Hans Torma, PhD,
Department of Medical Sciences/Dermatology at University Hospital in Uppsala, Sweden
|Dr. Anders Valquist||Dr. Hans Törmä|
Valquist and Törmä are studying EI due to mutations in keratin genes (KRT1 and 10) which code for structural proteins constituting the cell skeleton in skin cells. The model system used is cultured skin cells (keratinocytes) from a number of EI patients with different keratin mutations and different clinical symptoms. The clinical severity of the patients has been shown to be reflected in the cultured cells. They have found that such cells, when exposed to external heat stress, show a collapse of the cell skeleton, which is similar to that in EI skin. Pre-treatment of the cells with certain compounds (chemical chaperones) preserves an intact cytoskeleton, also, after heat-stress. The protective effect of these compounds not only suggests that they are putative drugs for treating EI, but also that EI resembles other protein misfolding disorders, e.g. neurodegenerative disorders, that have been experimentally treated with systemic administration of chemical chaperones.
In the continued search for improved therapeutic options in EI, they will utilize patients´ cells and a library of more than 8,000 chemicals to screen for other drugs which protect the cytoskeleton.
Cultured EI cells are exposed to the compounds and subsequently heat-stressed. The cells are fixed and stained with protein markers that are involved in the collapse of the cytoskeleton.
The expression of selected markers is studied by automated high-content screening methodology using fluorescence microscopy, which allows us to rapidly search for compounds that prohibit the collapse in vitro.
Importantly, the safety and pharmacokinetics of such new candidates need to be tested in pre-clinical trials before controlled clinical trials in patients can be commenced.
UPDATE: March 2011
The new ArrayScan was installed in November 2010. Prior to the installation we performed a number of experiments in order to optimize protocols with respect to fixation and keratin staining. We have obtained a few new antibodies for K1 and K10 which have been tested for immunocytochemistry using ordinary fluorescence microscopy and the ArrayScan.
All our cell lines need to be calcium-differentiated in order to express keratin 1 and keratin 10. The necessary switch from glass coverslips into 96-well plates required optimization with respect to seeding density, degree of confluency at differentiation start, calcium concentration and exposure time. What we regularly observe is that the cell lines from EI patients show lower expression of suprabasal keratins than the control cell lines under the same culture conditions. A new protocol which gives reproducible expression of K10 has been established. This part of the project has taken unexpectedly long time.
Using an immunofluorescence microscope equipped with an Apotome (optical sectioning) the aggregates of mutated K1/K10 are easily detected despite the fact that cells grow as a multilayer. The cell borders are not depicted with high accuracy by the software (Figure 1) but different settings is being tested. When we started to elaborate the protocol for detection of keratin 10 aggregates using the ArrayScan (Figure 1), we found that these aggregates are easily detected by manual focusing but when using automatic scanning (auto-focus) some of the specificity was lost. This is probably due to the fact that auto-focus highlight the brightest areas in the viewing field, which is usually around the nuclei and not in the cell periphery. This can probably be circumvented by equipping the ArrayScan hardware with an Apotome function for optical sectioning. At this stage we have requested a demonstration and a try-out period of such an Apotome hardware.
In parallel with these experiments, we have established protocols for in situ proximity ligation assay (in situ-PLA). This is used for detection of proteins which co-localize with suprabasal keratins in the aggregates. This technique might be used for screening of aggregates in case scanning by keratin immunofluorescence staining fails. Among the proteins so far examined for co-localization are HSP70, MAPK, and ubiquitin, all of which are involved in the intracellular processing of misfolded keratin polymers (Chamcheu et al, J Invest Dermatol in press.)
UPDATE: October 2012
« Back to Previous Page