Fig 1.
Anatomical locations of the investigated areas.
Photographic image of the isolated mouse skull cap with cranial dura mater is shown. White squares labeled 1, 2, 3 and 4 in the left and right coronal suture regions designate investigated areas.
Fig 2.
In the investigated area, PDPL and LYVE1 do not colocalize.
Representative immunofluorescence images showing blood vessel staining with WGA lectin (magenta), PDPL-associated immunofluorescence (blue) and LYVE1 expression (green) in the CS investigation area taken with 20x objective. Note that in all 4 groups of animals LYVE1 expression (green) does not overlap with PDPL expression (blue). Scale bar in D, 50 μm.
Fig 3.
Dura mater stromal and vascular responses to alkaline corneal injury.
Representative immunofluorescence images of FSP1 (magenta), PDPL (blue) and α-SMA (yellow) expression from investigation area 3 taken with 20x objective. Note significantly elevated stromal FSP1 expression in WT injury group (B) compared with WT naïve (A), DCN-/- (C) and DCN-/- injury (D) animals. Also note endothelial (white arrows in B, D and F) FSP1 expression in WT injury (B) and DCN-/- injury mice two weeks post initial insult. In F, area marked with the white square in D was imaged using 40x objective to show in more detail post injury endothelial FSP1 expression (white arrows), as well the loss of VSMC coverage of the blood vessels and acquisition of bead-like morphology by VSMCs (white asterisk). E, box diagram depicting fold change of in FSP1-associated immunofluorescence in four experimental groups compared with WT naïve animals. Whiskers, minimal to maximal, P–Ordinary One-Way ANOVA, *** P < 0.0001. Scale bars in D and F, 50 μm. For black and white images depicting each channel individually for FSP1, PDPL, and α-SMA please see S1–S3 Figs correspondingly.
Fig 4.
Changes in blood vessel VSMC coverage two weeks post alkaline corneal injury.
A through F, blood vessel segmentation stages are depicted. In A through F, red channel represents blood vessel wall WGA lectin staining and blue channel represents α-SMA VSMC staining. The ensemble of deep learning cascades was used to perform blood vessel segmentation to enable quantitation of VSMC coverage. In A, the original RGB image is shown. In B and C, masked red and blue channels are shown respectively. In D. vessel mask is shown whilst in E, vessel tree boundary is applied over the original image. Blood vessel region on the original image is shown in F with gray color marking the background. G, box diagram depicting fold change in α-SMA-associated immunofluorescence over blood vessel area in four experimental groups compared with WT naïve animals. Whiskers, minimal to maximal, P–Ordinary One-Way ANOVA, * P < 0.05. Scale bar in F, 50 μm.
Fig 5.
Changes in the number of PDPL positive dura mater lymphatic sprouts two weeks post alkaline corneal injury.
PDPL+ meningeal lymphatics were visualized using primary anti-PDPL antibody followed by the secondary Alexa Fluor® 647-conjugated goat anti-Syrian hamster IgG. Note significant increase in PDPL+ lymphatic vessel index in WT injury (B) group compared with WT naïve animals (A), as well as the number of PDPL+ lymphatic sprouts (white arrows) in WT injury (B) and DCN-/- (C) groups, but not in DCN-/- injury (D) mice compared with WT naïve (A) mice. E and F, box diagrams showing changes in PDPL+ dura mater lymphatic vessel index and the number of PDPL+ sprouts two weeks post alkaline corneal injury. Whiskers, minimal to maximal, P–Ordinary One-Way ANOVA, * P < 0.05; *** P < 0.0001. Scale bar in D, 50 μm.
Fig 6.
Altered LYVE1 expression within CS area of the cranial dura mater in DCN-/- animals.
The expression of lymphatic vessel endothelial hyaluronan receptor LYVE1 (green) was visualized using AlexaFluor 488-conjugated anti-mouse LYVE1 antibody. Note that in addition to nonvascular stromal LYVE1 expression and individual nonvascular LYVE1+ cells, perivascular blood vessel LYVE1 expression (yellow arrowheads in A through D) could be observed. However, there was no LIVE1 positive lymphatic vessels found in DM areas analyzed in this study. Compared with WT naïve animals (A) overall LYVE1 associated immunofluorescence did not change significantly in WT injury group (B). However, compared with WT naïve and WT injury animals, it was significantly reduced in DCN-/- mice (C). Interestingly, two weeks post alkaline corneal injury, dura mater nonvascular tissue stroma LYVE1 expression increased significantly in DCN-/- injury group (D, white arrow). In addition to the overall LYVE1 associated immunofluorescence, the number of nonvascular LYVE1 positive cells was also analyzed. Compared with WT naïve animals, the number of nonvascular LYVE1+ cells did not change significantly in WT injury group. However, it was significantly reduced in DCN-/- and DCN-/- injury mice. E and F, box diagrams showing fold change differences in the overall LYVE1 associated immunofluorescence compared with WT naïve animals (E) and the number of nonvascular LYVE1+ cells (F). Whiskers, minimal to maximal, P–Ordinary One-Way ANOVA, * P < 0.05; ** P < 0.01; *** P < 0.0001. Scale bar in D, 50 μm.