Three dimensional multicolor direct Stochastic Optical Reconstruction Microscopy and Applications in Membrane Biochemistry
Lampe, André

HaupttitelThree dimensional multicolor direct Stochastic Optical Reconstruction Microscopy and Applications in Membrane Biochemistry
TitelvarianteDreidimensionale Mehrfarben direkte stochastische optische Rekonstruktonsmikroskopie und Anwendungen in der Membran-Biochemie
AutorLampe, André
Geburtsort: Herford
GutachterDr. Jan Markus Schmoranzer
weitere GutachterProf. Dr. Volker Haucke
Freie Schlagwörtersuper-resolution; dSTORM; spectral demixing; 3D; multi-color; noise-reduction; temperature stabilization; microscopy; microscope
DDC572 Biochemie
ZusammenfassungFluorescence microscopes are limited by the diffraction barrier of light. This limit is ultimately defined by the wave-nature of light itself. Approximately 200 nm is the resolution limit for fluorescence microscopes using visible light. In this thesis 3D SD-dSTORM (3D spectral demixing direct stochastic optical reconstruction microscopy) is described, which is capable of resolving structures with sizes below the diffraction barrier. While dSTORM is already a widely used technique, spectral demixing is a novel approach to multicolor super-resolution and its development represents a major part of this thesis. When splitting the emission of a single molecule signal spectrally, a pair of localizations will be detected. Using two different fluorophores, the ratio of the detected intensities of each localization pair should be different and allows a distinction between the two fluorophores. By this, multicolor super-resolution without registration errors or aberration is achieved. How the physical hardware for this work was set up is described in detail and the software needed to acquire and analyze the data is discussed. A conceptual introduction to spectral demixing is given and the capabilities and functionality of the developed software to analyze the data is shown. Multicolor imaging is not the only benefit of spectral demixing, but also its noise-reducing potential is demonstrated. Furthermore, the registration error-free reconstruction and low crosstalk of multicolor data is presented. Those three features are rarely found in other super-resolution approaches. Spectral demixing is compatible with existing solutions to super-resolve structures in three dimensions. In this thesis, the approach to 3D by astigmatism with spectral demixing is demonstrated and the required temperature stabilization of the setup is discussed. The capability of 3D SD-dSTORM is demonstrated on examples from membrane biochemistry. Cluster sizes below the diffraction barrier of STOML3 in sensory neurons could be measured, the spatial correlation of GIT1 and bassoon at the neuronal active zone could be depicted as well as SNX9 rich tubular structures attached to clathrin coated pits in dynamin2 depleted cells. The versatility of the setup was demonstrated by using it as a single molecule total internal reflection microscope investigating the oligomerization-state of GPCRs on the plasma membrane. Finally the 3D SD-dSTORM is compared and positioned in relation to the multitude of current super-resolution techniques, acknowledging its novelty and advantages but not concealing the specific drawbacks, which every super-resolution approach possesses. An extensive discussion of improvements is given, as well as a comprehensive review of potential biological targets.
InhaltsverzeichnisAffidavit 4
Acknowledgments 5
Table of contents 6
1. Summary 11
2. Zusammenfassung 13
3. Introduction 15
3.1. Microscopy and biology 15
3.1.1. The diffraction barrier of microscopes 15
3.1.2. Light and the excitation of electrons 16
3.1.3. Fluorescence 18
3.1.4. Fluorescent ligands 22
3.2. Super-resolution - circumventing the diffraction barrier 24
3.2.1. Single molecule localization-based super-resolution microscopy 24
3.2.2. Stimulated emission depletion 29
3.2.3. Structured illumination microscopy30
3.2.4. Other approaches 30
3.3. Labeling density and sampling rate in SMLM 32
3.4. Multicolor and three dimensions in SMLM 34
3.4.1. Multicolor 34
3.4.2. Three dimensions 35
3.5. Limitation of current super-resolution approaches 37
4. Materials & Methods 41
4.1. Materials 41
4.1.1. Chemicals and consumables 41
4.1.2. Tissue culture 41 NIH 3T3 cell line 41 BS-C-1 cell line 41
4.1.3. Antibodies 42
4.1.4. Fluorescent ligands 42 Fluorophores coupling to antibodies 43 Degree of labeling (DOL) 44
4.1.5. Buffers, media and solutions 44
4.1.6. Staining protocol 46
4.2. Devices and equipment 46
4.3. Microscopy 47
4.3.1. Cover slip preparation 47
4.3.2. Fluorescent bead immobilization on cover slips 48
4.4. Software 48
5. Results 51
5.1. The SD-dSTORM Setup 51
5.1.1. Layout of the microscope 51
5.1.2. Sample drift 56
5.2. Multicolor dSTORM by spectral demixing 58
5.2.1. Spectral demixing and pair-finding 58
5.2.2. The effect of heavy water (D 2 O) 64
5.2.3. Accuracy of multicolor registration 66
5.2.4. Offset optimization for the pair-finding algorithm 66
5.2.5. Improved signal-to-noise with pair-finding 69
5.3. Three dimensions 72
5.3.1. 3D by astigmatism 72
5.3.2. Temperature stability 78
5.4. Experimental localization precision and optimally obtainable resolution 82
5.5. Data structure 83
6. Cell biology applications 87
6.1. GIT1 at the neuronal active zone 87
6.2. SNX9 recruitment 89
6.3. STOML3 in sensory neurons 90
6.4. GPCRs and the setup as a single molecule TIRF microscope 93
7. Discussion 97
7.1. Advantages and limitations of the SD-dSTORM platform 97
7.1.1. Multicolor with SD-dSTORM 99 More than two colors 99
7.1.2. Fluorescent dyes, buffers and labeling density 100 Different labeling strategies 101
7.1.3. Three dimensions 102 Thermal stability 103 Biplane vs. astigmatism 104
7.1.4. Data analysis 105
7.1.5. A comparison to electron microscopy 106
7.2. Advantages of using SD-dSTORM in cell biology 106
7.2.1. Membrane cell biology 106
7.2.2. Further biological targets 107
7.3. Further optimization of SD-dSTORM 109
7.3.1. Camera 109
7.3.2. Higher laser powers 109
7.3.3. Automated color filters and time analysis 110
7.3.4. Multi-candidate pairs 110
8. Conclusion and outlook 113
List of figures 115
List of tables 123
Nomenclature 124
Bibliography 128
Appendix 147
A. Publications by the author 147
B. SDmixer software 149
C. Script for laser control 151
PDF-Datei von FUDISS_thesis_000000104244
Falls Ihr Browser eine Datei nicht öffnen kann, die Datei zuerst herunterladen und dann öffnen.
Seitenzahl159 Seiten
Fachbereich/EinrichtungFB Biologie, Chemie, Pharmazie
Rechte Nutzungsbedingungen
Tag der Disputation21.11.2016
Erstellt am21.02.2017 - 14:22:29
Letzte Änderung24.02.2017 - 10:19:49
Statische URLhttp://www.diss.fu-berlin.de/diss/receive/FUDISS_thesis_000000104244