Final Activity Report Summary - QDOTS (Novel Quantum Dot Imaging technologies for the study of morphogenesis and other biological processes)
We postulated that near infra-red (NIR) emitting Quantum Dots would allow the labelling and tracing of labelled cells at much deeper parts of the embryos than with the use of traditional organic and protein fluorophores. Initial tests evaluated the threshold of toxicity of these QD´s in the living embryo as well as the phototoxicity and photodamage in long term imaging experiments. Relatively high amounts of QD´s were found to be tolerated by Xenopus embryos, which were only imaged for short periods of time to evaluate the successful introduction of the nanocrystals. These embryos developed normally to the tadpole stage. However embryos, which were used for long term imaging of morphogenetic movements showed visible signs toxicity within one or two hours of observation. We concluded that use of optimal filters sets for QD imaging which use excitation wavelengths centred around 400nm are not appropriate for use in in vivo experiments especially when imaging deep tissues where high intensity excitation light must be used.
We thus created custom sets with excitation filters cantered at 500nm. These filter sets excited the QDs suboptimally but embryos showed no signs of photodamage or phototoxicity even after 24 hours of time lapse imaging. These experiments confirmed that the initial observed toxicity was not due to phototoxicity from the emitted NIR light of the QDs but was photodamage stemming directly from the near UV excitation light. During the time frame of this project a number of other parameters were optimised resulting in the successful imaging of deep movements during the entire gastrulation and neurulation of embryos. Despite the high signal to noise ratio and high penetration achieved with NIR QDs imaging of the deepest tissues was still unattainable with the use of non-targeted QDs.
Another problem was the inability to clearly differentiate individual cells within a tissue due to the homogeneous localisation of the QDs within cells as well as the inability to create a mosaic pattern of labelling through microinjections. This problem was overcome with the creation of nuclear localisation peptide-QD(NLS-QD) conjugates. A custom NLS peptide was used to carry out a reaction with carboxyl QDs thus creating QD nanocrystals coated with multiple NLS signal peptides. These NLS-QD conjugates were effectively taken up and concentrated in the cell nuclei.
The localised high concentration of QDs in the nuclei led to a significant increase of the penetration as well as the resolution of our imaging. With the use of the NLS-QD conjugates we have been able to track individual cells and document deep movement in Xenopus with single cell resolution something impossible with other methods. Time lapse images of these movements have been created and are currently being analysed. These 4D movies were improved through the use of spectral un-mixing algorithms using custom un-mixing matrices we created for the specific QD conjugates.
Further improvement was attained with the application of de-convolution software using custom experimentally determined point spread functions (PSF) for the QD conjugates we used.
Although this is a time consuming method we conclude that it is the best suited method for imaging of morphogenetic movements in vivo, excluding the use of extremely costly multi photon confocal microscopes. As part of this project we also evaluated the potential use of structured illumination as a method for the creation of 3D stacks from gastrulating Xenopus embryos.