l 3‘ . .(5? :1 iatflnxn z... 7. {i .. 1...; ....v...5..!. : r . U.— a I . 3. y 1;: . v7.2: 1. y. .. . v IL. ..l....v..J . 1a.; .. .1 11.34:“: 25‘.“ , ‘ .5.” . lfl..w.!..tgl .23... a . 1.5.3:. on. Er... .:. ‘ 2.! 5. . 1. V 533:; x .. .. I: .u: v x. 23: 11:2... .333 i .11. 2:? $.25? 1:... r5-.. . .29 p , .3. ”a .1. , :‘niix .. In H. . .9». 2.»... it... “fink“.wrrhnp? \ s . 1.3!..l4u2. ] v ,. .l ., . .1 . x . . full: . 21.1.... 3.3 '35:}.sy: 3;: it .. : Jr",£:.: :. THESIS 2220 LIBRARY“ Michigan State Unlverslty PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE mm mm.“ GALECTIN-3: GENE STRUCTURE, REGULATION OF EXPRESSION AND SUBCELLULAR LOCALIZATION By Mark Martin Kadrofske A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry l 999 ABSTRACT GALECTIN-3: GENE STRUCTURE, REGULATION OF EXPRESSION AND SUBCELLULAR LOCALIZATION by Mark Martin Kadrofske Galectin-3 is a B-galactoside-specific lectin that is a pre-mRNA splicing factor. Galectin-3 is coded by a single gene in the human genome and the structure of this gene (LGALS3) has now been determined. LGALS3 is composed of six exons and five introns spanning a total of ~17 kilobases; there are two transcription initiation sites located 52 (+1a) and 50 (+1b) nucleotides upstream of the exon I-intron 1 border; the translation start site is in exon 11. Clues to the regulation of expression of LGALS3 come from functional characterization of the promoter. A genomic fragment encompassing -836 to +141 nucleotides, relative to +la, has significant promoter activity when transiently transfected into HeLa human cervical carcinoma cells and human diploid fibroblasts (HDFs). Serum-starved quiescent HDFs have low promoter activity and low levels of galectin-3 protein. Both promoter activity and protein levels increase following serum addition. Putative serum-responsive activation regions in the promoter have been identified. The subcellular localization of galectin-3 is also dependent on the proliferative state of the cell. Galectin-3 is found in both the nucleus and the cytoplasm of young, proliferating HDF. In contrast, galectin-3 appears excluded from the nuclei of senescent HDF that have lost their replicative capacity through in vitro culture. In heterodikaryons derived from fusion of young and senescent cells, galectin-3 localized in both nuclei is the predominant phenotype. These data suggest that senescent HDF lack factor(s) specifically required for galectin-3 nuclear import. Results from these studies help to provide a basis for determining the molecular mechanisms of proliferation—dependent galectin-3 transcriptional activation and subcellular localization. To my mother and father, and to Gwynne, for their love and support iv ACKNOWLEDGEMENTS I wish to thank my major professor Dr. John L. Wang for his support and guidance. He displayed great patience in me, and for that I am eternally indebted and grateful. I have developed as a scientist under his tutelage through the many scientific discussions and by learning his keen skill in breaking down and attacking a research problem. I've particularly enjoyed his many excellent stories and the enthusiasm in which they were told! I‘ve also enjoyed the many political and sports conversations over the years ..... I still dislike a certain player from Gonzaga and I continue to maintain Isiah was robbed. I also wish to thank the members of my graduate committee for their invaluable advice and suggestions: Drs. Sue Conrad, Pam Green, Justin McCormick, and Steve Triezenberg. I wish to thank Dr. Mel Schindler for his help and suggestions with the confocal microscopy experiments, as well as the many excellent discussions of MSU, U- M, Piston, and Knick basketball. Thanks to Mr. Joe Leykam for his suggestion of using extended PCR to isolate 5’-end genomic clones, and to Dr. Annette Thelen for her encouragement and suggestions during the library screening. Special thanks to Mr. Kyle Openo for his expert technical assistance (and friendship) and whose efforts were invaluable to the completion of this thesis. I also wish to thank Ms. Patty Voss for her assistance, especially with the nuances of molecular biology techniques and by helping get my project off the ground. Thanks to all my friends and colleagues in the Wang/Schindler labs and throughout the Department: Drs. Eric Arnoys, Anandita Vyakarnum, Neera Agrwal, Sung-Yuan Wang, Yeou-Guang Tsay, Jamie Laing, John Loh, John Ho, Glenn Gaunt, Mike Washbum, Tim Smith, Mark Sutton, Carla Margulies, Peter Horn, Susan Sullivan, John Stebbins, Ms. Sharon Grabski, Ms. Nancy Lin, and Mr. Andy Lenneman. You have all made this experience worthwhile. Eric and Anandita, thanks. Thanks to my parents and to Alan and Sarah for their love and support. Dad, I know you thought this M.D.lPh.D. thing was a crazy idea, but you supported my decision nonetheless. Mom, I miss you very much. And finally, I wish to thank my wife, Gwynne. Her love, understanding and patience amazed me. She spent far too many evenings alone while I was working late at the lab. Gwynne, without you, I never could have done this. vi TABLE OF CONTENTS LIST OF TABLES .......................................................................................................... xi LIST OF FIGURES ....................................................................................................... xii LIST OF ABBREVIATIONS ....................................................................................... xiv CHAPTER 1. LITERATURE REVIEW ..................................................................... 1 A. GALECTINS AND GALECTIN-3 ............................................................................ 2 The Galectin Family of Animal Lectins ................................................................. 2 Polypeptide Architecture of Galectins and Identification of a B-Galactoside-Binding Cassette ............................................................................ .2 Galectin-3 Saccharide Binding Specificity and Polypeptide Structure ................. 4 Galectin-3 Subcellular Localization ...................................................................... 7 Galectin-3 (and Galectin-1) are pre-mRNA Splicing Factors in a Cell-Free System ........................................................................................... 9 Biochemical Association of Galectin-3 with Nuclear Complexes ...................... 10 Tissue Distribution and Developmental Regulation of Galectin-3 ...................... 11 Expression of Galectin-3 is a Function of the Proliferative State of the Cell and as a Function of the Cell Cycle ................................................... 12 B. CELLULAR SENESCENCE ................................................................................... 14 Cellular Senescence may be a Model of Aging ................................................... 14 Mechanism(s) of Cellular Senescence ................................................................. 15 Genomic Instability ............................................................................................. 16 The Werner Syndrome ...................................................................................... 16 Telomeres ............................................................................................................ 17 Gene Expression Changes in Cellular Senescence .............................................. 20 vii Dominance of Senescent Phenotype with Respect to DNA Synthesis ................ 20 Alterations in pre-mRNA Processing in Cellular Senescence ............................. 23 Is Cellular Senescence Reversible? ..................................................................... 23 C. NUCLEAR IMPORT ................................................................................................ 25 The Nuclear Pore Complex ................................................................................. 25 Signals and Receptors Involved in Nuclear Import ............................................. 26 Three Distinct Import Pathways ......................................................................... 27 Mechanism of Nuclear Import ............................................................................ 28 Regulation of Nuclear Import .............................................................................. 29 D. REFERENCES ......................................................................................................... 30 CHAPTER II. THE HUMAN LGALS3 (GALECTIN-3) GENE: DETERMINATION OF THE GENE STRUCTURE AND FUNCTIONAL CHARACTERIZATION OF THE PROMOTER ...................................................... 40 ABSTRACT .................................................................................................................... 41 INTRODUCTION ........................................................................................................... 42 MATERIAL AND METHODS ...................................................................................... 45 Nomenclature of Galectin-3 cDNAs ................................................................... 45 Cell Culture and Synchronization ....................................................................... 45 Antibody Reagents and Other Materials ............................................................. 46 Genomic Library Screening ................................................................................ 47 Polymerase Chain Reaction (PCR) Amplification of Genomic DNA ................ 47 Subcloning and DNA Sequencing ....................................................................... 49 Isolation of HeLa cell poly(A)+ RNA .................................................................. 50 viii Primer Extension Analysis .................................................................................. 50 Ribonuclease Protection ...................................................................................... 51 cDNA Isolation by RT-PCR ................................................................................ 52 Southern Analysis of Genomic DNA ................................................................... 53 Promoter-reporter Gene Construct Preparation ................................................... 53 Determination of Promoter Activity .................................................................... 54 Preparation of LG] Cell Extracts and Immunoblot Analysis .............................. 57 RESULTS ........................................................................................................................ 58 Isolation and Characterization of Human Galectin-3 Genomic Clones .............. 58 Structure of the Human Galectin-3 Gene ............................................................ 58 Southern Blot Analysis of Genomic DNA .......................................................... 63 Detemiination of the Transcription Initiation Site .............................................. 66 Comparison of Human LGALS3 to other LGALS3 Gene Structures ................... 70 Comparison of the Genomic Sequence Flanking the Human and Mouse Galectin-3 Transcription Initiation Sites ............................................................. 73 Functional Characterization of the Promoter Region .......................................... 76 Sequence of the Promoter Region ....................................................................... 82 DISCUSSION ................................................................................................................. 86 ACKNOWLEDGEMENTS ............................................................................................ 90 REFERENCES ................................................................................................................ 91 CHAPTER III. GALECTIN-3 EXPRESSION AND SUBCELLULAR LOCALIZATION IN SENESCENT HUMAN FIBROBLASTS .............................. 96 ABSTRACT .................................................................................................................... 97 ix INTRODUCTION ........................................................................................................... 98 MATERIALS AND METHODS .................................................................................. 101 Cell Culture ........................................................................................................ 101 Antibody Reagents ............................................................................................. 102 Immunofiuorescence Microscopy ..................................................................... 103 Preparation of Cell Extracts and Immunoblotting Analysis .............................. 104 Polyethylene Glycol-Mediated Cell Fusions ..................................................... 105 RESULTS ...................................................................................................................... 107 Galectin-3 Expression in LGl Human Fibroblasts ............................................ 107 Nuclear versus Cytoplasmic Localization of Galectin-3 in Young and Senescent LGl Cells .......................................................................................... 111 Formation of Heterokaryons Derived from Fusion between Young and Senescent Cells ........................................................................................... 115 Analysis of Galectin-3 Localization in Heterokaryons ..................................... 121 Analysis of Nuclear and Cytoplasmic Markers and of DNA Synthesis in Heterokaryons ................................................................................................ 129 DISCUSSION ............................................................................................................... 135 ACKNOWLEDGMENTS ............................................................................................. 140 REFERENCES .............................................................................................................. 141 CHAPTER IV. CONCLUDING STATEMENTS .................................................... 144 LIST OF TABLES CHAPTER I. Table 1. Summary of Galectins ............................................................................. 3 Table H. Comparison of Gene Expression between Young and Senescent Cells ..................................................................................... 21 CHAPTER II. Table I. Exon-Intron Borders of the Human LGALS3 Gene .............................. 62 CHAPTER 111. Table I. Bromodeoxyuridine Labeling of P19 and P27 Cells Before and After Fusion ...................................................................... 108 Table 11. Nuclear Localization of Galectin-3 in P19 and P27 Cells Before and After Fusion ...................................................................... 124 xi LIST OF FIGURES CHAPTER I. Figure 1. Schematic Diagram Illustrating the Polypeptide Architecture of Galectins ............................................................................................. 5 CHAPTER 11. Figure 1. Genomic Organization of the Human LGALS3 gene ........................... 60 Figure 2. Southern Blot Analysis of Genomic DNA ............................................ 65 Figure 3. Identification of the Transcription Initiation Site by Ribonuclease Protection and Primer Extension Analyses .................... 68 Figure 4. Comparison of the Human LGALS3 Gene to the Structures of Other Mammalian Galectin Genes ................................................... 72 Figure 5. Comparison of the Human LGALS3 and Mouse LGALSS Genomic Sequences Flanking the Transcription Initiation Sites .......... 75 Figure 6. Promoter Activities of Human Galectin-3-Luciferase Constructs in HeLa Cells (A) and Quiescent and Serum-Stimulated LGl * Fibroblasts (B) ...................................................................................... 78 Figure 7. Galectin-3 Protein Levels in LG] Cells as a Function of Serum and Calcium Withdrawal and After Readdition of Complete Growth Medium ................................................................................... 81 Figure 8. Human LGALS3 Promoter Sequence and Identification of Putative Transcription Factor Binding Sites ......................................... 84 CHAPTER III. Figure 1. Galectin—3 Expression as a Function of Serum Starvation and Restimulation in LG] Cells ................................................................. 110 Figure 2. Double Immunofluorescence Staining for Galectin-3 Localization and for Determining Cells Competent for DNA Synthesis ................ 113 xii Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Figure 8. Analysis of Representative Fluorescent Cells by Laser Scanning Confocal Microscopy .......................................................................... 117 Tagging LGl Cells with Beads .......................................................... 120 Analysis of Homo- and Heterokaryons for Galectin-3 Localization ..123 Comparison of the Theoretical and Actual Phenotypes of Galectin-3 Localization ...................................................................... 127 Analysis of the Localization of Nuclear and Cytoplasmic Markers 131 Comparison of the Levels of Expression of Karyopherin-B and Ran in P19 and P27 Cells ................................................................... 134 xiii dNT P FACS FITC Gal-3 HDF IEG IgE BP Kar-B Kb kDa LDH LGALS3 Luc MEM Mr mRN A 4-MU NLS Nt N+ N- P PBS PCR PEG PI Poly(A)+ R RNA RNase RNP SDS-PAGE LIST OF ABBREVIATIONS B-galactosidase Basic helix-loop-helix Basepairs Bromodeoxyuridine Carbohydrate binding protein 35 Complementary deoxyribonucleic acid CAMP-response element binding factor Cumulative population doubling Dulbecco’s modified Eagle’s media with high glucose Deoxyribonucleic acid Deoxynucleoside triphosphate Fluorescence activated cell sorting Fluorescein isothiocyanate Galectin-3 Human diploid fibroblasts Immediate early gene Irnmunoglobulin B binding protein Karyopherin B Kilobasepairs Kilodaltons Lactate dehydrogenase Galectin-3 gene Luciferase Minimum Essential Eagle‘s Medium Relative molecular weight in daltons Messenger ribonucleic acid 4-methylumbelliferyl-B-D-galactoside Nuclear localization signal or sequence Nucleotides Nuclear staining for galectin-3 No nuclear staining for galectin-3 Passage Phosphate-buffered saline Polymerase chain reaction Polyethylene glycol Propidium iodide Polyadenylated Random Ribonucleic acid Ribonuclease Ribonucleoprotein Sodium dodecyl sulfate polyacrylamide gel electrophoresis xiv SIE Sm tRNA T-TBS 5’-UTR WRN WS Sis-inducible element Smith antigen of small nuclear ribonucleoprotein complex Transfer ribonucleic acid Tween, Tris saline buffer 5’-untranslated region Werner syndrome gene Werner syndrome XV CHAPTER I. Literature Review CHAPTER I. Literature Review A. Galectins and Galectin-3 The Galectin Family of Lectins. Galectins are soluble proteins defined by two characteristics: (a) affinity for B- galactosides and (b) a specific carbohydrate-recognition domain (CRD) with conserved amino acid residues (1). Mammalian galectins have been numbered (galectin-l, galectin- 2, etc.) and ten members have thus far been identified (2) (see Table 1). Lower vertebrates and invertebrates also have galectins, but the extensive sequence differences between them and the mammalian counterparts has made their assignment into galectin- 1, 2, etc. categories difficult, and precludes a nomenclature for all species. Nevertheless, because of the sequence similarity within the CRD, galectins are known to exist as far down the evolutionary scale as sponges and fungi (3, 4). Polypeptide Architecture of Galectins and Identification of a B-Galactoside-Binding Cassette. Galectins have been divided into three categories based on their respective polypeptide architectures (5, 6): (a) proto type (galectins-1, 2, 5, 7, 10) have one CRD (Mr ~ 14-16,000); (b) chimera type (galectin-3) consists of a CRD in the carboxy- terminal half and an amino terminal half rich in proline and glycine residues (Mr ~ 29- Table 1. Summary of Galectins Designations MW Structural Quaternary Source Organisms Tissue/Cell Distribution (kDa) Type Structure Mammals Galectin-1 14.5 Proto Dimer Human, rat, mouse, Muscle, heart, lung, hamster, monkey, ox, placenta, brain, spleen, pig liver, lymph node, thymus, prostate, colon Galectin-2 14.5 Proto Dimer Human, mouse Small intestine Galectin—3 29-35 Chimera Monomer Human, rat, mouse, Macrophage, colon, dog, hamster leukemia cell, fibroblasts Galectin-4 36 Tandem Monomer Human, rat, pig, Alimentary tract epithelium repeat mouse Galectin-5 17-18 Proto Monomer Rat Erythrocyte Galectin-6 34 Tandem Monomer Mouse Castro-intestine repeat Galectin-7 14.5 Proto ? Human, rat Skin Galectin-8 34 Tandem Monomer Human, rat Liver, lung, kidney repeat Galectin-9 35 Tandem ? Human, rat, mouse Kidney, thymus, Hodgkin’s repeat lymphoma Galectin-10 17 Proto Dimer Human Birds Chick 14K 14 Proto Monomer Chick Skin, intestine, etc. Chick 16K 16 Proto Dimer Chick Muscle, liver, etc. Chick 30K 30 Chimera ? Chick Chondrocm Amphibians Xenopus 16K 16 Proto Dimer Xenopus laevis Skin Bufo 15K 14.5 Proto Dimer B. arenarum Oocyte Fish Electrolectin 16 Proto Dimer Electric eel Electric—organ Con erin 16 Proto Dimer Conger eel Skin mucus Nematodes Nematode 32K 32 Tandem Monomer C. elegans Cuticle, pharynx repeat Nematode 16K 16 Proto Dimer C. elegans ? OvaGaIBP 32 Tandem ? 0. volvulus ? repeat Sponge: Gth1/2 13-18 Proto Dimer G. cydonium Plasma membrane Fungi; Cgl-I/II 15.5/ Proto Dimer C. cinereus Fruiting body 17 35,000); and (c) tandem-repeat type (galectins—4, 6, 8, 9) have two CRDs, one in the amino terminal half linked to another one in the carboxyl terminal half (Mr ~ 32-36,000) (Figure 1). Proto type galectins 1 and 2 exist as dimers under physiologic conditions; all other galectins thus far identified are monomers. The galectin CRD consists of non-contiguous amino acid residues (Figure 1). Studies using site-directed mutagenesis, as well as X-ray crystallography results from galectins 1 and 2, have revealed the critical amino acid residues in the CRD necessary for carbohydrate binding. The X—ray structures of bovine galectin-1 (7) and human galectin- 2 (8) were determined in the presence of N-acetyllactosamine and lactose, respectively. Both the galectin-1 and -2 monomer is composed of a 5- and 6-stranded anti-parallel [3- sheet arranged in a B-sandwich motif. The B-strands are connected by short loop regions and there are no oc-helical segments. All of the amino acid residues that interact directly with the disaccharide are contained on adjacent anti-parallel B-strands. These strands are continuous in the primary structure and analysis of the genomic DNA structure from a number of galectins (including mouse galectin-3) indicates they are derived from a single exon (9). This suggests that the galectin CRD represents a B-galactoside-binding cassette evolutionarily conserved by all members of the galectin family. Galectin-3 Saccharide Binding Specificity and Polypeptide Structure. Galectin-3 is the only member of the chimera type thus far identified. It was initially isolated from extracts of cultured mouse Swiss 3T3 fibroblasts and human SL66 fibroblasts by passing the extracts over galactose-containing glycoconjugate affinity Proto Type: I fi-fipfiw-fime-é-fidmm I Chimera Type: I (PGAYPGXXX). 1 Fr-r'uPé~V-ir~we-é-§~F-G~R I Tandem Repeat Type: fiipfi~vfi~weéé~ne~a I fi-fiPE~V-fi~we-é-§~F-G~R I Figure 1. Schematic diagram illustrating the polypeptide architecture of galectins. The Proto Type is composed of a single lectin domain that contains the CRD. The Chimera Type galectin has two parts, a C-terminal half containing the galectin CRD and an N—terminal half rich in proline and glycine residues. The Tandem Repeat Type has two homologous CRD domains. Conserved amino acid residues in galectin CRD are indicated. Dark circles denote residues that directly interact with the carbohydrate by hydrogen bonding. In the N-terminal domain of Chimera Type, the nine amino acid sequence motif that is tandemly repeated 8-12 (n) times is shown. The single letter amino acid code is used. X, any amino acid. columns and elution with lactose or galactose, but not mannose, sucrose or N- acetylglucosamine (GlcNAc) (10, 11). Hydropathy analysis of the amino acid sequence based on the cDNA clone for mouse galectin-3 demonstrated two distinct structural domains (12, 13). The carboxy- terminal half (residues 118-264) containing the CRD has a globular structure. The amino-terminal half (residues 1-117) is neither hydrophobic nor hydrophilic and contains a proline-glycine-alanine-tyrosine (PGAY) motif. This PGAY motif is repeated five times in the human polypeptide and eight times in the mouse polypeptide. The amino- terminal domain has structural features seen in certain polypeptides from ribonucleoprotein (RNP) complexes, including hnRNP-Cl (12) and SAP62 (14), the homologue of the yeast splicing factor PRPl 1. A third domain at the amino terminus has been suggested given that amino acid residues 10-39 of the human galectin-3 have 46% sequence identity with residues 18-47 of the human serum response factor (16). As might be predicted, the Pro-, Giy-rich amino-terminal domain is sensitive to digestion by collagenase D. Following digestion of mouse galectin-3 with collagenase D, a Mr ~ 16,000 polypeptide remains, corresponding to the carboxy-terminal half of the polypeptide with retention of carbohydrate-binding actitivy. The individuality of the carboxy- and amino-terminal domains was further demonstrated using differential scanning calorimetry (15). Each domain folds independently and the carboxy-terrninal domain is therrnodynamicaliy stabilized when bound to lactose. The pI of galectin-3 is 8.7 based on a calculated value from the deduced amino acid sequence and experimentally from isoelectric focusing studies of the recombinant mouse galectin-3 (17). Extracts of both human and mouse fibroblasts subjected to two- dimensional electrophoresis reveal two isoelectric species: the pI 8.7 (non- phosphorylated) form and a pI 8.2 (singly phosphorylated) form (17). Based on sensitivity to alkaline phosphatase, the single phosphate is likely O-linked. Indeed, one study has shown that either serine-6 (the major site at 90%) or serine-12 (a minor site at ~10%) are the sites of phosphorylation in Madin-Darby canine kidney cells (18). Although the function of the phosphorylation is unknown, as well as the point in the cell cycle at which it is phosphorylated and the enzyme responsible, these two isoelectric variants have differential expression and subcellullar localization (see below). Galectin-3 Subcellular Localization. Despite the lack of a hydrophobic signal sequence, a small fraction (~5%) of galectin-3 can be found outside the cell (at cell surface or in medium). This has provoked investigators to study "non-classical" pathways of extemalization (19). In any case, the cell surface galectin-3 stimulated studies on the role of the protein as a tumor marker, in cell-cell recognition and metastasis (20, and references therein). Despite the presence of galectin-3 at the cell surface, the majority of galectin-3 is localized to the cytoplasm and nucleus, based on immunofluorescence (IF) studies with fixed and permeabilized 3T3 fibroblasts (21), the human cervical carcinoma cell line HeLa (22), and normal human fibroblasts (23, 24). Subcellular fractionation studies were consistent with the IF results, and demonstrated that galectin-3 is found prominently in the nuclear pellet and the postnuclear soluble fraction (25). The levels of galectin-3 expression were analyzed under quiescent and proliferating conditions in Swiss 3T3 cells. It was found that sparse proliferating cells have higher levels of galectin—3 than either contact-inhibited or serum-starved quiescent cells, based on overall IF intensity and percentage of positively stained cells (21). In addition, galectin-3 was found to localize predominantly to the nucleus under proliferating conditions, and the percentage of cells with distinct punctate nuclear staining reached a maximum just prior to the onset of the S phase of the cell cycle (21). These data suggested that galectin-3 may have an endogenous nuclear function and play a role in the proliferative state of the cell. The punctate nuclear distribution of galectin-3 represents diffuse staining throughout the nucleoplasm and "black holes" with little or no staining in areas presumed to be nucleoli (26). If permeabilized cells are extracted with ammonium sulfate (which extracts the majority of nonhistone nuclear proteins, leaving chromatin, nuclear matrix, and associated RNAs ) prior to fixation, the staining pattern is less diffuse and a distinct speckled pattern is observed (26, 27). The emergence of this speckled pattern is likely secondary to an overall decreased antigen level since similar results can be obtained with more dilute antibody concentrations. With loss of some galectin-3 from the nuclei during extraction, staining of discrete structures/regions becomes more apparent. Interestingly, the galectin-3 speckled pattern of staining is similar to that observed for the Sm antigen, an epitope defined on certain polypeptide components of the snRNPs. The punctate and speckled staining suggests that galectin-3 is associated with subnuclear structures. To directly test this, cells were treated with nucleases following perrneabilization. Nuclear staining was completely lost when cells were treated with ribonuclease, but not deoxyribonuclease (26). This suggests that galectin-3 is associated with RNP structures, consistent with galectin-3 being present in nuclear fractions enriched in hnRNPs. To more precisely localize galectin-3 in the nucleus, immunogold electron microscopy was performed. Galectin-3 was localized to interchromatin spaces (but not interchromatin granule clusters), at the border of condensed chromatin, and surprisingly, in the dense fibrillar component and at the periphery of the fibrillar centers of nucleoli (26). Although there is general agreement between the light microscope IF results and the electron microscope results, this apparent discrepancy between the lack of immunofluorescence staining in nucleoli and presence of immunogold localization to nucleoli cannot be explained at present. One explanation might be that the epitope is masked in the fixed cell but not in thinly sliced sections. Galectin-3 (and Galectin-1) are pre-mRN A Splicing Factors in a Cell-Free System. Galectin-1 and -3 are pre-mRNA splicing factors in a cell-free system from a HeLa cell nuclear extract (NE) (28, 29). This conclusion is derived from the following key findings: (a) NE capable of carrying out pre-mRNA splicing contain both galectins-1 and -3; (b) depletion of both galectins from the NE, either by lactose affinity adsorption or by double antibody depletion, resulted in the concomitant loss of splicing activity; (0) depletion of either galectin-l or galectin-3 individually, by specific antibody adsorption, failed to remove all of the splicing activity and the residual activity was still sensitive to saccharide-specific inhibition; (d) either galectin-l or galectin-3 alone is sufficient to reconstitute, at least partially, the splicing activity of NE depleted of both galectins. The COOH-terminal domain of galectin-3 alone (generated by collagenase D digestion and isolated by lactose affinity chromatography) was sufficient to restore splicing activity (28). Nevertheless, the exact role of the CRD in splicing is unknown at 10 present. Lactose-induced inhibition of splicing activity can be alternatively explained by: (a) a conformation change in the polypeptide after lactose binding (as discussed above-- see ref 15) or (b) molecular mimicry, with the carbohydrate mimicking an endogenous nuclear molecule that binds to the CRD (30). In either case, lactose binding would render galectin-3 (or galectin-l) unable to bind to a putative splicing partner(s) with subsequent disruption of spliceosome assembly and activity. To test for the role of the CRS, the ability of site-directed mutants of the galectin-3 (or galectin-1) CRS to restore splicing activity and spliceosome assembly must be determined, but these studies have not yet been performed. The functional redundancy of galectin-1 and galectin-3 suggests redundancy throughout the entire galectin family. It is not known, however, whether other members of the galectin family participate in pre-mRNA splicing, and it would be especially interesting to determine if a tandem repeat type galectin (-4, -6, and -8) has splicing activity. Again, the splicing activity may (or may not) be related to carbohydrate binding specificity per se (as indicated above) or due to the polypeptide structure per se. Biochemical Association of Galectin-3 with Nuclear Complexes. The IF and ultrastructural studies clearly demonstrate a nuclear localization for galectin-3. Because galectin-3 is released from permeabilized nuclei after ribonuclease (but not deoxyribonuclease) treatment, its association with nuclear components, specifically RNA-containing components or structures, was sought. It was determined that galectin-3 is a component of heterogeneous nuclear ribonucleoprotein (hnRNP) complexes based on the following observations in 3T3 cells: (a) fractionation of 11 nucleoplasm by means of a cesium sulfate gradient placed galectin-3 in fractions with the same densities as those reported for hnRNP (~1.3 g/ml); (b) galectin—3 co-isolates with hnRNP by sucrose gradient centrifugation (40S); and (c) saccharide—specific affinity chromatography of nucleoplasm yielded a bound fraction with both galectin-3 and polypeptides with molecular weights matching those reported for certain hnRNP proteins (31). Indeed, the primary structure in the NH2-terminal half of galectin-3 has sequence similarity to hnRNP A and C1 (12). Given the role of galectin-3 in in vitro pre-mRNA splicing, its ability to restore higher order spliceosome complexes, its nuclear co-localization with splicing factors (SC35, Sm) in discrete foci/speckles, and its localization at the ultrastructural level in regions believed to be rich in splicing factors or in the assembly of pre-mRNA transcripts, studies are underway to identify specific spliceosome components or other cellular components that interact with galectin-3. Unfortunately, there is presently no good biochemical evidence to suggest what specific polypeptide(s) or nucleic acid(s) interact with galectin-3. Tissue Distribution and Developmental Regulation of Galectin-3. The notion of functional redundancy in pre-mRNA splicing between galectins-1 and -3 is consistent with the results of experiments using transgenic mice in which a null mutation in the gene encoding galectin-1 has been introduced by homologous recombination in embryonic stem cells (32). Homozygous animals carrying the mutant allele lack galectin-l but development was not affected and the mice were viable and 12 fertile. It is possible that the functional role of galectin-1 was assumed by a closely related protein(s) expressed in these cells, perhaps galectin-3. The pattern of tissue specific expression of galectin-1 and galectin-3 has been studied carefully during mouse embryogenesis by Poirier and her colleagues (33-35). Both galectins are first detected on day 4 of mouse development and their expression appears to be limited to the trophoectoderm of the hatched blastocyst. Thus, galectins-1 and -3 overlap in terms of their expression during early embryogenesis. Following gastrulation, however, their patterns of expression appear to diverge. Galectin-1 is found in muscle cell precursors of somites while galectin-3 is restricted to the notochord. During the later parts of mouse embryogenesis, galectin-1 expression can be detected in many tissues of the kidney, gut, lung, liver and muscle but not in the chondrocytes of cartilage. In contrast, galectin—3 could be found in the cartilage of the vertebrae, with the hypertrophic chondrocytes exhibiting higher levels of expression than differentiated chondrocytes. Galectin-3 is also found in the suprabasal layer of the epidermis while no transcripts for galectin-1 could be detected. Finally, while galectin-1 is found in the motomeurons and in the sensory neurons of the dorsal root ganglia, galectin-3 could not be observed in the central nervous system. Expression of Galectin-3 is Sensitive to the Proliferative State of the Cell and as a Function of the Cell Cycle. As previously noted, galectin-3 protein levels were found to be high in proliferating but not in quiescent tissues, and IF staining for galectin-3 was increased in proliferating cells. Furthermore, galectin-3 expression is increased in transformed cells in 13 culture (36), and in cells with increased metastatic potential from both human breast and colon carcinoma (37, 38). To further characterize the expression of galectin—3 as a function of the proliferative state of the cell and as a function of the cell cycle, northern blot and nuclear run-off analyses were performed. Galectin-3 mRNA levels are nearly undetectable when mouse 3T3 fibroblasts are made quiescent by serum starvation (36). Following the addition of serum, which allows these quiescent cells to synchronously enter the cell cycle, galectin-3 mRNA levels increase rapidly (within 30 minutes). This rapid increase in galectin—3 expression is insensitive to cycloheximide, indicating that prior protein synthesis is not required for activation of the LGALS3 (galectin-3) gene. Thus, galectin-3 expression is regulated at the level of transcription and LGALS3 is characterized as an immediate early gene (IEG) in 3T3 cells. The initial increase in galectin-3 mRNA is followed by a decrease at about 3 hours, but then levels again rise and continue to increase throughout the mid and late 61 phase. Interestingly, two distinct galectin—3 mRNA transcripts are formed by alternative splicing at the 5’ end and each transcript is differentially expressed as a function of the cell cycle (discussed in detail in Chapter 2). Proliferation-dependent expression and subcellular localization of galectin-3 was further studied in 8166 normal human diploid fibroblasts (HDF) as a function of the cell’s proliferative capacity (23, 24). In senescent HDF, incapable of carrying out DNA synthesis and cell division, galectin-3 expression and subcellular distribution is altered. Following serum addition to serum-starved synchronized senescent HDF, the characteristic nuclear localization of galectin-3 is absent. In contrast to early passage HDF with a high replicative capacity, galectin-3 protein and mRNA levels are elevated in l4 senescent HDF during serum starvation and decrease after serum addition. Although many genes are specifically expressed or exhibit either an increased or decreased expression in cellular senescence (see below), the observation that galectin-3 expression increases and decreases in senescent HDF due to the absence or presence, respectively, of serum stimulation, places galectin-3 in a category distinct from other known senescent- specific genes. Given this altered expression and subcellular distribution of galectin-3 in senescent HDF, a more detailed discussion of cellular senescence is warranted. B. Cellular Senescence Cellular Senescence. The loss of the ability to carry out DNA synthesis and cell division is observed with all primary cells passaged in culture. This loss of proliferative or replicative capacity is termed cellular senescence. Senescence occurs after a certain number of cell divisions and is not secondary to the time spent in culture (chronological age). For example, normal HDFs senesce after approximately fifty population doublings, and for any given culture, each successive or cumulative population doubling (CPD) will lead to a decline in the percentage of proliferating cells. Cellular senescence is often considered a model of organismal aging based on the following observations: a) cells from older individuals senesce after fewer CPDs than cells obtained from younger individuals (39, 40); b) cells from shorter-lived species senesce after fewer CPDs than cells obtained from shorter-lived species (41); c) cells from people with certain human progerias (premature aging syndromes), such as Werner 15 Syndrome, senesce after fewer CPDs than cells obtained from age-matched controls (42); and d) there is an increased number of senescent cells in older individuals (43, 44). Thus, this evidence suggests that cultures of early passage cells (low CPD) are "young," whereas late passage or senescent cultures (high CPD) are "old." The use of cellular senescence as a model for aging, however, is controversial. For example, a recent study found no significant correlation between donor age and proliferative capacity of cultured fibroblasts (45), and the correlation between species age and proliferative capacity in culture has been alternatively explained to be due to species size per se (and hence an increased inherent requirement for more cells via more cell divisions) (46). Furthermore, reduced proliferative capacity of cells is not found in all progerias (47). In this regard, the yeast Saccharomyces cerevesiae may be useful to study and connect cellular senescence with aging because the cell IS the organism, and indeed, yeast cells do exhibit a finite lifespan in a fashion analogous to primary cells in culture. In any case, regardless of whether cellular senescence is an accurate model of organismal aging or not, it certainly is a unique growth arrest phenotype of all primary cells in culture. Mechanism(s) of Cellular Senescence. Potential mechanisms of cellular senescence include accumulation of genomic mutations (nuclear and/or mitochondrial), gene expression changes, genomic instability, telomere shortening, apoptosis, and alterations in oxidative or heat shock stress (for recent reviews, see ref 47-50). In order to extend this discussion to organismal aging, additional factors/mechanisms such as endocrine control and caloric intake would also 16 have to be considered, as well as a thorough review of the genes demonstrated to control life-span in Drosophila melanogaster and in the nematode Caenorhabditis elegans. Many of these topics are beyond the scope of the present review. Therefore, this review will focus on those aspects most closely related to the experimental chapters of this thesis, with particular emphasis on changes associated with senescence that occur within the nucleus in the HDF and S. cerevesiae. Genomic Instability. Genomic instability has long been proposed as a potential cause for aging and cellular senescence. Illigitimate recombinations, rearrangements, chromosome fragmentations and changes in number, accumulation of mutations, and loss of repeated DNA sequences have all been described and thought to be potential causes of aging. Using the model system S. cerevisiae, Sinclair et. al. have shown that extrachromosomal ribosomal DNA circles (ERC) increase with population doubling number (51). The rDNA is present in 100-200 tandem copies on chromosome XII. It is believed that a single ERC is generated by homologous recombination after approximately 50% of the yeast’s lifespan is complete. Copies of the ERC are produced with each subsequent cell cycle by replication from an origin present in the rDNA. Interestingly, the ERCs segregate into the mother cell, not the daughter cell, with each round of replication, so that there is an exponential increase in ERCs within the mother cell with each cell division. The accumulation of the ERCs within the nucleolus is believed to cause nucleolar enlargement and fragmentation near or at senescence (47). This may lead to altered rDNA transcription and ribosome assembly directly and/or indirectly alter DNA 17 replication and/or transcription by tying up required factors. It has been proposed that ERCs function as an "aging clock" in S. cerevisiase, since each round of replication results in an exponential increase in nucleolar ERCs (and since organismal mortality in general increases exponentially with age) (47). ERCs, however, have not been identified in other eukaryotic cells. Werner Syndrome. Werner syndrome (W S) is an autosomal recessive human progeria characterized by a decreased lifespan (average age of death, 46 i 11 years) and certain phenotypic features of the aging process, including skin atrophy, hair graying, loss of hair, cataracts, non-insulin dependent diabetes mellitus, arteriosclerosis, osteoporosis, and neoplasms (52). WS fibroblasts in culture senesce at fewer CPDs than age-matched controls. In addition, and similar to senescent fibroblasts, WS fibroblasts are characterized by chromosome instability with increased frequency of deletions and rearrangments (47). The gene responsible for WS (WRN) was found by positional cloning and is localized to 8p] l-12 (53). The coding region of WRN has sequence similarity to the RecQ DNA helicase family (53), and indeed, when expressed in a baculovirus/insect cell system, the protein has ATP-dependent 3’-->5’ DNA helicase activity (54). There is controversy as to the exact subnuclear localization of WRN, with one group finding it in the nucleolus and the other in the nucleoplasm in fibroblasts (55, 56). Most of the mutations in WRN identified thus far lead to a truncated helicase lacking its nuclear localization signal (NLS) in the carboxy-terminal region (57). Indeed, although the WRN mRNA is only slightly decreased in WS fibroblasts and B-cells (it is believed to be 18 constitutively expressed and is a "housekeeping" gene under Spl transcriptional regulator control), the protein appears excluded from the nucleus as determined by immunofluorescent staining at the single cell level (57, 58). It is believed that the WRN helicase functions in the cell to maintain genomic stability via maintenance of normal DNA unwinding and replication and repair. Thus the exclusion of WRN from the nucleus explains the similar clinical phenotype found in WS despite the many different mutations. Telomeres. Telomeres are repeated DNA sequences ('ITAGGG for vertebrates) at the ends of all eukaryotic linear chromosomes. They function to prevent the loss of the 5’ end on the lagging strand during replication due to the requirement for RNA priming and primer erasure (59). They also function to prevent chromosome fusions, recombinations, and degradation (59). Telomeres shorten with each successive cell division in HDF and most somatic tissues (100 bp per cell division). For this reason, telomeres have also been proposed to act as a molecular clock, counting each cell division. Unlike stem, gerrnline and embryonic mammalian cells, HDF and most somatic tissues lack telomerase, a ribonucleoprotein DNA polymerase that functions to add the telomeric repeat sequence (60). The role of telomere shortening in the senescence of cells has been controversial. Correlative and direct evidence in favor of a central role for telomere shortening and cellular senescence is as follows: a) telomere length decreases with each successive CPD in primary human cells in culture and with aging of human tissues in vivo; b) in most 19 human primary cells, telomerase activity is not detected. If the cells are transfected with SV4O T-antigen and senescence can be bypassed, telomerase activity is detected in the post-crisis cultures and in subsequent immortalized cells (61); and c) the replicative capacity of normal retinal pigment epithelial cells, HDF, and vascular endothelial cells increased following transfection and expression of telomerase. Many of the cell lines had proceeded for 20 CPDs beyond their normal senescence point and show no signs of decreased growth rate, changes in karyotype, or alterations in morphology (62). Conversely, there is evidence against a direct role for telomerase activity and maintainance of telomere length in cellular senescence: a) S. cerevisiae also senescece but the average telomere length is maintained (63, 64); b) telomerase activity is detectable in only 80-90% of human tumor samples, and c) the telomerase knockout mouse is viable for six generations and cells from these mice could be immortalized in culture, transformed by viral oncogenes, and generated tumors in nude mice following transformation despite having shortened telomeres (65). Most surprisingly, the cells from the fourth generation onward possessed no aneuploidy or chromosome abnormalities, including end-to-end fusions, despite having no detectable telomere repeats. These results clearly suggest the existence of an alternative mechanism that maintains chromosome integrity, at least in these cases. Unfortunately, the mouse knock out results may be somewhat difficult to extrapolate to the human setting because mouse cells undergo immortalization and malignant transformation much easier than human cells, suggesting the existence of alternative mechanism(s) and/or pathways involved. 20 Gene Expression Changes in Cellular Senescence. Unlike the quiescent state of young cells, such as those induced by low serum or high density, senescent cells are refractory to mitogenic stimulation by serum or growth factors. The expression of many genes is altered in senescent cells (Table 2) (66-79, and see 80-82 and references therein). Table 2 is not a comprehensive compilation, but rather it is meant to highlight certain categories of genes that exhibit changes in expression. Of particular importance may be the the CipI/WAFI/Sdil gene product or p21 (74, 83, 84). p21 inactivates cyclin-dependent kinases (CDK5) and blocks cell cycle progression, possibly by inactivating cyclin-CDK-induced phosphorylation of Rh. p21 is overexpressed in senescent cells and disruption of the gene leads to increased replicative capacity of HDF (see below). In addition, another CDK inhibitor, p16 (85), is overexpressed in senescent HDF (75). Thus the overexpression of CDK inhibitors and other genes in senescent cells is consistent with senescence having a dominant phenotype with respect to DNA synthesis (see below). Dominance of Senescent Phenotype with Respect to DNA Synthesis. A clue to the mechanism and phenotypic characterization of cellular senescence emerged during cell fusion studies. For example, restoration of DNA synthesis does not occur in the nucleus of a senescent cell when fused with an actively replicating HDF, but rather the initiation of DNA synthesis is inhibited in the nucleus of the actively replicating HDF (86). Ongoing DNA synthesis is not inhibited and the inhibition occurs only if the replicating cell is in the early to mid 61 phase of the cell cycle, at least 3 hours prior to the onset of the S phase (87). The inhibition of DNA synthesis is delayed by 24- 21 Table II. Comparison of Gene Expression between Young and Senescent Cells . Cam; Gene/Protein Egression’ System Comment Reference c-myc No change HDF 51 c-fos Decreased HDF/HDF extract: Decreased AP-l activity 51-54 has been demonstrated Dunscript‘ion Factors and Related Proteins E2F ? HDF extracts Binding activity decreased 55 in DHFR promoter CRE—BP ? HDF extracts Binding Activity Decreased 54 CTF ? HDF extracts Binding Activity Decreased 54 ld-l Decreased HDF Ids inhibit bHLH 56 Id-2 Decreased HDF transcription factors 56 Cyclin C No change HDF 57 eye A Decreased HDF 58 cyc B Decreased HDF 58 Cyclins and cyclin- dependent kinases (CDKs) CDK-4 No change HDF 57 CDK-5 No change HDF 57 CDC 2, cdc 2 Decreased HDF 53 CDK inhibitors p21 Increased HDF 59 p16 Increased HDF 60 humor suppressor p53 No change HDF 51, 61 genes Rb No change HDF Remains in persistent activated 62 (hypophosphorylated) state RAS-l Decreased S. cerevisiae if overexpress RAS-2, but not 63 Ce” signaling proteins RAS-2 Decreased S. cerevisiae RAS-l, get increased lifespan 63 MAPK Decreased Rat hepatocytes Isolated from young & old rats 53 Other PCNA Decreased HDF DNA polymerase 8 64 co-factor "' in senescent cells vs young cells HDF, human diploid fibroblast 22 48 hours when the heterodikaryon cell is transiently treated with a protein synthesis inhibitor (either cycloheximide or puromycin) shortly after fusion (88). These data suggest that the senescent cell has not lost a particular factor required for DNA synthesis, or that a critical factor has been diluted to sub-optimal concentrations, but rather that a polypeptide is produced in these non-replicating cells which inhibits (either directly or indirectly) cell-cycle transit and DNA synthesis. Therefore, based on cell fusion studies between pre-senescent and senescent HDF, cellular senescence is considered a dominant phenotype with respect to DNA synthesis. The genetic dominance of the senescent phenotype (and the recessive nature of immortality) has also been demonstrated in cell fusion studies between normal and immortal cells (89-91). Furthermore, because multiple genes are believed to cause senescence, fusion of different immortal cell lines should sometimes result in immortal hybrids, indicating that the two parental cell lines had the same genetic defect, and these parental cell lines could be assigned the same complementation group. Similarly, cell fusions between different immortal cell lines should sometimes result in hybrids which exhibit replicative senescence, indicating that the parental cell lines were of a different complementation group. By this cell fusion approach with many different immortal human cell lines, four complementation groups have been suggested, indicating that at least four genes or gene pathways contribute to the phenotype of senescence (92, 93). However, assignment into distinct complementation groups is controversial and may be a result of the particular experimental methods employed (94). 23 Alterations in pre-mRN A Processing in Cellular Senescence. No direct evidence for a defect in pre-mRNA splicing exists in cellular senescence. However, the proliferating cell nuclear antigen (PCNA) mRNA levels are undetectable in senescent WI—38 HDF and yet little or no difference in the PCNA transcription rate or hnRNA levels exists between the young and senescent cells, based on run-off transcription analysis and RT-PCR, respectively (95). These data clearly suggest a post-transcriptional block in the expression of PCNA and provides indirect evidence of a potential splicing defect in senescent WI-38 HDF. In addition, alternative splicing variants are specifically expressed in WI—38 HDF for histone 3 (51) and in [MR-90 HDF for fibronectin (96). Also, differential expression of fibronectin splice site variants exists between tissue isolated from young and old rats (96). Any role for galectin-3 in alternative splicing, or targeting certain hnRNAs for processing (e.g., PCNA) is purely speculative, but given its altered expression and subcellular distribution in senescent SL66 HDF, galectin-3 certainly becomes an intriguing target in searching for a mechanism for altered pre-mRNA processing in senescent cells. Is Cellular Senescence Reversible? Clearly, cells are able to escape senescence and become immortalized. The frequency with which cells from different species escape cellular senescence and give rise to immortal cell lines varies extensively, and while mouse cells immortalize with a high probability, immortalization of HDF by chemical carcinogens or oncogenes is an extremely rare event (97). Given this difficulty in immortalizing HDF and given the above described multiple gene alterations in the senescent cell, it is not surprising that 24 introduction of single genes c-fos (98, 99) and cdc2 (100) by microinjection, or other protooncogenes by transfection (101), fails to induce senescent cells to resume DNA synthesis. Nevertheless, inactivation of p21 by two sequential rounds of targeted homologous recombination was sufficient to bypass senescence and extend the CPD of LFl HDF in culture (102). When introduced into early passage cells, the SV40 virus large T-antigen extends the replicative life-span about 40% (20 additional CPDs for HDF) (101, 103). In senescent cells, introduction of T-antigen can reactivate one round of DNA synthesis, but not cell division. The T-antigen can interact with the tumor suppressor p53, as well as with Rb, inactivating the growth suppressive activities of both proteins: SV40 large T- antigen mutants lacking either the Rb or p53 binding domains are unable to reactivate DNA synthesis in senescent HDF (104). Using lMR-90 HDF transfected with a steroid- inducible T-antigen, it was shown that T-antigen is required for proliferation during the extended life span prior to crisis (a period when cell number remain constant or declines as successful cell division is balanced by cell death), as well as following immortalization (101). On the basis of these results, the escape from senescence was delineated into two stages: M1, which produces what is commonly regarded as "senescence" and which can be bypassed/overcome by T-antigen until crisis occurs, and M2, whose rare inactivation gives rise to immortalization following crisis. 25 C. Nuclear Import. The Nuclear Pore Complex. The nucleus is isolated from the cytoplasm by the nuclear envelope, which consists of inner and outer nuclear membranes, nuclear pore complexes, and nuclear lamina (105). Proteins enter the nucleus through the nuclear pore complex (NPC) which spans the double bilayer nuclear envelope; it is estimated that the nuclear envelope of a eucaryotic cell contains approximately 2000-4000 pores (106). The NPC allows passive diffusion of proteins less than ~40-60 kDa, although many small nuclear proteins do, in fact, enter the nucleus via a mechanism distinct from simple diffusion (107, 108). Larger proteins enter the nucleus by a facilitiated process (109), meaning import has the following characteristics: a) requires the presence of a nuclear localization signal (NLS) which is both necessary and sufficient for transport; b) requires a receptor/cytosolic factor; and c) is energy and temperature dependent. It has also been shown that some proteins enter the nucleus by a "piggyback method" where they are co-transported with an NLS-containing protein (110). Furthermore, nuclear import can be divided into two discrete steps: a) a relatively rapid binding of proteins to the nuclear pore; and b) a slower ATP-dependent (and wheat germ agglutinin-inhibitable) translocation step into the nucleus (111). The NPC is a large (125 MDa) cylindrical assembly with eight-fold rotational symmetry (for recent reviews, see ref. 112, 113). The NPC core is a tripartite structure: a spoke-central plug assembly framed by a nuclear ring facing the nucleoplasm and a cytoplasmic ring facing the cytoplasm (114). The central plug attains the shape of a 26 cylinder with a tapered center and can expand to a diameter of ~25 nm, large enough for the passage of large macromolecules, including RNA (and RNP). The central plug is open transiently and although it physiologically acts as a gated transport channel, there is no evidence for a discrete gate. The NPC also contains continuously open ~9 nm aqueous channels, and may be involved in water and electrolyte nuclear-cytoplasmic homeostasis. Extending out from the cytoplasmic ring into the cytoplasm are short filaments or fibers, and extending out from the nuclear ring into the nucleus are long nuclear fibers which form a basket-like structure. The fibers may serve to recognize and bind components to be transported. The NPC is composed of over 100 different polypeptides called nucleoporins (112, 113). Some of these nucleoporins are modified with an O—linked N- acetylglucosamine (O-GlcNAc), and many contain a characteristic short repeat motif sequence XFXFG, where X is any amino acid with a small or polar side chain. Irnmunoelectron microscopy has been used to determine the locations of the nucleoporins within the NPC. The O-GlcNAc nucleoporins are believed to be docking sites for nuclear transport. Specifically, all B-type karyopherins, a family of transport factors, bind to these NPC glycoproteins. For example, evidence suggests NUP358, localized to one of the cytoplasmic fibers, is a site where the protein import complex is initially assembled and contains docking sites for Ran GTPase and karyopherin al/a2-B complexes. Signals and Receptors Involved in Nuclear Import. Nuclear localization signals on polypeptides are generally short sequences (8-10 amino residues) of mostly basic residues, usually lysine or arginine. The polypeptide 27 NLS is contained within one sequence or may be divided into a bipartite signal and can occur in any position within the polypeptide. The NLS of the SV40 large T-antigen (PKKKRKV) is well-characterized and is considered the "classic" unipartite signal; the NLS for nucleoplasmin, a major nuclear protein in Xenopus oocytes and embryos, requires a bipartite signal (KRPAAIKKAGQAKKKK) for import (115). The NPC does not contain all the components necessary for transport, as evidenced in studies of isolated nuclei (116, 117) and in vitro systems of cultured cells treated with digitonin to selectively perrneabilize the plasma membrane (117). Digitonin- perrneabilized cells retain the capacity to import nuclear proteins only with the supplement of soluble factors from fractionated cytosol (117). These factors do not contain RNA and cannot be pelleted by centrifugation at 100K x g (8100 fraction). Also, the factors are not species specific since 8100 fractions from one species can support in vitro nuclear transport in another species. Cytosolic factors/receptors have now been purified, and include karyopherins a1, a2, [31, and transportin (or karyopherin [32). Three Distinct Import Pathways. Different NLSs and their respective cytosolic receptors have defined three distinct import pathways: (a) classic (basic) NLS/karyopherin b1; (b) M9 NLS/transportin; and (c) KNS NLS/unknown receptor. (a) nuclear proteins with the classic (basic) single or bipartite NLS bind karyopherin Oil/Bl and a2/B1 receptor complexes; (b) hnRNP A1 contains an NLS designated M9 which is a contiguous stretch of 38 amino acids containing 12 glycine residues (118). The M9 import receptor has been identified 28 in a variety of organisms, in vertebrates it is designated transportin, MIP (M9 interacting protein), and karyopherin I32 (119, 120), and in yeast, it is named Kap104p (121). Transportin appears homologous to karyopherin B1, binds to nucleoporins with repeat sequence motifs, but does not mediate import of the "classic" SV40 or nucleoplamin-type NLS proteins (119, 120); (c) the hnRNP K contains a so-called KNS sequence NLS. Surprisingly, and distinct from the above two pathways, the KNS sequence can dock proteins to the nuclear pore without supplemented cytosolic factors (122). The nucleoporin involved in binding is presently unknown. It is likely that many more cytosolic receptors, NLS signals, and nucleoporins (and hence pathways) will be identified in the future. Mechanism of Nuclear Import. The classic NLS pathway is best characterized. The first step is the binding of the NLS to karyopherin or] (or a2) in the cytoplasm (123, 124), which also augments the interaction between karyopherin 0t and [31 subunits (125). Karyopherin [31 contains binding sites for repeat sequence motif-containing nucleoporins, and mediates the docking of the complex onto the nuclear pores (ATP-independent) (126). This protein/NLS-karyopherin or/Bl (docking ) complex is then translocated through the pore by an energy-dependent mechanism involving Ran, a Ras-GTPase (128, 129) and NTF2 (130, 131). Current evidence suggest that a Ran GTPase cycle coordinate the import pathway, with Ran-GDP being more abundant on the cytoplasmic side where it promotes 29 the formation of the docking complex and Ran-GTP being more abundant on the nuclear side where it stimulates dissociation of the complex (132, 129). Regulation of Nuclear Import. Post—translational modifications of polypeptides and cytosolic "anchor" proteins are two general mechanisms that may be involved in the regulation of nuclear import (reviewed in ref 133). For example, the yeast transcription factor SW15 is cytoplasmic in the S, G2, and M phases of the cell cycle, as phosphorylation of three serine residues near the NLS by the CDC28 kinase prevents nuclear entry. Dephosphorylation of these residues at the end of the M phase (when CDC28 is inactive) causes SW15 to enter the nucleus (134). One of the best studied examples of cytosolic anchors is work with the transcription factor NF-icB (135, 136). In quiescent cells, NF-KB is associated with its anchor IKB in the cytoplasm and upon stimulation of the cells, IKB is phosphorylated resulting in dissociation of the NF-KB/IKB and NF-KB nuclear translocation (137). 10. ll. l2. 13. 30 REFERENCES. Kasai, K. and Hirabayashi, J. (1996) J. Biochem. 199, 1-8. Hirabayashi, J. (ed.) (1997) Trends Glycosci. Glyotechnol. 9 (see especially p 180). Muller, W.E.G., Blumbach, B., Wagner-Hulsmann, C., and Less, U. (1997) Trends Glycosci. Glyotechnol. 9, 123-130. Cooper, D.N.W., Boulianne, R.P., Charlton, S., Farrell, E.M., Sucher, A., and Lu, B.C. ( 1997) J. Biol. Chem. 272, 1514-1521. Hirabayashi, J., and Kasai, K. (1993) Glycobiology 3, 297-304. Barondes, S.H., Cooper, D.N.W., Gitt, M.A., and Leffler, H. (1994) J. Biol. Chem. 269, 20807-20810. Liao, D.I., Kapadia, G., Ahmed, H., Vasta, GR, and Herzberg, O. (1994) Proc. Natl. Acad. Sci. USA 91, 1428-1432. Lobsanov, Y.D., Gitt, M.A., Leffler, H., Barondes, SH, and Rini, J .M. (1993) J. Biol. Chem. 268, 27034-27038. Leonidas, D.D., Elbert, B.L., Zhou, Z., Leffler, H., Ackerman, S.J., and Acharya, KR. (1995) Structure 3, 1379-1393. Roff, C.F., Rosevear, P.R., Wang, J .L., and Barker, R. (1983) Biochem. J. 211, 625-629. Roff, CF and Wang, J .L. (1983) J. Biol. Chem. 258, 10657-10663. Jia, S., and Wang, J.L. (1988) J. Biol. Chem. 263, 6009-6011. Wang, J .L., Laing, 1.6., and Anderson, R.L. (1991) Glycobiology 1, 243-252. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 31 Bennett, M. and Reed, R. (1993) Science 262, 105-108. Agrwal, N., Sun Q., Wang, S-Y., and Wang, J.L. (1993) J. Biol. Chem. 268, 14932-14939. Oda, Y., Leffler, H., Sakakura, Y., Kasai, K., and Barondes, S. (1991) Gene 99, 279-283. Cowles, E.A., Agrwal, N., Anderson, R.L., and Wang, J.L. (1990) J. Biol. Chem. 265, 17706-17712. Huflejt, M.E., Turck, C.W., Lindstedt, R., Barondes, S.H., and Leffler, H. (1993) J. Biol. Chem. 268, 26712-26718. Lindstedt, R., Apodaca, G., Barondes, S.H., Mostove, K., and Leffler, H. (1993) J. 'Biol. Chem. 268, 11750-11757. Akahani, S., Inohara, H., Nangia-Makker, P., and Raz, A. (1997). Trends Glycosci. Glcoytechol.. 9, 69-75. Moutsatsos, I.K., Wade, M., Schindler, M., and Wang, J.L. (1987) Proc. Natl. Acad. Sci. USA 84, 6452-6456. Vyakarnam, A., Lenneman, A.J., Lakkides, K.M., Patterson, R.J., and Wang, J.L. (1998) Exp. Cell Res. 242, 419-428. Cowles, E.A., Moutsatsos, I.K., Wang, J.L., and Anderson, R.L. (1989) Exp. Geront. 24, 577-585. Hamann, K.K., Cowles, E.A., Wang, J.L., and Anderson, R.L. (1991) Exp. Cell Res. 196, 82-91. Moutsatsos, I.K., Davis, J.M., and Wang, J .L. (1986) J. Cell Biol. 102, 477-483. Hubert, M., Wang, S-Y., Wang, J.L., Seve, A-P., and Hubert, J. (1995) Exp. Cell 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 32 Res. 220, 397-406. Wang, S-Y., Voss, P.G., Patterson, R.J., and Wang, J .L. (1995) Antib. Immunoconjug. Radiopharm. 8, 311-324. Vyakarnam, A., Dagher, S.F., Wang, J.L. and Patterson, R.J. (1997) Mol. Cell Biol. 17, 4730-4737. Dagher, S.F., Wang, J .L., and Patterson, R.J. (1995) Proc. Natl. Acad. Sci. USA 92,1213-1217. Kaur, K.J., Khurana, S., and Salunke, D.M. (1997) J. Biol. Chem. 272, 5539- 5543. Laing, J. G. and Wang, J. L. (1988) Biochemistry 27, 5329-5334. Poirier, F., and Robertson, E.J. (1993) Development 119, 1229-1236. Colnot, C., Ripoche, M.A., Scaerou, F., Fowlis, D., and Poirier, F. (1996) Trans. Biochem. Soc. 24, 141-146. Fowlis, D., Colnot, C., Ripoche, M.A., and Poirier, F. (1995) Dev. Dyn. 203, 241- 251. Poirier, F., Timmons, P.M., Chan, C.-T. J ., Guenet, J.-L., and Rigby, P.W.J. (1992) Development 115, 143-155. Agrwal, N., Wang, J .L., and Voss, PG. (1989) J. Biol. Chem. 264, 17236-17242. Schoeppner, H.L., Raz, A., Ho, SB, and Bresalier, R.S. ( 1995) Cancer 75, 2818- 2826. Nangia-Makker, R, Thompson, E., Hogan, C., Ochieng, J., and Raz, A. (1995) Intl. J. Oncology 7, 1079-1087. 31 39. 40. 41. 42. 43. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 33 Martin, G.M., Sprague, C.A., and Epstein, C]. (1970) Lab. Invest. 23, 86-92. Schnedier, E.L., and Mitsui, Y. (1976) Proc. Natl. Acad. Sci. USA 73, 3584-3588. Rhome, D. (1981) Proc. Natl. Acad. Sci. USA 78, 5009-5013. Salk, D. (1982). Hum. Genet. 62, 1-15. Dimri, G.P., Lee, X., Basile, G., Acosta, M., Scott, G., Roskelley, C., Medrano, E.E., Linskens, M., Rubelj, 1., Pereira-Smith, O., Peacocke, M., and Campisi, J. (1995) Proc. Natl. Acad. Sci. USA 92, 9363-9367. Effros, RB. (1998) Am. J. Hum. Genet. 62, 1003-1007. Cristofalo, V.J., Allen, R.G., Pignolo, R.J., Martin, B.G., and Beck, J.C. (1998) Proc. Natl. Acad. Sci. USA 95, 10614-10619. Promislow, D.E.L. (1993). J. Gerontol. 48, B115-B123. Johnson, F.B., Sinclair, DA, and Guarente, L. (1999) Cell 96, 291-302. Goldstein, S. (1990) Science 249, 1129-1133. Campisi, J. (1996) Cell 84, 497-500. Holbrook, N.J., Martin, G.R., Lockshin, R.A. (eds.) (1996) Cellular Aging and Cell Death. Wiley-Liss, Inc. New York, NY. Seshadri, T., and Campisi, J. (1990) Science 247, 205-209. Riabowol, K., Schiff, J., and Gilman, M.Z. (1992) Proc. Natl. Acad. Sci. USA 89, 157-161. Liu, Y., Guyton, K.Z., Gorospe, M., Xu, Q., Kokkonen, G.C., Mock, Y.D., Roth, GS, and Holbrook, N .J . (1996) J. Biol. Chem. 271, 3604-3607. Dimri, G.P., and Campisi, J. (1994) Exp. Cell Res. 212, 132-140. Dimri, G.P., Hara, E., and Campisi, J. (1994) J. Biol. Chem. 269, 16180-16186. 56. 57. 58. 59. 61. 62. 63. 65. 66. 34 Hara, E., Yamaguchi, T., Nojima, H., Ide, T., Campisi, J ., Okayama, H., and Oda, K. (1994) J. Biol. Chem. 269, 2139-2145. Rittling, S.R., Brooks, K.M., Cristofalo, V.J., and Baserga, R. ( 1986) Proc. Natl. Acad. Sci. USA 83, 3316-3320. Stein, G.H., Drullinger, L.F., Robetorye, R.S., Pereira-Smith, O.M., and Smith, JR. (1991) Proc. Natl. Acad. Sci. USA 88, 11012-11016. Noda, A., Ning, Y., Venable, S.F., Pereira-Smith, D.M., and Smith, JR. (1994) Exp. Cell Res. 211, 90-97. Alcorta, D.A., Xiong, Y., Phelps, D., Hannon, G., Beach, D., and Barrett, J.C. (1996) Proc. Natl. Acad. Sci. USA 93, 13742-13747. Afshari, C.A., Vojta, P.J., Annab, L.A., Futreal, P.A., Willard, TB, and Barrett, J.C. (1993) Exp. Cell Res. 209, 231-237. Stein, G.H., Beeson, M., and Gordon, L. (1990) Science 249, 666-668. Sun, J ., Kale, S.P., Childress, A.M., Pinswasdi, C., and Jazwinski, SM. (1994) J. Biol. Chem. 269, 18638-18645. Chang, C-D., Phillips, P., Lipson, K.E., Cristofalo, V.J., and Baserga, R. (1991) J. Biol. Chem. 266, 8663-8666. Cristofalo, V.J., Phillips, P.D., Sorger, T., and Gerhard, G. (1989) In Growth Control During Cell Aging, (H. Warner and E. Wang, eds.) CRC Press, Boca Raton, FL, pp 5-18. McCormick, A., and Campisi, J. (1991) Curr. Opinion Cell Biol. 3, 230-234. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 35 Cristofalo, V.J., Pignolo, R.J., Cianciarulo, F.L., DiPaulo, BR, and Rotenberg, MD. (1992) Exp. Geront. 27, 429-432. Harper, J.W., Adami, G.R., Wei, N., Keyomarsi, K., and Elledge, SJ. (1993) Cell 75, 805-816. El-Deiry, W.S., Tokino, T., Velculescu, V.E. (1993) Cell 75, 817-825. Serrano, M., Hannon, G.J., and Beach, D. (1993) Nature 366, 704-707. Sinclair, D.A., and Guarente, L. (1997) Cell 91, 1033-1042. Salk, D., Fujiwara, Y., and Martin, G.M. (eds.) (1985) Wemer’s syndrome and human aging. Plenum Press, New York, NY. Yu, C.-E., Oshima, J., Fu, Y.-H., Wijsman, E.M., Hisama, R., Alisch, R., Matthews, S., Nakura, J., Miki, T., Ouais, S., Martin, G.M., Mulligan, J., and Schellenberg, GD. (1996) Science 272, 258-262. Suzuki, N., Shimamoto, A., Imamura, O., Kuromitsu, J ., Kitao, S., Goto, M., and Furuichi, Y. (1997) Nucleic Acids Res. 25, 2973-2978. Shiratori, M., Sakamoto, S., Suzuki, N., Tokutake, Y., Kawabe, Y., Enomoto, T., Sugimoto, M., Goto, M., Matsumoto, T., and Furuichi, Y. (1999) J. Cell Biol. 144, 1-9. Marciniak, R.A., Lombard, D.B., Johnson, RB, and Guarente, L. (1998) Proc. Natl. Acad. Sci. USA 95, 6887-6892. Matsumoto, T.A., Shimamoto, A., Goto, M. and Furuichi, Y. (1997) Nat. Genet. 16, 335-336. Yamabe, Y., Shimamoto, A., Goto, M., Yokota, J., Sugawara, M., and Furuichi, Y. (1998) Mol. Cell. Biol. 18, 6191-6200. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 36 Greider, CW. (1996) Annu. Rev. Biochem. 65, 337-365. Blackburn, E. (1991) Nature 350, 569-573. Counter, C.M., Avilion, A.A., LeFeuvre, C.E., Stewart, N .G., Greider, C.W., Harley, CB, and Bacchetti, S. (1992) EMBO J. 11, 1921-1929. Bodnar, A.G., Ouellette, M., Frolkis, M., Holt, S.E., Chiu, C-P., Morin, G.B., Harley, C.B., Shay, J.W., Lichtsteiner, S., and Wright, W.E. (1998) Science 279, 349-352. D’Mello, NP, and Jazwinski, SM. (1991) J. Bacteriol. 173, 6709-6713. Smeal, T., Claus, J., Kennedy, B., Cole, F., and Guarante, L. (1996) Cell 84, 633- 642. Blasco, M.A., Lee, H-W., Hande, M.P., Samper, E., Lansdorp, P.M., DePinho, R.A., and Greider, CW. (1997) Cell 91, 25-34. Norwood, T. H., Pendergrass, W. R., Sprague, C.A., and Martin, GM. (1974) Proc. Natl. Acad. Sci. U.S.A. 71, 2231-2235. Rabinovitch, PS, and Norwood, TH. (1980) Exp. Cell Res. 130, 101-109. Burmer, G.C., Zeigler, C.J., and Norwood, TH. (1982) J. Cell Biol. 94, 187-192. Bunn, C.L., and Tarrant, GM. (1980) Exp. Cell Res. 37, 385. Pereira-Smith, C.M., and Smith, J .R. (1981) Somatic Cell Genet. 7, 411. Pereira-Smith, C.M., and Smith, J .R. (1983) Science 221, 964. Pereira-Smith, C.M., and Smith, J .R. (1988) Proc. Natl. Acad. Sci. USA 85, 6042. Duncan, B.L., Whitaker, N.J., Moy, B.L., and Reddel, RR. (1993) Exp. Cell Res. 205, 337. 94. 95. 96. 97. 98. 99. 100. 101. l 02. 103. 104. 105. 1 O6. 107. 37 Ryan, P.A., Maher, V.M., and McCormick, J .J . (1994) J. Cell Physiol. 159, 151- 160. Chang, C-D., Phillips, P., Lipson, K.E., Cristofalo, V.J., and Baserga, R. (1991) J. Biol. Chem. 266, 8663-8666. Magnuson, V.L., Young, M., Schattenberg, D.G., Mancini, M.A., Chen, D., Steffensen, B., and Klebe, R.J. (1991) J. Biol. Chem. 266, 14654-14662. McCormick, J .J ., and Maher, V.M. (1988) Mutat. Res. 199, 273-284. Dimri, G.P., and Campisi, J. (1994) Cold Spring Harbor Symp. Quant. Biol. 59, 67. Rose, D.W., McCabe, G., Feramisco, J .R., and Adler, M. (1992) J. Cell Biol. 119, 1405-1411. Stein, G.H., Drullinger, L.F., Robetorye, R.S., Pereira-Smith, C.M., and Smith, J .R. (1991) Proc. Natl. Acad. Sci. USA 88, 11012. Wright, W.E., Pereira-Smith, C.M., and Shay, J.W. (1989) Mol. Cell Biol. 9, 3088-3092. Brown, J .P., Wei, W., and Sedivy, J.M. (1997) Science 277, 831-834. Ide, T., Tsuji, Y., Nakashima, T., and Ishibashi, S. (1984) Exp. Cell Res. 150, 321- 328. Shay, J .W., Pereira-Smith, C.M., Wright, W.E. (1991) Exp. Cell Res. 196, 33-39. Gerace, L., and Burke, B. (1988) Ann. Rev. Cell. Biol. 4, 335-374. Maul, G.G. (1997) Int. Rev. Cytol. Suppl. 6, 75-186. Moreland, M., Langevin, G.L., Singer, R.H., Garcea, R.L., and Hereford, L.M. (1987) Mol. Cell. Biol. 7, 4048-4057. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 38 Breeuwer, M., and Goldfarb, D.S. (1990) Cell 60, 999-1008. Silver, P. (1991) Cell 64, 489-497. Zhao, L-J., and Padmanabhan, R. (1988) Cell 55, 1005-1015. Richardson, W.D., Mills, A.D., Dilworth, S.M., Laskey, R.A., and Dingwall, C. (1988) Cell 52, 655-664. Davis, LL (1995) Ann. Rev. Biochem. 64, 865-896. Pante, N., and Aebi, U. (1996) Crit. Rev. Biochem. Mol. Biol. 31, 153-199. Daneholt, B. (1997) Cell 88, 585-588. Robbins, J., Dilworth, S.M., Laskey, R.A., and Dingwall, C. (1991) Cell 64, 615- 623. Newmeyer, D.D., Finlay, DR, and Forbes, DJ. (1986) J. Cell Biol. 103, 2091- 2102. Adam, S.A., Marr, RS, and Gerace, L. (1990) J. Cell Biol. 111, 807-816. Siomi, H., and Dreyfuss, G. (1995) J. Cell Biol. 129, 551-560. Pollard, V.W., Michael, W.M., Nakielny, S., Siomi, H., Wang, F., and Dreyfuss, G. (1996) Cell 86, 985-994. Fridell, R.A., Truant, R., Thorne, L., Benson, RE, and Cullen, BR. (1997) J. Cell Sci. 110, 1325-1331. Aitchison, J .D., Blobel, G., and Rout, MP. (1996) Science 274, 624-627. Michael, W.M., Eder, PS, and Dreyfuss, G. (1997) EMBO J 16, 3587-3598. Gorlich, D., Prehn, S., Laskey, R.A., and Hartmann, E. (1994) Cell 79, 767-778. Irnamoto, N., Tachibana, T., Matsubase, M., and Yoneda, Y. (1995) J. Biol. Chem. 270, 8559-8565. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 39 Moroianu, J., Blobel, G., and Radu, A. (1996) Proc. Natl. Acad. Sci. USA 93, 6572-6576. Moroianu, J., Blobel, G., and Radu, A. (1995) Proc. Natl. Acad. Sci. USA 92, 2008-2011. Melchior, F., Paschal, B.M., Evans, J ., and Gerace, L. (1993) J. Cell. Biol. 123, 1649-1659. Moore, M.S., and Blobel, G. (1993) Nature 365, 661-663. Kutay, U., Izaurralde, E., Bischoff, F.R., Mattaj, I.W., and Gorlich, D. (1997) EMBO J. 16, 1153-1163. Moore, M.S., and Blobel, G. (1994) Proc. Natl. Acad. Sci. USA 91, 10212-10216. Paschal, B.M., and Gerace, L. (1995) J. Cell Biol. 129, 925-937. Rexach, M., and Blobel, G. (1995) Cell 83, 683-692. Hunter, T., and Karin, M. (1992) Cell 70, 375-387. Moll, T., Tebb, G., Surana, U., Robitsch, H., and Nasmyth, K. (1991) Cell 66, 743-758. Baeuerle, P.A., and Baltimore, D. (1995) Cell 87, 13-20. Verrna, I.M., Stevenson, J., Schwarz, E.M., Van Antwerp, D., and Miyamoto, S. (1995) Genes Devel. 9, 2723-2735. Ghosh, S., and Baltimore, D. (1990) Nature 344, 678-682. CHAPTER II. The Human LGALS3 (Galectin-3) Gene: Determination of the Gene Structure and Functional Characterization of the Promoter 40 AB TRACT Galectin-3 is a B-galactoside-specific lectin that is a pre-mRNA splicing factor. We report here the genomic organization of the human LGALS3 (galectin-3) gene and functional characterization of the promoter. Southern blot analysis of genomic DNA revealed that galectin-3 is coded by a single gene in the human genome. The gene is composed of six exons and five introns, spanning a total of ~17 kilobases (kb). Based on primer extension and ribonuclease protection analyses, there are two transcription initiation sites located 52 and 50 nucleotides (nt) upstream of the exon I-intron 1 border, and defined here as +1a and +1b, respectively. The translation start site is in exon II. The ribonucleoprotein-like N-terrninal domain, containing the proline-glycine-alanine- tyrosine (PGAY) repeat motif, is found entirely within exon III. The carbohydrate recognition sequence is found entirely within exon V. Genomic fragments encompassing -836 to +141 nt (relative to +1a) have significant promoter activity when linked to the luciferase reporter gene and transiently transfected into HeLa cells or human diploid fibroblasts. Quiescent fibroblasts have low promoter activity but the activity increases 100 fold following serum addition. Serum responsive activation regions in the promoter are located between -513 to -339 nt and between -339 to -229 nt; an additional activation region may be located between -105 to -15 nt. Because galectin-3 is an immediate-early gene whose expression is dependent on the proliferative state of the cell, this study provides the basis for determining the molecular mechanisms of transcriptional regulation in neoplasia or cellular senescence. 42 INTRODUCTION Galectin-3 is a member of a family of B-galactoside-specific lectins found in many species and cell types (1). The human galectin-3 polypeptide (Mr ~29,000) consists of two domains: a carboxyl-terminal domain that contains the conserved galectin carbohydrate recognition sequence and an amino-terminal domain that contains a proline- glycine-alanine-tyrosine (PGAY) motif repeated five times (2-6). The amino-terminal domain has structural features seen in certain polypeptides from ribonucleoprotein (RNP) complexes, including hnRNP-Cl and SAP62, the homologue of the yeast splicing factor PRP11 (7, 8). Subcellular fractionation studies showed that the majority of galectin-3 is found in the cytoplasm and nucleus associated with RNP complexes (9, 10). Within the nucleus, galectin-3 is found throughout the nucleoplasm as determined both by immunofluorescence and by immunogold electron microscopy (10). The nuclear staining is sensitive to ribonuclease, but not deoxyribonuclease treatment. In nuclear matrix preparations enriched in RNP, galectin-3 yielded a staining pattern similar to that of the snRNP antigen Sm (10), and the non-snRNP splicing factor SC35 (11). Using a cell-free pre-mRNA splicing assay, we demonstrated that splicing activity in nuclear extracts of human HeLa cells was inhibited by disaccharides with high affinity for galectin-3 (lactose and thiodigalactoside) (12). Moreover, splicing activity is lost when galectin-3 was depleted from the extract by either lactose-Sepharose affinity chromatography (12), or by selective antibody depletion (11). Splicing activity is restored when recombinant galectin-3 is added back to the depleted extract. These data demonstrate that galectin-3 is a pre-mRNA splicing factor. 43 Galectin-3 expression is dependent on the proliferative state of the cell and as a function of the cell cycle. There is intense immunofluorescence staining for galectin-3 in sparse proliferating cultures of both mouse 3T3 fibroblasts and normal human fibroblasts, whereas staining is absent from quiescent growth-arrested cells (13-15). Galectin-3 expression is increased in transformed cells in culture (16), and in cells with increased metastatic potential from both human breast and colon carcinoma (17, 18). When serum- starved, quiescent mouse 3T3 fibroblasts are stimulated with serum, immunofluorescence staining for galectin-3 increases within eight hours and precedes DNA synthesis (14). Based on northern blot and nuclear run-off analyses, galectin-3 mRNA levels are nearly undetectable during serum starvation but increase rapidly (within 30 minutes) following the addition of serum (16). This increase in galectin-3 expression is insensitive to cycloheximide, indicating that prior protein synthesis is not required for activation of the LGALS3 (galectin-3) gene. Thus, galectin-3 expression is regulated at the level of transcription and LGALS3 is characterized as an immediate early gene (IEG). Proliferation-dependent expression of galectin-3 was further studied in normal human diploid fibroblasts (HDF) as a function of the cell’s replicative capacity. In senescent HDF, incapable of carrying out DNA synthesis and cell division, galectin-3 expression and subcellular distribution is altered. Following serum addition to serum- starved synchronized senescent HDF, the characteristic nuclear localization of galectin-3 is absent. In contrast to early passage HDF with a high replicative capacity, galectin-3 protein and mRNA levels are elevated in senescent HDF during serum starvation and decrease after serum addition (15). Although many genes are specifically expressed or exhibit either an increased or decreased expression in cellular senescence (see, for 44 example, references (19-22)), the observation that galectin-3 expression increases and decreases in senescent HDF due to the absence or presence, respectively, of serum stimulation, places galectin-3 in a unique category from other known senescent-specific gene expression variants. The human galectin-3 cDNA has been cloned in many different laboratories (2-6). However, the gene structure and characterization of the promoter is unknown. We report here the structure of the human LGALS3 gene and establish functional promoter activity for the 5’ flanking region in transient transfections of HeLa cells and in response to serum activation in HDF. This study provides the basis for determining the mechanisms of transcriptional regulation of the human LGALS3 gene, and in particular, for studying altered galectin-3 expression in neoplasia and/or cellular senescence. 45 MATERIALS AND METHODS Nomenclature of galectin-3 cDNAs. A more uniform nomenclature, i.e., galectins, was recently adopted to identify these B-galactoside-specific lectins of mammalian origin (1). Galectin-3 has been studied in various laboratories under different names: CBP35 (23), Mac-2 (3) and IgE binding protein (IgE BP) (2). Because slight differences exist in the cDNA nucleotide sequences cloned from the different labs, the previous name (CBP35, Mac-2, IgE BP) of galectin-3 will be used in this manuscript when making reference to a specific cDNA or a specific nucleotide sequence. Basepair numbers associated with an oligonucleotide are given relative to the beginning of the sequence given in the GenBankTM/EMBL database. Cell culture and synchronization. Human cervical carcinoma HeLa cells (American Type Culture Collection (ATCC) CCL2) were kindly provided by Dr. R. Patterson (Department of Microbiology, Michigan State University) and grown as monolayer cultures in Dulbecco’s modified Eagle’s medium with high glucose (DME-HG) containing 10% fetal bovine serum, 100 U/ml penicillin, and 100 ug/ml streptomycin at 37°C in 5% C02. The HeLa cells were grown to ~80% confluence and serially passaged at a split ratio of about 1:8. The diploid human foreskin fibroblast cell strain LGl was kindly provided by Dr. J. McCormick (Carcinogenesis Laboratory, Michigan State University) (24). These cells were cultured in complete growth medium (MEM, 1.8 mM CaClz, 10% bovine calf serum) at 37°C in 5% C02. The LGl cells were seeded (3.5 x 103 cells/cmz, approximately 20% confluent) and serially passaged at a split ratio of 1:4. The cells were made quiescent by remvoing the complete growth medium, washing twice 46 with phosphate-buffered saline (PBS; 140 mM NaCl, 2.68 mM KCl, 10 mM NazHPO4, 1.47 mM KH2P04, pH 7.4), and incubating in starvation medium (MEM, 0.1 mM CaClz, 0.2% bovine calf serum) for 72 h (25). Quiescence was verified in separate experiments by cell counting and using fluorescence activated cell sorting analysis (FACS), carried out by standard methods after staining the cells with propidium iodide (26). Cells were stimulated to reenter the cell cycle in a synchronous fashion by the addition of complete growth medium. The onset of DNA synthesis was determined by the incorporation of [3H]thymidine. Briefly, cell cultures seeded in 15 x 80 mm dishes were pulse-labeled for two hours with 20 uCi [3H]thymidine (2 Ci/mmol), washed extensively with PBS, detached from the growth surface with 0.5% trypsin, and collected by vaccum filtration over a Whatman GF/C filter. The cells were washed three times with ice-cold PBS and then washed sequentially with 10% ice-cold trichloroacetic acid and methanol. The filters were air dried and subjected to liquid scintillation counting. Antibody reagents and other materials. Anti-Mac-2 is a rat monoclonal antibody directed against an epitope mapped to the amino-terminal domain of galectin-3 (27, 28). The antibody was purified from a cell culture supernatant derived from the hybridoma line M3/38.1.2.8.HL.2, obtained from ATCC (T IB 166). Anti-actin is a polyclonal rabbit antiserum raised against calf thymus actin. Radionucleotides [or-32P]-dCTP (3000 Ci/mmol), [or-32P]-CTP (800 Ci/mmol), [y-3ZP]- ATP (6000 Ci/mmol), and [a3SS]-dATP (~1500 Ci/mol) were purchased from New England Nuclear, Boston, MA. Oligonucleotides used as probes and primers were custom synthesized in the Macromolecular Structure Facility in the Department of Biochemistry at Michigan State University. 47 Qnomic libm screening. A human genomic library (Stratagene, La Jolla, CA) contained fragments of genomic DNA cloned into the kFixII vector after partial digestion with the restriction endonuclease Sau3AI of genomic DNA isolated from the human lung fibroblast cell line WI-38. Two probes were used to screen the library: (i) the 880 bp mouse CBP35 cDNA (23), which begins at the second codon and extends in a 3’ direction 80 bp past the TAA stop codon, and (ii) a 440 bp fragment from the 5’ end of clone hPCRl (see below) generated by PCR amplification using the primer pairs 5’- GAGCCAGCCAACGAGCGGT-3’ and 5’-GCCGTCTGG'ITTGCTGAGCGAG—3’. Each probe was labeled with a32P-dCTP to a specific activity of ~1 x 109 cpm/ug using a random primed DNA labeling kit (Boehringer-Mannheim, Indianapolis, IN) and the method of Feinberg and Vogelstein (29). Plaque lifts were made using BA-S nitrocellulose membranes (Schleicher and Schuell, Keene, NH) and processed according to the manufacturer. Prehybridization, hybridization, and washing conditions were performed by standard methods (30). Putative positive clones were identified and purified by a secondary and tertiary screen. Polmerase chain reaction (PCR) amplification of genomic DNA. All reactions were carried out in a Perkin-Elmer-Cetus 9600 Thermal Cycler using MicroAmp tubes, 25 picomoles of each primer, 200 M of each dNT P, in a final volume of 100 pl. Products from each reaction were visualized by ethidium bromide staining following agarose gel electrophoresis and isolated by using either GeneClean glassmilk beads (BiolOl, Vista, CA) or Qiaex resin (Qiagen, Chatsworth, CA). Two distinct PCR methods were used to amplify genomic DNA containing portions of the galectin-3 gene and specific primers 48 were chosen based on a predicted structural conservation between the human and mouse galectin-3 genes: (i) four pg of human genomic DNA, isolated from disease-free whole blood (Novagen, Madison, WI), were amplifed by the method of Cheng et al. (31) using the XL PCR kit (Perkin-Elmer) and the following oligonucleotide primer pair: (Mac-2; bp 2 to 18, with two additional nucleotides (GT) at the 3’ end; 5’- GAGCCAGCCAACGAGCGGT-3’) and (IgE BP; reverse-complement bp 287 to 308; 5’-GCACTTGGCTGTCCAGAAGATG-3’). The reaction mixture was incubated at 80°C for 5 minutes, 2.5 Units thh DNA polymerase XL was added, followed by 14 cycles of the following: 94°C x 15 seconds for denaturation, 68°C x 10 minutes for annealing and extension. This was followed by 16 cycles of the following: 94°C x 15 seconds, 68°C x 10 minutes (with an increase of 15 seconds per cycle thereafter). After a total of 30 cycles, the reaction was allowed to extend at 68°C for 15 minutes and then immediately ramped to 4°C. (ii) Four pg human genomic DNA were amplified using the following oligonucleotide primer pairs: (IgE BP; bp 43 to 67; 5’-GATGCGTI‘ATCI‘GGGTCTG- GAAACC-3’) and (IgE BP; reverse-complement bp 405 to 429; 5’-CGTGCCCA- GAATI’G'ITATCAGCATG-3’). The genomic DNA template was denatured at 95°C for 5 minutes, 2.5 Units Taq polymerase (Gibco/BRL, Gaithersberg, MD) was added, followed by 30 cycles of the following: 97°C x 1 minute, 55°C x 1 minute, 72°C x 2 minutes. After 30 cycles, the reaction was allowed to extend at 72°C for 5 minutes and then immediately ramped to 4°C. One microliter (1/ 100 volume) was next subjected to a second round of PCR amplification using the following nested primers: (IgE BP; bp 90 to 111; 5’-CGCATGGGGGAACCAGCCTGCT-3’) and (IgE BP; reverse-complement bp 49 392 to 413; 5’-ATCAGCATGCGAGGCACCACTC-3’). The same cycling protocol was used as above. Subcloning and DNA gcguencing. The purified genomic clones identified by library screening were amplified and mapped with the restriction endonucleases BamHI, EcoRI, EcoRV, HindIII, and Natl. Individual restriction fragments were probed with the mouse CBP35 cDNA and various Oligonucleotides complementary to the human IgE BP cDNA. Based on the hybridization patterns obtained, these restriction fragments were subcloned into pBluescript II SK+ or pBluescript II KS+ vectors (Stratagene). In some cases, nested deletion mutants were prepared by standard methods using exonuclease III and SI nuclease digestions (30). The genomic clones obtained by PCR amplification were initially subcloned into the pCRII vector (Invitrogen, San Diego, CA) and restriction fragments were then subcloned into pBluescript II SK+. Sequencing of double-stranded DNA was carried out by the dideoxy chain termination method (32) using a35S-dATP and either deltaTaq Version 2.0 DNA polymerase (United States Biochemical/Amersham Life Sciences, Cleveland, OH) or Sequenase (USB/Amersham), or by automated cycle sequencing in the DNA Sequencing Facility at Michigan State University. Sequence data were compiled and analyzed using Genetics Computer Group (GCG) (Madison, WI) software. The nucleotide sequence has been submitted to the GenBank/EMBL Nucleotide Sequence Database under Accession Numbers AF031421, AF031422, AF031423, AF031424, and AF031425. Transcription factor binding sites were found by visual inspection, by downloading the database (fetch tfsites.dat) and using Findpattems_-dat=tfsites.dat in GCG, or from the web site http://www.gsf.de/cgi- bin/matsearch.pl (33). 50 Isolation of HeLa cell mlyIA)+ RNA. HeLa cells, known to express galectin-3 (2, 12), 2. The cells were washed were grown in T-150 flasks to a density of ~1 x 105 cells/cm twice in ice-cold PBS and isolated by scraping with a rubber policeman. A total of ~2 x 108 cells were collected by centrifugation at 300 x g for 5 minutes, washed once with ice- cold PBS, and the poly(A)+ RNA was isolated by using the PolyATtract System 1000 kit (Promega, Madison, WI). The concentration and purity were determined by UV spectrophotometry at 260 and 280 nanometers. The poly(A)+ RNA was stored in H20 at - 80°C until use. Primer extension analysis. An oligonucleotide primer corresponding to the reverse complement of the sequence from 11 nt at the 5’ end of exon III and extending to 16 nt at the 3’ end of exon 11 (5’-GCATCATGGAGCGAAAAATTGTCTGCC-3’) was used. The primer was 5’ end-labeled with 'fZP-ATP using T4 polynucleotide kinase (to a specific activity of ~2 x 107 cpm/ug). Following overnight sodium acetate/ethanol precipitaion at -20°C, the labeled primer was pelleted, washed twice in 70% ethanol, and dissolved in 20 I11 H20. An aliquot (2 x 105 cpm) of the labeled primer was hybridized with 6 pg of HeLa cell poly(A)’r RNA or 6 ug yeast tRNA for 18 hours at 30°C in 1 M NaCl, 10 mM Tris pH 7.4, 1 mM EDTA. The hybridization reaction was diluted to 200 pl with H20 and sodium acetate/ethanol precipitated. The annealed primer/RNA was pelleted, washed twice in 70% ethanol, and dissolved in 12 It] H20. The annealed primer was extended using 8 Units Tth DNA polymerase (Boehringer-Mannheim) at 70°C for 15 minutes in 1 mM MnCl2 and 250 M of each dNTP. The reaction was diluted in 10 mM Tris pH 8, 0.1 mM EDTA, extracted with Tris-buffered (pH 8) phenol:chloroform:isoamyl alcohol and the aqueous phase isolated and sodium 51 acetate/ethanol precipitated. The pellet was washed with 70% ethanol, dried for 2 minutes in a Speed-Vac, and dissolved in 2 It] H20 and 3 ul formamide gel loading buffer. The sample was heated for 3 minutes in a 95°C sand bath and analyzed on a 7 M urea/6% polyacrylamide gel. The gel was fixed in 12% methanol/10% acetic acid and autoradiographed at -80°C using an intensifying screen. An antisense oligonucleotide primer (IgE BP; reverse-complement bp 90 to 111; 5’-AGCAGGCTGG'ITCCCCCAT- GCG-3’) in exon III, prepared and extended exactly as described above, was also used for primer extension analysis. An unrelated sequencing reaction was used to determine the size of the extended product for two reasons: (i) the primer used for extension is ~8 kb downstream in the genomic sequence, and (ii) this region of the gene is refractory to manual DNA sequencing. However, given the difference in DNA mobility on a denaturing polyacrylamide gel due to nucleotide composition (1' 1-2 nt), the exact size of the extended product reported is within this range. Ribgnuclease protection. Two Oligonucleotides, (5’-TI'I'I'I'ITCTAGACCTGGTCCG- GGGAGAGGACTG-3’) and (5’-TI‘I'I'ITGAATI‘CCGCTCG'ITGGCTGGCTCCG-3’), were used to PCR-amplify a fragment extending from the exon I-intron 1 border to 166 nt upstream. The fragment generated was double-digested with the restriction endonucleases XbaI and EcoRI and directionally subcloned into pBluescript H SK+. The authenticity of the subclone was verified by DNA sequencing. An antisense strand radiolabeled RNA probe was prepared by linearizing the subclone with EcoRI and transcribing with T7 RNA polymerase (Boehringer-Mannheim) in the presence of a32P- CTP. Similarly, sense strand RNA probes (radiolabeled and unlabeled) were prepared by linearizing the subclone with XbaI and transcribing with T3 RNA polymerase 52 (Boehringer-Mannheim) with or without a32P-CTP. The radiolabeled probes (1 x 10° cpm) were annealed to 4 pg HeLa cell poly(A)+ RNA, 4 ug yeast tRNA, or unlabeled sense-strand RNA probe, in a hybridization buffer (80% formamide, 40 mM PIPES pH 6.4, 0.4 M NaCl, 1 mM EDTA) for 16 hours at 40°C. The reaction mixture was digested with 36 units RNase One (Promega) for 30 minutes at 37°C. Protected fragments were visualized on a denaturing 7 M urea/6% polyacrylamide gel. RNA probes of known size (51 and 44 ribonucleotides) (generated by run-off transcription as described above), end- labeled pBR322/Msp1 fragments, and a DNA sequencing ladder were used as size markers. Because RNA migrates more slowly than similarly sized single-strand DNA through a denaturing polyacrylamide gel, the mobility of the RNA run-off transcript markers was used as a control for sizing. Compared to the DNA sizing ladder, the RNA markers migrated between 3-6% slower, similar to what is commonly reported (34). cDNA isolation by RT-PCR. Eighty ng of the HeLa cell poly(A)+ RNA was reverse- transcribed and PCR-arnplified using AMV reverse transcriptase (5 Units) and Tfl DNA polymerase (5 Units). The reactions were carried out in a Perkin-Elmer-Cetus 9600 thermal cycler using MicroAmp tubes and a final volume of 100 pl. The buffer contained 1.5 mM MgS04, 200 M of each dNTP, and 25 pmol each of an oligonucleotide primer complementary to an exon I sequence (5’-GCAGCCGTCCGGAGCCAGCC-3’) and an exon VI sequence (5’-CATGGTATATGAAGCACTGGTGAGG—3’). The following cycling protocol was used: 48°C x 5 minutes; add enzymes, 48°C x 45 minutes; 94°C x 2 minutes; then 40 cycles of 94°C x 30 seconds, 65°C x 45 seconds, 72°C x 90 seconds; ending with 72°C x 5 minutes. The 783 bp band was excised from a 1.5% agarose/TEE gel with GeneClean glassmilk beads (BiolOl) and cloned into the pGEM-T vector 53 (Promega). The product was authenticated as the human galectin-3 cDNA by DNA sequencing. MW Ten pg of human genomic DNA (Novagen) were digested individually with the restriction endonucleases BamHI, EcoRI, or HindIII. The digests were electr0phoresed through a 1% agarose/TAB gel and Southern transferred onto a Nytran membrane (Schleicher and Schuell). The DNA was UV-crosslinked with a Stratalinker Model 1800 (120,000 pJoules/cmz) and incubated in prehybridization solution (6x SSPE, 37% formamide, 5x Denhardt’s, 0.1% SDS, and 100 pg/ml salmon sperm DNA; see reference (30)) for 4 hours at 42°C. A 256 bp probe was isolated by PCR-amplification of the human galectin-3 cDNA by standard methods using the following primer pair in exon IH: 5’-GATGCGTI‘ATCTGGGTCTGGAAACC-3’ and 5’-GCACTTGGCTGTCCAGAAGATG-3’. The probe was labeled by random priming (specific activity = 109 cpm/pg) and added to the hybridization solution (same as the prehybridization solution without the salmon sperm DNA) at 107 cpm/ml and incubated for 24 hours at 42°C. The membrane was then washed to high stringency (0.1x SSC, 0.5% SDS, 42°C, 30 minutes; see reference (30)), dried, and autoradiographed at -80°C with an intensifying screen for 3 days (and an additional 14 days). Lambda DNA digested with HindIII was used as a size marker. Promoter-remrter gene construct preparation. A 1.2 kb NotI-HindIII fragment from the genomic clone M6 was subcloned into pBluescript II SK+. This fragment extends ~400 bp upstream of the exon I/intron I border. To study 5’ flanking sequences that extend further upstream, a NotI-NotI fragment from clone M6 (the 5’ NotI site in this fragment is from the XFixII vector; it is not located in the human galectin-3 gene) which 54 extends 0.6 kb 5’ to the 1.2 kb NotI-HindIII M6 fragment was isolated. This 0.6 kb NotI- Natl M6 fragment was ligated into the 1.2 kb Notl-HindIII M6 subclone to generate a 1.8 kb NotI-NotI-HindIII subclone in pBluescript II SK+ (the correct orientation was determined by DNA sequencing). For this NotI-NotI-HindIII subclone, a AFixII SalI site lies 28 bp 3’ to the 5’-most NotI site; a pBluescript II SalI site lies 16 bp 3’ to the HindIII site. In order to remove the remaining Mixll sequence and to take advantage of convenient restriction enzyme sites for nested deletion mutagenesis, the ~1.8 kb SalI-SalI fragment was removed, blunt-ended, and inserted into the SmaI site of pBluescript II SK+. Nested deletions were made at both the 5’ and 3’ ends of the genomic fragment using exonuclease III and S1 nuclease. The 5’ end of each deletion mutant was blunt- ended by Klenow fill-in; the 3’ end of each mutant was flanked by a Sad site 4 bp downstream in pBluescript II SK+. The deletion mutants were then directionally subcloned into the SmaI and SacI sites in the luciferase reporter vector pGL2-basic (Promega). The 5' and 3' ends of the genomic fragment in the reporter construct were determined by DNA sequencing and are designated relative; to the transcription start site, defined as +1. Determination of promoter activity. Each human galectin-3 promoter-reporter construct was transiently transfected into HeLa cells. To normalize for differences in transfection efficiency, each construct was co-transfected with the pSV-B-gal vector (Promega), a plasmid expressing B-galactosidase (B-gal) driven by the SV40 promoter/enhancer. Differences in luciferase expression (after normalization with B-gal activity) are given relative to the pGL2-basic vector which contains no human galectin-3 promoter sequence. For each HeLa cell transfection, 0.5 pg pGalectin-3-luc construct (or 55 pGL2-basic), 0.5 pg pSV-B-gal, and 6 pl LipofectAMlNE (Life Technologies, Gaithersburg, MD) were added to antibiotic-free and serum-free DME-HG and incubated for 30 minutes at room temperature to allow liposome formation. One ml of this transfection solution was added to a 35 mm well of HeLa cells (4.4 x 104 cells/cmz, plated 16 hours prior) and incubated at 37°C, 5% CO2 for 6 hours. One ml of antibiotic-free 20% fetal bovine serum/DME-HG was then added to each 35 mm well (final serum concentration = 10%) and incubated for 18 hours. This transfection media was removed and replaced with 10% fetal bovine serum/DME-HG containing penicillin (100 Units/ml)-streptomycin (100 pg/ml) and incubated for an additional 24 hours. The transfected HeLa cells were then incubated in the 35 mm well with 200 pl of 1x reporter lysis buffer (Promega) for 15 minutes at room temperature, collected by scraping with a rubber policeman, and sonicated for 15 seconds on ice. A similar set of transient transfections, using the same promoter-reporter constructs, was also carried out on diploid human foreskin fibroblasts, M31 (24). The cells used were pre-senescent with a cumulative population doubling of ~30. For these transfections, 2.5 pg pGalectin-3-luc construct (or pGL2-basic) and 5 pl SuperFect reagent (Qiagen) were added to 300 pl antibiotic-free and serum-free minimum essential medium (MEM) and incubated for 10 minutes at room temperature. Three ml of complete growth media (MEM, 1.8 mM CaCl2, 10% fetal calf serum) were added and 3.3 ml of this DNA/SuperFect transfection media were added to each 80 x 15 mm plate of LG] cells (7 x 103 cells/cmz, approximately 40% confluent, plated 24 hours prior) and incubated at 37°C, 5% CO2 for 6 hours. The transfection media was removed, the cells washed twice with PBS, and then replaced with 8 ml of complete growth media and 56 incubated for 12 hours. These cells were then made quiescent (as determined in separate experiments using propidium iodide staining and fluorescence activated cell sorting analysis) by removing the complete growth media, washing twice with PBS, and incubating in starvation media (MEM, 0.1 mM CaCl2, 0.2% fetal calf serum) for 72 hours. Complete growth media was then added back to these cells for 24 hours. The transfected LG] cells were harvested by scraping with a rubber policeman and lysates prepared by incubating in 100 pl of 1x reporter lysis buffer followed by sonication. The HeLa and LG] cell lysates were centrifuged at 12,000 x g for 2 minutes at 4°C and the supernatant isolated for luciferase and B-gal assays. The luciferase activity was determined by adding 20 pl of the cell lysate to 100 pl of the assay reagent (20 mM tricine, 1.07 mM [MgC03]4Mg[0H]2'5H20, 2.67 mM MgSO4, 0.1 mM EDTA, 33.3 mM DTT, 270 pM coenzyme A, 530 pM ATP, 470 pM luciferase, pH = 7.8) and luminescence was measured in a Model TD-20e luminometer (Turner Designs, Sunnyvale, CA). The B-gal activity was measured fluorometrically using an assay system with 4-methylumbelliferyl-B-D—galactoside (4-MU) as the substrate. Briefly, 40 pl of cell lysate were incubated with 160 pl of 4-MU reaction buffer (25 mM Tris-HC1 pH 7.5, 125 mM NaC], 2 mM MgCl2, 12 mM 2-mercaptoethanol, 0.3 mM 4-MU) at 37°C for 30 minutes. The reaction was stopped with 50 pl of 25% (w/v) trichloroacetic acid and the samples were centrifuged at 12,000 x g for 2 minutes at room temperature. A 100 p] aliquot of the supernatant was then added to 2 ml of stop buffer (133 mM glycine, 83 mM sodium carbonate, pH = 10.7), mixed, and B-gal activity was measured in a TKO 100 mini-fluorometer (Hoefer Instruments, San Francisco, CA). 57 Preparation of LGl cell extracts and immunoblot analysis. The LG] cells (~8 x 103 cells/cmz, ~50 confluent) were washed twice with PBS, detached from the growth surface with 0.5% trypsin, and collected by centrfugation. The cells were washed three times with PBS, incubated in a modified Laemmli buffer (10% glycerol, 2 % SDS, 0.25 M sodium phosphate, pH 6.8, 10 mM B-mercaptoethanol), sonicated 4 x 15 sec on ice, and then incubated for 15 min in a sand bath at 95°C. Extracts from an equal number of cells were loaded into each lane, resolved by 12.5% SDS-PAGE, and electrophoretically transferred to a PVDF membrane (Irnmobilon-P, Millipore, Bedford, MA) using a transfer buffer containing 25 mM Tris, 193 mM glycine, and 10% methanol. The membrane was blocked for several hours in T-TBS (10 mM Tris, pH 7.5, 0.5 M NaCl, 0.05% Tween 20) containing 10% non-fat dry milk. After several brief washes with T- TBS, the membrane was incubated with the primary antibody, diluted in 1% non-fat dry milk/T-TBS for one h at room temperature. After five washes (15 min each) with T-TBS, the membrane was incubated in 1% non-fat dry milk/T-TBS containing secondary antibody conjuated with horseradish peroxidase for one h at room temperature. After extensive washing with T-TBS, the polypeptides were visualized using the Renaissance chemiluminescence detection system (New England Nuclear Life Sciences). 58 RESULTS Isolation and characterization of human galectin-3 genomic clones. Two positive clones were identified by screening a genomic library with the mouse CBP35 cDNA, and are designated M3 and M5 (Figure 1). Because clone M3 did not extend into the 5’ end of the gene, human genomic DNA was amplified by PCR using a primer complementary to the 5’ end of the human Mac-2 cDNA (bp 2 to 18) and a primer complementary to a sequence in the middle of the human IgE BP cDNA (bp 287 to 308), which was believed to be in exon III of the human gene, based on a predicted structural similarity between the mouse gene (whose structure is known; references (35-37)) and the human gene. An ~8.9 kb fragment was amplified, cloned, and is designated hPCRl. Similarly, a 2.3 kb fragment, designated hPCR2, was amplified by using a primer complementary to a sequence in the IgE BP cDNA thought to be in exon 1]] (bp 90 to 111) and to a primer from exon IV (bp 393 to 413). A 440 bp genomic fragment at the 5’ end of hPCRl, generated by PCR amplification, was used as a probe to re-screen the genomic library and several positive clones were identified, including M6. Individual M5, 7th3, hPCR], hPCR2, and M6 restriction enzyme fragments were probed with either the mouse CBP35 cDNA or with Oligonucleotides complementary to the human IgE BP cDNA sequence. Based on the hybridization patterns, fragments were subcloned and sequenced. Structure of the human galectin-3 gene. Comparison of the genomic sequence with the five previously published human galectin-3 cDNA sequences (2-6) reveals that the gene consists of six exons and five introns and spans a total of approximately 17 kb (Figure 1). Two Alu family middle repetitive sequences were found (38). As will be described 59 Figpre l. Genomic organization of the human LGALSJ gene. The Roman numerals indicate the positions of exons. The dark boxes indicate exon sequences corresponding to the open reading frame. The hatched box represent the 3’ untranslated region (UTR). The white boxes represent the 5’-UTR, and the white ovals represent Alu-family middle repetitive sequences. Introns and flanking sequences are represented by thin lines. The entire gene is encompassed by five genomic clones, M6, hPCRl, hPCR2, M3, and M5, and each is represented by a shaded rectangle. The (~) indicates a gap in Ah3 and M5-- the 3’ ends of M3 and M5 are omitted for clarity. A partial restriction endonuclease map of these clones is given: H, HindIII; R, EcoRI, RV, EcoRV, B, BamI-II. The EcoRV site indicated is the only one mapped—there may be additional sites. l numeral iondinglo ll] (17R); ly middle ncs. lit and iii. nd illi- PSSC map )R V Slit 60 2kb 39 R I hh6 lthS —~_ hPCR2 hh3 hPCRl 61 below, the transcription initiation site defines the beginning of exon 1. The first ATG or CT G trinucleotide sequence encountered by scanning downstream from the transcription initiation site, and by comparing the sequence to the cDNA, is an ATG sequence found in exon 11. This ATG sequence forms a consensus Kozak box (CGGAAAATGG) (39) and is designated the translation start site (Table I). The stop codon (TAA) and 3’ untranslated sequence is in exon VI. Exon VI extends 130 bp to an AAUAAA sequence, designated here as the human galectin-3 polyadenylation signal. There is another AAUAAA sequence located ~400 bp further downstream. None of the published human galectin-3 cDNAs possess the sequence between the first and second AAUAAA sequence, suggesting that only the 5’ most AAUAAA is used as the polyadenylation signal. Furthermore, the size of the human galectin-3 mRNA on northern blots (~1.1 kb) is consistent with the use of the 5’-most AAUAAA. In each case, near-consensus exon- intron donor (5’-GTAAGT) and acceptor [~6 pyrimidines-NCAG-3’] splice site sequences are present, obeying the GT/AG rule (40) (Table I). There is one split codon; the 3’-most codon in exon IV is incomplete. Splicing of the intron between exon IV and exon V makes the three nucleotide codon (AGA) for an arginine residue. This restoration of a codon by a splicing event is seen in one exon in other members of the galectin gene family (35, 41-43). The human LGALS3 genomic sequence reported here, using exon sequences covering the translation start site to the stop codon, gives an open reading frame of 750 bp corresponding to a deduced polypeptide of 250 amino acids. Three of the five previously published cDNA sequences also code for a protein of 250 amino acids. However, one published cDNA (3) codes for 248 amino acids and another cDNA (5) codes for 242 _> > shavessémueeaaaaez..finu--uauauuaaeuem—u<< 28 a :88 50,53 :88 5% 05 830%.: Hoe—oak. 2F .26 gm coma—mg 05 3:889... OH< poem—hove: 2: d :88 E .5 :88 we use .m 05 85.er Rama cows—aeovaba 05 5. Semi oomv H :88 we use .m 05 3:an 86 Sam 553.855 25. .306 283232.5— 5 Sim mm ESE come we 36 0382x293 05 :5 £8339 E eczw m_ :83 some mo 8% 2:. £320— 83332 .3 moose—com .555 05 ”E38— omaocoqm: .3 3.28058 2a mooeonvom opus—oz: :98 2F .eeou MES 53::— 2: ..e Evie: :ehfiéexm — oznh. 63 amino acids. The differences between these two cDNAs and the LGALS3 genomic sequence (and the other three cDNAs) are due to nucleotide differences contained almost exclusively within GC-rich stretches in exon 111. Therefore, it is likely that these discrepancies are due to sequencing errors only. Indeed, the published cDNA coding for 242 amino acids (5) has been corrected in GenBankTM/EMBL and does code for 250 amino acids. One nucleotide difference has been previously noted 274 nt from the 5’ end of exon III (cytosine or adenine), and is considered to be an allelic variant (6). This nucleotide is located in overlapping genomic clones: sequencing of hPCRl gives an adenine at this position whereas sequencing of hPCR2 gives a cytosine. Interestingly, a cytosine at this position would give a CCC codon for a proline residue and another PGAY motif would be formed (for a total of six repeats), whereas an adenine would give an ACC codon for a threonine. Unfortunately, however, the significance of the PGAY repeat motifs is unknown. Southern blot analysis of genomic DNA. A 256 bp PCR fragment from exon III was used as a probe and hybridized to human genomic DNA digested individually with various restriction enzymes. The exon III probe was chosen (instead of the complete cDNA) in order to avoid cross-hybridization with other members of the galectin family. Based on the partial restriction map of the genomic clones (see Figure 1), the exon III probe should hybridize and detect bands of approximately 3.2 kb for BamHI, 2.6 kb for EcoRI, and 2.0 kb for HindIII. Furthermore, if galectin-3 is a single gene, only these bands will hybridize. Indeed, when genomic DNA is digested, Southern blotted and probed, only bands of 3.2 kb (for BamHI), 2.6 kb (for EcoRI), and 2.1 kb (for HindIIl) are present, even after long exposure (14 days) with an intensifying screen (Figure 2). Along Figpre 2. Southern blot analysis of genomic DNA. Human genomic DNA (10 pg) was digested individually with BamHI, EcoRI, and Hindlll. The digests were electrophoresed on a 1% agarose gel, transferred to a supported nitrocellulose membrane, and hybridized with a 256 bp 32P-iabeied probe in exon III. Lane 1, BamHI digest; lane2, EcoRI digest; lane 3, Hindll] digest. Sizes (in kb) of fragments from a HindIII digest of A DNA are indicated to the left. 65 x, _ 3 ‘ .. e ‘0 .P'L.” .l- — .0 L _ _ _ . B 9 r0 4 22 66 with the fact that the cDNA detects a single ~1.1 kb band on northern blotting of poly(A)+ RNA from human fibroblasts (15) and HeLa cells (2), as well as from other human cell types (2), these data are consistent with galectin-3 being a single gene in the human genome. Determination of the transcription initiation site. Because the size of the human galectin-3 message is ~1.] kb, the 5’-untranslated region (5’-UTR) was estimated to be < 220 bp in total length [~1.1 kb - 880 bp (translation start site to the polyadenylation signal) = ~ 220 bp]. In order to verify the length of the 5’-UTR (and making the assumption that no additional intron exists upstream of the putative exon I-intron 1 border), two probes were hybridized to a northern blot of HeLa cell poly(A)’r RNA: a NotI-Ber fragment extending 350 to 124 bp upstream of the putative exon I-intron 1 border and a Ber-XhoI fragment which extends from 124 bp upstream to 141 bp downstream of the putative exon I-intron 1 border (the XhoI site is located in vector sequence an additional 20 bp downstream and is not in the genomic sequence) (see Figure 8 and Figure 8 legend). Only the Ber-Xhol fragment gave a positive hybridization signal (data not shown). Therefore, the 5’-UTR does extend less than ~124 bp upstream of the putative exon I-intron 1 border. The information derived from these data were used to design the RNA probes for RNase protection analysis. An antisense strand RNA probe, extending from the putative exon I-intron 1 border to 166 nt upstream, was synthesized. The total length of this probe is 225 nt (166 nt + 59 nt pBluescript) (Figure 3A, lane 1]) and it is completely digested with RNase (Figure 3A, lane 10). The probe was annealed to HeLa cell poly(A)+ RNA, digested with RNase, and fragments with sizes between 51-53 nt were protected (Figure 3A, lane 6). 67 Fimre 3. Identification of the transcription initiation site by ribonuclease protection and primer extension analyses. (A) Ribonuclease protection analysis. RNA probes corresponding to the genomic sequence extending 166 nt upstream from the exon I-intron 1 border were hybridized to HeLa cell poly(A)+ RNA, yeast tRN A, or cold sense RNA, and digested with ribonuclease. The protected fragments were analyzed on a 7 M urea/6% polyacrylamide gel (see Methods). Lane 1, RNA run-off transcript markers with sizes (in ribonucleotides, rnt) indicated to the left; lanes 2-5, DNA sequencing ladder (M13mp18, -40 primer); lane 6, antisense strand RNA probe annealed to 4 pg poly(A)+ RNA; lane 7, sense strand RNA probe annealed to 4 pg poly(A)+ RNA; lane 8, antisense strand RNA probe annealed to 4 pg yeast tRNA; lane 9, antisense strand RNA probe annealed to cold sense strand RNA (diluted 1:1000 as compared to lanes 6, 7, 8, 10); lane 10, antisense strand RNA probe digested with RNase; antisense strand RNA probe undigested (diluted 1:100 as compared to lanes 6, 7, 8, 10); lane 11; pBR322/Msp1 markers with sizes (in bp) indicated to the right. (B) Primer extension analysis. A 32P- end-labeled antisense primer composed of 16 nt from the 3’ end of exon II and 11 nt from the 5’ end of exon III was used in this experiment, and is shown in the schematic diagram. The primer was annealed to 6 pg HeLa cell poly(A)+ RNA or 6 pg yeast tRNA and extended with Tth DNA polymerase. The extended products were analyzed on a 7 M urea/6% polyacrylamide gel and the sizes determined by comparison to an unrelated DNA sequencing reaction (see Methods). Lanes 1-4, DNA sequencing ladder used for sizing; lane 5, primer annealed to poly(A)+ RNA and extended; lane 6, primer annealed to yeast tRN A and extended. 68 In M N v- e m van. fizzaaha ll .-.E 3 ti-lr III lllv rt Ural.- mm 1. 5.- H lMd llll IIHIIIlI ll mwaaama % V0 MVX ..dx fxm? Vx over? a ; ... we. acre 9 Ideas: 101 nor \t‘llt i \l i ll [[11 adder 1 it an: robe Int mic {ml P m 69 Neither a sense strand 205 nt RNA probe annealed to HeLa cell poly(A)+ RNA and digested, nor the antisense strand probe annealed to yeast tRNA and digested, gave similarly-sized protected bands (Figure 3A, lanes 7 and 8, respectively). In addition, the antisense strand RNA probe was annealed to cold sense probe and digested. Only the expected fragment of 166 nt was protected—no bands corresponding to 51-53 nt in size were observed (Figure 3A, lane 9). Together, these controls indicate that the RNase- protected fragments from the antisense probe annealed to poly(A)+ RNA are not artifacts of the experimental procedure. Furthermore, there is no consensus intron-exon splice acceptor site adjacent to the region ~50 nt upstream of the exon 1- intron 1 border and, therefore, it is unlikely that an intron-exon border has been mapped. The multiple bands of the protected fragments may be due to closely spaced multiple start sites or due to incomplete RNase digestion close to the double-stranded RNAzRNA termini. The position of the transcription start site as determined by RNase protection was confirmed by primer extension analysis (Figure 3B). A exon II-exon III hybrid antisense primer (27 nt) was used. Two bands, 83 and 85 nt in size, were detected (Figure 3B, lane 5). This maps the transcription start site to 50 and 52 nt upstream of the putative exon 1- intron 1 border [(83 or 85 nt) - 6 nt (exon 11 GAAATG) - 27 nt (primer sequence) = 50 and 52 nt]. No primer extended bands with sizes of ~ 83 or 85 nt were observed when the the exon II-exon III hybrid primer was annealed to yeast tRNA (Figure 3B, lane 6). Longer exposure of the gel (21 days) revealed no additional bands, thus indicating that there are no additional start sites further upstream, at least in HeLa cells. These data are in agreement with the RNase protection analysis. Moreover, the genomic sequence in 70 this region is now confirmed as exon 1. The transcription start sites are designated hereafter as +1a and +1b (52 and 50 nt, respectively, from the exon I - intron 1 border). For the primer extension analysis, we had also used an antisense probe from within exon III. However, in the course of this work, studies by Raimond et al. suggested that intron 2 may have promoter activity (44). Thus, it was possible that the exon III antisense primer may map to a putative transcription initiation site within intron 2. Our results showed that this primer yielded, nevertheless, two extended fragments of 147 and 149 nt (data not shown), mapping to the same positions as the exon II-exon III hybrid antisense primer. There were no additional (smaller or larger) bands present on this extension analysis and hence, no transcription initiation sites were found in intron 2. This suggests that the putative intron 2 promoter may not be used in HeLa cells, or its use may occur only at a particular point in the cell cycle. Our failure to detect additional bands may also be due to the low sensitivity of primer extension analysis, as compared to the method used by Raimond et al., which involved two rounds of amplification (RT-PCR, followed by standard PCR) (44). Comparison of human LGALS3 to other LGALS gene structures. A comparison between the known gene structures for members of the galectin family is shown in Figure 4. The only striking difference between the mouse and the human LGALS3 gene is the sizes of intron 1 (~8 kb for the human and ~5 kb for the mouse) and intron 3 (~23 kb for the human and ~1 kb for the mouse). The galectin-3 amino-terminal domain possessing the RNP-like sequence and PGAY repeat motif is contained entirely within exon III. The carboxy-terrninal domain contains the consensus galectin (or S-type) carbohydrate recognition sequence. All of the amino acid residues directly involved in carbohydrate 71 Figpre 4. Comparison of the human LGALS3 gene to the structures of other mammalian galectin genes. Human LGALS3 gene structure compared to other mammalian galectins with known gene structures: mouse LGALS3 (35-37); mouse LGALS] (41); human LGALS] (42); and human LGALSZ (43). The exons are represented by shaded boxes, the introns by thin lines, and untranslated sequences by open boxes. The sizes of introns and polypeptide coding sequences are given in kb and bp, respectively. 72 .5 a: 5 e3 .5 8 558 .5 e a e. 3 «3.5.— .55... E n u .5 v: .5 a: .5 8 5:8 .5 a e. 3 e. 3 5.23 9.8... >— E n _ .5 3; .5 a: .5 8 ”58... .5 a .3 3 e. 3 e. 3 5.23 5.8.... E n _ 3.5.8 .5 a: .5 mm. .5 e3 .5 a. .518». .5 «a ngwd 095:. amt—{Ut— =35:— 73 binding, as determined from the crystal structure of bovine galectin-1 (45) and human galectin-2 (46), are contained in exon V. There is greater nucleotide and amino acid identity between galectin-3 and galectin-1 than between galectin-3 and galectin-2 (data not shown). Therefore, based on the gene structures and the sequence similarities, it is likely that the LGALS3 gene evolved following the duplication of the LGALS] gene and insertion of the exon III/RNP-like sequence by exon shuffling. Comparison of the genomic muence flanking the human and mouse galectin-3 transcription initiation sites. For the mouse galectin-3 gene, alternative transcription initiation sites are used which lead to two distinct differentially-expressed mRNAs (37): if transcription initiation occurs at the upstream or start site, a more proximal splice donor site in exon I is used, and this mRNA (designated as the Type 11 message) is constitutively expressed; however, if transcription initiation occurs at the downstream B, 8, 7 start site region, a more distal splice donor site in exon I is used, resulting in the retention of a 27 bp genomic sequence in the 5’-UTR (designated the Type I message) (Figure 5). The Type 1 message is expressed as a function of the cell cycle. The +1a/+1b transcription start site in the human gene appears to correspond to a position similar to the B, 8, 7 start site region in the mouse galectin-3 gene, given the remarkable sequence identity between the human and mouse exon I sequences beginning at the +1al+1b start site (B, 8, y for the mouse) and extending ~100 nt upstream (Figure 5). However, the sequence between the +1a/+1b start sites and the exon I-intron 1 splice site in the human gene contains no sequence similar to the 27 bp sequence in the mouse galectin-3 exon 1. Based on the RNase protection and primer extension analysis of HeLa cell mRNA, there is no evidence of a start site for exon 1 in the human gene corresponding to the upstream 74 Figpre 5. Comparison of the human LGALS3 and mouse LGALS3 genomic sequences flanking the transcription initiation sites. The alignment was done using the Bestfit program in the GCG software package and by visual inspection. The thin vertical lines indicate nucleotide identity and (.) denotes an inserted gap to increase the alignment between the sequences. The shaded region indicates the 27 bp genomic sequence present in the 5’-UTR when the distal splice donor site is used in the mouse. The 5’ ends of the splice donor sites (5’-GTGAG) are indicated by the thick horizontal lines. The human LGALS3 transcription initiation sites are double-underlined and designated +1a and +1b. The mouse LGALS3 transcription inititation sites are designated 01., B, 5, y in keeping with the previously reported nomenclature (37). H, human LGALS3 sequence; M, mouse LGALS3 sequence. I 75 CCCAGCCTGGTCCGGGGAGAGGACTGGCTGGGC.AGGGGCGCCGCCCC II 11111 I || I II II |||| |||| II lllll TGGGGCGCGGTCCAGCCAGGCGCCT.GCAGGGCGAGGGATGCTGCCCC GCC.TCGGGAGAGGCGGGCCGGGCGGGGCTGGGAGTATTTGAGGC.TC III I 11111 llllllll lllll Illll I II II I GCCCT.GGGAG.GGCGGGCC...CGGGGAAAAGAGTACTAGAAGCGGC a (+1a) (+lb) GGAGCCACCGCCCCGCCQGQGCCCGCAGCACCTCCTCGCCAGCAGCCG llllllllllll II I III I II I III CGAGCCACCGCCCAGCTCTGACAGCTAGCGGAGCGGCGGGTGGAGCAC B r 5 intron 1 splice donor (H) T ............. CCGGAGCCAGCCAACGAGCGQIQAGCTGCGCGGG l I II III lllllll I lllll TAATCAGGTGAGCGGCACAGAGAGCACTACCCAGGTGAGCGGTGCGGG Wartime] splice donor (M) War splice donor (M) GCGCGGGGGACGCGGCTCCGGCCG Illlllll II || | l I GCGCGGGGACCGGTACTGC.TCAG 76 (or) start site of the mouse gene. Similarly, a sequence corresponding to the proximal splice donor site is not present in the human exon 1. Functional characterization of the promoter reg'on. In order to determine if the 5’ flanking region (relative to the transcription initiation site) has promoter activity, reporter gene constructs were prepared. A genomic fragment encompassing exon 1 was subcloned and nested deletion mutants prepared. These deletion mutants were directionally subcloned into a luciferase reporter vector (Figure 6). Each construct was then transiently transfected into HeLa cells. To correct for transfection efficiency, a B-gal reporter vector driven by the SV40 promoter/enhancer was co-transfected with each galectin-3 promoter- reporter construct. Differences in luciferase expression (and, hence, differences in promoter activity) are given relative to the luciferase reporter plasmid alone containing no human galectin-3 genomic sequence. The genomic region encompassing -836 to +14] nt has significant promoter activity as indicated by the ~250 fold increase in luciferase activity relative to the reporter vector alone (Figure 6A). Promoter activity remains nearly constant until the removal of a 110 nt fragment between -339 and -229 which results in a ~50% decrease in activity, suggesting that a transcriptional activator may exist in this region. Another significant decrease is detected as the region from -105 to -15 is deleted. A second transcriptional activator sequence may also be located in this region, or the decrease may be due to the loss of the DNA binding sites for the basal transcription complex. The -15/+141 region exhibits a ~2-13 fold increase in promoter activity versus vector alone. We have also determined the activities of the promoter-reporter constructs transfected into human diploid fibroblasts that can be made quiescent. LG] cells were 77 F'gpre 6. Promoter activities of human galectin-3-luciferase constructs in Hela cells (A) and quiescent and serum-stimulated LGl fibroblasts (B). Human galectin-3 promoter-reporter constructs are schematically shown. The transcription initiation sites (+la and +1b) are designated by the arrow for clarity, and the sizes of each construct are defined relative to +1a. Each construct was co-transfected into HeLa cells and LG] fibroblasts with pSV40-B-gal, a plasmid expressing B-galactosidase (B-gal) driven by the SV40 promoter/enhancer. Differences in transfection efficiencies are corrected by dividing the luciferase activity by the B-galactosidase activity. Data are expressed relative to the luciferase reporter plasmid containing no human LGALS3 sequence. The results shown represent the means (i standard deviation) of triplicate samples from one experiment. Another experiment gave identical results. 78 A. +141 uc +141 6:94 uc _513 +141 = uc uc +141 1111': .299 uc 5 +141 40:: DE] +141 "SP—4E1 0 5'0 1001s02002'50300 Relative luciferase activity B . Quiescent Serum-stimulated +141 “C _513 +141 -339 +141 3 no +141 1' Inc I 45 +141 'l'l'l‘l -2g9 0 25 0 is 5'0 7'5 100 Relative luciferase activity 79 cultured for 72 h in MEM containing 0.2% fetal calf serum, 0.] mM Ca2+. We determined the cell number as a function of time during this period. In addition, FACS analysis was carried out on parallel cultures after staining with propidium iodide. Both methods indicated that the LG] cells were made quiescent by simultaneous lowering the concentrations of serum and calcium ions (data not shown). Finally, a low level of DNA synthesis was observed in these cells, as assayed by the incorporation of [3H]thymidine (Fig. 7). Upon readdition of serum and Ca2+, the level of [3H]thymidine incorporation increased dramatically between 12 and 18 h, representing the first S-phase of the synchronized cell p0pulation. Galectin-3 expression, revealed by immunoblotting of the polypeptide, decreased as the cells were made quiescent. The reinitiation of the cell cycle resulted in the increased expression of galectin-3, beginning as early as 1 h and peaking at about 12 h following medium change (Fig. 7). Galectin-3 is an IEG; previous Northern blot and nuclear run-off analyses have documented that the increased expression of the galectin-3 polypeptide is due, at least in part, to an increased transcription of the gene ( 16). Thus, by comparing serum-starved cells versus cells reactivated by the addition of serum, the region of the promoter responsive to serum induction could be analyzed. In quiescent LG] cells, promoter activity is low, ranging between ~2-12 fold over vector alone (Figure 6B). When these cells re-enter the cell cycle following serum addition, there is a ~100 fold increase in promoter activity for the -836 to +14] and the - 513 to +141 nt constructs. When the region —513 to -339 is removed, serum activation of the promoter decreased ~50%. There are two other significant drops in promoter activity: (a) when the region -339 to ~229 is removed; and (b) when the region-105to-15 is 80 Figpre 7. Galectin-3 protein levels in LG] cells as a function of serum and calcium withdrawal and after readdition of complete growth medium. The cells were made quiescent by incubation in starvation medium (MEM, 0.2% fetal calf serum, 0.] mM Ca2+). These cells were then stimulated to synchronously reenter the cell cycle by adding complete growth medium (MEM, 10 fetal calf serum, 1.8 mM Ca2+). Extracts from an equal number of cells were prepared at various times and subjected to SDS-PAGE and immunoblotting with antibodies directed against galectin-3 and actin. The binding of the primary antibodies were revealed with horseradish peroxidase-conjugated secondary antibodies and detected using a chemiluminescence system. DNA synthesis was measured by [3H]thymidine incorporation. GO (7'01 x urdo) amid“ GUIPEWKID-HE 81 w . A.,...W;A._.v ._ 1.... meson—aw 3+ w~+ 2+ w+ m+ ~+ Nb- we- 9N- ~— 25 25a. A .8523 85.—om A3355; E38 ... . .t . u it . in . r {I 82 deleted. The importance of these latter two regions to the promoter activity parallels that observed for HeLa cells (Figure 6A and Figure 6B). Sguence of the promoter region. The sequence of the human galectin-3 promoter from -836 nt to +141 relative to the transcription initiation site is shown in Figure 8. From - 314 to +1a. the sequence is GC rich (74%) and contains many sequence motifs which are potential binding sites for transcription factors. The sequence lacks TATA and CCAAT boxes immediately proximal to the transcription start site, although both are present between ~500-8OO nt further upstream. Again, however, there is no evidence of a transcription start site located in this region. Overall, the 5’- flanking region contains five putative Spl binding sites (GC boxes), common to TATA-less promoters (47), two NF- KB-like (core consensus = GGGGAC'IT CC) sites, and multiple CAMP-dependent response element (CRE) motifs (core consensus = TGACGTCA) (for core consensus sequences, see reference (48)). In this regard, the human LGALS3 promoter sequence is similar to the murine LGALS3 promoter. Surprisingly, the -513/-339 promoter region contains no exact matches with binding sites of transcriptional activators with known activity in fibroblasts or upon serum induction. The -339/-229 promoter region contains one GC box (GC Box 2), one AP-4-like site (core consensus CAGCTGTGG), four AP-l- like sites (core consensus TGANTCA), one sis-inducible element (SIB) ['I'I‘CC(C/'I')GT(C/A)A], and a consensus basic helix-loop-helix (bHLH) core sequence (CANNT G). Within the AP-l/SIE box, the SIE flanks one AP-l site and overlaps another; within the AP-l/bI-ILH box, the bHLH core sequence is flanked by and overlaps two AP-l-like sites. Additional individual AP-l and CRE sites may be present since their cognate transcription factors cross-react with each other’s binding motifs (49). 83 Figure 8. Human LGALS3 promoter sequence and identification of putative transcription factor binding sites. The genomic sequence extending -848 nt upstream and +141 downstream relative to the transcription initiation site +la is shown. The 5’ end of each reporter gene construct is indicated with the distance from +la given above the nucleotide. The DNA sequences for putative transcription factor binding sites are indicated by the shaded boxes. The restriction endonuclease sites for NotI and Ber are shown by the dotted underline. In the AP-l/SIE box, AP-l sequences are indicated by a single underline and the SIE sequence is double underlined; similarly, in the AP-l- like/bHLH box, AP-l sequences are indicated by the single underline and the bHLH core sequence is double underlined. 84 P§15) beads (Fig. 4A, bottom) are shown. LG] cells at P26 were induced to ingest non-fluorescent beads. These beads appear as black dots (highlighted by a white arrow) when viewed in the fluorescein channel, against a green fluorescent background representing galectin-3 in the cytoplasm (Fig. 4B). The nucleus, whose boundary is defined by P1 labeling in the red channel, is devoid of galectin-3 in these senescent cells. A cell with a few (~8) beads (Fig. 4B, top) and a cell with many (>25) beads (Fig. 4B, bottom) are shown. 119 Figpre 4. Tagging LG] cells with beads. (A) P17 LG] cells tagged with green fluorescent beads (average diameter ~l.75 um). (B) P26 LG] cells tagged with non- fluorescent beads (average diameter ~2 pm). The cells were seeded at a density of ~5 x 103 cells/cm2 and beads were added at a ratio of ~500 beads/cell. After overnight culture, the cells were washed to remove the non-ingested beads. The cells were fixed and stained simultaneously with rat anti-galectin-3 plus FIT C-goat anti-rat immunoglobulin and PI. The nucleus is defined by the PI staining in the "red" channel. In panel A, both the green fluorescent beads (highlighted by white arrowheads) and the fluorescein yielded fluorescence in the “green” channel. The non-fluorescent beads appear as black dots and are highlighted by white arrows in panel B. Bar, 20 um. 120 121 The non-fluorescent (black) beads used to tag P26 cells have an average diameter of ~2 pm; the fluorescent (green) beads used to tag P17 cells have an average diameter of ~1.75 um. However, in the fluorescence micrographs shown in Figure 4 (and in Figure 5 as well), the green beads appear larger than the black beads. The is due to the halo effect or fluorescent flare of the green beads. In any case, we had surveyed a number of other beads, varying in size and in fluorescent properties, and found that ingestion of beads of approximately the same size represented the best way to tag the young and senescent LG] cells. Analysis of Galectin-3 Localization in Heterokagyons The various panels of Figure 5 provide examples of the galectin-3 localization patterns and the scoring system that will be used to analyze the results detailed in Table II. Panel A shows: (a) a homodikaryon derived from fusion of P18 cells because only green fluorescent beads are observed (highlighted by white arrowhead); (b) the positions of the two nuclei are defined by P1 staining; and (c) both nuclei contain galectin-3. Similarly, panel B shows: (a) a homodikaryon derived from fusion of P27 cells because only black beads are observed (highlighted by white arrow); (b) the positions of the two nuclei are defined by PI staining; and (c) both nuclei are devoid of galectin-3. Panels C - B represent heterodikaryons formed by fusion of P18 and P27 LG] cells; both green and black beads are observed, highlighted by the arrowheads and arrows, respectively. We score a heterodikaryon only if it contains at least seven of each kind of beads. Both nuclei of the heterodikaryon shown in panel C contain galectin-3. In panel D, one of the two nuclei of the heterodikaryon contains galectin-3 while the other nucleus is devoid of 122 F'gpre 5. Analysis of homo- and heterokaryons for galectin-3 localization. Galectin-3 (Gal-3) was detected with rat anti-galectin-3 plus FITC-goat anti-rat immunoglobulin. The nucleus is defined by P1 staining. Homo- versus heterokaryons are distinguished by the kind(s) of beads present in the double nucleated cell. Green fluorescent beads are highlighted by the white arrowheads; non-fluorescent black beads are highlighted by the white arrows. (A) Homodikaryon derived from fusion of P18 LG] cells, with galectin-3 in both nuclei; (B) Homodikaryon derived from fusion of P27 cells, with galectin-3 in neither nuclei; (C - E) Heterodikaryons derived from fusion of P18 and P27 cells. In panel C, both nuclei contain galectin-3; in panel D, one of the two nuclei is positive for galectin-3; and in panel B, neither of the two nuclei contain galectin-3. Bar, 20 pm. 124 .Emm. $8 .45.. $2 $5... $$ .89.. $ m. .95... $8 33. $... .3... $9 39.... $..... 6&8. $2. ES... $3 629$... 32:3... 823.. $2. .85... $8 .35.. $8. -z~ .2. +2.. -zm +2. «cor-.628... :23. .23 new 20.2. «:3 5.. 2.33:3 new 95: 2. 0.5.8.8 .0 5.35.30. 30.2.2 .83. $.. .25.. $8. .3... $.. .38. $6 .03. $N. at... $4. .85. $2 .5... $8 .82.... $.. .83.... $8. .32.... $3 seam. $8 -2. +2». +2. gorguofio: .53. $6 .ENN. $8 63...... $9 .95... $5... .82.... $8 .33. $9. .5... $.. .58. $8 .528. $5 .55.... $8 .8 .3. $8 .8 .3... $9. .33. $2 .53 $3 .83.... $8 .33. $8 .85....5... .$.Bm.$mm ..éu... $t. ..éum. $8 .828. $3 823. .. $8 6%.... $8 89$. $2 :2 +2 «coinage: 0mm+0Nm +5.3. Own. + 0N1 own. 8.. + a... tn 0 £2...on own + an + a... Own. + hNn. Rn. own. + a... a... a 2.2.225 own... 5. + a... 0mm + hm... bun. 0m... + an. m... <33...on .= 0.90... 125 the protein. Finally, panel B shows a heterodikaryon in which neither of the nuclei contain galectin-3. The results of three fusion experiments, which yielded essentially the same results, are presented in Table 11. Even within a single experiment, there are many numbers to consider. In the interest of brevity (and clarity) we state the major conclusions by analyzing the data in Experiment A of Table H and we provide an analysis of the theoretical and actual results in Figure 6. We monitored the nuclear versus cytoplasmic localization of galectin-3 in: (a) P19 cells not exposed to any fusogen (PBG), in which the overwhelming majority of the cells are monokaryons; (b) P19 cells treated with PEG, in which only homokaryons can be formed and observed (along with unfused monokaryons); (c) P27 cells not exposed to PEG; ((1) P27 treated with PEG; and (e) P19 and P27 cells in co-culture treated with PEG, in which we scored only heterodikaryons (although monokaryons and homokaryons are also present). Higher order karyons (tri-, tetra-, etc.) comprise only a very small fraction of the total number of fused cells and are excluded from the analysis. The following general statements can be derived from analysis of the data in Table II: (a) the majority of LG] cells at P19 exhibited nuclear galectin-3 (N+), while the majority of P27 cells showed no galectin-3 in the nucleus (N'); (b) this difference is retained after exposure to PEG, comparing the unfused monokaryons in the fusogen- treated culture. The percentages of cells showing N+ or N' in these monokaryons (under column labeled Monokagons in rows labeled P19 + PEG and P27 + PEG) form the basis for the calculation of theoretical values presented in Figure 6; (c) the majority of 126 Figpre 6. Comparison of the theoretical and actual phenotypes of galectin-3 localization. The theoretical values were calculated from the percent of homokaryons exhibiting nuclear localization of galectin-3 when P19 or P27 cells were treated with PEG (see Table II). The actual values were determined from heterodikaryons derived from PEG-mediated fusion of P19 and P27 cells. * Chi square analysis showed difference from the theoretical value was significant (p< 0.001). ** Not significant (p<0.025) by chi square analysis. 127 nausea 9%: axe: gene agent sewu 33$. 1.9.3.523. 128 homodikaryons derived from fusion of P19 cells show N+ in both nuclei (under column labeled Homodikaryons in row labeled P19 +PEG). (d) On the other hand, the majority of homodikaryons derived from fusion of P27 cells show N- in both nuclei (under column labeled Homodikaryons in row labeled P27 + PEG). and (e) Most strikingly, heterodikaryons derived from fusion carried out with P19 and P27 in co-culture showed predominantly the N+ phenotype in both nuclei (under column labeled Heterodikgons in row labeled P19 + P27 + PEG). Because both N+ and N' cells exist in P19 as well as in P27 cultures, it is useful to consider the probability of the phenotype(s) derived from all possible combinations of fusion to give rise to a heterodikaryon: (a) a P19 N+ cell fused with a P27 N+ cell; (b) a P19 N- cell fused with a P27 N+ cell; (c) a P19 N+ cell fused with a P27 N' cell; and (d) a P19 N- cell fused with a P27 N' cell. Following the rules of conditional probability, we can calculate a theoretical value for the percentage of cells showing a particular phenotype in the double nucleated cell. Thus, a heterodikaryon showing N+ in both nuclei is predicted to represent about 28% (.86 x .33) of the population (Fig. 6). Approximately 63% of the population is predicted to show the 1 N+ and 1 N- phenotype ([.86 x.67] + [.14 x .33]). Finally, ~9% (.14 x .67) of the heterodikaryons should show N- in both nuclei. The experimental results yielded far more heterodikaryons exhibiting N+ in both nuclei, far fewer of the 1 N+ and l N' phenotype, and the expected percentage of cells with N' in both nuclei (Table II, Experiment A and Fig. 6). The difference 129 between the theoretical and actual results points to an enhancement of nuclear localization of the galectin-3 polypeptide in the heterokaryons. On this basis, we conclude that the components of the P19 cells can somehow alter the regulation of nucleocytoplasmic distribution of the galectin-3 in the fused cells. Analysis of Nuclear and Cfloplasmic Markers and of DNA Synthesis in Hetero ons We have carried out a similar analysis on markers of the nucleus and cytoplasm. The Sm antigen is an epitope recognized by human autoimmune serum and found on the core polypeptides of the snRNPs. Immunofluorescence staining for Sm in LGl cells always localized the epitope to the nucleus, irrespective of CPD (P17 or P26), with or without PEG treatment, and in heterodikaryons (Fig. 7). Likewise, staining for LDH always localized the enzyme to the cytoplasm. Therefore, the exclusion of galectin-3 from the nucleus of senescent LGl cells and its altered nucleocytoplasmic distribution in heterodikaryons derived from P19 and P27 fusions does not reflect a general loss of nuclear:cytoplasmic segregation. Because our experimental results suggested that senescent cells might be deficient in factor(s) required for nuclear import and that the cytoplasm of young cells can rectify this apparent deficiency in heterokaryons, we tested two candidate factors, Ran and karyopherin-[3. Ran is a small GTP-binding protein (Mr ~25,000) involved in nucleo- cytoplasmic transport, and karyopherin-B is one subunit (Mr ~97,000) of the receptor for nuclear localization signal (NLS) such as that identified on the SV40 large T-antigen (see, for example, the review by Izaurralde and Adam (23)). Using immunoblotting, we found little or no difference in the levels of Ran and karyopherin-B expressed in P18 and P27 130 Figgge 7. Analysis of the localization of nuclear and cytoplasmic markers. The localization of Sm was determined with human autoimmune serum reactive against snRNPs and the localization of lactate dehydrogenase (LDH) was determined using antiserum raised against LDH. The experimental conditions for PEG treatment of P17 and P26 cells are the same as that described in legend to Figure 5. --- ID.— -.-- Em Own. + Own— + 0mm + ©Nn. + 5n. ©Nn_ omn— tfi. tn. 132 LG] cells (Fig. 8). It should be noted, however, that the amount of galectin-3 is higher in the larger P27 cells than in the smaller P18 cells (Fig. 8; see also Fig. 1A and 1B). On this basis, the ratio of karyopherin-B to galectin-3 would be lower in P27 cells than in P18 cells. Similar arguments apply to the Ran to galectin-3 ratio. Finally, we have also monitored DNA synthesis in heterokaryons derived from fusion of P19 and P27 LGl cells. Our BrdU labeling experiments show that, in the majority of the dikaryons analyzed, neither of the nuclei were labeled (Table I). This is in agreement with previous reports (21, 22, 24), documenting that the senescent cell nucleus is dominantly suppressive when fused with a cell capable of undergoing DNA replication. 133 Figure 8. Comparison of the levels of expression of karyopherin-[3 and Ran in P19 and P27 cells. Cells from non-synchronized cultures were harvested and extracts from an equal number of cells were prepared for SDS-PAGE as described in legend to Figure 1. Mouse monoclonal antibodies against karyopherin-B (Kar—B) and Ran were used in conjunction with horseradish peroxidase-conjugated goat anti-mouse immunoglobulin and were detected using a chemiluminescence system. Procedures for immunoblotting of galectin-3 (Gal-3) and actin were as described in legend to Figure l. ar- Ran Gal-3 Actin .. : 3: r, . . ‘ f. I Y ' 51'; , - ' \ - . , . , - v in. ‘ ‘ ‘ " ‘ . ~ .. '.'. '1' ‘twrfifiuqu- . . n», if, - ‘ . - , ‘ ,. , . , 3’: )n M l “ ._. «. r I" ‘l ‘1 .§ Us: ..,~ .., '~ a’;~-. 135 DISCUS ION The key findings of the present report include: (a) In the [.61 strain of HDF, galectin-3 expression is diminished during quiescence brought about by serum starvation and is induced upon readdition of serum. This pattern is observed independent of the replicative capacity of the cells. (b) Although expressed in senescent LG] cells, the protein appears to be excluded from the nuclei of these cells. This is in contrast to young LGl cells, in which 80—90% of the cells showed galectin-3 in the nucleus (and in the cytoplasm). (c) In heterodikaryons derived from fusion of young and senescent LG] cells, the predominant phenotype was galectin-3 in both nuclei. ((1) The results of these analyses, together with the control experiments, suggest that the components of young LG] cells can alter the nucleocytoplasmic distribution of galectin-3 by supplying some factor(s) required for the nuclear localization of the protein. Comparison of the results of the present study with a previous study performed on another strain of HDF (SL66) highlights the interesting fact that nuclei of senescent HDF, no longer capable of undergoing DNA synthesis, do not contain galectin-3. This is achieved by one of two ways. In the case of SL66, the expression of galectin-3 is simply downregulated, as assayed at the level of the mRNA by Northern blotting or at the level of the protein by Western blotting (13). In the case of LG], on the other hand, the regulation appears to be much more subtle (and perhaps more exquisite): while the polypeptide of galectin-3 continues to be expressed in LG] cells no longer undergoing DNA synthesis, the protein appears to be excluded from the nuclei of these cells, 136 resulting in a predominantly cytoplasmic localization. At least three examples can be cited to draw analogies to this phenomenon of nuclear exclusion. First, patients with Werner syndrome prematurely develop a variety of the major age-related diseases, including atherosclerosis, osteoporosis, type II diabetes, and malignant neoplasms (6). Early graying and hair loss, skin atrophy, and aged appearance are also manifested by these individuals. Fibroblasts from these patients have shorter replicative life-spans, which are similar to the life-spans of corresponding cells taken from elderly individuals. The WRN gene has been cloned; the amino acid sequence of the encoded polypeptide shows significant similarity to DNA helicases (25). Indeed, ATP- dependent 3’—)5’ helicase activity has been demonstrated in vitro (26, 27). The WRN gene product is normally found in the nucleolus (28, 29) or nucleoplasm (30) but there is impaired nuclear localization of defective DNA helicases in Werner syndrome cells (31). Second, the wild-type p53 protein can act to negatively regulate cell proliferation and function as a suppressor of transformation and tumorigenesis. Mutations at the p53 locus have been thought of as the most common mechanism to inactivate the negative regulatory effects of p53 on cell proliferation. However, breast cancers contain p53 mutations in only about 30% of the cases. Moreover, many breast cancers show a reduction to homozygosity at the p53 locus, but more than 60% of such breast cancers retain the wild-type p53 allele. On this basis, it was reasoned that additional mechanisms exist, in addition to mutations, which affect the activity of the p53 protein and its negative regulation of cell growth (32). Indeed, it appears that some breast cancers that contain the wild-type form of p53 protein may inactivate its tumor-suppressing activity by sequestering this protein in the cytoplasm, away from its site of action in the cell nucleus. 137 More strikingly, the same study also detected cytoplasmic p53 in normal lactating breast tissue, suggesting that estrogen-mediated cell proliferation in this particular physiological situation may use nuclear exclusion to inactivate p53. Finally, the BReast CAncer 1 (BRCA-l) gene is expressed in many cell types, including normal breast epithelial cells, breast cancer cell lines, as well as tumor lines derived from bladder, cervical and colon cancers. In normal cells and in tumor cells derived from tissues other than breast or ovary, BRCA—l is localized in the nucleus. However, it is has been reported that BRCA-l is found mainly in the cytoplasm of all breast cancer cell lines, in primary cells from malignant pleural effusions and biopsy sections from patients with breast cancer (33). Moreover, sporadic (not familial or heritable) cases, which account for ~95% of the breast cancers, do not harbor BRCA-l mutations. Thus, BRCA-l may be inactivated by intragenic mutations in hereditary breast cancer. In most non-hereditary breast cancers, however, BRCA-l may be inactivated indirectly, by mislocation in the cytoplasm (33). The mechanism by which galectin-3 is excluded from the nucleus of senescent LG] cells is not known. However, several key pieces of information seem particularly important. The first is that the galectin-3 polypeptide, derived from P27 LG] cells, is not intrinsically impaired in terms of nuclear import. This was demonstrated by monitoring selectively the human galectin-3 polypeptide in heterodikaryons derived from fusion between mouse 3T3 and human LG] (at P27) cells (K.P. Openo, unpublished observations). NCL—GAL3 is a mouse monoclonal antibody that: (a) immunoblots galectin-3 from extracts of human but not mouse cells; and (b) stains human HeLa cells and LG] fibroblasts but not mouse 3T3 fibroblasts under immunofluorescence. Thus, it 138 appears that the NCL—GAL3 antibody specifically recognizes human galectin-3 but not its murine counterpart. Using this antibody to stain heterodikaryons derived from fusion between mouse 3T3 cells and human LG] cells (at P27), we found human galectin-3 in both nuclei. This indicates that the galectin-3 polypeptide in senescent human fibroblasts can be imported into the nucleus. The second fact is our present observation that heterokaryons derived from fusions between young and senescent LG] cells show a phenotype in which both nuclei contain galectin-3. This strongly suggests that the senescent LG] cells might lack factor(s) required for nuclear import, which must be supplied by the young cells. One possibility is that there might be a deficiency in Ran or karyopherin-B, inasmuch as the ratio of Ran (and karyopherin-B) to galectin-3 is much lower in senescent cells than in young cells. In previous studies, Agrwal (34) had conjugated recombinant galectin-3 with rhodamine and the resulting fluorescent derivative was tested for nuclear localization after: (a) microinjection into intact mouse 3T3 fibroblasts (in vivo assay); or (b) incubation with 3T3 cells permeabilized by digitonin (in vitro assay). Microinjected galectin-3 accumulated in the nucleus of some recipient cells whereas fluorescently labeled human serum albumin (rh-HSA) failed to do so in any of the cells. Fluorescently labeled human serum albumin conjugated with a peptide containing the nuclear localization signal of SV40 large T antigen (rh-HSA-NLS) was translocated into the nucleus in 100% of the microinjected cells. Thus, while galectin-3 introduced into 3T3 cells can be imported into the nucleus, its nuclear localization properties were different from those documented for HSA-NLS. Galectin-3 and HSA-NLS were also different in 139 terms of requirements and conditions of nuclear import in the in vitro assay. While rh- HSA-NLS was imported into the nucleus in an ATP and cytosol dependent fashion, little or no nuclear transport of rh-galectin-3 could be demonstrated. Thus, it appears that the 8100 cytosolic fraction (containing the nuclear import factors such as karyopherins, Ran, and NTF2) used in the in vitro assay could not supply the factor(s) necessary for the import of galectin-3 while the cytosol of microinjected cells could indeed fulfill the requirements of nuclear transport. This suggests that the nuclear import of galectin-3 may require specialized components and/or mechanisms. In our analysis of heterodikaryons formed by fusion of young and senescent LG] cells, the dominant phenotype is galectin-3 in the nucleus (N+ in both nuclei), corresponding to the phenotype of the young cells. This observation is in contrast to the dominance of the senescent phenotype in heterokaryons between replicative and post- replicative HDFs when assayed for nuclear DNA synthesis. Entry into S phase is blocked in nuclei in heterokaryons containing at least one senescent cell (2], 22, 24). Using P19 and P27 LG] cells, we have performed parallel analyses for galectin-3 localization and for BrdU incorporation in heterodikaryons derived from a single fusion experiment. We have ascertained that, indeed, there is a divergence of dominant phenotypes: (a) the young cell phenotype is dominant with respect to galectin-3 nuclear localization; and (b) the senescent cell phenotype is dominant with respect to DNA Synthesis and cell cycle progression. This phenotype divergence suggests that galectin-3 nuclear localization is not sufficient for restoration of nuclear DNA synthesis in the senescent HDF. Thus, the significance of excluding galectin-3 from the nucleus of senescent cells remains to be explored, particularly in terms of its role in processing nuclear RNA 140 molecules. Although no direct evidence has been reported to link cellular senescence and problems in pre-mRNA splicing, it has been shown that senescent WI-38 HDFs have a post-transcriptional block in the expression of the Proliferating Cell Nuclear Antigen (PCNA) gene (35) that could result from a splicing defect. In addition, alternative splicing variants of fibronectin are specifically expressed in senescent IMR-90 HDFs (36). Given our documentation on the loss of galectin-3 from nuclei of senescent SL66 (13) and LG] cells, this protein certainly becomes an intriguing candidate for altering the pre-mRNA processing events in senescent cells. W This work was supported by grants from the National Institutes of Heath (GM-38740), the National Science Foundation (MCB 97-23615), and the Elsa U. Pardee Foundation. M.M.K. was supported by fellowships from the Medical Scientist Training Program at Michigan State University and from the Glenn Foundation/American Federation for Aging Research. Special thanks to Mr. Kyle Openo for his outstanding technical skills in performing the cell fusion and immunoblotting experiments which resulted in production of Figures 4-8. We thank Drs. Justin J. McCormick and Veronica M. Maher for gifts of the human diploid fibroblast LG]. Chapter III has been written in the format for the Experimental Cell Research and will soon be submitted for publication. 10. ll. 12. 13. 141 REFERENCES Hayflick, L. (1965). Exp. Cell Res. 37, 614-636. Goldstein, S. (1990). Science 249, 1129-1133. Martin, G.M., Sprague, C.A., and Epstein, CJ. (1970). Lab. Invest. 23, 86-92. Schneider, EL. and Mitsui, Y. (1976). Proc. Natl. Acad. Sci. USA 73, 3584- 3588. Cristofalo, V.J., Allen, R.G., Pignolo, R.J., Martin, B.G., and Beck, J .C. (1998). Proc. Natl. Acad. Sci. USA 95, 10614-10619. Salk, D. (1982). Hum. Genet. 62, 1-15. Rhome, D. (1981). Proc. Natl. Acad. Sci. USA 78, 5009-5013. Gutsmann-Conrad, A., Heydari, A.R., You, 8., and Richardson, A. (1998). Exp. Cell Res. 241, 404-413. Dimri, G.P., Lee, X., Basile, G., Acosta, M., Scott, G., Roskelley, C., Medrano, E..,E Linskens, M., Rubeli, 1., Pereira-Smith, O., Peacocke, M., and Campisi, J. (1995). Proc. Natl. Acad. Sci. USA 92, 9363-9367. Sinclair, D.A., Mills, K., and Guarente, L. (1997). Science 277, 1313-1316. Patterson, R.J., Dagher, S.F., Vyakarnam, A., and Wang, J .L. (1997). Trends Glycosci. Glycotechnol. 9, 77-85. Moutsatsos, I.K., Wade, M., Schindler, M., and Wang, J.L. (1987). Proc. Natl. Acad. Sci. USA 84, 6452-6456. Hamann, K.K, Cowles, E.A., Wang, J .L., and Anderson, R.L. (1991). Exp. Cell Res. 196, 82-91. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 142 Morgan, T.L., Yang, D. Fry, D.G., Hurlin, P.J., Kohler, S.K., Maher, V.M. and McCormick, J.J. (1991). Exp. Cell Res. 197, 125-136. Steck, P.A., Voss, P.G., and Wang, J .L. (1979). J. Cell Biol. 83, 562-575. Vindelov, L.L., Christensen, 1.1., and Nissen, NJ. (1983). Cytometry 3, 328-331. Ho, M.K., and Springer, T.A. (1982). J. Immunol. 128, 1221-1228. Agrwal, N., Sun, Q., Wang, S.-Y., and Wang, J.L. (1993). J. Biol. Chem. 268, 14931-14939. Metcalf, T.N., Szabo, L.J., Schubert, K.R., and Wang, J.L. (1980). Nature (London) 285, 171-172. Laemmli, UK. (1970). Nature (London) 227, 680-685. Stein, G.H. and Yanishevsky, RM. (1979). Exp. Cell Res. 120, 155-165. Pereira-Smith, OM. and Smith, JR. (1982). Somatic Cell Genet. 8, 731-74. Izaurralde, E. and Adam, S. (1998). RNA 4, 351-364. Norwood, T.H., Pendergrass, W.R., Sprague, C.A., and Martin, GM. (1974). Proc. Natl. Acad. Sci. USA 71, 2231-2235. Yu, C.E., Oshima, J., Fu, Y.H., Wijsman, E.M., Hisama, F., Alisch, R., Matthews, 8., Nakura, J., Miki, T., Ouais, S., Martin, G.M., Mulligan, J., and Schellenberg, GD. (1996). Science 272, 258-262. Suzuki, N., Shimamoto, A., Imamura, O., Kuromitsu, J ., Kitao, 8., Goto, M., and Furuichi, Y. (1997). Nucleic Acids Res. 25, 2973-2978. Gray, M.D., Shen, J.C., Kamath-Loeb, A.S., Blank, A., Sopher, B.L., Martin, G.M., Oshima, J., and Loeb, LA. (1997). Nat. Genet. 17, 100-103. 28. 29. 30. 31. 32. 33. 34. 35. 36. 143 Gray, M.D., Wang, L., Youssoufian, H., Martin, G.M., and Oshima, J. (1998). Exp. Cell Res. 242, 487-494. Marciniak, R.A., Lombard, D.B., Johnson, RB, and Guarente, L. (1998). Proc. Natl. Acad. Sci. USA 95, 6887-6892. Shiratori, M., Sakamoto, 8., Suzuki, N., Tokutake, Y., Kawabe, Y., Enomoto, T., Sugimoto, M., Goto, M., Matsumoto, T., and Furuichi, Y. (1999). J. Cell Biol. 144, 1-9. Matsumoto, T., Shimamoto, A., Goto, M., and Furuichi, Y. (1997). Nat. Genet. 16, 335-336. Moll, U.M., Riou, G., and Levine, A.J. (1992). Proc. Natl. Acad. Sci. USA 89, 7262-7266. Chen, Y., Chen, S.-F., Riley, D.J., Allred, D.C., Chen, P.-L., Von Hoff, D., Osborne, C.K., and Lee, W.-H. (1995). Science 270, 789-791. Agrwal, N. (1992). Doctoral dissertation, Michigan State University. Chang, C.-D., Phillips, P., Lipson, K.E., Cristofalo, V.J., and Baserga, R. (1991). J. Biol. Chem. 266, 8663-8666. Magnuson, V.L., Young, M., Schattenberg, D.G., Mancini, M.A., Chen, D., Steffensen, B., and Klebe, R.J. (1991). J. Biol. Chem. 266, 14654-14662. CHAPTER IV. Concluding Statements 144 145 CHAPTER IV. Concluding Statements Knowing the structure of the galectin-3 gene, along with data from this initial functional characterization of the promoter, can enable future investigators to further study the molecular mechanism(s) regulating transcriptional activity in growth-arrested versus proliferating cells. In particular it would be of interest to further examine how a housekeeping-like promoter might also be proliferation or serum inducible. The fact that expression of galectin-3 remained serum-inducible in senescent LG] human fibroblasts came as a surprise, especially given our previous findings of decreased galectin-3 expression after serum addition in another strain of human fibroblast, SL66. Nevertheless, both strains do exhibit nuclear exclusion of galectin-3 when senescent. The present study extended the previous findings, however, and we were able to detemiine galectin-3 subcellular localization and simultaneously distinguish whether that cell was proliferating or not by using double antibody staining and co-immunofluorescence. Also, by taking advantage of the scanning confocal microscopy, we unequivocally identified the exact subcellular localization of galectin-3 in human fibroblasts. The nuclear exclusion of galectin-3 in senescent human fibroblasts can be used as a probe to identify factors required for nuclear import. Generation of heterodikaryons can allow us to further study the mechanism of nuclear transport in general. It would be 146 interesting to test whether fusion of a senescent LG] cell with its immortalized counterpart MSU 1.] could also provide the missing transport factor(s) and produce galectin-3 nuclear localization, suggesting that in the process toward tumorigenesis, certain nuclear import factors are re-activated. The mechanism of altered galectin-3 nuclear localizatioin is unknown. Future studies could address whether the localization factor is a protein or a nucleic acid, and whether this factor(s) is regulated as a function of the cell cycle. When the galectin-3 nuclear localization signal is delineated and the cytosolic receptor(s) required for nuclear import identified, a more thorough study of which factor(s) are lost in senescent cells could be carried out. Furthermore, the cell fusion method along with the antibody that is able to distinguish mouse from human galectin-3 can be used to study nucleo- cytoplasmic shuttling and the factors involved.