Study Of Multi-scale Bio-imaging Based On Synchrotron Radiation X-rays

Understanding of the microstructures of biological systems including individuals, organs, tissues and cells even smaller, ranges from meter to micrometer. To realize the imaging of bio-system with multiscale is of vital importance for the development of biological science. X-ray imaging techniques fill the gap of optical microscopy and electron microscopy, owing to the facts that X-rays have the properties of short wavelength and high penetration power. The specific properties make it a best candidate for nondestructive three-dimensional (3D) high resolution imaging of biological specimen. Recently, with the development of worldwide synchrotron radiation light sources and fabrication of X-ray optical elements, synchrotron based X-ray microscopy has been playing an important role in the structural investigation of biology science. However, some technical difficulties remain including the radiation damage of biological specimens, the fidelity and accuracy of tomographic algorithms and post. imaging processing of reconstructed images for better data interpretation. In the thesis, we perform the study of multi-scale bio-imaging based on synchrotron radiation X-rays. The main contents are follower:(1) High resolution cellular imaging based on synchrotron radiation X-rays with low radiation dose and intracellular imaging of nanomaterials with dual-energy contrast X-ray imaging.In combination with scanning transmission X-ray microscopy (STXM) and a new iterative tomographic algorithm termed equally sloped tomography (EST), we performed the quantitative 3D imaging of a macrophage which is cultivated in a potential nanomedicine [Gd@Cs2(OH)22]n. Two datasets with 46 equally sloped projections with range of ±79.4°were acquired at the energy of 1186 eV and 1189 eV just before and above the absorption edge of Gd M2. The micro-focused spot of X-rays was 50 nm in diameter and the step size of scanning is 50 nm. The microstructures of the macrophage with size of ～18.1×15.5×3 μm3 with 50 nm pixel resolution were achieved by EST algorithm and we observed the characteristic structures of the cell including flat structures, nucleus and lysosomes and so on. In the experiment, the radiation dose of the cell was greatly reduced to 4.43×106 Gy by combining STXM and EST algorithm. There were no structural changes observed in the cell before and after the data acquisition in the experiments. Meantime, the accuracy and fidelity of the tomographic reconstruction was improved when the number of projections was limited and the missing wedge was partially solved owe to the iterative algorithm used.The dual-energy contrast X-ray microscopy was systematically investigated, which is based on the fact that the absorption of specific elements changes dramatically when the energy of X-ray change from before absorption edge to above it. By this method, the quantitative 3D intracellular imaging of [Gd@Cs2(OH)22]n was realized. We analyzed the two strategies to achieve the 3D imaging of elemental-specific nanomaterials in dual-energy contrast X-ray microscopy. In the first method, the two-dimensional (2D) elemental mapping of specific elemental at tilt angles was calculated, and then the series of 2D projections were used for tomography after intensity normalized and the rotation axis calibrated with center of mass intensity method. Then the quantitative 3D elemental mapping of specific element was realized. In the second method,3D imaging of the macrophage at two energies was calculated using EST algorithms and then the quantitative 3D distribution of specific elements was calculated by subtracting of the two 3D images. X-ray fluorescence (XRF) mapping of the same macrophage further show better performance of the proposed method.By comparing the two methods, we find that the second method can achieve much better image quality showing a much higher signal to noise ratio. Based on the 3D distribution of Gd achieved by the second method, we analyzed the subcellular distribution of [Gd@C82(OH)22]n in the macrophage. The macrophage that were exposed to a potential antitumor agent [Gd@C82(OH)22]n were investigated. And large amount of nanoparticles aggregated within the cell and mainly located in phagosomes. No nanoparticles was observed in the nucleus. Imaging of the nanomedicine within whole cells advanced the understanding of its antitumor activities with high efficiency and low toxicity. The image technique could probe nanomaterials within intact cells at nanometer resolution in three dimensions, and may greatly benefit the field of biomedicine and nanotoxicology.Meantime, cellular imaging of an Ag-labelled lymphocyte was performed by combining STXM and EST algorithm, in which 37 equally sloped projections with angle range of -63.44°to 69.44°were acquired at 1060 eV. The micro-focused spot of X-rays was 50 nm in diameter and the step size of scanning is 50 nm. And projections were reconstructed with EST algorithm after the calibration of rotation axis and normalization of intensity. The external morphology of the cell with size of～20×17x5 μm3 was achieved successfully. But the internal details cannot distinguished owe to maybe the fact that the cell thickness is thicker than the depth of focus; the ratio of transmission of X-rays is so little; the size of Ag nanoparticles is far less than the recent spatial resolution that it cannot be observed. The experiment helps to optimize the parameter for further cellular imaging.(2) Study of phase contrast X-ray imaging of living insects with low radiation dose and improved resolution capability. In the study, two parts are included that (1) a dried yellow mealworm Tenebrio molitor was imaged based on absorption contrast used to compare the performance of EST algorithm and the conventional tomographic algorithm (Filtered Back Projection, FBP); (2) a living yellow mealworm T. molitor was imaged based on in-line phase contrast imaging.Firstly, absorption contrast imaging of a dried yellow mealworm T. molitor was preformed based on synchrotron X-rays. In the experiment,320 equally sloped and 600 equally angled projections were acquired separately from the same specimen at 12 KeV and exposure time of a single projection was 55 ms. And the projections were reconstructed using EST and FBP separately. In the tomographic process, the number of projections was changed to symmetrically compare the performance of the two tomographic algorithms. The metrics to characteristic the image quality such as signal to noise ratio, contrast to noise ratio, and noise power spectrum all show the better performance of EST algorithm. EST algorithm with 320 projections perform better than FBP with 600 projections. Even EST with 180 projections could achieve recognized internal microstructural details. EST algorithm could achieve the same or better image quality with one to third projection compared with conventional tomographic method. With the decrease of number of projections used for reconstruction, the radiation dose reduced greatly. Based on the microstructures achieved by EST algorithms with 320 projections, the internal microstructures and external morphology were analyzed and recognized, such as different muscle tissues.Then a freezing-fixed living yellow mealworm T. molitor with similar size of the dried one was imaged with in-line phase contrast X-ray microscopy. The internal microstructures and external morphology of the insect was analyzed in living state. By the optimization of experimental setup such as the use of fast detectors and mechanical shutter just before the sample, and application of EST algorithm, the total exposure time of 320 projections was 1.92 s and the radiation dose in the data acquisition was decreased dramatically to 1.2 Gy which is far less than the lethal dose of insects. The mealworm stay alive after the experiment. The 3D images could be analyzed quantitatively and qualitatively at 9 μm pixel resolution. Based on the high-quality 3D images, branching tracheoles and different tissues of the insect in a natural state were identified and analyzed, demonstrating significant advantage of the technique over conventional X-ray radiography or histotomy. The improved static tomographic method demonstrated shows advantage in the non-destructive structural investigation of living insects in three dimensions due to the low radiation dose and high resolution capability, and offers potential applications in biological science.(3) Noninvasive 3D structural analysis of arthropod by synchrotron X-ray phase contrast tomography. Equally angled 450 projections were acquired at 12 KeV with each projections of 6 s exposure time. Three-dimensional microstructure of arthropod with high contrast has been demonstrated by synchrotron X-ray in-line phase contrast tomography. The external morphology and internal structures of an earthworm were analyzed based upon tomographic reconstructions without and with phase retrieval by phase-attenuation-duality Born algorithm (PAD-BA) at 9 μm pixel resolution. We also identified and characterized various fine structural details such as the musculature system, the digest system, the nervous system, and the circulatory system. Conventionally the microstructures were achieved by histological section, in which serial sections are stained and observed by light microscopy or scanning and transmission electron microscopy. However, the anatomical approaches are destructive and time consuming, making measurements of large numbers of samples impractical. Here, we used in-line phase contrast X-ray tomography, the simplest and the most straightforward among all type of the phase contrast microscopy, to noninvasively investigate the external morphology and internal microstructures, by which we hope to significantly advance the understanding of the species and the evolution. This work exhibited the high efficiency, high precision and wide potential applications of synchrotron X-ray phase contrast tomography in nondestructive investigation of low-density materials and biology.(4) Ring artifact correction with an improved sinogram processing method for synchrotron X-ray microtomography. Ring artifacts are common in tomographic images which complicate the analysis and interpretation of object from synchrotron X-ray microtomography. Here we proposed a novel method to calibrate the ring artifacts by processing sinogram data. Before tomographic reconstruction, the ring artifacts were identified and located as line artifacts in sinograms domain. Then the addressed line artifacts were corrected as single point noises by filtering in the original projections. As a result, the corresponding line artifacts were removed and thus the ring artifacts were corrected in the reconstructed tomographic images. The proposed method provides a new approach for removing ring artifacts. Firstly, we simulated the process of ring artifacts corrections with phantom model showing the fidelity and feasibility of proposed method; secondly, the method was further checked by the experimental data that a dried yellow mealworm T. molitor was imaged based on absorption contrast imaging in which tomographic slices are corrupted to some extent. Both simulations and experiments verified the efficiency of the proposed method. And much better performance could achieve by combining better line artifacts detection methods and denoising algorithms.