Peptide Tags and Domains for Expression and Detection of Mammalian Membrane Proteins at the Cell Surface

Abstract

Normal functions of cell-surface proteins are dependent on their proper trafficking from the site of synthesis to the cell surface. Transport proteins mediating solute transfer across the plasma membrane constitute an important group of cell-surface proteins. There are several diseases resulting from mutations in these proteins that interfere with their transport function or trafficking, depending on the impact of the mutations on protein folding and structure. Recent advances in successful treatment of some of these diseases with small molecules which correct the mutations-induced folding and structural changes underline the need for detailed structural and biophysical characterization of membrane proteins. This requires methods to express and purify these proteins using heterologous expression systems. Here, using the solute carrier (SLC) transporter NaCT (Na⁺-coupled citrate transporter) as an example, we describe experimental strategies for this approach. We chose this example because several mutations in NaCT, distributed throughout the protein, cause a severe neurologic disease known as early infantile epileptic encephalopathy-25 (EIEE-25). NaCT was modified with various peptide tags, including a RGS-His₁₀, a Twin-Strep, the SUMOstar domain, and an enhanced green fluorescent protein (EGFP), each alone or in various combinations. When transiently expressed in HEK293 cells, recombinant NaCT proteins underwent complex glycosylation, compartmentalized with the plasma membrane, and exhibited citrate transport activity similar to the non-tagged protein. Surface NaCT expression was enhanced by the presence of SUMOstar on the N-terminus. The dual-purpose peptide epitopes RGS-His₁₀ and Twin-Strep facilitated detection of NaCT by immunohistochemistry and western blot and may serve useful tags for affinity purification. This approach sets the stage for future analyses of mutant NaCT proteins that may alter protein folding and trafficking. It also demonstrates the capability of a transient mammalian cell expression system to produce human NaCT of sufficient quality and quantity to augment future biophysical and structural studies and drug discovery efforts.

1. Introduction

1.1. Peptide tags and domains

Recent advancements in peptide fusion technologies to maximize recombinant expression of proteins that retain the functional properties of the wild-type protein have enabled the study of a growing number of membrane transport proteins and the impact of disease-causing mutations in these proteins on their structure and function [1]. For comprehensive information on peptide fusion tags and domains, the reader is referred to recent reviews on the topic [1–6].

Here we present a strategy for engineering of select peptide tags and domains that we have successfully used for detection of mammalian ATP binding cassette (ABC) and Solute Carrier (SLC) transporter proteins in eukaryotic expression systems, and that also serve to facilitate purification of the target protein. The epitope tags we describe here include the Arg-Gly-Ser-His10 (RGSH₁₀) and Twin-Strep, and enhanced Green Fluorescence Protein (EGFP) suitable for detection and cellular localization studies, as well as the SUMOstar domain that demonstrates enhanced expression of fully glycosylated mature protein.

Polyhistidine affinity tags (His₆, His₈ or His₁₀) are the most commonly used affinity purification tags probably because of their small size and the simple purification strategy on Immobilized-Metal Affinity Chromatography (IMAC) resin leading to relatively pure proteins (>80%) in a single step. Polyhistidines fused to the N- or C-terminus of a target protein are typically weak epitopes for recognition by antibodies and thus commercially available anti-His tag antibodies often fail in sensitivity. An exception is the Arg-Gly-Ser-His4 epitope that is recognized by a monoclonal anti-RGS-His₄ antibody (Qiagen) with high avidity and selectivity resulting in strong signals with very low background noise in western blots and immunofluorescence analyses.

Twin-Strep tag includes the Strep II tag, an eight-amino acid epitope (Trp-Ser-His-Pro-Gln-Phe-Glu-Lys) two times in series connected by a linker of at least eleven amino acids (https://www.iba-lifesciences.com/strep-tag/). The binding affinity of StrepII tag to Strep-Tactin^® resin (Iba-Lifesciences) is in the nM range and nearly 100 times higher than that of streptavidin whereas the binding affinity of the Twin-Strep tag is in the pM range. The anti-Strep antibody (Iba Lifesciences) does not cross-react with yeast proteins but does weakly recognize several proteins in mammalian whole cell lysates resulting in some background signals.

The Small Ubiquitin-like Modifier [7] also known as the Saccharomyces cerevisiae Smt3 protein, when fused to the N-terminus, exerts a chaperone-like effect, enhancing folding, solubility and stability of target proteins [8–11]. The SUMOstar domain contains the two interfacial mutations, R64T and R71E that render the domain resistance to cleavage by intrinsic eukaryotic proteases [12]. The SUMOstar polypeptide can be removed from its fusion partner with the highly specific SUMOstar protease (Lifesensors, Inc.) that precisely identifies tertiary structure of the domain and removes it to generate the recombinant protein with its native N-terminus [8,9]. Additional tobacco etch virus (TEV) cleavage sites maybe engineered to afford cleavage in the presence of detergent [13]. TEV protease is a ~27 kDa member of the cysteine proteases family possessing highly specific proteolytic activity towards the seven-amino acid recognition sequence (ENLYFQS/G) and is more resilient in detergent solutions than most other proteases.

Enhanced Green Fluorescence Protein (EGFP), developed for expression in mammalian cells, may be included to facilitate detection [14,15,12,11]. Fusion of EGFP is especially useful for both real-time monitoring of expression of the target gene in live cultures and rapid evaluation of subcellular distribution by fluorescence microscopy. EGFP has proven to be invaluable for quantitation of the target gene in crude cell lysates and successive protein purification steps. With the example of NaCT, quantitative in-gel fluorescence methods showed that total cell extract of transiently transfected HEK293 cells contained 4 μg of HSS*-GFP-NaCT per 10⁶ cells and was comparable to the Cystic Fibrosis Conductance Regulator (CFTR) expressed in stable HEK293 cells [11].

We have successfully used these epitope/affinity tags and protein domains, alone or in combination, for detection and localization of membrane proteins in mammalian HEK293 and CHO cells, and in yeast (S. cerevisiae and P. pastoris) [16,11,17]. Indeed, we have demonstrated that tandem affinity chromatography of His_6–10 and TwinTrep-tagged targets yielded highly purified proteins (>95%) that are fully functional and suitable for biochemical and biophysical studies [18–21]. In this chapter, we focus on mammalian expression systems using the popular HEK293FT cell line that afford native posttranslational modifications (including glycosylation) and allow assessment of in vivo transport function in whole cell transport assays.

1.2. Mammalian expression vectors and cloning

Mammalian expression vectors:

Many expression vectors have been developed for transfection of mammalian cells, including pcDNA3.1 (ThermoFisher), pCI (Promega), pCAG (FUJIFILM Wako Pure Chemical Corporation), pTT5 (ThermoFisher) and pCMV (Clontech). We focus here on the pcDNA^™3.1 vector designed for nonviral transient and stable gene expression in a variety of mammalian cell lines. It contains a cytomegalovirus (CMV) promoter that drives high level constitutive gene expression. A bacterial ampicillin resistance gene and a pUC origin are for DNA propagation and maintenance in E. coli. Three untagged versions of the pcDNA^™3.1 were originally created by Invitrogen/ThermoFisher with different selectable markers (Geneticin^®, Zeocin^™, or Hygromycin) for use alone or in co-transfections; most often the neomycin resistance gene is used for selection of stable cell lines with neomycin (Geneticin). The vector features a large multiple cloning site in either forward (+) or reverse (−) orientations. An SV40 enhancer origin serves for episomal replication in cell lines expressing the large T antigen (e.g., HEK293T cells).

Cloning:

A scheme of different tag variants that were engineered at the N-or C-terminus of the NaCT gene is displayed in Fig. 1. RGS-His₁₀, TwinStrep and GFP were placed either alone or in combination at the N-or C-terminus while SUMOstar is effective only at the N-terminus. The nucleotide and protein sequences of our most successful HSS*-NaCT construct and the enhanced GFP insert are given in Fig. 2. For recombinant protein production in mammalian cells, it is advisable to insert a Kozak translation initiation sequence around the ATG initiation codon for proper initiation of translation, which has been shown to significantly increase protein expression levels [22–24]. An example of a Kozak consensus sequence is illustrated in Fig. 2a (highlighted in red) with a G or A at position –3 and a G at position +4 (coding for a glutamate) giving strong consensus. The target gene must also contain a stop codon for proper termination of your gene. Please note that the NcoI restriction site contains an internal start codon (CCATGG) and a G at position +4. Cloning strategies to fuse epitope tags to a target gene have been described in detail elsewhere [25]. Nowadays, many companies offer de-novo gene synthesis and deliver the gene in the desired expression vector ready for transfection for a small cost. For epitope tags and domains larger than 100 bp, we recommend DNA strings and In-Fusion cloning strategies as described in Method 3.1.

Fig. 1.

Schematic representation of variants introducing either N- or C-terminal peptide tags Arg-Gly-Ser-His10 (RGS-His₁₀) and twin Strep II (TwinStrep), and the domains SUMOstar and enhanced Green Fluorescence Protein (GFP) that are cleavable by Tobacco Etch Virus (TEV) protease. Restriction enzyme sites compatible with cloning into the pcDNA3.1(−) vector are 5’-XbaI and 3’-EcoRI. Restriction enzyme sites flanking the gene (5’-SpeI and 3’-XhoI) were included for convenient exchange of target genes. Plasmid constructs harboring the NaCT cDNA (named pcDNA3.1(−)-SLC13A5) are available upon request.

Fig. 2.

Example of the nucleotide sequence of the (a) HSS* variant and (b) the GFP domain. Color coding is as in Fig. 1. The tags were designed with a 15 bp N-terminal extensions homologous to the pcDNA3.1 vector ends when cleaved with 5’-XbaI and 3’- EcoRI restriction enzymes (dashed line). The Kozak translation initiation sequence around the ATG initiation is highlighted in red. Sequences encoding RGS-His₁₀ (pink), TwinStrep, SUMO* and GFP are separated by flexible Gly-Ser-Ala hinge peptides (teal) to improve accessibility of the tags; in addition, a thrombin protease cleavage site (dark grey) was introduced between RGS-His₁₀ and Twin Strep tags. The TwinStrep tag includes two times in series the StrepII epitope (cyan) connected by an 11 amino acids linker (teal). The SUMOstar domain (yellow) contains the two interfacial mutations R64T and R71E highlighted in dark blue [12]. Primer sequences with 15 bp extensions homologous to the TEV cleavage site (light grey) of the HSS*-tag and to the 3’- vector sequences for amplification of the target cDNA (NaCT in this case) are indicated by an arrow above the sequence (forward primer) and below the sequence (reverse primer). The primers are designed to anneal to the target cDNA with a melting temperature (Tm) of at least 55 °C. (b) If desired, sequences encoding the enhanced GFP protein developed for expression in mammalian cells [12] can be introduced in the HSS*-tag by In-Fusion cloning. The GFP sequence contains the fluorescence enhancing mutations V64L/S65T (red box), and the A206K mutation (red box) that minimizes GFP dimerization [15]. A design of GFP flanked by six amino acid Gly-Ser-Ala hinge peptides (teal) and including a 15 bp overlap homologous to the SUMO* domain (5’, yellow), as well as a 15 bp overlap homologous to the TEV sequence (3’, light grey) is shown. The entire ORF sequence of HSS*–NaCT and HSS*GFP-NaCT were deposited in GenBank (accession numbers MZ367587, MZ367588).

1.3. Transient transfection

Transfection is a powerful tool to introduce foreign genetic material (DNA or RNA) into cells to study gene and protein functions. Transfection methods are broadly classified as biological (virus-mediated), physical (electroporation [26]), or chemical [27]. Chemical transfections are most popular because of ease and reproducibility. They employ cationic reagents that form complexes with negatively charged nucleic acids to cross the cell membrane. Some examples are calcium phosphate, polyethylenimine (PEI), Fugene, and liposomal reagents such as Lipofectamine 3000 (Fig. 3). Lipofectamine forms liposomal vesicles after mixing with the pcDNA3.1 plasmid DNA effectively encapsulating the DNA that is then taken up into the cell by endocytosis (Fig. 3) [28,29]. Transiently transfected DNA reaches the nucleus but is not integrated into the genome. If cell lines stably expressing the protein are desired, the plasmid DNA should be linearized to favor homologous recombination into the genome; individual, highly expressing clones can then be selected as described in [30]. Chemical transfection methods are influenced by multiple factors that can be optimized including cell type, confluency of cells, pH, temperature and plasmid/reagent ratio [27]. In our experience, mixing about 0.5 μg plasmid DNA with 0.75 μl Lipofectamine 3000 per 1 ml culture volume in 24-well plates yielded the highest protein expression, but needs to be optimized for each gene empirically (see Note 7).

Fig. 3.

Transient DNA transfection with liposomal vehicles. DNA plasmid is mixed with cationic lipids that coat the negatively charged nucleic acids. The DNA-liposome complex particle, now positively charged on the surface, fuses with negatively charged plasma membranes allowing the complex to overcome the electrostatic repulsion of the cell membrane [29]. The DNA-liposome particle is engulfed by the plasma membrane, and upon endocytosis the DNA is released from the endosome and reaches the nucleus for transcription. After transcription, the mRNA is transported back to the cytoplasm, and is translated on the ribosomes into active proteins that shuttle to the cellular destinations, i.e., the plasma membrane in the case of the NaCT transporter protein.

1.4. SLC family

The solute carrier (SLC) family comprises one of the largest groups of membrane protein transporters found in archaea, bacteria, and eukaryotes [31] with more than 400 members that are divided into around 65 families based on sequence homology and function [32]. They are passive transporters, ion transporters, and exchangers that transport different solutes (e.g., amino acids, inorganic ions, lipids, neurotransmitters, drugs) across the membranes [33,34,31]. Many diseases are associated with the dysfunction of SLC transporters, making them an important target for drug development. The SLC13 family or divalent anion/Na⁺ symporter (DASS) protein family co-transports Na⁺ with either dicarboxylates, tricarboxylates, inorganic sulfates or phosphates. One of the DASS family members is the Na⁺-coupled citrate transporter (NaCT), which is expressed primarily in the liver, testis, brain, bone, and teeth. In the liver, its function is related to the generation of energy, synthesis of cholesterol and fatty acids, and reciprocal regulation of glycolysis and gluconeogenesis. In the testis, it helps with the nutritional needs of the sperm as seminal fluid contains millimolar concentrations of citrate [35]. In the brain, it is mostly expressed in neurons where citrate serves as an energy source and as a precursor for synthesis of acetylcholine, GABA, and glutamate [36]. Bone and teeth contain most of the citrate in the body where it serves as a chelator of calcium and hence in the mineralization of these tissues [37].

A number of SLC transporters have been expressed with various N- and C-terminal tags to facilitate detection and purification for structure-function analyses; these tags include His_8–10 [38–41], FLAG [38,39], Strep-tag [42–44], and V5 [45]. However, systematic analyses of tag placements are still rare. Thus, lessons learned here from placement of tags in the example of NaCT might be applicable to other family members. In the case of NaCT so far, the lack of highly selective antibodies capable of recognizing the human NaCT protein in immunofluorescence and/or western blot analyses has impaired progress.

To explore the utility and performance of various in-frame domain fusions of recombinant NaCT, the HEK293FT cell line was transiently transfected with the various constructs shown in Fig. 1, and NaCT expression was analyzed by western blot (Fig. 4). The NaCT constructs can be detected by anti RGS-His antibody (Fig. 4a, Fig. 5) as well as anti-Strep, anti-SUMOstar and anti-GFP antibodies, and by GFP fluorescence (Fig. 4b). The highest expression levels were observed with the HS*- and HSS*-NaCT variants that carry the SUMOstar domain at the N-terminus of NaCT. Expression was very low in the variants HS-NaCT (far left), and NaCT-HS and NaCT-GFP-S that are lacking the SUMOstar domain. NaCT resolved as a double band that collapsed into a single band of higher electrophoretic mobility upon treatment with endoglycosidase PNGase F, indicating expression of both core-glycosylated (empty arrowhead) and complex-glycosylated (filled arrowhead) protein. Cells expressing HS*-NaCT and HSS*-NaCT were analyzed using immunofluorescence confocal microscopy to compare NaCT expression levels and surface compartmentalization. The recombinant proteins exhibited a strong pattern particularly evident along the cell surface/plasma membrane.

Fig. 4.

Detection of NaCT variants with different tags in transiently transfected HEK293FT cells. (a) Protein expression levels of NaCT variant were revealed using the monoclonal anti RGS-His antibody and enhanced chemiluminescence. The highest expression levels were observed with the HS*- and HSS*-NaCT variants, while expression was very low in variants lacking the SUMOstar domain (e.g. HS-NaCT, far left, and NACT-HS, NaCT-GFP-S, middle). Complex glycosylated NaCT migrates as multiple fuzzy bands with an apparent molecular weight of ~85 kDa, and collapses into a sharp protein band with a calculated molecular mass of 75 kDa after treatment with PNGase that cleaves N-linked oligosaccharides. Core-glycosylated (empty arrowhead) and complex-glycosylated (filled arrowhead) (b) In-gel fluorescence of the same gel; the GFP domain adds approximately 27 kDa to the NaCT protein and may affect expression efficiency. (c) Ponceau-S total protein staining of the nitrocellulose membrane used for the Western blot in (a) served as assessment for protein loading.

Fig. 5.

Immunocytochemistry revealed plasma membrane and ER distribution of HS*-NaCT and HSS*-NaCT expressed in HEK293FT cells. HEK293FT cells, transiently transfected with HS*-NaCT, HSS*-NaCT and the “empty” pcDNA3.1 vector, were grown on glass coverslips. Cells were fixed, permeabilized, and NaCT visualized with the monoclonal anti RGS-His antibody and secondary Alexa fluor-568 antibody (shown in red). Nuclei were stained with DAPI (blue). HS*-NaCT and HSS*-NaCT variants showed high protein expression levels with strong plasma membrane surface staining, and some intracellular staining likely resembling ER localization.

In conclusion, the presence of the SUMOstar domain proved advantageous for NaCT synthesis in our system since the comparative expression analysis indicated a 3 to 5-fold increase in surface expression levels. If necessary, for downstream applications, the SUMO* tag is removable with the highly specific SUMOstar and TEV proteases [8].

1.5. Transport assay

For small, charged transport substrates, the preferred means to validate function is still the transport assay using radioactive substrates by monitoring their uptake into whole cells or membrane vesicles. Alternatively, electrophysiology [46] can be used to assess quantitatively fast reaction-transport processes in live cells but requires specialized equipment and is still applicable only for those transporters that are electrogenic [47]. Membrane transporters are important for the passage of biological molecules across the lipid bilayer. Functional fluorescence imaging is used to quantify fast reaction-transport processes in solution and in live cells using ligand-specific sensor such as molecular fluorescent indicators with substrate specificity [48,49]. The most widely used method to measure transport activity utilizes radiolabeled substrates due to its high sensitivity. Typically, cells expressing the target protein are incubated with a suitable radiolabeled substrate under controlled conditions (pH, ligands, and temperature,) and the radioactivity associated with the cells is measured after an indicated time period. This involves washing of the cells to remove any radiolabeled substrate that has not entered the cells. The time of incubation with the substrate should also be standardized for initial uptake rates and for ensuring lack of significant metabolism of the transported substrate inside the cells. Working with radioactive materials requires an institutional license and specialized training and equipment (protective shields, scintillation counter), and must be planned out carefully. Fig. 6 shows Na⁺-coupled [¹⁴C]-citrate uptake into HEK293FT cells transiently transfected to express the NaCT variants described in Fig. 1.

Fig. 6.

Na⁺-coupled uptake of citrate into HEK293FT cells expressing NaCT variants. Cells expressing NaCT are incubated for 30 min with [¹⁴C]-citrate in Na⁺-free 140 mM NMDG-Cl buffer (white), in 140 mM NaCl buffer (grey) or 140 mM NaCl buffer in the presence of 10 mM LiCl (black). NMDG-Cl buffer displayed a transport activity similar to the empty vector in all NaCT variants. Addition of 10 mM LiCl in the presence of Na⁺ significantly increased citrate uptake suggesting that Li⁺ stimulates the transport activity of human NaCT. The highest transport activity was detected in the HS*-NaCT and HSS*-NaCT variants with uptake rates and characteristics very similar to the non-tagged wild-type NaCT (far left). The higher transport rates coincide with higher expression levels and surface expression seen with the HS*-NaCT and HSS*-NaCT variants containing the N-terminal SUMO* domain. Variants containing a GFP domain at the N-terminus show somewhat (~2-fold) decreased transport activity, but kinetic characteristics are retained. Each column represents the mean ± SEM, with *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001 and ****P ≤ 0.0001. Data was analyzed by GraphPad Prism (GraphPad, San Diego, CA).

Experimental conditions can be designed to confirm that the observed uptake activity is solely due to the expressed exogenous transporter. In the case of NaCT, we showed that the observed citrate uptake is significantly enhanced in the presence of Na⁺ based on the fact that it is a Na⁺-coupled transporter. We also studied the effect of Li⁺ on the Na⁺-dependent citrate uptake as an additional feature to confirm the involvement of NaCT in the observed citrate uptake. It has been shown that the function of NaCT involves co-transport of 4 Na⁺ ions along with one trivalent citrate anion [50]. Li⁺ exhibits a much higher affinity than Na⁺ to some but not all of these four Na⁺-binding sites [51,52]. As such, NaCT is not functional when Na⁺ is replaced in its entirety with Li⁺, but the transport function is enhanced by Li⁺ when studied in the presence of Na⁺. These experimental manipulations enable us to assign the observed transport function solely to the heterologously expressed NaCT. A similar strategy needs to be devised with appropriate experimental conditions suitable for the given transporter. Variants containing SUMOstar increased transport activity (Fig. 6), which correlates with increased expression levels (Fig. 4 and 5) and surface expression (Fig. 5), seen in HS*-NaCT and HSS*-NaCT variants. While variants containing GFP seem to decrease transport activity, expression, and surface levels. However, kinetic characteristics are retained between variants and non-tagged wild type NaCT.

2. Materials

It is essential that you consult the appropriate Material Safety Data Sheets and your institution’s Environmental Health and Safety Office for proper handling of equipment and hazardous materials used in this protocol.

2.1. Equipment

1. Thermal cycler programmed with desired amplification protocol

2. Microtube centrifuge 0 to 13000 rpm

3. Agarose electrophoresis equipment

4. Sterile laminar flow hood

5. 5% CO₂ Incubator at 37 °C.

6. Shaking incubator

7. Multipurpose scintillation counter

8. Microscope with A1 confocal and STORM super resolution software

2.2. Cloning of the target gene

1. Forward and reverse primers

2. pcDNA3.1(−) vector

3. DNA with desired N-terminal and C-terminal epitope and domain tags, and combinations thereof as shown in Figs. 1 and 2.

4. Phusion II DNA polymerase kit (New England Biolabs)

5. Deoxyribonucleotide (dNTP) mix: 10 mM stock solution

6. 1% Agarose gel

7. In-Fusion HD Cloning Kit for seamless DNA cloning (Takara Bio)

8. XL10-Gold Ultracompetent E. coli cells (Agilent Technologies)

9. Plasmid DNA purification miniprep kit

10. Plasmid Plus DNA purification Midiprep Kit

2.3. Tissue culture consumables

1. 0.22 μm filters

2. Sterile 24-well plates

3. Collage-coating solution: 714 μl of collagen Type I, 47 μl acetic acid, diluted in 49 ml of filtered sterilized water

2.4. Transfection and adherent cell culture

1. HEK293FT cells (RRID: CVCL_6911, ThermoFisher Scientific), a fast-growing, highly transfectable clonal isolate derived from human embryonal kidney cells transformed with the SV40 large T antigen

2. DMEM media: High glucose Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10 % fetal bovine serum (FBS) and 1% penicillin/streptomycin

3. Dulbecco’s Phosphate Buffered Saline (PBS)

4. Opti-MEM 1X, Reduced Serum Medium (Gibco)

5. Lipofectamine 3000 Reagent (ThermoFisher Scientific)

2.5. Expression of target protein

1. Radioimmunoprecipitation assay (RIPA) buffer: 25 mM Tris HCl, pH 7.4, 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS).

2. 10 mg/ml Leupeptin in water

3. 2.5 mg/ml E-64 in water

4. 100 mM 4-(2-Aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF) in water

5. 250 mM Benzamidine in water

6. 10 mg/ml Pepstatin A in dry DMSO

7. 2.5 mg/ml Chymostatin in dry DMSO

8. 0.1–1 M Phenylmethylsulfonyl fluoride (PMSF) in dry DMSO, ethanol, or isopropanol (see Note 8).

9. Ethylenediaminetetraacetic acid (EDTA)

10. Anti RGS-His antibody (Qiagen), Strep-tag antibody (IBA Lifesciences), goat anti-mouse IgG (heavy and light chain) antibody conjugated to horseradish peroxidase (HRP, Qiagen) for use as secondary antibody

11. The SuperSignal Pico Enhanced Chemoluminescence kit (Pierce/ ThermoFisher)

12. Pierce^™ Bicinchoninic Acid Kit for Protein Determination

2.6. Cellular localization of target protein

1. Fixing solution: 4% paraformaldehyde (PAF) diluted in PBS

2. Goat anti-Mouse IgG Alexa Fluor 568

3. Prolong Diamond Antifade Mountant with DAPI (Invitrogen/ThermoFisher)

2.7. Transport assays with whole cells

1. NMDG-Cl buffer: 25 mM Hepes/Tris, pH 7.4, 140 mM N-methyl-D-glucamine (NMDG) chloride, 5.4 mM KCl, 1.8 mM CaCl₂, 0.8 mM MgSO₄, and 5 mM glucose

2. NaCl buffer: 25 mM Hepes/Tris, pH 7.4, 140 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl₂, 0.8 mM MgSO₄, and 5 mM glucose

3. 10 mM LiCl in NaCl buffer

4. [¹⁴C]- Citrate with a specific radioactivity of ~113 mCi/mmol (Moravek Biochemicals)

5. Lysis buffer: 0.5% SDS in 0.2 N NaOH

6. Scintillation cocktail

3. Methods

3.1. Construction of an expression plasmid containing the recombinant gene with different epitope tags and domains

1. Design the desired N-terminal and C-terminal epitope and domain tags, and combinations thereof; include GS-linkers between individual tags as shown in Figs. 1 and 2. For tags >50 bp, we recommend ordering synthetic “DNA strings” for easy cloning; it is important to include at least 15 bp extensions homologous to the flanking vector and target gene for In-Fusion cloning.

2. Design gene-specific primers with at least 15 bp extensions homologous to the tags or vector ends. PCR amplify your target gene using a high-fidelity polymerase such as Phusion II and isolate the correct size DNA fragment from an agarose gel (see for example [53]).

3. Digest the pcDNA3.1 plasmid vector with appropriate restriction enzymes in the multiple cloning site (for example 5’- XbaI and 3’-EcoRI), or PCR amplify the vector. Isolate the correct size DNA fragment from an agarose gel.

4. Mix equal amounts (50 – 100 μg each) of the DNA string containing the epitope/domain tags and the gel-purified DNA fragments of the amplified gene and of the vector DNA and add the In-Fusion enzyme premix according to manufacturer’s recommendations. Incubate for 15 min at 50 °C.

5. Let the mixture cool and transform 5 μl into 50 μl of XL10-Gold ultracompetent E. coli cells. Plate aliquots of transformed cells on LB-agar containing 100 μg/ml ampicillin and incubate overnight at 37 °C. Pick single colonies and purify plasmid DNA minipreps. Screen plasmids by restriction enzyme analysis and DNA sequencing to confirm the presence of tags and the integrity of the gene and entire open reading frame (see Note 1). Be sure to make glycerol stocks of the positive E. coli colonies for safekeeping.

6. For transfections it is advisable to purify larger batches (50–200 μg) of selected plasmids that are free of endotoxin using DNA midiprep kits (see Note 2 and 3) of DNA.

3.2. Transient transfection of target protein in HEK293FT cells

Transient transfection procedure closely follows protocols described in [54]. HEK293FT cells (see Note 4) are maintained (see Note 5) in DMEM media in T75 flasks and split once or twice a week at around 70–80% confluency. Transient transfections for expression analysis, confocal microscopy and transport assays are conveniently done in 24-well cell culture plates, but transfections can be scaled to any well/plate size (please consult manufacturer’s recommendations).

1. Coat 24-well cell culture plates with 300 μl of collagen-coating solution at room temperature for 20 min. Wash twice with PBS to remove residual acetic acid.

2. Seed HEK293FT cells (see Note 6) at 2 × 10⁵ cells per well in DMEM media.

3. Next day, visualize cells under the microscope; cell confluency should be around 70–90 %.

4. Replace media with 1 ml of DMEM media before transfection.

5. Mix 0.5 μg of DNA per well with 25 ul of Opti-MEM in a 1.5 ml Eppendorf tube and add 1 μl of P3000 Reagent (DNA solution), as recommended in the Lipofectamine 3000 Reagent protocol.

6. Dilute 0.75 μl Lipofectamine 3000 reagent (see Note 7) into 25 μl Opti-MEM.

7. Mix DNA solution with diluted Lipofectamine 3000 Reagent at a 1:1 ratio.

8. Incubate for 20 min.

9. Add DNA-lipofectamine mix to HEK293FT cells dropwise mix by gently swirling the plate.

10. Incubate at 37 °C in a 5% CO₂ incubator.

11. After 24h replace media with 1 ml of DMEM media.

12. Estimated transfection efficiency is >70% even for difficult to transfect cells.

3.3. Cell harvest and detection of the recombinant protein

1. After 48 hours post-transfection, wash the cells twice with 0.5 ml of ice-cold PBS on ice.

2. Incubate cells with 100 μl/well of ice-cold RIPA buffer with protease inhibitors (see Note 8) for 30 min on ice with slow shaking.

3. Transfer the cell lysate to a 1.5 ml microfuge tube, and centrifuged at 14,000 × g for 20 min at 4 °C.

4. Transfer the supernatant a new microfuge tube and discard the pellet.

5. Assay ~10 μl for protein content using a BCA Protein Assay (see Pierce BCA Protein Assay Kit protocol) (see Note 9). The total protein yield ranges between 50 and 200 μg per well (see Note 10).

6. Mix 10 μg of the cell lysate (~10 μl) with 10–20 μl of 5x SDS gel-loading buffer (see Note 10). Resolve on SDS-PAGE gels (see Note 11). On each SDS-PAGE gel, load molecular weight markers, a negative control (cells transfected with the “empty” vector) to assess the cross-reactivity of the antibody with other proteins present in the cell lysate, and a positive control for comparison of expression levels. A control sample of target protein expressed from another system, if available, is also desirable for comparison of molecular weight, posttranslational modification, and expression level.

7. To check for glycosylation of the target protein, tread 10 μg of cell lysate (~10 μl) with 0.5 μl (500units/μl) of endoglycosidase PNGase F for 60 min at 37°C, before mixing with 5x SDS gel-loading buffer, and resolving on SDS-PAGE gels.

8. After SDS-PAGE, transfer the polyacrylamide gel into water and image for GFP fluorescence (see example in Fig. 4B).

9. Immediately thereafter, transfer the proteins to a nitrocellulose membrane by electroblotting, and stain the membrane with the reversible protein stain Ponceau-S (see example in Fig. 4C).

10. Destain the nitrocellulose membrane in water and block the membrane with 1% milk in buffers compatible with immunoblotting. In Fig. 4A, the membrane was analyzed by western blotting using the anti RGS-His antibody with the enhanced chemo luminescence (Pico-ECL) following manufactures recommendations. For detailed protocols on protein assay, SDS-PAGE, and western blotting, see [55–59].

3.4. Protein detection and localization using confocal microscopy

1. For confocal microscopy, grow HEK293FT cells on polylysine-treated coverslips in 24-well plates (or larger depending on the size of the coverslip), and transfect with Lipofectamine 3000 as described above in 3.2.

2. After 48 hours post transfection, wash the cells with ice-cold PBS.

3. Fix the cells with fixing solution for 15 min at room temperature and wash three times with PBS.

4. Permeabilize the cells with 0.1% Triton X-100 for 5 min and wash twice with PBS.

5. Block the cells with ice-cold 1% BSA in PBS for 20 min.

6. Next, incubate the cells are with anti-RGSH antibody (1:500 dilution in 1% BSA in PBS) for 30 min at 4 °C. Then, wash the cells three times with ice-cold PBS.

7. Incubate with Alexa 568 anti-mouse secondary antibody (1:500 dilution in 1% BSA in PBS) for 30 min and wash twice with ice-cold PBS.

8. Mount the coverslips using Prolong mounting solution and cure for 6 hours or overnight at room temperature.

9. Next day, visualize the cells using a Confocal microscope as shown in Fig. 5.

3.5. Transport assays with whole cells expressing different constructs

1. After 48 hours post transfection, the culture medium in the individual wells of the 24-well plate is carefully removed by aspiration.

2. Wash the cells twice with NMDG-Cl or NaCl buffer.

3. Incubate with either 250 μl of NMDG-Cl -buffer, or NaCl buffer, or 10 mM LiCl in NaCl buffer each containing [¹⁴C]-citrate (0.1 μCi; 4 μM citrate) for 30 min at 37 °C. Each condition is assayed in triplicates.

4. At the end of incubation, remove the buffer from each well, and wash the cells with ice-cold NMDG-Cl buffer, or NaCl buffer, or NaCl buffer, respectively.

5. Lyse the cells in 500 μl lysis buffer by gently shaking the plate for 1 hour at room temperature.

6. Transfer the lysed cells to a scintillation vial containing 4 ml of scintillation cocktail and count the ¹⁴C radioactivity in a liquid scintillation counter. Compare the counts per min to a reference standard to determine the pmol citrate transported in each well.

7. Determine total protein content in cell lysates of parallel wells (not treated with radioactive citrate) using the BCA Protein Assay Kit. Normalize uptake values determined above to total protein per well in each of the sample (see Fig. 6).

4. Notes:

1. The pcDNA3.1 plasmids containing the NaCT (SLC13A5 gene) with the various tags/domains are available from our lab upon request.

2. Plasmid DNA transfected into eukaryotic cells must be free of endotoxins that may reduce cell viability and transfection efficiencies [60]. Endotoxins can have complex effects on mammalian systems, such as nonspecific activation of the immune system, induction of toxic-shock syndrome and stimulation of cytokine overproduction, which may all influence reproducibility and outcome of a transfection experiment.

3. Plasmid DNA for transfection into eukaryotic cells must also be free of chemical and microbial contaminants, as these are toxic to cells, and interfere with the lipid complexing of the transfection reagents and thereby decreasing transfection efficiency.

4. HEK293 cells are the most popular mammalian transient production since HEK293 cells are amenable to a variety of transfection methods. However, they have poor growth and protein production efficiency. On the other hand, CHO cells have robust growth, hardiness in suspension culture, stellar protein production and secretion, but they are hard to transfect. Therefore, CHO cells are primarily used for generation of stable cells.

5. HEK293FT should be maintained using proper mammalian tissue culture techniques to keep cells healthy at high viability for good results. All tissue culture work must be performed inside a sterile laminar flow hood. Solutions should be sterile if in contact with cells. All media must be warmed to 37 °C prior to use. All cells are grown by incubating them in an incubator set at 37 °C and 5 % CO₂.

6. HEK293FT cells are used between passages 3 to 10 since their viability and growth potential decrease. Passage cells when confluent; do not let them to overgrow.

7. Before conducting the transfection analysis, the transfection conditions should be evaluated first by using different ratios of DNA to lipofectamine. For a 24-well plate, 0.5 μg of DNA was used with different concentrations of lipofectamine, with ratios from 1:1 to 1:5 (DNA:lipofectamine). Optimization of the DNA:lipofectamine ratio is particularly important if the culture medium contains serum. Next, different amounts of DNA (0.5, 1, 2 and 3 μg) were tested with the optimum lipofectamine ratio. Refer to manufacturer’s recommendations for further details.

8. PMSF is a widely used protease inhibitor that is inexpensive and very effective against serine and cysteine proteases. PMSF stocks are prepared at 0.1–1 M concentrations in dry DMSO, ethanol, or isopropanol; its effective range is 0.1–1 mM. However, PMSF has a relatively short half-life in aqueous solutions (~35 min at pH 8) and needs to be replenished frequently. It is often substituted or supplemented with the less toxic 4-(2-aminoethyl) benzenesulfonyl fluoride (AEBSF, also called Pefabloc; Roche), which is more stable at physiological pH values and is effective against a broad range of serine proteases. The typical concentration range used for AEBSF is 0.1–1 mM. In addition to PMSF or AEBSF, protease inhibitors commonly used during lysis of E. coli, insect cells or mammalian cells may include metalloprotease inhibitors (EDTA), additional serine and cysteine protease inhibitors (leupeptin, benzamidine, and E-64), and, if the pH becomes acidic during lysis, the aspartic protease inhibitor pepstatin A [25].

   Inhibitor stocks are prepared as follows: 10 mg/ml leupeptin in water, 2.5 mg/ml E-64 in water, and 10 mg/ml pepstatin A in dry DMSO. Stocks are used at a 1000-fold dilution. Benzamidine is pared at 250 mM in water and used at a 100-fold dilution. All stocks should be frozen in aliquots to avoid repeated freeze-thaw cycles.

9. BCA assay is used because Bradford reagent is incompatible with detergents found in RIPA buffer. The detergent tends to bind strongly to the protein, inhibiting the protein binding sites for the dye reagent.

10. The yield is ~50–200 μg of total protein per well in a 24-well plate depending on the confluency of the cells. The anti RGS-His antibody gives a strong signal and 10 ug total protein or less loaded per lane is often sufficient for detection on western blots using ECL.

11. Membrane proteins cannot be boiled in SDS gel-loading buffer. Incubate the samples in SDS gel-loading buffer for 5 – 10 min at room temperature (or 37 °C) before loading on SDS-gels.