By Kristen M. Harris (June 26, 2024)

Sample Readings about sampling biases from Harris lab

• Harris KM, Jensen FE, Tsao B (1992) Three-dimensional structure of dendritic spines and synapses in rat hippocampus (CA1) at postnatal day 15 and adult ages: Implications for the maturation of synaptic physiology and long-term potentiation. J. Neurosci. 12:2685-2705. PMCID: PMC6575840. (6.8MB PDF)

• Effects of field composition

• Fiala JC, Harris KM (2001) Extending unbiased stereology of brain ultrastructure to three-dimensional volumes. Journal of the American Medical Informatics Association. 8(1):1-16. PMCID: PMC134588. (1.4MB PDF)

• Fiala JC, Harris KM (2001) Cylindrical diameters method for calibrating section thickness in serial electron microscopy. J Microscopy 202(3):468-472. (0.2MB PDF)

• Fiala JC, Kirov SA, Feinberg MD, Petrak LJ, Goddard CA, George P and Harris KM (2003) Timing of Neuronal and Glial Ultrastructure Disruption During Brain Slice Preparation and Recovery During Incubation In Vitro. J Comp Neurol. 465(1):90-103.  (1MB PDF) –

  • Has some interesting MT measures – and how we calculated CA1 MT length.

• Harris KM, Hubbard DD, Kuwajima M, Abraham WC, Bourne JN, Bowden JB, Haessly A, Mendenhall JM, Parker PH, Shi B, Spacek J (2022) Dendritic Spine Density Scales with Microtubule Number in Rat Hippocampal Dendrites. Neuroscience, 489: 84-97. PMCID: PMC9038701. (PDF)

Choosing Sample Regions

1. Choose central section by following criterion:

   • A block of 10 serial sections near the central section that have NO flaws.

2. Prepare a rectangle grid with 5 µm by 5 µm squares using PyReconstruct grid tool. (Light blue in the adjacent picture)

   • Place the grid in the center of the central section of the series

3. Identify four 5 µm by 5 µm squares for picking small dendrites to reconstruct.

   • Move these 4 to avoid large dendrites from spanning the grids.

   • Yellow squares in adjacent tSEM image.

Open

Stamp candidate dendrites with unique D### - capital D for ID stamps

1. View/stamp each candidate dendrite D### through adjacent 5-10 sections to confirm/deny dendrite category

2. To automagically increment each stamp, you can use D<001>

3. Note highly visible color, transparent fill, so it can be seen when zoomed out.

4. Fill when unselected so easy to see.

Open

Stamping dendrites

1. Place D### stamp on the central section.

2. Copy it.

3. Move to next section and paste it with

4. ctrl v – move stamp with arrows to keep it centered (approximately).

If not a dendrite, make a red X

1. Indicate in the exclude column what the item is

   • axon, glia, other

   • Rename to that object name.

2. Note I named rectSAMP1a for the upper corner.

3. The lower corner only has a couple of dendrites.

4. Identify all dendrites in all 4 squares.

Rationale for choosing sample locations

• Local circuits versus broad field selection.

• Rationales for other ways to choose

Assess length criterion

• Rapidly stamp entire dendrite to estimate dendrite lengths

  • Don’t use the old strategy of multiple dendrites at a time – instead do one dendrite at a time.

  • Make the stamp to encircle the dendrite on each section – or every 5th section

    • You can use the new PyRecon stamp tool to draw the stamp diameter to match that of the dendrite, super quickly in each section.

  • Estimate the z length through the dendrite stamps.

  • If less than criterion length (suggest 10 µm) for your datasets, exclude.

• Note you can copy/paste the stamp and wait to change it’s size until the underlying dendrite changes.

• Then copy/paste the new stamp going forward.

Determine the semi-Quantitative Drift for specific objects

For cluster analysis, protrusion identification and annotation

1. Blend first and last section of the stamped dendrite.

2. Measure center to center distance of the dendrite.

3. Divide that length by the number of sections spanned.

4. Set a Criterion for this value (distance/#sections)

   • Suggest less than 50 nm drift per section as the maximum drift criterion.

5. If greater than 50 nm / section the dendrite is excluded

6. This criterion is dendrite-specific and could be as great as you are willing to accept for your specific work

   • e. g. interneuron dendrites.

Perform Microtubule counts on the dendrites and determine criteria for inclusion

1. Adjust section contrast to optimize quality of the MT

   • sort of a darkish gray central core with a bit of fuzz around it

2. Find three locations where the microtubules (MT) are well stained and appear cross-sectioned or slightly obliquely sectioned in the dendrite. These should be near the beginning, middle, and end of the dendrite segment.

3. Stamp each MT on each of 2 – 3 adjacent sections.

   • Feel free to view/stamp additional sections to gain confidence that the object is a MT and not another small structure in cross section.

4. Once the stamps are done on 2-3 sections – rename the mt to d###MT_Section# --

   • then the counts will be unique on each section, and you can see whether they match.

   • If they do not, you should recheck to be sure none were missed, or over stamped.

   • It is also possible for a MT to start/end on a single section, so the count might be actually different

5. See readings above.

If not enough dendrites, or too many

• Not enough then pick another 5 µm square from the whole rectangle and evaluate every dendrite in it as above.

• If more than enough dendrites meet criteria then add another uniform criterion for exclusion --

  • Exclude dendrites with the largest drift value(s) even if they met other criteria.

Adding large dendrites

1. Start with those intersecting your 4 rectangles.

2. Then add from adjacent rectangles until you reach your sample size that meets criterion.