Basal epithelial tissue folding is mediated by differential regulation of microtubules

ABSTRACT The folding of epithelial tissues is crucial for development of three-dimensional structure and function. Understanding this process can assist in determining the etiology of developmental disease and engineering of tissues for the future of regenerative medicine. Folding of epithelial tissues towards the apical surface has long been studied, but the molecular mechanisms that mediate epithelial folding towards the basal surface are just emerging. Here, we utilize zebrafish neuroepithelium to identify mechanisms that mediate basal tissue folding to form the highly conserved embryonic midbrain-hindbrain boundary. Live imaging revealed Wnt5b as a mediator of anisotropic epithelial cell shape, both apically and basally. In addition, we uncovered a Wnt5b-mediated mechanism for specific regulation of basal anisotropic cell shape that is microtubule dependent and likely to involve JNK signaling. We propose a model in which a single morphogen can differentially regulate apical versus basal cell shape during tissue morphogenesis.

wnt5b MO (C) injected embryos. Embryos were fixed at 24 ss in 4% paraformaldehyde in phospho-buffered saline with 0.1% Tween-20 (PBT) for 2 hours and immunostained for phospho-histone 3 (1:800, Millipore, #06-570) and counter-stained with propidium iodide. (D) Quantification of PH3 positive cells as a percentage of total cells in the region indicated by the white box. Data represented as mean ± SEM from three independent experiments. Control MO,n=3;wnt5b MO,n=3. (E,F)   First, midbrain and hindbrain intensity is averaged. MHB intensity is then divided by the averaged midbrain-hindbrain intensity. Final comparisons are made between Control and wnt5b MO injected embryos. (D,E) Representative images of a 10 μm Z-series average intensity projection of -tubulin immunostaining. Boxed areas indicate where -tubulin average intensity was quantified for apical, middle, and basal regions of the MHBC within a single embryo. The average intensity at either the apical MHBC region or the basal MHBC region is divided by the average intensity at the middle MHBC region in the same embryo to acquire apical or basal MHBC -tubulin intensity, respectively. (F) Formula for normalizing MHBC apical or basal average intensity. This normalization was used for comparison of basal MHBC intensity across embryos and for comparisons between Control MO and wnt5b MO injected embryos. (G) Quantification of the normalized Apical -tubulin intensity at the MHBC in Control versus wnt5b morphants. Apical MHBC intensity was divided by the intensity in the middle of the cell. Box plots indicate the 25 th and 75 th percentiles and the median. Three independent experiments are represented. Control MO, n=7; wnt5b MO, n=6. Scale bars: 10 μm. Boxed areas indicate where average intensity of either MRLC-GFP or phalloidin was quantified for apical, middle, and basal regions of the MHBC cells. (C) Formula is shown for normalizing apical or basal MHBC average intensity. Briefly, either apical or basal MHBC average intensity is divided by the average intensity of the middle MHBC region within the same embryo, for each embryo. This normalization was used for comparison of apical or basal MHBC intensity across embryos and experiments. (D,E) Representative confocal images of 10 μm average intensity projections at 24 ss of wild-type embryos co-injected with 150 pg/embryo MRLC-GFP, 50 pg/embryo memCherry, wnt5b morphants demonstrate no significant difference apically or basally. Box plots indicate the 25 th and 75 th percentiles and the median. Three independent experiments are represented. Control MO, n=9; wnt5b MO, n=8. Scale bars: 10 μm. (G,H) Representative confocal images of 10 μm average intensity projections at 24 ss of phalloidin stained wild-type embryos that were co-injected with 3 pg/embryo p53 MO and either 3 pg/embryo of Control MO (G) or 3 pg/embryo wnt5b MO (H). For phalloidin staining, embryos were fixed at 24 ss in 4% paraformaldehyde in phosphate buffered saline with 0.1% Tween (PBT) for two hours, washed with PBT three times for 10 minutes each, deyolked, and incubated in phalloidin (1:40, A12379, Invitrogen) in PBT overnight at 4 degrees C. Next, embryos were washed three times for ten minutes each in PBT, flat-mounted in glycerol, and imaged using confocal microscopy.    S7. wnt5b knockdown affects basal but not apical localization of -catenin. In order to examine the effects of wnt5b signaling on subcellular variations of the pathway, we examined the localization of -catenin at the MHBC using immunohistochemistry. Embryos were fixed in Dents (80% Methanol, 20% DMSO) for 1 hour at room temperature and washed in PBT three times for 30 minutes each. Embryos were deyolked, washed three times for 10 minutes in PBT, and incubated in block solution (1% Boehringer Mannheim Blocking Reagent, 10% lamb serum, 80% Maleic Acid Buffer) overnight at 4 degrees C, followed by incubation with 2.5 g/ml catenin primary antibody (ab6301, Abcam) in PBT overnight at 4 degrees C. The next day, embryos were washed four times for 1.5 hrs each in PBT and incubated in secondary antibody overnight at 4 degrees C, 1:2500 goat anti-mouse AlexaFluor-555 (A21422, Invitrogen) in PBT. Next, embryos were washed in PBT three times for 30 minutes each, flat mounted in glycerol, and imaged using confocal microscopy. (A,B) Average intensity projections of a 10 μm Z-series of 24 ss embryos immunostained for -catenin. Boxed areas indicate where average intensity was quantified for apical, middle, and basal regions of the MHBC and the outside basal region posterior to the MHBC. (C) Formula for normalizing MHBC apical and basal average intensity. Apical or basal MHBC average intensity is divided by the average intensity of the middle MHBC region within the same embryo. MHBC:Outside ratios are acquired by dividing the basal MHBC intensity by the average intensity directly adjacent and posterior to the MHBC region. Control and wnt5b MO injected embryos were compared for each normalized intensity. (D,E) Average intensity projection of 10 μm Z-series of 24 ss wild-type embryos co-injected with 3 pg/embryo p53 MO and either 3 pg/embryo Control MO (D) or 3 pg/embryo wnt5b MO (E) and immunostained for -catenin.   Figure 3. Representative live confocal timelapse at the MHB of an embryo co-injected with memCherry, EB3-GFP mRNA, and Control MO. Timelapse data were acquired at 21-24 ss for 10 minutes at 1 frame per 4 seconds and cropped to 100 seconds. Video plays at 5 frames per second for 5 seconds and shows raw EB3-GFP comets. Arrowhead indicates MHBC. Scale bars: 10 μm. Figure 3. Representative live confocal timelapse at the MHB of an embryo co-injected with memCherry, EB3-GFP mRNA, and Control MO. Timelapse data were acquired at 21-24 ss for 10 minutes at 1 frame per 4 seconds and cropped to 100 seconds. Video plays at 5 frames per second for 5 seconds and shows processed EB3-GFP comets and tracks. Images have been processed using the OTSU thresholding method and FIJI Trackmate plugin for particle tracks. Arrowhead indicates MHBC. Scale bars: 10 μm.

Movie 2. Control MO EB3-GFP Tracks Timelapse, related to
Movie 3. wnt5b MO EB3-GFP Raw Data Timelapse, related to Figure 3. Representative live confocal timelapse at the MHB of an embryo co-injected with memCherry, EB3-GFP mRNA, and wnt5b MO. Timelapse data were acquired at 21-24 ss for 10 minutes at 1 frame per 4 seconds and cropped to 100 seconds. Video plays at 5 frames per second for 5 seconds and shows raw EB3-GFP comets. Arrowhead indicates MHBC. Scale bars: 10 μm.
Movie 4. wnt5b MO EB3-GFP Tracks Timelapse, related to Figure 3. Representative live confocal timelapse at the MHB of an embryo co-injected with memCherry, EB3-GFP mRNA, and wnt5b MO. Timelapse data were acquired at 21-24 ss for 10 minutes at 1 frame per 4 seconds and cropped to 100 seconds. Video plays at 5 frames per second for 5 seconds and shows processed EB3-GFP comets and tracks. Images have been processed using the OTSU thresholding method and FIJI Trackmate plugin for particle tracks. Arrowhead indicates MHBC. Scale bars: 10 μm.
Movie 5. DMSO EB3-GFP Raw Data Timelapse, related to Figure 6. Representative live confocal timelapse at the MHB of an embryo co-injected with memCherry, EB3-GFP mRNA, and treated at 18 ss with DMSO. Timelapse data were acquired at 21-24 ss for 10 minutes at 1 frame per 4 seconds and cropped to 100 seconds. Video plays at 5 frames per second for 5 seconds and shows raw EB3-GFP comets. Arrowhead indicates MHBC. Scale bars: 10 μm.
Movie 6. DMSO EB3-GFP Tracks Timelapse, related to Figure 6. Representative live confocal timelapse at the MHB of an embryo co-injected with memCherry, EB3-GFP mRNA, and treated at 18 ss with DMSO. Timelapse data were acquired at 21-24 ss for 10 minutes at 1 frame per 4 seconds and cropped to 100 seconds. Video plays at 5 frames per second for 5 seconds and shows processed EB3-GFP comets and tracks. Images have been processed using the OTSU thresholding method and FIJI Trackmate plugin for particle tracks. Arrowhead indicates MHBC. Scale bars: 10 μm.
Movie 7. SP600125 EB3-GFP Raw Data Timelapse, related to Figure 6. Representative live confocal timelapse at the MHB of an embryo co-injected with memCherry, EB3-GFP mRNA, and treated at 18 ss with SP600125. Timelapse data were acquired at 21-24 ss for 10 minutes at 1 frame per 4 seconds and cropped to 100 seconds. Video plays at 5 frames per second for 5 seconds and shows raw EB3-GFP comets. Arrowhead indicates MHBC. Scale bars: 10 μm. at 1 frame per 4 seconds and cropped to 100 seconds. Video plays at 5 frames per second for 5 seconds and shows processed EB3-GFP comets and tracks. Images have been processed using the OTSU thresholding method and FIJI Trackmate plugin for particle tracks. Arrowhead indicates MHBC. Scale bars: 10 μm.