Femur statistical atlas construction based on two-level 3D non-rigid registration.

The statistical atlas is a 3D medical image analysis tool to enable more patient-oriented and efficient diagnosis. The atlas includes information on geometry and its variation across populations. The comparison with information from other patients is very useful for objective quantitative diagnosis. The statistical atlas can also be used to solve other challenging problems such as image segmentation. As a key to the construction of statistical atlases, 3D registration remains an important yet unsolved problem in the medical image field due to the geometrical complexity of anatomical shapes and the computational complexity arising from the enormous size of volume data. In this work we developed a two-level framework to efficiently solve 3D non-rigid registration, and applied the method to the problem of constructing statistical atlases of the femur. In contrast to a general multi-resolution framework, we employed an interpolation to propagate the matching instead of repeating the registration scheme in each resolution. The registration procedure is divided into two levels: a low-resolution solution to the correspondences and mapping of surface models using Chui and Rangarajan's thin-plate spline (TPS)-based algorithm, followed by an interpolation to achieve high-resolution matching. Next, principal component analysis (PCA) is used to build the statistical atlases. Experimental results show the shape variation learned from the atlases, and also demonstrate that our method significantly improves the efficiency of registration without decreasing the accuracy of the atlases.

Variations in acetabular anatomy with reference to total hip replacement.

Three-dimensional surface models of the normal hemipelvis derived from volumetric CT data on 42 patients were used to determine the radius, depth and orientation of the native acetabulum. A sphere fitted to the lunate surface and a plane matched to the acetabular rim were used to calculate the radius, depth and anatomical orientation of the acetabulum. For the 22 females the mean acetabular abduction, anteversion, radius and normalised depth were 57.1 degrees (50.7 degrees to 66.8 degrees ), 24.1 degrees (14.0 degrees to 33.3 degrees ), 25 mm (21.7 to 30.3) and 0.79 mm (0.56 to 1.04), respectively. The same parameters for the 20 males were 55.5 degrees (47.7 degrees to 65.9 degrees ), 19.3 degrees (8.5 degrees to 32.3 degrees ), 26.7 mm (24.5 to 28.7) and 0.85 mm (0.65 to 0.99), respectively. The orientation of the native acetabulum did not match the safe zone for acetabular component placement described by Lewinnek. During total hip replacement surgeons should be aware that the average abduction angle of the native acetabulum exceeds that of the safe zone angle. If the concept of the safe zone angle is followed, abduction of the acetabular component should be less than the abduction of the native acetabulum by approximately 10 degrees .

Accuracy of ultrasound to MR registration of the knee.

BACKGROUND: Ultrasound-based registration to 3D surfaces segmented from MR imaging is proposed as a non-invasive alternative to point-based registration for image-guided surgery. By relying upon diagnostic MR imaging, the expense of additional CT imaging (and exposure to radiation) is avoided. The technique would enable navigation in arthroscopic and other minimally invasive procedures. METHODS: Optically tracked registrations using point-based and ultrasound-based methods to MR and CT imaging volumes for two cadaveric specimens were acquired and analysed. RESULTS: The average RMS distance between fiducials was 0.27 mm for CT and 0.72 mm for MR utilizing point-based registration. The average RMS distance for ultrasound-based registration to CT was 0.59 mm and 0.76 mm to MR. The RMS distance for fiducials co-located in MR and CT imaging volumes was 0.74 mm. The end-to-end error of ultrasound registration to MR imaging was 2.98 mm, as compared to 1.65 mm for CT. CONCLUSIONS: Ultrasound registration to MR imaging data is a viable non-invasive alternative to point-based registration.

An androgen response element mediates LNCaP cell dependent androgen induction of the hK2 gene.

Human glandular kallikrein (hK2) is an androgen regulated protein primarily expressed in the prostate and recently identified as a novel prostate cancer marker. A 5 kb 5' flanking region of the hK2 gene was isolated and sequenced to characterize the regulatory mechanisms for the expression of hK2 in the androgen responsive prostate cell line, LNCaP. Using gene transfer, gel shift, and mutagenesis assays we have identified an ARE in the 5' far upstream promoter region of the hK2 gene that is crucial for its regulation in LNCaP cells. This study further demonstrated that the hK2 upstream ARE plays a predominant role in androgenic response. More interestingly, previously identified AREs in the prostate specific antigen promoter and the hK2 proximal promoter exert little activity in LNCaP cells. This study for the first time identifies a unique ARE that alone mediates the function of the androgen receptor in LNCaP cells in a cell dependent manner. This study also examines the activity of this ARE with 1alpha, 25 dihydroxy-vitamin D3 on the expression of the hK2 gene in LNCaP cells.

Effects of Ca++ mobilization on expression of androgen-regulated genes: interference with androgen receptor-mediated transactivation by AP-I proteins.

BACKGROUND: The adult prostate is maintained through an equilibrium between cell growth and death rates. Androgen deprivation induces an increase in intracellular Ca++, AP-1 gene expression of androgen-inducible genes. METHODS: Northern blot analysis, band-shift assays, and transient cotransfection assays were used to study the effects of Ca++ mobilizer A23187 on gene expression in human prostate cancer cells. RESULTS: A23187 repressed androgen-upregulated mRNAs for prostate-specific antigen (PSA) and hKLK2, and rapidly induced mRNA levels for c-fos and c-jun. AP-1 protein-DNA binding activities were elevated after A23187 treatments. Androgen receptor (AR)-mediated induction of chloramphenicol acetyltransferase (CAT) reporter was repressed by AP-1 proteins. CONCLUSIONS: The repression of AR-mediated induction of PSA and hKLK2 genes by Ca++ mobilizers is due to the interference of AR transactivation activity by AP-1 proteins.

Identification of two novel cis-elements in the promoter of the prostate-specific antigen gene that are required to enhance androgen receptor-mediated transactivation.

A monomeric androgen responsive element (ARE) is not sufficient to mediate significant androgen induction of the prostate-specific antigen (PSA) gene. Co-transfection experiments using a series of 5'deletion fragments of the proximal promoter region of the PSA gene linked to bacterial chloramphenicol acetyltransferase (CAT) as a reporter have identified two motif sequences which are indispensable for androgen receptor (AR)-mediated transactivation of the PSA promoter and have been designated as motifs A and B respectively. Of note, motif B alone has very little independent enhancer activity regardless of the presence or absence of androgen, whereas multi-copies of motif A exert androgenic inducibility for a heterologous promoter independent of the presence of ARE. Nucleotide substitutions in either motif significantly decrease the androgen inducibility and the nuclear protein binding ability. Furthermore, gel band shift experiments consistently demonstrate that nuclear proteins can bind these motifs, and they are non-receptor factors. Our data indicate that these two DNA motifs are novel cis -regulatory elements and exhibit different mechanisms in cooperation with ARE for AR-mediated transactivation.

Defining a functional androgen responsive element in the 5' far upstream flanking region of the prostate-specific antigen gene.

Previously, an androgen responsive element (ARE or promoter ARE) was identified in the proximal promoter region of the prostate-specific antigen (PSA) gene. The proximal promoter fragment could mediate androgen induction of expression of a heterologous reporter gene in androgen receptor (AR)-less cells with exogenous AR in co-transfection assays. However, it exerted little androgen inducibility in androgen sensitive human prostate cells, LNCaP, which produce PSA mRNA and protein. In this study, we have identified a second functional ARE (or upstream ARE) approximately 4 kb upstream of the cap site of the PSA transcript. Interestingly, although the AREs are necessary for androgen induction, the DNA sequences surrounding the upstream ARE are also required for androgen induction by the PSA promoter in LNCaP cells. The results indicated that the upstream DNA sequences can cooperate with either ARE to mediate androgen induced gene expression in LNCaP cells.

Calpain inhibitor-induced apoptosis in human prostate adenocarcinoma cells.

Although it has been shown that calpains may play a positive role in causing apoptosis of T cells, we report here that, on the contrary, the inhibition of calpain-like activities can induce apoptosis in human prostate cancer cells. Two calpain inhibitors were used to test growth response on prostate cancer cells and showed remarkable cytotoxicity. The cytotoxicity was due to apoptosis as judged by large genomic DNA fragmentation, chromatin condensation and nuclear fragmentation. Furthermore, using gel band shift assays we have demonstrated that calpain inhibitor 1 causes a prolonged elevation of AP-1 protein activity in human prostate cancer cells. The elevation of AP-1 activity appears to be specific, because calpain inhibitor 1 only stimulates AP-1 but not AP-2 and SP-1 activities. We postulate that the sustained increase in AP-1 activity may be involved in apoptosis induced in prostate cells by calpain inhibitors. Our study thus suggests that calpain-like activity may be a potentially therapeutic target for cancer.

Antagonism of androgen action in prostate tumor cells by retinoic acid.

Recent studies have demonstrated that retinoic acid (RA) can repress the growth of human prostatic epithelial cells. Since the proliferation of prostate cells is highly dependent on androgen stimulation, presumably via its cognate receptor, we investigated the effects of RA on the expression of the androgen receptor and other androgen-regulated genes in the human prostatic adenocarcinoma cell line LNCaP. Using a radioligand binding assay, we found that androgen-binding activity was reduced 30-40% in cells treated with 10(-5) M RA plus 6 nM dihydrotestosterone (DHT), as compared to cells with the androgen alone. Moreover, the reduction of the androgen receptor (AR) was not accompanied by alteration of the ligand-binding affinity. Concomitant changes in the function of AR were manifested by a dramatic reduction in AR-mediated transcription activity in a transfection experiment. Androgen-induced levels of both prostate-specific antigen (PSA) and human glandular kallikrein-1 (hKLK2) mRNAs were significantly repressed by RA in a dose- and time-dependent manner. Consistent with this finding, androgen induction of PSA glycoprotein was also repressed by RA, with maximal inhibition occurring at 10(-5) M. These data suggest that the suppression of proliferation and function of prostatic cells by RA may be via modulatory effects on the AR.

Tumor-promoting phorbol ester-induced cell death and gene expression in a human prostate adenocarcinoma cell line.

12-O-tetradecanoyl phorbol ester (TPA) has profound cytotoxic effects on a human prostate cancer cell line, LNCaP. The TPA effect may be mediated via a protein kinase C (PKC) pathway, since staurosporine, a potent PKC inhibitor, could reverse the cell-killing effect. Our studies, based on cellular fragmentation, chromatin condensation, and nuclear fragmentation, suggest that the cell-killing effect is due to apoptosis. Moreover, we also examined expression of early growth response genes and androgen-induced genes in association with TPA-induced apoptosis. Northern blot analysis demonstrated that androgen induction of human glandular kallikrein-1 (hKLK2) mRNA was repressed by TPA in a concentration-dependent manner. A time course study showed that both hKLK2 and c-myc mRNAs were repressed by TPA as early as four hours. In contrast, the steady state mRNA levels for c-fos, c-jun, nerve growth factor induced gene A, and the orphan steroid receptor nur77 were rapidly induced within the first two hours of the treatment. Furthermore, transient co-transfection experiments demonstrated that c-fos and c-jun could repress androgen receptor-mediated gene induction. The above studies suggest that (1) the repression of androgen induction of gene expression by TPA-activated PKC is at least in part due to overexpression of c-jun and c-fos and (2) PKC may be a negative growth regulator in prostate cells.

Androgen induction of a human prostate-specific kallikrein, hKLK2: characterization of an androgen response element in the 5' promoter region of the gene.

The human prostate-specific kallikreins, human glandular kallikrein-1 (hKLK2) and prostate-specific antigen (hKLK3), have been shown to be regulated by androgens. To determine whether the androgen induction of these genes is transcriptionally regulated via an androgen response element, an hKLK2 promoter DNA fragment was linked to a promoterless chloramphenicol acetyltransferase (CAT) reporter gene and cotransfected with an androgen receptor expression vector in an androgen receptor-less human prostate cell line, PC-3. Dose response and steroid specificity experiments showed that the hKLK2 promoter confers androgen receptor-mediated gene induction in a ligand-specific manner. Moreover, 5' deletion constructs of the hKLK2 promoter DNA and internal deletion constructs devoid of the 5' half-site of the putative androgen responsive element (ARE) were used to show that the putative ARE is indeed acting as a functional ARE in prostate cells. In addition, multiple AREs from both hKLK2 and hKLK3 were able to reconstitute androgenic induction, further strengthening the argument that the AREs are functional. Although previous studies have shown that hKLK3 mRNA is expressed at a higher level than that of hKLK2, our results suggest that the hKLK2 ARE may have higher androgenic inducibility than the hKLK3 ARE. These results suggest that other cis-acting elements may be involved in coordinating in vivo androgenic induction of hKLK2 and hKLK3 genes.

The importance of N- and O-linked oligosaccharides for the biosynthesis and in vitro and in vivo biologic activities of erythropoietin.

Erythropoietin (EPO) plays a critical role in stimulating the proliferation and differentiation of erythroid precursor cells. EPO is heavily glycosylated with three asparagine (N)-linked tetraantennary oligosaccharides that may contain N-acetyl-lactosamine repeats and a single serine (O)-linked oligosaccharide. EPO expressed in Chinese hamster ovary cells exhibits biologic properties and amino acid and carbohydrate composition similar to natural urinary EPO. The importance of the complex N-linked and the O-linked carbohydrate was studied by expressing EPO in cells that are deficient in UDP-galactose/UDP-N-acetylgalactosamine 4-epimerase activity. In these cells, the ability to add galactose and N-acetylgalactosamine to glycoproteins can be controlled by the addition of these sugars to the culture medium. The results demonstrate that a block in O-linked glycosylation and/or the ability to process N-linked carbohydrate to completion does not alter EPO secretion. EPO produced without O-linked carbohydrate exhibits normal in vitro and in vivo biologic activity and in vivo clearance. However, EPO produced with incompletely processed N-linked oligosaccharides exhibits normal in vitro activity but is at least 500-fold less effective in stimulating erythropoiesis in vivo. Studies on the survival of bioactive EPO remaining in the circulation demonstrated that EPO with incomplete N-linked oligosaccharides exhibits a sevenfold increased rate of clearance. However, this increased clearance may not fully account for the 500-fold loss of in vivo activity. These results suggest a potentially important unique requirement for appropriate complex N-linked oligosaccharides for the intrinsic biologic activity of EPO in vivo.

Source of the aeroallergen of soybean dust: a low molecular mass glycopeptide from the soybean tela.

Airborne soybean allergens in the dust generated during the unloading of soybeans in the harbor caused asthma epidemics in Barcelona, Spain. The major allergen causing the epidemics was a glycopeptide less than 14 kd molecular mass abundant in soybean dust. This allergen occurs in all parts of the soybean plant at all stages of growth, but the telae (hulls) and pods are by far the richest source. Small amounts of a similar cross-reacting allergen are found in some other grain dusts. The botanical function and significance of this soybean plant component is not known nor is the potential for airborne dispersion of this allergen at other grain-handling sites.

Ragweed-specific IgA in nasal lavage fluid of ragweed-sensitive allergic rhinitis patients: increase during the pollen season.

Because the secretions of asthma and rhinitis contain toxic eosinophil granule proteins and because secretory IgA is the most potent immunoglobulin stimulus for eosinophil degranulation, we measured eosinophil-derived neurotoxin and ragweed-specific IgA and IgE antibodies in nasal lavage before and during the ragweed pollen season in 44 hay fever patients. We found IgA antibody in nanogram/milliliter concentrations before the season and rising 20-fold by the end of the season. IgE antibody was present in picogram/milliliter concentrations and did not change. Eosinophils and eosinophil-derived neurotoxin also increased. We conclude that IgA is the predominant antibody in allergic nasal secretions and increases with allergen exposure. The hypothesis that secretory IgA antibody-allergen complexes contributes to allergic inflammation by stimulating eosinophil degranulation warrants further study.

Translational control mediated by eucaryotic initiation factor-2 is restricted to specific mRNAs in transfected cells.

The translational efficiency of mRNA molecules transcribed from plasmid DNA transfected into COS-1 monkey cells can be increased 10- to 20-fold by the coexpression of the adenovirus virus-associated RNAs I and II. Experiments described here demonstrate a similar increase in translational efficiency by the addition of 2-aminopurine, an inhibitor of double-stranded RNA-activated protein kinase, to the culture medium. Both virus-associated RNA and 2-aminopurine presumably exert their effect by alteration of the functional level of eucaryotic initiation factor-2. The translational stimulation mediated by both means is shown to be restricted to the plasmid-derived mRNAs because there is no qualitative or quantitative alteration in host protein synthesis. The results are consistent with models invoking a localized activation of double-stranded RNA-activated kinase leading to a translational block.

Human CSF-1: molecular cloning and expression of 4-kb cDNA encoding the human urinary protein.

A 4-kilobase complementary DNA (cDNA) encoding human macrophage-specific colony-stimulating factor (CSF-1) was isolated. When introduced into mammalian cells, this cDNA directs the expression of CSF-1 that is structurally and functionally indistinguishable from the natural human urinary CSF-1. Direct structural analysis of both the recombinant CSF-1 and the purified human urinary protein revealed that these species contain a sequence of at least 40 amino acids at their carboxyl termini which are not found in the coding region of a 1.6-kilobase CSF-1 cDNA that was previously described. These results demonstrate that the human CSF-1 gene can be expressed to yield at least two different messenger RNA species that encode distinct but related forms of CSF-1.

Translational efficiency of polycistronic mRNAs and their utilization to express heterologous genes in mammalian cells.

The translation of polycistronic mRNAs in mammalian cells was studied. Transcription units, constructed to contain one, two or three open reading frames (ORFs), were introduced stably into Chinese hamster ovary cells and transiently into COS monkey cells. The analysis of mRNA levels and protein synthesis in these cells demonstrated that the mRNAs transcribed were translated to generate multiple proteins. The efficiency of translation was reduced approximately 40- to 300-fold by the insertion of an upstream ORF. The results support a modified 'scanning' model for translation initiation which allows for translation initiation at internal AUG codons. High-level expression of human granulocyte-macrophage colony stimulating factor was achieved utilizing a vector that contains a polycistronic transcription unit encoding an amplifiable dihydrofolate reductase marker gene in its 3' end. Thus, polycistronic expression vectors can be exploited to obtain high-level expression of foreign genes in mammalian cells.

A large region (approximately equal to 95 kDa) of human factor VIII is dispensable for in vitro procoagulant activity.

Factor VIII (antihemophilic factor) is a high molecular weight plasma glycoprotein that participates in the blood clotting cascade. The recent cloning and sequence analysis of the cDNA encoding human factor VIII revealed an obvious domain structure for the protein, which can be represented as A1-A2-B-A3-C1-C2. We now report the DNA sequence analysis of porcine exons encoding the entire B domain and part of the A2 and A3 domains. We found an unusually high degree of porcine-human amino acid sequence divergence in the B region compared with the limited sequence available for other regions of the porcine factor VIII molecule. In addition to sequence divergence, there are numerous gaps in the porcine B domain totalling over 200 amino acids. Recombinant DNA techniques were used to effect the removal of large segments of DNA encoding the B domain from the full-length human factor VIII cDNA. These constructs directed the synthesis of biologically active factor VIII when introduced into mammalian cells despite the deletion of up to 38% of the factor VIII molecule.

Selection and amplification of heterologous genes encoding adenosine deaminase in mammalian cells.

We demonstrate that an adenosine deaminase (ADA) cDNA gene can function as a dominant selectable and amplifiable marker for gene transfer experiments in mammalian cells. Cells that incorporate the gene can be selected by growth in the presence of low concentrations of the ADA inhibitor 2'-deoxycoformycin with cytotoxic concentrations of adenosine or its analogue 9-beta-D-xylofuranosyl adenine. The DNA copy number of the transfected ADA minigene in the isolated transformants of Chinese hamster ovary cells can be amplified greater than 100-fold by growth in ADA selection media and increasing concentrations of 2'-deoxycoformycin. This selection scheme may allow for the introduction and subsequent amplification of heterologous DNA in a variety of mammalian cells.