Mammalian Two-Hybrid Assay Showing Redox Control of HIF-Like Factor

David Lando; Daniel J. Peet; Ingemar Pongratz; Murray L. Whitelaw

Introduction

A growing number of transcription factors including activating protein-1 (AP-1), Myb, activating transcription factor/cAMP response element-binding protein (ATF/CREB), nuclear factor-κB (NF-κB), and p53 have been shown to have their DNA-binding activities modulated by changes in redox status (reviewed by Morel and Barouki1). For the Fos–Jun heterodimer that constitutes the AP-1 complex, the nuclear redox protein Ref-1 imparts DNA-binding activity by reducing specific cysteine sulfhydryl groups in the basic DNA-binding regions of Fos and Jun. Although the precise mechanism by which Ref-1 reduces Fos and Jun is not known, chemical cross-linking studies with bacterially expressed proteins have suggested that a direct cysteine-mediated interaction may occur between Ref-1 and Jun.2 However, in vivo cellular interactions between Ref-1 and Fos or Jun have not been demonstrated, nor have interactions been shown in cell extracts by standard biochemical techniques such as coimmunoprecipitation. This is most likely due to the transient nature of such interactions.

The mammalian hypoxia-inducible factors HIF-1α and HIF-like factor (HLF) are two highly related basic helix–loop–helix/Per–Arnt–Sim homology (bHLH/PAS) transcription factors that are rapidly activated by oxygen deprivation to induce a network of genes responsible for maintaining oxygen homeostasis (reviewed by Semenza3). The molecular mechanisms of oxygen sensing and signaling that lead to the activation of HIF-1α and HLF are poorly understood. However, evidence suggests the involvement of an oxygen-regulated redox signal that may activate a kinase cascade or modify the HIFs directly.4 In support of this, over-expression of redox factors Ref-1 and thioredoxin enhances the hypoxic response of HIF-1α and HLF.5–7 We have shown that Ref-1 can specifically enhance the in vitro DNA-binding activity of HLF by reducing a cysteine residue in the basic DNA-binding domain of HLF.8

Here we describe a cell-based mammalian two-hybrid assay that is capable of detecting the intracellular interaction of Ref-1 with the basic DNA-binding domain of HLF. By using this assay we are able to demonstrate that the interaction between Ref-1 and HLF is dependent on a redox-sensitive cysteine in the basic region of HLF.

Methods and Procedures

Principle

The mammalian two-hybrid system is a simple and inexpensive, yet powerful technique capable of assaying for protein–protein interactions within a cellular environment. Like the yeast two-hybrid system originally described by Fields and Song,9 this is an in vivo assay based on the functional reconstitution of a transcription activator. Typically, in this system, one protein of interest is expressed as a fusion protein with the yeast Gal4 DNA-binding domain (DBD), and the other protein is expressed as a fusion to the activation domain of the VP16 protein of the herpes simplex virus. The vectors that encode these fusion proteins are then cotransfected along with a luciferase reporter gene into a mammalian cell line. The expression of the luciferase gene within the reporter plasmid is under the control of Gal4-binding elements. If the two fusion proteins interact, this will bring together the Gal DBD and VP16 domains, forming a hybrid transcription activator capable of binding the Gal4 response elements and increasing the expression of the luciferase reporter gene. The increase in luciferase can then be quantitated with a luminometer.

Plasmids

The mammalian expression vectors we use to analyze Ref-1 interaction with HLF and HIF-1α were constructed from pCMX/GalDBD and pCMX/VP16 vectors described previously8,10 and schematically outlined in Fig. 1. A description of general molecular biology techniques such as cloning, mutagenesis, and plasmid propagation and isolation can be found elsewhere.11 Briefly, the Gal4DBD chimeras contain the yeast DBD of Gal4 (amino acids 1–147) fused in frame with either full-length Ref-1 (GalDBD/Ref-1), HLF amino acids 1–265 (GalDBD/HLF 1–265), or HIF-1α amino acids 1–245 (GalDBD/HIF-1α 1–245). The VP16 chimeras contain the 78-amino acid activation domain from the herpes simplex virus fused in frame with either full-length Ref-1 (VP16/Ref-1), HLF amino acids 1–265 (VP16/HLF 1–265), or HLF 1–265 cysteine (amino acid residue 25)-to-serine point mutant (VP16/HLF 1–265 C25S). The luciferase reporter construct G5E1b-Luc contains five copies of the Gal4-binding site upstream of the minimal E1b promoter-driven firefly luciferase gene.

Two-Hybrid Assay

Several methods to transfect plasmid DNA into mammalian cells are available, including calcium phosphate, DEAE-dextran, and liposome-mediated transfection. To analyze the in vivo interaction of Ref-1 with HLF we use the monkey kidney COS-1 cell line in conjunction with the liposomal transfection reagent LipofectAMINE 2000 (Life Technologies, Rockville, MD) and the Dual-Luciferase reporter assay system (Promega, Madison, WI). Although we have obtained similar results with other transfection reagents (1,2-dioleoyl-3-trimethylammonium-propane, DOTAP; Boehringer Mannheim, Germany) and cell lines (human embryonic kidney, HEK293T), in our hands COS-1 cells and the LipofectAMINE 2000 reagent routinely give high transfection efficiencies, resulting in consistent two-hybrid results. A description of general mammalian cell tissue culture techniques is beyond the scope of this chapter; readers who require further information on these techniques are therefore referred to methods and protocols given elsewhere.11

For two-hybrid assays COS-1 cells are cultured in a humidified incubator (95% air/5% CO2) at 37° in Dulbecco’s modified Eagle’s medium (DMEM, GIBCO-BRL, Gaithersburg, MD) supplemented with 10% (v/v) fetal calf serum (FCS; GIBCO-BRL). Before transfection, cells are trypsinized and plated onto 24-well tissue culture plates (Falcon; Becton Dickinson Labware, Lincoln Park, NJ) at a density of 5 × 104 viable cells per well in a final volume of 450 μ1 of DMEM–FCS. After 24 hr (50–60% cell confluency), transfections are carried out with LipofectAMINE 2000 reagent as follows.

1. For each transfection (or well of cells), 425 ng of plasmid DNA consisting of 300 ng of G5Elb-Luc reporter, 50 ng of GalDBD or GalDBD chimera, 50 ng of VP16 or VP16 chimera, and 25 ng of Renilla internal control pRL-TK plasmid (Promega) is diluted into DMEM without FCS to a final volume of 25 µl. The pRL-TK control plasmid provides constitutive expression of Renilla luciferase from the herpes simplex virus thymidine kinase (TK) promoter and its addition serves as an internal control of transfection efficiency.

2. For each well of cells to be transfected, dilute 1 µ1 of LipofectAMINE 2000 reagent into 25 µ1 of DMEM without FCS.

3. Combine the 25 µ1 of plasmid-DMEM mix (step 1) with 25 µ1 of diluted LipofectAMINE 2000 reagent (step 2). Incubate the mixture at room temperature for 20 min to allow plasmid allow plasmid DNA–LipofectAMINE 2000 reagent complexes to form. Note: Diluted LipofectAMINE 2000 from step 2 must be combined with plasmid DNA mix within 10 min of preparation.

4. After incubation add the plasmid DNA–LipofectAMINE 2000 reagent complexes (50 µ1) directly to each well. To obtain consistency and high transfection efficiencies it is important to disperse the plasmid DNA–LipofectAMINE 2000 reagent complexes thoroughly and evenly throughout each well. This is best done by gently rocking the plate back and forth three or four times.

5. Return the transfected cells to the incubator and incubate for 24 to 48 hr. We have found that it is not necessary to remove the complexes or change the medium during the assay, as prolonged cell exposure to LipofectAMINE 2000 does not affect the transfection activity.

6. After transfection wash all wells once with 500 µ1 of phosphate-buffered saline solution and harvest with Dual-Luciferase lysis buffer (100 µ1/well; Promega). Lysed extracts (3 µ1) are then analyzed for luciferase activity in a Turner Designs (Sunnyvale, CA) T20/20 luminometer, using the Dual-Luciferase reporter assay system, and measured G5E1b-Luc reporter activity is normalized to the activity of the internal control (pRL-TK).

Results and Discussion

Typical results of a two-hybrid assay showing Ref-1 interaction with HLF are...