QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 9 September 2003
doi: 10.1242/jcs.00720

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hathaway, H. J. Articles by Shur, B. D. Search for Related Content PubMed PubMed Citation Articles by Hathaway, H. J. Articles by Shur, B. D.

Mutational analysis of the cytoplasmic domain of ß1,4-galactosyltransferase I: influence of phosphorylation on cell surface expression

Helen J. Hathaway^1,2,3,*, Susan C. Evans^3,*,, Daniel H. Dubois³, Cynthia I. Foote¹, Brooke H. Elder¹ and Barry D. Shur^1,3,

¹ Department of Cell Biology, Emory University School of Medicine, Atlanta, GA, USA
² Department of Cell Biology and Physiology, University of New Mexico School of Medicine, Albuquerque, NM, USA
³ Department of Biochemistry and Molecular Biology, University of Texas M.D. Anderson Cancer Center, Houston, TX, USA

View larger version (86K): [in a new window]   Fig. 1. Indirect immunofluorescence of live cells shows TL-CAT localized to the cell surface. NIH 3T3 cells stably transfected with CAT or CAT fusion genes under the control of the MT-1 promoter were induced with Zn2+. Equal expression of the transgene in the different clones was confirmed by CAT activity. TL-CAT is localized to the cell surface, whereas mock- and CAT-transfected cells show minimal staining with anti-CAT antibody. Cells transfected with TS-CAT also show background levels of surface fluorescence (not shown).

View larger version (23K): [in a new window]   Fig. 2. TL-CAT is localized to the cell surface in stably transfected 3T3 cells. (a) As a positive control for flow cytometric analysis of GalT I expression on the cell surface, nontransfected 3T3 cells were incubated with antibodies raised against recombinant GalT I catalytic domain (-GalT I) and FITC-conjugated secondary antibody, and subsequently analyzed by flow cytometry as described in Materials and Methods. Control assays were incubated with preimmune serum (PI) rather than anti-GalT I antiserum. (b) 3T3 cells were stably transfected with CAT fusion constructs under the control of the inducible MT-1 promoter. Stably transfected 3T3 cells, induced with Zn2+, were labeled with anti-CAT antibody and FITC-conjugated secondary antibody, and analyzed by flow cytometry. The peak positions represent fluorescence intensity on the cell surface. Mock-transfected, or cells transfected with CAT alone or with TS-CAT show minimal fluorescent signals in both the uninduced (–) and induced (+) samples. Two populations of cells transfected with TL-CAT are shown in the mixed sample after Zn2+ induction (M+), representing adherent and nonadherent populations. The adherent cells (A+) show little surface staining above control levels (–) with anti-CAT antibody when induced with Zn2+ and assayed separately. However, fluorescence intensity increases dramatically in the population of live, nonadherent TL-CAT-expressing cells (F+).

View larger version (30K): [in a new window]   Fig. 3. TL-CAT constructs are associated with the detergent-insoluble cytoskeleton. 3T3 cells were transiently transfected with CAT, TS-CAT or TL-CAT under the control of the constitutive PGK promoter, and assayed for CAT immunoreactivity in detergent-extracted cell lysates as described in Materials and Methods. CAT protein expression on the cell surface was approximately 7-fold higher in TL-CAT transfectants as compared with CAT alone transfectants, and CAT expression on the surface of TS-CAT transfectants remained near control levels. Eighty-one percent of the CAT immunoreactivity on the cell surface of TL-CAT transfectants was associated with the detergent-insoluble cytoskeleton. In contrast, detergent-insoluble immunoreactivity was similar between CAT and TS-CAT transfectants. The results represent the average (±s.e.m.) of 4 experiments, each performed in triplicate, and each assay normalized to the TL-CAT values.

View larger version (47K): [in a new window]   Fig. 4. A portion of GalT I partitions with detergent-resistant lipid rafts. (a) Nontransfected 3T3 fibroblasts were lysed in detergent and insoluble material isolated and applied at the bottom of Optiprep gradients. After equilibrium centrifugation, the lighter, caveolin-containing lipid rafts float to near the top of the gradient (fraction 2), whereas membrane and matrix proteins (GM130, p115) not associated with lipid rafts remain in the heavier density fractions near the bottom of the gradient (fraction 10,11). The long GalT I isoform is found in both the detergent-insoluble fractions and the caveolin-containing lighter lipid rafts. (b) When the entire detergent-treated cell lysate (i.e. detergent soluble and insoluble pools) was subjected to the gradient, a larger percentage of the total cellular GalT I remained at the bottom of the gradient, as assayed with antibodies that recognize both GalT I isoforms (i.e. anti-GalT I catalytic domain). This is consistent with the presence of an additional pool of GalT I that is detergent-soluble, and therefore not associated with lipid rafts.

View larger version (41K): [in a new window]   Fig. 5. The cytoplasmic and transmembrane domains of the long (TL-GFP) or short (TS-GFP) GalT I were fused to GFP and expressed in 3T3 cells. (a) Cultures with similar levels of transgene expression, as judged by GFP immunoblotting, were used for subsequent analysis. Representative examples are presented of cells transfected with either TL-GFP, TS-GLP, GFP or mock-transfected controls. The TS-GFP fusion protein migrates slightly slower than does GFP alone, because of the addition of the short isoform cytoplasmic and transmembrane domains, whereas the TL-GFP fusion protein migrates slightly larger than TS-GFP because of the additional residues in the long cytoplasmic domain. (b) TS-GFP and TL-GFP fusion proteins were detectable in the Golgi complex, as well as mutated TL-GFP fusions [S11D and F3G,F7G (SD and FF) shown]. GFP fluorescence was also detectable on filopodia and lamellipodia of a significant percentage of cells expressing TL-GFP and specific mutated TL-GFP constructs (TL, FF shown), but not on TS-GFP or other mutated TL-GFP constructs (TS, SD shown). Bar, 25 µm (c) TL-GFP-expressing cells demonstrated a dominant-negative phenotype by inhibiting cell spreading on laminin matrices (three cells shown). In contrast, cells expressing TS-GFP spread extensively on laminin matrices (four cells shown). Bar, 25 µm.

View larger version (17K): [in a new window]   Fig. 6. Mutation of putative phosphorylation sites in the long GalT I cytoplasmic domain affects the expression of the GFP reporter on the cell surface. GFP fluorescence on processes of cells cultured on fibronectin was quantified as described in Materials and Methods. As expected, TS-GFP gave significantly lower surface fluorescence than did TL-GFP (P=0.0001). TL-GFP and mutations that change S11 or T18 to alanine (S11A, T18A and S11A,T18A) show the highest levels of surface GFP fluorescence, which are not significantly different from one another. In contrast, the S11D and T18D mutants both resulted in significantly decreased cell surface GFP fluorescence compared with TL-GFP. Mutating F3 and F7 to glycine (F3G,F7G) resulted in normal levels of surface GFP fluorescence. Data represents the mean ±s.e.m. from 3 to 6 individual determinations. P values were determined by Student's t test, and are shown for each construct. Using the Bonferroni method, statistical significance is calculated as P=0.05 divided by the number of mutants (in this case 6), therefore P<0.008 is statistically significant (asterisks).

View larger version (23K): [in a new window]   Fig. 7. Mutation of putative phosphorylation sites in the long GalT I cytoplasmic domain affects its ability to act as a dominant-negative. Spreading of 3T3 cells transiently transfected with TS-GFP, TL-GFP or mutant constructs was measured in cells plated on laminin matrices as described in Materials and Methods. Spreading of TS-GFP-expressing cells was similar to mock and GFP transfectants, whereas cells expressing TL-GFP spread poorly, demonstrating the dominant-negative phenotype. Mutating S11 or T18 to aspartic acid (S11D or T18D) inhibited the dominant-negative phenotype, whereas substituting S11 and/or T18 with alanine (S11A or T18A and S11A,T18A) had no effect on the ability of the long cytoplasmic domain to produce a dominant-negative phenotype. Substituting F3 and F7 with glycine (F3G,F7G) also inhibited the dominant-negative phenotype, although this construct was expressed normally on the cell surface (Fig. 6). Data represents the mean ±s.e.m. from 3 individual determinations. P values were determined by Student's t test, and are shown for each construct. Using the Bonferroni method, statistical significance is calculated as P=0.05 divided by the number of mutants (in this case 6), therefore P<0.008 is statistically significant (asterisks).

View larger version (49K): [in a new window]   Fig. 8. Phosphorylated GalT I is not expressed on the cell surface. 3T3 fibroblasts were metabolically labeled in either [35S]methionine (a,c) or [32P]orthophosphate (b), lysed and subjected to anti-GalT I immunoprecipitation. Radiolabeled GalT I is readily detectable from both 35S- and 32P-labeled cells. Immunoprecipitation using either antibodies against the GalT I catalytic domain (-GalT I) or against a peptide unique to the long cytoplasmic domain (-long GalT I) produced similar results. Immunoprecipitation of Gal T I was inhibited by inclusion of the immunizing peptide (+ peptide). Surface-associated GalT I was determined by immunoprecipitating GalT I from cell surface fractions isolated by biotinylation of intact cells, application to streptavidin agarose, and elution of biotinylated material with ß-mercaptoethanol (BME released). Aliquots of the unbound and BME-released surface fraction are shown. Approximately 7% of the total cellular 35S-GalT I was present in the cell surface fractions (determined by quantitative enzyme assay), whereas none of the phosphorylated GalT I could be found on the cell surface. The arrowheads denote the relative migration of GalT I. Molecular weight markers are shown to the left of each panel.