SPECIFIC DNA LABELLING


In oncology, the detection of the pathological cell is the most developed application. This detection is essentially based on the measurement of abnormal DNA content in the nucleus of the tumor cell. Many studies now use DNA measurement in cytometry: in plants, animals, parasitology...

Specific DNA Labeling

There are many DNA dyes available for FCM studies.

Hoechst is a benzimidazole derivative that emits blue fluorescence when excited in the UV. It has a high affinity for DNA and binds preferentially to A-T bases, but it is not an intercalator. DAPI has the same characteristics as Hoechst and can be interchanged with it.

Mithramycin, chromomycin are antibiotics excitable at 445nm and emit at 575 nm (orange). They bind to G-C regions by non-intercalating mechanisms.

Propidium iodide and ethidium bromide are the most commonly used dyes. These two dyes are intercalators of both DNA and RNA double strands. Both are excitable in the blue and emit in the red (615 nm).

There are many other molecules allowing to mark DNA with very different spectral characteristics allowing to use many different excitation sources like Sytox (blue, green, orange, red), Vybrant DyeCycle (purple, green, red)...

Some of them do not cross the membrane barrier like PI, DAPI and Sytox. This property can be exploited to differentiate living cells from dead cells since the latter have lost their membrane inequity. Living cells do not take up the dye while dead cells become fluorescent. It is also possible to permeabilize them so that the dye can access the DNA.

Other dyes, such as Vybrant, can diffuse through the cell membrane to label the DNA, allowing us to obtain information on the DNA content of living cells.

Stichiometry of labeling

It is important that the fluorescence emitted by the probe is proportional to its binding to the DNA. Otherwise it is not possible to measure an accurate DNA content.

Intercalators, although their binding is related to the accessibility of the DNA, are more used to measure DNA content because the content of A-T and G-C does not affect their binding.

Elimination of doublets

The reliability of the results relies on the possibility of not taking into account the cellular aggregates. Indeed, how to distinguish an aggregate of 2 or more cells since the quantity of DNA measured will be the multiple of the quantity of a single cell.

The method used is based on the possibility of analyzing the profile of the signal collected, its area, its height or its time of flight (TOF, width depending on the machine) (Figure 1).

Format Signal
Figure  1 : Signal profile


Using combinations of these parameters, it is possible to distinguish doublets from singlets (Figure 2). This methodology is also used for cell sorting to avoid contamination of the sorting by unwanted cells. Indeed, during the sorting, the cytometer sees the aggregate as a single cell, if this aggregate is composed of a negative cell and a positive cell, it will be seen as a positive cell. If this aggregate is sorted as such, there will be a negative cell sorted at the same time as the positive cell, resulting in a decrease in the purity of the sorted cells.


Elimination des doublets

Figure 2 : Principle of doublet elimination

Cell viability

The use of DNA markers that do not cross the membrane barrier makes it very easy to determine the percentage of living/dead cells. Dead cells with altered membranes that allow the dye to pass become fluorescent (Figure 3).

Figure 3 : Dead cells labeling by Sytox

When labeling complex populations, this type of labeling is important because dead cells very often show non-specific labels that can distort the results. In the case of cell sorting, it avoids sorting dead cells if the sorted population is to be put back into culture.

Evaluation of DNA quantity

By permeabilizing the cells, DNA labelling will allow to quantify the quantity present in the cell. Be careful however, double stranded RNA can also be marked by the intercalants and will be added to the measurement of the DNA, to avoid this it will be necessary to use a marking buffer degrading this RNA (RNase). It is also necessary that the accessibility of the DNA to the dye is optimized. There are many DNA marking protocols depending on the cell type, it is advisable to test one's own cells with several protocols so that the measurement is optimized.

This application is used in many fields such as agricultural research where it has been found, for example, that the amount of DNA in certain plants varies according to altitude or geographical location (the higher the altitude, the more DNA, gradient of DNA quantity between coffee trees in West Africa and East Africa) (Figure 4), and in fish farming, where it was observed that triploid males had better growth rates than their diploid counterparts, hence the search for rearing conditions to obtain a majority of triploid males (temperature, pH of the environment). ..

Figure 4 :Measurement of the amount of DNA in coffee plants. Top, 2X and 3X control plane. Middle 2X and 3X mixture. Bottom 3X control mixture and unknown sample measured at 4X.


Cell cycle

"The cell cycle represents the entire period of division, that is, all the biochemical and morphological events that are responsible for cell proliferation".

Measurement of the cell cycle by conventional FCM methods divides the cycle into three phases: G0/G1, cell activation phase, S, DNA synthesis phase, G2/M mitosis phase (Figure 5 and 6). The distinction between G0 (quiescent phase) and G1 (preparation phase for DNA synthesis) as well as G2 (preparation for mitosis) and M (mitosis) is impossible with a method using an intercalant as propidium iodide.

Figure 5 : The different phases of cell cycle

Figure 6 : DNA quantity evolution during cell cycle


FCM offers a fast and easy to implement methodology for cell cycle analysis. It allows to follow the distribution of cells in the different phases of the cycle according to various stimuli or to the addition of certain drugs. It also allows to see the presence of cells with abnormal DNA content...
Most of the applications concerning the cell cycle use only one parameter, the DNA content (Figure 7). Mathematical programs calculate the different phases of the cycle. The main applications concern pharmacology: study of the effect of drugs on the cell cycle, cancerology: to determine the proliferation of a tumor and to see its DNA content compared to normal cells.


Exemple de cycle cellulaire
Figure 7 : Cell cycle analysis of a tumor cell line.

There are other methods of cell cycle assessment that can calculate more cell cycle parameters. These methods use classical DNA labelling with an intercalant but add the incorporation of thymidine analogues. The BrdU technique requires the cells to be cultured with this analog (Figure 8) which is incorporated during the DNA synthesis phase. It will then have to be revealed by means of a specific antibody capable of penetrating the nucleus to bind to BrdU. This method required the denaturation of the DNA with hydrochloric acid which denatured the proteins and prevented the coupling of the cycle label with a membrane marker. There is now a milder variant allowing to study in addition the markers by immunofluorescence (Carayon et al.). By adjusting the incubation times with the thymidine analogue, information on the kinetics of the cell cycle can be calculated.

Figure 8 : Cell cycle labeling with BrdU. In X, propidium iodide labeling, in Y BrdU labeling. On the left, single IP labeling, in the middle, double labeling on standard cells, on the right, double labeling on S-phase blocked cells

New methods that do not require DNA denaturation have appeared on the market, such as the ClickIt method from ThermoFischer, which uses a modified thymidine analogue (modified EDU). The revelation is done without denaturation because the revelation molecule is small enough to bind to the modified EDU after fixation and permeabilization by a detergent (Figure 9).

Figure 9: Cell cycle labeling by the ClickIt technique. On the left the principle, on the right double ClickIt labeling (in Y) with FxCycle DNA label (in X).

Back