Lineage Tracing

Adult tissues are dynamic structures whose maintenance depends on the same process of development as embryos. While stem cells persist indefinitely, many cells live only briefly and must be replenished from new cells. The ultimate source of cells for this renewal process are the stem cells, which are defined by their ability to produce daughter cells by dividing asymmetrically and possess self-renewal ability. Although stem cells are important for tissue homeostasis, finding them, characterizing their cellular properties, and mapping the daughters they produce is difficult. This is essentially because tissues contain different types of cells that constantly undergo complex movements and renewal that makes it generally difficult to identify stem cells.

Lineage analysis represents the most powerful and reliable method for stem cell identification and for deciphering many aspect of normal tissue behaviour. Ideally, a single cell or group of cells is labelled in a way that is transferred to all daughter cells arising from the initial cell, resulting in labelled descendants. By analysing the phenotype, number, location and type of cell generated by the stem cells, their lifetime and movement can be deduced. Classically, lineage tracing is divided into prospective analysis, mosaic analysis, and retrospective analysis. Each of these methods can be used independently or simultaneously with each other to map stem cell fates and determine cell lineage.

Prospective Lineage Analysis

Prospective lineage analysis involves labelling cells at a specific stage and at a known position, and subsequent analysis of the cells’ contribution to a structure or tissue. This technique generally requires pre-existing knowledge of the cells and their location in the tissue of interest.

Stem cells can also be labelled with several dyes in vitro and introduced into a particular organ where they can undergo cell division, and the descendants are monitored over time. Grafting of labelled cells was utilized in mapping the heart-forming fields in a chicken embryo and to map the fate of cells in the epiblast of a mouse embryo. However, this approach is usually invasive in nature. A less-invasive approach compared to grafting has also been used; this involves introducing lypophilic carbocyanine dyes, like Dil, into a specific region of an embryo. These dyes have the capability of intercalating into the cell membrane and can easily be visualized using a fluorescent microscope. This technique allows the tracing of a group of cells during the development process of the tissue of interest. One of the important studies performed by this technique includes the deduction of migration pathways of neural crest cells and how their temporal order contribute to cell derivatives in a chicken embryo.

One disadvantage of the above methods is that multiple cells are labelled, complicating analysis at single-cell resolution. However, a single cell can also be labelled by microinjection of horseradish peroxidase (HRP) or dextrin linked to fluorescent dyes, which are too large to diffuse between adjacent cells and can therefore serve as an intracellular marker of the cell of interest. HRP is a glycoprotein that can be conjugated to a labelled molecule such as a protein, RNA, or DNA. When incubated with a proper substrate, HRP produces a coloured, luminescent, or fluorimetric derivative of the initial labelled molecule, allowing detection and quantification using a spectrophotometer or fluorescent microscope. HRP injection into a leech embryo was used to demonstrate that teloblasts, the founder cells of segments, give rise to topographically invariant lineages of different cells in leech segments.

In the last 20 years, the advent of GFP (Green Fluorescent Protein) has had a major impact in cell fate mapping and tracing. GFPs are proteins that exhibit bright green fluorescence when exposed to ultraviolent light in the blue range and comprise 238 amino acid residues. Genetically encoded fluorescent protein can be targeted to the nucleus or plasma membrane using microinjection, providing more information such as cell shape dynamics, mitotic status, and clearer resolution. Following the same basis, microinjection of mRNAs encoding membrane-bound labelled proteins into a single cell, followed by time-lapse imaging, can allow for the analysis of cell movement and position-dependent induction. The biopotent ability of chicken somite and the ability of its descendants to progressively acquire a muscle or dermal cell fate was demonstrated using the microinjection of GFP into the chicken embryo.

In addition to GFP and fluorescent dyes, other techniques have been used for prospective lineage analysis. Advances in the knowledge of chromophore photochemistry has allowed for genetic engineering of photo-modulatable fluorescent proteins. These labelled proteins undergo a spectral change after light activation. After introduction of these proteins into the cells, they can be photo-converted by using either a fluorescent microscope or confocal laser microscopy. In some instances, two-photon microscopy may be applied for single-cell labelling to achieve even greater information. There is the advantage of deeper tissue penetration and less phototoxicity using
 this approach. Studies in Drosophila using photo-modulatable proteins demonstrated clonal restriction at anterior but not posterior segments when an Engrailed I-positive parasegment is activated.

Several problems are associated with using fluorescent dyes in lineage tracing, including brightness of the fluorescence produce, stability, rapidity of photoconversion, and its toxicity, thus making it imperative that experimental conditions are optimized and these parameters tested for the developmental stage and specific organisms before use in an individual experiment.

In general, the prospective approach of lineage tracing using microinjected markers and encoding photo mouldable fluorescent proteins allows scientists to draw cell fate maps and reconstruct cell lineages. A major merit of fluorescent proteins is that they can be utilized after a stable integration into the genome has been achieved and thus can provide permanent cell labelling. However, the use of these methods is limited by marker dilution at each stage of cell division and differentiation, hence only allowing for short-term experiments. Equally important, the introduction is often invasive and leads to tissue and cellular damage.

Genetic Tracing

In order to trace the lineage of a cell in a non-invasive manner, the genetic recombinase technique has been used to achieve permanent genetic labelling through specific activation of a reporter gene. FLP-FRT (flippase recognition target) recombination system involves the recombination of DNA sequences that lie between the FRT sites by the FLP recombinase plasmid. On the other hand, Cre-Lox recombination system uses the Cre recombinase enzyme to recombine a pair of Lox sequences, known as short target sequences. The recombination activity can be targeted or triggered by an external stimulus, such as chemical signal or heat shock, to a specific organ and allows for consistent alteration in DNA of a particular cell type as well as very specific labelling of the cell of interest. The improvements in several variants of FLP and Cre recombinase systems, coupled with identification of more efficient specific target site variants including FRT and Lox, has increased the effectiveness of this technique in lineage tracing. This approach is generally regarded as rapid, specific, and efficient. The recombinase technique has been used extensively in lineage tracing, including in the mapping of cells originating at the mouse midbrain-hindbrain constriction.

An alternative to the recombinase approach is the permanent cell labelling by auto-induction of the Gal4 gene under the control of UAS in the reporter cassette without the need for the tissue-specific promoter. Normally, a single marker is activated when an effective recombination is achieved. However, current improvements have been made possible for switchable reporter lines to be activated, in which recombination inverses a first reporter cassette and allows the expression of a second cassette, hence marking
of cell types before recombination and after successful recombination. To effectively follow all derivatives of a cell, there is a need for a ubiquitously expressed sequence regulating the conditional reporter. The techniques that underlie genetic cell labelling
are spatially regulated by the use of a tissue-specific promoter sequence to target a progenitor cell of choice. The genetic labelling approach provides information on the activity of a promoter.

It is often important to remember that in these approaches all the cells that had expressed the specific promoter driving Cre or FLP are labelled, without any distinction between different progenitor cells. Any results from this tracing method should be looked at as identification of a structure that came from a gene expression domain.

Retrospective Lineage Analysis

Prospective analysis is usually based on the knowledge about the progenitor cell population, potential stage of development, and location of the progenitor cells of interest. However, these parameters are not always known and limit the conclusion
on lineage of a cell. In contrast, retrospective analysis is based on analysis of marked cells at the final stage of an experiment and deduction of their history, lineage, and interrelationships. This approach involves random genetic labelling of progenitor cells and mapping their potential to colonize a specific structure.

Spatially Random Labelling

Clones induced at a specific time from random spatial labelling have been produced by high doses of x-irradiation-induced genetic recombination. The irradiation induces the formation of fusion genes characteristic of somatic cells originating from different progenitor cell. This allows for deduction of clonal, compartment, and boundaries.

Another approach to spatial random labelling is by infection of progenitor cells with a retrovirus defective in replication. In addition, infection with GFP encoding lentiviruses and retroviruses has been used to label single cells. Tamoxifen inducible recombinase systems based on the use of ubiquitous reporter proteins have also been used to provide spatially random labelling, by adjusting the duration and dose of the tamoxifen used for induction. The use of tamoxifen induction of recombination led to the important conclusion that stem cells maintain their turnover through cell division that is independent of asymmetric cell divisions.