Transcriptomal analysis of failing and nonfailing human hearts.

Heart failure is a multifactorial disease that may result from different initiating events. To contribute to an improved comprehension of normal cardiac function and the molecular events leading to heart failure, we performed large-scale gene expression analysis of failing and nonfailing human ventricle. Our aim was to define and compare expression profiles of 4 specific pathophysiological cardiac situations: 1) left ventricle (LV) from nonfailing heart; 2) LV from failing hearts affected by dilated cardiomyopathy (DCM); 3) LV from failing hearts affected by ischemic CM (ICM); 4) right ventricle (RV) from failing hearts affected by DCM or ICM. We used oligonucleotide arrays representing approximately 12,000 human genes. After stringent numerical analyses using several statistical tests, we identified 1,306 genes with a similar expression profile in all 4 cardiac situations, therefore representative of part of the human cardiac expression profile. A total of 95 genes displayed differential expression between failing and nonfailing heart samples, reflecting a reversal to developmental gene expression, dedifferentiation of failing cardiomyocytes, and involvement of apoptosis. Twenty genes were differentially expressed between failing LV and failing RV, identifying possible candidates for different functioning of both ventricles. Finally, no genes were found to be significantly differentially expressed between failing DCM and failing ICM LV, emphasizing that transcriptomal analysis of explanted hearts results mainly in identification of expression profiles of end-stage heart failure and less in determination of expression profiles of the underlying etiology. Taken together, our data resulted in identification of putative transcriptomal landmarks for normal and disturbed cardiac function.

Exclusion of muscle specific actinin-associated LIM protein (ALP) gene from 4q35 facioscapulohumeral muscular dystrophy (FSHD) candidate genes.

Facioscapulohumeral muscular dystrophy (FSHD) is an autosomal dominant disorder for which no candidate gene has yet been identified. The gene corresponding to one of the novel human cDNAs that we cloned on the basis of a muscle restricted expression pattern [Piétu G, Alibert O, Guichard B, et al. Genome Res 1996;6:492-503] was mapped in the region of the FSHD1A genetic locus, i.e. one of the loci involved in this muscular dystrophy. The corresponding encoded protein contains a PDZ and a LIM domain, two protein-protein interaction domains, and was very recently shown to bind alpha-actinin-2 and was named ALP (actinin-associated LIM protein) [Xia H, Winokur S, Kuo W, Altherr M, Bredt D. J Cell Biol 1997;139:507-515]. We raised a specific polyclonal anti-ALP serum against an ALP recombinant polypeptide to evaluate the size, level of expression and subcellular localization of ALP in three patients, clearly diagnosed with FSHD disease. Quantitative or qualitative alterations of ALP expression have not been detected in any of them, thus prompting us to exclude ALP as a FSHD gene candidate.

Molecular cloning and functional expression of a novel human gene encoding two 41-43 kDa skeletal muscle internal membrane proteins.

Systematic analysis of gene transcript repertoires prepared from libraries made with various specific human tissues permitted isolation of many partially sequenced cDNA clones. A few of these represented novel genes with limited or no similarity to known genes from humans or other species. The present study set out to isolate and sequence the full-length cDNA corresponding to one of these novel human transcripts, and identify the corresponding protein product at the subcellular level. Current sequence analyses have revealed that the protein contains a hydrophobic N-terminal segment and an internal leucine-zipper motif. Numerous sites of putative post-translational modifications, such as N-linked glycosylation, myristoylation and phosphorylation sites, were also identified. Using one monoclonal antibody raised against a recombinant fragment, two different 41-43 kDa proteins were detected in human skeletal muscle, heart and placenta homogenates at various ratios. Both immunodetected protein products of the novel human gene were distributed in the transverse tubules and/or near the junctional sarcoplasmic reticulum within skeletal muscle cells. Both proteins had physical properties believed to be attributable to integral membrane components. Finally, the GENX-3414 gene was chromosomally localized at position 4q24-q25.

CholecystokininB receptor from human Jurkat lymphoblastic T cells is involved in activator protein-1-responsive gene activation.

The aim of this study was to analyze the role of cholecystokinin (CCK(B)) receptor in human lymphoblastic Jurkat T cells. We investigated the trophic effect resulting from activation of such a receptor by using the reporter gene strategy. For this purpose, we transiently transfected Jurkat T cells with the reporter plasmid p[(TRE)3-tk-Luc] and found that CCK-8 was able to dose-dependently induce luciferase expression related to activator protein-1 (AP-1) activation with a maximal response identical to that obtained with compounds known to activate AP-1 complex (quantitatively, the same level of induction was obtained with 1 nM 12-O-tetradecanoylphorbol-13-acetate, 100 microM diacylglycerol, or 4 nM epidermal growth factor). The involvement of the CCK(B) receptor in such a stimulation was demonstrated by the inhibiting effect of the selective CCK(B) receptor antagonist PD-135,158. This effect was confirmed in COS-7 cells transfected with the cDNA of CCK(B) receptor cloned from Jurkat T cells. To better understand the AP-1-dependent luciferase expression in Jurkat T cells, we tested two specific inhibitors of serine/threonine phosphatases-1 and -2A: okadaic acid and calyculin A. These compounds strongly increased the phorbol-12-myristate-13-acetate response, whereas we have not observed a contribution of phosphatase inhibitors on a CCK-8-induced luciferase activity. To confirm that CCK(B) receptors are involved in AP-1 response, we investigated the CCK-8 effect on interleukin-2 expression, a natural endogenous gene regulated by several factors, including AP-1. In Jurkat T cells activated by phorbol-12-myristate-13-acetate and phytohemagglutinin, CCK-8 induced IL-2 expression. This induction was abolished by PD-135,158. Our results indicate that CCK-8 exerts a trophic effect in Jurkat T cells through stimulation of CCK(B) receptors by modulation of expression of AP-1-regulated genes.

The cardiac myosin heavy chain Arg-403-->Gln mutation that causes hypertrophic cardiomyopathy does not affect the actin- or ATP-binding capacities of two size-limited recombinant myosin heavy chain fragments.

Our aim was to investigate the potential functional consequences of myosin heavy chain (MHC) mutations identified in patients with familial hypertrophic cardiomyopathy. We observed the presence of a mutated beta-MHC mRNA in a formalin-fixed paraffin-embedded myocardial tissue of a proband from family A, which Geisterfer-Lowrance et al. [Geisterfer-Lowrance, Kass, Tanigawa, Vosberg, McKenna, Seidman and Seidman (1990) Cell 62, 999-1006] identified as carrying the Arg-403 to Gln mutation. Recombinant DNA methods were then used to obtain size-limited, soluble and undenatured fragments of mutated myosin subfragment 1 focused around the 403 mutation. The present analysis indicated that the 403 mutation did not quantitatively alter the actin- or ATP-binding capacities of two 246-residue or 524-residue-long recombinant MHC fragments containing this mutation. The absence of any apparent impact of the 403 mutation in the recombinant MHC fragments on interactions between actin and ATP is discussed in relation to numerous biochemical and structural reports which demonstrate the crucial role of the central MHC segment, where the 403 mutation occurs, in myosin functions.

Mapping of the actomyosin interfaces.

Recombinant DNA methods were used to obtain soluble, undenatured fragments of the heavy chain of myosin subfragment 1 (S-1). These fragments were of preselected lengths and could include protease-sensitive segments that are destroyed when other preparation methods are used. Actin binding by each of the three contiguous segments (residues 1-248, 249-524, and 518-722, essentially spanning the entire S-1 heavy chain) was demonstrated. ATP binding, comparable to that of native S-1, was obtained only with a segment consisting of residues 1-524. Competition among the various fragments for actin was also studied. The data are discussed in relation to the recently reported resolved structure of S-1 [Rayment, I., Rypnieski, R. W., Schmidt-Bäse, K., Smith, R., Tomchick, D. R., Benning, M. M., Winkelmann, D. A., Wesenberg, G. & Holden, H. M. (1993) Science 261, 50-58].

Probing myosin light chain 1 structure with monoclonal antibodies.

Five monoclonal antibodies that react with different regions of myosin light chain 1 from human ventricular myocardial muscle were used to obtain information on interactions between the light chain 1 and heavy chains and generally on the tertiary structure of the light chain 1 within the myosin head. We performed Western blot assays of the five antibodies with myosins from different cardiac and skeletal muscles, with different proteolytic fragments of bovine ventricular myosin light chain 1 (LC1) and to different recombinant fragments of human ventricular LC1 and rat fast skeletal light chain LC1/LC3. The five antibodies were mapped in three different regions of the light chain 1: two antibodies mapped within the first eight amino-terminal residues, two between residues 71 and 74, and one between residues 129 and 134. The apparent dissociation constants of the last three antibodies, determined by antibody-antigen equilibria in solution, were lower than when isolated light chains were used as antigens. It is probable that the corresponding amino acids involved in the antibody epitopes were either involved in interactions between the light and heavy myosin subunits, or somehow hindered by the myosin heavy chain bulk. In contrast, the apparent dissociation constants measured for both other antibodies were higher when myosin, rather than isolated light chains, was used as antigen. Thus LC1 fixation to heavy chains within the myosin molecule induced conformation changes at the amino-terminal end of the light chain 1. No difference in the accessibility of this mobile LC1 segment was detected in the presence of actin. Finally, observed differences in epitope accessibility on the light chain LC1 in myosin, as compared with chymotryptic subfragment 1 (SF1), indicated conformational differences between native myosin and extensively studied SF1 molecules.

Expression of human beta-myosin heavy chain fragments in Escherichia coli; localization of actin interfaces on cardiac myosin.

A cDNA clone coding for an internal fragment of slow-cardiac beta-myosin heavy chain was isolated from a lambda gt10 human skeletal muscle library. Six overlapping cDNA subclones, which span myosin heavy chain subregions and presumably interact with actin, were derived from this clone, fused to a beta-galactosidase vector and expressed in Escherichia coli. Three of the subclones were obtained by PCR (polymerase chain reaction) which enables gene or cDNA fragments to be amplified independently of preexisting restriction sites. Initially, various experiments were carried out using a long MHC (myosin heavy chain) fusion protein containing the 50 kDa-20 kDa connecting region, the whole 20 kDa region and the short subfragment 2 region. This MHC fusion protein was chemically or proteolytically cleaved in the same conditions as the native myosin molecule. Whole and truncated forms of the MHC fusion protein were separated on polyacrylamide gels, electroblotted on nitrocellulose sheets and renatured. They were then assayed in overlay experiments with F-actin and/or myosin light chains in solution. Specific antibodies were used to detect interactions between heavy chain fragments and F-actin or light chains. We thus observed that one long heavy chain fragment synthesized by E. coli behaved like proteolytic or chemical MHC preparations made from native myosin molecules. Two chymotryptic fragments of the MHC fusion protein, which are soluble at low ionic strength, cosedimented with F-actin in solution. Our results demonstrate that, in actin overlay experiments with whole fusion proteins, interactions seem to be due to the heavy chain fragment, not to the bacterial component. All interactions were non ATP-sensitive. We further investigated the possible participation of the six recombinant MHC fragments in contributing to the actomyosin interfaces on the 50 kDa-20 kDa regions of the human cardiac beta-MHC. The present procedure, which enables the synthesis of any MHC fragment independent of any protease site, is a powerful new tool for studying structure-function relationships within the myosin molecule family.

Characterization of a casein kinase which interacts with the rabbit progesterone receptor. Differences with the in vivo hormone-dependent phosphorylation.

Previous in vivo studies have shown that the rabbit progesterone receptor undergoes two phosphorylation reactions: one basal and a second one which is hormone-dependent. We report here on the presence and characteristics of a kinase activity found in receptor preparations highly purified by immunoaffinity chromatography. 1. This kinase activity is not due to the receptor molecule itself since the two proteins may be separated by several chromatographic and immunological methods. 2. The presence of the kinase in receptor preparations is not an artefact of the purification procedure. The kinase binds to the receptor as shown by coelution in immunoaffinity experiments and during various chromatographies. This interaction probably takes place in vivo and is not artefactually formed during solubilization of the receptor since the kinase also copurifies with receptors isolated from the uterine nuclei of progestin-treated rabbits. 3. This enzyme may be classified as a casein kinase since it readily phosphorylates the latter substrate (Km approximately equal to 0.15 mg/ml) and is not regulated by cyclic nucleotides, Ca2+ and calmodulin or phospholipids. Its classification as a casein kinase I or II is difficult since on the one hand it is inhibited by heparin, activated by polyamines and may use both ATP and GTP, but on the other hand it modifies only serine residues, and is not inhibited by heparin when the receptor itself is employed as a substrate. 4. The kinase which copurifies with the receptor does not mimic in vitro the effects of the hormone-dependent phosphorylation of the receptor observed in vivo: there is no enhancement of kinase activity by the hormone, and the phosphorylated receptor does not exhibit the characteristic "upshift" in its electrophoretic mobility. Thus either this kinase is not the enzyme responsible for the hormone-dependent receptor phosphorylation or, during purification, a factor has been lost which is necessary for retaining the hormone dependency of the reaction.

The nuclear-bound form of the progesterone receptor is generated through a hormone-dependent phosphorylation.

The solubilized ("cytosolic") receptor present in the rabbit uterus in the absence of hormone and the chromatin-bound ("nuclear") receptor obtained after injection of a progestin were compared. Crude cellular extracts were analyzed by immunoblotting and receptors were purified by immunoaffinity chromatography. With both methods it was observed that the electrophoretic mobility of the "nuclear" receptor was slower than that of the "cytosolic" receptor. This difference in mobility appeared to be due to the existence of variably phosphorylated forms of receptor. The phosphorylation reaction was examined in uterine slices. In the absence of hormone the cytosolic receptor was phosphorylated. When hormone was added the phosphorylation of receptor was markedly enhanced and the electrophoretic mobility of the "nuclear" receptor was decreased. These experiments thus show that the receptor in its "cytosolic" form is a phosphoprotein. Under the effect of the hormone the receptor is further phosphorylated on some supplementary site(s). This polyphosphoprotein is the chromatin-bound, putatively active, form of the receptor. In this respect the intracellular progesterone receptor is similar to various membrane receptors for hormones and growth factors which are phosphorylated upon binding of their ligand.

Cloning of a gene expressed in human breast cancer and regulated by estrogen in MCF-7 cells.

Messenger RNAs (mRNAs) were prepared from MCF-7 breast cancer cells grown in the presence of estradiol. Complementary DNAs (cDNAs) were inserted into pBR322 plasmid and a library of 4400 recombinant bacterial clones was prepared. The clones were screened by in situ differential hybridization with cDNAs prepared from RNAs of MCF-7 cells grown either in the presence or absence of estradiol. Several estrogen-induced or estrogen-repressed clones were identified. One of them corresponded to a relatively frequent mRNA (0.8% of recombinant plasmids) of 650 nucleotides. The concentration of this mRNA was increased by estradiol (half maximal induction approximately 0.05 nM) but not by progesterone, dexamethasone, or dihydrotestosterone. Tamoxifen inhibited the effect of estradiol but was devoid of any agonistic activity when administered separately. This messenger was present in biopsies of breast cancer, but not in endometrium or liver. The cloned cDNA was sequenced. An open reading frame was found corresponding to a protein of less than 100 amino acids. A search of data banks showed no identity or marked similarity to previously published DNA or protein sequences, particularly to those of growth factors evoked by some characteristics of the coded polypeptide. The cloned cDNA probe was used to screen a library of Charon 4A phage containing human genomic fragments. Screening of 300,000 phages yielded two different recombinants hybridizing to the cDNA. Southern blot experiments using DNA from recombinant phage, MCF-7 cells, and placenta showed the presence of a unique gene exhibiting a similar restriction pattern in DNAs from malignant and nonmalignant tissues.

Interferon inhibits transformation of mouse cells by exogenous cellular or viral genes.

Specific genes can be introduced into cultured mammalian cells by DNA transfer, sometimes to be stably integrated and expressed. Because interferon inhibits the transformation initiated by DNA viruses, we wondered if it might also affect transformation induced by recombinant plasmids and cellular genes. We show here that in mouse L cells interferon treatment prevents stable integration and expression of the transfected plasmids containing the cloned herpesvirus thymidine kinase (tk) gene, or the dihydrofolate reductase (dhfr) gene from hamster ovary cells. In contrast, interferon does not prevent transient expression of the tk gene in unintegrated form.