Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Oct 1.
Published in final edited form as: J Cell Biochem. 2014 Oct;115(10):1840–1847. doi: 10.1002/jcb.24861

The Basic Helix-Loop-Helix/Leucine Zipper Transcription Factor USF2 Integrates Serum-Induced PAI-1 Expression and Keratinocyte Growth

Li Qi 1, Craig E Higgins 1, Stephen P Higgins 1, Brian K Law 2, Tessa M Simone 1, Paul J Higgins 1,3
PMCID: PMC4134751  NIHMSID: NIHMS603222  PMID: 24905330

Abstract

Plasminogen activator inhibitor type-1 (PAI-1), a major regulator of the plasmin-dependent pericellular proteolytic cascade, is prominently expressed during the tissue response to injury although the factors that impact PAI-1 induction and their role in the repair process are unclear. Kinetic modeling using established biomarkers of cell cycle transit (c-MYC; cyclin D1; cyclin A) in synchronized human (HaCaT) keratinocytes, and previous cytometric assessments, indicated that PAI-1 transcription occurred early after serum-stimulation of quiescent (G0) cells and prior to G1 entry. It was established previously that differential residence of USF family members (USF1→USF2 switch) at the PE2 region E box (CACGTG) characterized the G0→G1 transition period and the transcriptional status of the PAI-1 gene. A consensus PE2 E box motif (5′-CACGTG-3′) at nucleotides -566 to -561 was required for USF/E box interactions and serum-dependent PAI-1 transcription. Site-directed CG→AT substitution at the two central nucleotides inhibited formation of USF/probe complexes and PAI-1 promoter-driven reporter expression. A dominant-negative USF (A-USF) construct or double-stranded PE2 “decoy” attenuated serum- and TGF-β1-stimulated PAI-1 synthesis. Tet-Off induction of an A-USF insert reduced both PAI-1 and PAI-2 transcripts while increasing the fraction of Ki-67+ cells. Conversely, overexpression of USF2 or adenoviral-delivery of a PAI-1 vector inhibited HaCaT colony expansion indicating that the USF1→USF2 transition and subsequent PAI-1 transcription are critical events in the epithelial go-or-grow response. Collectively, these data suggest that USF2, and its target gene PAI-1, regulate serum-stimulated keratinocyte growth, and likely the cadence of cell cycle progression in replicatively-competent cells as part of the injury repair program.

Keywords: PAI-1, USF, transcription, keratinocytes, gene regulation, expression profiling, SERPINE1

INTRODUCTION

Upstream stimulatory factor-1 and -2 (USF1/2) are members of the conserved basic helix-loop-helix/leucine zipper (bHLH-LZ) MYC family of E box-binding transcription factors [Littlewood and Evan, 1995; Corre and Gailbert, 2005]. Multiple signaling networks influence USF1/2 function largely through site-specific phosphorylation; dimer composition and recruited co-factors dictate target gene expression and growth control [Pawlus et al., 2013; Qi et al., 2006; Gailbert et al., 2001]. The cell cycle-related anti-proliferative functions of USF1/2, moreover, are context-dependent and involve specific mitogen-activated (p38, PKC, PKA) and cyclin-dependent (CDK1, CDK4) kinases as well as the APC, p27, BRCA2, p53 and TGF-β1 tumor suppressor pathways [Corre and Gailbert, 2005; Luo and Sawadogo, 1996; Jaiswal and Naravan, 2001; Qyang et al., 1999; Kim et al., 2008; Jung et al., 2007]. Inhibition of cell proliferation by USF family members likely requires down-regulation of cyclin B1 and CDK1, c-MYC suppression and p27 and p53 induction [Corre and Gailbert, 2005; Kim et al., 2008; Jung et al., 2007].

USF1/2 regulate growth state-dependent transcription of the plasminogen activator inhibitor-1 (PAI-1, SERPINE1) gene [Qi et al., 2006], a major TGF-β1 and p53 target [Akiyoshi et al., 2001; Kortlever et al., 2006; Freytag et al., 2010; Allen et al., 2005]. Physiologically, PAI-1 modulates the pericellular plasmin-generating cascade and is a prominent member of the serum-induced “wound-response” transcriptome (Iyer etr al., 1999; Qi et al., 2008]. In immortalized keratinocytes, PAI-1 is required for TGF-β1-stimulated planar migration and Matrigel barrier invasion perhaps by LRP1-dependent engagement of the Jak/Stat pathway [Freytag et al., 2010; Providence et al., 2008; Degryse et al., 2004]. Matricellular (i.e., anchored) PAI-1, in fact, promotes a mesenchymal-to-amoeboid transition with activation of signaling pathways required for efficient 3-D “stromal” migration [Cartier-Michaud et al., 2012]. The consistent association of PAI-1 expression with the global program of tissue injury [Iyer et al., 1999; Providence and Higgins, 2004; Fitsialos et al., 2007] suggests that this SERPIN integrates cycles of cell-to-substrate adhesion/dis-adhesion with repair “scaffold” remodeling to meet the requirements for effective cellular migration [Czekay et al., 2011]. The specific phenotypic impact of PAI-1, however, likely depends on a balance of its intracellular (caspase-3 inhibition) and extracellular (protease inhibitory, receptor signaling, suppression of CUB domain-containing protein 1 cleavage by plasmin) activities that may be both cell type- and stimulus-specific [Law et al., 2013].

PAI-1 transcription is an early event in serum-stimulated quiescent (G0) keratinocytes. Expression occurs prior to G1 entry and involves a USF subtype switch (USF1→USF2) at the PE1/PE2 site E box motifs (5′-CACGTG-3′) in the PF1 region (nucleotides -794 to -532) of the PAI-1 promoter [Qi et al., 2006]. The typical rapid decline in PAI-1 mRNA levels several hours prior to the onset of DNA synthesis [Qi et al., 2006; Mu et al., 1998], however, suggests that PAI-1 may have a negative influence on cell growth. PAI-1 knockdown, in fact, leads to escape from both senescence-associated proliferative arrest and TGF-β1-induced cytostasis in primary fibroblasts and human (HaCaT) keratinocytes [Kortlever et al., 2006,2008]. Certain growth suppressive genes, particularly those responsive to the USF family of anti-proliferative bHLH-LZ transcription factors, may actually function as key regulators of the “go or grow” proliferation/migration dichotomy. The current paper addresses this issue; the evidence implicates USF2 as an integrator of human keratinocyte growth and PAI-1 gene control.

MATERIALS AND METHODS

Cell lines and expression vectors

Human HaCaT keratinocytes were cultured as described [Qi et al., 2006]. A HA-tagged dominant-negative USF construct (A-USF) [Qyang et al., 1999; Krylov et al., 1997] was cloned in the Mlu I/EcoR V site of pBl and co-transfected with pTRE2 (for Tet-Off induction). Mock transfectants (empty vector+pTRE2) served as controls. Insert expression after oxytetracycline removal was confirmed by RT-PCR and western blotting. Full-length human PAI-1 cDNA, excised from pcDNA3.1 using 5′-KpnI and 3′-NotI, was cloned into AdTrack-CMV to produce recombinant adenovirus [Law et al., 2013]. The pCMV-USF2 vector was detailed previously (4). Regions of the human PAI-1 promoter were PCR-amplified using primers to add 5′ SacI and 3′ XhoI restriction sites. PCR products were gel-purified, ligated into TOPO TA and cloned into the SacI/XhoI site of the pGL3 luciferase expression vector to create a full length (nucleotides -806 to +72) construct [Allen et al., 2005]. The dinucleotide substitution CACGTG→CAATTG in the PE1 and PE2 region E boxes was created by site-directed mutagenesis. HaCaT cells were co-transfected with individual reporter constructs and a SV40-driven β-galactosidase-expressing plasmid (for signal normalization) using Lipofectamine, FBS-deprived for 1 day then stimulated with serum or TGF-β1 for 8 hours prior to extraction and reporter analysis.

Microarray analyses

Transcript profiling utilized Affymetrix Human Genome U133 Plus 2.0 arrays [Qi et al., 2008] and Affymetrix GCOS software; probe signal outputs (pivot tables) were imported into GeneSpring v6.1. Expression profiles of empty vector vs. A-USF transfectants after oxytetracycline removal used the PAHS-028A human tumor metastasis and PAHS-012A human apoptosis RT2 PCR arrays (SABiosciences). Expression levels of selected genes (2ΔΔCt based-fold change calculations), normalized to housekeeping controls, was determined by real-time PCR used a MyiQ Cycler system (Bio-Rad, Hercules, CA).

Northern blotting and RT-PCR

Cellular RNA was denatured at 55°C for 15 minutes in 1X MOPS, 6.5% formaldehyde and 50% formamide, size-fractionated on 1% agarose/formaldehyde gels using 1X MOPS, transferred with 10X SSC to positively-charged nylon membranes and UV crosslinked. 32P-labeled cDNA probes to PAI-1, uPA or GAPD were hybridized overnight at 42°C in 50% formamide, 2.5X Denhardt’s solution, 1% SDS, 100 μg/ml sheared/denatured salmon sperm DNA, 5X SSC, 10% dextran sulfate and washed 3 times with 0.1X SSC/0.1X SDS for 15 min each at 42°C then at 55°C prior to exposure to film. For RT-PCR, total RNA was isolated with Qiagen RNeasy mini-columns (Qiagen, Valencia, CA) and first strand cDNA synthesized by addition of MMLV RNase H+ iScript reverse transcriptase (BioRad, Hercules, CA) to a mixture of 2–10 μg RNA and oligo(dT)/random primers. Forward and reverse primer sets for selected genes are described in Table 1.

Table 1.

Primer sets for RT-PCR validation of microarray profiling.

Primer Sequence
human beta-actin, forward primer 5′-GAGAAAATCTGGCACCAC-3′
human beta-actin, reverse primer 5′-CTAGAAGCATTTGCGGTG-3′
human uPAR, forward primer 5′-GGGAAGAAGGAGAAGAGC-3′
human uPAR, reverse primer 5′-TTTCGGTTCGTGAGTGCC-3′
human HSPC159, forward primer 5′-CCCACGACTGATAGTTCC-3′
human HSPC159, reverse primer 5′-TGGTGATCTGGAGGTCTC-3′
human MAP3K8, forward primer 5′-CACCTCATGAGACTCTCC-3′
human MAP3K8, reverse primer 5′-GTTGCTAGGTTTAATATC-3′
human EGR3, forward primer 5′-GGCTACAGAGAATGTAAT-3′
human EGR3, reverse primer 5′-GAATGCCTTGATGGTCTC-3′
human dual specificity phosphatase 1, forward primer 5′-CAAAGGAGGATACGAAGC-3′
human dual specificity phosphatase 1, reverse primer 5′-GATGGAGTCTATGAAGTC-3′
human 18 S rRNA, forward primer 5′-TTCAAAGCAGGCCCGAGC-3′
human 18 S rRNA, reverse primer 5′-CGTCTTCTCAGCGCTCCG-3′
Open in a new tab

Electrophoretic mobility shift assay

Keratinocytes were disrupted in cold hypotonic buffer (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 1 mM DTT, 0.1 mM EGTA, 0.5 mM PMSF, 0.6% NP-40), nuclei collected at 15,000 xg for 1 minute, lysed on ice for 30 minutes (in 20 mM HEPES, pH 7.9, containing 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 1 mM PMSF) and nuclear extracts clarified at 15,000 x g for 5 minutes. The 32P-end-labeled double-stranded PAI-1 promoter PE2 E box region probe: 5′-CCAAGTCCTAGACAGACAAAACCTAGACAATCACGTGGCTGGCTGC-3′ was incubated with 2–10 μg nuclear extract protein in binding buffer (4% glycerol, 1 mM MgCl2, 0.5 mM EDTA, 0.5 mM DTT, 50 mM NaCl, 1 μg/ml poly dI:dC, 10 mM Tris-HCl, pH 7.5) at room temperature for 20 minutes followed by an additional 30 minute incubation (where indicated) with antibodies to USF2, MAX, TEF3 or E2F1 (Santa Cruz Biotechnology) and complexes separated on 4% polyacrylamide gels in TBE buffer. PE2 sequence requirements for PE2 region probe binding were determined using wild-type and mutant PE2 competing constructs and mobility shift assay as detailed [Qi et al., 2006].

Western Blotting

Trypsin-released cells (at the indicated time points after serum reintroduction) were collected by centrifugation, lysed in 4% SDS (in Ca2+/Mg2+-free PBS) and extracts boiled for 5 minutes. Following electrophoretic separation (of 30 μg of protein) and transfer to nitrocellulose, membranes were blocked in 5% milk in 0.05% Triton-X 100/PBS, incubated overnight with antibodies to the HA tag, PAI-1 or c-MYC in 5% milk in 0.05% Triton-X 100/PBS, washed three times in 0.05% Triton-X 100/PBS and incubated with appropriate secondary antibodies. Immunoreactive proteins were visualized with ECL reagent. Membranes were stripped and reprobed with ERK2 antibodies (Santa Cruz Biotechnology) to assess loading levels.

RESULTS

Kinetics of serum-induced PAI-1 expression

The 2 to 6 hour window after serum-stimulation, based on c-MYC and cyclin D1 induction (Figure 1A–C) as well as prior cytometric staging [Qi et al., 2006], appears to be a commitment point in the entry of quiescent HaCaT keratinocytes into a cycling G1 state. Microarray analysis of serum-starved and 2 hour FBS-“activated” HaCaT cells served, therefore, to define the transcriptional signature of the early G0→G1 transition [Qi et al., 2008]. Prominently up-regulated transcripts included those encoding proteins involved in the initial growth response (EGR1-4), transcription (Myc, Fos, Jun, KLF4, AT3, p300/pTAF) and ECM remodeling (uPA, uPAR, tPA, SERPINE1 [PAI-1], SERPINB1 [leukocyte elastase inhibitor], SERPINB2 [PAI-2], CTGF, MMP-2, MMP-12, CYR61) (Figure 2A). Rank ordering of microarray, northern blot and RT-PCR data indicated that PAI-1 and the uPA receptor (PLAUR) (Figure 2A–C) partitioned to the most significantly-elevated group. PAI-1 transcripts, both the 3.0- and 2.2-kb species, were low to undetectable in quiescent HaCaT cells, peaked approximately 2 hours after serum-addition (Figure 2D) which correlated with residence in an early activated G0 substate and declined prior to synchronous entry into S phase [Qi et al., 2006]. PAI-1 protein synthesis closely followed induction of PAI-1 mRNA (Figure 2E). uPA levels increased (by 12-fold) as well although maximal expression occurred several hours later than PAI-1 mRNA (Figure 2D).

Figure 1. Kinetics of c-MYC and cyclin D1 induction following serum-stimulation of quiescent HaCaT keratinocytes.

Figure 1

Open in a new tab

The fraction of DNA-synthesizing (i.e., % BrdU+) cells in non-synchronous proliferating HaCaT cultures approximated 24% of the total population (A). The switch to FBS-free medium resulted in a rapid decrease in S phase cells (to 3.8% within 24 hours) which further declined to just 0.2% by day 3 of serum-deprivation. Data plotted in (A) represents the mean±SD of the % BrdU+ HaCaT keratinocytes in 20, randomly-selected, fields for each culture condition. c-MYC was not detectable in quiescent (Q) HaCaT cells but was evident as early as 2 hours post-serum addition; by 6 hours, c-MYC expression was significantly elevated and subsequently increased upon progression through G1 (B). Serum re-introduction induced expression and nuclear accumulation of the G1 cyclin D1, but not the S phase cyclin A, within 5 hours after initial stimulation (C).

Figure 2. Expression profiling of serum-responsive genes.

Figure 2

Open in a new tab

Microarray analysis identified several significantly up-regulated transcripts within 2 hours after serum addition (A). The clustergram (B) illustrates functional and locational groupings for proteins encoded by several highly-expressed (red to orange shading), relative to more moderately-stimulated, genes (map constructed with Ingenuity Pathways software). A complete list of 79 genes that are >10-fold serum-increased in HaCaT cells is available [Qi et al., 2008]. RT-PCR confirmed induction of several transcripts including HSPC159, EGR3, uPAR, MAP3K8 and DUSP1 (C). PAI-1 mRNAs (both the 3.0 and 2.2 kb species) are low to undetectable in quiescent (Q) HaCaT keratinocytes and induced within 1 to 2 hours of serum re-introduction (D). uPA transcripts, while also elevated at the 2 hour time point, were not highly expressed until 6 hours post-FBS stimulation. Induction of PAI-1 mRNA was closely followed by a marked increase in immunocytochemically-detectable PAI-1 protein (E). Collectively, these data indicate that 3 important modulators of the pericellular plasmin generating system (uPA, uPAR, and the pathway negative regulator PAI-1) are prominent members of the early response to serum-stimulation.

Interference with USF-DNA binding impacts PAI-1 expression

The PE2 CACGTG hexanucleotide motif resides 3 nucleotides downstream of a trio of SMAD-binding elements located at positions -595 to -569 upstream of the transcription start site [Dimova and Kietzmann, 2006] (Figure 3A). Mutation of the central nucleotides (CG→AT) (Figure 3A), which ablates binding of USF to PE2 region probes [Qi et al., 2006], established the importance of an intact PE2 E box consensus sequence in serum-stimulated PAI-1 promoter-driven reporter transcription (Figure 3B). Mobility shift data indicated that PE2 probe recognition by USF2 is constitutively present and active regardless of cellular growth state (Figure 3C,D). Prior chromatin immunoprecipitation analyses established that the growth factor-responsive PE2 E box sites in the PAI-1 gene are, indeed, USF target sequences in vivo and that serum-stimulated PAI-1 expression reflected a dynamic USF1→USF2 switch at this CACGTG site [Qi et al., 2006]. Expression control by USF family members, therefore, is distinct from simple motif recognition [Dimova and Kietzmann, 2006; Samoylenko et al., 2001; Qi et al., 2006]. Mobility shift assesments, moreover, confirmed that constructs that retained an intact E box site functioned as effective competitors while those lacking the CACGTG motif or containing the transcription-attenuating CG→AT mutation (as in Figure 3A,B) failed to block complex formation (Figure 3E). Based on these collective findings, it was important to determine if disruption of USF function would specifically affect induced PAI-1 expression. Two approaches were selected to evaluate this possibility. Initially, keratinocytes were transfected with a dominant-negative A-USF construct in which replacement of the wild-type DNA-binding domain with acidic residues (Figure 3F) inhibited formation of endogenous USF/DNA complexes [Furbass et al., 2010] with the PE2 (human) and HRE-2 (rat) PAI-1 target probes [Gailbert et al., 2001; Kutz et al., 2006]. A-USF expression significantly attenuated both serum- and TGF-β-induced PAI-1 levels (Figure 3G). Secondly, using the identified sequence restraints for reporter activation (Figure 3A,B) and DNA binding [Qi et al., 2006], a double-stranded 45-bp PE2 DNA construct was designed based on the previously identified requirements for an intact CACGTG motif for probe recognition by USF [Allen et al., 2005] Transfection of these double-stranded USF-binding, “decoys” (Figure 3C) effectively reduced both serum- and TGF-β1-induced PAI-1 transcript levels in HaCaT keratinocytes (Figure 3H).

Figure 3. The PE2 region E box is an important USF2-binding, PAI-1 expression-regulating motif, in HaCaT cells.

Figure 3

Open in a new tab

Position of the PE2 region E box motif (including the introduced CG→AT mutations) and the 3 SMAD-binding AGAC sites just upstream of the PE2 E box are indicated (A). Luciferase activity for cells transfected with the wild-type and CG→AT mutated PE2 E box reporter constructs, or one lacking PAI-1 promoter sequences (basal; pGL3 only=1.0), were determined after a 12 hour serum exposure (B). Normalization was to a co-transfected SV40 promoter-β-galactosidase expression vector. Data plotted (B) is the mean+SD of 3 independent experiments and expressed as fold-increase relative to basal luciferase signal. Mobility shift assays using a double-stranded 32P-labeled PE2 construct (C; only top strand is illustrated) were incubated with nuclear extracts from proliferating (P), quiescent (Q) and 2 as well as 24 hour (2, 24) serum-stimulated HaCaT cells. Positions of the original protein-probe complex and the USF antibody-induced supershift are indicated with arrowheads (D); (–) = absence of nuclear extract; none = no IgG added. Antibodies (2 μg) to MAX, TFE3 or E2F1 did not produce supershifts (not shown). The PE2 sequence requirements for PE2 region probe binding were determined using the topographic map in (C) to design wild-type and mutant PE2 competing constructs for mobility shift assay (E). Nuclear extracts were incubated with 32P-labeled PE2 probe in the presence or absence of a 100-fold molar excess of the following unlabeled competing DNAs: Self (wild-type 45-mer PE2 sequence), PE1 (upstream region; only homology to PE2 is a consensus E box), PE2SBE-3′ (proximal SBE+AAT spacer+E box), PE2-3′ (AAT+E box), SBE mutant (45-mer PE2 sequence with all 3 SBEs mutated), -AAT (45-mer PE2 sequence with the AAT spacer deleted) and non-competing constructs: PE2SBE-5′ (2 5′-SBEs), CA→AT mutant (45-mer PE2 sequence with the 2 central E box nucleotides mutated [Qi et al., 2006, for details as to sequences used for mobility shift assay]. (E) adapted from [Qi et al., 2008]. To determine if interference with USF-PE2 site recognition had an effect on PAI-1 induction, Keratinocytes were transfected with the dominant-negative (A-USF) USF construct (F), cultured under quiescence (Q) conditions then stimulated with FBS (10% final concentration) or TGF-β1 (1 ng/ml) for 5 hours prior to extraction for western analysis of cellular PAI-1 protein. A-USF expression significantly attenuated (by approximately 80%) PAI-1 induction in response to both serum and TGF-β1 (G). Transfection of HaCaT cells with a double-stranded USF-binding 45-bp PE2 region decoy construct (C) markedly decreased the levels of both serum- and TGF-β1-induced PAI-1 transcripts (H). Insert (in H) is a representative northern blot. Graphed data (G,H) is the mean±SD of 3 independent experiments. sDNA = sheared, double-stranded, control DNA; SBE = SMAD-binding elements.

Inducible A-USF expression attenuates PAI-1 expression and stimulates human keratinocyte proliferation

Although USF1/2 regulate the expression of genes that impact cell growth [Corre and Gailbert, 2005], the actual cause-effect relationship between specific genomic outputs and the proliferative program is largely unknown. Initial experiments confirmed that transfection of a wild-type USF2 expression vector reduced colony-forming efficiency of HaCaT cells by approximately 80% (Figure 4A). To more specifically implicate USF in proliferative control, a DOX-dependent A-USF expression system was developed. Keratinocytes were engineered to express an HA-tagged dominant-negative A-USF construct (DN-USF) in an inducible Tet-OFF system. Insert-containing clones were selected based on RT-PCR and HA-tag immunoreactivity following oxytetracycline removal (Figure 4B, insert). Drug removal and subsequent A-USF induction significantly increased cell number (Figure 4B), the fraction of Ki-67+ keratinocytes (Figure 4C,D) and Ki-67 transcripts while suppressing both PAI-1 and PAI-2 mRNA levels (Figure 4E). Adenoviral-mediated PAI-1 overexpression, moreover, dramatically inhibited keratinocyte colony expansion (Figure 4F) mimicing the results of USF2 transfection (Figure 4A).

Figure 4. Expression of pCMV-USF2, a DOX-responsive A-USF construct and adenoviral-mediated PAI-1 over-expression inhibit HaCaT cell proliferation.

Figure 4

Open in a new tab

Transfection of pCMV-USF2 effectively attenuated HaCaT keratinocyte growth, relative to neomycin controls, as assessed by colony-forming efficiency in 10% FBS-containing medium (A). For HaCaT cells engineered to express an inducible A-USF insert upon DOX removal (confirmed by western blot detection of the HA tag), a significant increase in growth was evident within 4 days after drug withdrawal compared to mock (empty vector+pTRE2) transfectants (B). Induction of A-USF increased the fraction of Ki-67+ cells compared to empty vector controls (C,D). Tet-OFF A-USF expression augmented Ki-67 mRNA abundance approximately 8-fold, relative to mock transfectants, while inhibiting expression of both PAI-1 and PAI-2 transcripts (E). Growth inhibition (i.e., reduced colony expansion) was evident in HaCaT populations infected with CMV-PAI-1/GFP encoding adenoviruses but not in keratinocytes infected with a control GFP-only adenovirus (F).

DISCUSSION

The PE2 E box is 5′-flanked by three SMAD-binding sites and a minimum of two SMAD-recognition (AGAC) sequences is necessary for the maximal response of the PAI-1 gene to TGF-β1, at least in cells in which TFE3 is a major E box-binding trans-activator [Hua et al., 1999]. The two proximal AGAC motifs, however, are not required for USF occupancy of a PAI-1 PE2 region E box target since non-SMAD site E box constructs or SMAD site-deleted or mutated constructs effectively compete for USF-PE2 probe complex formation [Allen et al., 2005; Qi et al., 2006; this paper]. Successful competing sequences include the 18-bp region of the rat PAI-1 proximal promoter containing the USF/HIF-1α-binding HRE-2 E box (5′-TACACACACGTGTCCCAG-3′) and a 23-bp consensus E box construct with no PAI-1 flanking homologies (5′-CACCCGGTCACGTGGCCTACACC-3′) but not the 23-bp central nucleotide E box mutant (5′-CACCCGGTCAATTGGCCTACACC-3′) [Qi et al., 2006; this paper]. Use of separate PE1/PE2 region probes confirmed that nuclear USF1/2 binding to PE1/PE2 probe targets in serum-stimulated cells was independent of growth state; transcriptional activation of the PAI-1 gene, moreover, reflects a USF1→USF2 switch at both the PE1 and PE2 E box sites [Qi et al., 2006]. USF1→USF2 dimer replacement at the critical PE2 E box motif and induced PAI-1 expression occurs early after cellular “activation”. Collectively, these findings as well as the luciferase reporter data (Figure 3A,B) highlight the importance of an intact PE2 region E box in PAI-1 gene control.

While a consensus E box motif at the PE2 site is both necessary and sufficient for USF binding, the role of co-factors in PAI-1 gene control reflects whether transcription is TGF-β1- or serum-initiated. pSMAD2/3 involvement, at the PE2 region 5′ SMAD sites, for example, is stimulus-dependent as inhibition of SMAD3 phosphorylation with SIS3 effectively attenuates PAI-1 expression in response to TGF-β1 but not serum [Samarakoon et al., 2013]. Similarly, Yin Yang 1 (YY1) represses some SMAD-dependent transcriptional targets (the early-response PAI-1, Id1 genes) but not others (p15, p21, c-myc). YY1, in fact, inhibits SMAD occupancy of the PAI-1 promoter likely via Interactions with the N-terminal MAD homology domain in SMAD4 and SMAD2/3 with suppression dependent on the number of SMAD-binding elements in the specific target gene [Kurisaki et al., 2003]. Importantly, USF2 appears to contribute to C/EBPβ target gene expression by interacting with YY1, either via the repressive domain or by blocking formation of YY1/co-regulator complexes to antagonize YY1 [Viart et al., 2013]. Interactions between USF2 and YY1 release YY1 inhibition stimulating C/EBPβ transactivation of target genes [Viart et al., 2013]. Other mechanisms have implications with regard to cell growth control. USF1/2 interact with components of the basal transcription complex [Garcia-Sanz et al., 2013] and guide recruitment of the hSET1A histone methylase complex to chromatin as part of lineage-specifc cell differentiation [Deng et al., 2013]. Recent findings suggest, moreover, that USF2 is required for induction of HIF2α responsive genes (including PAI-1) under hypoxic conditions likely involving recruitment of the histone acetylases p300 and CBP [Palmeri et al., 2002]. While introduction of a USF2 expression vector alone stimulated PAI-1 promoter activity, co-transfection of a construct encoding a stabilized form of HIF2α along with the USF2 vector potentiated PAI-1 transcription more than either alone [Pawlus et al., 2013].

PAI-1 transcriptional activation has important phenotypic consequences. Several SERPINS (SERPINE1 [PAI-1], SERPINB1, SERPINB2) are prominent members of the “tissue repair” transcriptome functioning in the integrated control of focalized matrix restructuring, cell-to-substrate adhesion/detachment, migration and proliferation [Qi et al., 2008; Iyer et al., 1999; Providence et al., 2008; Degryse et al., 2004; Providence and Higgins, 2004; Fitsialos et al., 2007; Palmeri et al., 2002; Rossignol et al., 2004; Wang et al., 2005; Chan et al., 2001; Ebrahimian et al., 2012; Dayem et al., 2003]. PAI-1 effectively limits pericellular plasmin generation to maintain a supporting “scaffold” for cell movement [Czekay et al., 2011] while also regulating urokinase-dependent growth factor activation attenuating, thereby, the associated proliferative response [Kortlever et al., 2006]. Indeed, PAI-1 is highly-expressed in cells undergoing replicative senescence [Kortlever et al., 2006; Mu et al., 1998] and certain USF target “growth arrest-associated” genes (i.e., p16INK4a, PAI-1) may function in the wound repair program to inhibit proliferation while promoting migration. Perhaps not unrelated is the finding that most genes activated by USF family members typify differentiated, largely non-cycling, cells [Liang et al., 2009]. The USF1→USF2 transition at the PAI-1 PE2 E box, and subsequent PAI-1 transcription, appears to be a critical aspect of the go-or-grow response during epidermal wound healing. PAI-1 may regulate the temporal cadence of cell cycle progression in replicatively-competent cells as part of the injury repair program.

Acknowledgments

Grant Sponsors: NIH grant R01 GM057242 (PJH), an AHA Fellowship 14PRE18170012 (TMS) and the generous support of the Friedman Family Cancer Fund, the Graver Family Endowed Fund for Cancer Research and the Butler Family Foundation (PJH).

Abbreviations

USF

upstream stimulatory factor

PAI-1

plasminogen activator inhibitor-1

TGF-β1

transforming growth factor-β1

References

  1. Akiyoshi S, Ishii M, Nemoto N, Kawabata M, Aburatani H, Miyazono K. Targets of transcriptional regulation by transforming growth factor-β: expression profile analysis using oligonucleotide arrays. Jpn J Cancer Res. 2001;92:257–268. doi: 10.1111/j.1349-7006.2001.tb01090.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allen RR, Qi L, Higgins PJ. Upstream stimulatory factor regulates E box-dependent PAI-1 transcription in human epidermal keratinocytes. J Cell Physiol. 2005;203:156–165. doi: 10.1002/jcp.20211. [DOI] [PubMed] [Google Scholar]
  3. Cartier-Michaud A, Malo M, Charriere-Bertrand C, Gadea G, Anguille C, Supiramaniam A, Lesne A, Delaplace F, Hutzler G, Roux P, Lawrence DA, Barlovatz-Meimon G. Matrix-bound PAI-1 supports cell blebbing via RhoA/ROCK1 signaling. PLoS One. 2012;7:e32204. doi: 10.1371/journal.pone.0032204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chan JC, Duszczyszyn DA, Castellino FJ, Ploplis VA. Accelerated skin wound healing in plasminogen activator inhibitor-1-deficient mice. Am J Pathol. 2001;159:1681–1688. doi: 10.1016/S0002-9440(10)63015-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Corre S, Gailbert M-D. Upstream stimulating factors: highly versatile stress-responsive transcription factors. Pigment Cell Res. 2005;18:337–348. doi: 10.1111/j.1600-0749.2005.00262.x. [DOI] [PubMed] [Google Scholar]
  6. Czekay RP, Wilkins-Port CE, Higgins SP, Freytag J, Overstreet JM, Klein RM, Higgins CE, Samarakoon R, Higgins PJ. PAI-1: an integrator of cell signaling and migration. Int J Cell Biol. 2011;2011:562481. doi: 10.1155/2011/562481. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dayem MA, Moreihon C, Turchi L, Magnone V, Christen R, Ponzio G, Barbry P. Early gene expression in wounded human keratinocytes revealed by DNA microarray analysis. Comp Funct Genom. 2003;4:47–55. doi: 10.1002/cfg.239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Degryse B, Neels JG, Czekay RP, Aertgeerts K, Kamikubo Y, Loskutoff DJ. The low density lipoprotein receptor-related protein is a motogenic receptor for plasminogen activator inhibitor-1. J Biol Chem. 2004;279:22595–22604. doi: 10.1074/jbc.M313004200. [DOI] [PubMed] [Google Scholar]
  9. Deng C, Li Y, Liang S, Cui K, Salz T, Yang H, Tang Z, Gallagher PG, Qiu Y, Roeder R, Zhao K, Bungert J, Huang S. USF1 and hSET1A mediated epigenetic modifications regulate lineage differentiation and HKoxB4 transcription. PLoS Genet. 2013;9:e1003524. doi: 10.1371/journal.pgen.1003524. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dimova EY, Kietzmann T. Cell type-dependent regulation of the hypoxia-responsiveness plasminogen activator inhibitor-1 gene by upstream stimulatory factor-2. J Biol Chem. 2006;281:2999–3005. doi: 10.1074/jbc.M512078200. [DOI] [PubMed] [Google Scholar]
  11. Ebrahimian TG, Squiban C, Roque T, Lugo-Martinez H, Hneino M, Buard V, Gourmelon P, Benderitter M, Milliat F, Tamarat R. Plasminogen activator inhibitor-1 controls bone marrow-derived cells therapeutic effect through MMP9 signaling: role in physiological and pathologic wound healing. Stem Cells. 2012;30:1436–1446. doi: 10.1002/stem.1126. [DOI] [PubMed] [Google Scholar]
  12. Fitsialos G, Chassot AA, Turchi L, Dayem MA, LeBrigand K, Moreihon C, Meneguzzi G, Busca R, Mari B, Barbry P, Ponzio G. Transcriptional signature of epidermal keratinocytes subjected to in vitro scratch wound reveals selective roles for ERK1/2, p38, and phosphatidylinositol 3-kinase signaling pathways. J Biol Chem. 2007;282:15090–15102. doi: 10.1074/jbc.M606094200. [DOI] [PubMed] [Google Scholar]
  13. Freytag J, Wilkins-Port CE, Higgins CE, Higgins SP, Samarakoon R, Higgins PJ. PAI-1 mediates the TGF-β1+EGF-induced “scatter” response in transformed human keratinocytes. J Invest Dermatol. 2010;130:2179–2190. doi: 10.1038/jid.2010.106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Furbass R, Tomek W, Vanselow J. Upstream stimulating factors 1 and 2 enhance transcription from the placenta-specific promoter 1.1 of the bovine cyp19 gene. BMC Mol Biol. 2010;11:5. doi: 10.1186/1471-2199-11-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gailbert MD, Carreira S, Goding CR. The Usf-1 transcription factors is a novel target for the stress-responsive p38 kinase and mediates UV-induced tyrosine expression. EMBO J. 2001;20:5022–5031. doi: 10.1093/emboj/20.17.5022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Garcia-Sanz P, Fernandez-Perez A, Vallejo M. Differential configurations involving binding of USF transcription factors and Twist1 regulate Alx3 promoter activity in mesenchymal and pancreatic cells. Biochem. 2013;450:199–208. doi: 10.1042/BJ20120962. [DOI] [PubMed] [Google Scholar]
  17. Hua X, Miller ZA, Wu G, Shi Y, Lodish HF. Specificity in transforming growth factorβ-induced transcription of the plasminogen activator inhibitor-1 gene: interactions of promoter DNA, transcription factor muE3, and Smad proteins. Proc Natl Acad Sci USA. 1999;96:13130–13135. doi: 10.1073/pnas.96.23.13130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Iyer VR, Eisen MB, Ross DT, Schuler G, Moore T, Lee JC, Trent JM, Staudt LM, Hudson J, Jr, Boguski MS, Lashkari D, Shalon D, Botstein D, Brown PO. The transcriptional program in the response of human fibroblasts to serum. Science. 1999;283:83–87. doi: 10.1126/science.283.5398.83. [DOI] [PubMed] [Google Scholar]
  19. Jaiswal AS, Naravan S. Upstream stimulation factor-1 (USF1) and USF2 bind to and activate the promoter of the adenomatous polyposis coli (APC) tumor suppressor gene. J Cell Biochem. 2001;81:262–277. doi: 10.1002/1097-4644(20010501)81:2<262::aid-jcb1041>3.0.co;2-r. [DOI] [PubMed] [Google Scholar]
  20. Jung HS, Kim KS, Chung YJ, Chung HK, Min YK, Lee MS, Lee MK, Lim KW, Chung JH. USF inhibits cell proliferation through delay in G2/M phase in FRTL-5 cells. Endocrine J. 2007;54:275–285. doi: 10.1507/endocrj.k06-120. [DOI] [PubMed] [Google Scholar]
  21. Kim KS, Jung HS, Chung YJ, Jung TS, Jang HW, Lee MS, Kim KW, Chung JH. Overexpression of USF increases TGF-β1 protein levels, but G1 phase arrest was not induced in FRTL-5 cells. J Korean Med Sci. 2008;23:870–876. doi: 10.3346/jkms.2008.23.5.870. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kortlever RM, Higgins PJ, Bernards R. Plasminogen activator inhibitor-1 is a critical downstream target of p53 in the induction of replicative senescence. Nat Cell Biol. 2006;8:877–884. doi: 10.1038/ncb1448. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kortlever RM, Nijwening JH, Bernards R. Transforming growth factor-β requires its target plasminogen activator inhibitor-1 for cytostatic activity. J Biol Chem. 2008;283:24308–24313. doi: 10.1074/jbc.M803341200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Krylov D, Kaasi K, Echlin DR, Taparowsky EJ, Arnheiter H, Vinson C. A general method to design dominant negatives to B-HLHZip proteins that abolish DNA binding. Proc Natl Acad Sci USA. 1997;94:12274–12279. doi: 10.1073/pnas.94.23.12274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kurisaki K, Kurisaki A, Valcourt U, Terentiev AA, Pardali K, ten Dijke P, Heldin CH, Ericsson J, Moustakas A. Nuclear factor YY1 inhibits transforming growth factor β- and bone morphogenetic protein-induced cell differentiation. Mol Cell Biol. 2003;23:4494–4510. doi: 10.1128/MCB.23.13.4494-4510.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kutz SM, Higgins CE, Samarakoon R, Higgins SP, Allen RR, Qi L, Higgins PJ. TGF-β1-induced PAI-1 expression is E box/USF-dependent and requires EGFR signaling. Exp Cell Res. 2006;312:1093–1105. doi: 10.1016/j.yexcr.2005.12.027. [DOI] [PubMed] [Google Scholar]
  27. Law ME, Corsino PE, Jahn SC, Davis BJ, Chen S, Patel B, Pham K, Lu J, Sheppard B, Norgaard P, Hong J, Higgins PJ, Kim J-S, Luesch H, Law BK. Glucocorticoids and histone deacetylase inhibitors cooperate to block the invasiveness of basal-like breast cancer cells through novel mechanisms. Oncogene. 2013;32:1316–1329. doi: 10.1038/onc.2012.138. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Liang SY, Moghimi B, Crusselle-Davis VJ, Lin IJ, Rosenberg MH, Li X, Stouboulis J, Huang S, Bungert J. Defective erythropoiesis in transgenic mice expressing dominant-negative upstream stimulatory factor. Mol Cell Biol. 2009;29:5900–5910. doi: 10.1128/MCB.00419-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Littlewood TD, Evan GI. Transcription factors 2: helix-loop-helix. Protein Profile. 1995;2:612–702. [PubMed] [Google Scholar]
  30. Luo X, Sawadogo M. Functional domains of the transcription factor USF2: atypical nuclear localization signals and context-dependent transcriptional activation domains. Mol Cell Biol. 1996;16:1367–1375. doi: 10.1128/mcb.16.4.1367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Mu XC, Staiano-Coico L, Higgins PJ. Increased transcription and modified growth state-dependent expression of the plasminogen activator inhibitor type-1 gene characterize the senescent phenotype in human diploid fibroblasts. J Cell Physiol. 1998;174:90–98. doi: 10.1002/(SICI)1097-4652(199801)174:1<90::AID-JCP10>3.0.CO;2-S. [DOI] [PubMed] [Google Scholar]
  32. Palmeri D, Lee JW, Juliano RL, Church FC. Plasminogen activactor inhibitor-1 and -3 increase cell adhesion and motility of MDA-MB-435 breast cancer cells. J Biol Chem. 2002;277:40950–40957. doi: 10.1074/jbc.M202333200. [DOI] [PubMed] [Google Scholar]
  33. Pawlus MR, Wang L, Ware K, Hu CJ. Upstream stimulatory factor 2 and hypoxia-inducible factor 2α (HIF2α) cooperatively activate HIF2 target genes during hypoxia. Mol Cell Biol. 2012;32:4595–4610. doi: 10.1128/MCB.00724-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Pawlus MR, Wang L, Murakami A, Dai G, Hu C-J. STAT3 or USF2 contributes to HIF target gene specificity. PLoS One. 2013;8:e72358. doi: 10.1371/journal.pone.0072358. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Providence KM, Higgins PJ. PAI-1 expression is required for epithelial cell migration in two distinct phases of in vitro wound repair. J Cell Physiol. 2004;200:297–308. doi: 10.1002/jcp.20016. [DOI] [PubMed] [Google Scholar]
  36. Providence KM, Higgins SP, Mullen A, Battista A, Samarakoon R, Higgins CE, Wilkins-Port CE, Higgins PJ. SERPINE1 (PAI-1) is deposited into keratinocyte migration “trails” and required for optimal monolayer wound repair. Arch Dermatol Res. 2008;300:303–310. doi: 10.1007/s00403-008-0845-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Qi L, Allen RR, Lu Q, Higgins CE, Garone R, Staiano-Coico L, Higgins PJ. PAI-1 transcriptional regulation during the G0→G1 transition in human epidermal keratinocytes. J Cell Biochem. 2006;99:495–507. doi: 10.1002/jcb.20885. [DOI] [PubMed] [Google Scholar]
  38. Qi L, Higgins SP, Lu Q, Samarakoon R, Wilkins-Port CE, Ye Q, Higgins CE, Staiano-Coico L, Higgins PJ. SERPINE1 (PAI-1) is a prominent member of the early the early G0→G1 transition “wound repair” transcriptome in p53 mutant human keratinocytes. J Invest Dermatol. 2008;128:749–753. doi: 10.1038/sj.jid.5701068. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Qyang Y, Luo X, Lu T, Ismail PM, Krylov D, Vinson C, Sawadogo M. Cell-type-dependent activity of the ubiquitous transcription factor USF in cellular proliferation and transcriptional activation. Mol Cell Biol. 1999;19:1508–1517. doi: 10.1128/mcb.19.2.1508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Rossignol P, Ho-Tin-Noe B, Vranckx R, Bouton MC, Meilhac O, Lijnen HR, Guillin MC, Michel JB, Angles-Cano E. Protease nexin-1 inhibits plasminogen activation-induced apoptosis of adherent cells. J Biol Chem. 2004;279:10346–10356. doi: 10.1074/jbc.M310964200. [DOI] [PubMed] [Google Scholar]
  41. Samarakoon R, Dobberfuhl AD, Cooley C, Overstreet JM, Patel S, Goldschmeding R, Meldrum KK, Higgins PJ. Induction of renal fibrotic genes by TGF-β1 requires EGFR activation, p53 and reactive oxygen species. Cell Signal. 2013;25:2198–2209. doi: 10.1016/j.cellsig.2013.07.007. [DOI] [PubMed] [Google Scholar]
  42. Samoylenko A, Roth U, Jungermann K, Kietzmann T. The upstream stimulatory factor-2a inhibits plasminogen activator inhibitor-1 gene expression by binding to a promoter element adjacent to the hypoxia-inducible factor-1 binding site. Blood. 2001;97:2657–2666. doi: 10.1182/blood.v97.9.2657. [DOI] [PubMed] [Google Scholar]
  43. Viart V, Varilh J, lopez E, Rene C, Claustres, Taulan-Cadars M. Phosphorylated C/EBPβ influences a complex network involving YY1 and USF2 in lung epithelial cells. PLoS One. 2013;8:e60211. doi: 10.1371/journal.pone.0060211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wang Z, Sosne G, Kurpakus-Wheater M. Plasminogen activator inhibitor-1 (PAI-1) stimulates human corneal epithelial cell adhesion and migration in vitro. Exp Eye Res. 2005;80:1–8. doi: 10.1016/j.exer.2004.06.006. [DOI] [PubMed] [Google Scholar]

RESOURCES