Introduction

Mammalian neurovascular endothelial cells (ECs) have unique functional and structural properties to regulate substance exchange between the blood and the brain, establishing the blood-brain barrier (BBB)^1,2. Loss of normal BBB function is associated with neuropathological states, including neurodegeneration, infection and cancer^1,2. Mural cells are vascular support cells that include pericytes and vascular smooth muscle cells (vSMCs) and control vascular development, function and BBB permeability^3–9. PDGFB/PDGFRβ signaling is required for the development of mural cells^3,4,10–12 and their depletion in PDGFB- or PDGFRβ-deficient rodents causes altered angiogenesis, aneurysm and vessel leakiness^3–6,13. Strikingly, rodent models of pericyte loss have an open BBB, with free transport of tracer dyes from the blood stream into the brain parenchyma but an otherwise intact vasculature^5,6. These and other observations have led to a current model whereby pericytes control BBB function by suppressing EC transcytosis, via a mechanism not yet fully understood. Such a mechanism holds significant translational promise to safely "open" the BBB in therapeutic settings, by discovering ways to disrupt pericyte control of ECs^5,14.

Studies of the BBB using invertebrates like Drosophila and some basal vertebrate fishes (e.g. some Elasmobranchii and Chondrostei) have suggested that the makeup of the BBB can vary significantly across the phylogeny. These species display a primarily glial cell barrier, rather than an EC barrier^15,16. However, in the modern teleost Danio rerio (zebrafish) the BBB cellular composition is more similar to that seen in mammals^17–21. As in mammals, the zebrafish BBB is controlled by tightly adherent vascular ECs that share many characteristics with the ECs of the mammalian brain. Zebrafish BBB endothelium expresses conserved molecular markers (e.g. glut1, cldn5, mfsd2a)^18,20–23 and displays similar tight cell-cell junctions as observed at the mammalian BBB^17,20,22. The membrane transport protein Mfsd2a has been suggested to have a conserved role in control of BBB function in zebrafish and mice^21,24, likely acting to reduce the transfer of substances between bloodstream and parenchyma by suppressing EC transcytosis^25,26. Furthermore, during development the mechanisms that control the formation of BBB ECs appear to show a high level of conservation between zebrafish and mice. Recent studies have identified major roles for Wnt–β-catenin signaling, brain EC specific MMPs, Vegfs and chemokine signaling in early BBB vascular EC development^23,27–29.

After the initial development of zebrafish BBB vasculature, mural cell populations develop and mature. Transcriptionally, developing pericytes and vSMCs express many genes that are highly conserved with mammals, for example zebrafish pericytes express well known markers such as pdgfrb, kcne4 and abcc9 ³⁰. As in mice, pdgfrb mutants fail to form brain pericytes, exhibit loss of vSMCs and display aneurysms as adults^13,31. As well as conserved pdgfrb control of mural cell development, Notch signaling³² also plays a role in mural cell development in zebrafish and in mice, with Notch 1 and 3 controlling the formation of pericytes in mice and Notch 2 and Notch 3 paralogues important in zebrafish^33,34. Overall, studies have suggested a high level of conservation of the molecular control of mural cell development, yet how mural cells control BBB function has not been interrogated in detail in zebrafish.

BBB function is first established in zebrafish between 3 days post fertilization (dpf, no barrier) and 5 dpf (established barrier)²¹. During this period, developing mural cells (pericytes and vSMCs) are establishing and expanding their interactions with brain ECs^31,34,35. The first glial cell neurovascular interactions are also observed during this period³⁶. Throughout subsequent larval, juvenile and adult development vSMC, pericyte and glial cell coverage of the vasculature increases^31,36,37. This is similar to mammals, albeit with some differences in the relative timing that lineages arise, and the percentage of coverage of neurovasculature by pericytes and glia may be lower in zebrafish. In mammals, microglia are known to play important roles in the control of the BBB and neurovascular unit³⁸. Zebrafish have a microglial/macrophage population, but it is yet to be explored in detail in the control of the BBB^39–41 and differences in other innate immune cells in the CNS has been recently reported⁴².

Overall, while there are some areas remaining to be fully explored, the development, cellular composition and the central role of ECs in the control of the zebrafish BBB is very similar to mammals. As such, zebrafish are accepted as a highly accessible model to study BBB function, as well as mural cell control of the BBB, relevant to mammalian biology¹⁹.

Here, we explored mural cell control of BBB function in zebrafish. We used pdgfrb mutants with loss of both pericytes and vSMCs, coupled with carefully optimized intravascular tracer dye injections. We analyzed barrier function from developmental to adult stages and found that at larval stages up to 14 dpf, well after the establishment of the BBB, loss of mural cells does not lead to measurable BBB leakage or loss of BBB integrity. In adult animals, we observed loss of BBB integrity that was consistent with leakage at hotspots, found to be aneurysm associated hemorrhages. These hotspots were also observed in juvenile animals (1 month old) and appeared to increase in number as animals aged. Our observations provide no definitive evidence of BBB leakage at capillary networks that might be consistent with a broad increase in EC transcytosis or loss of barrier integrity independent of focal hemorrhages. This work demonstrates that young animals can have a functional BBB in the absence of mural cells and highlights the central role for the EC in control of barrier function across the zebrafish neurovasculature.

Results

To investigate the influence of mural cells upon the neurovasculature, we generated a pdgfrb mutant (pdgfrb^uq30bh, hereafter pdgfrb^-/-), which possesses an early deletion leading to a frameshift, predicted to cause a premature stop codon and a putative null allele (Extended Data Fig. 1a). Using both the TgBAC(pdgfrb:EGFP)^uq15bh and TgBAC(abcc9:abcc9-T2A-mCherry)^um139 transgenic marker strains, we observed that pericytes were lost in the cerebral central arteries of pdgfrb mutants (Fig. 1a, Extended Data Fig. 1b–c), as observed in previous studies^13,31. These mutants showed indistinguishable development from their wild-type siblings up to 30 days post fertilization (dpf) and progressively reduced survival and craniofacial defects thereafter (Extended Data Fig. 1d–f), consistent with previous studies of pdgfrb null loss-of-function mutants^13,35. To assess how brain vasculature develops without pericytes, we imaged, traced and quantified central arteries in the midbrain (Fig. 1a–c). At 7 dpf, mutants displayed significantly reduced vascular complexity compared to siblings, with reduced vessel length and branch points (Fig. 1d–e). This was more severe at 14 dpf, demonstrating the phenotype is progressive (Fig. 1d–e). Timecourse imaging revealed the reduction in vascular complexity was apparent from 5 dpf (Fig. 1f–g). These results suggest that the pericytes (as vSMCs mature after 5 dpf³⁴) regulate cerebral angiogenesis or remodeling from early in larval development.

Fig. 1:

We next tested BBB function in mutants at 7 and 14 dpf. We performed intravenous injection of fluorescently conjugated tracers of various molecular weights (1-kDa NHS, 10- and 70-kDa Dextran), and live imaged fish at 2 hours post injection (hpi, the same timing used in²¹) (Extended Data Fig. 2a). 2000-kDa Dextran was co-injected to normalize for the amount of vascular tracer delivered in each animal, and we measured intravascular versus extravascular tracer intensity to assess leakage (see methods for full details). In control animals, quantification confirmed that 2000-kDa tracer remained stably within the vasculature, while 1 kDa and 10 kDa were more readily detected in the brain parenchyma and 70 kDa provided intermediate sensitivity (Extended Data Fig. 2b–c). This validated the use of 2000-kDa tracer to normalize for amount of tracer injected, as reported elsewhere⁴³. To provide additional confidence in our measurements, we also normalized the extravasated tracer intensity to the EC marker kdrl (Tg(kdrl:EGFP)^s843 or Tg(kdrl:Hsa.HRAS-mCherry)^s916). We found no evidence of elevated 10- or 70-kDa Dextran tracer extravasation into the brain parenchyma of pdgfrb mutants compared to siblings at either timepoint (7 or 14 dpf) using any of our normalization methods (Fig. 2a–d, Extended Data Fig. 3a–d). To confirm that our quantification methods were sufficiently sensitive to detect BBB leakage, we assessed 10- and 70-kDa tracer leakage before and after the establishment of BBB function (at 3 and 7 dpf²¹). With both 10- and 70-kDa tracers, we could detect an open BBB at 3 dpf that was closed at 7 dpf (Fig. 2e–f). Together, these results show that we can accurately measure BBB leakage and integrity in zebrafish and that based on these measures neither pericytes nor vSMCs are necessary for BBB function during larval stages.

Fig. 2:

To investigate how mural cells control cerebral vasculature at later stages of life, we tissue-cleared brains of adult stage pdgfrb mutants and siblings⁴⁴ and performed whole brain imaging. We confirmed loss of pericytes and vSMCs at this stage using TgBAC(pdgfrb:EGFP)^uq15bh and TgBAC(acta2:EGFP)^uq17bh (Fig. 3a–b). We examined a consistent region of the left optic tectum (midbrain) and detected decreased branching of capillaries and increased diameter of major vessels (Fig. 3d–f). The observed vessel dilation was restricted to larger caliber arteries that would normally be invested with vSMCs (Extended Data Fig. 4a–b) and was not seen in capillaries in this region of the brain (Fig. 3e–f). The loss of vSMCs was observable from the earliest stages of vSMC development and artery dilation was observed by 21 dpf, consistent with previous studies¹³ (Fig. 3c, Extended Data Fig. 4c–d). This suggests that major vessel dilation seen in pdgfrb mutants is due to loss of vSMCs and that reduced branching of capillaries is due to loss of pericytes (given the timing observed in Fig. 1).

Fig. 3:

To investigate BBB integrity at later stages, we injected 3-month-old adult zebrafish with 10-kDa Dextran. Brightfield imaging of vibratome-sectioned brains revealed severe aneurysms across large caliber vessels (Fig. 4a) with blood pooling that was potentially consistent with vessel ruptures (Fig. 4c) as previously reported³⁵. Furthermore, we detected parenchymal accumulation of tracer dye closely associated with large caliber vessel aneurysms (Fig. 4b–c). To quantify the distribution of parenchymal tracer dye from deep (around major artery aneurysms) to superficial (capillary beds) brain regions, we measured fluorescence intensity in 10 brain regions approximately 100 microns apart from medial to lateral locations. This demonstrated accumulation of tracer dye in medial regions associated with aneurysm in mutants (Fig. 4d–e). These animals showed little evidence of leakage in capillary beds, with marginally higher tracer intensity in mutants than controls in lateral locations that could be due to diffusion from medial locations with high concentrations of tracer (Fig. 4e). To better understand this extravasated tracer accumulation, we conducted whole brain imaging after tracer injection in 5-month-old adults. Interestingly, we observed highly localized 70-kDa tracer accumulation in regions that could be considered leakage "hotspots" at large caliber vessel aneurysms (Fig. 5a; Supplementary Videos 1 and 2).

Fig. 4:

Fig. 5:

As 10-kDa tracer showed higher sensitivity than 70-kDa tracer in our control experiments (Extended Data Fig. 2b–c), we examined 10-kDa tracer injections at earlier stages and found that hotspots were readily detectable from as early as 1 month using this lower molecular weight tracer (Fig. 5b). The hotspots were not readily detectable using 70-kDa Dextran in 2-month-old zebrafish, consistent with increased sensitivity of 10-kDa tracers (Extended Data Fig. 5a–d). Interestingly, the quantification of hotspots in whole cleared brains at 1 and 5 months of age showed that the number increased with aging, suggesting a progressive loss of BBB integrity at focal sites of leakage (Fig. 5c). To further understand the nature of leakage hotspots, we examined red blood cells in pdgfrb loss-of-function adults. We used pdgfrb Crispant knockouts, which we validated by finding that F0 CRISPR consistently reproduced the adult mutant phenotype (Extended Data Fig. 5e–f). Using the Tg(gata1:DsRed)^sd2 transgenic line to label erythrocytes, together with tracer injection, we found that knockout animals showed tracer accumulation concomitant with extravasated red blood cells at hotspots (Fig. 5d). This identified hotspots as sites of focal hemorrhage. Notably, these sites were not detected at capillaries (Fig. 5e). Taken together, this shows that as mutants age, loss of BBB integrity is associated with an increasing number of hotspot sites of focal hemorrhage, that are found along aneurysmal vessels.

The quantification of tracer dye intensity at 2 hpi in adult mutants from medial (major vessels) to lateral (capillaries) locations indicated leakage most closely associated with hotspot focal hemorrhages. However, the intensity of tracer dye found in the lateral capillary regions was measurably higher than in control animals (Fig. 4e). To test if this was likely due to progressive diffusion of tracers from medial to lateral locations or if there may be some local leakage at capillaries, we examined 10-kDa tracer leakage at 0.5 and 6 hpi. While hotspots could be detected at 0.5 hpi, we detected little tracer accumulation in medial locations and no evidence of parenchymal tracer leakage at lateral capillary locations (Fig. 5f, h, Extended Data Fig. 6a–a′′). However, after 6 hours of diffusion of tracer within the brain, we clearly observed tracer at both medial and lateral locations, with higher intensity in medial locations (Fig. 5g, i, Extended Data Fig. 6b–b′′). While we cannot rule out the possibility of slow leakage of tracer at capillaries occurring concurrently with leakage at hemorrhages, these observations seem consistent with a primary source of leakage at focal hemorrhages, followed by diffusion of tracer throughout the parenchyma.

Pericytes are proposed to regulate the BBB by suppressing transcytosis through brain capillary ECs^5,6. This model is built on the observation by electron microscopy of the induction of increased numbers of vesicular structures and caveolae in leaky pericyte-deficient vessels^5,6. To investigate if the loss of BBB integrity is associated with ultrastructural changes, we performed electron microscopy. We imaged large caliber vessels and capillaries (Fig. 6a–b, Extended Data Fig. 7a) and examined cellular junctions, basement membrane and caveolae (defined as uncoated vesicular profiles of <100 nm diameter). While tight junctions between endothelial cells remained intact, caveolae numbers along large vessel aneurysms in pdgfrb mutants were significantly increased (Fig. 6a′′–c). This increase was almost exclusive to the abluminal side of the ECs (Fig. 6c). This abluminal accumulation of caveolae may be more associated with changes in the mechanical state of ECs (as caveolae are highly regulated by membrane tension^45,46) than changes in intracellular transport. Furthermore, we detected thickening of basement membrane (BM) (Fig. 6d), gaps in the BM and fluid accumulation outside the vessel wall at aneurysms in mutants. We found no obvious difference in the caveolae density in capillary ECs of pdgfrb mutants compared with siblings (Extended Data Fig. 7b). Thus, leakage hotspots at aneurysms are associated with a loss of the normal flattened morphology of ECs, caveola induction and disruption of the BM, all factors that could contribute to or be observed concurrent with vessel wall rupture.

Fig. 6:

Discussion

Brain vasculature is endowed with mural cells that control vascular development and stability^3,4,13,47. Studies in rodents have shown that pericytes regulate BBB integrity, with knockout of Pdgfb or Pdgfrb leading to vessel dilations and brain hemorrhages from early in development^3,4. However, hypomorphic mutant alleles impacting Pdgfb or Pdgfrb show increased leakage of the BBB, with the severity of the leakage phenotype correlating with the expressivity of the pericyte phenotype^5,6. BBB leakage is associated with increased vesicular structures in ECs and the "open" BBB state in the absence of mural cells is thought to be caused by hotspots of increased EC transcytosis^2,48. Other potentially relevant pathways in ECs that could be altered by pericytes include altered cell-cell adhesion (e.g. Cldn5¹⁴), control of Mfsd2a^21,26, Netrin1-Unc5B signaling⁴⁹ or vitronectin regulation of integrin receptors⁵⁰, but direct links are yet to be established. Some studies using other models to deplete pericytes in mice have reported loss of pericytes without vessel leakage. For example, optical or chemical ablation of pericytes in the brain or retina of mice can be achieved without compromising the integrity of BBB or blood-retinal barrier^51,52, however these methods may induce inflammatory or compensatory mechanisms that also influence local vasculature. Taken together, these studies show that mural cells and in particular pericytes control the BBB in mammals, yet the precise downstream mechanisms by which pericytes regulate EC function require further study.

We show using zebrafish that mural cells are involved in cerebrovascular patterning in larval and juvenile animals, but these vessels maintain BBB function in the absence of mural cells. While the zebrafish BBB has been shown to be established between 3 and 5 dpf, we find that our leakage assays at 7 and 14 dpf in mural cell-deficient pdgfrb mutants indicate intact BBB function and wildtype vessel integrity. This is unlikely to be due to technical issues with detecting leakage, as multiple measures and methods of tracer normalization produced consistent results. Furthermore, we can readily detect the "open" BBB before it closes during development (Fig. 2e–f). This demonstrates that a functional vertebrate BBB can exist in the absence of pericytes. Importantly, while other species have a glial cell BBB (e.g. Drosophila, some sharks and sturgeon), the zebrafish is well established to have an EC barrier and the glial cell coverage at the stages analyzed here is far from complete³⁶, making glial cell compensation for a loss of pericytes unlikely. There are other evolutionary differences (e.g. developmental timing, innate immune differences^42,53) between zebrafish and mammals and our observation of intact BBB function without mural cells could represent an additional divergence. However, the high level of conservation of developmental pathways in BBB ECs and the very similar transcriptional signatures of both neurovascular ECs and mural cells between species^19,30,34,54 may make this surprising. Once additional mechanisms of pericyte-EC crosstalk are established in mice and zebrafish, deeper mechanistic comparative studies will help to better understand if our observations identify major differences in BBB control between these organisms or not.

Ultimately, in older mutants from 1 month of age onwards, BBB integrity defects are observable with major arteries progressively harboring aneurysms and focal hemorrhages (leakage hotspots). These aneurysms are likely due to the loss of vSMCs seen in our pdgfrb mutants and other zebrafish pdgfrb mutant models^13,35. Adult mutant hotspots show extravascular erythrocytes and tracer dye, demonstrating that focal hemorrhages are a cause of vessel leakage. Similar to this, in Pdgfb or Pdgfrb mutant mice with strong phenotypic expressivity, hemorrhages are also reported (although not in mild or hypomorphic mutants used in BBB studies)^3,4,12. These observations suggest a highly conserved role for mural cells in the regulation of aneurysm and hemorrhage in mice and zebrafish. In our zebrafish model, BBB leakage in juvenile and adult animals appears to only be observed once hemorrhages or hotspots are present. Coupled with the fact that larval zebrafish can have no mural cells, but intact barrier function, we suggest that any roles for mural cells in BBB integrity beyond control of aneurysm and hemorrhage are likely to be minor roles in this setting. It remains possible that there is slow undetectable capillary leakage, but future models that specifically lack capillary pericytes and not vSMCs will be needed to test this. Understanding the nature of hotspot leakage more broadly in various mammalian and non-mammalian models of mural cell deficiency would seem likely to uncover further insights into BBB regulation. Deeper understanding of the fundamental biology of the BBB is still needed and will pave the way for increasingly sophisticated efforts to target the BBB in disease in the future.

Methods

Zebrafish husbandry

Zebrafish work was conducted in compliance with animal ethics committees at the Peter MacCallum Cancer Centre, The University of Melbourne and The University of Queensland. The following previously published transgenic lines were used in this study: TgBAC(pdgfrb:EGFP)^{uq15bh 55}, Tg(kdrl:Hsa.HRAS-mCherry)^{s916 56}, Tg(kdrl:EGFP)^{s843 57}, Tg(gata1:DsRed)^{sd2 58}, TgBAC(acta2:EGFP)^{uq17bh 55}, and Tg(5xUAS:RFP)^{nkuasrfp1a 59}.

To ensure optical transparency, live imaging was performed in slc45a2 (ENSDARG00000002593) F0 knockout animals as previously described⁶⁰ or by adding 0.003% 1-phenyl-2-thiourea (PTU) to embryo water at 1-day post-fertilization (dpf).

BAC recombineering and transgenesis

The TgBAC(abcc9:abcc9-T2A-mCherry)^uom139 transgenic line was generated by BAC recombineering as previously described^61,62. Briefly, T2A-mCherry insert with a kanamycin cassette was amplified by PCR using pCS2-T2A-mCherry-KanR plasmid as template. The primers used for amplifying the insert were:

5′-atggagcaggaggacggcctgtttgcatcttttgtcaaagccgacatgGAGGGCAGAGGAAGTCTGCTA-3′

5′- aaaatggcttttattgatctgttaaggccaaaagtggtgtaaagtggggaTCAGAAGAACTCGTCAAGAAGGCG-3′ (lowercase letters indicate homology arms to the BAC vector; uppercase letters indicate primer binding sites on the template plasmid).

The amplified insert was introduced into CH211-58C15 (abcc9) BAC clone containing a pRedET plasmid (GeneBridge, Heidelberg, Germany) and iTol2 cassette. Approximately 1 nL of purified BAC DNAs (100 ng/μL) mixed with tol2 transposase mRNA (25 ng/μL) were injected into single-cell stage embryos. Injected fish were raised to adulthood and screened by fluorescence and PCR for germline transmission. The pdgfrb:gal4FF BAC construct was generated previously³¹ and was injected to produce the TgBAC(pdgfrb:gal4FF)^um140 transgenic line. The pdgfrb:gal4FF BAC construct was kindly provided by Dr. Nathan Lawson (Department of Molecular, Cell, and Cancer Biology, University of Massachusetts Chan Medical School, USA).

Genome editing and genotyping

The pdgfrb^uq30bh mutant strain was generated using CRISPR/Cas9-mediated genome editing. The resulting allele carries a 39-bp deletion and a 5-bp insertion in exon 3 of pdgfrb (ENSDARG00000100897), leading to a predicted frameshift and premature stop codon.

The gRNA sequence and genotyping primers used were:

• gRNA binding site: 5′-GATGGTGACTAAGACGCGA-3′

• Forward genotyping primer: 5′-CTTCCTTAGATCCTGACGTGTG-3′

• Reverse genotyping primer: 5′-TATTGATGGGTTCGTCACCAG-3′

F0 pdgfrb crispants used were generated using Alt-R CRISPR-Cas9 reagents (IDT). The gRNA sequence and genotyping primers used were:

• Dr.Cas9.PDGFRB.1.AA: 5′-GATGGTGACTAAGACGCGAG-3′

• Forward genotyping primer: 5′-CTTCCTTAGATCCTGACGTGTG-3′

• Reverse genotyping primer: 5′-TATTGATGGGTTCGTCACCAG-3′

• Dr.Cas9.PDGFRB.1.AB: 5′-CTCGGTGCACACATAAACCC-3′

• Forward genotyping primer: 5′-GACGAGAACATCCCAGACTTTC 3′

• Reverse genotyping primer: 5′-GCGTGTAAACAAATCCTAACGG-3′

Throughout this study, sibling control fish were pooled homozygous wild-type and heterozygous animals, as no significant differences were detected between these genotypes (e.g. Extended Data Fig. 1c-d).

Fluorescent tracer injections

Fluorescently conjugated tracers used were: 1 kDa NHS Ester–Alexa Fluor™ 405 (Thermo Fisher: A30000), 10 kDa Dextran–Cascade Blue (Thermo Fisher: D1976), 10 kDa Dextran–Alexa Fluor™ 647 (Thermo Fisher: D22914), 70 kDa Dextran–Fluorescein (Thermo Fisher: D1822), 2000 kDa Dextran–Tetramethylrhodamine (Thermo Fisher: D7139) each at 5 mg/mL.

Larvae (3–7 dpf) were anesthetized with tricaine and embedded in 0.5% low-melting agarose (Bio-Rad 1613112) in E3 water on 35-mm glass-bottom dishes (MatTek P35G-1.5-20-C). Approximately 5 nl was injected into the posterior cardinal vein (PCV). At 14 dpf, tracers were injected into the PCV with ∼10 nL tracer on 3% agarose pads. In juveniles and adults retro-orbital injections were performed⁶³ on 3% agarose pads, with 0.2 μL or 0.5 μL tracer respectively. Tracers were injected individually or co-injected (1–70 kDa with 2000 kDa) depending on experimental design.

Imaging

Sample preparation

For live imaging at 3-14 dpf, larvae were mounted in 0.5% low-melting point agarose in E3 embryo water on a 35-mm glass bottom dish as previously described⁶⁴. For fixed tissue imaging, zebrafish were euthanized with tricaine overdose (10 g/L), fixed (as whole or after brain extraction) in 4% paraformaldehyde (PFA) at 4°C overnight and washed in PBS. Brains were embedded in 7% low-melting agarose in PBS (Tissue-Tek cryomolds) and coronally sectioned at 200 μm thickness using a vibrating microtome (Leica VT1000 S). Sections were mounted in RapiClear 1.52 solution (SunJin Lab, RC152201) for 3 hours at room temperature prior to imaging. Whole fish or brain samples were cleared using CUBIC clearing adapted from a previously described protocol⁴⁴. Briefly, samples were incubated in CUBIC-L solution at 37°C overnight after fixation, washed and incubated in PBS at 4°C overnight, then incubated in CUBIC-R solution at room temperature overnight. Samples were embedded in 2% agarose in CUBIC-R on a 35-mm glass bottom dish for imaging.

Image acquisition

Imaging was performed using the following microscopes and objectives: Olympus FV-MPERS upright multiphoton, 25x water dipping 1.05 NA (Fig. 1a-b,f; Extended Data Fig. 5e), Olympus FV3000 confocal 10x air 0.4 NA (Fig. 3a-a′; Fig. 4b-b′,c′,d-d′; Fig. 5a-a′; Extended Data Fig. 5f) and 40x oil immersion 1.4 NA (Fig. 5d-e), Olympus/Evident FV4000 confocal 10x air 0.4 NA (Extended Data Fig. 2d-e; Fig. 3b-b′; Extended Data Fig. 4b,d; Fig. 5b,f,g; Extended Data Fig. 6a-b′′) and 30x silicone immersion 1.05 NA (Extended Data Fig. 1b; Fig. 2a,c,e; Extended Data Fig. 2b; Fig. 3c; Extended Data Fig. 4a-a′′,b′,c-c′; Fig. 5b′) Nikon TiE with Yokogawa CSU-W1 spinning disk, Andor Sona sCMOS camera, 10x air 0.45 NA (Extended Data Fig. 5a,b), 20x air 0.75 NA (Extended Data Fig. 3c; Extended Data Fig. 5a′) and 40x water immersion 1.15 NA (Extended Data Fig. 3a). For tracer leakage assays, live imaging was performed at 2 hours post-injection (hpi), with a maximum variation in timing of 15 minutes due to the sequential imaging. Fixed tissue samples were collected at 0.5 hpi (Fig. 5f, Extended Data Fig. 6), 2 hpi (Extended Data Fig. 2d–e, Fig. 4b-d′, Fig. 5a–b′,d–e, Extended Data Fig. 5a–b) or 6 hpi (Fig. 5g). Stereomicroscope images were acquired using an NSZ-606 Zoom Stereomicroscope with a Tucsen Photonics Co. camera using TCapture/IS Capture software.

Transmission electron microscopy

Brains were extracted from euthanized 5-month-old zebrafish, fixed in 2.5% glutaraldehyde for 1 hour at room temperature and washed in PBS. For sectioning, the brains were embedded in 7% low-melting point agarose in PBS in cryomolds (Tissue-Tek) and were sectioned coronally using a vibrating microtome (Leica VT1000 S) with 200 μm thickness. Tissue sections were processed as described previously⁶⁵. Briefly, tissue sections were immersed consecutively in a series of aqueous solutions: 1.5% potassium ferricyanide and 2% osmium tetroxide, 1% thiocarbohydrazide, 2% osmium tetroxide, 2% uranyl acetate and 0.06% lead nitrate using a Pelco Biowave at 80W under vacuum for 3 min each. Vibratome sections then underwent serial dehydration in increasing concentrations of ethanol, before serial infiltration with increasing concentrations of Procure 812 resin. Ultrathin sections were obtained using a Leica UC64 ultramicrotome and imaged on a JEOL JEM-1011 at 80 kV.

Image processing

Images were processed using Fiji⁶⁶ or Imaris 10.1 (Bitplane). Fluorescent images are shown as maximum intensity projections (MIP), unless otherwise indicated. Vascular tracing (Fig. 1c) and masked head regions (Extended Data Fig. 1b) are presented as snapshots, and videos highlighting the hotspot leakage sites were generated using Imaris 10.1 and annotated in Adobe Premier Pro 2024. Schematics were prepared in Adobe Illustrator 2025, and pseudocolor overlays on transmission electron microscopy images were generated in Adobe Photoshop 2024.

Quantification of phenotypes

Gross anatomy

Prior to imaging and genotyping, fish body lengths were measured from the tip of the head to the tail base. Body length datasets were compiled from multiple independent experiments. Survival was quantified using separately housed pdgfrb^uq30bh mutants and siblings to enable continuous tracking. Sibling and mutant groups were initially sorted based on pdgfrb:EGFP-positive cells in the brain using TgBAC(pdgfrb:EGFP)^uq15bh, and were genotyped by PCR at the endpoint for each animal. Cerebrovascular aneurysms in juvenile pdgfrb mutants were qualitatively classified as mild or severe on the basis of aneurysm size relative to sibling controls.

Mural cell number, vessel length, diameter and branching points

Mural cell number was quantified by counting pdgfrb:EGFP-positive cells (TgBAC(pdgfrb:EGFP)^uq15bh) in the brain associated with blood vessels (Tg(kdrl:Hsa.HRAS-mCherry)^s916) in 3D view using Imaris 10.1. Vessel length, diameter and branching points were quantified in Imaris 10.1. At larval stages, the lumenized midbrain central arteries (Tg(kdrl:Hsa.HRAS-mCherry)^s916) branching from the middle mesencephalic central arteries (MMCtA) was traced using the Filament tool and the total filament length (μm) was plotted. Branching points within the traced vasculature were counted manually in 3D view. For adults, 100-μm thick maximum intensity projections (MIP) were generated from whole brain imaging, using the continuation of the left MMCtA as an anatomical landmark to define a consistent region of interest. For diameter measurements, 10 capillary diameters were measured per sample. Vessel anatomy was assigned using Isogai et al⁶⁷ as reference.

Tracer intensity

The quantification of tracer intensity was performed on genotype blinded image sets using Imaris 10.1 (Bitplane) and Fiji⁶⁶ in 50-μm thick MIPs, generated starting from 50 μm below the dorsal longitudinal vessel (DLV) to avoid including potential leakage from surface vessels. For live imaging quantifications, two different methods were used to validating findings. In the first method, the midbrain region was masked using the tracer accumulation in the skin as an outer boundary and metencephalic vessels as the border with the hindbrain. Vasculature was then masked and removed using Tg(kdrl:EGFP)^s843 or Tg(kdrl:Hsa.HRAS-mCherry)^s916, and mean intensity of extravasated tracer within the midbrain was measured. Relative leakage by molecular weight in wild-type animals was assessed by normalizing extravasated midbrain tracer intensity to intravascular tracer intensity. For comparisons between groups (siblings versus pdgfrb mutants, or for wild-type at 3 versus 7 dpf), extravasated midbrain tracer intensity was normalized to vascular reporter intensity (kdrl:EGFP or kdrl:Hsa.HRAS-mCherry). In the second method, in experiments including 2000-kDa dextran as an injection control, extravasated 70- or 10-kDa dextran intensity in midbrain parenchyma was calculated from the average intensity in four defined parenchymal regions. These values were normalized to the vascular intensity of 2000-kDa dextran, calculated from the average intensity in four regions within the DLV lumen. As such, for each data point: intensity=parenchymal tracer (10 or 70 kDa) intensity/luminal 2000-kDa tracer intensity. For datasets assessing leakage between siblings versus pdgfrb mutants at 7 and 14 dpf (Fig. 2, Extended Data Fig. 3), we assessed the raw (non-normalized) tracer intensity in the parenchyma and compared these values to normalized values. We found that both normalization strategies improved the analysis, reducing variance and controlling for differences in tracer delivery and image variation between animals.

For tracer intensity quantification in sectioned brains, 50-μm thick MIPs were generated. Tracer intensity outside the vasculature was measured at ∼70-μm intervals from medial to lateral within a 0.7-mm region in juveniles, and at ∼100-μm intervals within a 1-mm region in adults, due to differences in brain size. For tracer intensity quantification in sectioned brains, 50-μm thick MIPs were generated and dextran intensity outside the vessels was measured at approximately every 70 μm from medial to lateral direction within 0.7 mm in juveniles and at every 100 μm within 1 mm in adults due to size difference. These measurements were normalized to vascular reporter intensity (kdrl:EGFP). MMCtAs served as anatomical landmarks to ensure consistent region selection.

Hotspot leakage sites

Hotspot leakage sites were identified as prominent focal accumulations of fluorescent tracer adjacent to and outside the vasculature. Quantification was performed in 3D using Imaris 10.1.

TEM quantification

Imaged sections were quantified using Fiji⁶⁶. Caveolae were defined as circular profiles of less than 100 nm in diameter and were scored as luminal or abluminal based on proximity to each surface membrane (within 500 nm of each surface or in a thin-walled vessel the caveolae closest to each surface). Basement membrane thickness was scored in six different locations per vessel that were selected at random.

Statistical analysis

Graphic representations and statistical analyses were performed using GraphPad Prism 10. Shapiro–Wilk test was applied to test normal distribution of the data. Two-tailed Student’s t-tests were used for two-group comparisons with normal distribution, and Mann–Whitney U tests for non-normal data. One- or two-way ANOVA followed by Tukey’s post hoc test was used for multiple group comparisons. P<0.05 was considered statistically significant.

Data availability

Source datasheet 1 contains the numerical data used to generate the figures.

Acknowledgements

This project was supported in part by funding from Bright focus (A2018807S), the Australian Research Council (ARC) (DP210102712) and The Brain Cancer Centre (founded by Carrie’s Beanies for Brain Cancer). B.M.H was supported by a National Health and Medical Research Council Senior Research Fellowship (1155221) and Investigator grant (2033008). R.G.P was supported by an ARC Laureate Fellowship (FL210100107). We thank the Centre for Advanced Histology and Microscopy (RRID:SCR_025432) at the Peter MacCallum Cancer Centre and Microscopy Australia Research Facility at the Centre for Microscopy and Microanalysis at The University of Queensland. We thank Koji Ando and Nathan Lawson for providing constructs and protocols for BAC recombineering. We also thank Kelly Smith and Marcos Sande Melon for academic discussions and technical suggestions.

Additional information

Author contributions

O.F.B performed, analyzed experiments and co-wrote manuscript. B.M.H, conceptualized experiments, analyzed data and co-wrote manuscript. A.U.G, S.D, W.W, S.P, M.C.R.G, Y.W.L, J.R, R.P, A.L and A.F performed and analyzed experiments.

Funding

Department of Education and Training | Australian Research Council (ARC) (DP210102712)

• Benjamin M Hogan

DHAC | National Health and Medical Research Council (NHMRC) (1155221)

• Benjamin M Hogan

DHAC | National Health and Medical Research Council (NHMRC) (2033008)

• Benjamin M Hogan

Department of Education and Training | Australian Research Council (ARC) (FL210100107)

• Robert G Parton

Additional files

Source datasheet 1. Numerical data used to generate all plots presented in the figures. Source data in an excel spreadsheet in which each separate tab provides the individual measurements and numbers used to generate the plots presented in each figure.

Supplementary Video 1

Supplementary Video 2