Key words

1 Introduction

Intercellular communication through gap junction channels is known to play important roles in cell homeostasis and synchronization of physiological functions of tissues and organs [1–3]. In vertebrates, junctional coupling is mediated by connexins, a family of proteins that oligomerize to form intercellular gap junction channels [4]. Since malfunction of junctional coupling and mutation of connexins can cause human diseases [5], it is important to study and to understand mechanisms governing intercellular gap junction communication in different biological systems. Methods designed to follow changes in coupling strength or the rates of molecular transfer among coupled cells would certainly facilitate such investigations.

Besides electrophysiological approaches, a number of optical methods have been developed to assay gap junction coupling. Essentially all these methods are based on fluorescence imaging of a soluble fluorescent tracer that can diffuse through gap junction channels, i.e., tracers with a molecular weight below 1000. In addition, imaging methods that indirectly assess the extent of cell coupling have also been reported. For example, Ca2+ imaging and subsequent correlation analysis of Ca2+ activity among individual cells has been used to infer cell coupling in cell populations [6].

This chapter describes two fluorescence imaging assays, LAMP [7] and infrared-LAMP [8], that employ a caged and cell membrane permeable coumarin dye NPE-HCCC2/AM [9] to selectively mark a cell or a subpopulation of cells by using the technique of photoactivation (uncaging) [10]. NPE-HCCC2/AM is the neutral ester of a caged coumarin dye NPE-HCCC2 (Fig. 1). It can diffuse across cell membranes and becomes trapped in the cytosol once the AM ester is hydrolyzed by cellular esterases. Upon photoactivation , either with UV light (LAMP assay) or with two photon laser excitation (infrared-LAMP assay), NPE-HCCC2 is converted to a highly fluorescent coumarin dye HCCC2. HCCC2 is fairly small (molecular weight 449 Da) so it can rapidly diffuse to neighboring cells through gap junction channels. The process of intercellular dye diffusion from the donor cell to the recipient cell is then captured by digital fluorescence microscopy using either wide field, confocal, or two photon imaging . Subsequent analysis of time lapse imaging data provides quantitative information on the kinetics of cell–cell dye transfer [7, 8, 11, 12]. To extent this technique to assay cell coupling in vivo, we have also developed another technique termed Trojan-LAMP and applied it to a model organism, the nematode C. elegans [13]. The study revealed that early during embryonic development , the pattern of cell coupling in the developing embryo underwent dramatic remodeling and germ cell precursors were isolated from the somatic cell communication compartment early on.

Fig. 1
figure 1

Structures of a caged and cell permeable coumarin, NPE-HCCC2/AM; its intracellular hydrolysis product, NPE-HCCC2; and its photolyzed product, HCCC2

Full size image

The LAMP assay possesses several salient features that are desirable for assaying gap junction coupling in live cells. First, because NPE-HCCC2/AM is cell membrane permeable, the method is applicable to fully intact populations including tissues. Second, the uncaging efficiency of NPE-HCCC2 is remarkably high, about two orders of magnitude higher than other caged fluorophores either by UV or two photon photolysis [9]. This extraordinarily high uncaging efficiency makes it possible to photolyze NPE-HCCC2 with a very low dose of UV light or two photon excitation light, hence minimizing phototoxicity or cell damage during uncaging. Together, the above two features makes LAMP assay a truly noninvasive imaging technique. Third, since HCCC2 is a photo-stable and bright fluorescent dye with high fluorescence quantum yield, it yields excellent signal for cell imaging. This in turn produces good signal to background ratio to facilitate quantification of intercellular dye transfer rate. Finally, HCCC2 dye emits blue light that spectrally complements many popular fluorescent sensors emitting at green, yellow or red. This facilitates multicolor imaging to allow imaging cell coupling and other biochemical events in cells concurrently. For instance, we demonstrated that Ca2+ influx through store operated Ca2+ channel potently inhibited cell coupling by combining the LAMP assay with Fluo-3 calcium imaging [7, 11].

Both the LAMP assay and infrared-LAMP assay have been successfully applied to cultured cell lines [7, 11], freshly isolated primary cells [14], or freshly dissected tissues [8]. We describe herein the LAMP assay and infrared-LAMP assay performed in cultured cells expressing Cx43 and GFP, providing detailed procedures for dye loading, local uncaging and imaging, as well as post-acquisition data analysis and quantification.

2 Materials

2.1 Cell Preparation

  1. 1.

    Cell lines: In addition to primary cells expressing endogenous connexins, cell lines, e.g., Hela cells (see Note 1 ) can also be used to study gap junction communication after transfecting these cells with plasmids containing connexin genes (see Note 2 ).

  2. 2.

    Plating cells: cells are seeded at about 30 % confluence on 3.5 cm imaging dishes containing a glass bottom (MatTek, see Note 3 ).

  3. 3.

    Cell transfection : A number of transfection reagents including TransIT-LT1 (Mirus Bio Corporation) can be used to transfect adherent cells. After transfection cultured cells for 24–36 h to allow sufficient expression of connexin proteins (see Note 4 ).

  4. 4.

    Cell incubator (37 °C ± 1 °C, 90 % ± 5 % humidity, 5 % ± 1 % CO2).

  5. 5.

    Cell culture medium: DMEM medium containing 10 % FBS and 1 % penicillin–streptomycin.

  6. 6.

    Trypsin–EDTA and DPBS.

2.2 Dye Loading Solution

  1. 1.

    10 % Pluronic stock solution: dissolve 50 mg pluronic in 0.5 mL anhydrous DMSO to prepare a stock solution. The resulting solution can be stored at 20–25 °C for a couple of months.

  2. 2.

    2 mM NPE-HCCC2/AM stock solution: Dissolve 1 mg NPE-HCCC2/AM in 670 μL anhydrous DMSO. Store the stock solution at −20 °C or leave it on ice during experiments (see Notes 5 and 6 ).

  3. 3.

    2 mM Fluo-3/AM or calcein /AM stock solutions: Dissolve 1 mg Fluo-3/AM or calcein /AM in anhydrous DMSO (440 μL or 500 μL respectively). Cells are typically loaded with 0.5 mM of calcein /AM, or with 1–2 mM of Fluo-3/AM (see Note 7 ).

  4. 4.

    1× HBS solution: 10× Hank’s Balanced Salt Solution (Gibco) is diluted ten times with water. During dilution, Hepes buffer is added to a final concentration of 20 mM, and the pH is adjusted to 7.35 with NaOH or HCl. The solution is then sterilized by filtering through a 0.22 m cellulose filter, aliquoted in 50 mL tubes and stored at 4 °C.

  5. 5.

    DMSO, >99.8 %.

  6. 6.

    Pluronic F127, powder.

  7. 7.

    Plasmid DNA for connexins.

  8. 8.

    Mirus transfection reagent.

  9. 9.

    An inverted fluorescence microscope equipped with a field diaphragm, CCD camera, and excitation and emission filter wheels controlled by computer and imaging software. Two photon uncaging was performed on a Zeiss LSM510 imaging system equipped with a Chameleon-XR laser (Coherent) [12].

3 Methods

3.1 Preparation of Dye-Loading Solution

  1. 1.

    Mix 1.2 μL of the pluronic stock solution (10 %,) and 1.2 μL of NPE-HCCC2/AM in a 1.5 mL centrifuge tube.

  2. 2.

    Add 0.5 mL of HBS to the mixture and briefly vortex the solution to ensure thorough mixing.

  3. 3.

    If multiple LAMP assays need to be performed on a given day, a larger volume of loading solution can be prepared at once.

  4. 4.

    Once prepared, the loading solution should be stored on ice or at 4 °C and used on the same day (see Notes 8 and 9 ).

3.2 Dye Loading

  1. 1.

    Remove the culture medium from the imaging dish by using a disposable Pasteur pipette.

  2. 2.

    Wash the cells twice with 1× HBS solution

  3. 3.

    Remove the HBS solution from the imaging dish and add 0.5–1 mL of the preprepared dye-loading solution to the imaging dish (see Note 10 ).

  4. 4.

    Cover the imaging dish with a lid to minimize water evaporation.

  5. 5.

    Incubate cells in the loading solution in the dark for 20–45 min (see Note 11 ).

  6. 6.

    Remove the loading solution with a Pasteur pipette and rinse cells once with HBS solution.

  7. 7.

    Add 1.5 mL of HBS solution to the imaging dish.

  8. 8.

    Incubate the cells in the dark for 10 min to allow complete hydrolysis of AM ester and to trap NPE-HCCC2.

3.3 Imaging Assays Using the LAMP Assay

  1. 1.

    The imaging dish containing cells loaded with NPE-HCCC2 is placed on the stage of a microscope.

  2. 2.

    Bring cells into focus by observing cellular fluorescence of a fluorescent marker, e.g., GFP, calcein , or Fluo-3 (excitation 490 ± 10 nm, emission 525 nm ± 10 nm).

  3. 3.

    Once a fluorescent cell pair in close contact is identified, move the stage to center the cell pair in the field of view (Fig. 2a, b).

    Fig. 2
    figure 2

    LAMP assay of gap junctional communication. (a–c) DIC (a) and GFP images (b–c, Excitation 490 ± 10 nm, Emission 525 ± 10 nm) with the iris of the field diaphragm fully open (b) or minimized (c). The local uncaging area (outlined by the dashed yellow circle in c) covered a portion of the donor cell and was away from the cell–cell interface. Hela cells were transfected with plasmid containing Cx43-IRES-GFP construct 24 h earlier. Two cells in the field of view were transfected and expressed Cx43 and GFP. (d) Coumarin image (Excitation 425 ± 5 nm, Emission 460 ± 10 nm) taken at the end of the experiment after global uncaging showing all cells were loaded with NPE-HCCC2. Two regions of interests (ROIs) representing the donor cell (d) and recipient cell (R) are outlined by dashed yellow circles. (e) Time course of fluorescence intensity of HCCC2 (FHCCC2, arbitrary units) in donor (filled triangle) and recipient cells (open circle). F 0 denotes FHCCC2 immediately after uncaging, and F e represents FHCCC2 when dye transfer reaches equilibrium. Intensity values between F 0 and F e are FHCCC2 at different time points of dye transfer (F t ). (f) Quantification of cell–cell dye transfer kinetics by fitting the first half of the dye transfer data from e. The slope of the fitted line gave the rate of dye transfer . Adapted from Ref. [12]

    Full size image
  4. 4.

    To perform local uncaging in a single cell among a coupled cell pair, place a field diaphragm in the excitation light path to control the diameter of excitation beam.

  5. 5.

    Reduce the field diaphragm to a minimum, such that only a portion of the chosen donor cell is covered by the excitation beam. The recipient cell of the coupled cell pair should be away from the reduced excitation beam (Fig. 2c).

  6. 6.

    Make fine adjustments to position the coupled cell pair with respect to the reduced excitation light beam. The goal is to selectively uncage the donor cell by UV excitation while avoiding photolyzing the recipient cell. Ensure that the cell–cell contact region between the donor cell and the recipient cell stays clear from the reduced uncaging beam (Fig. 2a–c).

  7. 7.

    After adjusting the stage position, open the iris of the field diaphragm and start image acquisition. Acquire two-color image pairs (blue/coumarin channel: Excitation 425 ± 5 nm, Emission 460 nm ± 10 nm; and green/GFP channel: Excitation 490 ± 10 nm, Emission 525 nm ± 10 nm) every 6–10 s for about a minute. We typically set the exposure time for each channel to 50–200 ms. This provides a time course for the baseline cellular fluorescence, which should remain fairly stable.

  8. 8.

    Prior to local uncaging of the donor cell, increase the image acquisition frequency to about one image pair every 2 s. Reduce the iris of the field diaphragm to a minimum to illuminate only a portion of the donor cell as you set up earlier (Fig. 2c). Viewing from the green channel, reconfirm that the reduced light beam still only targets the donor cell.

  9. 9.

    With the iris of the field diaphragm closed, the donor cell is locally uncaged with a short pulse of UV light (360 ± 20 nm) lasting for 0.1–1 s (see Note 12 ).

  10. 10.

    Open the iris of the field diaphragm immediately to continue image acquisition.

  11. 11.

    Depending on the cell coupling strength or the rate of dye transfer , adjust the acquisition frequency by taking a dual-color image pair every 2–10 s.

  12. 12.

    Continue monitoring NPE-HCCC2 transfer from the donor cell to the recipient cell until the equilibrium is reached. At this time the cellular coumarin fluorescence intensities in both the donor and recipient cells come close to each other and remain fairly stable over time (Fig. 2e).

  13. 13.

    Additional uncaging episodes can be performed by repeating steps 8–12.

  14. 14.

    The duration of UV exposure should be increased in subsequent uncaging to compensate for the consumption of caged coumarin from the previous photolysis.

  15. 15.

    Before ending the experiment, perform a global uncaging with the field diaphragm fully open to photolyze caged coumarin in all cells. Cells in the field of view should display intense blue coumarin fluorescence (see Note 13 ).

3.4 Imaging Assays Using the Two Photon Uncaging and Imaging (Infrared-LAMP Assay)

  1. 1.

    The imaging dish containing cells loaded with NPE-HCCC2 is placed on the stage of a microscope equipped with a two-photon laser (e.g., Zeiss LSM510, or LSM 780 or other equivalent laser scanning imaging systems).

  2. 2.

    Bring cells into focus by observing cellular fluorescence using a fluorescent marker (e.g., GFP, calcein , or Fluo-3) and confocal imaging (Excitation 488 nm laser, Emission 525 nm ± 20 nm, see Note 14 ).

  3. 3.

    Similar to the LAMP assay, move the stage to position the coupled cell pair in the center or the field of view.

  4. 4.

    Under 488 nm laser excitation, acquire a z-stack of confocal GFP images.

  5. 5.

    After choosing a donor cell, adjust the focus drive to a z-plane across the middle height of the chosen donor cell. Take a single confocal image at this height.

  6. 6.

    Using LSM510 imaging software, under the “Bleach Control” module specify two photon uncaging laser wavelength (normally between 730 and 760 nm), laser power (~10 mW), uncaging repetitions (20–40) and define the two-photon uncaging area from the confocal image taken above.

  7. 7.

    Use “Define Region” function to draw a circle along the cell periphery of the donor cell (see Note 15 ).

  8. 8.

    Start taking time lapse two photon images of cell coumarin fluorescence by exciting the cells at 820 or 830 nm. Take an image every 6–10 s for about a minute. This provides a time course for the baseline cellular fluorescence, which should remain fairly stable.

  9. 9.

    Initiate two-photon uncaging at the predefined donor cell by activating the “Bleach” button. The uncaging laser (760 nm or below) scans the defined uncaging area repeatedly 20–40 times in a total of a few seconds.

  10. 10.

    The laser excitation wavelength is changed from the uncaging wavelength (760 nm or shorter) to the imaging wavelength (820 or 830 nm) as soon as photo uncaging is finished (see Note 16 ).

  11. 11.

    Continue two photon imaging of coumarin fluorescence to monitor dye transfer from the donor to the recipient cell (see Note 17 ).

  12. 12.

    To follow dye diffusion in 3D in dissected tissues or in organotypic cultures, set up a z-stack to center the donor cell along the z-axis.

  13. 13.

    Sample 20–40 z-slices spaced ~2 μm apart either below or above the donor cell. This provides dynamic information of dye transfer from the donor cell to neighboring recipient cells in 3D.

  14. 14.

    Since NPE-HCCC2 can be loaded into cells to fairly high concentrations, additional episodes of dye transfer between coupled cells can be generated by repeating two-photon uncaging in a cell of a coupled cell pair.

  15. 15.

    Between each uncaging episode, cells can be treated with pharmacological or biochemical agents to alter the coupling strength. Their effects on gap junction coupling can then be assessed by comparing the rates of dye transfer before and after drug treatment [7, 12].

3.5 Data Analysis of Dye Transfer Between a Coupled Cell Pair

  1. 1.

    Fick’s equation is applied to quantify rates of intercellular coumarin dye transfer . Fick’s equation describes kinetics of molecular movement between two compartments separated by a membrane: ln[(F e −F t )/((F e −F 0)] = −kt, where F e , F 0, and F t are cellular NPE-HCCC2 signal at equilibrium, zero time and time t, respectively.

  2. 2.

    To quantify cellular NPE-HCCC2 signal, draw separate regions of interest (ROI) above the bulk cytoplasm of the donor cell and the recipient cell to analyze cellular NPE-HCCC2 fluorescence intensity (Fig. 2d). Many imaging software, including OpenLab and ImageJ, can be used to perform such post-acquisition analysis.

  3. 3.

    Plot the measured fluorescence intensities of donor and recipient cells against time. Determine F 0 and F e from the time course of NPE-HCCC2 signal (Fig. 2e).

  4. 4.

    Replot the data according to Fick’s equation (Fig. 2f). Fit the data using the linear equation to obtain the kinetic constant of dye transfer , k (in units of sec-1) (see Note 18 ).

4 Notes

  1. 1.

    Primary cells frequently express more than one type of connexins. To study a specific connexin protein, cultured cell lines with very low or undetectable connexin expression can be used, once they are infected with a connexin gene of interest. Cultured cell lines also offer the convenience of easy accessibility and availability.

  2. 2.

    Many plasmids containing connexin genes are available from the open plasmid repository Addgene (https://www.addgene.org/). We have applied LAMP assay in Hela cells infected with plasmids containing Cx43, Cx43-eGFP, Cx43-IRES-eGFP, or Cx26-IRES-DsRed [11].

  3. 3.

    Adhesion of cells to the glass surface usually takes at least 6 h. At low seeding confluence, it is easy to form small cell clusters containing just a few cells (doublets).

  4. 4.

    In this protocol, we used Hela cells expressing Cx43-IRES-eGFP.

  5. 5.

    For long-term storage, make 20–50 μL aliquots of the stock solution and wrap them with aluminum foil to minimize light exposure.

  6. 6.

    NPE-HCCC2/AM is available from VitalQuan LLC or can be synthesized as previously described [9].

  7. 7.

    Calcein normally gives rise to fairly bright fluorescence in cells, so less calcein is needed for cell loading.

  8. 8.

    To aid cell visualization and to assist focusing during fluorescence imaging, a membrane-permeable fluorescent dye such as Fluo-3/AM or calcein /AM can be added to the loading solution. These dyes emit green fluorescence so they will not interfere with imaging blue coumarin signal.

  9. 9.

    Pluronic is a mild detergent that helps to solubilize hydrophobic compounds in aqueous solutions. It greatly improves the cell loading efficiency of AM esters or lipophilic fluorescent probes. The pluronic stock solution may turn cloudy during storage but it can be clarified by warming it at 37 °C or higher temperature for several minutes.

  10. 10.

    Make sure that cells on the center glass coverslip are covered by the dye-loading solution. If necessary, cells on the plastic surface and away from the center glass can be wiped off with a piece of folded Kimwipes. The dried plastic surface will help to retain the loading solution to the center glass.

  11. 11.

    When the confluence of cells is high (>60 %), longer incubation may be required to load sufficient amount of caged coumarin into cells. Alternatively, higher concentration (>2 mM) and/or larger volume of stock solution of NPE-HCCC2/AM can be used to load more cells.

  12. 12.

    The optimal duration of UV photolysis varies with the UV light intensity, or the amount of caged coumarin loaded into cells. Ideally, UV uncaging should produce a sizable increase of coumarin fluorescence that is at least three times above the baseline level. Excess photolysis may generate too strong a coumarin signal that saturates the detector, which should be avoided.

  13. 13.

    If coumarin fluorescence appears to be too weak or too strong after complete photolysis of NPE-HCCC2, the amount of NPE-HCCC2 loaded into cells needs to be adjusted accordingly by, for example, changing the concentration of NPE-HCCC2/AM in the loading solution.

  14. 14.

    Fluo-3 fluorescence tends to be fairly dim at resting cellular Ca2+ level. Signal of calcein or GFP is much stronger.

  15. 15.

    The defined uncaging area should be restricted within the donor cell to avoid photolyzing neighboring cells.

  16. 16.

    The old LSM510 imaging software usually takes about 30 s to tune the laser before showing that the laser is mode-locked at the new wavelength. Newer versions of imaging system (LSM780 or more recent ones) tune the laser much faster.

  17. 17.

    The time interval of image acquisition is typically set to be 10–20 s since cell–cell dye transfer usually taking several minutes to reach equilibrium.

  18. 18.

    We normally only fit the first half of dye transfer data because changes in fluorescence intensity over time become smaller and noisier as the dye transfer approaches equilibrium (Fig. 2f).