-T than to other diseases involving abnormal processing of DNA damage.

Chromosomal assignment of the heparin-binding cytokine genes MDK and PTN in mouse and man.

MDK and PTN are two members of a family of heparin-binding cytokines thought to be involved in a number of developmental processes. The locations for these genes were determined in man and mouse using somatic cell hybrid analysis and interspecific backcross analysis. Human MDK was mapped to 11p13-->p11. MDK in the mouse (Mdk) was mapped to a syntenic region of mouse Chromosome 2. A pseudogene of Mdk was mapped to mouse Chromosome 11. The closely related human gene PTN was mapped to a separate location on human chromosome region 7q22-->qter.

Assignment of the gene for human carbonic anhydrase VIII(CA8) to chromosome 8q11-->q12.

The human carbonic anhydrase (CA) VIII gene (CA8) has been mapped to chromosome 8 at q11-->q12 by human-mouse hybrid mapping and by fluorescence in situ hybridization. The closely-linked human CA isozyme genes, CA1, CA2 and CA3, are also located on chromosome 8, but at q22, and therefore not closely linked to the CA8 locus.