Engineering genetic reporters for in vivo photoacoustic imaging of mammalian tissues

After gaining much understand about how individual cells work, biologists are moving their attention to understand how many different cells work together in tissues and organs. Imaging cells functioning in a living animal gives us tremendous insight into how these complex systems work. This is particularly true if the imaging not only tells us where a particular cell is but also gives us some information about what is going on inside it over time. With this grant, we plan to use a new technique called photoacoustic imaging along with genetic engineering to so we can image particular cells functioning within a living animal. Currently, we can image clumps of cells in small animals using imaging techniques used in medicine. Magnetic resonance imaging (MRI) and computed tomography (CT) can produce detailed imaging of a whole small animal. However, it is difficult to identify rare cells we are particular interested in. On the other scale, researchers can use microscopes to observe cells in living animals. Considerable information has been gained by this kind of work. Although allowing very detailed images, this kind of imaging can only observe a small portion of tissue at a time. Further, very often surgery must be performed on the animal to allow the piece of tissue of interest to fit into the microscope apparatus. What we really need is something which can image at resolutions between microscopy and whole animal imaging so we can observe a whole section of tissue function in a non-invasive way. We are currently working on a technique called 'photoacoustic imaging' which can meet these requirements. This non-invasive technique is a unique mixture of optical imaging and ultrasound. We first shine laser light into animal tissue. This light is not powerful enough to cause any damage, but it causes tiny shock waves within the tissues whenever the laser hits something which absorbs light. For instance, small blood vessel packed full of red blood cells absorb light very well. These small shock waves are very similar to the signals received by Ultrasound imagers, and we use an ultrasound receiver to generate 3D images of a section of tissue in a living small animal. The scale of the imaging is precisely what we desire to study complex tissues in action. While we can see small blood vessels clearly with this technique, up until recently we have had to inject chemical dyes into the animals to try and label certain cell types. This is difficult to do as the dyes do not always go just to the cells of interest, and in a living animal the dyes are broken down. We have thought of the idea of genetically engineering cells we are interested in studying so that they make their own dyes. We have taken two genes from coral and introduced them into bacteria. These genes instruct the cells to make a blue pigment allowing us to image these particular cells. We have performed many experiments to show that these blue proteins should be ideal for photoacoustic imaging, particularly since the red light they absorb can penetrate deeply into tissues and is not absorbed so much by red blood cells. Before this can become practically useful, a lot of work needs to be done to modify these coral proteins and establish exactly how they behave when imaged. For example, these coral proteins have to be engineered so that they can be more easily made by mammalian cells and not aggregate or clump. Next, again by engineering, we can alter the colour of these proteins potentially allowing us to image many different cell types at the same time. Next, we can alter these proteins to act as sensors, so they only have colour when a particular event is happening within the cell. Finally, we need to perform a lot of work with cells expressing these proteins in small animals to establish how best to use them. These new proteins will greatly increase the usefulness of photoacoustic imaging and will help us understand how complex collections of cells work together.

Functional imaging of living tissue is indispensable to understanding biological systems. The use of engineered genetically encoded fluorescent proteins (FPs) for microscopy has revolutionized this field allowing the study of cellular activity in living organisms. The utility of these FPs has increased through protein engineering: for instance, the optical range available has widened allowing imaging of many fluorescent proteins simultaneously; functionalization of these proteins allows them to act as biological sensors etc. While optical techniques allow detailed imaging below 1mm, until recently we lacked a means of imaging tissue at depth with a cellular resolution - scatter denies deeper in vivo study even in advanced microscopy. With Photoacoustic imaging (PAI), non-ionizing laser pulses are delivered into biological tissues. Some of the delivered energy is absorbed leading to thermoelastic expansion and wideband ultrasonic emission. These are detected by ultrasonic transducers to form images at a dept of centimeters with a possible resolution under 100uM. This modality bridges the gap between microscopy and radiology but is still in its infancy. Notably we lack genetically encoded markers which have proved so valuable with functional microscopy. FPs are unsuited for PAI: their fluorescence property decreases thermoelastic response; further, FPs absorb in the blue/green range while for deep imaging far red absorbance above that of Heme is needed. We have demonstrated feasibility of PAI with two blue chromoproteins (CP) cjBlue and aeCP597. These CPs are related to FPs but only absorb light, thus converting more energy to acoustic signal. Further, these proteins have good absorbance above that of heme. We plan to build a set of optically and functionally distinct CPs based on cjBlue and aeCP597. These proteins will form powerful tools to dynamically image genetically labelled tissues in living animals and to track engineered cells in human subjects.

We propose to develop chromoproteins (CP), as genetically encoded markers and labels for noninvasive photoacoustic imaging (PAI) of living tissue. If successful, this work will impact many aspects of fundamental and applied biological research, reduce the amount of animals required for experimentation, allow imaging of engineered cells in human subjects, and would create valuable intellectual property as well as adding to the utility and value of photoacoustic imaging itself. *Scientific Applications. This work could potentially impact on all cell biology research utilizing small animal models. Rather than an exhaustive list, I will provide some examples: Embryology: It may be possible to distinctly label differently differentiating cells and their lineages, or track expression of particular genes or signalling events such as apoptosis in developing murine embryos. Neurolobiology: it is possible to image the brain of a young mouse with PAI - it should also be possible to engineer a CP to report electrolyte flux allowing real time functional brain imaging perhaps using different colour CP in different neuron types. Cancer biology: it should be possible to engineer cancer models so they report on e.g. stemness, cell cycle state, expression of metastasis genes, apoptosis and so forth. A particularly important area may be therapeutics where transgenic animals or engineered tumour xenografts express reporters which report biological effects of experimental pharmaceuticals giving real-time information on drug effects. Finally, with more information obtainable per animal, with longitudinal non-invasive imaging, this work should lead to less small animals for experimentation and less invasive protocols. *Translational Applications. This technology would impact clinical research in the tracking and study of genetically modified cellular therapeutics in patients. Increasingly, we and others are using genetically modified T-cells to treat cancer. Beyond this, cellular therapeutics with engineered cells is likely to greatly expand following the recent demonstration that terminally differentiated cells can be re-programmed into induced progenitor cells. Notably, modern multi-cistronic gene-therapy vectors used in both above applications make expression of an additional small protein, such as a chromoprotein, facile. While PAI is not a modality for deep or whole body imaging, detailed tracking of engineered T-cells to superficial sites e.g. skin cancers, lymph nodes would be possible. Development of endoscopic or laparoscopic photoacoustic imaging probe may extend utility to viscera and intra-abdominal imaging. Detailed imaging at a cellular resolution would provide invaluable insight into the function of such therapeutics. *Commercial Impact. A set of novel CPs with application for PAI are easy to patent and relatively easy to commercialize. For instance plasmids coding for most fluorescent proteins can be purchased through companies such as Invitrogen, New England Biolabs etc. Should PAI become more widely adoped, this is a likely route to commercialization. Such genetically encoded markers are likely to make PAI more attractive to a wide variety of researchers and this work would impact on PAI field with increased commercial interest and funding. Further: design, construction and testing custom CP sensors for pharmaceutical industry would represent the high end of commercialization, as would custom transgenic mice. Together, engineered CP along with PAI should impact biotechnological and imaging hardware companies with considerable revenues.