I'll preface this post by indicating that I work primarily in bioinformatics, but I need to demonstrate some wet-lab skills in order to graduate with a degree in neuroscience.
I'm proposing a study that seeks to count various immunohistochemically-labeled cell counts and to quantitatively assess dendritic morphology/spine features from serial sections of the **same individuals**. I'm insistent on using the same individuals, because the opposite hemisphere is being fresh-frozen and used for RNAseq studies, so we want to correlate these changes with morphological/neuroanatomical features. The tissue in question will be saline perfused, immersion fixed in 4% PFA. I had planned to section the tissue at roughly 40-50 microns; hopefully just large enough to permit the study of a few intact dendritic arbors per slice (and also thin enough to permit antibody penetration for IHC).
Ideally, some variation of Golgi staining seems would have been great for dendritic arbors (e.g. Sholl analysis) and spine quantification. However, I've read that these approaches are generally poorly suited for pre-sectioned slices at my planned thickeness (e.g. they tend over-stain and become very rigid).
So I'm wondering if:
1) Anyone is aware of a modified Golgi protocol for thin sections (e.g. 40 microns; they can be free floating or mounted as needed to accomodate).
2) Anyone can suggest an alternative approach to Golgi-type staining that would facilitate the quantitative morphological study of the dendritic arbor and dendritic spine density/features.
Hi Daniel,
I did extensive work on reconstructing the dendritic morphology of layer V pyramidal neurons in mice for my masters work. I performed sparse single cell iontophoresis in 200um sections to label neurons with fluorescent dye. This enabled faster and more accurate reconstruction. I subsequently performed IHC to determine the genotype of the cells that were filled with dye.
I agree that depending on the cell type you want to reconstruct, 40um will likely not be sufficient.
I now work in biotech but I wrote up a blog about the free programs I used to trace and extract complex morphological parameters. I highly suggest you checkout those programs. They were a lifesaver for a lab with no budget for fancy software. Here's the link:
http://www.stressmarq.com/Blog/Stress-Reducers/Top-5-Free-Image-Analysis-Software-Tools-for-Micro.aspx
The protocol for the single cell electroporation can be found in my paper just published. http://journal.frontiersin.org/article/10.3389/fncel.2015.00145/abstract
Let me know if you have any questions!
Best of luck!
You may want to look up immunostaining for dendrite-specific or -selective proteins like CRMP or MAP2.
Any presectioning will cause significant underestimation of neuron morphologies, especially when you are planning to assess the morphologies within 40micron, its not even come closer to a complete neuron. Even with slice preparations with thickness of 300µm, you will not get entire structure.
In vivo neuron labelling with biocytin and ABC staining is a good alternative. Read recent papers from our lab..
(1) Beyond Columnar Organization: Cell Type- and Target Layer-Specific Principles of Horizontal Axon Projection Patterns in Rat Vibrissal Cortex.
(2) Juxtasomal biocytin labeling to study the structure-function relationship of individual cortical neurons.
(3) Cell type-specific three-dimensional structure of thalamocortical circuits in a column of rat vibrissal cortex.
If you go through these papers you will get an idea how people used to underestimate neuron features.
Golgi staining is a good method, however, this method only stain a small population of neurons so the sampling is biased. And additinally, this approach make significant tissue shrinkage.
If you have any questions on protocols, we are happy to offer our help.
Rajeev
Hi Daniel,
I did extensive work on reconstructing the dendritic morphology of layer V pyramidal neurons in mice for my masters work. I performed sparse single cell iontophoresis in 200um sections to label neurons with fluorescent dye. This enabled faster and more accurate reconstruction. I subsequently performed IHC to determine the genotype of the cells that were filled with dye.
I agree that depending on the cell type you want to reconstruct, 40um will likely not be sufficient.
I now work in biotech but I wrote up a blog about the free programs I used to trace and extract complex morphological parameters. I highly suggest you checkout those programs. They were a lifesaver for a lab with no budget for fancy software. Here's the link:
http://www.stressmarq.com/Blog/Stress-Reducers/Top-5-Free-Image-Analysis-Software-Tools-for-Micro.aspx
The protocol for the single cell electroporation can be found in my paper just published. http://journal.frontiersin.org/article/10.3389/fncel.2015.00145/abstract
Let me know if you have any questions!
Best of luck!
Hi Daniel,
I second Leslie's suggestion to label with fluorescent hydrazine if you have the equipment (or can borrow a stimulator and pipette puller). Many neurobiology labs will have a stimulator hanging around, and it is an easy protocol to implement in fixed tissue. The hydrazines come in a variety of colors, so you could even label different cell types/layers with different Alexa Fluor hydrazines.
And Leslie, thanks for the link to your blog! I have been looking for an automated tracing software. XP is no problem--we have many computers that run on that platform that we use for our image analysis.
Best,
Jill
Our lab typically utilizes diolistic labeling of 1.5% perfused tissue, but it's typically wise to slice at 200+um to allow for full reconstruction (as everyone else has stated). Diolistic labeling is extremely cheap, but requires a confocal microscope to image the fluorescent lipophiloic dye, DiI (excitation spectra around that of TRITC, about 532 nm). This allows for 3D reconstruction, which Golgi staining doesn't.
- Badges
- Science topic
- Similar topics
- Biological Science
- Neuroscience
- Neuroanatomy