Micro- and nanoanatomy of human brain tissues

Khimchenko, Anna. Micro- and nanoanatomy of human brain tissues. 2017, Doctoral Thesis, University of Basel, Faculty of Medicine.


Official URL: http://edoc.unibas.ch/diss/DissB_12501

Downloads: Statistics Overview


“Better to see something once than to hear about it a thousand times.”
The human brain is one of the most complex organs in the body, containing billions of neurons of hundreds of types. To understand its properties and functionality at the most fundamental level, one must reveal and describe its structure down to the(sub-)cellular level. In general, three-dimensional (3D) characterisation of physically soft tissues is a challenge. Thus, the possibility of performing non-destructive label-free 3D imaging with the reasonable sensitivity, resolution and increased manageable specimen sizes, especially within the laboratory environment, is of great interest.
The focus of the thesis relies on the non-destructive 3D investigation of the micro and nanoanatomy of human brain tissues. The ambitious challenge faced was to bridge the performance gap between the tomography data from laboratory systems, histological approaches employed by anatomists and pathologists, and synchrotron radiation-based tomography, by taking advantage of recent developments in X-ray tomographic imaging.
The main reached milestones of the project include (i) visualisation of individual Purkinje cells in a label-free manner by laboratory-based absorption-contrast micro computed tomography (LBμCT), (ii) incorporation of the double-grating interferometer into the nanotom® m (GE Sensing & Inspection Technologies GmbH, Wunstorf, Germany) for phase-contrast imaging and (iii) visualisation and quantification of sub-cellular structures using nano-holotomography (nano-imaging beamline ID16A-NI, European ynchrotron Radiation Facility (ESRF), Grenoble, France).
Hard X-ray micro computed tomography (μCT) in the absorption-contrast mode is well-established for hard tissue visualisation. However, performance in relation to lower density materials, such as post mortem brain tissues, is questionable, as attenuation differences between anatomical features are weak. It was demonstrated, through the example of a formalin-fixed paraffin-embedded (FFPE) human cerebellum, that absorption-contrast laboratory-based micro computed tomography can provide premium contrast images, complementary to hematoxylin and eosin (H&E) stained histological sections. Based on our knowledge, the detection of individual Purkinje cells without a dedicated contrast agent is unique in the field of absorption-contrast laboratory-based micro computed tomography. As the intensity of H&E staining of histological sections and the attenuation contrast of LBμCT data demonstrated a correlation, pseudo colouring of tomography data according to the H&E stain can be performed, virtually extending two-dimensional (2D) histology into the third dimension. The LBμCT of FFPE samples can be understood as a time-efficient and reliable tissue visualisation methodology, and so it could become a method of choice for imaging of relatively large specimens within the laboratory environment. Comparing the data acquired at the LBμCT system nanotom® m and synchrotron radiation facilities (Diamond-Manchester Imaging Branchline I13-2, Diamond Light Source, Didcot, UK and Microtomography beamline ID19, ESRF), it was demonstrated that all selected modalities, namely LBμCT, synchrotron radiation-based in line phase-contrast tomography using single-distance phase reconstruction and synchrotron radiation-based grating interferometry, can reach cellular resolution.
As phase contrast yields better data quality for soft tissues, and in order to overcome the restrictions of limited beamtime access for phase-contrast measurements,a commercially available advanced μCT system nanotom® m was equipped with an X-ray double-grating interferometer (XDGI). The successful performance of the interferometer in the tomography mode was demonstrated on a human knee joint sample. XDGI provided enough contrast (1.094 ± 0.152) and spatial resolution (73 ± 6) μm to identify the cartilage layer, which is not recognised in the absorption mode without staining. These results suggest that the extension of a commercially available absorption-contrast μCT system via grating interferometry offers the potential to fill the performance gap between LBμCT and phase-contrast μCT using synchrotron radiation in the visualising soft tissues.
Although optical microscopy of stained tissue sections enables the quantification of neuron morphology within brain tissues in health and disease, the lateral spatial resolution of histological sections is limited to the wavelength of visible light, while the orthogonal resolution is usually restricted to the section´s thickness. Based on the data acquired from the ID16A-NI, the study demonstrated the application of hard X-ray nano-holotomography with isotropic voxels down to 25 nm for the three dimensional visualising the human cerebellum and neocortex. The images exhibit a reasonable contrast to noise ratio and a spatial resolution of at least 84 nm. Therefore, the three dimensional data resembles the surface images obtained by electron microscopy (EM), but in this case electron dense staining is avoided. The (sub-)cellular structures within the Purkinje, granule, stellate and pyramidal cells of the FFPE tissue blocks were resolved and segmented. Micrometre spatial resolution is routinely achieved at synchrotron radiation facilities worldwide, while reaching the isotropic 100-nm barrier for soft tissues without applying any dedicated contrast agent, labelling or tissue-transformation is a challenge that could set a new standard in non-destructive 3D imaging.
Advisors:Müller, Bert
Faculties and Departments:03 Faculty of Medicine > Departement Biomedical Engineering > Imaging and Computational Modelling > Biomaterials Science Center (Müller)
UniBasel Contributors:Müller, Bert
Item Type:Thesis
Thesis Subtype:Doctoral Thesis
Thesis no:12501
Thesis status:Complete
Number of Pages:1 Online-Ressource (102 Seiten)
Identification Number:
edoc DOI:
Last Modified:08 Feb 2020 14:42
Deposited On:22 Mar 2018 09:27

Repository Staff Only: item control page