Live-Cell FRET Imaging of Phosphorylation-Dependent Caveolin-1 Switch

Abstract

The detection of dynamic conformational changes in proteins in live cells is challenging. Live-cell FRET (Förster Resonance Energy Transfer) is an example of a noninvasive technique that can be used to achieve this goal at nanometer resolution. FRET-based assays are dependent on the presence of fluorescent probes, such as CFP- and YFP-conjugated protein pairs. Here, we describe an experimental protocol in which live-cell FRET was used to measure conformational changes in caveolin-1 (Cav-1) oligomers on the surface of plasmalemma vesicles, or caveolae.

1. Introduction

Formation and function of caveolae require caveolin-1 (Cav-1) and cavin-family proteins inserted or associated with cholesterol and phospholipid-enriched microdomains of the plasma membrane [1–3]. Caveolae play critical roles in a variety of cellular functions, including endocytosis and transcytosis [4], the establishment of signaling hubs [5], storage of cholesterol [6], and buffering of mechanical stress [7]. The caveolar coat is far from static [8], but uncovering the intricate mechanisms by which Cav-1 molecules rearrange themselves to form tighter or looser associations upon receiving relevant cellular stimuli requires monitoring at the nanometer scale. One of the methods capable of reaching such nanometer resolution is FRET, which depends on fluorescence-based probes directly conjugated to proteins and biosensors. The most commonly used fluorescent probe is green fluorescent protein or GFP, which was originally derived from the jellyfish, Aequorea victoria [9]. Since then, the original GFP protein has undergone numerous mutations to produce enhanced GFP (eGFP), blue fluorescent protein (BFP), cyan fluorescent protein (CFP), and yellow fluorescent protein (YFP) (see Note 1). Variants of enhanced CFP and YFP (ECFP and EYFP) have become the most popular FRET pairs (see Note 2).

FRET occurs in the range of 1–10 nm, due to collision-free, resonance-based energy transfer between chromophore probes, without transmission of photons [10], when the emission band of the donor exhibits spectral overlap with the absorption band of the acceptor (Fig. 1). FRET has been used extensively to study conformational changes in molecules, cis- and trans-conversions, and the assembly of macromolecules [11–13]. Traditional approaches to FRET, such as acceptor photobleaching to quantify the donor fluorescence before and after photodestruction of the acceptor, requires the sample to be exposed to high-intensity light, often resulting in cellular phototoxicity [14]. In contrast, the live FRET technique, described later, assesses FRET efficiency from brief low-intensity light pulses, enabling rapid and dynamic measurements with relatively minimal photodamage in living cells (see Note 3). Live FRET microscopy can be employed to assess protein-protein interactions, such as those exhibited by Cav-1 monomers in their natural environment, and to measure spatial and temporal changes as they are occurring between binding partners in a single living cell. We have used this nondestructive technique to study changes in protein-protein interactions at the caveolar coat due to phosphorylation of Cav-1 at tyrosine 14 (Y14) by Src in the process of endocytic uptake of albumin in live cells [8, 15].

Fig. 1.

Idealized absorption and emission spectra for FRET donor and acceptor pair, such as ECFP (donor) and EYFP (acceptor) fluorescent proteins. Absorption spectra are illustrated as blue and yellow curves, while the emission spectra are presented as green and red curves. The region of overlap between the donor emission and acceptor absorption spectrum is represented by a gray area. For FRET to occur, there should be a direct excitation of the acceptor by the emission spectrum of the donor, and the donor and acceptor molecules need to have the appropriate dipole alignment and be positioned within 10 nm of each other

2. Materials

2.1. Plasmids and Primers

1. Cav-1 gene (NM_001753) (Addgene, plasmid #55468)

2. pAMCyan1-N1 (Takara, #632440) and pEYFP-N1 (Clontech #6085-1) vectors,

3. Primer pair: Cav1-WT-Fwd: 5′-ACTAGCTAGCGACCGCCA TGTCTGGGGGCAAA-TAC-3′ and Cav-1-WT-Rvrs: 5′-ACTGGGTACCGTTATTTCTTTCTGC AAGTTGATGCG-3′

2.2. General DNA Manipulation

1. DNA Phusion High-Fidelity Polymerase.

2. Nhe1 and Kpn1 restriction enzymes.

3. T4 DNA ligase.

4. Chemically competent cells, such as MAX Efficiency DH5α T1 (Invitrogen).

5. Agarose gel, electrophoresis apparatus for agarose gels, agar plates, Mini-Prep and Midi-Prep kits.

2.3. Generation of Cell Line Stably Expressing Fluo-Tagged Cav1

1. Lipofectamine 3000 (Invitrogen).

2. G418 or Geneticin.

2.4. FRET Picture Acquisition

1. Glass-bottom dishes (MatTek).

2. Confocal microscope such as Zeiss 710 BIG (Carl Zeiss).

3. Methods

3.1. Labeling of the Cav-1 Protein at the C-Terminus with EYFP or ECFP Fluorescent Protein Tags

1. Combine the primer pair lacking the stop codon: Cav1-WT-Fwd Cav-1-WT-Rvrs: with the full-length Homo sapiens Cav-1 gene as a template in a PCR reaction using DNA Phusion High-Fidelity Polymerase according to the manufacturer’s instructions [8, 15].

2. Purify the resulting PCR fragment by running agarose gel.

3. Digest the purified PCR fragment, as well as pEYFP-N1 vector, with restriction enzymes 5′-NheI and 3′-KpnI (restriction sites underlined in the primers) according to the manufacturer’s instructions.

4. Ligate digested PCR fragment with a digested pEYFP-N1 vector using T4 DNA ligase, according to the manufacturer’s instructions.

5. Subsequently, transform DH5α bacteria with ligated DNA and spread them on agar plates.

6. Isolate DNA from single colonies on the agar plate and purify them using Mini-prep kit.

7. Verify the correct insertion of the fluorescent tags by Sanger sequencing using standard CMV fwd sequencing primer. Prepare Midi-prep of the correct DNA for future transfections.

8. Generate the Cav-1-CFP construct by repeating steps 1–7 but by substituting pAmCyan vector for pEYFP-N1.

3.2. HEK Cell Transfections

1. Split and culture HEK cells on 6-cm dishes for 2 days until they reach ~50–80% confluence (see Note 4). Transfect cells with either 2 or 4 μg of Cav-1-EYFP and Cav-1-ECFP cDNA (4 μg alone or a combination of 2 μg each) using Lipofectamine 3000 according to the manufacturer’s protocol.

2. After 24 h, add G418 (geneticin) to the cells at a concentration of 0.5 mg/mL. Stable expression of cDNA in HEK cells should be obtained within 10–14 days.

3. As a negative experimental control, and for determination of optimal conditions for minimizing photobleaching of fluorophores with time, cotransfect one batch of HEK cells with a combination of pEYFP-N1/pAmCyan empty vectors by repeating steps 1–2.

4. To measure DER and AER constants (see Note 3), prepare a batch of HEK cells transfected with Cav-1-EYFP construct only (an acceptor-only control), and a nd batch transfected with Cav-1-ECFP construct only (a donor-only control), following steps 1 and 2.

3.3. Plating Cells for the Experiment

1. Confirm equivalent coexpression of EYFP- and ECFP-tagged Cav-1 constructs in HEK cells intended for FRET by using fluorescent microscopy.

2. Plate four batches of HEK cells (FRET sample, a negative control, a donor-only control, and an acceptor-only control) on glass-bottom dishes and keep them growing on glass for 48 h prior to the experiment (see Note 5).

3. Deprive HEK cells of serum by changing medium to phenol-red free DMEM for 2 h before the beginning of Live FRET measurements.

3.4. Microscope Measurements

1. During long-term live-cell imaging experiments, aim to maximize the photon collection and minimize the laser power of the confocal microscope, such as Zeiss 710 BIG.

2. Use Plan Apochromat objective, at least ×63 magnification with numerical aperture 1.40 or higher.

3. Open the pinhole to about 3.3 Airy units to collect more photons, and to have optical sections of ~2 μm, so it is less likely that the caveolin-positive vesicle leaves this optical section during the course of the experiment.

4. Use low laser power, in the 1–4% range.

5. Before the beginning of the experiment, collect fluorescence (I^da and I^dd, followed by I^da* and I^aa*; see Note 3) for the two control samples: HEK cells transfected with Cav-1-ECFP (donor only) and HEK cells transfected with Cav-1-EYFP (acceptor only) to calculate DER and AER constants. Use the absorption/emission settings described in Table 1.

6. Begin the actual FRET experiment by equilibrating HEK cells cotransfected with a mixture of Cav-1-EYFP/Cav-1-ECFP for 1 h in the environmental PeCon-heated chamber with CO₂ controller attached to the microscope stage.

7. Generate images (for ~5 min) to obtain basal fluorescence before the addition of the stimulus from ROIs (Regions of Interest) situated in the peripheral areas of the cell. Use the 458-nm laser line to excite and collect simultaneously the emission spectra in 526–608-nm region for FRET (I^DA), and 463–516-nm region for CFP (I^DD). In addition, at each time point, measure I^AA by exciting the FRET sample with 514-nm laser line and collecting emission fluorescence in the 526–608-nm range.

8. After ~5 min, stimulate cells with BSA (3 mg/mL; see Note 6) and then continue the measurements at each time point for FRET, CFP, and I^AA for an additional 20–30 min.

9. Repeat steps 2–4 for the negative control using HEK cells cotransfected with the mixture of pEYFP-N1/pAMCyan empty vectors.

Table 1.

List of excitation and emission wavelengths needed to convert raw FRET reading to a corrected FRET (cFRET)

	Sample	Excitation (nm)	Emission (nm)
I da	Donor-only control	458	526–608
I dd	Donor-only control	458	463–516
I da*	Acceptor-only control	458	526–608
I aa*	Acceptor-only control	514	526–608
I DA	FRET sample	458	526–608
I DD	FRET sample	458	463–516
I AA	FRET sample	514	526–608

The donor spectral bleed-through and acceptor crossexcitation are accounted for, according to the formula: cFRET = I^DA − DER * I^DD − AER * I^AA, where $\text{DER}\text{=}\frac{I^{\text{da}}}{I^{\text{dd}}}$, and $\text{AER}\text{=}\frac{I^{\text{da*}}}{I^{\text{aa}*}}$

3.5. Analysis

1. Use ImageJ (Fiji) to quantify fluorescence intensities for I^da, I^dd, and I^da*, I^aa*, as well as FRET (I^DA), cyan (I^DD), and I^AA, in the following way:

   1. Download and install Fiji from https://imagej.net/Fiji/Downloads
   2. Go to Fiji analysis→ set measurements→ then, check: mean gray value, standard deviation, and area
   3. Draw a rectangle around the area whose fluorescence you want to measure using the rectangle tool
   4. Go to Analyze → measure, or use command “ctrl-M”
   5. Express Intensity as a mean gray value ± standard deviation

2. First, calculate coefficients DER and AER according to the formula in Note 3.

3. Next, for each time point, calculate corrected FRET (cFRET) by using DER and AER constants obtained earlier:

   $$\text{cFRETt}=I^{\text{DA}}-\text{DER}\ast {\text{I}}^{\text{DD}}-\text{AER}\ast {\text{I}}^{\text{AA}}$$

4. To account for temporal fluctuations in donor fluorescence, express changes in cFRET at each time point as the FRET index by dividing cFRET intensity at each time point by CFP (I^DD) intensity at a given time point: $\mathrm{F}\mathrm{R}\mathrm{E}\mathrm{T}\text{index}=\frac{\text{cFRET}}{I^{\text{DD}}}$.

5. Subsequently, normalize to 1 the FRET index before BSA stimulation, and represent any changes in FRET after BSA stimulation relative to it (Fig. 2; see Note 7).

Fig. 2.

Dynamic changes in caveolin-1 upon BSA treatment, monitored in real time by FRET. HEK cells were cotransfected with either wild-type or Y14F Cav1-YFP and Cav1-CFP constructs to create FRET pairs, followed by selection with G418 for stable expression. Prior to the experiment, cells were serum deprived for 2 h, and images were then acquired with a Zeiss 710 BIG confocal microscope in a temperature- and CO₂/humidity-controlled chamber. FRET was recorded every 15 s using the 458-nm laser line for excitation; the 526- to 608-nm emission range was measured for FRET and the 463- to 516-nm emission range for CFP, recorded as an internal control. After temperature equilibration and stabilization of fluorescence, recordings began and basal FRET was collected for 2–5 min before stimulation with BSA (3 mg/mL), after which FRET was collected for an additional 9 min. At each time point, the FRET index was calculated (FRET/CFP) to account for temporal fluctuations in fluorescence. The FRET index before BSA stimulation (basal) was normalized to 1, and all subsequent time points were expressed as a fraction of the basal fluorescence. The FRET trace is shown for WT-Cav-1 and Y14F-Cav1. Data are shown as mean ± SD

4. Notes

1. It is known that many fluorescent proteins (FPs) can easily form oligomers at high concentrations (mM), which may lead to unforeseen artifacts. For example, noncovalent dimerization of GFP/eGFP is a common [10] occurrence [16, 17]. Other FPs can form even higher-order oligomeric structures, such as dsRed, which has the propensity for tetramerization [18]. Many of these wild-type FPs, however, can be engineered into monomers only, such as by the introduction of the A206K mutation on the dimerization interface of the GFP protein [19, 20]. For typical FRET studies, the standard EYFP/ECFP pairs are used with the assumption that their concentration in the cell is less than the threshold needed for oligomerization [10].

2. The sensitivity of biosensors based on ECFP and EYFP is generally limited [21]. A high-efficiency FRET pair, CyPet and YPet, has been developed to significantly enhance the dynamic range of FRET [22]. However, CyPet is not suitable for live-cell imaging as it folds very poorly at 37 °C [23].

3. This method [24] requires correction for crosstalk, which consists of the donor spectral bleed-through (SBT) and acceptor cross excitation. Crosstalk is corrected by two reference measurements, in addition to the actual FRET experiment [24]. The first is performed using a control sample containing only donor fluorophores in order to account for donor bleed through (DER). The second is performed on a sample containing only acceptor fluorophores to account for direct acceptor excitation (AER). FRET can be corrected for the crosstalk according to the following formula:

   $$\text{cFRET}=I^{\text{DA}}-\text{DER}\ast {\text{I}}^{\text{DD}}-\text{AER}\ast {\text{I}}^{\text{AA}},\text{where}$$

   $$\mathrm{DER}=\frac{I^{I^{\text{d}}}}{I^{\text{d}}},\text{measured}\text{from}\text{donor-only}\text{control}\text{sample};$$

   $$\text{AER}=\frac{I^{\text{da}*}}{I^{\text{a}*}},\text{measured}\text{from}\text{acceptor-only}\text{control}\text{sample};$$

   while I^DA, I^DD, and I^AA are all measured from the actual FRET sample, containing the donor/acceptor pair. To calculate DER, the donor-only sample is illuminated with donor excitation and fluorescence intensity is collected at acceptor emission (I^da); next, the donor-only sample is illuminated with donor excitation and fluorescence is collected at donor emission (I^dd).

   To calculate AER, the acceptor-only sample is illuminated with donor excitation and fluorescence intensity is collected at acceptor emission (I^da*); next, the acceptor-only sample is illuminated with acceptor excitation and fluorescence intensity is collected at acceptor emission (I^aa*).

   To measure I^DA, the fluorescence intensity is obtained with donor excitation and collected at acceptor emission in FRET sample; I^DD fluorescence intensity is obtained with donor excitation collected at donor emission in FRET sample; and finally, I^AA fluorescence intensity is obtained with acceptor excitation collected at acceptor emission in FRET sample.

4. All cell experiments reported here were performed in HEK cells. Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 10 mM HEPES, 1% penicillin/streptomycin (10,000 U/mL), and, as required to maintain stable expression, G418 (Geneticin). Cells were cultured at 37 °C and 5% CO₂ and maintained in 10-cm tissue culture dishes.

5. You may want to culture cells on poly-d-lysine–treated glass-bottom dishes, if the attachment of cells to the glass surface is a problem. Prepare 1 mg/mL stock solution by dissolving 100 mg Poly-d-lysine hydrobromide in 100-mL water. Filter sterilize the solution, then store aliquots at −20 °C. The working solution is obtained by diluting the stock solution with sterile water to 0.1 mg/mL. The working solution is then added to the dish in such a way that the glass surface is completely covered and incubated for 1 h at room temperature in the hood. Subsequently, the poly-d-lysine solution is suctioned off and dishes are washed twice in PBS.

6. Addition of BSA was performed to the side of the dish via tubing connected to a syringe, without opening of the environmental chamber, with as little disruption to the cells as possible.

7. Figure 2 depicts FRET measured continuously in live HEK cells before and after stimulation with 3 mg/mL BSA. HEK cells were cotransfected with YFP- and CFP-conjugated Cav1 cDNA coding for either wild-type Cav1, or Y14F mutant. Within 10 min of BSA stimulation, there is a decrease in FRET signal (Cav1-YFP/Cav-1-CFP ratio) to about ~60% of the initial value for WT Cav-1. No significant changes in FRET were observed when the irreversible, phosphodefective Y14F mutant was stimulated with BSA. This indicates that phosphorylation-dependent changes at tyrosine 14 introduced by mutation trump any changes in caveolar coat packing due to phosphorylation.