Ratiometric Imaging Needs a Specialized Setup

December 12, 2011

Ratiometric imaging is widely used to study highly dynamic intracellular ion, voltage or pH changes. The most common application, however, is calcium imaging. Ratiometric imaging is also used for investigating cellular networks, where e.g. relative calcium concentrations are passed among cells or different cell types dynamically change ion, voltage or pH levels. Also FRET assays can considerably benefit from ratiometric imaging, as the signal-to-noise-ratio is greatly improved.

Typical applications of ratiometric imaging

Ratiometric imaging is widely used to study highly dynamic intracellular ion, voltage or pH changes. The most common application, however, is calcium imaging. Calcium dynamics are of special interest for many researchers due to its role as one of the most important 2nd messengers, where it regulates a wide range of cellular functions e.g. synaptic transmission (exocytosis), muscle contraction, apoptosis, and as a cofactor for many enzymes. Calcium fulfills these functions in a highly specific, highly dynamic and sometimes also spatially restricted manner. Minute changes of intracellular calcium concentrations can have a profound effect on the cellular state. Therefore, measuring absolute calcium concentrations (or generally ion concentrations) is of huge interest for cell biologists. Additionally, it has to be kept in mind that many ion, voltage or pH changes are not uniform throughout a cell and have different functions in different cell compartments. In such investigations, the absolute or relative ion concentrations in the different parts of the cell are of interest for researchers. Calcium concentrations in morphologically different areas of a cell can only be reliably compared using ratiometric imaging. Ratiometric imaging is also used for investigating cellular networks, where e.g. relative calcium concentrations are passed among cells or different cell types dynamically change ion, voltage or pH levels. To understand network activity it is mandatory to know about absolute and relative changes in ion, voltage or pH levels. Again, only ratiometric imaging methods deliver reliable quantitative and comparable results. Last but not least, it should also be mentioned that FRET assays can considerably benefit from ratiometric imaging, as the signal-to-noise-ratio is greatly improved.

Special fluorophores are the key to ratiometric imaging

To perform ratiometric imaging, certain prerequisites have to be fulfilled concerning the imaging setup. First of all, suitable fluorophores or fluorophore combinations have to be selected. These might be synthetic dyes or protein constructs. There are several well described dyes designed for ratiometric imaging, as they exhibit an emission shift upon ion binding, voltage change or pH change. There are two ways of detecting this shift: Firstly, dual excitation dyes that are alternately excited by light with two independent wavelengths and the emission is detected at a single wavelength (e.g. fura-2, Figure 1a). Secondly, dyes that are excited with light of one wavelength and the emission is detected at two independent wavelengths (e.g. indo-1, Figure 1b). In principle, it is also possible to simultaneously load cells with two different fluorophores which respond with an intensity change instead of an emission shift (e.g. fluo-3 and fura red, Figure 1c). It is also described that cells can be co-loaded with ion-, voltage- or pH-insensitive fluorophores, which then act as reference dyes. An advantage of combining dyes is that dyes with longer excitation wavelengths can be used. This usually causes less harm to the cells than using ratiometric dyes that are excited with UV- or near UV-light (e.g. Fura-2), as light at visible wavelengths is less phototoxic. The drawback of this approach is that the cells have to be loaded with two dyes. However, it is important that the combined fluorophores have a similar emission or excitation wavelength. Finally, many protein constructs designed for FRET assays are also suitable for ratiometric imaging.

Fig. 1: Spectra of some calcium-sensitive dyes or dye combinations. (A) Emission of fura-2 at 510 nm for different excitation wavelengths (usually dually excited with 340 nm and 380 nm light wavelength). Depending on the calcium concentration the fluorescence intensity at 340 nm will increase, whereas it will decrease at 380 nm. (B) Emission spectrum of indo-1 excited with 338 nm. The emission intensity will increase around 400 nm and decrease around 500 nm. (C) Combination of fluo-3 and fura red excited at 488 nm. The intensity of fluo-3 will increase around 520 nm, whereas the intensity of fura red will decrease around 660 nm, allowing ratiometric imaging. (D) Emission spectrum of the non-ratiometric dye calcium green excited at 488 nm. Light intensity around 510 nm increases upon calcium binding. Ratiometric imaging is not possible. Source: The Molecular Probes® Handbook – Online Edition.

Chemical dyes, dye combinations and protein constructs

The following lists provide some described chemical dyes or dye combinations and protein constructs that meet the requirements of ratiometric imaging:

Tab. 1: Fluorophores/fluorophore combinations for ratiometric calcium imaging

Name of fluorophore

Kd (µM)

Excitation (nm)

Emission (nm)

What can be imaged?

Type

fura-2

0.224

340/380

510

Calcium

synthetic

indo-1

0.250

350

405/485

Calcium

synthetic

BTC

7

465/400

530

Calcium

synthetic

Premo Cameleon Calcium Sensor

0.240

435

480/540

Calcium

protein

fluo-3/Fura Red

0.390/1.6

488

530/670

Calcium

synthetic

Calcium Green-1/Lucifer yellow

0.19/-

420/488

515

Calcium

synthetic

 

Tab. 2: Fluorophores for ratiometric ion imaging

Name of fluorophore

Kd (mM)

Excitation (nm)

Emission (nm)

What can be imaged?

Type

mag-fura-2

1.9

340/380

500

Magnesium

synthetic

mag-indo-1

2.7

340

420/480

Magnesium

synthetic

SBFI

11.3

340/380

510

Sodium

synthetic

PBFI

44

340/380

510

Potassium

synthetic

 

Tab. 3: Fluorophores for ratiometric pH imaging

Name

pH range

Excitation (nm)

Emission (nm)

Type

LysoSensor™ Yellow/Blue

alternatively

3.5-6

340/400

365

520

450/510

synthetic

BCECF

<7

440/490

530

synthetic

SNARF-5F

6.8–7.4

488 or 514

580/640

synthetic

SNARF-1

7-8

488 or 514

580/640

synthetic

 

Tab. 4: Fluorophores for ratiometric voltage imaging

Name

Excitation (nm)

Emission (nm)

Type

di-4-ANEPPS/

440/505

630

synthetic

di-8-ANEPPS

(alternatively)

475

488

560/620

540/610

synthetic

 

All excitation and emission wavelengths in Tables 1–4 are approximations. The Kd values shown can be different, heavily depending on experiment conditions (in vitro/in situ, buffer solutions) and individual crosstalk with other ions and their respective concentration. For detailed spectra and Kd values see: The Molecular Probes® Handbook – 11th Edition.

A specialized setup is a prerequisite for ratiometric imaging

For ratiometric imaging it is essential that the time between the acquisition of the two gray scale images needed for ratio calculation is as small as possible. To fulfill this requirement, a short excitation duration, fast image acquisition and a fast switch of excitation or detection wavelengths is mandatory. To make excitation duration as short as possible, a powerful light source (e.g. strong DC lamps (>200 W), lasers) and optics with high emission for the desired wavelengths are recommended. A fast switch between the excitation/emission wavelengths can be achieved by employing fast filter wheels in combination with a strong light source, which manage to switch excitation/emission wavelengths within milliseconds. Also, monochromators or lasers combined with an AOTF (acousto-optic tunable filter) are suitable for a fast switch between excitation wavelengths. Last but not least, a good CCD camera (if possible EM-CCD) that combines reasonable resolution and high sensitivity to enable fast image acquisition is a prerequisite for up-to-date ratio imaging experiments. A high frame rate in ratiometric imaging per se is a clear advantage, as many of the processes that are investigated are highly dynamic and conditions can change within milliseconds.

Factors influencing detected fluorescence light intensity

Concentration of a fluorophore
The concentration of a fluorophore (c) inside a cell has a massive impact on emitted fluorescence light intensity. One can easily imagine that the more fluorophore is inside a cell, the more light is emitted. Unfortunately, the fluorophore concentration in the cell can hardly be controlled. The membrane passage of synthetic dyes (e.g. Fura-2) can be influenced via incubation length and the concentration of the fluorophore outside of the cell. However, the amount of dye that finally gets into the cell strongly depends on the type of preparation (e.g. slice preparation), cell type or cell line. Generally, dye penetration into cells in acute slices is worse than in cell lines, primary cultures or isolated single cells. Additionally, individual cells can take up dyes in very different amounts (depending on e.g. health state, metabolic state etc.). When using protein fluorophores instead of synthetic dyes, the amount of fluorophore in the cytosol depends on the transcription/translation rate and the protein turnover. The transcription and translation rate of a protein can be controlled via the usage of specific promotors and by insertion of specific DNA or amino acid sequences. However, this gives the experimenter only limited control, as the actual rate of protein depends on numerous factors (e.g. a cell’s health state, age, cell type). Again, within one cell culture or even one single cell different transcription/translation rates, transport rates and protein turnover rates can occur. These facts render an estimation of actual fluorophore amounts in a cell or a part of a cell almost impossible.

Diameter of the specimen
Another issue in conventional fluorescence imaging is differences in the diameter of the specimen (d) in z-axis. Specimen diameter can range from some nanometers (parts of single cells, e.g. dendrites, axons, sperm flagellum), via some microns (cell monolayers, soma of single cells) to several hundreds of microns (e.g. in acute slice preparations). In fluorescence imaging all dyes in the light path are excited and detected (exceptions: confocal microscopy, TIRF), resulting in much higher detected fluorescence light intensity in “thick” parts of the specimen compared to other parts. This can easily lead to misinterpretations, as thinner structures, which might be not as “bright” as other parts of the cell, might have a much higher concentration of the ion (e.g. calcium) that is investigated. Nevertheless, these parts might appear much less bright than the amount of fluorophore in the light path and therefore the detected light intensity is lower.

Optical constants
Last but not least, optical constants (K) in different specimens (cell types; slice preparation, cell cultures, single cells etc.) and imaging setups vary. Different cell types can have different refractive indices and transmission, mainly depending on their morphology. On the setup side, different manufacturers have different transmission in their optical components (dichroic mirrors, filters, objectives etc.). Additionally, different light sources have specific light emission intensities (e.g. laser, xenon arc lamps, mercury arc lamps), which has a massive impact on the efficiency of fluorophore excitation and therefore the amount of light emitted by the fluorophore. Finally, there are also huge differences in the detection of the emitted light, depending on whether e.g. a conventional CCD camera, an EM-CCD camera or a photomultiplier is used. Moreover, there are also notable differences in photon sensitivity between manufacturers and also between the models that are offered by these manufacturers. It should also be mentioned that individual differences can occur even within the same model of e.g. an objective or a camera by the same manufacturer. All these circumstances make every single imaging setup slightly different and give every single imaging setup its own optical constant (K). This again makes estimations of the amount of e.g. calcium ions in a cell impossible by simply measuring mere light intensity.

Comments