Neurons need resolution
Neuroscience, one of the most fascinating topics in biology, still harbors the most complex secrets within the science of life, including the mind-body problem. For a compact overview on the history of this matter, see [1, 2]. Speculations about the functional significance of the brain go back to ancient Egypt, although at that time the brain was not assumed to play a vital role in humans. About 200 A.D., Galen formulated the "ventricle doctrine" (Ventrikellehre), which suspected the nerves of being hollow tubes, through which sensations and stimuli were transported (the ventricles acting as control authority). Soon, these ideas led to the assumption that the ventricles host the immortal soul. Indeed, the concepts were "accepted" (i.e. dogmatically established by the clergy) until the 18th century.
The operation of a huge and complicated channel system requires the hollow nerves to be joined together and allow material transport amongst the various structures already described. With the invention and improvement of microscopes, it was shown by works of Matthias Schleiden and Theodor Schwann in 1839 that all living material (including nerves) consists of cells . The title of the publication already demonstrates the important role that the microscope had assumed in biology. Nevertheless, there was no clear proof of whether or not these nerve cells stay separate individuals or connect with each other, thus creating a huge syncytium spanning the whole organism. The decision on whether nerves are permeable for a liquid "spiritus animalis", or whether that concept was wrong after all was finally reached by Santiago Ramón y Cajal  between 1890 and 1900. He was able to prove that nerve structures (later called neurons), are isolated by a distinct gap. This breakthrough was possible after the development of appropriate specimen preparation (fixation by ethanol and chromic acid) and staining techniques (Golgi silver staining).
These techniques allowed the preparation of thin and solid (i.e. manageable) brain slices with good contrast, high specificity and superior resolution. However, the main breakthrough was the employment of high-resolving microscopes that were able to image the tiny details in these preparations. The key that opened the door to the new world was the oil immersion technique, invented in 1878 by Ernst Karl Abbe, which raised (ordinary) microscopic resolution significantly from about 1/3 µm to 1/5 µm – where it was more or less halted until 1994, when Stefan Hell invented the STED microscope.
The third dimension – confocal and multiphoton microscopy
Widefield microscopic images, also when using high-resolution immersion techniques, contain information both from the focal plane and from all layers above and below that plane. If the sample is very thin, these contributions are negligible. In thick samples, especially with dense fluorescent staining of complex structures, the relevant information may be covered by out-of-focus haze. This problem led Marvin Minsky to the invention and design of the first confocal microscope in an attempt to solve the problems he had when examining thick Golgi-stained brain slices . How a confocal microscope is capable of generating optical sections is described in the "confocal microscopy – optical path" tutorial .
Most cells have comparable sizes in the spatial dimensions. Neurons are different. The main body – the soma – is shaped similar to many other cells. But the complex branching of dendrites and the axon plus its branches makes the nerve cell an extraordinarily three-dimensional object. The thin dendritic or axonal extensions with diameters in the range of less than a micrometer span lengths ranging from fractions of a millimeter to even meters. In the brain itself, billions of such cells are each connected by some thousand synapses. At a rough estimate, each human being has a 100 billion neurons , which would amount to some 1021 for all human neurons in the current world population of 7 billion, the "neurome of mankind". With the Internet as a connecting network and the display-retina system as artificial synapse, these indeed form an unimaginable computing system, including your and my brain – whether you like it or not. Although the brain is a highly ordered and organized system, the plain anatomical view is that of a huge bowl of spaghetti. To sort out the trajectory of single neurons, staining techniques were invented and applied. With specialized dyes, mostly lipid-soluble fluorescent markers, it is possible to stain a single neuron on its entire way through the spaghetti bowl. The rest is microscopy.
Confocal imaging gives high contrast 3D data, but for very deep imaging, a different technique is employed: multiphoton excitation . This is also an optical sectioning scanning technique, but is much less prone to optical deteriorations in deeper layers that occur due to scattering. Obviously, the field of imaging and research with fluorescent staining is expanding incredibly with fluorescent proteins and biosensors. Just to name brainbow  techniques, where a set of dyes are expressed in varying intensities, allowing not only a single, but a few hundred different neurons to be distinguished just by their different tones. Newer techniques like STED or localization microscopy allow higher definition in optical microscopy, exceeding the 1/3 µm to theoretically unlimited resolution. Other techniques, like CARS (Coherent Antistokes Raman Spectroscopy) and derivatives offer possibilities of imaging nervous tissue without specific labeling.
Tools and functional imaging
Besides structural analysis, microscopy is also important in other areas of neuroscience. To begin with, it is an indispensible tool in electrophysiology. Here, it is used just to observe the micropipette when attempting to prick the pipette tip into a neuron. It is even more important when doing patch-clamp experiments, where differential interference contrast (Nomarski ) is a standard technique. But not only as an aid for electrophysiological experiments: phase contrast and differential interference contrast have become indispensable for research on life cell preparations, including neurons. One result of these experiments is the finding of anterograde and retrograde vesicle transport in axons of vesicles to and from the soma. Although material transport has now been proven to occur in the hollow tubes (axons and dendrites), this does not reinstate Galen’s ventricle doctrine.
The transport is not related to the conduction of nervous stimuli but rather concerned with supplying the relevant structures for signal transduction.
From these early functional investigations, it was not a long way to employ fluorescence microscopy for neuroscience. On the one hand, structural information is retrieved by applying fluorescently labeled antibodies to neuron-specific proteins. Fluorescent proteins then enabled scientists to study the genetic regulation of proteins during development and differentiation. An application is the recording of calcium concentration changes or changes in the electrical potential across neuronal membranes by means of fluorescent indicators. These indicators were first designed as direct sensors. Meanwhile, many FRET-based biosensors for many different metabolites, ions and cell parameters are available and in use; including endogenically expressed sensors  as derivatives of fluorescent proteins.
These techniques allow the brain to be monitored at work without injecting chemicals and dyes or importing them by other invasive means. In-vivo brain investigations became possible by mounting a real glass window onto the opened skull for high-resolution deep imaging. This is shown in the image on top of this page, where a head of a mouse is positioned by modeling clay in a Multiphoton-microscope. In vivo imaging in the intact mouse brain using a dedicated objective lens. The objective is a 20x water dipping lens with 1.0 NA for highest resolution at a large observation area.
- Wikipedia: Neuroscience, Oct. 30th (2011).
- Wikipedia: Geschichte der Hirnforschung, Oct. 30th (2011).
- Schwann T: Mikroskopische Untersuchungen über die Übereinstimmung in der Struktur und dem Wachstum der Thiere und Pflanzen. Sander’schr Buchhandlung Berlin (1839).
- Ramon y Cajal SF: Revista Trimestral de Histología Normal y Patológica (May 1888).
- Masters BR: Confocal Microscopy and Multiphoton excitation microscopy. P. 85, SPIE press (2006).
- Borlinghaus RT: Confocal Microscopy – Optical Path. Leica Science Lab (2011).
- Diaspro A: Confocal and Two-Photon Microscopy; Foundations, Applications and Advances. Wiley & Sons (2002).
- Lichtman JW, Livet J and Sanes JR: A technicolour approach to the connectome, Nature Reviews Neuroscience 9:6 (2008) 417–422; doi:10.1038/nrn2391.
- Gamse JT et al.: The parapineal mediates left-right asymmetry in the zebrafish diencephalon. Development 130 (2003) 1059–68 (2003).
- Nomarski G: Microinterféromètre différentiel à ondes polarisées. J Phys Radium 16 (1955) 9–11 (Paris).
- Steinmetz I: FLIM-FRET in solutions: In vitro measurement of a Cerulean-Citrine biosensor. Leica Science Lab (2011).