Figure 1.
Geometric characteristics of dendritic spines allow for objective identification.
(A) Common dendritic spines types found in the cortex. Spine maturity progresses (from left to right) from long, thin filopodia type structures (red) to wide-headed mushroom spines (blue) and the occasional branched spine (purple). Geometric characteristics of spines, listed below each type, are incorporated into the rapid spine analysis method. (B) Golgi-cox stained secondary dendritic branch of a Layer II/III pyramidal neuron in mouse primary visual cortex. Different spine types are indicated by arrowheads, color-coded to match A. Scale bar, 5 µm.
Figure 2.
Flowchart for the rapid Golgi spine analysis method.
Sample preparation and data collection (purple) are performed according to established protocols and manufacturer’s instructions. The rapid Golgi analysis method is split into two main parts. The first part, Image Analysis (teal), involves the importation of images into RECONSTRUCT and the measurement of dendrites and spines. These steps are repeated for each dendrite that is analyzed in a single Z-stack. The second part, Data Analysis (blue), uses custom formulas in Microsoft Excel to categorize spines and quantify numerous dendrite/spine variables. Figures corresponding to various sub-steps are listed on the right hand side of the chart.
Figure 3.
‘Import images’ dialog box in RECONSTRUCT.
In this example, images from a Z-stack titled “3355_3_07 Cell 15″ are being loaded in sequential order. Pixel size (i.e. µm per pixel), once calibrated by the user, can be entered here or at any time after image importation.
Figure 4.
Dendrite identification and length measurements.
(A) Main window in RECONSTRUCT. The tool panel can be seen in the upper right corner. The red rectangle indicates the dendritic segment chosen for analysis in this example. (B) Zoomed in image of the chosen dendritic segment. The ‘Draw Line’ tool has been selected to create the straight length measurement (∼10 µm) for this segment, with ‘start’ and ‘stop’ positions indicated. (C) The ‘Draw Z-Trace’ tool must be used to measure the Z-length of the dendritic segment. The ‘start’ and ‘stop’ positions created in B are used to guide the Z-trace.
Figure 5.
(A) Due to the nonlinear nature of dendrites and spines, single optical sections will have a mixture of in focus (blue) and out-of-focus (red) spines. Spine width measurements, made by drawing a straight line across the widest part of the spine head, should only be drawn on sections where the spine is in focus. (B) Once all spines on the segment of interest have been found and measured, cutting and pasting all traces onto the same section as the reference line (orange) enables the creation of Export List A from the Trace List. Note that spines were marked in order going counter-clockwise around the reference line. (C) Drawing accurate Z-length measurements for spines often requires scrolling up and down through the Z-stack. In this example, the Z-trace starts at the base of the spine on Section a, continues down the neck of the spine through Section b, and terminates at the tip of the spine on Section c. (D) The Z-Trace List (used to create Export List B) yields the Z-length measurements for all analyzed dendrites and spines within the series. Z-traces can be visualized in the 3D Scene window by double-clicking on the name of the trace. Z-length measurements for this dendritic segment followed the same order for spines established in B. (E) Visualization of all Z-traces (red and orange) and straight line width traces (blue) for this segment.
Figure 6.
Categorizing spines and characterizing dendrites in Excel.
The provided spreadsheet template (Spreadsheet S1) contains all of the formulas required to utilize the measurements obtained from RECONSTRUCT. Identifying information (red) allows the user to specify each analyzed dendrite according to their own conventions. The ‘DEN ID’ column must be unique to each dendrite in a data set for the proper working of the included formulas. Values imported from RECONSTRUCT (blue) are obtained from Export Lists A & B. “By Spine” formulas (gold), including ‘LWR’, or length-to-width ratio, and ‘Type’, which classifies spines according to a custom hierarchical formula, should be dragged down and repeated for each row (i.e. spine) of the data set. “By Dendrite” formulas (green), which measure average protrusion width, length, LWR, and protrusion density, should be copied and pasted only onto the first line of each new DEN ID value.
Figure 7.
The rapid Golgi spine analysis method accurately reports spine proliferation and maturation.
(A) Diagram of the right hemisphere of the mouse brain, sectioned coronally, showing the location of V1. Secondary and tertiary dendrites of Layer II/III pyramidal neurons were analyzed via the rapid spine analysis method. (B) Representative images of Golgi-Cox stained dendritic spines at P14 and P25. Spines at P25 appear shorter and more abundant than their P14 counterparts. Scale bar, 2 µm. (C) Protrusion density significantly increases between P14 and P25 (P<0.05) while (D) average LWR decreases, reflecting shorter, wider spines (P<0.05). (E) The percentage of immature filopodia-type spines sharply decreases between P14 and P25 (P<0.05), offset by (F) an increase in the proportion of mature mushroom spines (P<0.05).