Chapter 5 - Measuring lysosomal pH by fluorescence microscopy

https://doi.org/10.1016/bs.mcb.2014.10.021Get rights and content

Abstract

The technique of dual-wavelength ratio fluorescence microscopy provides a powerful tool to measure organellar pH. Unlike single-wavelength measurements, this method is unaffected by changes in focal plane, dye volume, and fluorophore bleaching, providing a quantitative and dynamic readout of the pH of subcellular compartments. This chapter describes the application of dual-wavelength ratio fluorescence microscopy to the measurement of lysosomal pH, highlighting its advantages and limitations. Probe selection, calibration methods, and salient aspects of the required hardware are discussed in detail.

Introduction

Lysosomes are involved in functions as diverse as the degradation of material delivered via the endocytic and autophagic pathways, provision of membranes for wound repair, formation and maturation of phagosomes, and the generation of immunogenic peptides for presentation by antigen-presenting cells (Huynh et al., 2004, Platt et al., 2010, Samie et al., 2013, Settembre et al., 2013). A feature that is characteristic of lysosomes and is central to their function is their uniquely acidic luminal pH. The highly acidic (pH  5) lumen of lysosomes promotes the activity of various hydrolytic enzymes, thus endowing the organelle with a potent degradative capacity. In addition, the proton-motive force associated with the luminal acidification is harnessed by intrinsic transporters to drive the transfer and recycling of degraded material across the lysosome membrane (Coffey and De Duve, 1968, Gunshin et al., 1997, Tabuchi et al., 2000). As such, the dissipation of the luminal pH has numerous adverse effects on lysosome function, highlighting the importance of maintaining the acidity of this compartment.

The pH gradient across the lysosomal membrane is generated by the vacuolar-type ATPase (V-ATPase), a multimeric pump that transports protons from the cytosol into the lysosomal lumen using energy derived from the hydrolysis of cytosolic ATP. The translocation of protons across the lysosomal membrane is an electrogenic process and, in order to prevent the buildup of an inhibitory electrical potential (lumen-positive), counter-ion conductive pathways exist to dissipate the voltage generated by the V-ATPase (Cuppoletti et al., 1987, Dell'Antone, 1979, Graves et al., 2008, Harikumar and Reeves, 1983, Ohkuma et al., 1983, Ohkuma et al., 1982, Steinberg et al., 2010, Van Dyke, 1993). The necessary charge compensation can be provided by the inward movement of anions (mainly Cl), or by efflux of cations (mainly K+ and Na+) (Steinberg et al., 2010).

Historically, studies addressing the acidic nature of lysosomes relied initially on the preferential partitioning of membrane-permeant basic probes that become protonated and hence trapped in acidic compartments. This approach was pioneered by de Duve and colleagues, who demonstrated that weak bases such as chloroquine accumulate in lysosomes (de Duve et al., 1974). The realization that weak bases accumulate in acidic compartments prompted the use of fluorescent organic bases, such as acridine orange, to detect organellar acidification by optical means (Allison & Young, 1964). This principle is still widely used and is the basis of the popular LysoTracker® probes. The same principle was also extended to identify acidic organelles by electron microscopy. Like the optical probes, the basic form of dinitrophenol, 3-(2,4-dinitroanilino)-3’-amino-N-methyldipropylamine (DAMP) concentrates in acidic compartments, where it can be fixed and immuno-labeled using monoclonal antibodies directed to the dinitroarene moiety. Gold labeling of the primary antibodies using protein A or secondary antibodies, enables identification and detailed analysis of acidic organelles at the ultra-structural level (Anderson et al., 1984, Anderson and Orci, 1988). Although the application of weak bases has proven useful for the visualization of acidic compartments within cells, these techniques remain largely qualitative, suffer from a lack of organellar specificity, and can exhibit adverse effects on cell physiology even at subtoxic concentrations (Palmgren, 1991, Tsien, 1989a).

A more quantitative—and therefore more accurate—estimation of lysosomal pH can be obtained by applying ratiometric fluorescence imaging techniques. This approach, first applied by Ohkuma and Poole for lysosomal pH measurements (Ohkuma & Poole, 1978), involves the use of pH-sensitive fluorophores attached to large biomolecules that are targeted to lysosomes using the cells' own endocytic pathway. A commonly used pH-sensitive fluorophore is fluorescein. When excited at 490 nm, its peak wavelength, the fluorescence emitted by fluorescein is exquisitely pH-sensitive. Importantly, the fluorescence change is not identical at all excitation wavelengths; for instance, at 440 nm fluorescein is much less pH-sensitive (Figure 1(A) and (B)). This property, which is shared by most fluorescein derivatives, allows for dual-excitation ratio fluorescence imaging, which is insensitive to changes in the focal plane (Figure 1(E)), photobleaching (Figure 1(D)), and the amount of fluorophore, but is exquisitely sensitive to changes in pH near the pKa of the fluorophore (6.4 in the case of fluorescein) (Figure 1(B)). The insensitivity to alterations in focal plane, bleaching, changes in cell thickness, leakage of dye, etc. stems from the fact that these variables affect the fluorescence at all wavelengths equally; calculation of the ratio of the two wavelengths corrects automatically for changes introduced by these variables.

The ratio of the fluorescence measured at λex = 490/λex = 440 can be converted to absolute pH values by performing an in situ calibration; this requires the organellar pH to be clamped at various known pH values, which is often accomplished using ionophores. Nigericin, monensin, and similar ionophoric antibiotics exchange protons for monovalent cations and, by setting the concentration of the latter, it is possible to ensure that the luminal pH of organelles attains a value that closely approximates the cytosolic pH, which in turn approximates that of the extracellular medium.

Fluorescence ratio can be measured at the population, single cell or subcellular level. Macroscopic techniques, such as spectrofluorometric analysis (Ohkuma & Poole, 1978) are useful for assessing the average pH of all the lysosomes in a population of cells, but do not allow for the resolution of heterogeneity among individual lysosomes or even between individual cells. Heterogeneity among cells can be assessed by flow cytometry (Marchetti et al., 2009, Murphy et al., 1982a, Murphy et al., 1982b), but cytometers are rarely endowed with the necessary laser lines to perform ratio measurements and, besides, cannot resolve individual organelles within each cell. For these reasons, the protocol described herein focuses on dual-excitation ratio fluorescence microscopy using fluorescein-labeled dextran loaded into lysosomes, which allows not only for the resolution of individual cells, but also of individual lysosomes (Carraro-Lacroix et al., 2011, Steinberg and Grinstein, 2007, Steinberg et al., 2010).

The hardware necessary for the fluorescence ratio imaging techniques described in this protocol is outlined in Figure 2. Use of an inverted microscope is recommended, to facilitate access to the sample while recording with high-resolution oil-immersion objectives. An external high-intensity arc lamp (e.g. X-Cite® 120, EXFO Photonic Solutions Inc.) is required as a source of light and is an important consideration for ratiometric imaging. Two commonly used light sources are mercury arc lamps and xenon arc lamps (often abbreviated as HBO and XBO, respectively). The output from mercury lamps is characterized by sharp and intense emission peaks across the UV–visible spectrum, whereas xenon lamps provide a more even output spectrum (Webb & Brown, 2013). As such, mercury lamps are only suited to dual-excitation ratio imaging if the two wavelengths required for the fluorophore of choice are properly excited by the peaks of the lamp. In addition to having a more even spectrum, xenon lamps have more stable emission over time, rendering them more suitable for quantitative and ratiometric imaging techniques. They are, however, much less powerful than mercury sources. For these reasons, some manufacturers have introduced light-emitting diode sources that have improved features.

Light is carried from the source via a fiber optic cable to 485 ± 10 nm and 438 ± 12 nm excitation filters mounted on a computer-controlled filter wheel (e.g. the Lambda 10-2 Optical Filter Changer, Sutter Instrument Company) that allows for the rapid transition between excitation wavelengths. Because the wheel is equipped with a shutter, the incident light can be blocked off between acquisitions to prevent excessive illumination that can result in bleaching of the fluorophore and phototoxic damage to the cells. The filtered excitation light is then directed to a dichroic mirror (505 nm) mounted on a filter cube (Chroma Technology Corp®) and forwarded to the sample. The emitted light is selected through a 535 ± 20 nm emission filter. The latter can be mounted onto the cube or contained in a second computer-controlled filter wheel. A filter wheel on the emission port enables dual-emission ratio measurements, which are recommended for pH-sensitive dyes like SNARF. The emitted light is detected by a back-illuminated electron-multiplied charge-coupled device camera (e.g. the Cascade II EMCCD camera, Photometrics®) that is thermoelectrically cooled to reduce thermal noise. The high sensitivity and large dynamic range of such cameras is optimally suited for quantitative ratiometric imaging of cells/organelles. When required, binning (electronically merging a cluster of pixels into a single pixel) can be used to improve the signal-to-noise ratio, at the expense of spatial resolution. The entire workstation, including the shutters and two filter wheels, is controlled by software such as the MetaFluor® Fluorescence Ratio Imaging Software (Molecular Devices).

A number of fluorescent indicators undergo large spectral changes when they bind protons and therefore function as useful pH probes. The emission intensity of the indicator at two selected wavelengths can be used to generate a ratio that, after subtraction of the background/autofluorescence, reflects the state of protonation of the indicator (Grynkiewicz et al., 1985, Tsien, 1989b, Tsien et al., 1985). An important first consideration when selecting a pH sensor is its pKa, which dictates the pH range where the probe will be most sensitive and accurate. The pH of the lysosomal lumen has been reported to range from 4.5 to ≈5 (Christensen et al., 2002, Coen et al., 2012, Lange et al., 2006, Poët et al., 2006, Tabeta et al., 2006, Trombetta et al., 2003); therefore, an indicator with a pKa in this range is ideal for lysosomal pH measurements. Two commonly used probes for measuring lysosomal pH are Oregon Green and fluorescein, which have pKa values of approximately 4.7 and 6.4, respectively. These probes, which can be used singly or in combination, have been shown to reliably report the lysosomal pH (Carraro-Lacroix et al., 2011, Coen et al., 2012, Haggie and Verkman, 2009a, Haggie and Verkman, 2009b, Steinberg et al., 2010). Both fluorescein and Oregon Green are amenable to dual wavelength ratiometric microscopy (Figure 1(A) and (B)), as they are not equally pH-sensitive at all wavelengths.

Another important consideration is the ability to target the probe to the lysosomal compartment. There are commercially available ratiometric dyes, such as 2-(4-pyridyl)-5-((4-(2-dimethylaminoethyl-aminocarbamoyl) methoxy)phenyl) oxazole (also known as LysoSensor™ Yellow/Blue DND-160) that, by virtue of being weak bases, partition into acidic compartments without the need for active uptake and targeting by the cell (Diwu, Chen, Zhang, Klaubert, & Haugland, 1999). These dyes are taken up rapidly, which is advantageous; however care should be taken in interpreting the results, because the dyes also label acidic compartments other than lysosomes (Bankers-Fulbright et al., 2004, Diwu et al., 1999, Zhang et al., 2012). Moreover, since they are weak bases, these dyes can exert alkalizing and osmotic effects as they accumulate in acidic compartments (Ohkuma and Poole, 1981, Poole and Ohkuma, 1981). In addition, acidotropic dyes like LysoSensor™ cannot be used for studies where lysosomal pH is manipulated; the dye will not be retained in lysosomes when the pH gradient is dissipated, as often occurs during determinations of buffering capacity or when performing calibrations.

A better approach involves linking the pH sensor covalently to large biomolecules that can be internalized by the cells by fluid-phase uptake and subsequently trafficking lysosomes via the endocytic pathway. Dextrans are commonly used for this purpose; they are commercially available, not destroyed by the degradative enzymes of the lysosome (Ohkuma & Poole, 1978), and do not induce excessive osmotic swelling, as single dextran molecules can bear multiple fluorophores, so that comparatively small amounts need to be loaded into the cells to provide a strong signal.

The two fluorescent signals used to generate the ratio need not originate from the same fluorophore. In recent studies, dextrans labeled with both a pH-sensitive fluorophore and a second, pH-insensitive fluorophore have been used. The ratio of the fluorescence emitted by the pH-sensitive probe to that emitted by the pH-insensitive probe is used as a readout of the pH. This approach can provide a larger dynamic range, because unlike the fluorescein derivatives where the reference wavelength is nevertheless somewhat sensitive to pH, a second dye chosen as reference can be entirely pH-insensitive or even change fluorescence in the opposite direction. However, this approach is prone to error, because differential photobleaching of the fluorophores is inevitable and will result in changes to the ratio that are not indicative of pH (Figure 1(F)).

Although obtaining fluorescence measurements at two different wavelengths and generating a ratio is indicative of relative differences in pH, ultimately a calibration of the signal must be performed to convert the fluorescence ratio into absolute pH values. A number of methods exist for calibration. The most straightforward procedure is to perform a calibration (i.e. establish the relationship between fluorescence ratio and the pH of the surrounding medium) of the free dye in buffers of varying pH, in vitro. However, this approach assumes that the dye behaves identically in the cellular compartment of interest as it does in solution in vitro; this is often not the case, due to interactions of the fluorophore with cellular constituents and/or with itself in the confined space of the organellar lumen. A preferred method of calibration involves the use of the K+/H+ antiporter nigericin to effectively “clamp” the pH at known values and to calibrate in situ (Thomas, Buchsbaum, Zimniak, & Racker, 1979). This technique involves bathing the cells in a high [K+] buffer (containing approximately the same concentration of K+ as the cytosol and presumably also similar to that of the organellar lumen) titrated to the desired pH and containing nigericin. This allows for the rapid equilibration of the pH across biological membranes. By determining the fluorescence ratio in buffers of various known pH values, a calibration curve can be generated that can then be used to convert experimental fluorescence ratios into absolute pH values. This technique is advantageous in that it is relatively quick, and makes no assumptions about the behavior of the dye under different physiological environments, as the calibration is performed in situ. However, it is worth mentioning that there are caveats to this approach: the cytosolic and the luminal [K+] of the lysosome (or other organelle of choice) are assumed to be 140 mM. This estimate is likely to approximate the concentration in the cytosol of most (though not all) mammalian cells, but the [K+] inside lysosomes has only been measured in few instances. Moreover, the responsiveness of individual cell types to nigericin is known to vary (Chow, Hedley, & Tannock, 1996). For these reasons, discrepancies in the absolute pH values obtained using the high K+/nigericin technique have been reported (Boyarsky et al., 1996a, Boyarsky et al., 1996b).

Perhaps the most accurate method for calibrating involves the use of varying ratios of weak acids and weak bases to determine the absolute pH value. As discussed earlier, small weak bases and acids move freely across cellular membranes, producing an alkalosis or acidification, respectively. The magnitude of the change depends not only on the concentration and pK of the weak electrolytes, but also on the pH of the target compartment. At the appropriate ratio of concentrations, a combination of a defined weak acid and weak base can produce equal and opposite changes, resulting in no net change in pH and hence in the fluorescence ratio. This point is referred to as the “null point” (Eisner et al., 1989, Szatkowski and Thomas, 1986). By exposing cells to buffers containing a range of weak acid to weak base ratios, the null point can be found or interpolated. The pH of the compartment can then be determined using the equation:pHx=pHe0.5log[(Aτ)/(Bτ)]where pHx is the pH of the compartment containing the pH sensor, pHe is the extracellular pH, (Aτ) is the concentration of the weak acid, and (Bτ) is the concentration of the weak base (Chow et al., 1996). Importantly and unlike the high K+/nigericin clamping technique, this approach makes no assumptions about the ion composition of the compartment being assessed and has been used to measure both cytosolic pH (Boyarsky et al., 1996a, Chow et al., 1996) as well as the pH of other internal compartments (Schapiro et al., 2000). It must be borne in mind that the null point method assumes that only the uncharged form of the weak acid and weak base move across the membrane (or that they do so very much faster than the charged species), that the pKa and the pKb of the weak acid and weak base are the same outside and inside the compartment of interest, and that cellular pH-regulating mechanisms do not interfere with the null point determination (Chow et al., 1996). Clearly, the method of calibration chosen requires careful consideration of the nature of the compartment being assessed and of the inherent limitations of each procedure.

Section snippets

Cell Lines

  • RAW264.7 murine macrophage-like cells (ATCC® TIB-71™)

  • Cells are grown in RPMI-1640 medium (Wisent Bioproducts) containing l-glutamine, bicarbonate-buffered and supplemented with 5% heat-inactivated fetal bovine serum (Wisent Bioproducts)

Reagents

  • Fluorescein-dextran, 10,000 M.W. (Sigma–Aldrich®) dissolved in sterile 1X phosphate-buffered saline (Wisent Bioproducts) at a concentration of 25 mg/mL

  • Hank's balanced salt solution (HBSS) with calcium and magnesium, without phenol red (Wisent Bioproducts)

  • 1X

Loading Cells with Fluorescein-Dextran

This protocol takes advantage of the high constitutive endocytic activity of RAW264.7 cells, but is applicable to other cell types as well. Incubating the cells in medium containing fluorescein-dextran allows for its uptake into early endosomes, along with the fluid phase. The excess extracellular dextran is then washed away and the internalized dextran chased to the lysosomal compartment. A sufficient chase time is a key aspect of this assay, as imaging too soon after loading may result in

Conclusion

Dual-wavelength ratio imaging provides a powerful tool for assessing pH and provides considerable advantages over macroscopic whole-population techniques by allowing for the resolution of individual cells and organelles. By using the appropriate hardware, pH indicator, calibration procedure, and overall experimental setup it is possible to obtain accurate and reproducible results. We believe that this approach, which is true and tested, will continue to provide valuable information on the role

Acknowledgments

J.C. is supported by a Cystic Fibrosis Canada postdoctoral fellowship. Research in the authors' laboratory is supported by the Canadian Institutes for Health Research grants MOP7075, MOP102474 and MOP4665.

References (51)

  • M.G. Palmgren

    Acridine orange as a probe for measuring pH gradients across membranes: mechanism and limitations

    Analytical Biochemistry

    (1991)
  • M. Samie et al.

    A TRP channel in the lysosome regulates large particle phagocytosis via focal exocytosis

    Developmental Cell

    (2013)
  • M. Tabuchi et al.

    Human NRAMP2/DMT1, which mediates iron transport across endosomal membranes, is localized to late endosomes and lysosomes in HEp-2 cells

    Journal of Biological Chemistry

    (2000)
  • R.Y. Tsien

    Fluorescent indicators of ion concentrations

    Methods in Cell Biology

    (1989)
  • R.Y. Tsien et al.

    Measurement of cytosolic free Ca2+ in individual small cells using fluorescence microscopy with dual excitation wavelengths

    Cell Calcium

    (1985)
  • R.G. Anderson et al.

    Visualization of acidic organelles in intact cells by electron microscopy

    Proceedings of the National Academy of Sciences of the United States of America

    (1984)
  • R.G. Anderson et al.

    A view of acidic intracellular compartments

    Journal of Cell Biology

    (1988)
  • J.L. Bankers-Fulbright et al.

    Platelet-activating factor stimulates cytoplasmic alkalinization and granule acidification in human eosinophils

    Journal of Cell Science

    (2004)
  • G. Boyarsky et al.

    Inadequacy of high K+/nigericin for calibrating BCECF. I. Estimating steady-state intracellular pH

    American Journal of Physiology

    (1996)
  • G. Boyarsky et al.

    Inadequacy of high K+/nigericin for calibrating BCECF. II. Intracellular pH dependence of the correction

    American Journal of Physiology

    (1996)
  • L.R. Carraro-Lacroix et al.

    A weak base-generating system suitable for selective manipulation of lysosomal pH

    Traffic

    (2011)
  • S. Chow et al.

    Flow cytometric calibration of intracellular pH measurements in viable cells using mixtures of weak acids and bases

    Cytometry

    (1996)
  • K.A. Christensen et al.

    pH-dependent regulation of lysosomal calcium in macrophages

    Journal of Cell Science

    (2002)
  • K. Coen et al.

    Lysosomal calcium homeostasis defects, not proton pump defects, cause endo-lysosomal dysfunction in PSEN-deficient cells

    Journal of Cell Biology

    (2012)
  • D.A. Eisner et al.

    A novel method for absolute calibration of intracellular pH indicators

    Pflugers Archiv

    (1989)
  • Cited by (0)

    View full text