Single-molecule labeling for studying trafficking of renal transporters

Abstract

The ability to detect and track single molecules presents the advantage of visualizing the complex behavior of transmembrane proteins with a time and space resolution that would otherwise be lost with traditional labeling and biochemical techniques. Development of new imaging probes has provided a robust method to study their trafficking and surface dynamics. This mini-review focuses on the current technology available for single-molecule labeling of transmembrane proteins, their advantages, and limitations. We also discuss the application of these techniques to the study of renal transporter trafficking in light of recent research.

INTRODUCTION

Transmembrane transport proteins play a critical role in translocation of various physiological substances, chemicals, and drugs into and out of cells via active and passive mechanisms. The establishment and maintenance of this polarized expression of these transport proteins are affected by diverse protein-trafficking complexes. In the last decade, single-molecule labeling of membrane proteins combined with high-resolution microscopy has contributed enormously to our understanding of their function, localization, behavior, and trafficking in polarized epithelium. They have proven to be very useful in studying how protein-protein interactions (6, 26, 53, 77), protein motifs/domains (90), spatial arrangement, mobility and organization (14, 16), and signaling pathways (7, 18) regulate the trafficking of membrane proteins. Surprisingly, application of these techniques to study trafficking of renal transporters has not been widespread. Characterizing the single-molecule behavior of renal transporters and their regulation will provide new information on the mechanisms of ion and water transport and help identify novel pathways, potentially subject to therapeutic intervention. This mini-review describes the current methods available for single-molecule labeling of transmembrane proteins along with their advantages and limitations.

FLUORESCENT PROTEINS

The development of green fluorescent proteins and other red/yellow/enhanced green fluorescent proteins (FPs), along with pH-sensitive versions of the tags (3), provided the initial breakthrough in the ability to tag transmembrane proteins, such as the renal sodium-bicarbonate cotransporter, renal sodium-phosphate cotransporter-2a), renal glucose transporter-4 (or GLUT4), among others (2, 6, 18, 26, 36, 46, 50, 90, 91, 95, 96) (Fig. 1A). Subsequent modifications of FPs coding sequence improved their spectral and biochemical properties. Fluorophore properties like extinction coefficient, quantum yield, photostability and pH sensitivity (72) need to be considered for different imaging applications like fluorescence resonance energy transfer (24), confocal microscopy (79), total internal reflection fluorescence microscopy (51), or super-resolution microscopy (62, 78). However, there are major disadvantages of using FPs to tag membrane proteins, such as interference with endogenous activity of the tagged protein due to overexpression (1), oligomerization caused by FPs (68), and the large size of the tag (~30 kDa) (100). These difficulties have been circumvented to some degree by strategies that reduce overexpression (4), optimization of the linker’s design and length, and the choice of site for insertion of the FP sequence (5, 48, 100). Despite the enormous initial contribution of standard FPs in making super-resolution microscopy possible, their property to undergo irreversible photobleaching, low photostability, and quantum yield limits their use for long-time scale experiments. However, newly engineered fluorescent proteins, such as phytochrome-based near-infrared fluorescent proteins, have much higher signal-to-noise ratio and photostability, thus allowing imaging deeper into tissues, which is, thus, relevant for live-animal or excised-tissue models (33).

Fig. 1.

Diagrammatic representation of single-molecule labeling methods discussed in the review. The rectangle in gray represents a transmembrane protein, double lines represent phospholipid bilayer, and a yellow line represents a fluorescent entity. ACP, acyl carrier protein; BAD, biotin acceptor domain; BC2, short linear epitope corresponding to 16–27 residues of β-catenin; eGFP, enhanced green fluorescent protein; FAP, fluorogen-activating peptide; GFP, green fluorescent protein; GST-O1, glutathione-s-transferase omega1; HA, hemagglutinin; LAP, lipoic acid acceptor peptide; PCP, peptidyl carrier protein; RFP, red fluorescent protein; SBP, streptavidin-binding peptide; TC, tetracysteine; UAA, unnatural amino acid; YFP, yellow fluorescent protein; α-BTX, α-bungarotoxin. Other tag names used in the examples are commercial trademark names such as Halo Tag, CLIP tag, SNAP Tag, and FLAG tag.

CHEMICAL TAGS

An alternative to FPs include Halo Tag (87), SNAP Tag (80, 89), CLIP Tag (101), and glutathione-S-transferase-omega-1 tag (GST-O1) (59, 88) (Fig. 1B). These tags are genetically incorporated into the membrane protein and have been used to study trafficking of the platelet-derived growth factor receptor (87), glucagon-like peptide-1 receptor, and membrane trafficking in kidney epithelial cell lines (80, 89), β₂-adrenergic receptor and orexin receptor-1 (101), cannabinoid receptor, and epidermal growth factor receptor (59, 101) in other cell models. The tags are then subsequently modified to allow binding by covalent or high-affinity bonds to an organic fluorophore with high photostability and brightness (10, 20). These tags have been used to simultaneously image two-membrane proteins (30, 101). More recently, releasable SNAP-tag probes were developed to add versatility in the study of endocytosis and recycling (19). These chemical approaches usually require multiple washing steps to remove the unreacted fluorophore to minimize background signal. To circumvent this issue, a fluorogenic SNAP probe that only fluoresces upon binding the tag has been developed (75, 92). Additional limitations remain, such as overexpression, which may be decreased by the use of low-expression promoters (69) and large-molecular-weight of the tag (~20–30 kDa). This latter issue has been improved by development of smaller tags (~8 kDa), such as the acyl carrier protein tag (15, 50) and the peptidyl carrier protein tag (108), among others (109) (Fig. 1C). These have been used to study trafficking of transferrin receptor (108) and odorant receptors in a kidney cell line where their roles are beginning to be appreciated (50). However, these smaller tags still require the addition or expression of an exogenous enzyme to covalently attach the fluorophore.

ACCEPTOR SEQUENCE TAGS

Significant reduction in the size of the tag has been achieved by adding short peptide acceptor sequences (~15 amino acids) to extracellular domains of transmembrane proteins. These sequences can be a biotin acceptor domain (BAD) (31, 51, 60), lipoic acid acceptor peptide (LAP) (31) (Fig. 1H), the coiled-coil tag (105–107), streptavidin-binding peptide tag (SBP), or the α-bungarotoxin (α-BTX) binding tag, which uses high-affinity binding to α-BTX-conjugated fluorophores (12, 55, 66, 84, 104) (Fig. 1D). We used BAD as a tag to study protein trafficking and surface behavior of the renal Na-K-2Cl cotransporter NKCC2 (51), while the other tags have been used for β₂-adrenergic receptor (31), EP3 receptor (106, 107), and γ-aminobutyric acid A/B receptor (12, 104) in other mammalian cell lines. An advantage of these methods is the versatility provided by a variety of stable fluorophores like Alexa Fluor-conjugated streptavidin (51, 60). Streptavidin is tetravalent (binds four biotin molecules) and may induce oligomerization of transporters. However, monovalent streptavidin has been developed that allows single-molecule analysis and avoids issues with oligomerization (60). Alternatively, quantum dots can be used as fluorophores. Although they provide a higher quantum yield compared with other probes such as Alexa Fluor (45, 51), a major drawback is their intrinsic property to emit fluorescence in an intermittent “on” and “off” fashion (blinking) (56), which can be a problem to track real-time trafficking events (51). To circumvent this, nonblinking quantum dots have been developed and used to study real-time trafficking (57, 63). Another limitation of short-peptide tags is the need to perform an enzymatic reaction to label the sequence. For example, the BAD sequence needs to be biotinylated by the Escherichia coli biotin ligase BirA. To eliminate this step, BirA can be coexpressed to induce metabolic biotinylation of the protein of interest (51, 67). Alternatively, the SBP tag has been developed to directly bind to streptavidin. This method has been used to study trafficking of vesicle-associated membrane protein-2 (64). A similar method is based on E. coli lipoic acid (LplA) ligase, which adds an alkyl azide group from substrates to an engineered 13-amino acid LAP fused to the protein of interest (31). This method can be used with BirA to simultaneously visualize different membrane proteins (31). An improved one-step labeling method called probe incorporation mediated by enzymes uses an engineered lipoic acid ligase capable of catalyzing its covalent conjugation to LAP (99).

EPITOPE-BASED TAGS

They consist of genetically incorporating epitopes like hemagglutinin (HA), FLAG, or c-Myc in the extracellular domains and then using antibodies for labeling (Fig. 1E). HA tag was used to study membrane trafficking of renal potassium channel ROMK (29) and membrane proteins in other mammalian cell models (32, 52, 76). The FLAG tag was used to study membrane trafficking of renal vasopressin receptor type-2 and sweet taste receptor T1R2/3 in the kidney cell model, in which its role is beginning to be appreciated (11, 85) and c-myc tag to study surface accumulation of aquaporin-2 upon vasopressin stimulation in endocytosis-resistant membranes in kidney cells (11). However, prolonged labeling times and several washing steps are required to remove the unbound fluorophore with these methods, limiting their use in time-lapse microscopy.

FLUOROGENIC TAGS

They include fluorogenic labeling agents that result in fluorescence only upon binding to a genetically encoded short sequence (Fig. 1F). The first fluorogenic chemical tag was based on a tetracysteine sequence, recognized by a membrane-permeable ligand fluorescein arsenical hairpin binder (FIAsH) or its red-shifted analog resorufin arsenical hairpin binder (ReAsH). Binding of FIAsH or ReAsH to the TC tag results in fluorescence (8, 38). However, this method has only been sparsely used to study transmembrane proteins, such as agonist-induced conformational change in adenosine-2A receptor and α_2A adrenergic receptor (42) and the pannexin-1 channel in renal epithelial cell model (9). This method has been primarily used to study trafficking of intracellular proteins (17, 23) because strong reducing agents are needed in the extracellular space to facilitate the reaction between the tagged membrane protein and the fluorogens. Additional drawbacks include the facts that FIAsH and ReAsH may bind nonselectively to proteins with cysteine pairs, requiring prior destaining of cells with dithiols or is applicable only to cysteine-free background membrane proteins. The toxicity of biarsenical dyes may also be a concern (22). A new approach uses a genetically encoded fluorogen-activating peptide (FAP) that induces fluorescence upon binding to their conjugate nontoxic fluorogens (93). Several FAP/fluorogen pairs have recently been used to successfully study trafficking of various membrane proteins, such as β₂-adrenergic receptor (34) and cystic fibrosis transmembrane conductance regulator (44), among others (43, 74, 86). Coupling the fluorogen to a pH-sensitive dye also allowed studying pH changes in endocytic compartments (39). Combining advanced FAP-based approaches with high-resolution nanoscopy like stimulation emission depletion (or STED) allowed the long-term study of protein trafficking in live cells (35).

NANOBODY

They are recombinant or genetically encoded, antigen-specific, variable fragments of camelid heavy chain-only antibodies that are either genetically fused to FPs or conjugated to dyes or chromogenic enzymes (83) (Fig. 1G). Their small size, stability, reversible refolding, solubility in aqueous solution, and ability to recognize their endogenous targets in a specific transient conformation with subnanomolar affinity (73) makes them exploitable for single-particle tracking and manipulating membrane protein trafficking kinetics (25, 40, 41, 47, 58, 61, 71, 97, 83, 97, 103).

FLUORESCENT UNNATURAL AMINO ACID

A recent approach minimizes the undesirable effect of large-size tags on protein function by using site-directed mutagenesis to allow incorporation of a single fluorescent unnatural amino acid (UAA) (Fig. 1I). This method has been used to study trafficking of nicotinic ACh receptor (70). However, it requires coexpression of appropriate pairs of tRNAs and aminoacyl-tRNA synthetases that recognize the UAA (100).

TAG-FREE LABELING METHOD

This method allows labeling of proteins in their native state (94, 98) (Fig. 1J). It requires a known ligand for the surface protein (for example, bradykinin), a nucleophilic amino acid exposed to the surface, and a chemical-engineered ligand to transfer a probe (biotin or fluorescent) to the protein of interest, based on a typical nucleophilic acyl substitution reaction (65).

POTENTIAL APPLICATIONS AND CONCLUSION

Mutations or single-nucleotide polymorphisms leading to defective protein trafficking is a mechanism observed in about one-third of monogenic diseases affecting the kidney (82). A detailed understanding of these trafficking events will advance the characterization of better targets for pharmaceutical agents. A combination of single-molecule tagging coupled with high resolution has been very useful in studying how oligomerization (16), spatiotemporal organization (14), protein-protein interactions (6, 7, 26, 53), protein motifs/domains (90), and signaling pathways (18, 7) regulate the trafficking of membrane proteins in various mammalian cell models. Application of these novel technologies coupled with real-time tracking has provided information on apical targeting of receptors to highly specialized membrane domains, which can lead to clustering for efficient signal transduction (13, 21, 28, 37) and membrane receptor activation (58, 103) following stimulation by ligands (11). These methods can also be used to develop automated and high-throughput fluorescent drug screening (86) and study off-target effects of pharmacological agents (102) or can also be employed to study the cellular effects of precise inactivation of a molecule (49). Thus, most of the applications outlined above have a lot of potential to be exploited to enhance the understanding of renal transporter trafficking in live cells. While multiphoton microscopy has been used to image trafficking events in the kidney of live rats (27, 54, 81), very few have used this imaging modality to image membrane protein trafficking after single-molecule tagging. With the development of new and suitable epitope tags for renal transporters, regulation of their trafficking and surface behaviors could be studied in live animals, enabling a deeper understanding of renal physiology. In this review, we tried to provide an overview of single-molecule tagging methods to study transporter trafficking. Clearly, the choice of labeling method and imaging modality will be highly specific to each application. However, it is clear that these imaging techniques could be of immense value to renal physiologists.

GRANTS

This work was supported, in part, by American Heart Association-Predoctoral fellowship award 16PRE27510032 to Ankita B. Jaykumar and National Institutes of Health Grant 1R01DK107263 01A1, and an American Heart Association Grant-in Aid to Dr. Pablo A. Ortiz.

DISCLOSURES

No conflict of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.B.J. and P.S.C. prepared figure; A.B.J. drafted manuscript; A.B.J., P.S.C., and P.A.O. edited and revised manuscript; A.B.J. and P.A.O. approved final version of manuscript.