Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Apr 15;111(15):5574-9.
doi: 10.1073/pnas.1315034111. Epub 2014 Mar 27.

Sumoylation Differentially Regulates Sp1 to Control Cell Differentiation

Affiliations
Free PMC article

Sumoylation Differentially Regulates Sp1 to Control Cell Differentiation

Lili Gong et al. Proc Natl Acad Sci U S A. .
Free PMC article

Abstract

The mammalian small ubiquitin-like modifiers (SUMOs) are actively involved in regulating differentiation of different cell types. However, the functional differences between SUMO isoforms and their mechanisms of action remain largely unknown. Using the ocular lens as a model system, we demonstrate that different SUMOs display distinct functions in regulating differentiation of epithelial cells into fiber cells. During lens differentiation, SUMO1 and SUMO2/3 displayed different expression, localization, and targets, suggesting differential functions. Indeed, overexpression of SUMO2/3, but not SUMO1, inhibited basic (b) FGF-induced cell differentiation. In contrast, knockdown of SUMO1, but not SUMO2/3, also inhibited bFGF action. Mechanistically, specificity protein 1 (Sp1), a major transcription factor that controls expression of lens-specific genes such as β-crystallins, was positively regulated by SUMO1 but negatively regulated by SUMO2. SUMO2 was found to inhibit Sp1 functions through several mechanisms: sumoylating it at K683 to attenuate DNA binding, and at K16 to increase its turnover. SUMO2 also interfered with the interaction between Sp1 and the coactivator, p300, and recruited a repressor, Sp3 to β-crystallin gene promoters, to negatively regulate their expression. Thus, stable SUMO1, but diminishing SUMO2/3, during lens development is necessary for normal lens differentiation. In support of this conclusion, SUMO1 and Sp1 formed complexes during early and later stages of lens development. In contrast, an interaction between SUMO2/3 and Sp1 was detected only during the initial lens vesicle stage. Together, our results establish distinct roles of different SUMO isoforms and demonstrate for the first time, to our knowledge, that Sp1 acts as a major transcription factor target for SUMO control of cell differentiation.

Keywords: crystallin gene expression; eye development; transcription regulation.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.
Fig. 1.
SUMO1 and SUMO2/3 are differentially expressed and conjugated during lens differentiation. (A) Cryosections of mouse eye at embryonic day (ED) 16.5 were processed for IHC and observed under a confocal microscope. Lens epithelial (LE) cells and lens fiber (LF) cells are indicated by arrow and arrow head, respectively. (a and c) Overlapping of SUMO1 or SUMO2 (green) image with nuclei (DAPI staining, blue). (b and d) IHC staining of SUMO1 or SUMO2. (Magnification: a–d, 100×.) (b1, b2, d1, and d2) Magnifications of boxed areas in b and d, respectively. (B) WB showing conjugation of SUMO1 (Left) and SUMO2/3 (Right) in mouse lenses of indicated ages. NB, newborn; 7D, 7 d; 2M, 2 mo. The newly sumoylated protein species in the adult lens are indicated by red arrow head. (C) WB showing SUMO1 (Left) and SUMO2 (Right) in adult mouse lenses and bFGF-induced αTN4-1 cells. Protein species with enhanced SUMO1 modification or decreased SUMO2 modification under bFGF treatment were indicated by red asterisk or blue asterisk, respectively. Those also modified by SUMO1 or SUMO2 in vivo were labeled with red or blue arrow head, respectively.
Fig. 2.
Fig. 2.
Contrasting effects of SUMO1 and SUMO2/3 on bFGF-induced fiber differentiation. (A) αTN4-1 cells stably expressing GFP, GFP-SUMO1, GFP-SUMO2, or GFP-SUMO3 were left untreated (al) or treated with bFGF for 15 d (a′–l′). Cell morphology was observed under a phase contrast microscope (PH) or under a fluorescence microscope to detect GFP or nuclei (DAPI). (Magnification: 50×.) (B) WB to show β-crystallin in four stable cell lines without (Mock; 15 d) or with (FGF; 15 d) bFGF induction. (C) Established mock (Mock), SUMO1 knockdown (SUMO1 sh), SUMO2 knockdown (SUMO2 sh), or SUMO2/3 knockdown (SUMO2/3 sh) αTN4-1 cells were treated with bFGF to induce fiber differentiation for 8 d. (Magnification: 50×.) (D). WB to show β-crystallin expression in Mock, SUMO1 sh, SUMO2 sh, and SUMO2/3 sh cells during bFGF-induced fiber differentiation. (E) MTT assay to evaluate the cell proliferation in different cells indicated. Experiments were done in triplicate and represent mean values ± SD.
Fig. 3.
Fig. 3.
SUMO1 and SUMO2 differentially regulate Sp1 stability and interactions with other factors. (A) SUMO2 overexpression decreased Sp1 stability. (a) WB to show Sp1 level in nucleus and cytoplasm of the indicated cell lines without or with bFGF treatment. (b) Myc-tagged Sp1 (Myc-Tag-Sp1) was cotransfected with equal amount of HA-tagged SUMO1 or -SUMO2 into HLE cells. (c) mRNA level of Sp1 was examined by qPCR at the indicated transfections. (d) Cycloheximide (100 μg/mL) treatment to assess the effect of SUMO1 or SUMO2 on the steady level of Sp1 wild type (WT) or K16R mutant (K16R) in HLE cells. The numbers below the β-Actin lanes are the relative levels of Sp1 expression under different conditions. (B). qChIP showing that SUMO2 overexpression enhances Sp3 but suppresses p300 binding into βB1, βB2, and βB3 promoters. (C) SUMO2 overexpression inhibits interactions between Sp1 and p300. (a) Myc-Sp1 was cotransfected with SUMO1 or SUOM2 and then precipitated by Myc-tagged magnetic bead conjugate. Interactions with p300 under indicated transfections were detected by WB using anti-p300 and anti-IgG antibodies (an equal amount of normal IgG was also added into input samples for loading comparison). (b) SUMO2 overexpression inhibits interactions between Sp1 and p300 during bFGF induction. GFP, SUMO1, and SUMO2-αTN4-1 cells were untreated (Mock) or treated with 100 ng/mL bFGF to induce fiber differentiation. A total of 5 mg of proteins were used in each IP. The resulting precipitates were separated by 6% SDS/PAGE gel and analyzed by WB.
Fig. 4.
Fig. 4.
SUMO1 and SUMO2/3 modulates Sp1 in mouse lens. (A, Upper) Colocalization (yellow) between Sp1 (red) and SUMO1 (green) was analyzed by IHC in ED11.5 mouse lens. (Lower) Colocalization (yellow) between Sp1 (red) and SUMO2/3 (green) was analyzed by IHC in ED11.5 mouse embryonic lens. The overlapping coefficiency for SUMO1 and Sp1 is 0.953 and is 0.921 for SUMO2/3 and Sp1 using the thresholds method (SI Methods). (Magnification: 50×.) (B) Co-IP to show modifications of Sp1 by SUMO1 or SUMO2/3. Total proteins were extracted from ED11.5 eye or NB mouse lens, immunoprecipitated by anti-Sp1 (a), anti-SUMO1, or anti-SUMO2/3 antibodies (b). The precipitated samples were immune-blotted with anti-SUMO1 (a, Upper) or anti-SUMO2/3 antibodies (a, Lower), or anti-Sp1 antibody (b). Note that both SUMO1 and SUMO2/3 were detected in proteins precipitated by anti-Sp1 antibody at ED11.5, but only SUMO1 detected at NB lens (a). Similarly, at ED11.5, SUMO-Sp1 signal was found in anti-SUMO1-and SUMO2/3-pelleted samples but not in control IgG (b, Upper). In NB lens, however, SUMO-Sp1 signal was found only in anti-SUMO1–pelleted sample but not in control IgG or anti-SUMO2/3–pelleted samples (b, Lower). (C) SUMO1 but not SUMO2/3 occupies βB1, βB2, and βB3 promoters as analyzed by qChIP. (D) Schematic diagram to show that SUMO1 and SUMO2 differentially regulate lens differentiation through Sp1. +, positive regulation; −, negative regulation.

Similar articles

See all similar articles

Cited by 19 articles

See all "Cited by" articles

Publication types

MeSH terms

Substances

Feedback