Home

thick brain slice cultures and a custom-fabricated

1. Absorption Vibrational 10 Seconds fete States Internal Internal 2 Conversion Conversion and Vibrational S Relaxation 1014 1011 Sec SA 3 2 A 0 Delayed Fluorescence o anum 1 Fluorescence 9 ag Intersystem ae i Crossing if Intersystem Crossing Non Radiative ar Relaxation Triplet Quenching RA Phosphorescence Non Radiative 5 C989 38a Relaxation o GA o shen OR SA A Ground State Figure 3 2 Jablonski Energy diagram to show fluorescence fundamentals A fluorescent molecule has ground state S0 and excited singlet states S1 S2 and triplet states T1 The singlet and triplet states have several vibrational energy states Several closely located electronic transitions are possible from different vibrational energy levels of ground state to different vibrational energy level of excited states and vice versa These possibilities result in the spectral bandwidth of the excitation and emission spectra of the molecule Picture Courtesy Prof Michael Davidson Absorption Fluorescence emission 350 400 450 500 550 600 650 Wavelength nm Figure 3 3 An Example of Excitation and Emission spectra of a fluorophore Each fluorophore has an excitation and an emission spectra Generally in one photon excitation the excitation wavelengths are smaller than the emission wavelengths data courtesy Invitrogen Data Fluorescein molecule 20 ONE PHOTON VERSUS MULTIPHOTON EXCITATION
2. 110 26 Similarly slice the other hemispehere too and transfer it to the chilled nutrient medium From harvesting brain to cutting slices should not take more than 3 5 minutes to ensure high viability tissue to start with 27 Using two blunt microspatulas separate the slices and leave them in the chilled nutrient medium 28 Using a biopsy tool of appropriate diameter same as infusion chamber for best fit cut the slices into round discs Transfer these round discs of slices to a new dish containing the chilled nutrient medium and enclose it with a teflon lid to transfer the tissue to the laminar hood to ensure sterility Remove the teflon lid from the culture chamber 29 Very gently remove any excess laminin in the infusion chamber 30 Using two flat microspatuals gently transfer the tissue to the infusion chamber Make sure that there are no micro cuts accidentally happened to the tissue during this time The micro cuts will form path of low resistance for the infused medium resulting in hampering of perfusion of medium thorough the tissue thickness 31 Fill the withdrawal chamber with the nutrient medium and close the chamber with a teflon lid 32 Start the nutrient medium flow from the syringe pump immediately 33 Transfer the tissue in the control experiment chambers and the membrane insert chamber that are floating on the nutrient medium Close the culture dishes with the appropriate teflon lids 111 34
3. Development of a method to culture thick brain slices and a 3D microfluidic microelectrode neural interface system 10 will eventually enable us to define stimulation and recording from different cortical layers and would be a more in vivo like model yy M y 9 F j l WD38 9mm 20 0kV x25 a electrodes Fluidic Channel Fluidic ports Figure 2 5 Three dimensional microfluidic multielectrode arrays A Right Prototype structure of a three dimensional microfluidic multielectrode array for brain slice cultures Left picture of two towers with microfluidic ports and microelectrodes The size of the fluidic ports increases gradually with height of the tower to maintain uniform flow as a result of pressure drop The fluidic towers are 1mm tall hollow structures that contain fluidic port from bottom to top at every 100um The towers taper from bottom to top with tip dimension of 24umx80um B Right A prototypic structure of a three dimensional microfluidic multielectrode array with cross bars to support three dimensional dissociated neuronal cultures in bioactive gel scaffolds Left enlarged microtowers showing fluidic side ports fluidic channels and microelectrodes Image courtesy Laura Rowe BRP collaborator 11 SS i ONY Wy ae AN N AALL i ANNANN NY Figure 2 6 Pyramidal three dimensional microfluidic multielectrode arrays A Right prototypic s
4. Neuronal transfection in brain slices using particle mediated gene transfer Neuron 13 1263 1268 1994 70 London J A Biegel D and PACHTER J S Neurocytopathic effects of beta amyloid stimulated monocytes a potential mechanism for central nervous system damage in Alzheimer disease Proc Natl Acad Sci USA 93 4147 4152 1996 71 LUKSCH H GAUGER B and WAGNER H A Candidate Pathway for a Visual Instructional Signal to the Barn Owl s Auditory System J Neurosci 20 RC70 1 4 2000 72 LUNDSTROM K ET AL Semliki Forest virus vectors efficient vehicles for in vitro and in vivo gene delivery FEBS Lett 504 99 103 2001 73 LUNDSTROM K ABENAVOLI A MALGAROLI A and EHRENGRUBER M U Novel Semliki Forest virus vectors with reduced cytotoxicity and temperature sensitivity for long term enhancement of transgene expression Mol Ther 7 202 209 2003 74 MACLEAN J WATSON B AARON G and YUSTE R Internal Dynamics Determine the Cortical Response to Thalamic Stimulation Neuron 48 811 823 2005 125 75 MACLEAN JN FENSTERMAKER V WATSON BO and YUSTE R A visual thalamocortical slice Nat Methods 3 2 129 34 2006 76 MAJEWSKA A YIU G and YUSTE R A custom made two photon microscope and deconvolution system Pflugers Arch 441 2 3 398 408 2000 77 MALETIC SAVATIC M MALINOW R and SVOBODA K Rapid dendritic morphoge
5. 67 69 69 70 70 71 71 13 75 78 78 79 80 82 85 85 85 87 87 88 88 89 89 89 RESULTS 90 Viability of thick brain slice cultures over time 90 Organotypic organization of thick cortical slice cultures 91 Thickness preservation 93 Electrophysiological activity of cultured thick brain slices 94 DISCUSSION AND CONCLUSIONS 95 8 THICK BRAIN SLICE CULTURING METHOD RECOMMENDATIONS AND DIRECTIONS FOR FUTURE WORK 98 OPTIMIZATION OF PERFUSION PARADIGM 98 QUANTITATIVE ANALYSIS OF ORGANOTYPIC ORAGANIZAION 99 CHARACTERIZATION OF THICK BRAIN SLICE CULTURES FOR LONGER TERM 99 QUANTITATIVE ANALYSIS OF ELECTROPHYSIOLOGY 100 CHARACTERIZATION OF 1MM THICK BRAIN SLICES 100 MODIFICATIONS OF CULTURING CHAMBER 101 APPENDIX A A step by step user manual to operate the custom made multiphoton microscope 102 APPENDIX B A trouble shooting manual for the custom made multiphoton microscope 106 APPENDIX C A detailed protocol to set up the fluidic system and the thick brain slice culturing method 108 APPENDIX D Labeling and imaging the brain slice cultures for viability assessment and organotypic organization 113 APPENDIX E Data analysis using the imageJ software 115 APPENDIX F Published works 117 REFERENCES 120 LIST OF TABLES Page Table 3 1 Comparison of multiphoton and single photon imaging 24 Table 3 2 Advantages and disadvantages of multiphoton microscopy over confocal microscopy 25 Table 3
6. Autonomous Robots 11 305 310 2001 21 DEMARSE T B WAGENAAR D A BLAU A W and POTTER S M Interfacing neuronal cultures to a computer generated virtual world Proc 7th Joint Symposium on Neural Computation USC 36 42 2000 22 DENK W STRICKLER J H and WEBB W W Two photon laser scanning fluorescence microscopy Science 248 73 76 1990 23 DE PAOLA V ARBER S and CARONI P AMPA receptors regulate dynamic equilibrium of presynaptic terminals in mature hippocampal networks Nat Neurosci 6 491 500 2003 121 24 DE PAOLA V HOLTMAAT A KNOTT G SONG S WILBRECHT L CARONI P and SVOBODA K Cell type specific structural plasticity of axonal branches and boutons in the adult neocortex Neuron 49 861 875 2006 25 DE SIMONI A GRIESINGER C B and EDWARDS F A Development of rat CA1 neurones in acute versus organotypic slices role of experience in synaptic morphology and activity J Physiol 550 135 147 2003 26 DONG H W and BUONOMANO D V A technique for repeated recordings in cortical organotypic slices J Neurosci Methods 146 69 75 2005 27 DOUPE A J SOLIS M M KIMPO R and BOETTIGER C A Cellular Circuit and Synaptic Mechanisms in Song Learning Ann N Y Acad Sci 1016 495 523 2004 28 EGERT U OKUJENI S NISCH W BOVEN K H RUDORF R GOTTSCHLICH N and STETT A Optimized Oxygen Availab
7. In one photon excitation the electrons in the ground state absorb the impinging photons of energy equivalent to the energy band gap to reach the excited state These electrons eventually come back to ground state by releasing excess energy as photons of longer wavelengths figure 4B In 1930 Nobel Laureate Maria Goeppart Mayer first predicted in her theoretical calculations the possibility of fluorophore excitation by absorption of multiple photons of total sum energy equal to energy band gap figure 4B 44 The absorption of two photons simultaneously by the electrons in the ground state of fluorophores requires very high density of photons Owing to this condition the excitation of the fluorophore occurs only at the focal volume of the specimen compared to overall excitation of the specimen illuminated in one photon excitation figure 3 4 21 excitation excitation one photon two photon Figure 3 4 One photon and two photon excitation A Picture showing excitation of a fluorescent solution in a cuvette with one photon and two photon white arrow excitation Picture Courtesy Brad Amos B A cartoon depicting the excitation region of specimen with one photon and two photon excitation Cartoon by Author PULSED LASER FOR MULTIPHOTON EXCITATION Two photon excitation requires two photons of half the excitation energy double wavelength of the fluorophore to impinge on its electrons concurrently and get absorbed With
8. SU8 fluidic tower a Figure 5 11 A section of fluidic tower of three dimensional microfluidic neural interface device The device was tested for fluidic functionality using dilute fluorescent bead solution The flow rate was set to 1 ml hr This experiment demonstrated ability to accommodate complex experimental set ups and perform simultaneous imaging Inset SEM image of the prototypic microfluidic array device DISCUSSION AND DIRECTIONS FOR FUTURE WORK Here we demonstrated several common tests to evaluate the performance of our custom made imaging system With this system we could successfully image living and non living in vitro preparations in 2D 3D and time lapse mode However many more 64 features could be improved or added to this system In near term goals it is desired to test the system for 4D imaging X Y Z t Large two photon excitation cross section of several fluorophores and spectral bandwidth of the laser pulses allow excitation of multiple fluorophores with overlapping two photon excitation spectrum A carefully selected combination of such fluorophores allows multicolor red green blue imaging of specimens during one scan This feature greatly enhances the ability of long term time lapse imaging of living specimens with minimal photodamage to the specimen Currently the system is equipped with only one detection channel Addition of two more detection channels will greatly increase the utility of this m
9. The cultures plated on unperfused microperfusion devices control experiments were placed in a small culture dish containing nutrient medium In these control experiments medium supply is diffusion based from the bottom of the gold grid analogous to membrane insert method After two days of perfusion the viability of cultures was assessed by multiphoton imaging of the slices labeled with Hoechst and Propidium Iodide fluorescent nuclear labels Appendix D The results of my experiments indicate a statistically significant increase in viability of the perfused cultures compared to the unperfused sister cultures figure 6 4 78 100 90 80 70 60 50 40 Viability 30 20 10 Membrane Unperfused Perfused Figure 6 4 Perfusion of nutrient medium through the tissue thickness results in enhanced tissue viability Cortical slices of 700m thickness were cultured on microperfusion devices and on membrane inserts using Stoppini s method The unperfused microperfusion device cultures and membrane insert cultures were taken as control cultures Figure shows Mean SEM n 4 of each case viability of these cultures assessed after 2DIV The perfusion rate was 10ul hr A generalized linear model ANOVA test followed by Tukey s multiple comparison test was used to evaluate statistical significance The perfused cultures showed statistically significant increased in viability compared to the control sister cultures p lt 0 05 Optim
10. and Kuniyoshi Y New York Springer 3139 130 145 2004 7 BARRIA A and MALINow R NMDA receptor subunit composition controls synaptic plasticity by regulating binding to CaMKII Neuron 48 289 301 2005 8 BEACH R L BATHGATE S L and COTMAN C W Identification of cell types in rat hippocampal slices maintained in organotypic cultures Brain Res 255 3 20 1982 9 BECQ H BOSLER O GEFFARD M ENJALBERT A and HERMAN J P Anatomical and functional reconstruction of the nigrostriatal system in vitro selective innervation of the striatum by dopaminergic neurons J Neurosci Res 58 553 566 1999 10 BENEDIKTSSON A M SCHACHTELE S J GREEN S H and DAILEY M E Ballistic labeling and dynamic imaging of astrocytes in organotypic hippocampal slice cultures J Neurosci Methods 141 41 53 2005 11 BINDOKAS V P LEE C C COLMERS W F and MILLER R J Changes in Mitochondrial Function Resulting from Synaptic Activity in the Rat Hippocampal Slice J Neurosci 18 12 4570 4587 1998 120 12 BREWER G J and COTMAN C W Survival and growth of hippocampal neurons in defined medium at low density advantages of a sandwich culture technique or low oxygen Brain Research 494 65 74 1989 13 Bucus P A and MULLER D Induction of long term potentiation is associated with major ultrastructural changes of activated synapses Proc Nat
11. and robot body XV CHAPTER 1 INTRODUCTION The brain controls our actions by two way communication with the body and its interaction with the environment Since the past decade it is becoming more widely accepted that behaviors are encoded in particular activity patterns that underlie the cytoarchitecture of networks of neurons in the brain 18 25 27 32 33 34 However the correlation between the two properties the activity patterns and the morphological connectivity underlying a behavior is not understood This makes one of the fundamental questions in neuroscience At present current technologies do not allow high resolution multisite imaging and electrophysiology simultaneously on in vivo preparations Further use of any of these techniques is limited to restricted anesthetized or non behaving model animal preparations A hybrid system of a living neuronal network interfaced to a robotic body HYBROT via an electronic interface may provide a simpler in vitro model to study network properties underlying learning and memory using multiple non invasive technologies simultaneously Microelectrode arrays MEA and multiphoton laser scanning imaging MPLSM methods provide the means to study functional and morphometric network properties simultaneously using a HYBROT model Brain slice cultures preserve in vivo like cyto architecture and offer several advantages over in vivo models due to easy tissue accessibility and controllabilit
12. 0 01 compared to membrane and unperfused controls indicate p lt 0 05 compared to membrane and unperfused controls Viability of organotypic thick brain slice cultures after 5 DIV I further assessed the viability of the cultures after 5 days of continuous perfusion of nutrient medium at the non destructive range of flow rates that was determined in the 80 2DIV series of experiments The viability was assessed using the same methods as explained in the optimal flow rate assessment results The results indicate greater than twice the viability of the perfused cultures compared to the unperfused cultures Even after 5 days of perfusion the flow rate 20ul hr proves to be the most suitable flow rate for enhanced viability of the culture figure 6 6 This flow rate translates to three culture volume exchanges per hour 100 90 80 70 60 50 40 Viability 30 20 10 Unperfused Membrane 5ul fnr 10xl hr 20ul hr Figure 6 6 Viability of cultured brain slices after 5 days in vitro Thick brain slices show enhanced viability at various non invasive flow rates 5 10 20ul hr compared to unperfused cultures and standard membrane insert based cultures after 5 days of perfusion At 20ul hr flow rate gt 80 viability was observed even after 5 days in culture and proved to be the optimal flow rate for sufficient nutrient supply through the thickness of the tissue The viability is plotted MeantSEM n 3 4 for various flo
13. 35 36 Gently transfer the perfusion set up and the control cultures in the culture incubator that is set at 5 CO2 9 Oz 65 relative humidity and 35 C temperature Clean the dissection hood and discard the biohazards appropriately After completing the experiments clean the capillary tubing and chambers immediately with deionized water till all the medium is clearly exited from them Leaving them even for a couple of hours without cleaning allows contents of the nutrient medium to dry and block them Similarly clean the chambers and preferably store them in deionized water till next use This not only helps to prevent blocking of the grid with tiny particles or molecules but also helps to keep the hydrophobic nature of PDMS manifold material low This ensures slightly reduced experimental set up time for next experiment 112 APPENDIX D LABELING AND IMAGING THE BRAIN SLICE CULTURES FOR VIABILITY ASSESSMENT AND ORGANOTYPIC ORGANIZATION LABELING THE CULTURE FOR THE VIABILITY ASSESSMENT 1 Take out appropriate number of the aliquots of the propidium iodide and the hoechst stains and keep them in light tight condition in the laminar flow hood Allow them to thaw and adjust to the room temperature 2 Gently transfer the set up to the laminar hood without hindering disturbing the perfusion equipment 3 Mix one aliquot of each hoechst and propidium iodide in 2001 per culture chamber nutrient medium The nutrient medium sho
14. FABRICATED MULTIPHOTON LASER SCANNING IMAGING SYSTEM ABSTRACT INTRODUCTION SYSTEM VALIDATION OPTICAL PROPERTIES Achievable scanning speed Field of view and scan angle Excitation pathway efficiency Chirping of laser pulses Illumination of the image across the field of view SYSTEM VALIDATION IMAGING Two dimensional 2D imaging Three dimensional 3D imaging Time lapse imaging Compatibility with multiple techniques DISCUSSION AND DIRECTIONS FOR FUTURE WORK 6 CULTURING THICK ORGANOTYPIC BRAIN SLICES A PERFUSION BASED CULTURING METHOD viii 34 36 36 37 41 43 46 48 50 50 50 51 51 52 55 56 57 58 58 60 62 63 64 67 7 ABSTRACT INTRODUCTION MATERIALS AND METHODS Brain slice culture Adhesion methods Viability assessment DESIGN Convective flow based culturing method The microfluidic chamber and the closed loop perfusion set up Flow trajectories in the infusion chamber RESULTS Perfusion allows enhanced culture viability Optimal perfusion rates for enhanced viability Viability of organotypic brain slice cultures after 5 DIV DISCUSSION AND CONCLUSIONS CHARACTERIZATION OF THICK ORGANOTYPIC CORTICAL SLICE CULTURES ABSTRACT INTRODUCTION MATERIALS AND METHODS Brain slice cultures and perfusion set up Adhesion methods Viability assessment Tissue fixing and H amp E staining Functional activity recording Microwire electrode set up 67
15. ImageJ software In the last appendix abstracts of published works related to these projects are listed CHAPTER 2 BACKGROUND The human brain is a remarkably complex organ It not only produces and controls one s thoughts actions memories feelings experiences and habits but also plays vital role in voluntary and involuntary control of the vital organs to lead a normal life How the brain encodes learning and reliably retrieves information memory underlying various simple and complex behaviors is one of the fundamental questions in neuroscience research In the past few decades scientists have developed several techniques and experimental models both in vivo and in vitro to study various properties underlying the functioning of the brain From a century of neuroscience investigations it is evident that every behavior is encoded in a network of several neurons On one side several behavioral studies have shown morphological changes underlying learning and memory 25 32 39 52 on the other hand several in vivo functional recordings have shown characteristic activity patterns related to a behavior 18 33 34 Likewise other in vitro studies show there are morphometric changes associated with activity of neurons 13 At present we recognize several mechanisms underlying neuronal plasticity such as long term potentiation LTP long term depression LTD spike timing dependent plasticity STDP etc Current technologies d
16. Two Dimensional 2D Imaging We tested this system for 2D two photon imaging with specimen labeled with various fluorescent labels Figure 5 6 is an image of neuronal culture that was fixed after 16 days in vitro The culture is labeled with Alexa488 that is attached to antibodies 58 against microtubule associated protein MAP2 Figure 5 7 is image of living dissociated mouse cortical neuronal cultures 2 DIV that was labeled with Calcein AM fluorescent indicator The culture was imaged alive with 800 nm excitation wavelength These images indicate the ability of the microscope to image fine dendritic and tiny spine structures Figure 5 6 Multiphoton image of fixed dissociated rat cortical neuronal network The network of dissociated neurons is fixed and labeled with Alexa 488 to MAP2 with water immersion 40X lens N A 0 8 Image courtesy Mark Booth 40x W 0 8 NA excitation wavelength 800 nm 59 Figure 5 7 Multiphoton image of living mouse cortical neuronal network labeled with Calcein AM The dissociate culture is 2 days in vitro The image is taken with a 40x W N A 0 8 objective lens The frame size is 512x512 pixels The laser was tuned to 800 nm wavelength Three dimensional 3D Imaging Further we tested the system for three dimensional imaging A pollen grains slide was used to image auto fluorescence of pollen grains The specimen was excited at 800 nm wavelength and images of a spiny pollen grain were ca
17. a Ti saph Mira900 laser across its entire tuning range when pumped with pump lasers Verdi of different powers 37 Figure 4 5 Transmission curve of the custom made dichroic mirror with a sharp cut off at 700nm wavelength 39 Figure 4 6 Sensitivity of the detector 41 xii Figure 4 7 Schematic diagram of synchronized scanning and data acquisition control to form basis for software control 45 Figure 4 8 Screen shot of software user interface to operate the microscope in various imaging modes 46 Figure 4 9 Environment control chamber 47 Figure 4 10 Custom fabricated multiphoton microscope 48 Figure 5 1 Speed of scanning mirrors at different angles of deflection 52 Figure 5 2 Montage of hemocytometer grid images taken at different scan angles 54 Figure 5 3 Excitation pathway efficiency to transport laser power at focal plane 56 Figure 5 4 Spatial and temporal properties of laser pulse before the microscope and at the focal plane 57 Figure 5 5 Illumination across the field of view 58 Figure 5 6 Multiphoton image of fixed dissociated rat cortical neuronal network 59 Figure 5 7 Multiphoton image of living mouse cortical neuronal network 60 Figure 5 8 z projection of a pollen grain 6l Figure 5 9 z projection of hippocampal slice labeled with nuclear stain 62 Figure 5 10 Montage of time lapse images of living neuronal culture 63 Figure 5 11 A section of fluidic tower of three dimensional microfluidic neural interface system
18. analyze particles is used to adjust the range of the size of the particles to be counted This command returns detailed analysis of the particles and their size distribution One flaw in this analysis is that when the two nuclei are very close the binary image converts it into a single unit and the software counts them as single entity However the number of such nuclei is only fractional part of the total number of nuclei 115 The data was visually inspected for each frame of stack to omit possibility of wrong threshold criterion that may result from automated threshold by software Ea 9 k go 8 s 2 6 5s 8 s 5 go s 6 eg 8 s 6 a 8 4 u a oe O88 x oge s n By a 6 ogle oy amp dy ae 8 8 aes oS y Pa Sp m 8 a M v s aiik ga b 8 1e Bs 5 oO w 2 N p OS By EN 0O Gey ep S a a bd ey p te f v 18 1 a i y rag Be ge Boge P p we i we N ard i o i Fe ws vor 1 FS ag amp Ae i FO wg j b g 18 hy w os 8 18 we ws ka ah poe v ER B m 18 182 Sa 194 m E yu e FY o P g S g w Se S oo Yi 18 we ow B B y oes Boy A age fy al je 1 1 45 A 0 Be 27 E 3 v O Bie tat B a XP aay Zager a we Be eS ca 8 eee 5 4 q a es 8 ka By ae S wE y a a w a a 2 5 _ aH Waa iu 2w Cu ee OF Figure E 1 An Example of data analysis using ImageJ software Data was collected in two different detector channels Blue Hoechst and Red Propidium iod
19. cylinder The gold grid is 3mm in diameter and 50um in thickness with a 54 square micrometer pores forming a total fluid transmission area of 40 of its total area figure 6 3A To allow efficient 3D transport of the nutrient medium inside the chamber volume there are 350um deep microchannels in the wall of the infusion chamber that are separated 50um from each other 119 The infused medium from the infusion port located at the bottom of infusion chamber passes through the tissue that is adhered to the gold grid in the infusion chamber and exits to the outer cylinder referred as the withdrawal chamber The used nutrient medium is withdrawn from the outlet port 173 located in one side of the withdrawal chamber The sterility in the chamber is maintained using a transparent semi permeable FEP membrane containing teflon lid 97 The FEP membrane is permeable to gases but not to liquids which allows exchange of O2 and CO2 gases between the incubator environment and the nutrient medium across the membrane while preventing the evaporation of the medium from the chamber This helps to maintain pH of the medium and the osmolarity inside the chamber and also prevent infection To operate in the infusion withdrawal function the chamber is attached to a push pull type syringe pump KD Scientific Inc using FEP capillary tubing as separate infusion and withdrawal lines and associated micro connectors The infusion line contains an aerator close to t
20. example as in the Biorad scanhead design This makes alignment of the system cumbersome and additional optical elements add to chirping of the laser pulses These intermediate optical elements can be safely eliminated by using 2D orthogonal scanning mirrors mounted with a very small separation distance These two mirrors could be imagined as a single mirror moving in the XY direction with its center at the mid point of 37 these two mirrors To scan the laser beam fast in a raster pattern we have chosen a top of the line 2D scanning mirror system Model 6215H Cambridge Inc including driver boards These mirrors have small size x scan mirror 3mm and y scan mirror 5mm and the separation between them is only 5 2 mm Their small size translates to less mass which helps them to move faster A scan lens is a critical optical element which determines the uniform delivery of laser pulses across the raster scan A plan apo stereomicroscope objective lens with a large 8cm diameter would offer an equal thickness to the pulses of the laser beam moving in a raster pattern 118 This would stretch the laser pulses to almost the same extent across the raster pattern for more uniform excitation Currently we are using F 60mm D 25 4mm near IR achromatic doublet AC254 060 B Thorlabs Inc An 80mm stereoscope objective lens from Zeiss might be an appropriate option in the future due to its longer focal length 63 5cm compared to the scan
21. generated from these inputs Ideally these should not be in direct control of the user I wanted to modify it for more intuitive user inputs that translate in the background LabView code to the appropriate settings for the raster scan I suggest incorporation of this feature in the software as an essential near term goal Inappropriate settings may lead to damage of the scanning mirrors due to discontinuous waveform that inappropriate settings will generate 18 Other operations such as time lapse 2D imaging z stack 3D imaging can also be done using appropriate settings from step 2 on the image A 1 The corresponding desired inputs for the z step settings the time lapse settings and the XY position settings can be done from the Z series control panel the time lapse control panel and the position control panel respectively 19 Always set the PMT VOLTAGE to zero before stopping the software 20 Always press stop or stop all button before quitting LabView Otherwise the scanning mirrors will keep running due to the stored waveforms on the FIFO memory of the digital to analog cards The commands stop or stop all force 104 all the buffers the counters the clocks and the memories on all the interfacing cards to be clear to use them again in the next scanning imaging session 21 After stopping and quitting the program NEVER save any changes asked by the prompt window o
22. in diameter and diverges 1 7 mrad slightly which translates to increasing beam diameter as a function of distance from the laser exit aperture The smaller size of the scanning mirror 3mm and the requirement to overfill the objective lens back aperture makes it necessary to reshape and collimate the laser beam along its excitation path This is achieved by a two lens telescope beam expander focal length 100mm and 80mm before and two lens telescope the scan lens tube lens after the scanning mirrors Further a Pockels cell is required to modulate or block the laser intensity to avoid any unwanted exposure of specimen to the laser light during the raster scan regions where signal is not collected Together these additional but essential optical elements introduce group velocity dispersion of the pulses of laser Multilayer dielectric coatings on reflecting mirrors and antireflection coatings on the transmission optics also introduce chirping of laser pulses The pre chirp unit is a two prism and a retrograde mirror system SF10 glass CVI Lasers Inc that will be used to compensate for GVD caused by the optical elements in the excitation pathway For four dimensional XYZ t imaging it is required to move the objective lens or the specimen in the z direction We prefer to move the objective lens in the fine z steps because our specimens are coupled to heavy electrical recording and stimulation hardware A coarse motorized Z translator quickly
23. index n 1 phase velocity of all the wavelengths is same however in the materials with refractive index higher than air n gt 1 it varies for different wavelengths This results in non uniform distribution of different 23 frequencies wavelengths in the pulse called group velocity dispersion GVD or chirping figure 3 6 Group velocity dispersion can be linear or complex 1 45 Air n l n gt 1 glass Figure 3 6 Laser pulse andGroup Velocity Dispersion GVD of laser pulse after passing through glass Femtosecond laser pulses have spectral width Different wavelengths travel at different speeds inside a medium of refractive index different from air Therefore pulses get distorted Picture Courtesy Prof Rick Trebino FLUORESCENCE MICROSCOPY In the past few decades fluorescence microscopy has evolved from conventional to confocal and multiphoton laser scanning type microscopes LSM for three dimensional imaging Table 3 1 shows a brief comparison of few factors among conventional confocal and multiphoton imaging modes TABLE 3 1 COMPARISON OF MULTIPHOTON AND SINGLE PHOTON IMAGING MICROSCOPY MODE CONVENTIONAL CONFOCAL MULTIPHOTON Usually halogen arc EXCITATION SOURCE lamp could be CW CW Ar He Ne UV PULSED IR tunable laser 3D IMAGING No Yes Yes High Slightly Better High Slightly poorer iaepsiaaaks Average than multiphoton Than confocal PHOTOTODAMAGE High High Much less SIGNAL NOISE RATIO Poor Good
24. maximum possible fluorescent signal using minimum excitation laser intensity Special attention is paid to get uniformly illuminated images and the ability to use the entire bandwidth of the pulsed laser 700 1000 nm with the same set of optical components Flexibility of the design will allow us to easily change or incorporate other optical components suitable for different experimental needs This microscope will allow us to do electrophysiology and imaging concurrently while maintaining the optimum temperature and CO2 levels 117 Keywords Multiphoton microscope long term imaging learning in vitro cortical neuronal networks Maintaining Viable Thick Cortical Slices by Perfusion of Nutrient Medium Komal Rambani Jelena Vukasinovic Ari Glezer Steve M Potter Society for Neuroscience Atlanta 2006 Brain slice cultures are valuable in vitro models for various electrophysiological morphological pharmacological and ischemia studies These include roller tube method Gahwiler J Neurosci Meth 1981 and static cultures in which brain slices are grown on permeable membranes Stoppini J Neurosci Meth 1991 With these methods the nutrient medium reaches cells across the thickness of the slice by diffusion and the slice thins down from 500um thickness to around 150um within two days Culturing thick brain slices gt 400m has been a challenge due to necrosis in the middle of the slice due to ischemia We hypothesize that pe
25. method to other techniques 16 Multiphoton microscopy is a relatively new non linear fluorescent imaging method that is gaining popularity with every passing day in various research investigations due to its several advantages over other fluorescence imaging methods in some key areas of biological investigations In this chapter there is a brief introduction to basics concepts of fluorescence microscopy various fluorescence microscopes and differences in the various fluorescent imaging methods and advantages of multiphoton microscopy over the confocal microscopy This chapter is designed to introduce some fundamentals for understanding of fluorescence microscopy that will aid in understanding terminology used in the next two chapters of this thesis THE ELECTROMAGNETIC SPECTRUM Human eyes are sensitive to only a subset of the entire electromagnetic spectrum that is known as the visible spectrum The wavelengths of the visible spectrum range from 400 700nm figure 3 1 Wavelengthinum 10 105 104 10 10 10 1 10 10 10 104 105 10 10 10 Visible Spectrum Near Infrared Infrared Microwaves Radiowaves Gamma Rays X rays Ultra Violet Figure 3 1 The entire electromagnetic spectrum and the visible spectrum There are two fundamental laws of the electromagnetic radiation Planck s law and Rayleigh s scattering law that are related to the topics explained in next two chapters Planck s law s
26. moves the objective up and down to facilitate the placing and focusing of the specimen with lum precision To capture finer 3D details of the specimen it is important to image the specimen at smaller z increments for which a piezo controlled fine z translator mount for the objective lens is required A 42 practical problem with the commercially available z focuser is that it covers most of the objective length with its motor and would collide with our MEA recording and stimulation hardware For such spatially constrained conditions it would be advantageous to be able to use the entire length of objective lens To meet this requirement we decided to mount a fine z positioner PI Inc upside down using a custom quick release mounting plate For repeated imaging of the different regions of interest ROJ in the specimen in a long term experiment it is required to move the specimen repeatedly to those regions at desired times We use a computer controllable XY stage which can repeatedly go to the same ROI 1 um when used in closed loop mode Phytron Inc In addition to these features in our system while a small optical zoom in the region of interest can be obtained by decreasing the scan angle of the scanning mirrors a comparatively bigger optical zoom can be obtained by quickly changing the objective lens via the custom fabricated quick release objective lens mounting plate Software control Scanning Data Acquisition Control XY st
27. the wavelengths The uniform distribution of spectral components of a pulse is distorted when it passes through a piece of glass refractive index n glass 1 3 1 5 compared to air 1 This distortion of pulse spectra is called group velocity dispersion GVD or chirping and results in reduced the peak power figure 4 1 Generally 100 200 femtosecond long pulses in the infrared wavelength regime experience less chirping than shorter pulses and are best for good multiphoton excitation 118 The Ti saph laser is a good choice as an excitation source for multiphoton microscopy 113 as it provides a tunable range of wavelengths across the near IR regime 32 Air n 1 n gt 1 Glass Figure 4 1 Group velocity dispersion of a laser pulse Laser pulses have spectral width Speed of light depends on its wavelength and refractive index of medium Thus different wavelengths travel at different velocities in glass resulting in group velocity dispersion of pulse Picrture Courtesy Prof Rick Trebino In ideal conditions a pulse should have no chirping at the focal plane of the objective lens However in the real world chirping is unavoidable because of distortion of the laser pulses by lenses and multilayer dielectric coated mirrors used in the microscope Imaging with chirped pulses will result in more laser power required for the multiphoton excitation of the fluorophore with more one photon heating To counter this problem there are t
28. to these cultures 109 19 20 21 22 23 24 25 Use 35mm diameter teflon membranes to prevent evaporation and infection while allowing gaseous exchange HARVESTING AND PLATING THE BRAIN SLICES Autoclave all the required dissection instruments Warning Never use any unsterile tools or media to handle the tissue The following procedure needs to be completed in only a few minutes to ensure high viability of the tissue just after the slicing Following the NIH rules euthanize one mouse pup P11 P15 at a time and remove its brain from its skull quickly 1 min or less Transfer the brain in the chilled nutrient medium and wait for 30 seconds to 1 minute This step helps to reduce the metabolic activity of the tissue due to reduced temperature and make the tissue slightly stiffer brain is very delicate gel like tissue otherwise which eases the tissue slicing During the wait time make all the tools ready for separating the tissue slices under a dissection microscope Cut the brain into two hemispheres using a micro knife Keep one hemisphere in the chilled nutrient medium and transfer the other half quickly to the tissue chopper disc Make sure that the tissue chopper is preset for the desire slice thickness and a sterile blade is secured in place Using appropriate coordinates cut tangential saggital or coronal cortical slices of desired thickness Gently transfer the tissue in the chilled nutrient medium
29. unction of Best pinhole size IMAGING DEPTH Poor Good Best better than confocal SIMULTANEOUS MULTIPLE LABEL EXCITATION AT SAME May be May be Most probably EXCITAITON WAVELENGTH PIN HOLE SETTINGS Not present Required Not present TIME LAPSE IMAGING Detrimental Detrimental Longer term 24 Multiphoton microscopy has several advantages over confocal microscopy as a result of localized excitation and use of IR laser as the excitation source Both advantages and disadvantages of the multipho are listed briefly in Table 3 2 99 ton microscopy over confocal microscopy TABLE 3 2 ADVANTAGES AND DISADVANTAGES OF MULTIPHOTON MICROSCOPY OVER CONFOCAL MICROSCOPY ADVANTAGES DISADVANTAGES Longer excitation wavelengths are scattered much less Rayleigh s scattering law so with infrared IR laser one can achieve deeper imaging of specimen Due to localized excitation the photobleaching and photodamage is dramatically reduced Since no pinhole is required in multiphoton imaging more fluorescent signal can be collected from the same focal plane with a given objective lens compared to the confocal imaging This results in brighter image Since there is no excitation at out of focus planes in specimen there is much less background signal This improves signal to noise ratio Excitation and emission spectra are separated better so more signal could be collected using wider band pass filters or no filters at all Hence
30. viable 700um thick brain slices for long term Image shows viability assessment after 2DIV and 5DIV Mean SEM n 3 4 A two way ANOVA test followed by Tukey s multiple comparison test was used to evaluate statistical significance indicates p lt 0 01 significant change in viability at 2DIV compared to 5DIV for the same culture condition i e perfused at a given flow rate unperfused membrane indicates p lt 0 01 for viability compared to viability of unperfused and membrane controls for same day of perfusion indicate p lt 0 05 for viability compared to unperfused and membrane controls for same day of perfusion Organotypic organization of thick cortical slice cultures To evaluate the morphological organization of the cultures I fixed them after 5 day perfusion experiments in 2 paraformaldehyde in 0 5X phosphate buffered saline PBS This concentration of fixing solution helped to maintain the osmolarity and pH of the tissue that are important parameters in determining cell health and size The tissue was sliced perpendicular to its diameter into 20 50um thick slices to reveal its thickness 91 I also prepared specimens from freshly cut and fixed tissue as a baseline to compare with cultured slices The tissue was stained with H amp E stain using standard protocols and was mounted on gelatin slides to observe the specimen using bright field microscope Several cell types of characteristic morphology like pyramidal cel
31. 00um after 5 DIV This could be explained by considering washing away of dead cells from the culture volume that died initially during the cutting process A more comprehensive study will be required to better understand this observation I also could record spontaneously or chemically evoked activity from cultured slices This indicates their usefulness in electrophysiology and pharmacological studies However these recordings were done using a single microwire electrode A more comprehensive examination could be done using multielectrode arrays 26 28 29 96 Viable slices can be easily obtained from young animals however recently there is emerging evidence that it is possible to obtain healthy organotypic slices from adult animals 62 64 116 Hence our culturing technique could be used to culture thicker organotypic slices from other parts of the brain that are traditionally cultured using the membrane insert method for example cerebellum 40 striatum 9 spinal cord 50 94 115 olfactory epithelium 43 thalamus 81 and cortex 26 63 other rodent species rat and from a range of ages post natal day 11 16 These thick organotypic cultures may benefit a wide spectrum of neuroscience investigations including learning and memory 7 13 16 19 23 25 31 64 84 39 development 43 63 81 88 94 105 traumatic brain injury 82 regeneration 78 94 116 effect of pharmacological agents on network properties and dr
32. 3 Factors affecting laser scanning fluorescence imaging 26 Table 5 1 Field of view 53 xi LIST OF FIGURES Page Figure 2 1 A Hybrot model to study learning and memory in vitro 7 Figure 2 2 Multielectrode array dish for two dimensional neuronal cultures 8 Figure 2 3 Multielectrode electrophysiology recording and stimulation set up 9 Figure 2 4 Schematic diagram of microfluidic multielectrode neural interface system 10 Figure 2 5 Three dimensional microfluidic multielectrode arrays 11 Figure 2 6 Pyramidal three dimensional microfluidic multielectrode arrays 12 Figure 2 7 Schematic diagram of experimental set up for simultaneous imaging and electrophysiology on three dimensional organotypic brain slice cultures 13 Figure 2 8 An experimental fluidic set up to culture organotypic thick brain slices 13 Figure 3 1 The entire electromagnetic spectrum and the visible spectrum 17 Figure 3 2 Jablonski energy diagram to show fluorescence fundamentals 20 Figure 3 3 An example of excitation and emission spectra of a fluorophore 20 Figure 3 4 One photon and two photon excitation 22 Figure 3 5 Pulsed laser is required for two photon excitation 23 Figure 3 6 Laser pulse and group velocity dispersion of laser pulse after passing through glass 24 Figure 4 1 Group velocity dispersion of a laser pulse 33 Figure 4 2 Pulse compressor 34 Figure 4 3 Optical diagram of the multiphoton microscope design 35 Figure 4 4 A typical power curve of
33. 64 Figure 6 1 Comparison of different culturing methods 72 Figure 6 2 Microfluidic culture set up 75 Figure 6 3 Flow trajectories and velocity distribution of microjets at various elevations in the infusion chamber 71 Figure 6 4 Perfusion of nutrient medium through the tissue thickness results in enhanced tissue viability 79 Figure 6 5 Assessment of range of optimal flow rates for enhanced brain slice culture viability 80 Figure 6 6 Viability of cultured brain slices after 5 days in vitro 81 xiii Figure 6 7 Representative micrographs of tissue nuclei labeled with nuclear stains 83 Figure 7 1 Viability of brain slice cultures over time 91 Figure 7 2 Morphology assessment 93 Figure 7 3 Maintenance of thickness of culture after 5 DIV 94 Figure 7 4 Activity traces from cultured slices after 5 DIV 95 Figure A 1 User interface of software to operate custom built multiphoton microscope 105 Figure E 1 An example of data analysis using ImageJ software 117 xiv SUMMARY Development of a three dimensional 3D HYBROT model with targeted in vivo like intact cellular circuitry in thick brain slices for multi site stimulation and recording will provide a useful in vitro model to study neuronal dynamics at network level In order to make this in vitro model feasible we need to develop several associated technologies These technologies include development of a thick organotypic brain slice culturing method a three dimensional 3D m
34. 7 1 3 2 6 853 33 853 33 412 90 412 90 211 56 211 56 134 73 134 73 1 4 2 8 914 29 914 29 449 12 449 12 228 57 228 57 143 81 143 81 1 5 3 0 984 62 984 62 474 07 474 07 243 81 243 81 153 28 153 28 1 6 3 2 1024 00 1024 00 512 00 512 00 258 59 258 59 164 10 164 10 1 7 3 4 1113 04 1113 04 533 33 533 33 278 26 278 26 175 34 175 34 1 8 3 6 1163 64 1163 64 568 89 568 89 290 91 290 91 186 86 186 86 1 9 3 8 1219 05 1219 05 609 52 609 52 304 76 304 76 195 41 195 41 2 0 4 0 1347 37 1347 37 640 00 640 00 208 13 208 13 2 1 4 2 1422 22 1422 22 673 68 673 68 345 95 345 95 216 95 216 95 22 4 4 1505 88 1505 88 731 43 731 43 360 56 360 56 226 55 226 55 2 3 4 6 1600 00 1600 00 752 94 752 94 376 47 376 47 239 25 239 25 2 4 4 8 1706 67 1706 67 775 76 775 76 387 88 387 88 248 54 248 54 2 5 5 0 825 81 825 81 406 35 406 35 258 57 258 57 53 Figure 5 2 Montage of 512x512 pixel images of hemocytometer grid taken at different scan angles with the Achroplan 20X NA 0 5 water immersion objective lens The scan angle can be changed from the user panel in single digit decimal increments The input from the graphical user interface is voltage that is delivered to driving boards of the scanning mirrors The default value is set at 1 V which translates to 2 degrees of scan angle from the central position zero position The voltage increments are set to one decimal place 0 1 V of increase in voltage input from the front panel translat
35. 82 1991 115 TAKUMA H SAKURAI M and KANAZAWA I In vitro formation of corticospinal synapses in an organotypic slice co culture Neuroscience 109 359 370 2002 116 Tom V J DOLLER C M MALOUF A T and SILVER J Astrocyte associated fibronectin is critical for axonal regeneration in adult white matter J Neurosci 24 9282 9290 2004 117 TRULLIER O and MEYER J A Animat navigation using a cognitive graph Biol Cybern 83 3 271 85 2000 118 TSAI P S NISHIMURA N YODER E J DOLNICK E M WHITE G A and KLEINFELD D Principles design and construction of a two photon laser scanning microscope for in vitro and in vivo brain imaging In vivo optical imaging of brain CRC Press 2002 119 VUKASINOVIC J and GLEZER A A microperfusion chamber for neuronal cultures Proceedings of Bio2006 1 2 2006 120 WAGENAAR D A PINE J and POTTER S M An extremely rich repertoire of bursting patterns during the development of cortical cultures BMC Neuroscience 7 11 2006 121 WAGENAAR D A MADHAVAN R PINE J and POTTER S M Controlling bursting in cortical cultures with closed loop multi electrode stimulation J Neuroscience 25 680 688 2005 122 WAGENAAR D A and POTTER S M A versatile all channel stimulator for electrode arrays with real time control Journal of Neural Engineering 1 39 45 2004 123 WAGENAAR
36. D A PINE J and POTTER S M Effective parameters for stimulation of dissociated cultures using multi electrode arrays Journal of Neuroscience Methods 138 27 37 2004 129 124 WAGENAAR D A and POTTER S M Real time multi channel stimulus artifact suppression by local curve fitting J Neurosci Methods 120 113 120 2002 125 WAGENAAR D A PINE J and POTTER S M Searching for plasticity in dissociated cortical cultures on multi electrode arrays Journal of Negative Results in BioMedicine 5 16 2006 126 WoKosIN D L AMOs B G and WHITE J G Detection sensitivity enhancements for fluorescence imaging with multi photon excitation microscopy JEEE EMBS 20 1707 1714 1998 127 YAMAMOTO M Electrical activity in thin sections from the mammalian brain maintained in chemically defined media in vitro Journal of Neurochemistry 13 1333 1343 1966 130
37. ERIZATION OF 1MM THICK BRAIN SLICES I tested only 700um thick and 3mm in diameter brain slice cultures for these studies Optimization of Imm thick preparations may facilitate viable cultures of the entire thickness of adult mouse brains To achieve this goal it will be required to change the dimensions of the culture chamber Additionally longer term characterization of viability thickness organotypic organization and functional activity may be required to consolidate this study 100 MODIFICATIONS OF CULTURING CHAMBER The current version of culturing chamber has a gold grid as a porous seat to support cultures and as an infusion port of nutrients The currently used gold grid has 40 transparency to the infused medium resulting in formation of microjets figure 6 3 A microporous membrane millipore membrane inserts may be a more amenable tissue seat and perfusion substrate due to uniformly distributed smaller pores resulting in diffused and uniform flow pressure on tissue surface Further these membranes are shown to be of high strength biocompatible and support tissue adhesion as neurons naturally attach to them 114 101 APPENDIX A A STEP BY STEP USER MANUAL TO OPERATE CUSTOM MADE MULTIPHOTON MICROSCOPE Note The procedure to switch on the microscope should be followed sequentially as mentioned below It is required to follow this procedure step by step as given below to make sure that computer recognizes the National i
38. ERIZATION OF THICK BRAIN SLICE CULTURES FOR LONGER TERM In the current studies I examined viability of thick brain slice cultures only for five days in vitro Optimization of viable cultures for longer term may be desired in some experimental paradigms Currently viability of cultures is tested at two time points 99 2DIV and 5DIV My experiments indicated various levels of decrease in viability from 2DIV to 5DIV at different flow rates To optimize the perfusion paradigm for long term viable cultures it is desired to characterize the viability of cultures at more time points QUANTITATIVE ANALYSIS OF ELECTROPHYSIOLOGY I used only one microwire electrode to probe functional activity of the culture A multisite recording at different heights of the culture is desired to authenticate reliable recordability from the cultures Ideally patch clamp recordings would be best to validate their spontaneous electrically or chemically evoked electrical properties compared to established preparations such as acute slices or dissociated tissue culture To use these cultures for targeted recording and excitability for 3D hybrot projects it is required to be able to reliably record and electrically stimulate at the targeted layers of cortex It may be required to use tetanic stimulation of layer V VI and determine synaptic responses in layer I II to authenticate successful use of these cultures for learning and memory model development 4 CHARACT
39. MPL I FERSTER D and YUSTE R Synfire Chains and Cortical Songs Temporal Modules of Cortical Activity Science 304 5670 559 564 2004 54 IYER V LOSAVIO B E and SAGGAU P Compensation of spatial and temporal dispersion for acousto optic multiphoton laser scanning microscopy J Biomed Opt 8 3 460 71 2003 55 JOHANSSON B B and BELICHENKO P V Neuronal Plasticity and Dendritic Spines Effect of Environmental Enrichment on Intact and Postischemic Rat Brain Journal of Cerebral Blood Flow and Metabolism 22 89 96 2002 56 KAKEGAWA W TSUZUKI K YOSHIDA Y KAMEYAMA K and OZAWA S Input and subunit specific AMPA receptor trafficking underlying long term potentiation at hippocampal CA3 synapses Eur J Neurosci 20 101 110 2004 57 KARPOVA AY TERVO DGR GRAY NW and SVOBODA K Rapid and Reversible Chemical Inactivation of Synaptic Transmission in Genetically Targeted Neurons Neuron 48 727 737 2005 58 KESHISHIAN H ROSS HARRISON S The outgrowth of the nerve fiber as a mode of protoplasmic movement J Exp Zool A Comp Exp Biol 301 201 203 2004 59 Kim J A ET AL Cytoskeleton disruption causes apoptotic degeneration of dentate granule cells in hippocampal slice cultures Neuropharmacology 42 1109 1118 2002 60 KLAPSTEIN G J and COLMERS W F Neuropeptide Y Suppresses Epileptiform Activity in Rat Hippocampus Jn Vit
40. N 02 702 POSITION CONTROL Focus Step 10 micron s 50000 NE STOP 30000 s04 SCAN MIRROR PARAMETERS 20000 Ads 302 Number of Pixels Number of lines aor 202 Forward Tup Tup o J ez 500 250 0 0 250 500 7801000 E Value 0 5 Number of Pixels Backward Tdown number of Occurances 2 0 000 Number of Pixels Flyback Tdown feo eo Corner Pixels Cut off Tau Amplitude wave Range T ji Pixel clock Hz J2n0e 3 Scan delay s or cycles fea Upper Limit TIME LAPSE CONTROL PANEL ANAKA t k Time Lapse Interval 3 secs f Total Time Points t3 oy ido iso S00 285 Z SERIES CONTROL PANEL Start Position do micron s End Position ifo micran s Corner Pixels Cut off X Tan Amplitude S wave Range oh Scan Cut off A24 DAQ Analog Filter 2150 kHz Step Direction Brightness Contrast Gamma BCG Values Brightness i 128 Enter Step Sze 10 50 micron s DOWN T a ne 0 25 SO 75 100 125 150 175 200 225 255 Contrast 8 eae Cee eel me ee 0 10 20 3 40 ACQUISITION PROGRESSION Z Slice Frame Number 10 of J 0 total slices Time Point fo of 0 completed Figure 4 8 Screen shot of Software User interface to operate the microscope in various imaging modes The current version of software allows continuous display of images without sa
41. NURIYA M JIANG J NEMET B EISENTHAL K B and YUSTE R Imaging membrane potential in dendritic spines PNAS 103 786 790 2006 93 O BRIEN J A HOLT M WHITESIDE G LUMMIS S C and HASTINGS M H Modifications to the hand held gene gun improvements for in vitro biolistic transfection of organotypic neuronal tissue J Neurosci Methods 112 57 64 2001 94 OIsHI Y BARATTA J ROBERTSON R T and STEWARD O Assessment of factors regulating axon growth between the cortex and spinal cord in organotypic co cultures effects of age and neurotrophic factors J Neurotrauma 21 339 356 2004 95 PEREZ VELAZQUEZ J L FRANTSEVA M V and CARLEN P L In vitro ischemia promotes glutamate mediated free radical generation and intracellular calcium accumulation in hippocampal pyramidal neurons J Neurosci 17 9085 9094 1997 96 PORTERA CAILLIAU C WEIMER RM PAOLA VD CARONIP and SVOBODA K Diverse Modes of Axon Elaboration in the Developing Neocortex PLoS Biol 3 e272 2005 97 POTTER S M and DEMARSE T B A new approach to neural cell culture for long term studies J Neurosci Methods 110 17 24 2001 98 POTTER S M FRASER S E and PINE J Animat in a Petri Dish Cultured Neural Networks for Studying Neural Computation Proc 4th Joint Symposium on Neural Computation UCSD 167 174 1997 127 99 POTTER S M Vital Imaging two p
42. POGNA M DEBANNE D MCKINNEY R A and THOMPSON S M Organotypic slice cultures a technique has come of age Trends Neurosci 20 471 477 1997 38 GAHWILER B H Nerve cells in culture the extraordinary discovery by Ross Granville Harrison Brain Res Bull 50 343 344 1999 39 GALIMBERTI I ET AL Long term rearrangements of hippocampal mossy fiber terminal connectivity in the adult regulated by experience Neuron 50 749 763 2006 40 GIANINAZZI C ET AL Apoptosis of hippocampal neurons in organotypic slice culture models direct effect of bacteria revisited J Neuropathol Exp Neurol 63 610 617 2004 41 GHOUMARI A M ET AL Mifepristone RU486 protects Purkinje cells from cell death in organotypic slice cultures of postnatal rat and mouse cerebellum Proc Natl Acad Sci USA 100 7953 7958 2003 42 GLOVER C P BIENEMANN A S HEYWOOD D J COSGRAVE A S and UNEY J B Adenoviral mediated high level cell specific transgene expression a SYN1 WPRE cassette mediates increased transgene expression with no loss of neuron specificity Mol Ther 5 509 516 2002 43 GONG Q LIU W L SRODON M FOSTER T D and SHIPLEY M T Olfactory epithelial organotypic slice cultures a useful tool for investigating olfactory neural development Int J Dev Neurosci 14 841 852 1996 44 GOpPERT MAYER M Uber Elementarakte mit zwei Quanten
43. SET UP FLUIDIC SYSTEM AND 9 THICK BRAIN SLICE CULTURING METHOD SETTING UP THE FLUIDIC SYSTEM Test all the gold grid devices to confirm proper adhesion of the gold grid to its orifice in the infusion chamber Choose appropriate FEP membrane lids to fit them snugly on the withdrawal chamber Change the FEP membrane of each lid every time a new experiment is started Clean the infusion and the withdrawal capillary tubing with ethanol followed by deionized water at least 4 5 times Make sure there is no leak or blockage in the microcapillary tubing Test for the possible leaks at each connector by flushing deionized water using a syringe Autoclave all the parts including the experiment platform stand on which chambers are secured tight during experiment Fix the capillary infusion and the withdrawal tubing to the infusion and the withdrawal ports of the devices inside a sterile laminar flow hood Fill the HBS solution in the withdrawal side syringes and the nutrient medium in the infusion side syringes Attach these syringes to the infusion and the withdrawal lines Make sure that there are no air bubbles in these lines at this time 10 Cover the chambers with the teflon lids to avoid any accidental unsterility 108 11 12 13 14 15 16 T7 18 Set the syringe pump at the appropriate flow settings Carefully fix the syringes on the infusion and the withdrawal sides Run a test flow at a flow r
44. THICK BRAIN SLICE CULTURES AND A CUSTOM FABRICATED MULTIPHOTON IMAGING SYSTEM PROGRESS TOWARDS DEVELOPMENT OF A 3D HYBROT MODEL A Thesis Presented to The Academic Faculty by Komal Rambani In Partial Fulfillment of the Requirements for the Degree Master of Science in the School of Biomedical Engineering Georgia Institute of Technology MAY 2007 COPYRIGHT 2007 BY KOMAL RAMBANI THICK BRAIN SLICE CULTURES AND A CUSTOM FABRICATED MULTIPHOTON IMAGING SYSTEM PROGRESS TOWARDS DEVELOPMENT OF A 3D HYBROT MODEL Approved by Dr Steve M Potter Advisor School of Biomedical Engineering Georgia Institute of Technology Dr Ravi V Bellamkonda School of Biomedical Engineering Georgia Institute of Technology Dr T Richard Nichols School of Physiology Emory University Date Approved 3 January 2007 To my Grand Parents and my Family ACKNOWLEDGEMENTS There are several people to whom I would like to acknowledge for their support and help for my M S thesis research First of all I would like to thank my advisor Dr Steve Potter for giving me opportunity for my graduate research work in such an interesting research field I wish to acknowledge his tremendous support for my education and excellent guidance for my research work Under his supervision I learned fine instrumentation and experimental skills that will prove to be great assets to my future career I would like to acknowledge that as a result of the fine i
45. a continuous wave CW laser the probability of two photons striking together simultaneously on an electrode is very little even at the focal spot due to low photon density A B CW laser Pulsed laser y y Objective lens sie sl cee pele eet ee oc SER SHAG DUD SHES SRS NRIS SEAS Creer j Rx Ka Rx Ka Rx RA is oa een ieee of at a ie ee Re D AEs SEAS 2 ee ee RAs RA ie REA Rear Rebar Roe Fuel Specimen Figure 3 5 Pulsed laser is required for two photon excitation A Cartoons of photons density in CW and Pulsed lasers B Two photon excitation occurs with pulsed laser green spot and not CW laser due to low photon density of CW laser A femtosecond pulse laser carrying high photon density pulses ensures simultaneous impinging of two photons on an electron of fluorophores in the focal volume resulting in two photon absorption figure 3 5 After the invention of pulsed lasers the first two photon laser scanning microscope TPLSM was made and patented by a team of researchers at Cornell university W Denk J H Strickler and W W Webb lead by Prof Watt W Webb in 1990 22 GROUP VELOCITY DISPERSION CHIRPING Femtosecond pulses have spectral 9 nm and temporal width 100 200fs for multiphoton imaging with all the wavelengths distributed around the central wavelength frequency The phase velocity of different wavelengths depends on the refractive index of medium in which they travel In air refractive
46. ablonski diagram figure 3 2 These schematic diagrams were first proposed 18 by Professor Alexander Jablonski in 1935 to describe absorption and emission spectra of light Under normal conditions the electronic configuration of the molecule is described to be in the ground state in these diagrams The ground state and the excited states are separated by the characteristic band gap energy of the molecule Due to vibrational spectra these states are further split into different energy levels When a photon of energy equal to band gap energy of the molecule hits electrons in the ground state there is high probability that these electrons can absorb the energy of the impinging photons and are raised to a higher electronic energy states excited states a process that may only take a femto second i e 10 seconds The electrons in the higher energy states can transition to other higher vibrational singlet energy states or triplet states by non radiative energy loss depending on their quantum states Eventually they release the excessive energy in the form of photons of longer wavelength and return to the ground state of the molecule atom The emitted photons in the visible regime are called fluorescent signal Due to transition of electrons to different vibrational energy states from ground to excited levels each fluorophore has excitation and emission spectra figure 3 3 19 Jablonski Energy Diagram Excitation Excited Singlet States
47. age and Focal Control Custom software written in LabView 7 1 controls the synchronized movement of scanning mirrors and data collection from the PMT Since digital to analog cards have only one FIFO memory available to exploit the fastest scanning speed of the scanning mirrors I chose two digital to analog cards Model PXI 6733 National Instruments Inc to control the motion of the scanning mirrors The scanning mirrors are driven using custom waveform developed by Tsai et al to optimize the speed of the raster scan A real time system integration RTSJ bus is used to synchronize commands using counters and clocks pixel clock line clock and frame clock of the digital to analog cards to each 43 scanning mirror collect data information as time series through data acquisition card and update position of z controller figure 4 7 The data is collected only in the linear part of the driving waveform to avoid the non uniform speed of the mirrors at the corners of the waveform At the edges of the scanned region non uniform velocity will cause non uniform illumination and a distorted shape of the specimen The collected fluorescent signal is converted to voltage in the preamplifier circuit which is passed to a data acquisition card Model PXI 6115 National Instruments Inc from where it is read displayed and stored as an image using custom written LabView software The current version of the Lab View user interface includes the following featu
48. aging using UV excitable fluorophores without using expensive quartz optics and harmful UV light 99 In our laboratory we combine multiple technologies to investigate the basic mechanisms underlying learning and memory These technologies include multielectrode arrays MEAs embodiment of 2D and 3D neuronal cultures with animats or robots and multiphoton microscopy 100 117 For long term morphological dynamics investigations and thick brain slice viability studies we need an efficient microscope with special features Some of these features include deep tissue imaging flexibility of the design to accommodate our electrophysiology and fluidic systems reduction in unnecessary exposure of specimen to the lasers flexibility to change the frame size field of view and ability to zoom in on a region of interest on the fly both optically and digitally and most importantly the ability to efficiently detect very small fluorescent 29 signals from micron sized features such as dendritic spines in spite of noise Currently none of the commercial multiphoton microscope systems have all these features as they often include design compromises that retain the microscope s ability to also perform visible confocal microscopy This fact motivated us to fabricate a custom designed microscope optimized solely for long term multiphoton imaging In this chapter I describe design criteria chosen for our microscope based on the desired features selection
49. al perfusion rates for enhanced viability To assess the flow rates for optimal supply of nutrients to the tissue without shear stress induced damage to the tissue we performed a series of experiments to perfuse the tissue at different flow rates that include 5 10 20 30ul hr The viability was assessed after 2DIV The results of my experiments indicate that flow rates 20ul hr or less result in enhanced viability of the tissue while higher flow rates turned out to be 79 detrimental to the delicate tissue structures resulting in formation of channels inside the tissue and reduced overall viability Thus I found that a flow rate translating to 3 culture volume exchanges per hour is the optimal flow rate to maximize the viability of the tissue as a result of constant nutrient supply figure 6 5 100 90 80 70 60 50 Viability 40 30 20 10 Unperfused Membrane Sulfhr 1Ou hr 20u hr 30ed hr Figure 6 5 Assessment of range of optimal flow rates for enhanced slice culture viability Flow rates lt 20 ul hr are non invasive flow rates for enhanced tissue viability A flow rate 20 ul hr of 3 volume exchanges per hour is optimal for enhanced viability Higher flow rates 2 20ul hr are detrimental to tissue Figure shows viability at each flow rate Mean SEM n 3 4 A generalized linear model ANOVA followed by a Tukey s multiple comparison test was used to evaluate statistical significance indicates p lt
50. ame set of optics to achieve a good spatial resolution and a high imaging frame rate and to obtain uniform intensity of the image across a frame scan Optical design An upright microscope design along with careful selection of optical components would meet most of these desired features figure 4 3 To obtain optically good imaging data while minimizing the photodamage to specimen it is required to have efficient excitation and emission pathways Since biological specimens are highly scattering media and absorb less in infrared regime of the electromagnetic spectrum than the visible the longer wavelength regime of the tuning range of the Ti saph laser is more suitable for 34 imaging these specimens especially when repeated imaging of a specimen is desired Further it is required to have minimally chirped pulses of laser light at the focal plane and high transmission of the laser light By using as few optical elements as possible we minimize light losses due to reflections and chirping The excitation path includes a pump laser Coherent Verdi 10W 532nm pulsed IR laser Coherent Mira 900 Ti saph laser beam routing mirrors laser beam reshaping unit pre chirp unit Pockels cell scanning mirrors scan lens tube lens dichroic mirror and objective lens The excitation and emission paths share a path containing objective lens and dichroic mirror Along with these two the emission path also includes the collection optics and a detec
51. angles higher speed of scanning can be obtained At higher angles bigger field of view can be obtained Only a scan angle up to 4 is desired for the various imaging needs The default value is set to 2 degrees of deflection Field of view and Scan angle The angle of scanning of the laser beam decides the field of view of specimen for a given objective lens This property could be exploited to optically zoom in or out of the specimen at the expense of scan speed figures 5 1 5 2 without changing the objective lens This property can be controlled from the graphical user interface GUI To calibrate the field of view of different objective lenses at various scan angles a hemocytometer 52 grid was used to reflect a weak laser beam lt 2mW to the detector A montage image of images taken at different scan angles with a 20X W N A 0 5 objective lens is shown in figure 5 2 With increasing scan angles the edges of the raster scan of the laser beam are blocked by the back aperture of the objective lens leading to the vignetting of the image A complete table of field of view with different objective lenses at different scan angles is given in table 5 1 TABLE 5 1 FIELD OF VIEW IN MICROMETERS USER SCAN PANEL HALF 10X 20X 40X 63X INPUT ANGLE 1 0 2 0 640 00 640 00 312 20 312 20 162 03 162 03 103 23 103 23 1 1 2 2 775 56 775 56 350 68 350 68 179 02 179 02 113 78 113 78 1 2 2 4 387 88 387 88 195 42 195 42 123 67 123 6
52. ate 1ml hr to test for any potential leaks Fix any leaks in the lines at this point Sometimes it may require repeating the entire process depending on the leak spots and the leak severity Any leaks will ruin the experiment by invalidating the set flow rate and or by the infection Once assured that there are no leaks and the system works in infusion withdrawal flow conditions stop the flow and check for any bubbles in the infusion line once more Remove the teflon lid and aspirate all the medium in the infusion and the withdrawal chamber very carefully with a micro pipette without damaging the gold grid and its adhesion to the device orifice Pour 201 of laminin in the infusion chamber Due to surface tension it may look like a drop sitting in the infusion chamber The excess of laminin will exit to the withdrawal chamber upon closing the chamber with the teflon lid This will ensure coating of gold grid and the interior walls of the infusion chamber with laminin properly Leave the enclosed chambers in the sterile hood and prepare for harvesting of the tissue slices To obtain good anchoring of tissue with the chamber it is required to coat the chamber with laminin atleast 30 40 minutes before transferring the tissue in it Prepare the gold grid chamber and the membrane insert culture dishes 35 mm cultures dishes for the control experiments Coat them with the laminin and fill the culture bath to ensure diffusion based supply of nutrients
53. ave compromised design to use same set of optics for multiphoton and one photon microscopy to retain their ability to perform confocal imaging To meet our requirements for deep tissue imaging and or long term time lapse imaging of neuronal networks in dissociated cultures or brain slices I have custom fabricated a multiphoton only microscope to meet the desired efficiency and flexibility required for our studies In this chapter I report the performance of the microscope for some of these applications such as 2D 3D and time lapse imaging SYSTEM VALIDATION OPTICAL PROPERTIES Achievable scanning speed The scanning mirrors can deflect up to range of 20 degrees but for our purpose only 4 degrees of deflection is adequate The mirrors are driven by a custom made waveform developed by Tsai et al to optimize speed and linearity of scanning during data 51 collection The smaller 3mm x 3mm mirror is driven at a fast rate which forms the basis for the line scan and the octagonal longer mirror 5mm x 3mm moves at a much slower rate to move the line scan in a raster pattern The maximum velocity that can be achieved by our scanning mirrors for different scan angles is given in the figure 5 1 1400 1300 1200 1100 1000 Frequency Hz 900 800 700 2 2 5 3 3 5 4 45 5 SS 6 Angle of deflection from center to one side degrees Figure 5 1 Speed of scanning mirrors at different angles of deflection scan angles At smaller
54. bed construction of a user friendly and flexible custom made microscope which will allow us to conduct long term imaging and multi unit recording and stimulation experiments concurrently to investigate the information processing mechanisms in networks of cultured mouse neurons While our microscope is designed to achieve optimum excitation and emission efficiency our design offers a unique blend of features which when incorporated will help us and others combine long term imaging and electrophysiology These features include programmable XY stage for 4D imaging to investigate an ongoing dynamic phenomenon in the specimen quick release objective lens accessory to quickly change the objective to zoom on a region of interest maximum 48 collection of isotropically emanating fluorescent signal using a mirror under the specimen and ability to reduce the thermal noise to detect tiny fluorescent signals With our system we can use any wavelength from the entire tunable range of Ti saph laser efficiently with a single set of optics Moreover a minimum number of optical elements retain the temporal quality of the laser pulses which translates to efficient excitation of the fluorophore The small size of the scanning mirrors helps to achieve fast scanning rates and eliminates unnecessary optical elements to attain good alignment of the laser beam path Along with scan mirrors a careful selection of scan lens and a custom made dichroic mirror makes the ex
55. ber at a flow rate of 5 pl min a Magnified view of the gold grid b The radially spreading infused fluid jets near the surface of gold grid submerged in stationary fluid c Upon interaction with a relatively stagnant medium within the chamber radial outflows intensify with elevation as jets broaden and decelerate streamwise along their axes d Concomitant with reduction in streamwise momentum yjet interactions become apparent with peripheral jets turning at lower elevations e Peripheral jets merge and begin vectoring towards the perimeter of the cultured domain influenced by a strong wall jet developing along the membrane f The exit flow is biased towards the uchannel exits in the infusion chamber wall and a strong wall jet originating from the centrally located stagnation zone forms as jets issuing closer to the center axis of the chamber penetrate deeper into the cultured domain than those located closer to pchannels This study ensures ample amount of nutrient supply in the center of the cultured brain slice resulting from convective flow of nutrient medium Data Courtesy Jelena Vukasinovic TT RESULTS Perfusion allows enhanced culture viability 700um thick cortical slices harvested from the same brain were cultured on membrane inserts using Stoppini s method and microperfusion devices The perfusion in the experimental microperfusion chambers was started at 1 0ul hr infusion withdrawal rate using a push pull syringe pump
56. cimetry uPIV studies at various elevations in the infusion chamber These studies indicate that the 75 microjets emanating from each of the gold grid openings interact with adjacent ones and this interaction increases intensifies with elevation of the chamber fig 6 3b e resulting in more uniform flow trajectories at higher elevation The flow direction of infused medium near the periphery of the chamber is biased by the micro channels in the chamber wall while the infused medium microjets closer to the axis of the chamber penetrate deeper before getting biased by the microchannel openings starting at mid height 350um from the gold grid orifice of the infusion chamber wall figure 6 3b f 6 1f 6 1g On the other hand the microjet trajectories emanating close to the infusion chamber wall get biased laterally by the microchannel openings This design ensures the lateral supply of the nutrient medium closer to the wall of the chamber This geometry of the chamber allows axial and lateral flow of nutrient medium inside the culture volume infusion chamber The flow profile of nutrient medium thus achieved ensures ample nutrient supply in the deeper ischemia vulnerable area of the tissue gt Microjet A stream of infused medium from each opening pore of the gold grid 76 Figure 6 3 Flow trajectory and velocity distribution of micro jets at various elevations in the infusion cham
57. citation path efficient such that minimum laser power is required to obtain a uniform illumination in the specimen Collection of photons emanating from the specimen is optimized by using a sensitive detector efficient collection optics and a transmission path reflection mirror under the specimen This reduces the power to illuminate the specimen and will allow the ability to image deeper in the specimen by increasing the laser power as a function of depth If needed this microscope can be extended to collect second harmonic generation SHG signal in the transmission path even while imaging in multiphoton mode 88 Life support of the cultures during long term imaging is important we have built an environmental chamber around the microscope to maintain appropriate temperature and CO levels 101 49 CHAPTER 5 VALIDATION OF THE CUSTOM FABRICATED MULTIPHOTON LASER SCANNING IMAGING SYSTEM ABSTRACT An efficient multiphoton microscope with flexible design ought to allow imaging of wide range of experimental specimens and accommodate experimental needs We have fabricated a high throughput custom designed multiphoton microscope with desired flexibility and features that is optimized for long term time lapse and three dimensional 3D imaging of living specimen After constructing the microscope we tested and calibrated its various features In this chapter we report validation of our custom made imaging station for two dimensional 2D
58. ction of laser beam route will be required to ascertain that system is not misaligned Ifthe laser beam is reaching at the focal plane then check if the dichroic mirror is in place It is required to be in a position directly above the objective lens to route the beam towards detector Make sure that the detector and the preamplifier power supply is on Check if you are using the right emission filters and the excitation wavelength for your fluorophore Take a test image using a test slide such as a pollen grain slide 106 THE SOFTWARE IS NOT STARTING Sometimes due to its previous history of stopping the system inappropriately the software may not work 1 Try restarting the system Each and everything 2 If you have switched on the computer before the chassis the chassis will not communicate with the computer This can be tested from the color of the four LEDs on the front side of the chassis green if it was recognized by computer orange if it is not recognized by computer In this case restarting the chassis followed by restarting the computer will solve the problem 3 Software still may not start due to MXI interface failure if the chassis was not powered on before the computer Switch off the chassis and wait for 5 minutes There is a large capacitor in the chassis that takes time to discharge Hurrying up to restart the chassis and the computer system will not help 107 APPENDIX C A DETAILED PROTOCOL TO
59. ctrophysiological properties of the perfused cultures compared to fresh tissue slices and control unperfused cultures To study long term morphological dynamics in two and three dimensional networks of neurons along with concurrent multisite electrophysiology it is required to have a flexible efficient non invasive and high resolution imaging system Mark Booth and I constructed a multiphoton only microscope with desired flexibility of design to accommodate electrophysiology and life supporting fluidic paraphernalia for concurrent imaging electrophysiology and perfusion These two projects have contributed towards essential progress in the development of a 3D HYBROT model This thesis is organized as follows The first two chapters provide introduction and background to the projects undertaken Chapter three provides a framework to briefly introduce the reader to fluorescence microscopy and the advantages of multiphoton microscopy over other conventional and confocal imaging modes Chapters four and five discuss in detail the design construction and validation of a flexible custom fabricated multiphoton imaging system I also point towards future directions in the further development and improvement of this system In chapter six the design working principle fabrication and characterization of a novel forced perfusion based micro culture chamber are described I used this device to validate our hypothesis that forced perfusion of oxyge
60. d FROTSCHER M Fine structure of rat septohippocampal neurons I Identification of septohippocampal projection neurons by retrograde tracing combined with electron microscopic immunocytochemistry and intracellular staining J Comp Neurol 325 207 218 1992 86 NEISEWANDER J L BAKER D A FUCHS R A TRAN NGUYEN L T L PALMER A and MARSHALL J F Fos protein expression and cocaine seeking behavior in rats after exposure to a cocaine self administration environment J Neurosci 20 798 805 2000 126 87 NEWELL D W BARTH A PAPERMASTER V and MALOUF A T Glutamate and non glutamate receptor mediated toxicity caused by oxygen and glucose deprivation in organotypic hippocampal cultures J Neurosci 15 7702 7711 1995 88 NIKONENKO I ET AL Integrins are involved in synaptogenesis cell spreading and adhesion in the postnatal brain Brain Res Dev Brain Res 140 185 194 2003 89 NIKOLENKO V NEMET B A and YUSTE R A two photon and second harmonic microscope Methods 30 1 3 15 2003 90 NORABERG J ET AL Organotypic hippocampal slice cultures for studies of brain damage neuroprotection and neurorepair Curr Drug Targets CNS Neurol Disord 4 435 452 2005 91 NORABERG J KRISTENSEN B W and ZIMMER J Markers for neuronal degeneration in organotypic slice cultures Brain Res Brain Res Protoc 3 278 290 1999 92
61. der a dissection microscope This entire procedure from decapitation to slice separation took place with in 5 6 minutes The 700um thick cortical slices were cut using a biopsy tool into 3mm round disks to fit snugly in the infusion chamber These tissue slice discs were transferred to a new sterile culture dish containing 2ml of chilled nutrient medium and were transferred to the laminin coated infusion chamber The microfluidic culture chamber was then enclosed with FEP membrane sealed teflon lids 97 The syringe pump was started and the entire set up was transferred to culture friendly controlled environment of a laboratory incubator maintained at 5 CO2 9 Oz 65 Relative Humidity and 35 C temperature 12 Adhesion methods At lower flow rates lt 20ul hr laminin coating facilitates adhesion of the tissue to the gold grid and the inner walls of the infusion chamber An FEP membrane containing teflon lid provides a gas permeable surface from the top of the tissue to hold the tissue down to the chamber while allowing gaseous exchange to equilibrate nutrient medium with the incubator environment At higher flow rates a tissue culture compatible weight a Millipore membrane attached to an approximately 6mm diameter gold ring using thin layer of PDMS was used to facilitate tissue adhesion to the infusion chamber Viability assessment The tissue viability was assessed using cell permeant and non permeant fluorescent nuclear labels Ho
62. e I believe that the viability of the unperfused cultures decreased dramatically after 5 DIV compared to perfused slices for two reasons one dying cut cells release harmful 95 chemicals that can diffuse in the tissue and trigger cell death in the adjacent tissue areas 62 66 and second the cells in the deeper layers of the tissue do not get sufficient nutrient and oxygen supply resulting in starvation of tissue and leading to apoptosis figure 7 2c d On the other hand in the perfused tissue using our perfusion paradigm there is sufficient supply of nutrients throughout the thickness of the tissue Further due to one way flow from bottom to top of infusion chamber of the nutrient medium and domination of convection over diffusion process the harmful chemicals are washed away resulting in less harm to the adjacent cell layers in the tissue Together these are expected to result in much greater viability of the perfused cultures compared to the unperfused cultures Further I noticed increased thickness maintained in perfused cultures compared to unperfused ones This could be explained by additional mechanical support on the sides of the tissue by the biocompatible walls of the infusion chamber Confinement of tissue in the infusion chamber prevented any lateral spreading of tissue unlike membrane insert based cultures This resulted in better preservation of thickness of the tissue However even the perfused tissue thinned down by 1
63. ed in table 3 3 TABLE 3 3 FACTORS AFFECTING LASER SCANNING FLUORESCENCE IMAGING FACTOR HOW DOES IT AFFECT EXCITATION WAVELENGTH NUMERICAL APERTURE FLUOROPHORE QUALITY PREPARATION AND CONCENTRATION EFFICIENCY OF MICROSCOPE FILTERS PIXEL DWELL TIME DETECTOR SENITIVITY BACKGROUND AUTOFLUORESCNCE MODE LOCKING OF LASER ALIGNMENT OF SYSTEM Shorter wavelengths cause photodamage and thermal damage to the specimen Larger numerical aperture can collect signal from larger cone resulting in more signal collection even from scattering specimen Expired or light exposed fluorophores do not fluoresce to maximum efficiency or at all Choice of the right fluorophore that is stored and prepared as per instructions guarantees good fluorescent signal yield with an appropriate excitation wavelength Concentration of the fluorophore also determines amount of fluorescent signal and image quality Efficiency of microscope to excite fluorophores sufficiently and collect the maximum amount of emanating fluorescent signal photons translates to quality of the images and the photodamage to the specimen The more the excitation laser power required to collect a desired intensity of an image the more the photodamage is to the specimen and fluorophores which results in faster photobleaching Transmittivity and spectral properties of excitation and emission filters of microscope also decide image quality and authenticity of ima
64. ep Size Afo so micron s DOWN ojos 6 0 25 50 75 100 125 150 175 200 225 255 SE ACQUISITION PROGRESSION Scan delay DAQ lS ee 45 s or cycles Analog Filter oe 105 20 ce 190 P70 150 90 2 Slice Frame Number of Jo total slices G J50 kH a Ga Jeo 20 ke TERM Time Point Jo of o completed 0 00 1 00 200 300 400 5 00 4 14 Figure A 1 User interface of software to operate custom built multiphoton microscope 1 Initialize software 2 Select operation mode 3 Enter file directory name 4 Hit channel 1 to ON 5 Hit Start 6 Increase PMT gain PMT voltage Adjust PMT gain to obtain desire intensity of the image 105 APPENDIX B A TROUBLE SHOOTING MANUAL TO FOR THE CUSTOM MADE MULTIPHOTON MICROSCOPE PM NOT GETTING AN IMAGE There can be several reasons for not getting an image with the custom made multiphoton microscope Check the following list 1 First of all check that the pulsed laser is mode locked and it is routed to the custom made microscope by a flip mirror Make sure that the laser beam is reaching at the focal plane You can do so by using a special fluorescent detection card excited by the infra red laser light If it is not reaching at the focal plane check if the mirror in the trinocular slider on its right hand side is pulled out Again check if the laser beam is reaching at the focal plane now If it is still not reaching at the focal plane further inspe
65. es to 0 2 degrees of scan angle increment Ability to change scan angle from front panel allows small change in field of view and hence magnification to capture details of image without changing the objective lens With 20X W NA 0 5 objective lens the scan angles larger than 3 8 degrees 1 9 input from user control panel lead to vignetting of the raster scan of the laser beam at the corners of the image due to blocking of laser beam at back aperture 0 8 mm standard of objective lens An increase in the scan angle reduces the scanning speed 54 Excitation pathway efficiency The excitation pathway efficiency depends on four factors 1 diameter of laser light transported by optical components 2 reflection at each optical component due to refractive index mismatch 3 chirping of the laser pulse and 4 absorption If the cross sectional intensity profile of the laser beam is gaussian over 90 of the laser intensity power is located at its full width at half maximum FWHM length To make maximum use of laser power the laser should be mode locked in the gaussian mode and the aperture of various optical components should be as big as or greater than the diameter of the laser beam We use a two lens f 100mm and 80mm telescope to reshape the laser beam to fit on the aperture of the small mirror 3mm Secondly we minimized the number of optical components mirrors and lenses on the laser beam excitation pathway The lenses transmissi
66. escht and Propidium Iodide as explained in the chapter 6 88 Tissue Fixing and Haematoxylin and eosin H amp E staining Due to limited imaging depth with fluorescence microscopy the cultures were fixed and sliced along their thickness into thin slices These slices were stained with H amp E stain using the standard protocol and were mounted on gelatin coated slides to obtain the thickness parameters The tissue was fixed in 2 paraformaldehyde in 0 5X PBS with its pH and osmolarity adjusted similar to that of nutrient medium that is supplied to the cultures Adjustment of these parameters was crucial to prevent any major changes in tissue thickness This was ensured by fixing fresh tissue of various thicknesses The thickness was assessed using bright field images taken with camera operated upright microscope in a calibrated field of view Functional Activity recording Micro wire electrode set up The Axoclamp electrophysiology station was modified to adopt 50um thin steel wire electrode and a ground electrode The wire electrode was firmly held using a holder attached to a xyz stage that allowed precise movement of the electrode to facilitate probing of the tissue at various places and depths The spontaneous signal was measured relative to a ground electrode using commercial software at 10 kHz sampling rate In the absence or low level of spontaneous activity the culture was treated with 120 mM KCl solution warmed at incubator tempera
67. ges especially in overlapping excitation and emission spectra or overlapping emission spectra of two fluorophores and the choice of appropriate spectral bandwidth of filters Longer pixel dwell time allows more fluorophore molecules to get excited and collection of more photons per pixel resulting in cleaner and brighter images Detector sensitivity to the emission spectra of the fluorophore determines the brightness contrast of images Uniform sensitivity across visible spectra allows equal amount of signal collection from different color fluorophores excited with the same laser power Some specimens contain molecules that fluoresce with the excitation wavelength used Depending on application it can be useful signal or count towards background noise For such applications appropriate choice of excitation wavelength and emission filters is necessary Multiphoton imaging is not possible without mode locking of the pulsed laser Misaligned system may lead to poor excitation and emission efficiency of the microscope in addition to non uniform illumination of specimen across the field of view The 3D sectioning properties of laser scanning microscopy methods make them special for various investigations Some of the commonly used applications in biomedical research include live or fixed tissue imaging deep tissue imaging time lapse imaging 26 Spatiotemporal imaging Fluorescence Recovery After Photo bleaching FRAP Fluorescence Reso
68. get a bright image it is important to collect this entire signal We purchased a custom made dichroic mirror Chroma Inc with 4cm x 5cm dimensions This dichroic mirror has a sharp cut off at 700nm wavelength such that the entire bandwidth of the laser is transmitted and the fluorescence signal lt 700nm is reflected at gt 99 The transmission and reflection properties of our dichroic mirror are shown in figure 4 5 39 T 400 600 800 1000 Wavelength nm Figure 4 5 Transmission curve of custom made dichroic mirror with a sharp cut off at 700 nm wavelength This allows to transmit the laser light to the specimen and to redirect the fluorescent signal lt 700nm towards the collection pathway data courtesy Chroma Inc In our system a pair of collection lenses along with a liquid light guide helps to transport and focus the fluorescent signal on the active area of the detector A collection lens with large diameter placed as close to dichroic mirror as possible ensures maximum collection of fluorescent signal We use an aspheric lens with a 50 8mm diameter Oriel Inc which focuses the fluorescent signal on the 8 mm diameter aperture of a liquid light guide Oriel Inc The liquid light guide transmits 90 of the collected fluorescent light to another aspheric lens with numerical aperture of 0 7 This second lens helps to focus the signal onto the active area of the photomultiplier tube PMT To detect the col
69. he input port that serves three main purposes one it traps the air bubbles in the infusion line and allows them to escape the micro capillary tubing second it acts like a damper for the medium coming in pulses as a result of stepping motor operation of the syringe pump and last it allows oxygenation of the nutrient medium by exchange of environment across the FEP membrane The perfusion set up is such that the amount of infused medium equals the amount of the withdrawn medium to maintain a constant pressure difference across the tissue in the chamber More details to set up the device for experiments are mentioned in Appendix C 74 Tefion capillary tubing Infusion withdrawal Syringe pump Brain Slice Lid with FEP Membrane Gold grid Aerator Infusion port Withdrawal port Figure 6 2 Microfluidic culture set up A The computer Aided Design CAD of microfluidic chamber to mass produce molds from thermoplastic material The chamber has two coaxial cylindrical units the inner cylinder called infusion chamber is hollow and houses a gold grid at 700um depth from top to support tissue The cylinder contains 150um wide and 350um deep channels on its wall to achieve uniform infusion of medium in the deeper layers of the cultured tissue The bottom of this cylinder is used as a infusion port using micro capillary tubing The outer cylinder called withdrawal chamber carries the out flowing medium f
70. hine in core histology facility several slides can be labeled with H amp E at once following the standard procedure 14 The labeled slides are taken out of the xylene solution a medium used during H amp E staining and are covered with coverslips to prepare for imaging with a regular brightfield microscope 114 APPENDIX E DATA ANALYSIS USING THE IMAGE J SOFTWARE The data was imported in the ImageJ software available for free from the NIH website http rsb info nih gov 1j I downloaded several important and relevant plugins Some of the relevant plugins for the viability data processing are LSM reader Stacks and AVI reader The data was imported in the ImageJ software using the LSM reader plugin The data from the two detector channels one for the hoechst label and the other for the propidium iodide label open up in the two different image windows The data is cleaned to remove the background noise and to increase contrast for the entire stack wherever needed The cleaned data is then converted to a binary data set using the adjust threshold command This command allows adjustment of the threshold based on pixel intensity The nuclei in the out of focus planes are fainter and appear smaller in size compare to those that are in the focal plane The adjust threshold command allows to select nuclei that are in the same plane based on their intensity After obtaining a binary image the command
71. hotons are better than one Curr Biol 6 12 1595 1598 1996 100 PoTTER S M Distributed processing in cultured neuronal networks Prog Brain Res 130 49 62 2001 101 POTTER S M Two photon microscopy for 4D imaging of living neurons Imaging in Neuroscience and Development A Laboratory Manual Yuste R and Konnerth Cold Spring Harbor Laboratory Press 2004 102 POTTER S M WAGENAAR D A and DEMARSE T B Closing the Loop Stimulation Feedback Systems for Embodied MEA Cultures Advances in Network Electrophysiology Using Multi Electrode Arrays Taketani M and Baudry M New York Kluwer 2006 103 POTTER S M WAGENAAR D A MADHAVAN R and DEMARSE T B Long term bidirectional neuron interfaces for robotic control and in vitro learning studies Engineering in Medicine and Biology 25th Annual International Conference of the IEEE Cancun 2003 104 Pozzo MILLER LD NK MAHANTY JA CONNOR and DMD LANDIS Spontaneous pyramidal cell death in organotypic slice cultures from rat hippocampus is prevented by glutamate receptor antagonists Neuroscience 63 471 487 1994 105 RAINETEAU O RIETSCHIN L GRADWOHL G GUILLEMOT F and GAHWILER B H Neurogenesis in hippocampal slice cultures Mol Cell Neurosci 26 241 250 2004 106 RATHENBERG J NEVIAN T and WITZEMANN V High efficiency transfection of individual neurons using modified electrop
72. hy thick cortical cultures with maintained organotypic morphological organization and functional activity Thick organotypic cortical slice cultures may offer a novel experimental model for long term neuroscience investigations This culturing method is a step closer to advance technology to create a simpler embodied in vitro neuro robotic hybrid model to study long term learning memory and drug addiction in vitro using multiple non invasive techniques MATERIALS AND METHODS Brain slice culture and perfusion set up The brain slices were harvested from P12 P16 mouse pups of strains CS7BL 6J and B6 Cg Tg Thy1 YFP 16Jrs J Jackson Laboratory and were mounted on the microfluidic infusion chamber The pups were euthanized using isofluorane in accordance with approved protocols Under sterile conditions the euthanized pup was decapitated and brain was removed and immediately stored in chilled pre oxygenated nutrient medium for approximately 1 minute The nutrient medium contains 50 OptiMEM Invitrogen 25 equine serum Hyclone 25 Hank s Balanced Salt Solution HBSS Invitrogen 5001 of 0 5mM Glutamax Gibco and 0 45 g of D glucose Invitrogen per 100ml of nutrient medium The brain was cut into two hemispheres and each hemisphere was sliced to obtain 700um thick tangential or coronal cortical slices The sliced tissue was immediately transferred again to chilled nutrient 87 medium and the slices were separated using micro spatulas un
73. hysiology techniques J Neurosci Methods 126 91 98 2003 107 SAKATA J T and BRAINARD M S Real time contributions of auditory feedback to avian vocal motor control J Neurosci 26 9619 9628 2006 108 SCHMIDT H ET AL Organotypic hippocampal cultures A model of brain tissue damage in Streptococcus pneumoniae meningitis J Neuroimmunol 113 30 39 2001 109 SHEPHERD GMG and SVOBODA K Laminar and columnar organization of ascending excitatory projections to layer 2 3 pyramidal neurons in rat barrel cortex J Neurosci 25 5670 5679 2005 110 SHEPHERD GMG STEPANYANTS A BUREAU I CHKLOVSKII DB and SVOBODA K Geometric and functional organization of cortical circuits Nat Neurosci 8 782 790 2005 128 111 SMONI A D and Yu L M Preparation of organotypic hippocampal slice cultures interface method Nature Protocols 1 1439 1445 2006 112 SOBCZYK A SCHEUSS V and SVOBODA K NMDA receptor subunit dependent Ca2 signaling in individual hippocampal dendritic spines J Neurosci 25 6037 6046 2005 113 SOELLER C and CANNELL M B Construction of a 2 photon microscope and optimization of illumination pulse duration Pflugers Arch European Journal Of Physiology 432 555 561 1996 114 STOPPINI L BUCHS P A and MULLER D A simple method for organotypic cultures of nervous tissue J Neurosci Methods 37 173 1
74. ical support using the biocompatible walls of the infusion chamber Together these results reveal that thick organotypic cortical slices could be cultured successfully that maintain organotypic cellular level morphological organization and are functionally active Thick organotypic brain slice cultures provide an in vivo like platform for the creation of 3D hybrots an in vitro embodied culture model to study learning and memory INTRODUCTION Brain slices are well established in vitro experimental models to study mechanisms underlying learning and memory and several other neuroscience investigations since they preserve in vivo like cyto architecture while providing easier access to the desired cellular networks and controlled input output variables compared to 85 in vivo preparations Today most of our understanding of synaptic plasticity and mechanisms of LTP and LTD is attributed to hippocampal and cortical slice studies Recently there is flurry of investigations performed on thick cortical hippocampal or co culture slices preparations 5 11 59 60 66 However these studies are done on acute slice preparations due to their short life time and unavailability of a reliable method to culture thick brain slices Hence these important investigations are limited to only a few hours To date very little information is available on long term network properties of brain information processing Organotypic brain slice cultures offer an oppo
75. ick organotypic brain slices Currently the commercially available multiphoton microscopes have a compromised optical design because they retain their one photon confocal imaging mode Additionally these microscopes have inflexible designs which do not allow accommodation of other technical platforms needed for an experiment requiring simultaneous electrophysiology and or fluidic delivery to the tissue culture I have custom designed and fabricated a multiphoton only imaging system with flexible design based on Tsai et al 118 This system will allow us to image deeper in the thick specimens and perform long term live cell imaging in future The circulatory system of brain tissue gets destroyed during the slicing process thus hindering supply of sufficient nutrients to cells to meet their metabolic needs Thin organotypic brain slices can be cultured using the roller tube method and the static membrane insert method 35 114 However these methods provide diffusion limited supply of nutrients and oxygen to the interior layers of cells of the tissue resulting in necrosis of cells in the center of the slice We have developed and optimized a novel microfluidic device for convection based forced perfusion of oxygenated nutrient 14 medium through the thickness of the brain slice This method allowed us to optimize 700um thick cortical slice cultures figure 2 8 We further characterized these cultures by morphological and electrophysiological ac
76. icro fluidic multielectrode Neural Interface system uNIS and the associated electronic interfaces for stimulation and recording of from tissue development of targeted stimulation patterns for closed loop interaction with a robotic body and a deep tissue non invasive imaging system To make progress towards this goal I undertook two projects 1 to develop a method to culture thick organotypic brain slices and ii construct a multiphoton imaging system that allows long term and deep tissue imaging of two dimensional and three dimensional cultures Organotypic brain slices preserve cytoarchitecture of the brain Therefore they make more a realistic reduced model for various network level investigations However current culturing methods are not successful for culturing thick brain slices due to limited supply of nutrients and oxygen to inner layers of the culture We developed a forced convection based perfusion method to culture viable 700um thick brain slices Multiphoton microscopy is ideal for imaging living 2D or 3D cultures at submicron resolution We successfully fabricated a custom designed high efficiency multiphoton microscope that has the desired flexibility to perform experiments using multiple technologies simultaneously This microscope was used successfully for 3D and time lapse imaging Together these projects have contributed towards the progress of development of a 3D HYBROT A hybrid system of brain slice cultures brain
77. icroscope for multicolor imaging During time lapse imaging images are captured only at a fixed interval of time Ideally a shutter is required at the excitation port to block unnecessary exposure of specimen to the laser Also the scanning of the scan mirrors should be stopped when the image is not being taken to avoid heat induced damage to their circuitry during long term time lapse imaging The current version of the microscope does not have these features A Pockels cell could be used to block the laser beam during the time when no image is captured To use a pockels cell in this mode it would be required to modify the current version of the software by synchronizing the time lapse interval with the output of the Pockels cell Additionally it will be required to modify the software control to stop the scanning of the mirrors during this time period The Pockels cell contains a large electro optic crystal that introduces group velocity dispersion of the pulses Thus it will be required to introduce a pulse compressor in the beam path to counter chirping of the pulse introduced by the Pockels cell and other additional optical components In addition 65 the Pockels cell could be synchronized to the frame clock of the software to increase the laser intensity as a function of imaging depth in the specimen Addition of this feature in the software will enable hands off imaging of the thick biological specimen Further feedback from the detecto
78. ide A A frame of cleaned data from a control slice culture 5DIV experiments B To avoid double time counting the data was thresholded to count nuclei in the focal same plane The nuclei that are in deeper plane appear fainter and smaller in size in the current plane C Reliable counting of nuclei by the software D E F Similar analysis for data from other detection channel propidium iodide for the same sample and same focal plane 116 APPENDIX F PUBLISHED WORKS Custom made multiphoton microscope for long term imaging of neuronal cultures to explore structural and functional plasticity Komal Rambani Mark C Booth Edgar Brown Ivan Raikov and Steve M Potter Proc SPIE Vol 5700 p 102 108 Multiphoton Microscopy in the Biomedical Sciences V 2005 We use dissociated cultures of E 18 rat cortical neurons to study how they process the information To correlate the electrophysiological data with corresponding network structure we observe effects of the stimuli on the structural changes in this culture using multiphoton microscopy To keep our 2D and 3D cultures alive for long term studies it is necessary to protect them against photodamage At the same time we need a flexible microscope design to accommodate our multielectrode arrays MEAs electrophysiological station We have constructed a custom designed multiphoton microscope based on design of Tsai et al The microscope is optimized for the two photon mode to collect
79. ility and Signal to Noise Ratio in Brain Slice Recordings with Perforated Microelectrode Arrays 5 International Meeting on Substrate Integrated Micro Electrode Arrays Reutlingen 2006 29 EGERT U HECK D and AERTSEN A 2 dimensional monitoring of spiking networks in acute brain slices Exp Brain Res 142 268 274 2002 30 EHRENGRUBER M U ET AL Recombinant Semliki Forest virus and Sindbis virus efficiently infect neurons in hippocampal slice cultures Proc Natl Acad Sci USA 96 7041 7046 1999 31 ENGERT F and BONHOEFFER T Dendritic spine changes associated with hippocampal long term synaptic plasticity Nature 399 66 70 1999 32 FELDMAN D E and KNUDSEN E I An Anatomical Basis for Visual Calibration of the Auditory Space Map in the Barn Owl s Midbrain J Neurosci Volume 17 17 pp 6820 6837 1997 33 FELSEN G TOURYAN J HAN F and DAN Y Cortical sensitivity to visual features in natural scenes PLoS Biol 3 10 e342 2005 34 Fu Y Djupsund K GAo H HAYDEN B SHEN K and Dan Y Temporal specificity in the cortical plasticity of visual space representation Science 296 2002 35 GAHWILER B H Organotypic monolayer cultures of nervous tissue J Neurosci Methods 4 329 342 1981 36 GAHWILER B H Organotypic cultures of neural tissue Trends Neurosci 11 484489 1988 122 37 GAHWILER B H CA
80. inement of tissue in infusion chamber helps to preserve the thickness of tissue better compared to Stoppini s method The development of thick organotypic cortical brain slices will prove to be very useful model for various studies requiring multiple layers of the cortex especially when cut tangential to the cortical surface Keywords Organotypic brain slices Thick Cortical slice cultures microfluidic system 119 REFERENCES 1 AKTURK S GU X KIMMEL M WANG Z and TREBINO R Measuring Spatio Temporal Pulse Distortions Using Grenouille Commercial and Biomedical Applications of Ultrafast Lasers VII 2005 2 ARAYA R JIANG J EISENTHAL K B and YUSTE R The spine neck filters membrane potentials PNAS 103 17961 17966 2006 3 ARAYA R EISENTHAL K B and YUSTE R Dendritic spines linearize the summation of excitatory potentials PNAS 10 1073 2006 4 ARONIADOU V A and KELLER A Mechanisms of LTP Induction in Rat Motor Cortex in vitro Cereb Cortex 5 353 362 1995 5 BAKER R E CORNER M A and PELT J Spontaneous neuronal discharge patterns in developing organotypic mega co cultures of neonatal rat cerebral cortex Brain Res 1101 29 35 2006 6 BAKKUM D J SHKOLNIK A C BEN ARY G GAMBLEN P DEMARSE T B and POTTER S M Removing some A from AI Embodied Cultured Networks Embodied Artificial Intelligence Iida F Pfeifer R Steels L
81. infusion chamber by minimizing the space on the tissue surface that might result in tissue floating during the initial adhesion process when the infusion is started 82 A Figure 6 7 Representative micrographs of tissue nuclei labeled with Hoechst and Propidium iodide A z stack project of part of perfused tissue thickness indicating significantly higher number of live cells blue than the dead cells red B z stack project of part of control tissue thickness indicating significantly higher number of dead cells compared to perfused tissue The results of my experiments indicate that perfusion of nutrient medium allows increased viability of the cultured brain slice compared to two types of control cultures Further I examined viability for a series of flow rates 5 10 20 30u1 hr used to perfuse the tissue for two days in culture The long term 5 DIV perfusion experiments results suggest there is a linear increase in the viability of the cultures with increase in infusion rates Additionally both the 2 day and 5 day perfusion results demonstrate that approximately 3 culture volume exchanges per hour supply ample nutrients to the tissue resulting in over 80 viability of the tissue even after 5 days in culture figure 6 7 Some percentage of the cell death could be ascribed to the initial injury to the peripheral cell layers of the culture caused by the tissue cutting process This novel perfusion paradigm to culture thick b
82. l Acad Sci 93 8040 8045 1996 14 CHAO Z C BAKKUM D J WAGENAAR D A and POTTER S M Effects of random external background stimulation on network synaptic stability after Tetanization A Modeling Study Neuroinformatics 3 3 263 280 2005 15 CHEUSS V YASUDA R SOBCZYK A and SVOBODA K Nonlinear Ca2 Signaling in Dendrites and Spines Caused by Activity Dependent Depression of Ca2 Extrusion J Neurosci 26 8183 8194 2006 16 COLLIN C MIYAGUCHI K and SEGAL M Dendritic spine density and LTP induction in cultured hippocampal slices J Neurophysiol 77 1614 1623 1997 17 COLTMAN B W EARLEY E M SHAHAR A DUDEK F E and IDE C F Factors influencing mossy fiber collateral sprouting in organotypic slice cultures of neonatal mouse hippocampus J Comp Neurol 362 209 222 1995 18 DAvE A S and MARGOLIASH D Song Replay During Sleep and Computational Rules for Sensorimotor Vocal Learning Science Vol 290 no 5492 pp 812 816 2000 19 DEBANNE D GAHWILER B H and THOMPSON S M Cooperative interactions in the induction of long term potentiation and depression of synaptic excitation between hippocampal CA3 CAI cell pairs in vitro Proc Natl Acad Sci 93 11225 11230 1996 20 DEMARSE T B WAGENAAR D A BLAU A W and POTTER S M The Neurally Controlled Animat Biological Brains Acting with Simulated Bodies
83. lcein AM in a culture dish sealed with a baggy FEP membrane lid The images of the culture were captured every 5 minutes for 4 hours The images were imported in ImageJ software and cleaned to remove background noise A montage of first 42 time frames is shown in figure 5 10 The conversion of the 62 time lapse images to a movie clearly shows morphometric dynamics of interacting neurons and astrocytes in the developing culture Figure 5 10 Montage of 42 time lapse images of living neuronal culture 2DIV labeled with calcein AM The specimen was excited at 800 nm wavelength and fluorescent signal was collect with a band pass filter Images were taken every 5 minutes to make a time lapse movie Images were taken with 40x W 0 8 NA objective lens with a 512 x 512 pixel frame size The laser power was set at 7OmW laser power at turning mirror The scale bar is 20 um Compatibility with Multiple techniques In collaboration with our BRP partners Bruno Frazier and group we are developing a novel three dimensional microfluidic multielectrode neural interface system We tested the ability of our system to accommodate a complex fluidic set up to test delicate microfluidic arrays of prototype devices and characterize their fluidic properties 63 Figure 5 11 shows an image of such a fluidic array that was tested using dilute solution of 1 um fluorescent beads being pumped at the rate of 1ml hr through the device Fluorescent bead
84. lect the maximum possible fluorescent signal using the minimum excitation laser intensity Special attention is paid to get uniformly illuminated images and the ability to use the entire bandwidth of the pulsed laser 700 1000nm with the same set of optical components Flexibility of the design will allow us to easily change or incorporate other optical components suitable for different experimental needs Further this microscope will allow us to perform electrophysiology and imaging studies concurrently while maintaining the optimum temperature and CO levels using a life supporting environment chamber 28 INTRODUCTION Since the invention of the multiphoton microscope in the past decade multiphoton microscopy has revolutionized live cell imaging experiments to explore dynamic processes at molecular cellular and tissue level in the biology world With confocal microscopy both in vitro and in vivo imaging of living tissue is limited to a small number of scans of due to deleterious effects of photodamage resulting from repeated exposure to multiple lasers Multiphoton microscopy is a great tool to perform live cell imaging experiments It has several advantages over confocal microscopy which include reduced photodamage due to localized excitation of fluorophores in the specimen no need to focus the emitted fluorescent signal deeper imaging of scattering tissue due to lesser scattering of long wavelengths used for multiphoton excitation and im
85. lected signal we have selected a side on PMT with large active area of 8mm x 24mm Model R3896 Hamamatsu Inc This PMT has high quantum efficiency and a relatively flat plateau of sensitivity over the entire visible range of wavelengths as shown in figure 4 6 It is desirable to cool the PMT in a custom liquid cooled housing Photocool Inc to reduce the detector thermal noise to detect the small 40 fluorescent signal emanating from tiny morphological structures A high sensitivity PMT requires blocking of any residual laser light collected along with the fluorescent signal For this purpose an IR blocking BG39 glass is placed in front of the PMT Additional band pass filters can be mounted in a quick release mount to achieve enhanced contrast A careful selection of highly regulated power supplies for the PMT and a custom fabricated preamplifier along with cooled housing makes it an efficient low noise detection system 100 a zE gt gt ES Ow Go Mare iL Sui E 2E wS Qo I oe z O 0 01 100 200 300 400 500 600 700 800 900 1000 WAVELENGTH nm Figure 4 6 Sensitivity of the detector A nearly constant sensitivity across the visual range of spectrum ensures detection of all color photons equally An efficient detector unit translates to lower laser power to excite the specimen data Hamamatsu Inc 41 Laser Beam Shaper The laser beam of the femtosecond laser Coherent MIRA900 is 0 8mm
86. ls and star shaped cells could be identified in the baseline specimen indicating organotypic nature of the culture 8 These cell types could also be identified in perfused cultures Further perfused cultures showed similar cell density and cell sizes as baseline preparations while they were compromised in the unperfused control cultures Evidently the unperfused culture on the gold grid shows lesser cell density in the deeper layers of tissue In membrane insert cultures the cell density and size were found to be seriously compromised figure 7 2 92 Figure 7 2 Morphology Assessment A Baseline tissue showing common cell types found in cortex for example pyramidal cells cyan squares star shaped astrocytes green squares etc B Perfused cultured slice after 5DIV shows similar cell densities and cell shapes These cultures also show characteristic cortical cells like pyramidal cells cyan squares and star shaped astrocytes green squares etc C Unperfused control slice culture after 5 DIV shows reduced cell density approximately in the middle layers of the tissue D Unperfused membrane insert culture shows compromised health of cells and reduced cell sizes and cell densities Baseline Freshly cut tissue and stained and or fixed to assess the viability and the morphology at cellular level in the beginning of the experiment membrane insert culture means culture as described by Stoppini 1991 Thickness preservati
87. ly 35 36 51 114 Thick organotypic brain slice cultures will offer an advanced in vitro model for neuroscience research that requires larger portions of intact laterally and tangentially interacting stereotypic pathways with in or between different sections of the brain 5 75 However it has been a challenge to culture such thicknesses of nervous tissue It is generally believed that thick gt 300um brain slice cultures suffer necrosis in the center of the tissue due to diffusion limited insufficient supply of nutrients A convection based perfusion method that allows flow of oxygenated nutrient medium through the thickness of the tissue will provide nutrients to every cell in the tissue slice This may allow enhanced viability resulting from simple forced convective based artificial restoration of circulatory system of the tissue Here I report a unique convective flow based perfusion method to successfully culture 700um thick organotypic brain slices An infusion withdrawal type micro fluidic chamber was fabricated optimized and used to culture thick brain slices The results of 68 my experiments indicated that perfusion of oxygenated nutrient medium through the thickness of the tissue allows enhanced viability of thick brain slices compared to the slices cultured using traditional culturing methods after 2 days in vitro DIV and 5 DIV Further I assessed a range of non destructive perfusion rates that results in enhanced the tis
88. mber Burris Bryan Williams Chris Ruffin and Beth Bullock Spencer for several helpful events and their support to students I gratefully acknowledge excellent Georgia Tech student services I realized how warm and supportive the Georgia Tech community is when they supported me in my crisis times and helped me to bring my smiles back I especially thank to Maxine McClain Jelena Vukasinovic and Yoonsu Choi I will always remember all the good laughs and support you gave me in my good and bad times I thank to Potter group members for their constant support I am thankful to Zenas Chao and Douglas Bukkam from whom I learned a lot of things I also want to thank undergraduate researchers who worked with me or in Potter group during these projects I can never forget Bobby Thompson Ryan Hanes Herna Chris Grubb and Juan Estrada for good times that we shared while working in the lab You guys are the best and most hardworking undergraduates I have ever met I thank to all the people who helped me directly and indirectly in enhancing my personality Thanks to natural problem solvers for teaching me a lot of things and ingraining a new level confidence in me The laboratory of neuroengineering at Georgia Tech is a great place to get education Not only one learns about cutting edge research here but also it provides a dynamic environment of human behavior and interactions to an observer I especially am impressed with the skills of people who are n
89. mbodied system thus created is a simple in vitro model to study learning and memory Along with simultaneous multiphoton imaging this system provides non invasive techniques to record activity and morphology of the neuronal network at once Drawing Zenas Chao ie ee as Figure 2 2 Multielectrode Array dish for two dimensional neuronal cultures A A multielectrode array dish The central part of the dish contains an 8x8 array of 30um diameter microelectrodes that are separated by 200um from each other The electrode contacts are carried to the outer electrode pads for preamplifier contact B A multielectrode array dish sealed with a gas permeable teflon membrane This enclosed chamber allows long term viability of culture by preventing bacterial and fungal infections and maintaining osmolarity by preventing evaporation of medium C An image of high density dissociated cortical neuronal culture plated on multielectrode array er e e E gt gt ex a o e K e2 g P ote OEE DE E 8 K Pe Figure 2 3 Multielectrode electrophysiology recording and stimulation set up This set up contains preamplifier and custom made all channel stimulation system To perform simultaneous electrophysiology and imaging microscope should be able to accommodate this system along with microfluidic system in case of 3D cultures Although dissociated flat cultures represent neuronal networks that may encode informa
90. mirror separation Its large diameter will allow only the central almost constant thickness region for the scanning laser beam A tube lens is required to collimate the laser light going to the objective lens A binocular is also required to visually inspect the specimen during experimental setup We have use a trinocular head Zeiss Inc whose lens is used as the tube lens in our microscope system Ideally the fluorescent light collected from the focal plane of the objective should appear collimated at its back aperture In reality this is not true for highly scattering specimens especially living brain tissue A good objective lens with high numerical aperture collects the scattered emitted photons from a wider angle We use a set of four high numerical aperture 10X NA 0 30 20X NA 0 50 40X NA 0 80 63X 38 NA 0 95 Zeiss water immersion objective lenses with long 1mm working distance These lenses allow for deep imaging in the specimen while collecting more scattered photons A very important optical element which bridges the excitation and emission paths in a fluorescence microscope is the dichroic mirror DM It separates the excitation light from the fluorescence signal In our upright microscope it transmits the laser light and reflects the fluorescence signal Fluorescence signal collected from a scattering specimen via a 20X objective lens spreads to a bigger area 45mm by the time it reaches the dichroic mirror To
91. more information about fine details of specimen is obtained Two photon excitation spectra are usually broader than one photon excitation spectra so it is easier to excite multiple fluorophores with a single laser wavelength attributing to much less photo damage to the tissue Pulsed laser is not monochromatic like CW lasers Pulses contain 10 nm spectral bandwidth This property together with broader two photon excitation spectra contribute to get brighter images at the same laser power Biological tissues absorb less in IR regime of spectrum than UV and visible regime hence multiphoton microscope causes lesser thermal damage to specimen Only one tunable laser is required for entire range of fluorophores Above mentioned properties makes multiphoton imaging really advantageous for repeated imaging of a specimen live cell imaging deep tissue imaging and long term live tissue imaging 25 Pulsed lasers are expensive Complicated instrumentation due to pulsed laser Slightly poorer resolution than confocal imaging due to longer excitation wavelengths Snell s law of refraction and diffraction Tunable lasers can not achieve long enough wavelengths to excite fluorophores requiring red excitation wavelengths In fluorescence imaging there are several factors that may affect the quality of images acquired with any of the microscopy conventional confocal and multiphoton modes Some of these factors are list
92. morphological dynamics of micron level structures it is important to illuminate the specimen uniformly across its field of view Fluorescent plastics have homogeneous fluorescent properties I used a piece of green fluorescent plastic to image this property The specimen was imaged with a 20X W N A 0 5 objective lens at scan angle 2 degrees The image was normalized and a contour plot was obtained using Matlab I observed a gradual decrease of intensity from center to periphery The difference in the intensity from center to edges is up to 15 figure 5 5 The difference in illumination intensity across the field of view could be due to non uniform thickness of the scan lens or tube lens across the raster scan resulting in different level of stretching of pulses at different scan angles 57 20X W N A 0 5 illumination across Field of View of 512x512 frame ee TS apa 500 fo 450 400 350 300 250 200 150 100 50 50 100 150 200 250 300 350 400 450 500 Figure 5 5 Percentage of illumination across the field of view A green fluorescent plastic was imaged with 20X W N A 0 5 lens to estimate the illumination of the specimen across the field of view The image intensity is normalized and a contour plot is obtained using MatLab It is evident that the illumination is inhomogeneous across the field of view with more illumination in the center that falls by 15 towards the edges of the image SYSTEM VALIDATION IMAGING
93. multaneous multiphoton excitation of both fluorophores at 800nm wavelength The viability was assessed by counting dead and live nuclei using ImageJ software Appendix E 70 DESIGN Convective flow based culturing method Brain slices cultures are widely accepted and commonly used experimental paradigms over a century now 17 26 35 37 40 46 47 50 51 59 62 114 During this time several techniques were invented to culture viable brain slices These days the roller tube method and the membrane insert interface method are the most popular techniques to culture organotypic brain slices for wide gamut of molecular biology electron microscopy imaging electrophysiology pharmacology ischemia neurotoxicology traumatic brain injury and immunohistochemical studies 7 11 13 16 19 25 31 59 62 68 74 77 81 The roller tube method was the first successful organotypic brain slice culturing method pioneered by Houge It was later refined by Gahwiler 35 In this process 100um thick brain slices are harvested and glued to a cover slip using a plasma and thrombin clot The cover slip is in turn submerged in a test tube that is half filled with the nutrient medium For sufficient gaseous exchange the tubes are rotated at around 20 revolutions per hour rph such that the tissue is submerged in nutrient medium and then exposed to gases to supply oxygen in turns figure 6 1A B C With this method the tissue flattens down to monolaye
94. n the LabView software 22 Always first stop the PMT power supply and then the preamplifier supply 23 Switch off the chassis before switching off the computer to avoid failure of its recognition as a slave by the computer during its next session 24 Make sure that the power supply for the driving boards of the scanning mirrors is off before you leave gt TPLSM_ver1 File Edit Op Tools Browse Window Help Offset Gain Zoom Factor fo en Ji 100000 90000 ge POWER CONTROL POSITION CONTROL samo OE o 3 of MAX Focus Step 10 micron s 2 60000 Ne r g 80 uP S soooo 6 702 2 30000 ZERO ak SCAN MIRROR PARAMETERS 20000 ee Number of Pixels Number of lines tee 202 Forward Tup ap ee 102 DOWN z 4 500 250 0 250 500 750 1000 H536 Hfsiz ae ee 2 0 00 Number of Pixels Number of Pixels Backward Tdown Flyback Tdown Upper Limit TIME LAPSE CONTROL PANEL J Jeo e as RTT DT EAAN EPATRAS 255 Time Lapse Interval Ajs oses T aE 0 50 100 150 20 255 Corner Pi Corner Is a 2 Cut off Y Tan Cut off X Tan Lower Limit plni G E get g Pore at Z SERIES CONTROL PANEL Amplitude Amplitude a e LOG SO a 255 a en Sasa Range Range Start Position f 0 microns ie s Brightness Contrast Gamma BCG Values sale a p Ji J phe End Position io microns a C Pixel clock Hz Scan Cut off Comic a aie neeieng 128 Enter St
95. nance Energy Transfer FRET and photo uncaging Knowledge of these basic fundamentals principles and methods may prove to be helpful in imaging and troubleshooting for basic problems that naive fluorescent imagers may encounter 27 CHAPTER 4 CONSTRUCTION OF A DEEP TISSUE MULTIPHOTON LASER SCANNING IMAGING SYSTEM ABSTRACT Multiphoton microscopy is a relatively new imaging modality with submicron resolution that is becoming popular in a wide spectrum of investigations in life sciences due to its several advantages over confocal and conventional fluorescence microscopy Some of the featured advantages of multiphoton imaging include localized excitation of specimen dramatically reduced photobleaching and photodamage and greater imaging depth These features make multiphoton microscopy especially suitable to study living and thick biological specimen We are optimizing thick organotypic brain slices to extend our current 2D dissociated culture studies to 3D to understand mechanisms underlying learning and memory To keep our 2D and 3D cultures alive for long term studies it is necessary to protect them against photodamage At the same time we need a flexible microscope design to use multiple techniques on our specimen for example multielectrode array MEA set up and fluidic equipment We have constructed a custom designed multiphoton microscope based on design of Tsai et al 118 The microscope is optimized for two photon imaging to col
96. nated medium may allow enhanced viability of 700um thick brain slices compared to unperfused sister cultures Additionally I characterized the range of flow rates amenable to cultured tissue for enhanced viability The chapter seven explicates qualitative characterization of organotypic organization of the cultures and recording of spontaneous or chemically evoked electrical activity of these cultures using a single electrode In the last chapter I make recommendations for future work for further characterization of these cultures both morphologically and electrophysiologically I give directions to elaborate this study with other suggested flow rates and perfusion paradigms to complete flow characterization for enhanced tissue viability I also discuss possibilities to modify the culture chamber for larger and thicker cultures that may benefit several other neuroscience investigations requiring thick brain slice or co cultures This thesis has six appendices Appendix A provides a step by step user manual to operate our custom made multiphoton microscope Appendix B presents a trouble shooting manual to optimize the multiphoton microscope for its best performance In Appendix C a detailed protocol to set up the fluidic system and the details of the culturing method are given Appendix D provides detailed methods to label and image cultures for viability assessment and organotypic organization Appendix E describes methods to analyze data using
97. nesis in CA1 hippocampal dendrites induced by synaptic activity Science 283 1923 1927 1999 78 MCALLISTER A K Lo D C and KATZ L C Neurotrophins regulate dendritic growth in developing visual cortex Neuron 15 791 803 1995 79 MCALLISTER A K Biolistic transfection of cultured organotypic brain slices Methods Mol Biol 245 197 206 2004 80 MIYAGUCHI K MAEDA Y COLLIN C and SIHAG R K Gene transfer into hippocampal slice cultures with an adenovirus vector driven by cytomegalovirus promoter stable co expression of green fluorescent protein and lacZ genes Brain Res Bull 51 195 202 2000 81 MOLNAR Z and BLAKEMORE C Development of signals influencing the growth and termination of thalamocortical axons in organotypic culture Exp Neurol 156 363 393 1999 82 MORONI F ET AL Poly ADP ribose polymerase inhibitors attenuate necrotic but not apoptotic neuronal death in experimental models of cerebral ischemia Cell Death Differ 8 921 932 2001 83 MTCHEDLISHVILI Z and KAPUR J High affinity slowly desensitizing GABAA receptors mediate tonic inhibition in hippocampal dentate granule cells Mol Pharmacol 69 564 575 2006 84 NAGERL U V EBERHORN N CAMBRIDGE S B and BONHOEFFER T Bidirectional activity dependent morphological plasticity in hippocampal neurons Neuron 44 759 767 2004 85 NAUMANN T LINKE R an
98. nknown whether current flow rates provide sufficient time for such neuron glia life supporting material 98 exchange I did not attempt any study to characterize the necessity of supplying neurotrophic factors and their effect on viability Thus a more detailed study using pulsatile flow compared to continuous flow may be necessary to optimize nutrient supply in more detail while allowing sufficient time for physiologic neuron glial interactions Further measurements of oxygen level at various heights of the specimen at different flow rates may be a helpful parameter to optimize the perfusion paradigm flow rates continuous versus pulsatile flow 28 QUANTITATIVE ANALYSIS OF ORGANOTYPIC ORGANIZATION In the current study I characterized morphological organization and recording of electrical activity from these cultures only qualitatively Although H amp E staining tells much about the cellular organization size and cell types based on their morphology a more detailed organotypic characterization effect of perfusion on survival of neurons and astrocytes and physiologic network organization of neurons and astrocytes requires detailed studies using immunostaining methods A quantitative study of dendritic structures cell types inhibitory excitory pyramidal astrocytes and cellular organization in the characteristic layer structure of the cortex is desired to authenticate detailed organotypic organization of these cultures CHARACT
99. nstrumentation training that I acquired under his supervision I was invited to present in conference invited talk and teach a hands on workshop which is a great honor for a student Next I would like to thank my thesis committee members Dr Ravi Bellamkonda and Dr Richard Nichols for their marvelous guidance and support I gratefully acknowledge the NIH NSF CBN and the Whitaker funding to support this research work I am indebted to my Bioengineering Research Partnership BRP collaborators for accommodating me as a team member and for helping and advising me on my research work I especially thank Professor Ari Glezer and Jelena Vukasinovic for superb collaboration and support for thick brain slice cultures project It is a fantastic experience to work with them My special thanks also go to Maxine McClain and Eno Ekong for teaching me brain slicing I acknowledge Kacy Cullen and Zenas Chao for helping me with the statistical analysis of my data I am grateful to NIH for distributing an excellent software ImageJ for free that is very helpful for several image processing and imaging data analysis applications I acknowledge Johnafel Crowe for providing me training on multiphoton imaging and also to accommodate my long imaging or multiphoton construction schedules without which my projects would not have come to completion in a timely manner I would like to thank excellent administrative staff for their help My special thanks to A
100. nstruments chassis as its slave a condition for successful operation of the microscope and its software control ROUTING THE LASER BEAM TO THE MICROSCOPE 1 Switch on the Ti saph pulsed laser at least 30 minutes prior to starting the imaging session to obtain stabilized laser power that translates to the quality of the images 2 Check using a spectrometer that the femtosecond laser is mode locked 3 Flip the routing mirror in front of the femtosecond laser box aperture to switch path of the laser beam from the Zeiss 51OMETA microscope to the custom made microscope 4 Measure the laser power at the focal plane of the objective lens using power meter and adjust it between a range of 20 40mW using wave plate circular dial to obtain good images SWITCHING ON THE MICROSCOPE PARTS 5 Switch on the power supply of the scanning mirrors The driving boards of the scanning mirror are installed in a commercially available box On the side of that 102 10 11 12 13 14 box notice if both the LEDs have turned to green color When the power supply is turned on it takes 3 seconds for the LEDs to turn green from orange Switch on the power supply of the NI chassis containing the interfacing cards Notice if all the four LEDs are glowing on the front panel of the chassis Switch on the computer The green color of all the chassis LEDs ensures that computer has recognized chassis as its slave Switch on
101. o not allow one to study morphological and electrophysiological properties simultaneously on in vivo preparations Understanding how morphology and activity are correlated will inform much about the network properties of neurons underlying the encoding and storing of information An in vitro neuronal preparation allows non invasive multisite electrophysiology and imaging of the neuronal network simultaneously using extracellular microelectrode arrays and multiphoton imaging Due to controlled inputs outputs better accessibility absence of interference from peripheral inputs and their amenability to the use of multiple technologies concurrently in vitro preparations provide a much simpler platform to understand neuronal functioning Learning is defined as encoding of information that the brain receives from sensory organs when an animal behaves in its environment However in vitro preparations lack sensory inputs and do not have any meaning to motor outputs Our laboratory is developing a novel hybrid in vitro model of artificially embodied neuronal cultures with robotic bodies HYBROTS to study learning and memory at the network level figure 2 1 This ambitious project requires several technologies to work simultaneously in real time which include development of long term neuronal cultures hardware and software for real time closed loop recording and stimulation of neuronal networks pattern recognition in the recorded data and their assignmen
102. of optical and mechanical components and construction of the microscope from these parts DESIGNING A MULTIPHOTON MICROSCOPE Desired features of microscope A good multiphoton microscope design allows the instrument to closely approach theoretical limits of efficiency while having flexible features to be user friendly Our research goals include simultaneous imaging and electrophysiology investigations on cultured dissociated neuronal networks Our near future goals are to conduct these studies on thick brain slices as well Thus taking into account near and long term imaging goals we require a multiphoton microscope that has following features flexible opto mechanical design to accommodate multiple experimental techniques such as electrophysiology and fluidic platforms and in vivo preparations allows long term time lapse live cell imaging allows deep imaging of thick specimen 30 easy access to optical components for quick changes for different experimental needs Fine precision XYZ stages for repeated imaging of different regions of interest of the same specimen during long term experiments ability to zoom in on interesting parameters during time lapse imaging by quickly changing the objective lenses sub micron resolution high optical efficiency to cover entire range of Ti saph laser and collect fluorescent signal from dim structures To achieve these goals I adopted the basic optical design of Tsai et al and modified i
103. on I measured thickness of the cultures using specimens prepared with H amp E staining To evaluate change in thickness due to fixing medium I cut fresh cortical slices of various thicknesses from 250um to 700m and fixed them using 2 paraformaldehyde in 0 5X PBS The fixed tissue was sliced into thin slices along its thickness stained with H amp E and mounted on gelatin slides With a calibrated field of view I observed less than 10 change in thickness of these slices Using this method I observed that thickness is 93 preserved better in perfused cultures compared to the membrane insert cultures figure 7 3 Thickness Preservation 700p 600 500 400 300 Thickness micro meters 200 100 Membrane Perfused Figure 7 3 Maintenance of thickness of culture after 5 DIV A Baseline B Perfused cultured slice after 5DIV C Unperfused control slice culture after 5 DIV D Unperfused static membrane culture Baseline Tissue cut freshly and stained and or fixed to assess the viability and the morphology at cellular level in the beginning of the experiment Static membrane culture means culture as described by Stoppini 1991 Electrophysiological activity of cultured thick brain slices Further I attempted to record functional activity from the perfused slice cultures Using a single micro wire electrode I probed the cultures at different randomly chosen places throughout their thickness Most of the ac
104. on optics are coated with antireflection coatings except the scan lens and the mirrors are coated with multilayer dielectric coatings to avoid 4 loss of laser power at each air optics interface The laser was tuned at 800 nm wavelength and the laser power was measured before the microscope and at the focal plane of the objective lens figure 5 3 55 a oo Oo N rs L 4 3 Laser power before the periscope mirror mW L 0 2 4 6 8 Laser power at focal plane mW o Figure 5 3 Efficiency of excitation pathway to transport laser power at the focal plane Chirping of laser pulses Laser pulses are not monochromatic They possess spectral bandwidth of 9 nm with a temporal length of 100 200 fs with all the wavelengths gaussian distributed across the central wavelength Attributing to different velocities of different wavelengths in a medium of refractive index different than air n 1 for air the pulses get distorted due to group velocity dispersion The chirped pulses reduce the excitation efficiency of the system To counter this problem we chose a minimal number of optical lenses on the laser beam path Using a novel device the Grenouille constructed by our collaborators Prof Rick Trebino and Group we measured pulse properties of a 80fs pulse before the microscope and at the focal plane at 800 nm wavelength 1 44 Without the Pockels cell in the beam path the pulses got elongated by only 15 by the current optic
105. onment of a humidified incubator 5 CO2 9 O2 65 RH and 35 C temperature 12 Adhesion methods At lower flow rates lt 20ul hr laminin coating facilitates adhesion of the tissue to the gold grid and the inner walls of the infusion chamber A FEP membrane containing teflon lid provides a gas permeable surface from the top of the tissue to hold the tissue down to the chamber while allowing gaseous exchange to equilibrate the nutrient medium with the incubator environment At higher flow rates a tissue culture compatible weight a Millipore membrane attached to an approximately 6mm diameter gold ring using a thin layer of PDMS was used to facilitate tissue adhesion to the infusion chamber Viability assessment The tissue viability was assessed using cell permeant and non permeant fluorescent nuclear labels Hoescht and Propidium Iodide respectively The cultures were labeled with 20u1 of Propidium iodide and Hoescht mixed in 200u1 of nutrient medium The cultures were incubated with fluorescent labels for 30 40 minutes while they were being perfused at the regulated flow rate The flow was stopped and the infusion and withdrawal ports of the chamber were sealed to avoid evaporation of the nutrient medium during imaging The z stack images of dead and living nuclei of tissue were collected using 20X NA 0 5 Achroplan water immersion objective lens Zeiss in two separate detector channels when the specimen was excited and scanned for si
106. ot only keenly engaged their research but also promote responsible conduct in research and teach ethics to younger generations of scientists and engineers They are doing a great work to teach these to students which helps them to survive and do justice in the advanced level of their career I feel grateful to opportunities to interact with such great personalities from time to time Last but not least I acknowledge enormous support from my husband and family I thank to all my friends with whom I shared great times while studying at Georgia Tech It is really a great experience to be a part of Georgia Tech community vi TABLE OF CONTENTS ACKNOWLEDGEMENTS LIST OF TABLES LIST OF FIGURES SUMMARY CHAPTER 1 INTRODUCTION 2 BACKGROUND 3 INTRODUCTION TO FLUORESCENCE MICROSCOPY ABSTRACT INTRODUCTION The ELECTROMAGNETIC SPECTRUM PHYSICS OF FLUORESCENCE ONE PHOTON VERSUS MULTIPHOTON EXCITATION GROUP VELOCITY DISPERSION CHIRPING FLUORESCENCE MICROSCOPY 4 CONSTRUCTION OF A DEEP TISSUE MULTIPHOTON LASER SCANNING IMAGING SYSTEM ABSTRACT INTRODUCTION DESIGNING A MULTIPHOTON MICROSCOPE Desired features of microscope Challenges posed by pulsed laser vil Page iv Xi xii XV 16 16 16 17 18 21 23 24 28 28 29 30 30 31 Optical Design SYSTEM CONSTRUCTION Laser system Scanning and detection unit Laser beam shaper Software control Environment chamber DISCUSSION 5 VALIDATION OF A CUSTOM
107. ptured at increasing z level using a 40X water immersion lens N A 0 8 The images were imported into ImageJ software to make a z projection figure 5 8 Next we tested this system to image a living 60 hippocampal brain slice labeled with nuclear stain Hoechst figure 5 9 The image shows a Z projection of the entire stack of images Figure 5 8 Z projection of a pollen grain Pollen grains embedded in agar slide were imaged with 40X W 0 8 objective lens at different z elevations to obtain 3D image The images of the z stack were merged to an average mean z projection using ImageJ utilities The brightness and contrast are enhanced in imageJ software The frame size is 512x512 pixels Shown frame is subset cropped from the original image This image demonstrates ability of the microscope to image micron level details of the specimen like spines of this pollen grain 61 Figure 5 9 Z projection of Hippocampal slice labeled with nuclear stain Hoechst The images are taken with a 20X W N A 0 5 objective lens at different z elevations The frame size is 512x512 pixels The images were merged to an average mean z projection using imaged utilities The different elevations are reflected as difference in intensity level of the nuclei The image is raw data with background noise Time lapse Imaging Further we tested ability of the system for time lapse imaging A two day old mouse neuronal network was labeled with Ca
108. r thickness However the culture maintains some organotypic structure 71 Gold Grid Brain Slice Thinned _ Brain Slice with necrosis FEP lid g Millipore i membrane Tube containing tissue attached to 5 cover slip and nutrient medium Figure 6 1 Comparison of different culturing methods A B C Gawhlier s Roller tube method D E Stoppini s membrane insert culturing method F G Our perfusion based culturing method Later in 1990 Stoppini et al 114 invented a simpler interface type method to culture organotypic brain slices on submerged permeable membrane inserts In this method 150 450um thick brain slices are harvested and placed directly on a membrane insert Millipore The membrane insert is floated on the nutrient medium inside a small culture dish in such a way that the brain slice gets nutrients by diffusion from the bottom side The top surface of the tissue is covered with a thin layer of nutrient medium resulting from capillary action however it is in direct contact with the air to allow the gaseous exchange from the top surface fig 6 1D E In this method also the tissue thins down to 5 6 cell layer thickness and spreads laterally 111 These two methods use diffusion as the primary source of nutrient supply to the tissue and are not successful at culturing thicker gt 400um brain slices Organotypic brain slice cultures thin down significantly from their originally cut thicknes
109. r to the pockels cell could be used to modulate the power of the excitation laser within the image plane This feature would allow observation of tiny structures more clearly Together addition of more detection channels modification of software to synchronize the frame clock to the power modulation by the Pockels cell and stopping the scanning mirrors during lapsed time of time lapse imaging will allow automated imaging in time lapse and deep tissue imaging modes Addition of a high numerical aperture objective lens an appropriate bandpass filter and a detector in the transmission path will allow collection of second harmonic generation signal along with two photon imaging 66 CHAPTER 6 CULTURING THICK ORGANOTYPIC BRAIN SLICES A PERFUSION BASED CULTURING METHOD ABSTRACT Brain slices are widely accepted in vitro models for wide spectrum of neuroscience investigations today However culturing thick organotypic brain slices is a challenge due to necrosis that starts in the center of the tissue as a result of insufficient diffusion based supply of nutrients We hypothesize that a convection based supply of oxygenated nutrient medium through the thickness of the tissue will provide sufficient nutrients to every cell and may result in enhanced viability of the tissue In this chapter I report fabrication and optimization of a novel infusion withdrawal type micro perfusion chamber to culture 700um thick brain slices The results sugges
110. rain slice cultures could be extended to culture slices from other parts of the brain as well With the modified design 83 and fabrication of the gold grid device one could extend this technique to culture thick organotypic co cultures of any lateral dimensions Further these devices could be modified to culture even thicker brain slices We expect thick brain slice cultures will provide an advanced experimental platform to several neuroscience and neuroengineering researchers who are working in the fields of learning and memory traumatic brain injury neuropharmacological studies neurotoxicity studies ischemia studies neural implant in vitro biocompatibility studies for prosthesis applications neural implant glial scan formation studies and so on 84 CHAPTER 7 CHARACTERIZATION OF THICK ORGANOTYPIC CORTICAL SLICE CULTURES ABSTRACT Thick organotypic cortical slice cultures might prove to be valuable in vitro models in a wide spectrum of neuroscience investigations Viable 700um thick brain slices can be successfully cultured in vitro by perfusion of nutrient medium through the tissue In this chapter we present qualitative characterization of organotypic organization and functional activity of the organotypic 700um thick brain slice cultures Further we demonstrate that our culturing method maintains greater thickness of the tissue compared to the control cultures even after 5 days in vitro DIV by providing additional mechan
111. res image display without storing line scan custom frame size of images different scan speeds 2D imaging 3D imaging z stack specimen XY stage control Z positioner control 2D and 3D time lapse imaging optical zoom by changing scan angle of scanning mirrors saturated pixel display and laser power modulation by software control of Pockels cell figure 4 8 The collected signal is stored as data files that can be converted to readable image files such as tiff jpeg bmp and png using custom developed LabView user interface utilities package A step by step microscope operating manual and a trouble shooting manual are given in appendices B and C 44 To X Scan Board To Y Scan Board To PMTs Pixel Clk Gate RTSI BUS pins AO UPDATE AO DAC 0 AO DAC 0 CTR GATE DATA ACQ AI DAC 0 Al DAC 1 atd Acquisition Gate Digital to analog cards Data acquisition card Figure 4 7 Schematic diagram of synchronized scanning and data acquisition control to form basis for software control 45 E gt TPLSM_ver1 6 i T y Eile Edit Operate Tools Browse Window Help 3 be gt on Frame Imaging Mode Continuous Frame Current Directory X C data imagestack bin Zoom Factor Ji Offset Gain Ep 100000 90000 PMT VOLTAGE Vmax 1250 V SS SS av Perrepeene A 0 250 500 750 1000 1250 AMD POWER CONTROL 70000 WOE o a ofmax 60000
112. rfusion of oxygenated nutrient medium will allow better viability of thick brain slices We have developed a simple closed loop infusion withdrawal type microfluidic chamber to allow perfusion of 700um tangential or transverse cortical slices Vukasinovic Proc BIO2006 The chamber is covered with a teflon membrane lid to allow sterility and gaseous exchange with incubator Potter J Neurosci Meth 2001 The infusion and withdrawal of medium is done at the same flow rate via syringe pump and teflon u capillary tubing that connect the p fluidic chamber to the medium containing syringes The fluidic system is kept in a humidified incubator maintained at 9 O2 5 COz and 65 humidity The infusion capillary tubing 118 contains an aerator and a one way check valve to allow equilibrium of nutrient medium with the incubator environment and to ensure smooth flux of fluid that reaches the tissue The design of this system requires the infused medium to pass through the thickness of the tissue that is supported on a porous gold grid before reaching withdrawal chamber which encourages perfusion of oxygenated nutrient medium at all levels of tissue throughout its thickness The viability of the tissue was assessed by propidium iodide and Hoechst fluorescent probes and multiphoton microscopy We have observed better viability in perfused cortical slices compared to the unperfused slices at different flow rates of oxygenated nutrient medium Further conf
113. ro J Neurophysiol 78 1651 1661 1997 61 Kovacs R ET AL Free radical mediated cell damage after experimental status epilepticus in hippocampal slice cultures J Neurophysiol 88 2909 2918 2002 124 62 KRASSIOUKOV A V ET AL An in vitro model of neurotrauma in organotypic spinal cord cultures from adult mice Brain Res Brain Res Protoc 10 60 68 2002 63 LETINIC K ZONCU R and RAKIC P Origin of GABAergic neurons in the human neocortex Nature 417 645 649 2002 64 LEUTGEB J K FREY J U and BEHNISCH T LTP in cultured hippocampal entorhinal cortex slices from young adult P25 30 rats J Neurosci Methods 130 19 32 2003 65 Li Z ET AL Synaptic vesicle recycling studied in transgenic mice expressing synaptopHluorin Proc Natl Acad Sci USA 102 6131 6136 2005 66 Li and MclIlwain Maintenance of resting potentials in slices of mammalian cerebral cortex and other tissue in vitro Journal of physiology London 139 178 190 1957 67 LIM C BLUME H W MADSENJ R and SAPER C B Connections of the hippocampal formation in humans I The mossy fiber pathway The Journal of Comparative Neurology 385 3 325 351 1998 68 LINKE R HEIMRICH B and FROTSCHER M Axonal regeneration of identified septohippocampal projection neurons in vitro Neuroscience 68 1 4 1995 69 Lo D C McAllister A K and Katz L C
114. rom the infusion chamber This chamber contains a small hole on one side of the withdrawal chamber to attach the withdrawal micro capillary tubing B Perfusion chamber made out of PDMS material using thermoplastic mould C The infusion chamber holds a 3mm in diameter gold grid usually used for SEM TEM microscopy The pore size in the gold grid is 54 x 54 um making up a total of 40 area open for the fluid injection D The perfusion chamber is covered with FEP membrane containing teflon lid When attached to infusion and withdrawal lines this chamber provides means to control the tissue environment and helps to prevent infections E Perfusion set up to perfuse oxygenated nutrient medium while maintaining pH temperature and osmolarity of the culture Drawing Infusion withdrawal of medium at the same flow rate maintains the pressure gradient across the tissue The set up is kept in a humidified incubator controlled at 5 COs 9 Ox 65 RH and 35 C The infused medium passes through the aerator and gets equilibrated with the incubator environment before reaching the tissue The teflon membrane lid prevents evaporation of medium and thus maintains osmolarity of nutrient medium Inset Experimental set up for four slices simultaneously using a single syringe pump Flow trajectories in the infusion chamber To ensure nutrient supply throughout the tissue volume chamber volume our collaborator Jelena Vukasinovic conducted micro Particle Image Velo
115. rtunity to extend these studies over longer time periods to study a wide range of mechanisms that include neurogenesis 105 synaptogenesis 88 regeneration 68 protein expression using viral vectors 30 56 72 73 and simulated traumatic brain insults 62 Additionally the accessibility of the preparation permits multiple noninvasive techniques to be applied simultaneously 1 2 20 21 24 13 16 52 53 57 85 92 73 74 96 109 112 e g multielectrode arrays multiphoton imaging pharmacological manipulations etc However current culturing methods allow only 5 6 cell thick organotypic cultures for longer term viability 111 A method to culture thick brain slice cultures will provide a novel platform to extend these studies over wider and deeper areas of neuronal networks for longer time periods Here I report successfully cultured 700um thick cortical slices of 3mm diameter However the culture chamber could be easily modified to accommodate any smaller or larger radial dimensions of the tissue In this study I qualitatively evaluated organotypic organization of the culture after 5 days in vitro The culture appears to maintain healthy cells of characteristic morphological shapes Additionally I found that these cultures 86 maintain thickness gt 80 of the actual thickness of the tissue Further I was able to record spontaneous or chemically evoked activity in most of the cultures Together these results indicate healt
116. s This could be because of two possible reasons a lack of mechanical support that promotes lateral migration of cells or cell death due to insufficient nutrients It is believed that the tissue suffers necrosis in the middle of the slice due to insufficient 72 diffusion limited nutrient and oxygen supply Consequently the ischemic cells eventually die and the tissue thins down to 150um thickness within 2 days in culture 114 A method that supports active transport and exchange of ample nutrient medium and gases oxygen and carbon dioxide throughout the thickness of the tissue may allow one to culture viable thick organotypic brain slices To test this hypothesis we have devised a micro biofluidic chamber in collaboration with Ari Glezer and Jelena Vukasinovic This device not only allows the convective perfusion of nutrient medium throughout the thickness but also provides mechanical support to the tissue on its circumferential sides figure 6 1F G The micro Fluidic chamber and the closed loop perfusion Set up To support 3D flow through the tissue thickness a coaxial bicylindrical device with common base was designed and constructed by our collaborators Ari Glezer and Jelena Vukasinovic by molding PDMS in wax molds The wax molds were prototyped using 3D Systems Thermojet printer and a 3D CAD of the device The inner cylinder called the infusion chamber has a gold micro grid PELCO on an orifice 700um deep from the top surface of
117. s which ensures high efficiency to excite the fluorophore figure 5 4 56 Before Microscope At focal plane of objective lens Temporal intensity and Phase JB Spectral intensity and Phase E Temporal intensity and Phase F _Spoctral intensity and Phase Jei asey Intensity a u pea eseud oo T T T T T T T T T T T T T T T T T T aoo 200 100 20 100 200 90 780 790 ny cto us e 300 200 100 0 100 200 300 780 79 800 810 820 830 c ime fs favelength nm Time fs Wavelength nm Wavelength nm Wavelength nm Wavelength nm 395 0 100 200 300 200 100 0 100 200 30C Delay fs Delay fs 300 200 100 90 T T T T T 300 200 100 0 4100 200 300 300 200 100 0 100 200 300 Delay fs Delay fs Figure 5 4 Spatial and temporal properties of laser pulse before the microscope left and at the focal plane of objective lens right A B temporal and spectral properties of pulse before the microscope C D measured raw data and cleaned data E F temporal and spectral properties of pulse after the microscope at focal plane of the objective lens G H measured raw data and cleaned data There was no Pockels cell in the beam path Few elements in the excitation pathway introduced chirping only unto 15 for a 80 fs pulse of laser tuned at 800nm wavelength Illumination of the image across the field of view For several imaging applications requiring studies of
118. spriingen Ann Phys 9 273 295 1931 45 Gu X AKTURK S SHREENATH A CAO Q and TREBINO R The Measurement of Ultrashort Light Simple Devices Complex Pulses XFemtosecond Laser Spectroscopy ed P Hannaford Springer Science Business Media Inc 2005 46 HARRISON R G Observations on the living developing nerve fiber Proc Soc Exp Biol Med 4 140 143 1907 47 HARRISON R G The outgrowth of the nerve fiber as a mode of protoplasmic movement J Exp Zoo 142 5 73 1959 48 HEINEMANN U ET AL Cell death and metabolic activity during epileptiform discharges and status epilepticus in the hippocampus Prog Brain Res 135 197 210 2002 49 HEINEMANN U ET AL Coupling of electrical and metabolic activity during epileptiform discharges Epilepsia 43 5 168 173 2002 123 50 HILTON K J BATESON A N and KING A E A model of organotypic rat spinal slice culture and biolistic transfection to elucidate factors that drive the preprotachykinin A promoter Brain Res Rev 46 191 203 2004 51 HoGuE M J Human fetal brain cells in tissue cultures their identification and motility J Exp Zool 106 85 107 2006 52 HOLTMAAT A WILBRECHT L KNOTT GW WELKER E and SVOBODA K Experience dependent and cell type specific spine growth in the neocortex Nature 441 979 983 2006 53 IKEGAYA Y AARON G COSSART R ARONOV D LA
119. sue viability MATERIALS AND METHODS Brain slice culture The brain slices were harvested from P11 P15 mouse pups of strains C57BL 6J and B6 Cg Tg Thy1 YFP 16Jrs J Jackson Laboratory and were mounted on the microfluidic infusion chamber The pups were euthanized using isofluorane in accordance with approved protocols Under sterile conditions the euthanized pup was decapitated and brain was removed and immediately stored in chilled pre oxygenated nutrient medium for approximately 1 minute The brain was cut into two hemispheres using a micro knife and each hemisphere was sliced to obtain 700um thick tangential or coronal cortical slices using a tissue chopper The sliced tissue was immediately transferred again to chilled nutrient medium and the slices were separated using micro spatulas under a dissection microscope This entire procedure from decapitation to slice separation took place with in 5 6 minutes The 700um thick cortical slices were cut into 3 mm round disks to fit snugly in the infusion chamber using a biopsy tissue cutter These tissue slice discs were transferred to new sterile culture dish containing 2 ml of chilled nutrient medium and were transferred to a laminin coated infusion chamber under sterile conditions inside a laminar flow hood The microfluidic culture chamber was then 69 enclosed with teflon sealed lids 97 The syringe pump was started and the entire set up was transferred to culture friendly controlled envir
120. t that continuous perfusion of oxygenated nutrient medium through the tissue allows enhanced viability of 700um thick brain slice cultures Further I investigated range of flow rates to obtain enhanced viability The results suggest that approximately three culture volume exchanges per hour is the best perfusion rate for enhanced viability of thick brain slices INTRODUCTION The brain has stereotypic characteristic organization in terms of functional units and pathways with in each species Brain slices preserve in vivo like cyto architecture and offer several advantages over in vivo models due to easy tissue accessibility and controllability of input output variables while using multiple advanced techniques 67 simultaneously e g multiphoton imaging and multi site multielectrode recording In the past two decades neuroscience investigations using explants of brain slices have contributed a plethora of information underlying the functioning of the central nervous system Thus brain slice cultures are valuable in vitro models for various electrophysiological morphological pharmacological ischemic and traumatic brain injury studies Development of organotypic brain slice cultures has opened avenues to extend several investigations done on acute brain slices over longer time periods up to several weeks With current methods brain slices can be maintained in culture only up to few micrometers thickness due to diffusion limited nutrient supp
121. t to meet our needs 118 Unlike confocal microscopes the multiphoton microscope needs a pulsed laser which has a peak power of the order of few kilowatts required for the multiphoton excitation and an average power of a few milliwatts that reduces the damage due to heating of the specimen Multiphoton excitation is a nonlinear process which involves absorption of two or more photons by the fluorophore with sum total energy equivalent to the energy required to excite the fluorophore 22 44 The probability of absorbing two photons simultaneously requires a very high density of photons which is achieved by using pulsed lasers with low duty cycle and high peak power Challenges posed by a pulsed laser Group Velocity Dispersion As opposed to continuous wave lasers used in confocal microscopy a pulsed laser is not monochromatic The pulses of a laser have a spectral width A and a temporal length t that are related as 31 AA x Ay P Equation 4 1 where Ao is the central wavelength of laser pulse Thus a shorter pulse has greater spectral width In an ideal pulse all its spectral components are uniformly distributed across its temporal length A pulse with small temporal length and large spectral width is ideal to achieve the maximum fluorescence intensity resulting from multiphoton excitation of a fluorophore The limitation in the instrumentation of a multiphoton microscope arises due to dependence of the refractive index of a medium on
122. t to motor commands behavior translation of robot or simulated animal behavior to stimulation patterns to send sensory information to the cultured neuronal network and a non invasive sub micron resolution imaging system amenable to electrophysiology and fluidic equipment Our current and alumni laboratory members have successfully developed ANIMAT long term dissociated cortical neuronal cultures on flat multielectrode arrays figure 2 2 real time closed loop recording and stimulation of neuronal networks figure 2 3 and electrical stimulation patterns to control activity of the dissociated cultured neurons 20 21 97 98 102 103 121 125 We are currently working towards pattern recognition in the recorded data and their assignment to motor commands behavior translation of robot or ANIMAT behavior to stimulation patterns to send sensory information to cultured neuronal networks 14 Microscopy Electrophys Modeling Simulation a Gripper Figure 2 1 AHYBROT Model to study learning and memory in vitro A dissociated culture of a few thousand neurons plated on a planar multielectrode array MEA is interfaced to a robotic body The activity patterns of the neurons in MEA dish are encoded as simple motor commands that allow robotic body to behave in its environment The behavior of the robotic body is encoded in stimulation patterns that are delivered to neuronal network using multielectrode array A closed loop hybrid e
123. tates that the energy of a photon is inversely proportional to its wavelength i 17 E hv he Equation 3 1 where h Planck s constant and c speed of light in vacuum Therefore shorter bluer wavelengths have higher energy than longer redder wavelengths The Rayleigh scattering principle states that the intensity of the scattered photons I is inversely proportional to the fourth power of their wavelengths la as Equation 3 2 gt In other words the scattering of blue light is 9 4 times as great as that of red light for equal incident intensities PHYSICS OF FLUORESCENCE In accordance with its electronic configuration every molecule or atom has a ground state of its electrons and the singlet or triplet excited states The ground and excited energy levels are separated by a characteristic energy difference called band gap energy Each molecule or atom can absorb incident photons of energy equivalent to its energy band gap to reach one of its excited states and emits longer wavelength photons to come back to its ground state The molecules that can absorb in ultra violet UV visible or near infrared IR spectra and can emit photons in the visible regime are called fluorophores The process of emission of photons in visible regime after the excited molecule or atom returns back to its ground state is called fluorescence More formally the fluorescence activity can be schematically illustrated with the classical J
124. the heater fan The chamber is made completely double walled with bubble wrap The surface of the optical table and the X 95 rail frame of microscope are also insulated to prevent any heat loss Currently this chamber can be maintained at 20 31 C temperature A higher wattage heater will be required to maintain physiologic temperature 37 C inside the chamber The eye pieces of microscope were made external for convenient visual inspection without opening the door during long term time lapse imaging Undergraduate researcher scholars Bobby Thompson and Christopher Grubb contributed significantly in the construction and optimization of this chamber A commercial CO sensor and a CO tank supply is used to regulate CO level in the chamber while imaging While this chamber serves the purpose as a life support system for the cultures it also helps to prevent exposure of detectors to stray light in the room that might result in background noise in the images A complete microscope is shown with and without environment control in figures 4 9 4 10 47 Figure 4 10 Custom fabricated multiphoton microscope A Full front view of the microscope B Scan plate containing upper periscope mirror alignment pinholes and scanning mirrors Scanning mirrors are mounted on custom fabricated high conduction and thermal loss aluminum plate to ventilate heat generated during high scanning speeds or long term scanning DISCUSSION We have descri
125. the power supply of the preamplifier first It is required to switch it on before the power supply of the detector to avoid damage to the detector system and the preamplifier Now switch on the detector PMT power supply The correct power needed to operate the PMT is stored on the memory of its power supply Hit recall button twice to supply power to a custom made high voltage circuit enclosed in a custom box of the PMT Do not touch high voltage wires thick red wire it carries 1250V Switch on the Z control power supply from the Corvus controller Start the custom written LabView software A user interface will appear The MXI interface code will recognize the interface cards to operate the microscope remotely via the software Place the test specimen under the objective lens Make sure that the BG glass filter and the appropriate band pass filters are present in the path before the liquid light guide 103 15 After visual inspection of the specimen bring the dichroic mirror cube in the center to route the fluorescent signal to the detector Pull the mirror sliding rod on the upper right side of the trinocular out of the laser beam path OPERATING THE SOFTWARE FOR IMAGING 16 Figure A 1 and its caption show the essential six steps to set the software to obtain a 512x512 pixel image 17 Do not change the scan mirror parameters until familiar with their meaning that is how do they translates to scan driving waveform
126. three dimensional 3D and time lapse imaging Further we discuss future directions to add more software and hardware features INTRODUCTION Multiphoton imaging has revolutionized life science investigations requiring live cell imaging to study dynamics properties from molecular to cellular to network level due to its several advantages over confocal imaging 99 Over the past decade there is an increasing number of revolutionary investigations in life sciences from molecular to systems level that have benefited from multiphoton microscopy such as protein trafficking on living neurons using quantum dots attached to molecular cues morphometric network properties of neurons at submicron resolution calcium imaging 50 from spine to network level of intact in vivo or sliced brain in vitro molecular imaging to unravel molecular interaction and their localization in the cells effect of pharmacological agents on targeted structures of cells photouncaging of molecules to study the effect of pharmacological agents and so on 2 3 7 15 16 24 31 52 53 74 75 77 92 109 110 Because of the non linear nature of multiphoton excitation that demands high photon density in excitation source it requires femtosecond pulsed laser A pulsed laser poses several challenges in the instrumentation of the multiphoton microscope and requires special features of optical components for an efficient microscope Most of the commercial microscopes h
127. tion it is argued that three dimensional cultures are a better representation of the in vivo like network organization In fact it is believed that there is stereotypic network circuitry of neurons underlying an activity pattern Further some studies show evidence that activity patterns shape the micro architecture of these circuits 15 29 Neural Processing Chip NPC Microfabricated Neural Interface System NIS 16x16 mucrotower H array with electrodes and fluidic ports Figure 2 4 Schematic diagram of microfluidic multielectrode neural interface system proposed in a bioengineering research partnership grant Organotypic brain slice cultures preserve intact cytoarchitecture of the brain Thus a hybrid model that includes an entire cortical thickness slice embodied with a robot 3D HYBROT would be a more in vivo like model of learning and memory To advance our 2D HYBROT technology to 3D HYBROTS there are two potential technology developments required first a method to culture thick organotypic brain slices and second a three dimensional multielectrode array that supports healthy organotypic brain slice cultures In collaboration with our bioengineering research partners BRP we are constructing three dimensional microfluidic multielectrode arrays that will have 1000 electrodes for targeted interface of cortical slices for stimulation and recording from different cortical layers figures 2 4 2 5 2 6
128. tion unit Laser System Scanning amp Detection Unit Scanning mirrors Scan lens Tube Lens Binocular Secondary dichroic Collection lens mirror Collection Periscope Unit z i Lenses Prescanning Unit i Primary n a a Dichroic Prechirp Unit mt mirror rex 1 Wave Beam ii Liquid Plate Expander ti light Objective lens l guide Band Pass BG Glass Filters Figure 4 3 Optical Diagram of the multiphoton microscope design The system has three units laser system prescanning unit and scanning and detection unit Specimen 35 SYSTEM CONSTRUCTION Laser system Each fluorophore has two photon excitation cross section Shorter pulses could have spectral width extended beyond the two photon cross section resulting in reduced excitation of the fluorophore at a given power A powerful laser system with broader tunable wavelength range promises multiphoton microscopy of broad range of fluorophores It is generally accepted that a Ti saph femtosecond laser has optimal wavelength tuning range and laser pulse properties for two photon excitation 113 We use the Coherent Mira 900 laser which has a tuning range of 700 1000nm with 100 200fs pulses of 9 nm full width at half maximum FWHM spectral bandwidth Currently we are sharing a laser system Chameleon Coherent Inc with a commercial inverted type two photon microscope A flip out mirror or a beam splitter is used depending on the
129. tivity could be recorded by chemically evoking action potentials by addition of 120mM KCl in the culture bath There was no evident relation of activity recording as a function of thickness However the amplitude of the recorded signal depended on the electrode s position relative to spontaneously firing neurons 94 a 1 b 60 40H 20 60 40 207 1 gt V pV 16 L L L L L L 4 L 1 L L L L L L L 159 215 159 22 159 225 159 23 159 235 159 24 74 44 74 45 74 46 74 47 74 48 74 49 74 5 74 51 74 52 gt time sec Figure 7 4 Activity traces from cultured slices after 5 DIV Either spontaneous or chemically evoked activity could be recorded in 75 of the cultured slices There were different amplitudes of recorded activity traces resulting from relative position of electrode to the firing neurons DISCUSSION AND CONCLUSIONS Here I described characteristic organotypic morphology and activity properties of the viable thick brain slice cultures Our perfusion method allowed ample supply of nutrients and oxygen throughout the tissue that resulted in increased viability gt 80 of tissue even after 5 days in culture However the viability was decreased over time from 2DIV to S5DIV The decrease in viability of perfused slices over time could be explained by slowly dying cells that might have released chemicals from the cut surfaces of the tissu
130. tivity of these cultures We anticipate that thick organotypic brain slice cultures will also benefit several other neuroscience and neuroengineering studies 15 CHAPTER 3 INTRODUCTION TO FLUORESCENCE MICROSCOPY ABSTRACT Today fluorescence imaging spans wide spectrum in modern biological investigations from molecular to network level at submicron resolution To take advantage of full potential of advanced fluorescence imaging techniques it is important to understand some underlying physical principles This chapter provides a framework to understand basic principles underlying fluorescence fluorescence microscopy one photon and multi photon laser scanning microscopy and advantages of multiphoton imaging of live tissue over single photon imaging INTRODUCTION Knowledge of various experimental techniques is an important factor for performing successful research in the interdisciplinary biomedical research Fluorescence imaging is one of the great tools for various biological investigations Over the past century fluorescence microscopy has evolved from linear to various non linear imaging methods because of its recognized power in biological and non biological research requiring submicron resolution of imaging These imaging methods have revolutionized various investigations from molecular to cellular to network level in modern biological and biomedical research due to its advantages over other techniques or as a complementary
131. tructure of tapered microfluidic array with one oval shaped fluidic port Left SEM picture of towers showing oval microfluidic port B Prototypic structure of three dimensional multielectrode arrays Image courtesy Yoonsu Choi Teflon capillary tubing Infusion Infusion withdrawal Syringe pump Incubator Laser beam Brain Slice Lid with FEP Membrane Optical fibe Gold grid EA Preamplifier Aerator Infusion port Withdrawal port Electrophysiology Station Figure 2 7 Schematic diagram of experimental setup for simultaneous imaging and electrophysiology on three dimensional organotypic brain slice cultures This is possible with the flexible design of our custom fabricated multiphoton microscope that can accommodate other experimental setups too Figure 2 8 An experimental fluidic setup to culture organotypic thick brain slices 13 The other side of the HYBROT learning project involves the ability to image the network morphology at sub micron level non invasively along with electrophysiology A multiphoton microscope is ideal for such needs As a BRP team member my contribution to 3D HYBROT project involved 1 development of a flexible custom fabricated multiphoton imaging system that can accommodate other technology platforms like multielectrode electrophysiology recording stimulation platform and microfluidic setup while imaging figure 2 7 and 11 development of a method to culture th
132. ture to record chemically evoked activity 89 RESULTS Viability of thick organotypic brain slice cultures Current popular methods to culture brain slices include the roller tube method and the membrane insert method 2 10 These methods exploit diffusion as the source of nutrient supply to the tissue throughout its thickness The tissue eventually flattens down to a monolayer in the roller tube method and 5 6 cells thickness in the membrane insert method 111 We have invented a novel perfusion method to culture thick brain slices Chapter 6 Using this method I evaluated viability of the cultures at non invasive perfusion rates after 2 days of perfusion and 5 days of perfusion figure 7 1 The perfusion of cultures showed increased viability as a result of increased perfusion rates both after 2 days and 5 days of perfusion The viability decreases over time Sdays versus 2 days even for perfused slices However the decrease in viability of unperfused slices was much greater than that of the perfused slices 90 W2 days in culture E5 days in culture 100 90 i 80 70 60 50 40 Viability 30 20 10 0 Membrane Unperfused 5ul hr 10ul hr 20ul hr Figure 7 1 Viability of culture after 2 days and 5 days in vitro The viability of the perfused cultures at 5DIV versus 2DIV decreases much less at 20ul hr flow rate compared to viability at 5ul hr flow rate thus indicating 20ul hr to be optimal flow rate to culture
133. ug addiction 86 ischemia studies 87 95 and so on 97 CHAPTER 8 THICK BRAIN SLICE CULTURING METHOD AND THE CULTURE CHAMBER RECOMMENDATIONS In the previous two chapters I described fabrication characterization and validation of a perfusion enabled culturing chamber that provides three dimensional flow of oxygenated nutrient medium through the thick tissue cultures Further I used this device to culture 700um thick brain slice cultures for five days in vitro In this chapter I make some recommendations for future work on this project OPTIMIZATION OF PERFUSION PARADIGM I tested continuous flow rates 5 10 20 30ul hr to evaluate amenable flow rates to brain tissue s mechanical strength and viability The results of my experiments indicate that forced perfusion at flow rates lt 20ul hr supports viable organotypic cortical slice cultures while higher flow rates are detrimental to the tissue health The viability of the cultures may be further characterized for additional flow rates 15ul hr and 251 hr to complete the spectrum of flow rate optimization studies Additionally I used continuous perfusion of the nutrient medium Currently tested flow rates indicate that approximately three culture volume exchanges of oxygenated nutrient medium per hour are optimal for enhanced viability of the tissue Other studies describe secretion of neurotrophic factors by glial cells that are necessary for normal functioning of neurons It is u
134. uld be equilibrated with the incubator environment It is advised to use the same nutrient medium preparation to avoid any pH or osmolarity shock to the culture 4 Remove the teflon lid from the culture chamber very gently to ensure that no harm is done to the tissue 5 Aspirate medium from the withdrawal chamber and pour the fluorescent dye mixed the nutrient medium 6 Close the chamber and transfer the set up to the incubator and wait for at least 40 minutes to hour to ensure intake of the dyes by the entire thickness of the tissue 7 Prepare for imaging 113 FIXING THE BRAIN SLICES AND STAINING THEM WITH H amp E 8 After imaging the slices are transferred to a 2 paraformaldehyde solution made in 0 5X phosphate buffer solution pH 7 4 9 Before transferring the slices in the fixing medium the pH and the osmolarity of the nutrient medium in the culture bath should be measured and adjust to the same value as of the fixing solution 10 The tissue should be stored at 4 C for at least 24 hours before slicing it into thin axial slices for H amp E staining 11 The fixed slices can be cut into thin axial perpendicular to its diameter either using the tissue chopper or the cryostat depending on required thinness of the axial tissue slice 12 The thin perpendicular slices show entire thickness of the tissue They are mounted on the gelatin slides and are left overnight to dry and adhere to the slide 13 Using the automatic mac
135. use of commercial two photon microscope to route the laser beam to our system on the same optical table Due to our preference to use longer wavelengths of laser from its tuning range an efficient excitation pathway of the microscope is required The power curve of the laser falls off at these wavelengths figure 4 4 To make use of all the power available at the extremes of the tuning range it is important to have good excitation efficiency To meet this requirement we use multilayer dielectric coated mirrors between the laser and the microscope Coherent Inc which can reflect gt 98 at 45 degrees and gt 99 7 at zero degrees for p polarized laser light ranging from 675 1000nm wavelengths 36 FW Verdi pump Output Power W goo zo Wavelength nm Figure 4 4 A typical Power curve of a Ti saph Mira900 laser across its entire tuning range when pumped with pump lasers Verdi of different powers Ti saph lasers lave peak power around 800 nm wavelength that falls rapidly towards the extreme ends of its tuning range Data courtesy Coherent Inc Scanning and detection unit The frame rate depends on the speed of the scanning mirrors Ideally in a 2D scanning mirror system the centers of both the mirrors should be imaged to the back focal point of the objective lens The deviation from this condition leads to vignetting To meet the optimal condition intermediate optics are required between the X and Y mirrors for
136. ving 2D single plane 3D z stack 2D and 3D time lapse imaging Further it allows controlling image intensity by changing laser power and detector PMT voltage gain This software also allows locating and focusing the specimen using X Y and Z controls Environment chamber For time lapse imaging of living cultures it is desired to keep the tissue healthy while imaging For this purpose a controlled environment chamber is constructed around the microscope to regulate physiologic levels of temperature and CO2 The supporting frame and body of the environment controlling unit was constructed using wooden bars mylar bubble plastic insulation and metal foil tape figure 4 9 To make sealable doors and windows Velcro is used A common chicken egg incubator is used as a heating 46 element that heats and maintains the chamber at physiologic temperatures 33 C 37 C by circulating the heated air Figure 4 9 Environment control chamber Left Older version of environment chamber around the custom made multiphoton microscope The frame work for this chamber was made with coat hangers There was limited heating due to large thermal loss by the conducting surface of the optical table The surface of the table was not covered with insulating bubble wrap Right Current version of the environment chamber The frame work is made up of wooden bars and the door is made of plexiglass This sturdy frame minimizes vibrations transferred from
137. w rates A generalized linear model ANOVA test followed by Tukey s multiple comparison test was used to evaluate statistical significance indicates statistical significance p lt 0 01 compared to unperfused and membrane controls 81 DISCUSSION AND CONCLUSIONS Here I presented a successful technique to culture viable 700um thick organotypic cortical slices using a novel through the thickness perfusion paradigm To use this method successfully it is required to anchor the tissue to the gold grid substrate and the inner walls of the infusion chamber to block all the paths of low resistance for fluid flow In the absence of any path of low resistance of flow the infused medium is required to flow through extracellular space of the tissue throughout its thickness before entering the withdrawal chamber The adhesion of the tissue is achieved by coating the infusion chamber with laminin and incubated for 30 40 minutes Additionally the brain slices are cut into 3 mm circular discs using a sterile biopsy tool before transferring to the infusion chamber The same diameter of the infusion chamber and the brain slice allow perfect accommodation of tissue in the chamber that facilitates adhesion of the tissue to the walls of the infusion chamber via laminin coating ensuring that no paths of low resistance exist for the flow of infused medium Further the FEP membrane containing teflon lid of chamber also assists in confining the tissue in the
138. wo solutions i to use minimum number of optical elements and ii use a pre chirping unit figure 4 2 The chirping could be of two types positive and negative If all the components of the microscope are chosen such that they chirp in the same direction a pre chirp unit made up of two prisms having the opposite chirping direction can be used to compensate for this non uniform stretch in the pulses at the focal plane of the objective lens 54 113 33 Chirped input pulse Compressed output pulse gt gt Figure 4 2 Pulse compressor or pre chirp unit The velocity of different wavelengths is refractive index dependent other than air refractive index n 1 for air 1 3 1 5 for glass The velocity of all wavelengths is same in the air but is wavelength dependent in media having other refractive index The wavelengths with faster velocity in glass higher refractive index compared to slower moving wavelengths travel larger area of glass prism while the slower wavelengths travel smaller glass areas tip of the prism thus optical path length that results in compressing of the chirped pulse Picture courtesy Prof Rick Trebino Apart from chirping other challenges to construct a good multiphoton microscope are to obtain optically efficient excitation and emission pathways to meet the requirement of the excitation of fluorophore with minimal laser power the ability to cover the entire wavelength range of the laser with the s
139. y of input output variables while using multiple techniques simultaneously Development of a three dimensional neuro robotic hybrid model 3D HYBROT with targeted intact cellular circuitry of thick brain slices for stimulation and recording will allow understanding of neuronal dynamics in the brain at the network level underlying learning and memory With the current culturing methods one can successfully culture organotypic brain slices 36 114 However the thick brain slice cultures suffer necrosis in the center of the tissue due to insufficient supply of nutrients resulting from these diffusion based methods 114 A convection based perfusion method that allows flow of oxygenated nutrient medium through the thickness of the tissue will ensure nutrient supply to every cell This may allow enhanced viability resulting from the forced convection perfusion based restoration of the circulatory system of the tissue To test this hypothesis an infusion withdrawal type micro perfusion chamber that can be modified to incorporate three dimensional electrodes was fabricated optimized and used to culture 700um thick brain slices The perfusion of oxygenated nutrient medium successfully demonstrated enhanced viability of thick brain slice cultures I further investigated viability of the cultures as a function of flow rates to determine an optimal range of the non invasive perfusion rates Additionally I investigated qualitatively morphological and ele

thick brain slice cultures and a custom-fabricated

Contents

Download Pdf Manuals

Related Search

Related Contents