Figure S2.
Microtubule ends in the outer segment and microtubule density in the inner segment. (A) Cartoon schematics showing that in cross-sections, microtubules spanning across adjacent sections (along the z axis) can be aligned, and the length can be estimated, because only a short fragment of microtubule might be lost. The red box indicates the region that can be lost during the sample preparations and data processing. We estimated that the loss was ∼10–20% in our tomography data. (B) Cartoon schematics showing that in lateral sections, microtubules spanning across adjacent sections cannot be well aligned due to the loss of a relatively long fragment in the middle. Therefore, the length of microtubules cannot be well estimated. However, the location of microtubule ends can be determined (black arrowheads). (C) A representative cross-sectional view image from the ET volume data of the TB in a haltere campaniform mechanoreceptor. (D and E) Two sets of consecutive images (along the z axis) showing microtubule ends appearing within the tissue tomogram (black arrowheads). The adjacent images had a spatial interval of ∼4 nm along the z axis (from left to right). (F) A representative lateral view image from the ET volume data of the outer segment in a leg campaniform mechanoreceptor. (G and H) Two sets of consecutive images showing microtubule ends appearing within the tissue tomogram (black arrowheads). The adjacent images had a spatial interval of ∼6 nm. In C–H, scale bars, 100 nm. (I) A representative image from the FIB/scanning electron microscopy volume data of a haltere campaniform mechanoreceptor. Red arrowheads indicate the microtubules in the inner segment. Scale bar, 1 µm. (J) The segmented model of plasma membrane and microtubules of the inner segment. Two centrioles (MC and DC) were segmented as location markers. White arrows indicate the positions of three cross-sectional images used to quantify the density of microtubules in this cell. One of the three sections and the corresponding segmentation are shown in K–M as an example. (K) A representative image showing the cross-sectional view (FIB/scanning electron microscopy slice image) of dendritic inner segment. Scale bar, 1 µm. (L) Vesicles and other cellular structures (outlined using blue lines) were manually segmented and excluded in measuring the cross-sectional area of cytoplasmic space. Based on this measurement, we estimated that free cytoplasmic area takes up 51.3 ± 0.04% (n = 3 sections) of the total cytoplasmic area. (M) The segmented model of microtubules in the inner segment. Each dot represented one microtubule, and the number of microtubules can be counted. We then estimated the ratio between polymeric and soluble dimeric tubulin in the inner and outer segments as follows. Assuming the density of microtubules is d μm−2, for a 1-µm-long cylinder with a diameter of 1 µm, it contains 0.785d microtubules. Because a 1-µm-long microtubule contains 1,625 tubulin dimers (assuming 13 protofilaments), the total number of tubulin dimers in microtubules is 1,276 d. In the meantime, given that the cytoplasmic concentration of free tubulin dimers is around 10 µM (Howard, 2001), the number of soluble tubulin dimers in this volume is 4,728. Based on our ET data, we counted that the density of microtubules in TB is 156 ± 29 µm−2 (n = 5 receptors). If the above cylinder represents a part of TB, it would contain 199,056 polymeric tubulin and 4,728 soluble tubulin dimers. Therefore, polymeric and soluble tubulin dimers take up ∼98% and ∼2% of the total tubulin signal in the outer segment, respectively. Based on the data in K–M, we counted that the density of microtubules in inner segment is 25 ± 11 µm−2 (n = 3 receptors). This predicts that in the inner segment, there are 31,900 polymeric and 4,728 soluble tubulin dimers. Therefore, polymeric and soluble tubulin dimers take up ∼87% and ∼13% of the total tubulin signal in the inner segment, respectively. DC, daughter centriole; MC, mother centriole.

Microtubule ends in the outer segment and microtubule density in the inner segment. (A) Cartoon schematics showing that in cross-sections, microtubules spanning across adjacent sections (along the z axis) can be aligned, and the length can be estimated, because only a short fragment of microtubule might be lost. The red box indicates the region that can be lost during the sample preparations and data processing. We estimated that the loss was ∼10–20% in our tomography data. (B) Cartoon schematics showing that in lateral sections, microtubules spanning across adjacent sections cannot be well aligned due to the loss of a relatively long fragment in the middle. Therefore, the length of microtubules cannot be well estimated. However, the location of microtubule ends can be determined (black arrowheads). (C) A representative cross-sectional view image from the ET volume data of the TB in a haltere campaniform mechanoreceptor. (D and E) Two sets of consecutive images (along the z axis) showing microtubule ends appearing within the tissue tomogram (black arrowheads). The adjacent images had a spatial interval of ∼4 nm along the z axis (from left to right). (F) A representative lateral view image from the ET volume data of the outer segment in a leg campaniform mechanoreceptor. (G and H) Two sets of consecutive images showing microtubule ends appearing within the tissue tomogram (black arrowheads). The adjacent images had a spatial interval of ∼6 nm. In C–H, scale bars, 100 nm. (I) A representative image from the FIB/scanning electron microscopy volume data of a haltere campaniform mechanoreceptor. Red arrowheads indicate the microtubules in the inner segment. Scale bar, 1 µm. (J) The segmented model of plasma membrane and microtubules of the inner segment. Two centrioles (MC and DC) were segmented as location markers. White arrows indicate the positions of three cross-sectional images used to quantify the density of microtubules in this cell. One of the three sections and the corresponding segmentation are shown in K–M as an example. (K) A representative image showing the cross-sectional view (FIB/scanning electron microscopy slice image) of dendritic inner segment. Scale bar, 1 µm. (L) Vesicles and other cellular structures (outlined using blue lines) were manually segmented and excluded in measuring the cross-sectional area of cytoplasmic space. Based on this measurement, we estimated that free cytoplasmic area takes up 51.3 ± 0.04% (n = 3 sections) of the total cytoplasmic area. (M) The segmented model of microtubules in the inner segment. Each dot represented one microtubule, and the number of microtubules can be counted. We then estimated the ratio between polymeric and soluble dimeric tubulin in the inner and outer segments as follows. Assuming the density of microtubules is d μm−2, for a 1-µm-long cylinder with a diameter of 1 µm, it contains 0.785d microtubules. Because a 1-µm-long microtubule contains 1,625 tubulin dimers (assuming 13 protofilaments), the total number of tubulin dimers in microtubules is 1,276 d. In the meantime, given that the cytoplasmic concentration of free tubulin dimers is around 10 µM (Howard, 2001), the number of soluble tubulin dimers in this volume is 4,728. Based on our ET data, we counted that the density of microtubules in TB is 156 ± 29 µm−2 (n = 5 receptors). If the above cylinder represents a part of TB, it would contain 199,056 polymeric tubulin and 4,728 soluble tubulin dimers. Therefore, polymeric and soluble tubulin dimers take up ∼98% and ∼2% of the total tubulin signal in the outer segment, respectively. Based on the data in K–M, we counted that the density of microtubules in inner segment is 25 ± 11 µm−2 (n = 3 receptors). This predicts that in the inner segment, there are 31,900 polymeric and 4,728 soluble tubulin dimers. Therefore, polymeric and soluble tubulin dimers take up ∼87% and ∼13% of the total tubulin signal in the inner segment, respectively. DC, daughter centriole; MC, mother centriole.

This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal