Summer Colloquium 2003: Abstracts

Cytokinesis can complete in the absence of a spindle midzone

Koen Verbrugghe

In animal cells, cytokinesis, the physical process that partitions the duplicated genome and cytoplasm into two daughter cells, is driven by constriction of an acto-myosin ring. This ring pulls the plasma membrane towards the spindle midzone leaving a channel connecting the two daughter cells that must be cleared and then broken in a coordinated manner in order to completely separate the cells. Current models suggest that the spindle midzone, a bundle of anti-parallel microtubules and associated proteins that is the remnant of the mitotic spindle, is important for this process but its specific role is poorly understood. Mutations in the zen-4 and cyk-4 genes of Caenorhabiditis elegans disrupt the spindle midzone and give rise to failures in the terminal phase of cytokinesis. We show that loss-of-function of spd-1 causes similar midzone disruptions, but cytokinesis generally completes. SPD-1 is a conserved microtubule bundling protein located in the spindle midzone. ZEN-4 and CYK-4 localize to the ingressing furrow in addition to the spindle midzone in wild-type embryos; however, only furrow localization is preserved in embryos depleted of SPD-1. We conclude that the localization of ZEN-4 and CYK-4 to the furrow is sufficient to allow for the completion of cytokinesis.

The maize FLORICAULA/LEAFY homolog, zfl2, is a candidate gene for a maize domestication QTL

Kirsten Bomblies

Maize and its wild ancestor teosinte show dramatic morphological differences despite having diverged less than ten thousand years ago. Our lab uses this system to investigate the genetics underlying morphological evolution under selection. One of several major differences between maize and teosinte is an increase in complexity in the structure of the ear (female inflorescence). Quantitative Trait Loci (QTL) for these differences have been mapped, and a major-effect "domestication QTL" affecting this trait was identified on maize chromosome 2. We are currently investigating the candidacy of one of two duplicate maize FLORICAULA/LEAFY homologs, zfl2, for this QTL. We have used a number of approaches including mutant analysis, quantitative analysis in segregating populations, nucleotide diversity and molecular evolution studies, and transgenic overexpression studies to examine the candidacy of zfl2 for this QTL. Future experiments have been designed to address whether zfl2 function is necessary for the QTL effect.

gon-1 interactors in C. elegans organogenesis

Dan Hesselson

Animal organogenesis requires coordinated cell division, differentiation and migration to assemble the mature organ. In C. elegans, the migrations of gonadal leader cells shape the gonad. gon-1, a gene previously characterized in our lab, is required for gonadogenesis. Specifically, it promotes the migration of the gonadal leader cells and the expansion of the tissue in all axis. gon-1 encodes a secreted metalloprotease, that is a member of the ADAMTS family of enzymes. I am interested in the molecular functions of GON-1. I have taken two approaches to learn more about its role in gonadogenesis. First, I am characterizing a putative substrate, which was identified through physical interaction studies. The second approach is a genetic screen to identify other genes that suppress the gon-1 phenotype. I will discuss my progress in these areas.

Effect of Fnr on E.coli K12 MG1655 anaerobic growth revealed by Microarray experiment

Yisheng Kang

Upon depletion of oxygen, Escherichia coli turn off the expression of enzymes critical for aerobic respiration, such as component of the TCA cycle and terminal oxidase complexes. At the same time, genes responsible for anaerobic respiration or fermentation are turned on. The global regulator FNR which is an oxygen-responsive transcriptional regulator functions as a major control system for the switch from aerobic to anaerobic metabolism. To identify the whole component of the Fnr regulon, we compared the genome-wide transcription profile of isogenic wild-type and fnr deletion Escherichia coli strains by affymetic microarray under both aerobic and anaerobic conditions. The microarray measurements not only confirmed the expression pattern of the majority of known FNR regulated genes but also allow identify 3X more FNR regulon genes than previously indicated. The new identified Fnr regulon genes falls in 9 major function groups: energy metabolism, central intermediary metabolism, detoxification, osmotic adaptation, chemotaxis and mobility, surface structures, 2'-Deoxyribonucleotide metabolism, Transport of small molecules, and not classified. 94 of them are unknown genes.

Characterization of an Arabidopsis mutant with altered root wave pattern

Kai (Billy) Hung

A screen for seedlings with altered root wave pattern on titled agar surface was conducted in an Arabidopsis thaliana population mutagenized by a modified Ac/Ds system from maize. The mutant wvc16 was identified and isolated for its compressed root wave pattern. Besides defects in root waving on tilted agar surface, the mutant also exhibits other morphological defects such as curly rosette and cauline leaves, kinked siliques, curly pedicles, and extended petals. In addition, the stems of wvc16 curl, some of which then uncurl over time. The mutant responds to gravity in a manner comparable to wild type plants. Comparison of root growth rate under the influence of exogenous IAA and 2,4-D at various concentrations between wild type and mutant revealed no significant difference. Using adapter PCR, the Ds insertion site was identified within the first predicted exon of a novel and hypothetical gene. Northern blot analysis and RT-PCR results both suggested that the mutant is over-expressing a mutant form of WVC16. Genetic, molecular, and physiological studies are being carried out to further characterize the nature of the wvc16 mutation.

Elucidating the role of dlg-1 in formation and function of epithelial junctions

Christopher Lockwood

We are interested in the genes involved in formation and function of epithelial junctions. Recently two proteins have been identified which localize and act at the C. elegans septate like junction. AJM-1, a novel coiled coil protein, and DLG-1, a homologue of Drosophila Discs large, have been shown to colocalize to a domain of apical junctions distinct from the catenin-cadherin complex. Loss of function studies with these genes demonstrate their requirement for proper junctional integrity. Individuals deficient for either protein arrest during elongation and have disrupted apical junctions at the level of electron microscopy. In addition, these two genes have been shown to interact both genetically and physically. A genetic interaction is seen by disruption of A JM-1 localization along the apical junction in the absence of DLG-1. A physical interaction is seen by the binding of AJM-1 to the N-terminal half of DLG-1 in yeast 2-hybrid and GST pull down experiments.

In order to further characterize the formation and function of the DLG-1/AJM-1 domain we are undertaking a series of structure function experiments. A dlg-1 deletion mutant provided by the C. elegans knockout consortium confers a loss of function background in which rescue with constructs deleted for domains of dlg-1 can be performed. Constructs are being tested for their ability to rescue dlg-1 mutant phenotypes with respect to lethality and AJM-1 localization. In parallel we are performing directed 2-hybrid experiments with deletion constructs of dlg-1 and ajm-1 to further map the physical interaction domain. Experiments have allowed us to identify a N-terminal region essential for the AJM-1/DLG-1 interaction. These results along with initial characterization of candid ate DLG-1 interactors will be presented.

Modulation of the actin cytoskeleton during C. elegans epithelial morphogenesis.

Mark Sheffield

During C. elegans embryogenesis, the hypodermis undergoes three major morphogenetic processes that lead to formation of the worm's tubular shape: dorsal intercalation, ventral enclosure, and elongation. It has previously been demonstrated that proper completion of all three of these events requires an intact actin cytoskeleton in the hypodermis, but the mechanisms involved in regulation of actin during epithelial morphogenesis are not yet entirely clear. We have taken a candidate gene approach to examine whether molecules known to be involved in actin dynamics have roles in epithelial morphogenesis. Members of the WASP family have been shown in vitro to act downstream of CDC-42 as activators of the Arp2/3 complex to promote actin branching during cell locomotion. Surprisingly, animals null for wsp-1, a C. elegans homolog of the WASP family, undergo normal embryogenesis. Drosophila Enabled is known to affect epithelial morphogenesis, and the vertebrate Ena family members have been implicated in actin remodeling. unc-34 encodes an ortholog of Ena and has been shown to function in cell migration in C. elegans. As with wsp-1, a null allele of unc-34 results in virtually no defects in embryogenesis. However, RNAi of wsp-1 in unc-34(gm104) animals yields embryonic lethality. These embryos appear to be specifically defective for certain hypodermal cell migrations that occur during ventral enclosure, but not those during dorsal intercalation. In addition to its synergy with wsp-1, unc-34 has other morphogenetic roles. RNAi for a homolog of the WASP relative SCAR/WAVE generates hypodermal defects and lethality, and performing this experiment in an unc-34(gm104) background enhances this phenotype to yield embryos that are completely defective in morphogenesis. Finally, unc-34 enhances morphogenetic phenotypes associated with mutation in the cadherin/catenin complex and localizes along the junction, similar to results from Drosophila. Clearly, UNC-34 plays multiple roles during embryogenesis, but how this occurs is presently unclear. Future experiments are aimed at beginning to elucidate these roles and to uncover the molecules involved.

Intracellular membrane trafficking is involved in spindle alignment in C. elegans early embryos

Haining Zhang

Positioning of the mitotic spindle in asymmetric divisions is crucial for the proper inheritance of segregated cytoplasmic determinants to daughter cells. How the spindle detects the cellular polarity and aligns along it are not well understood. We use C. elegans to study these problems as its zygote undergoes two sequential asymmetric divisions. At each division, the P0 and P1 spindles are rotated to lie along the polarity axis.

The role of intracellular membrane trafficking in spindle alignment was explored first by applying Brefeldin A (BFA), a potent secretion inhibitor, to the early embryos of C. elegans. Embryos treated with this drug show vigorous rocking movements during nuclear-centrosomal centration as well as P0 spindle orientation defects and failure of cytokinesis. The spatial distribution of polarity markers, such as PAR-2 and PIE-1 is not affected in these embryos, indicating that the effect of BFA is downstream of cellular polarity. To address which parts of the intracellular membrane trafficking are required for spindle alignment, we used RNAi to disrupt several members of the Rab superfamily which regulates the specificity of vesicle trafficking. We found that several Rabs, including RAB-5, RAB-6.1/RAB-6.2, and RAB-11 are required for spindle alignment in P0 and P1. In particular, when rab-11 is knocked down by RNAi, rotation of P0 and P1 spindles are inhibited, and the anaphase spindle position is also defective. The C. elegans homolog of mammalian rab11BP (the effector of Rab-11), R03E1.1, gives viable progeny when disrupted by RNAi. While RNAi of npp-20 (the C. elegans homolog of mSec13), the partner of rab11BP, exhibits similar spindle alignment defects as to rab-11 RNAi. The relationship between RAB-11 and NPP-20 will be further determined by yeast two-hybrid interaction and cellular localization studies.

The Three Dishevelleds Act Redundantly in Spindle Positioning in the Early C. elegans Embryo

Timothy Walston

Cell division in the early embryo is a highly coordinatled event involving several signaling pathways. Components of Wnt signaling play a role in the positioning of the mitiotic spindle in the EMS and ABar blastomeres. Surprisingly, no Dishevelleds have been reported as having defects in spindle orientation in either cell. I will discuss redundancy between the three C. elegans Dishevelleds to orient the spindle in both blastomeres. Analysis of other members of canonical and non-canonical Wnt signaling pathways show that both a non-canonical and a canonical-like pathway control spindle orientation and spindle rotation timing in the early embryo.

spd-3 is a novel gene required for mitotic spindle alignment in C. elegans

Maria Vidal Dinkelmann

In order to develop a complex organism, cells must have the ability to divide asymmetrically to produce two daughter cells of differentdevelopmental potential. A cell accomplishes this by polarizing cytoplasmic components to opposite ends such that, upon cleavage, each daughter inherits distinct components. This segregation requires coordination of the axis of polarity and the cleavage plane. The cleavage plane is specified by the alignment of the mitotic spindle. To better understand the molecular mechanism of spindle alignment we are studying a temperature sensitive, maternal effect, lethal mutation in the spd-3 gene of C. elegans. Cytological analysis reveals that the spd-3(oj35) mutant is defective in nuclear and spindle positioning. During the first mitosis the spindle fails to align along the anterior/posterior axis leading to abnormal cleavage configurations. spd-3(oj35) mutants also exhibit a failure to extrude polar bodies, which could be attributed to a misaligned meiotic spindle. Post-embryonic defects, including uncoordination and sterility, are consistent with cell division defects indicating that SPD-3 may be involved in spindle alignment in all tissues. The smallest genomic DNA fragment that rescues spd-3(oj35) consists of an operon containing the predicted ORFs H34C03.2 and H34C03.1. The spd-3(oj35) strain contains no mutations in H34C03.2, and a mutation converting Leucine130 to Phenylalanine in H34C03.1 indicating that H34C03.1 is spd-3. It is possible that SPD-3 is a novel component, or involved in the regulation of the dynactin complex since spd-3(oj35) and dynactin RNAi embryos show similar phenotypes. In studying genes such as spd-3 we hope to reach a better understanding of factors involved in spindle alignment which is essential for defining the plane of cytokinesis and ensuring proper asymmetric cell division.

The Drosophila Antenna: from Imaginal Disk to Auditory Organ

Dominic Ebacher

The Drosophila antenna is a highly sophisticated structure that functions in both olfaction and audition. Previous studies have identified Homothorax and Distal-less, two homeodomain transcription factors, as required for specification of antennal identity. The cut gene, which encodes a homeodomain transcription factor, is an antennal-specific target of Homothorax but not Distal-less. Cut is the Drosophila homolog of the mammalian CAAT-displacement protein (CDP), a potent transcriptional repressor. Cut is required for normal development of external mechanosensory structures and the Malphigian tubules (Drosophila kidney analogs); however, the role of Cut in the antenna has not been characterized previously. I have employed the FLP/FRT system to generate cut null clones in the antenna and have used these clones to investigate the potential role of Cut in auditory organ development. cut mutant flies show disruption of the chordotonal subunits that make up the Johnson's Organ (JO), an antennal structure essential for fly audition. Electrophysiological experiments, carried out by our collaborators, confirm that flies with cut null antennae are deaf. I find that cut acts in parallel to the genes atonal and spalt, that encode two other transcription factors required for JO differentiation. I conclude that Cut is required for JO differentiation and/or function, but not for antennal fate specification. Based on biochemical studies of the vertebrate homologs, I speculate that Cut functions by repressing as yet unidentified genes in the developing antenna.

Generation and Characterization of Overexpressing Mitochondrial HSP70 Transgenic Mice

Radhika Puttagunta

Emerging research indicates that mitochondria play a central role in the aging process. The epicenter of oxidative damage is the mitochondrion, where radical oxygen species (ROS) are created as byproducts of energy metabolism. ROS can damage proteins, lipids and DNA. Severe damage can lead to mitochondrial dysfunction, such as alterations in mitochondrial structure, membrane potential, release of cytochrome c from the inner membrane matrix and induction of the apoptotic pathway. Age is a risk factor for coronary heart disease, which is the most common form of heart disease. Ischemia and reperfusion-induced damage (I/R injury) is a result of heart disease and causes ROS damage and mitochondrial dysfunction in the myocardium.

Prior studies indicate an increased heat shock response with aging, which is likely a response to accumulated protein damage with aging. Cytosolic inducible HSP70 (heat shock protein of 70kDa) has shown a protective role against protein damage, oxidative stress, I/R injury, apoptosis and mitochondrial dysfunction in cell lines, drosophila, and/or mice. Overexpression of Hsp70 protects against oxidative stress in cell lines and Drosophila by avoiding protein denaturation. Transgenic flies containing a 10% increase in cytosolic HSP70 have shown longer lifespans than wildtype control. The HSP70 family is large and highly conserved throughout evolution. First discovered as a response to heat shock, HSP70 is now known to have multiple functions. Of the human HSP70 family, mitochondrial HSP70 (MTHSP75) is most homologous to bacterial Hsp70 (DnaK). The MTHSP75 gene is constitutively transcribed in the nucleus but localized to the inner membrane of the mitochondria, where it serves as an ATPase transporter of proteins into the mitochondria. Once proteins arrive inside the mitochondria, MTHSP75 is able to fold them into their proper conformation. It also serves as a chaperone of proteins synthesized on mitochondrial ribosomes, and can help degrade misfolded proteins within the mitochondrion. Importantly, under stress MTHSP75 refolds denatured proteins and prevents aggregation.

Our hypothesis is that if oxidative damage is centered at the mitochondrion, and HSP70 overexpression reduces oxidative damage by refolding damaged proteins, then transgenic mice overexpressing MTHSP75 should be protected against oxidative protein damage and show retarded aging and protection against I/R injury in the heart. In order to define the role of MTHSP75 in the aging process, we have created a transgenic mouse overexpressing human MTHSP75. Northern analysis shows expression in muscle, heart and brain. Western analysis and immunohistochemistry confirm these results. Electron microscopy of the heart has shown that the protein is localizing to the mitochondria. Preliminary results from functionality assays to demonstrate protection by the MTHSP75 transgene in cardiomyocyte cell lines to hydrogen peroxide damage indicate a possible protection against hydrogen peroxide damage in the transgenic lines versus wildtype cardiomyocytes. Further assays will be run to test protection of the MTHSP75 transgene against hydrogen peroxide, as well as, varying stressors such as doxorubicin and paraquat both in cell lines and in vivo. Future experiments will include I/R injury stressing experiments and aging timepoint microarray analysis of the MTHSP75 transgenic mice.