PIN-G – A novel reporter for imaging and defining the effects of trafficking signals in membrane proteins

Background The identification of protein trafficking signals, and their interacting mechanisms, is a fundamental objective of modern biology. Unfortunately, the analysis of trafficking signals is complicated by their topography, hierarchical nature and regulation. Powerful strategies to test candidate motifs include their ability to direct simpler reporter proteins, to which they are fused, to the appropriate cellular compartment. However, present reporters are limited by their endogenous expression, paucity of cloning sites, and difficult detection in live cells. Results Consequently, we have engineered a mammalian expression vector encoding a novel trafficking reporter – pIN-G – consisting of a simple, type I integral protein bearing permissive intra/extracellular cloning sites, green fluorescent protein (GFP), cMyc and HA epitope tags. Fluorescence imaging, flow cytometry and biochemical assays of transfected HEK293 cells, confirm the size, topology and surface expression of PIN-G. Moreover, a pIN-G fusion construct, containing a Trans-Golgi Network (TGN) targeting determinant, internalises rapidly from the cell surface and localises to the TGN. Additionally, another PIN-G fusion protein and its mutants reveal trafficking determinants in the cytoplasmic carboxy terminus of Kv1.4 voltage-gated potassium channels. Conclusion Together, these data indicate that pIN-G is a versatile, powerful, new reporter for analysing signals controlling membrane protein trafficking, surface expression and dynamics.


Background
The identification of sequence motifs that direct proteins to appropriate intracellular and plasma membrane locales represents a fundamental objective of cell biology [1]. Once defined, such signals permit the identification of mechanisms underlying protein trafficking [2] and the targeted delivery of therapeutic agents to those subcellular domains where they are most effective [3]. Unfortunately, the identification of trafficking signals is not trivial as they may consist of non-contiguous amino acids, whose juxtaposition to form a binding site may only be apparent in the fully folded protein [4]. Even in cases where trafficking signals are comprised of contiguous amino acids, they may be hard to identify since they lie within discrete subunits of complex oligomeric proteins [5]. Moreover, trafficking signals may overlap [6], act hierarchically [7], or be capable of being switched from inactive to active states through second messenger regulation [8].
A well-established strategy to identify trafficking signals has been to infer their necessity and sufficiency from their ability to direct simpler reporter proteins to which they are fused, to the appropriate cellular compartment [9][10][11]. Unfortunately, available reporter proteins have numerous limitations including their endogenous expression in diverse cells and tissues, lack of convenient cloning sites and potential interactions with intra-or extracellular proteins that may affect their subcellular distributions [12]. Even more significant is the limited capacity for detection of presently available reporters, not least because of their reliance upon the limited range of antibodies that are commercially available and the indirect nature of the detection methodology. To circumvent such problems, we have engineered an alternative protein trafficking reporter system ( Fig. 1) consisting of a simple artificial type I integral protein bearing permissive intra-and extracellular multiple cloning sites, a green fluorescent protein moiety (GFP) that permits direct localisation by fluorescence imaging in live cells and epitope tags for biochemical detection of protein turnover and surface expression.

Vector construction
To generate a suitable trafficking reporter (Fig. 1), a region encoding EGFP and unique extracellular restriction sites from pEGFP-C1 (Clontech) was first inserted into the cell surface expression vector, pDisplay, (Invitrogen) to generate an intermediate EGFP-pDisplay construct (Fig. 1A). The entire protein coding region of EGFP-pDisplay was then PCR amplified and cloned into a pEGFP backbone lacking the EGFP insert, thereby introducing a second, distinct, intracellular multiple cloning site (MCS-2). The primary homologue, described henceforth as pIN-G, is a Cloning scheme for the preparation of the pIN-G construct and the transmembrane topology of the pIN-G reporter protein Figure 1 Cloning scheme for the preparation of the pIN-G construct and the transmembrane topology of the pIN-G reporter protein.

1A.
A fragment encoding EGFP, obtained by AgeI/SalI digestion of pEGFP-C1, was ligated into an XmaI/SalI digested pDisplay backbone and the entire protein coding region PCR-amplified and ligated into a host backbone generated through NheI/KpnI digestion of pEGFP-C1 to retain the 3' multiple cloning site but exclude the EGFP coding region. The final pIN-G construct, with the appropriate reading frame throughout, was generated via a subsequent round of PCR. See Methods for further details. 1B. Following expression and cleavage of the Igk-chain leader sequence, the pIN-G protein is proposed to adopt a transmembrane orientation where HA and cMyc epitope tags and GFP have an extracellular disposition at the cell surface (or intra-lumenal in organelles or vesicular trafficking intermediates). 5042bp pUC-based vector (see Table 1 for details) containing all the desired protein coding elements in the correct reading frame (Fig. 1B)(deposited in Genbank as Accession number AY841887).

Properties of pIN-G
Transfection of pIN-G into HEK293 cells, and subsequent immunoblotting with antibodies against EGFP and HA disclosed a novel 41.3 kDa band, which was absent from mock-transfected cells (Fig. 2a). Constructs encoding pIN-G fused to a TGN targeting motif and the carboxy terminus of the voltage-gated potassium channel Kv1.4 gave bands of (44 kDa) and (51 kDa), respectively.
Fluorescence images of pIN-G-transfected HEK293 cells ( Fig. 2b panel A) revealed strong green heterogeneous fluorescence within intracellular organellar structures and at the plasma membrane. In contrast, EGFP alone ( Fig. 2b panel B) was expressed homogeneously throughout the cell. Thus, the presence of an Igκ leader sequence and the PDGF receptor transmembrane region, but no added trafficking motifs, appeared to direct pIN-G via the secretory pathway to the cell surface. To confirm this, first, the surface expression of pIN-G was tested by immunostaining non-permeabilized fixed cells with anti-EGFP ( Fig. 2b panels D-F) and anti-HA ( Fig. 2b panels G-I), and subsequent detection with Cy3-labeled secondary antibodies.

pIN-G is a faithful trafficking reporter
Next, we sought to establish the potential of pIN-G as a reporter of protein trafficking by exploiting the observation that a contiguous carboxy terminal motif within TGN-38 is both necessary and sufficient to localise this protein to the Trans-Golgi Network (TGN) [20]. Thus, cDNA encoding the TGN-38 signal with a corresponding stop codon, was cloned into the putative intracellular multiple cloning site of pIN-G (MCS-2) to yield a construct, designated as pIN-TGN. Fluorescence imaging of   a) pIN-G and fusion protein constructs yield proteins of the anticipated sizes following expression in HEK293 cells

Analysis of pIN-G surface expression levels
To confirm the cell surface disposition of pIN-G and pINfusion constructs, and its facile resolution by cell sorting assays, FACS analysis was used. In mock-transfected cells (Fig. 5A), or those treated with Cy5 secondary antibodies alone (Fig. 5B), most cells showed low fluorescence (Fig.  4, quadrant R3 (red) and R6 (green)). However, upon transfection with pIN-G or pIN-TGN, a new population of cells was revealed that showed strong fluorescence in the green channel, (Fig. 5C, 5D, quadrant R6). After surface labelingof pIN-G and pIN-TGN transfected cells, with anti-HA and Cy5 primary and secondary antibodies, a novel cell population displaying both red and green fluorescence ( Fig. 5E and 5F, quadrant R4) emerged, corresponding to those expressing construct at the cell surface. By defining F as the fraction of transfected cells in the R4 quadrant (i.e. F = R4/(R4+R6)), pIN-G surface expression was much greater than that of pIN-TGN (F = 0.99 vs. 0.44, respectively).

A pIN-G fusion protein identifies internalization determinants in the Kv1.4 carboxy terminus
Available evidence suggests that Kv1 family potassium channels contain one or more carboxy terminal internalization determinants [15,21]. To identify such determinants, a construct was generated encoding pIN-G to which the entire carboxy terminus of rat Kv1.4 (Genbank accession no. NM 012971) had been fused (pIN-JHE811). In transfected HEK293 cells, pIN-JHE811 fluorescence was localized to intracellular membrane vesicles and the cell surface ( Fig. 6a, panel A). Surface expression of pIN-JHE811 was confirmed through immunocytochemistry of non-permeabilised cells treated at 4°C with antibodies raised against the HA epitope tag (see Fig. 6a, panel B). However, in contrast to pIN-G (Fig. 6b, panel C), pIN-JHE811 surface labeling appeared less intense as indicated by a decrease in green:red (yellow) surface staining (Fig.  6b, panel E). Since, pIN-JHE811 contains putative internalization motifs, we questioned whether such diminished surface labeling at steady-state compared to pIN-G, might reflect partial redistribution of pIN-JHE811 into the endocytic pathway. To test this notion, we determined the fate of surface-labeled pIN-JHE811 compared to pIN-G following a pulse chase-protocol with anti-HA antibody (

Discussion
In this paper, we describe a novel construct -pIN-G -for the facile detection of trafficking determinants within membrane proteins. Expression of pIN-G yields an artificial, GFP-tagged, protein -PIN-G -whose characteristics and topology conform to the type I protein depicted in Figure 1B. Thus, PIN-G is recognised by surface-applied antibodies corresponding to each extracellular moiety and is also re-distributed by sequences appended to the putative intracellular domain. The presence of PIN-G at the cell surface (and throughout membranes of the secretory pathway) also suggests that it does not contain any intrinsic targeting or trafficking determinants. However, the addition of sequences containing known or putative targeting determinants re-specifies the trafficking fate of the expressed fusion construct indicating that pIN-G is a faithful reporter for visualising and analysing trafficking mechanisms. Equally salient is the utility of the pIN-G system in identifying and dissecting motifs, such as the previously undocumented Tyr612 in Kv1.4, that control the dynamics and expression of cell surface proteins, especially when, like Kv1.4, they have no specific radioactive ligands or where biotinylation strategies may not be straightforward.
In contrast to existing reporters [9][10][11], the pIN vector is a considerably more versatile tool, not least because of the presence of intra-and extracellular multiple cloning sites, epitope and GFP tags and lack of endogenous expression. Moreover, unlike with present reporters, any aberrant expression of pIN-G or its fusion constructs (such as inclusion body formation), while absent in the present study, should be readily visualised. Such versatility could be extended even further to include the dynamic visualisation of interacting proteins in live cells [22], and the development of combinatorial library approaches for identifying intracellular or extracellular signalling mechanisms affecting trafficking [23]. Other potential applications of pIN technology include the generation of constructs with alternate fluorophores to GFP such as mRFP [24], thereby, permitting the development of biosensors or energy transfer assays between pIN-fusion constructs. Finally, the reasonably small size and simplicity of the pIN construct, especially when compared to most integral proteins mediating signal transduction, should facilitate the analysis of trafficking in cells such as neurons which are hard to transfect [25] and also the development of viral gene delivery methodologies where packaging size is critical [26].

Conclusion
Here we have described the construction and faithful expression of a novel trafficking reporter protein, specifically designed for the detection and characterization of autonomous trafficking motifs. By incorporating epitope Re-specification of the distribution of pIN-G by an autono-mous targeting signal and GFP tags into a simple type I transmembrane protein, our construct permits the direct visualization and biochemical dissection of protein signals and signalling domains transplanted from much more complex parent membrane proteins. The potential uses of the pIN-G reporter are, however, not limited to trafficking studies and could serve as a platform for the development of novel biosensors and the high-throughput screening of protein-protein interactions. All other chemicals were of reagent grade or higher purity.

pIN-G vector construction
Generation of the pIN-G construct (Fig. 1B) was achieved by first digesting the pEGFP-C1 vector (Clontech, Palo Alto, CA) with Age I (bp 601) and Sal I (bp 1370) to yield a 769 bp fragment, corresponding to the coding region of EGFP. The resulting fragment was then subcloned into the Xma I (bp 849) and Sal I (bp 869) sites of the pDisplay vector (Invitrogen, San Diego, CA) to give a 6074 bp intermediate EGFP-pDisplay construct. The entire protein coding region of EGFP-pDisplay (bp732-1803) was then subject to PCR amplification using the following forward and reverse oligonucleotide primers: 5'-ATC CGC TAG CGC TAC TAG TCG CCA CCA TGG AGA CAG ACA CAC TC-3' and 5'-AAA GAC CGG TGA ACG TGG CTT CTT CTG CCA AAG CAT-3' which also served to introduce Nhe I and Kpn I sites for subcloning. Separately, the entire EGFP-coding region was excised from pEGFP-C1 by digestion with Nhe I (bp592) and Kpn I (bp1380). The linearized pEGFP-C1 backbone was then used as a host into which the EGFP-pDisplay PCR fragment, generated above, was subcloned. The resulting 5037 bp pIN precursor construct (pIN-G') was then used to generate pIN construct derivatives. For example, pIN-G was constructed by generating two PCR fragments pIN-A (401 bp), containing SnaB I and Pvu I sites for subcloning, (forward primer: pIN-B were then digested with Pvu I, ligated to give a 1383 bp fragment which was then cloned into the corresponding Snab I and Kpn I sites in pIN-G' to yield a correctlyframed pIN-G product (5042 bp) containing a novel Pvu I site.

pIN-TGN-38 vector construction
The pIN-TGN-38 construct, was generated by PCR amplification of the sequence encoding the trans-Golgi-Network targeting signal (SEQYDRL: refs [13,14]) using p∆Mep4-TGN-38-GFP as a template and the following forward and reverse oligonucleotide primers

pIN-JHE811 vector construction
A pIN fusion construct, encoding pIN-G fused to the entire intracellular carboxy terminus of the Kv1.4 voltagegated potassium channel, was generated by PCR using pEYFP-Kv1.4 [15] as a template and the following forward and reverse oligonucleotide primers: 5'-AAG TCT CTC TCG GGC TTC AGC -3' and 5'-AAG TCT CTC TCG GGC TTC AGC -3' that contained flanking sequences corresponding to the Kpn I site of pIN-G. The resulting PCR fragment was then cloned into the Kpn I site in MCS-2 and ligated using T4 ligase as described [15]. pIN-JHE811-L-A and pIN-JHE811-Y-F were created using the Stratagene QuikChange site-directed mutagenesis kit with pIN-JHE811 as a template and the following mutagenic oligonucleotide primers: 5'-CA TAC CTA CCT TCT AAT TTG CTC AAG AAA TTT CGG AGC TCT AC -3' and 5'-CTG GGG GAC AAG TCA GAG TTT CTA GAG ATG GAA GAA GGG-3'. The identity of all constructs was confirmed by restriction mapping and DNA sequencing (ACGT, Toronto, Canada, Geneblitz, Sunderland, UK).
Tissue culture HEK293 cells were subcultured every 3-4 days, incubated at 37°C in a humidified incubator and maintained with 5% CO 2 . Cells were detached from the flasks with phosphate-buffered saline without calcium and magnesium (PBS -), quenched in fresh medium, and re-seeded into T75 cm 2 flasks. Cells were cultured in Dulbecco's minimal essential medium (DMEM) containing 0.11 g/L sodium pyruvate and pyridoxine and supplemented with 10% FBS (v/v), 2 mM L-glutamine (w/v) and 1% penicillin and streptomycin.

Transfection
All transfections were carried out using Fugene 6 (Roche, UK), according to the manufacturers' instructions. HEK293 cells were seeded at 1:40 from a confluent T75 into a 6-well plate and grown to 50% confluency overnight at 37°C. 2 µg cDNA was added to 150 µl DMEM and mixed in an eppendorf tube containing 6 µl of Fugene6 reagent and 150 µl DMEM. The cDNA and reagent solutions were combined to allow the formation of lipid complexes for 20 minutes at room temperature. Subsequent to the addition of fresh DMEM, to remove any antibiotics, the DNA-lipid complexes were added to the cells in a drop wise manner and incubated for 12 hours at 37°C. After this period cells were either transferred to a 10 cm dish for further propagation or re-seeded 1:4 onto collagen-coated coverslips in 12-well plates then incubated at 37°C overnight.

Immunoblotting
HEK293 cells were lysed in RIPA buffer containing a cocktail of protease inhibitors (Roche, UK), and passed 8X through a 22-gauge needle to shear nucleic acids. The sheared lysate was then centrifuged at 16,000 gav for 15' and aliquots of the supernatant electrophoresed, and transferred to nitrocellulose and processed as described previously [15] using monoclonal antibodies to HA (1:1000)), or polyclonal antibodies to GFP (1:10,000). Immunoreactivity was detected by incubating each membrane with horseradish peroxidase-conjugated donkey anti-mouse or anti-rabbit (1:1000; Amersham, UK) followed by enhanced chemiluminescence (ECL; Amersham Biosciences) and film detection (Amersham Bio-Max).

Immunostaining of transfected HEK293 cells
Image-based detection of PIN-fusion proteins was observed in HEK293 cells transiently transfected with pIN-G, pIN-TGN or pIN-JHE811, fixed in 4% paraformaldehyde for 20 minutes at room temperature, washed in PBS, quenched with 0.1 M glycine and mounted onto glass slides using ProLong Gold anti-fade (Molecular Probes). The surface expression of PIN proteins was determined by incubating non-permeabilised, fixed cells with anti-HA antibodies (1:1000) for 30 minutes at room temperature and detecting with Cy3-conjugated goat antimouse secondary antibody (1:200, Jackson Immunoresearch, Stratech Scientific, Cambs, UK). Intracellular staining of organelles was performed by fixing HEK293 cells in 4% paraformaldehyde (as above), permeabilising with 0.5% saponin for 5 minutes followed by incubation with antibodies raised against the appropriate markers at a concentration of 10 µg/ml in 0.01% saponin. Detection of specific staining was performed with Cy3-conjugated antimouse secondary at 1:200.

Internalization assays
Internalization of the PIN fusion proteins was determined by pulse-labeling transfected HEK293 cells at 4°C with anti-HA antibodies (1:1000) for 30 minutes. The cells were then either fixed directly using 4% paraformaldehyde and permeabilised with 0.5% saponin, or incubated for 1 hr at 37°C in fresh medium prior to fixation and permeabilisation. For imaging, internalized anti-HA antibody was detected with Cy3-conjugated goat anti-mouse secondary (1:200) and nuclei were DAPI-stained (0.05 µg/ml) prior to mounting. For FACS analysis, to distinguish surface labeling transfected cells were not permeabilised and external anti-HA antibody was detected with Cy5-conjugated goat anti-mouse secondary (1:200)

Fluorescent activated cell sorting (FACS)
For surface analysis, transfected cells were detached with PBS without calcium and magnesium (PBS -) and adjusted to a concentration of 5 × 10 7 /ml with DMEM containing 1% FBS (v/v). 100 µl aliquots of cells were incubated with primary antibody (anti-HA) at a concentration of 10 µg/ ml diluted in Dulbecco's phosphate-buffered saline with calcium and magnesium (PBS + ) for 30 minutes at 4°C. For pulse-chase assays, transfected cells were plated onto 10 cm cell-culture dishes and incubated overnight at 37°C. The dishes were placed on ice and the cells labeled with anti-HA at a concentration of 10 µg/ml diluted in PBS + for 30 minutes at 4°C. Cells were then detached and transferred to FACS tubes. After three washes with 300 µl PBS + , 1% FBS (v/v), all samples were incubated with Cy5conjugated goat secondary anti-mouse IgG 1 (1:200) diluted in 10% FBS and 90% PBS + (v/v) for 45 minutes at 4°C. Cells were fixed with 100 µl aliquots of 2% (w/v) formaldehyde, 100 µl PBS -, 300 µl PBS + . The dilutions of antibodies used were saturating as confirmed by titration. Cells (20,000 per sample) were analysed in a FACScaliber ® flow cytometer (Becton Dickinson, UK) and mean fluorescent intensity values taken. Internalization half-lives were determined assuming a first-order exponential.

Delta Vision deconvolution microscopy
Mounted coverslips were visualized on an Olympus IX-70 inverted microscope using a ×100 UPLAN objective (NA 1.35) and images captured on a Photometrics CH350L CCD camera. The Delta Vision restoration microscope (Applied Precision Instruments, Seattle) utilized a Unixbased computer system equipped with SoftWoRx version 2.50, acquisition software. Epifluorescence was recorded using filter sets for DAPI/FITC/Texas red and cy3/cy5. Using the high precision nanochassis stage, widefield optical sections of 0.2 µm were acquired throughout the z plane of the cells. Z stacks generated were then deconvolved using a constrained iterative algorithm assigned by Delta Vision.