home » libere » molec_model_vis_gen




Gian Michele Ratto - gianmichele.ratto@sns.it

Silvia Landi

Riccardo Parra

Francesco Trovato (Scuola Normale Superiore)


Genetic mosaicism occurs when the cells in a tissue express two different genomes. There are many examples of this situation that has very important consequences if one of the alternative genomes expresses a pathological protein. In this case the tissue can be pathological even if only a subset of cells is “diseased”. A striking example of genetic mosaics is present in all female mammalians: here each cell has two copies of the X chromosome as opposed to males that only have a single copy of the chromosome. To prevent this duplication, one of the two copies is randomly inactivated, and therefore in a tissue there is a mosaic of cells expressing one or the other X chromosome. If one chromosome carries a pathological mutation the disease that arises is characterized by an intermixing of normal and “diseased” cells. A prototypic example of this class of pathologies is the Rett syndrome (RS), a rare and untreatable disease characterized by the insurgence of autism, mental regression and ataxia after the second, third year of age. In most cases, this disease is caused by a loss of function mutation of the X-linked gene MeCP2. In females only half of the brain cells carries the mutation but, since the diseased cells cannot be identified, it is not possible to study the cellular physiology of the affected cells interspersed in a field of normal cells.

In this activity, we are developing a general-purpose tool that allows the creation of models of genetic mosaicisms and the differential visualization of normal and “diseased” cells.

     1) Creation of the mosaic
In this example we show the way to create a mosaic of MeCP2 knock out. The starting point is a mouse line in which the native MeCP2 gene has been substituted (knock in) with the same gene flanked by two lox sequences (Chen et al. Nature Genetics 2001). When a gene is flanked by these sequences it is normally expressed and behaves exactly as the wild type gene. However, in presence of the bacterial enzyme Cre-recombinase, the gene is excised irreversibly and the cells becomes knock out for the gene. By grading the activation of Cre-recombinase, it is possible to create a sparse activation of CRE leading to a cortex in which only a fraction of cells are KO for MeCP2. In this example the MeCP2 floxed mouse is crossed with a mouse line expressing Cre in neurons. In the resulting offsprings, Cre can be activated by the administration of a factor (Tamoxyfen, an exogenous extrogen) necessary for its operation, Figure 1 shows the immunostaining for MeCP2 in the brain of a control mouse (no expression of Cre), after 24 and 36 hours after Cre activation.





Fig 1. Ablation of MeCP2 in pyramidal neurons after activation of inducible Cre. Double immunohistochemistry for Cre recombinase (green) and MeCP2 (red) in the visual cortex (layer 2/3) of three female mice obtained crossing a homozygous floxed MeCP2 female with a mouse heterozygous for CaMKIIa-Cre/ERT2. A) The control mouse did not express Cre recombinase and MeCP2 distribution is normally present in the neuron nuclei and is particularly enriched in the chromocenters. B) A mouse positive for Cre was treated with a single IP injection of Tamoxifen and sacrificed 24 hrs later. Cre is activated as demonstrated by the strong nuclear localization of the recombinase. MeCP2 staining is still evident but in some cases it can be seen a decrease in fluorescence in the nucleus (white arrows) while chromocenters are still bright, which is consistent with the strong binding of MeCP2 to these sites (Marchi et al, 2007). After 36 hours Cre has left the nucleus and pyramidal cells are indicated by a faint ring in the perinuclear region (white arrow). MeCP2 staining is basically gone from a number of cells. Since this is a female heterozygous for floxed MeCP2 one expects a loss of signal in about 50% of pyramids. D) Quantification of MeCP2 fluorescence in the three mice. * p<0.01 t-test.



The titration of the activating drug leads to a system in which Cre is activated only in a subset of neurons but, as shown in the figure, the excision of MeCP2 can only be demonstrated in fixed tissue by immune staining or RT-PCR. In order to allow the identification of recombined cells in vivo we are designing a family of fluorescent sensors (named Beatrix) that modify their optical properties upon activation of Cre. Figure 2 shows the architecture and function of one element of the family (Beatrix 1.0). In brief, the gene expressing this sensor includes two fluorescent proteins (EGFP and dsRED) that are both expressed, so the cells shines in yellow. The EGFP gene is flanked by two lox sequences: when Cre is active, the gene expressing the EGFP is excised and the cell turn red, allowing to recognize the activation of Cre.




Fig 2. Sensing CRE activation in HEK cells. The Beatrix 1.0 construct is schematized in the upper panels. The left scheme shows the intact construct formed by the two fluorescent proteins placed downstream of the promoter CAG. The GFP is flanked by two LoxP sites: upon Cre recombination the GFP is excised (right panel). The central image show a culture infected with the Beatrix 1.0 plasmid and a very low concentration of constitutively active Cre recombinase (1:200 ratio). In this way only a fraction of cells expressing Beatrix 1.0 are also endowed with Cre activity. Most cells appear red (thus expressing only DsRED due to the excision of GFP), but a fraction of cells have yellow-green fluorescence indicating expression of the intact construct. Panels B, C show the quantification of the colorimetric data: each symbol represents one cell. The data have been collected 24 and 36 hours after the transfection with Cre. After 36 hours two clearly distinct cell populations are visible.



These constructs have been packed in an expression cassette designed to allow the transfection of brain cells by means of in utero electroporation. Recently, we have collaborated with Laura Cancedda (IIT, Genoa) to develop a major improvement of this important technique. Figure 3 shows a preliminary data demonstrating that our expression cassette is powerfully expressed in the adult cortex after in utero electroporation.



Fig 3. Expression of dsRED under the CAG promoter by means of enhanced IUE. The left panel shows the cortical surface through an optical window placed on an adult mouse after in utero electroporation. The bright dots are neurons placed near the surface of the cortex. The right panel shows the maximum projection of a image stack obtained at the 2-photon microscope at about 150 μm of depth.


Future perspectives


The initial goal of this project is to deplete MeCP2 in a sparse set of pyramidal neurons or astrocytes tagged with the fluorescent reporter that will allow to recognize wt and MeCP2-KO cells. Thus, we will study single diseased or wt cells within a network with a graded presence of altered cells. This result represents a novel model to study the cellular mechanisms of Rett syndrome. Changes in axonal projections, dendrite patterning and spine morphology will be studied in vivo by means of time-lapse imaging. In parallel, we will study excitability and synaptic plasticity. We expect that these studies will provide a framework to identify what are the direct biochemical and physiological consequences of the MeCP2 mutation. For the first time, this approach allows a comparative study of Rett onset and progression in a mosaic model of the disease, allowing the direct comparison of the functional properties of diseased and control cells. More generally, this tools might represent a novel tool for "in vivo" identification of cells whose genome is under control of CRE recombinase.


Relevant references


Spatio-temporal dynamics and localization of MeCP2 and pathological mutants in living cells

Epigenetics 2 (3), 187-197 (2007)


The short-time structural plasticity of dendritic spines is altered in a model of Rett syndrome

Landi S, Putignano E, Boggio EM, Giustetto M, Pizzorusso T, Ratto GM.

Sci Rep. 2011;1:45.


High-performance and site-directed in utero electroporation by a triple-electrode probe

dal Maschio M, Ghezzi D, Bony G, Alabastri A, Deidda G, Brondi M, Sato SS, Zaccaria RP, Di Fabrizio E, Ratto GM, Cancedda L.

Nat Commun. 2012 Jul 17;3:960.


Tagged Neurons during Calcium Imaging by Means of Two-Photon Spectral Separation

Brondi M, Sato SS, Rossi LF, Ferrara S, Ratto GM.

Identification of EGFP Front Mol Neurosci. 2012;5:96.

Valid XHTML 1.0 Transitional