New genetic tools for central and peripheral vascular endothelia

doi:10.21203/rs.3.rs-7421061/v1

The divergence between the central and peripheral vascular system, particularly the emergence of the blood-brain barrier (BBB), is central to the brain’s homeostasis and functions. However, the molecular and genetic constituents that separating the BBB vascular cells from the rest remain elusive. Using single cell transcriptomics, we identified new cerebrovascular markers, e.g. zinc finger protein Zic3 is explicitly found in adult brain endothelial cells and the Atp13a5 ATPase is only expressed in brain pericytes. Using new genetic models, we further confirmed the specificity of Zic3 in cerebrovasculature. Additionally, we developed a mouse model based on Plvap, and confirmed it is specific for endothelial cells of the peripheral tissue and circumventricular organs in brain. In-depth transcriptomics analysis between Zic3⁺ and Plvap⁺ endothelial cells revealed that genetic programs associated with lipid metabolism, transporter systems and tight junction signaling are critical drivers behind the separation of central and peripheral endothelia. These new murine genetic tools will further aid our understanding of vascular heterogeneity and BBB specialization.

The blood-brain barrier (BBB), blood-spinal cord barrier and blood-retinal barrier provide the physical barriers limiting the entrance of circulating pathogens and toxins, immune cells and body’s metabolic waste products into the central nervous system (CNS), and supplies critical energy metabolites such as glucose and lactate, essential amino acids, fatty acids, vitamins and growth hormones via selective transporter systems^1,2. The BBB also helps clear brain’s own metabolic wastes including excess of neurotransmitters and proteinaceous molecules such as Alzheimer’s amyloid-β species, providing neurons with a tightly controlled microenvironment³. On the other hand, the vasculatures in the peripheral organs are more permeable, particularly in organs such as liver and kidney⁴. This functional heterogeneity also suggests that tissue microenvironment may provide important guidance cues for the specialization of their own vasculatures⁵. For example, Wnt signaling regulates both neural and BBB development in the vertebrate brain⁶, while hepatocyte growth factor (HGF) stimulates liver angiogenesis⁷.

Endothelial cells are the bedrocks of the vasculature. Anatomically, CNS endothelia are similar to the ones in the peripheral organs, connected by intercellular junctions and embedded in basement membrane. Yet, they are in close interactions with perivascular pericytes and astrocytic endfeet⁸, resulting in more tightly sealed tight junctions and minimal level of transcytosis⁹. At the molecular level, brain endothelial cells carry high levels of tight junctional proteins such as claudins and occludins¹⁰, and are enriched with special transporters such as glucose transporter 1 (GLUT1, encoded by SLC2A1)¹⁰, and major facilitator superfamily domain containing protein MFSD2A, a sodium-dependent lysophosphatidylcholine (LPC) transporter¹¹. The peripheral endothelia on the other hand are often fenestrated¹², providing non-selective transcellular transport through pinocytosis. The plasmalemma vesicle-associated protein (PLVAP), functioning as the diaphragm of these fenestrae, is in general absent in the adult brain¹³. However, the intrinsic programs underlying the central and peripheral divergence and the heterogeneity of vascular system remain underexplored. In addition, existing genetic tools (e.g., reporters and Cre drivers) are inadequate to differentiate or specifically manipulate central or peripheral vasculature endothelia, posing a significant research barrier.

To address these gaps, we analyzed single-cell transcriptomic datasets from 11 mouse tissues to identify endothelial markers with organ-specific expression patterns and revealed that Zic Family Member 3 (Zic3) is explicitly expressed in brain endothelial cells. We generated a Zic3-T2A-tdTomato-IRES-CreERT2 knock-in model, with both fluorescent tdTomato reporter and CreERT2 recombinase expressed from the endogenous alleles, and confirmed both tdTomato reporter expression profiles and CreERT2 recombinase activity are specific to the endothelial cells of the central nervous system, including brain, spinal cord and retina. Furthermore, we generated a Plvap-T2A-EGFP-IRES-CreERT2 knock-in model, and found that it reliably targets the peripheral endothelial cells in heart, lung, liver and kidney, as well as the endothelial cells in the choroid plexus and meningeal vessels in the brain. Based on these two markers, we further analyzed the endothelial transcriptomics at single-cell level, which revealed that transporter system and tight junction signaling are the major divergent points between the central and peripheral vasculatures. In sum, these new Zic3 and Plvap genetic models represent unique tools towards a better understanding of the molecular and genetic underpinning of the complexity of our vascular system in development, aging and diseases.

Differential vascular permeabilities between CNS and peripheral organs

The divergence between the central and peripheral vascularsystems in terms of permeabilities has been known for more than a century¹⁴, since the early observations based on dye injections in embryos¹⁵. Using HRP as a traditional tracer administered in the circulation¹⁶, we compared the vascular permeabilities between the central and peripheral organs (Extended Data Fig. 1a). Briefly, mice received a single dose of 0.5 mg/g of body weight of HRP solution in saline, and tissues were harvested 2 h later after perfusion. With fluorescent Tyramide signal amplification (TSA)¹⁷, we can amplify the HRP tracer and compare the vascular permeabilities between different organs (Fig. 1a and Extended Data Fig. 1b). Among peripheral organs, kidney exhibited the highest vascular permeability, followed by liver, intestine and lung, while heart and spleen showed relatively low levels of HPR molecule accumulation. On the other hand, the brain and spinal cord had very minimal HRP signals, particularly in the cortex, or the grey matter of spinal cord (Fig. 1a).

Zic3 is a new marker for brain endothelial cells

To explore the molecular basis of vascular heterogeneity, we integrated and performed secondary analysis on three single cell RNA sequencing (scRNA-seq) datasets covering both mouse brain and peripheral vasculatures^18–20. 48,526 single-cell transcriptomes were collected for the secondary analysis, as we recently reported²¹. In Uniform Manifold Approximation and Projection (UMAP), four major vascular cell types were separated into 20 clusters across different tissues including endothelial cells, mural cells, Oligodendrocytes, and fibroblasts (Fig. 1b), based on their specific genetic markers. These include but are not limited to Cldn5, Podxl and Kdr for endothelial cells, Pdgfrb, Vtn and Cspg4 for mural cells. Interestingly, we found that Zic3 is unique to the brain endothelium (Fig. 1c), and confirmed that Atp13a5 is a brain mural cell marker²¹ (Fig. 1d).

Zic3 encodes a zinc finger protein and transcription factor, essential for mammalian embryonic cardiovascular and neural development²². It is highly specific to the brain based on the Tabula Muris dataset²³ (Extended Data Fig. 1c-d). To further confirm its specificity, we analyzed the brain vascular scRNA-seq dataset¹⁸ and found that Zic3 expression is exceedingly precise to brain endothelial cells of capillaries, arterioles and veins, but not expressed in any other vasculature cells including VSMCs, oligodendrocytes, fibroblasts, microglia, or astrocytes (Extended Data Fig. 1e). In addition, analysis of a large dataset of 1,093,785 cells combining adult mouse cortical and hippocampal areas²⁴ confirmed that Zic3 expression is exclusively in endothelial cells, but not in any neuronal and glial cells (Extended Data Fig. 1f). Besides Zic3, we examined transporters that are currently known to be enriched in brain endothelial cells. For example, lysolipid transporter Mfsd2a and glucose transporter Slc2a1 are well-known for their roles in cerebrovasculature^25,26, but their expressions are evident in testicular endothelial cells based on the integrated transcriptomics (Extended Data Fig. 1g). In addition, although organic anion transporter Slco1c1 is specific to brain endothelial cells in the integration dataset, it is highly expressed in astrocytes as well, based on datasets including the Tabula Muris²³ (Extended Data Fig. 1h).

Fluorescent in situ hybridization (FISH) with RNAscope probes validated that Zic3 transcripts are expressed throughout the brain regions including cortex, but not observed in peripheral organs such as liver or heart (Fig. 1e-f ), and Zic3 mRNA transcripts were colocalized exclusively within the lectin-positive brain endothelium (Extended Data Fig. 2a). In addition, immunohistochemistry analysis with antibodies that react with murine ZIC3 protein confirmed its presence in vasculature across brain regions (Fig. 1g), including cortex, hippocampus, cerebellum, hypothalamus, striatum, medulla, midbrain, thalamus and pons (Extended Data Fig. 2b). This indicates that Zic3 is potentially a new specific marker for CNS endothelial cells.

Zic3-T2A-tdTomato-IRES-CreERT2 knock-in model for CNS endothelial cells

Next, we generated a new transgenic model targeting the Zic3 locus. This Zic3-T2A-tdTomato-IRES-CreERT2 knock-in model carries expressions of both tdTomato reporter and CreERT2 recombinase through the endogenous Zic3 locus (Fig. 2a, also see Methods). One F0 founder was selected based on germline transmission and genome sequencing, and the F1 generation was further tested with southern blot analysis and validated by PCR (Extended Data Fig. 3a-c). As Zic3 is located on X-chromosome, the model is maintained as hemizygous (Zic3^tdT/Y for male, and Zic3^tdT/+ for female). Homozygous female Zic3^tdT/tdT mice can be obtained from breeding of hemizygous, and appear normal. Importantly, tdTomato is reliably expressed in adult hemizygotes and homozygotes (Fig. 2b).

The Zic3-tdTomato profiles overlay well with Lectin-labeled endothelium throughout the brain, including cortex, hippocampus and thalamus (Fig. 2c), but not in peripheral tissues such as liver, heart or kidney (Fig. 2d). To verify the expression of Zic3 marker in the CNS, we also examined the spinal cord and retina in our transgenic model. We found robust Zic3-tdTomato signals in both white and gray matter of the spinal cord (Fig. 2e), with a little higher coverage in gray matter. In the retina, Zic3-tdTomato expressing cells were found in most of the vessels (Fig. 2f and Extended Data Fig. 3d).

Additional immunostainings further confirmed that Zic3-tdTomato expressing cells are indeed brain endothelial cells, as they are CD31-positive (Fig. 2g and Extended Data Fig. 3e), but not other cell types in the brain such as CD13-positive pericytes (Fig. 2h and Extended Data Fig. 3f-g), NeuN-positive neurons, GFAP-positive astrocytes, or IBA1-positive microglia (Extended Data Fig. 3h-j). Zic3-tdTomato reporter is found in over 90% of capillary endothelial cells, as well as in ~40% of endothelial cells on VCAM1-positive veins and venules or SMA-positive arteries and arterioles (Fig. 2i-k). It is estimated to cover at least 80% of the cerebrovasculature (Extended Data Fig. 3k).

Characterization of the Zic3-CreER recombinase activity

To test the CreER activity in our Zic3-T2A-tdTomato-IRES-CreERT2 model, we crossed it with the Ai3-EYFP floxed reporter line²⁷, and induced the CreER activity with tamoxifen administration (Fig. 3a, also see Methods). With 7 injections of Tamoxifen, more than 90% of Zic3-tdTomato⁺ endothelial cells in the brain, retina and spinal cord expressed robust EYFP signals (Fig. 3b-d). Yet, no recombination or EYFP reporter signal was observed in peripheral tissues such as heart, kidney or liver (Fig. 3c-d). Hence, our data demonstrated that Zic3 marker is unique for the CNS endothelium, and Zic3-T2A-tdTomato-IRES-CreERT2 model may help us decode the molecular and genetic underpinning of its specification.

Plvap-T2A-EGFP-IRES-CreERT2 model for peripheral and non-BBB endothelial cells

MECA-32 (an antibody against PLVAP glycoprotein) is widely used for staining of endothelial cells in embryos and most adult mouse peripheral tissues²⁸, and it also label the vasculature in some of the circumventricular organs, such as the choroid plexus. This is highly consistent with single cell transcriptomic (Fig. 1c). Therefore, we also generated the Plvap-T2A-EGFP-IRES-CreERT2 knock-in mouse model, carrying the EGFP reporter and CreERT2 recombinase (Fig. 4a). One F0 founder was selected based on germline transmission and genome sequencing, and the F1 generation was further tested with southern blot analysis and PCR for the integrity of the knock-in allele (Extended Data Fig. 4a-c). Plvap-T2A-EGFP-IRES-CreERT2 hemizygous mice are healthy, and we observed robust Plvap-EGFP signal in the lung and heart vasculature (Fig. 4b-c), which is consistent with the transcriptomic profiles in the Tabula Muris database²³ (Extended Data Fig. 4d).

To access the CreER activity of Plvap-T2A-EGFP-IRES-CreERT2 model, we crossed it with the Ai14-tdTomato floxed reporter mice²⁷ (Fig. 4d). Following Tamoxifen injections, we analyzed Plvap-CreERT2 driven Ai14-tdTomato reporter (Plvap-Cre::Ai14-tdT), and found very robust labeling of peripheral vasculatures including kidney and liver (Fig. 4e). Interestingly, we also observed Plvap-EGFP signals in the choroid plexus within the brain ventricles, but not in the cortex or hippocampus (Fig. 4f). This is further confirmed with the Plvap-Cre::Ai14-tdT model, which showed robust labeling of choroid plexus vasculature following Tamoxifen injections (Fig. 4g-h).

In addition, thePlvap model also reliably targets the meningeal vessels. As meningeal vasculature is known to be vulnerable in head injuries²⁹, we conducted mild traumatic brain injury (mTBI) on this model using a controlled cortical impact system³⁰ (Fig. 4i). Even though the impact did not result in direct damage to the skull (Fig. 4j)³⁰, the meningeal vessel were severely damaged in the ipsilateral side after the injury, as shown by the dramatic loss of Plvap-Cre::Ai14-tdT reporter under the impact site(Fig. 4k). Interestingly, the Plvap-EGFP reporter showed an increased pattern 3 days after mTBI, indicating robust angiogenesis and vascular remodeling occurred shortly after the injury(Fig. 4k). This demonstrated the new Plvap knock-in model is suitable for studying peripheral endothelial cells, and the built-in reporter and CreERT2 recombinase can be used synchronously for effective lineage tracing in vivo.

Transcriptomic heterogeneity between CNS and peripheral endothelial cells

To further explore the differences between Zic3-positive and Plvap-positive endothelial cells at single cell transcriptomic level, we analyzed 37,083 single endothelial cells from various organs, including the brain, intestine, colon, heart, kidney, liver, lung, muscle, spleen, and testis (Fig. 5a-c). Specifically, we compared 1,903 Zic3-positive ECs with 20,517 Plvap-positive ECs. These two populations exhibit distinct transcriptomic profiles, and we identified 341 genes with significant differential expression: 277 genes are significantly enriched in Zic3-positive ECs, while 64 genes show higher expression in Plvap-positive ECs. For instance, transporter genes such as Slc2a1, Slco1a4, Slco1c1, Slc22a8, Slc7a5, Slc38a3 and Mfsd2a are more abundantly expressed in Zic3-positive ECs, whereas lipid metabolism associated genes such as Phospholipid Phosphatases Plpp1 and Plpp3, fatty acid binding protein Fabp4, glycosylphosphatidylinositol anchored high density lipoprotein binding protein 1 (Gpihbp1) are commonly expressed in Plvap-positive ECs (Fig. 5d-e). In addition, a few genes showed organ specificities in the peripheral, e.g. surfactant protein C (Sftpc) is only found in lung ECs, C-Type Lectin domain protein Clec4g is mainly expressed in liver ECs, Insulin-like growth factor binding protein Igfbp5 is exclusive to kidney ECs, Stabilin 2 (Stab2) is highly enriched in spleen ECs (Fig. 5d). With these differentially expressed genes (DEGs), we conducted pathway analysis based on Gene Ontology enrichment method. Based on biological processes, these DEGs are related to transporter systems and regulate brain development (Fig. 5f). Their cellular functions are connected to membrane functions and intercellular junctions (Fig. 5g), and their molecular functions are associated with lipid binding and protein transporting (Fig. 5h). This was further confirmed with Ingenuity canonical pathway analysis, which showed that transporter system and tight junction signaling are the top pathways differentiating the Zic3-positive ECs from Plvap-positive ECs (Extended Data Fig. 5e). Taken together, these data further confirmed Zic3 and Plvap as unique markers for the CNS and peripheral endothelial cell, respectively.

The mammalian BBB is central to the CNS health and functions. It is an evolutionary milestone for overall brain fitness, yet in same time became a major research challenge, as our understanding of how the highly specialized BBB structure is established during development and how different brain vascular cell types orchestrate together to support brain functions remain limited¹⁰. This obstacle can be partly attributed to the lack of specific markers and genetic tools that can separate the cerebrovascular cells from the peripheral ones. For example, current murine genetic tools of vascular cells are often based on the well-known endothelial cell markers such as podocalyxin (Podxl), claudin 5 (Cldn5), tyrosine kinase (Tek/Tie2) and tyrosine kinase receptor (Kdr), which all tend to express throughout the body. Therefore, transgenic mouse models based on these alleles, including Tek-Cre and Tek-Cre/ERT2, Cdh5-Cre and Cdh5-Cre/ERT2, Kdr-Cre, all exhibit limitations and constraints when applied to brain vasculature³¹. Recently, we used single cell transcriptomics and identified a brain pericyte specific marker — Atp13a5, and successfully generated a knock-in mouse model for the brain pericyte²¹. With similar approaches, we further identified Zic3 as a brain endothelial specific marker from single cell transcriptomics, and confirmed its expression with both in situ hybridization and antibody staining.

The finding of Zic3 markers is very meaningful, as it guided us to generate a new transgenic model — Zic3-T2A-tdTomato-IRES-CreERT2. Based on the Zic3-tdTomato expression and the corresponding CreER recombinase activity after tamoxifen induction, we further confirmed that the Zic3-T2A-tdTomato-IRES-CreERT2 model is specific to the endothelial cells in the CNS, and not for peripheral cells. For peripheral endothelial cells, we also generated a new Plvap-T2A-EGFP-IRES-CreERT2 knock-in model, and the characterization of Plvap-EGFP reporter and corresponding CreER recombinase activity showed it targets the endothelial cells of the peripheral organs such as lung, kidney, liver, etc., as well as the circumventricular organs in the brain which are known to be outside of the BBB. Therefore, the Zic3 and Plvap genetic tools are sufficient to separate and selectively target the CNS and peripheral endothelial cells, respectively. Interestingly, the presence of Plvap in choroid plexus and meningeal vessels also indicates that the heterogeneity of cerebrovascular endothelial cells, at least between the BBB and non-BBB areas, which could be potentially probed with the genetic tools we have generated.

There is a list of receptors and transporters known to be highly enriched at the BBB endothelium, including lysolipid transporter Mfsd2a, glucose transporter Slc2a1 and organic anion transporter Slco1c1. Their candidacy for generating genetic models for brain or BBB endothelial cells were proposed, but none of them are specific enough, as shown by single cell transcriptomics (e.g. Extended Data Fig. 1). Based on our data from RNAscope, immunostaining and reporter models, Zic3 is indeed a CNS endothelial specific genetic marker. Zic3 encodes a highly conserved C2H2 zinc finger domain protein and a member of the GLI superfamily of transcription factors³². Its mutations are found in patients with X-linked heterotaxy, and its deficiency in mice disrupts the left–right asymmetry of cardiovascular development and results in neural tube deficits³³. While ZIC3 plays an important role in early development, it is still surprising to see its expression remained in the BBB endothelial cells even in the adult brain. In human endothelial cells derived from pluripotent stem cells, ZIC3 can act downstream of Wnt signaling and promote the endothelial barrier functions³⁴. Ectopic expression of Zic3 in human umbilical vein endothelial cells could also induce BBB marker gene expression³⁵. Therefore, it is important to determine if Zic3 is a master regulator of an intrinsic program that maintains the BBB functions in future studies. Interestingly, Zic3 is also located on X-chromosome, and may contribute to the sex-difference in the cerebrovascular functions associated with aging and Alzheimer’s disease³⁶.

The cerebrovascular development is guided by some of the classic growth factor pathways, such as VEGF, Wnt and Notch⁸. Take Wnt signaling as an example, Wnt7a/7b and Norrin³⁷, Frizzled receptor³⁸, co-receptors low-density lipoprotein receptor-related protein (LRP) 5 and 6³⁹ and co-activator orphan G-protein coupled receptor Gpr124⁴⁰ are all indispensable the proper development of the BBB. To understand the vascular heterogeneity, we analyzed current single cell transcriptomics gathered from different organs. Besides Zic3, Plvap and Atp13a5, we identified a list of genetic markers that can potentially be utilized to develop next-generation tools. For example, the transcriptomic comparison between CNS Zic3 + endothelial cells with Plvap + peripheral ones pointed to hundreds of DEGs that changed preferentially in BBB endothelial cells (Fig. 5). These genes are linked to interesting pathways associated with the establishment of the blood-brain barrier, brain development, lipid metabolism, transport systems and tight junction signaling. While Zic3 has a preference for CNS endothelial cells, Clec4g, Igfbp5 and Stab2 may potentially be utilized to target endothelial cells of the liver, kidney and spleen, respectively, and worth further investigation. Nevertheless, this endothelial heterogeneity is proof that ECs are heavily influenced by the local environment of each individual organs.

Last but not the least, BBB dysfunctions with endothelial activation is commonly found in a spectrum of CNS disorders, including AD and other dementia¹⁰. Therefore, the identification of Zic3 as a BBB endothelial cells marker will likely advance our studies towards a better understanding of BBB vascular cell changes in aging and CNS diseases, including vascular contributions to AD and dementia. Beyond the application of these transgenic models, these specific markers can be used to develop other tools, such as BBB cell-type specific antibodies and new viral vector designs to target BBB endothelial cells more specifically with artificial promoters based on tissue and cell-type specific enhancers⁴¹.

Acknowledgement

We thank Dr. Fan Gao for his valuable bioinformatics support and insightful discussions that greatly contributed to the completion of this study. We thank the support National Institutes of Health (NIH) grant nos. R01AG061288, RF1NS122060, RF1NS135617, and R21AG085559, BrightFocus Foundation grant no. A2019218S, and Alzheimer’s Association grant no. AA-ADP 23-1051406.

Daneman R, Prat A (2015) The Blood–Brain Barrier. Cold Spring Harb Perspect Biol 7:a020412
Zlokovic BV (2008) The blood-brain barrier in health and chronic neurodegenerative disorders. Neuron 57:178–201
Zlokovic BV (2011) Neurovascular pathways to neurodegeneration in Alzheimer’s disease and other disorders. Nat Rev Neurosci 12:723–738
Félétou M (2011) The Endothelium: Part 1: Multiple Functions of the Endothelial Cells—Focus on Endothelium-Derived Vasoactive Mediators. Morgan & Claypool Life Sciences, San Rafael (CA)
Obermeier B, Daneman R, Ransohoff RM (2013) Development, maintenance and disruption of the blood-brain barrier. Nat Med 19:1584–1596
Liebner S (2008) Wnt/[beta]-catenin signaling controls development of the blood-brain barrier. J Cell Biol 183:409–417
Xin X et al (2001) Hepatocyte growth factor enhances vascular endothelial growth factor-induced angiogenesis in vitro and in vivo. Am J Pathol 158:1111–1120
Zhao Z, Nelson AR, Betsholtz C, Zlokovic BV (2015) Establishment and Dysfunction of the Blood-Brain Barrier. Cell 163:1064–1078
Ayloo S, Gu C (2019) Transcytosis at the blood–brain barrier. Curr Opin Neurobiol 57:32–38
Sweeney MD, Zhao Z, Montagne A, Nelson AR, Zlokovic BV (2019) Blood-Brain Barrier: From Physiology to Disease and Back. Physiol Rev 99:21–78
Ben-Zvi A, Lacoste B, Kur E, Andreone B (2014) MSFD2A is critical for the formation and function of the blood brain barrier
Augustin HG, Koh GY (2017) Organotypic vasculature: From descriptive heterogeneity to functional pathophysiology. Science 357:eaal2379
Denzer L, Muranyi W, Schroten H, Schwerk C (2023) The role of PLVAP in endothelial cells. Cell Tissue Res. 10.1007/s00441-023-03741-1
Saunders NR et al (2014) The rights and wrongs of blood-brain barrier permeability studies: a walk through 100 years of history. Front Neurosci 8:404
Goldmann EE (1913) Vitalfärbung am Zentralnervensystem: Beitrag zur Physio-Pathologie des Plexus chorioideus und der Hirnhäute. Königl. Akademie der Wissenschaften
Daneman R, Prat A (2015) The Blood–Brain Barrier. Cold Spring Harb Perspect Biol 7:a020412
Faget L, Hnasko TS (2015) Tyramide Signal Amplification for Immunofluorescent Enhancement. In: Hnasko R vol (ed) ELISA, vol 1318. Springer New York, New York, NY, pp 161–172
Vanlandewijck M et al (2018) A molecular atlas of cell types and zonation in the brain vasculature. Nature 554:475–480
Kalucka J et al (2020) Single-Cell Transcriptome Atlas of Murine Endothelial Cells. Cell 180:764–779e20
Muhl L et al (2020) Single-cell analysis uncovers fibroblast heterogeneity and criteria for fibroblast and mural cell identification and discrimination. Nat Commun 11:3953
Guo X et al (2024) Atp13a5 Marker Reveals Pericyte Specification in the Mouse Central Nervous System. J Neurosci 44:e0727242024
Hupe M et al (2017) Gene expression profiles of brain endothelial cells during embryonic development at bulk and single-cell levels. Sci Signal 10:eaag2476
Tabula Muris Consortium (2018) Single-cell transcriptomics of 20 mouse organs creates a Tabula Muris. Nature 562:367–372
Yao Z et al (2021) A transcriptomic and epigenomic cell atlas of the mouse primary motor cortex. Nature 598:103–110
Ben-Zvi A et al (2014) Mfsd2a is critical for the formation and function of the blood–brain barrier. Nature 509:507–511
Winkler EA et al (2015) GLUT1 reductions exacerbate Alzheimer’s disease vasculo-neuronal dysfunction and degeneration. Nat Neurosci 18:521–530
Madisen L et al (2010) A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nat Neurosci 13:133–140
Kaplan L, Chow BW, Gu C (2020) Neuronal regulation of the blood–brain barrier and neurovascular coupling. Nat Rev Neurosci 21:416–432
Russo MV, Latour LL, McGavern DB (2018) Distinct myeloid cell subsets promote meningeal remodeling and vascular repair after mild traumatic brain injury. Nat Immunol 19:442–452
Wu Y et al (2021) Mild traumatic brain injury induces microvascular injury and accelerates Alzheimer-like pathogenesis in mice. acta neuropathol commun 9:74
Payne S, De Val S, Neal A (2018) Endothelial-Specific Cre Mouse Models: Is Your Cre CREdibile? ATVB 38, 2550–2561
Gebbia M et al (1997) X-linked situs abnormalities result from mutations in ZIC3. Nat Genet. 10.1038/NG1197-305
Zipfel WR et al (2003) Live tissue intrinsic emission microscopy using multiphoton-excited native fluorescence and second harmonic generation. Proceedings of the National Academy of Sciences 100, 7075–7080
Gastfriend BD et al (2021) Wnt signaling mediates acquisition of blood–brain barrier properties in naïve endothelium derived from human pluripotent stem cells. eLife 10:e70992
Hupe M et al (2017) Gene expression profiles of brain endothelial cells during embryonic development at bulk and single-cell levels. Sci Signal 10:eaag2476
Zhu D, Montagne A, Zhao Z (2021) Alzheimer’s pathogenic mechanisms and underlying sex difference. Cell Mol Life Sci 78:4907–4920
Zhou Y et al (2014) Canonical WNT signaling components in vascular development and barrier formation. J Clin Invest 124:3825–3846
Wang Y et al (2012) Norrin/Frizzled4 signaling in retinal vascular development and blood brain barrier plasticity. Cell 151:1332–1344
Chen J et al (2012) Retinal expression of Wnt-pathway mediated genes in low-density lipoprotein receptor-related protein 5 (Lrp5) knockout mice. PLoS ONE 7:e30203
Kuhnert F et al (2010) Essential regulation of CNS angiogenesis by the orphan G protein-coupled receptor GPR124. Science 330:985–989
Jüttner J et al (2019) Targeting neuronal and glial cell types with synthetic promoter AAVs in mice, non-human primates and humans. Nat Neurosci 22:1345–1356
Wu Y et al (2021) Mild traumatic brain injury induces microvascular injury and accelerates Alzheimer-like pathogenesis in mice. Acta Neuropathol Commun 9:74
Malong L et al (2023) Characterization of the structure and control of the blood-nerve barrier identifies avenues for therapeutic delivery. Dev Cell 58:174–191e8
Nikolakopoulou AM et al (2019) Pericyte loss leads to circulatory failure and pleiotrophin depletion causing neuron loss. Nat Neurosci 22:1089–1098

KEY RESOURCES TABLE

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Genomic data sets
Mouse brain vasculature	Gene Expression Omnibus	GSE98816 GSE99058
Atlas of Murine Endothelial Cells	Array Express	E-MTAB-8077
Fibroblast and mural cell in muscular organs	Gene Expression Omnibus	GSE150294
Tabula Muris	Gene Expression Omnibus	GSE109774
Single cell RNA-sequencing data processing
Seurat Package (v.3.1.5)	R	v.3.6.2
FindVariableFeatures function	R	v.3.6.2
FindAllMarkers function	R	v.3.6.2
FindMarkers function	R	v.3.6.2
Principal component analysis (PCA)	R	v.3.6.2
Uniform Manifold Approximation and Projection (UMAP)	R	v.3.6.2
Shared Nearest Neighbor (SNN) clustering	R	v.3.6.2
Experimental Models: Organisms/Strains
Zic3-tdTomato-CreERT2	This study
Plvap-tdTomato-CreERT2	This study
*DNA isolation and genotyping*
GoTaq Green Master Mix	Promega	M7122
Zic3-tdTomato-CreERT2 forward primer	AATGGCTCTCCTCAAGCGTATTC
Zic3-tdTomato-CreERT2 reverse primer	GTTATTCAACTTGCACCATGCCG
Plvap-EGFP-CreERT2 forward primer	GCTGTGTAGCAGAGACAAACCTTA
Plvap-EGFP-CreERT2 reverse primer	GGTGGTGCAGATGAACTTCAGG
Antibodies
Goat polyclonal anti-CD13	R&D systems	AF2335
Rat monoclonal anti- CD31	BD Pharmingen	550274
Rabbit polyclonal anti-ZIC3	Thermo Fisher Scientific	PA5-29073
Rat monoclonal anti-VCAM1	Millipore Sigma	CBL1300
Mouse monoclonal anti-SMA	Millipore Sigma	A5228
Rabbit polyclonal anti-Iba-1	Wako	019-19741
Rabbit polyclonal anti-GFAP	Dako	z0334
Rabbit polyclonal anti-NeuN	Millipore Sigma	ABN78
Rabbit anti-mouse Olig2	Millipore Sigma	AB9610
Dylight 488-conjugated L. esculentum lectin	Thermo Fisher Scientific	L32470
Dylight 649-conjugated L. esculentum lectin	Thermo Fisher Scientific	L32472
Immunohistochemistry
Vibratome	Leica	VT1200
Cryostat	Leica	CM3050 S
Normal Donkey Serum	Jackson ImmunoResearch	AB_2337258
VECTASTAIN ABC universal kit	Vector laboratories	PK6200
ImmPRESS Universal Polymer kit	Vector laboratories	MP-7500
*in situ* hybridization
SuperFrost Plus Micro Slide	VWR	48311-703
Zic3 RNAscope probe	Advanced Cell Diagnostics	Cat. #480351
Positive control RNAscope probe	Advanced Cell Diagnostics	Cat. #320881
Negative control RNAscope probe	Advanced Cell Diagnostics	Cat. #320871
HybEZ Oven	Advanced Cell Diagnostics
RNAscope® 2.5 HD Reagent Kit-RED	Advanced Cell Diagnostics	322350
RNAscope® Multiplex Fluorescent V2 Assay	Advanced Cell Diagnostics	323100
Nikon A1R MP+ confocal/ multiphoton microscope	Nikon	Nikon A1R
Revolve 4 brightfield and fluorescence microscope	Echo	RVL-100-G
Software
GraphPad Prism	GraphPad Software	GraphPad Prism 8
Metascape	http://metascape.org
ClueGO		2.5.4
Cytoscape		3.8.1
Image J	NIH

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and request for resources and reagents should be directed to and will be fulfilled by Lead Contact Zhen Zhao ([email protected]).

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Animals

Mice were housed in plastic cages on a 12 h light/dark cycle with access to water ad libitum and a standard laboratory diet. All procedures were approved by the Institutional Animal Care and Use Committee at the University of Southern California and followed National Institutes of Health guidelines. All animals were included in the study. Male and female animals of 2–3 months of age were used in the experiments. All animals were randomized for their genotype information. All experiments were carried out blind: the operators responsible for the experimental procedures and data analysis were blinded and unaware of group allocation throughout the experiments. For all experiments, at least three independent mice were analyzed, which included both sexes and no apparent sex difference were observed.

Generation of the Zic3-tdTomato-CreERT2 knock-in model

To generate Zic3-T2A-tdTomato-IRES-CreERT2 knock-in mouse, donor DNA templates encoding self-cleaving T2A peptide, tdTomato, internal ribosome entry site and CreERT2 were synthesized. These sequences were flanked by 375bp sequences and 4606bp sequences homologous to the third exon and 3’ UTR region of Zic3 gene. Next, these donor vector containing the T2A-tdTomato-IRES-CreERT2 cassette, and gRNA (TTTAACGAATGGTACGTCTGAGG) were co-injected into fertilized eggs to generate targeted conditional knock-in offspring. The F0 founder animals were genotyped by PCR and sequence analysis, and four F1 mice were generated and further confirmed with southern blotting for both 5’ arm and 3’ arm insertion sequences.

Generation of the Plvap-EGFP-CreERT2 knock-in model

We generated the Plvap-T2A-EGFP-IRES-CreERT2 knock-in mouse (Fig. 4b), with a donor DNA template encoding self-cleaving T2A peptide, EGFP, internal ribosome entry site, CreERT2. These sequences were flanked by 405bp sequences and 3834bp sequences homologous to the 6^thexon and 3’ UTR region of Plvap gene. These donor vector containing the T2A-EGFP-IRES-CreERT2 cassette, and gRNA (GCAGCTGGGTCCTCAACCGCTGG) were co-injected into fertilized eggs to generate targeted conditional knock-in offspring. The F0 founder animals were genotyped by PCR and sequence analysis, and three F1 mice were generated and further confirmed with southern blotting for both 5’ arm and 3’ arm insertion sequences.

Mild Traumatic Brain Injury model

To induce mild traumatic brain injury (mTBI) in mice, we followed a previously described protocol⁴². Briefly, we used the KOPF stereotaxic system to position the mouse’s head under the impactor at a specific angle, targeting a point 2 mm posterior and 2.5 mm lateral to Bregma. A 4 mm flat plastic tip (RWD Life Science) was used to deliver a controlled impact using a brain injury device (RWD #68099). Mice were anesthetized with ketamine and xylazine (90 mg/kg and 9 mg/kg, i.p.). After exposing the skull, we delivered an impact at a velocity of 3 m/s, a depth of 1 mm, and a duration of 180 milliseconds. Mice were then placed in warmed cages to recover.

METHOD DETAILS

Bioinformatics

ScRNA-seq data for mouse brain vasculature and multiple organs

For scRNA-seq integration dataset, we obtained the cell count matrix from Gene Expression Omnibus (GEO) with the series record GSE98816, GSE150294 and Array express E-MTAB-8077, and did secondary analysis after integration. Then we performed secondary analysis of 3 single cell sequencing (scRNA-seq) datasets on brain and peripheral vascular cells, 48,526 single-cell transcriptomes were collected using R Seurat Package.

For scRNA-seq dataset for mouse brain vasculature, we obtained the cell count matrix from GEO with the series record GSE98816 and GSE99058¹⁸. The data represent the expression levels of 18435 genes in 3186 cells. The mouse brain tissue was harvested for Smart-seq2 and sequencing was performed on a HiSeq2500 at the National Genomics Infrastructure (NGI), Science for Life Laboratory, Sweden, with single 50-bp reads (dual indexing reads).

For scRNA-seq dataset for multiple organs, we obtained the cell count matrix from GEO with the series record GSE109774²³. The data represent the expression levels of 23433 genes in 53760 cells. All organs were single-cell-sorted into plated using flurescence-activated cell sorting. Libraries were sequenced on the NovaSeq 6000 Sequencing System (Illumina) using 2 x 100-bp paired-end reads.

ScRNA-seq data preprocessing

The data processing of the scRNA-seq data were performed with the Seurat Package (v.3.1.5) in R (v.3.6.2). The basic scRNA-seq analysis was run using the pipeline provided by Seurat Tutorial (https://satijalab.org/seurat/v3.0/immune_alignment.html) as of June 24, 2019. In general, we set up the Seurat objects from different groups in experiments for normalizing the count data present in the assay. This achieves log-normalization of all datasets with a size factor of 10,000 transcript per cell. For different Seurat objects, FindVariableFeatures() function was used to identify outlier genes on a ‘mean variability plot’ for each object. The nFeatures parameter is 2000 as the default for the selection method called ‘vst’. These resulted genes serve to illustrate priority for further analysis.

Data processing

The dataset on all cells were used to scale and center the genes. First of all, principal component analysis (PCA) was used for linear dimensionality reduction with default computes the top 30 principal components. By applying the JackStraw() function, JackStrawPlot() function and ElbowPlot() function, we identified the principal components for further analysis. Then, PCA results were used as the input for the Uniform Manifold Approximation and Projection (UMAP) dimensional reduction.

We identified clusters of cells by a shared nearest neighbor (SNN) modularity optimization-based clustering algorithm. The algorithm first calculated k-nearest neighbors and computed the k-NN graph, and then optimizes the modularity function to determine clusters.

Determination of cell-type identity

To determine the cell type, we used FindAllMarkers() function with parameters min.pct and thresh.use set to 0.25 to find markers in each cluster and known marker genes that have been previously reported could be used to determine cell-type identity. These include, but are not limited to Snap25 for Neuron, Cldn10 for Astrocyte, Mbp for Oligodendrocyte, Cldn5 for EC, Kcnj8 for PC, Acta2 for VSMC, Ctss for microglial, Col1a1 for Fibroblast-like cell.

Pathway analysis and visualization by Metascape, ClueGO and Cytoscape

Using the Metascape online tool (http://metascape.org), we performed functional enrichment analysis of ZIC3-positive enriched genes. Enrichment of pathways from KEGG, GO Biological Process and GO Molecular Function was analyzed by Metascape. The terms with P-value < 0.01, minimum counts of 3, and enrichment factors of > 1.5 would be considered. The ClueGO Cytoscape pluin 2.5.4 and Cytoscape version 3.8.1 will be used for secondary KEGG pathway analysis and network visualization.

Cellular Biology Related Procedures

HPR injection and Lectin injection and mapping

The HRP solution was prepared by dissolving 125 mg (0.125 g) of HRP Type II (Sigma, P8250) in 2.5 ml of PBS, yielding a concentration of 0.5 mg/10 µl. Each animal was injected with a single dose of 0.5 mg/g of body weight and harvested 2 hours later. The mice were then sacrificed and perfused with PBS and PFA at 5 min after lectin (ThermoFisher Scientific, #L32470) injection. To visualize HRP in the injected samples under light microscopy, the samples were washed with PBS and incubated in Tris buffer containing 0.1% tyramide reagent and 0.0015% H₂O₂ for 10 minutes at room temperature, in the dark⁴³. All sections were then scanned using the Li-Cor Odyssey Dlx at a resolution of 21 µm, or on a NikonTi2 confocal microscope. The ImageJ plugin ‘Neuro J’ length analysis tool was used to measure the length of lectin-positive or HRP-positive endothelial capillary profiles. The capillary length was quantified and expressed as mm of lectin+ endothelial capillary profiles per mm2 of brain tissue. The HRP-occupied vascular area ratio was calculated by measuring the HRP-positive signal within lectin-positive regions.

Fluorescence in situ hybridization

Fluorescence in situ hybridization was performed using the RNAscope technology (Advanced Cell Diagnostics, Hayward, CA). Tissue sample preparation and pretreatment were performed on fixed brains cut into 15 µm sections mounted onto SuperFrost Plus glass slides following the manufacturer’s protocol (ACD documents 323100). After dehydration and pretreatment, slides were subjected to RNAscope Multiplex Fluorescent Assay (ACD documents 323100). RNAscope probes for mouse Zic3, positive control and negative control were hybridized for 2h at 40ºC in the HybEZ Oven and the remainder of the assay protocol was implements. Subsequently, the slides were subjected to immunohistochemistry. The fluorescent signal emanating from RNA probes and antibodies was visualized and captured using a Nikon AIR MP+ confocal/ multiphoton microscope (Nikon). All FISH images presented are projection of 10-image stacks (0.5 µm intervals) obtained from cerebral cortex, and a smoothing algorithm was applied during image post-processing (Nikon NIS-Elements Software).

Immunohistochemistry

Animals were anesthetized, perfused and brains were removed and postfixed as we described previously⁴⁴. Brain, spinal cord, kidney, liver, and heart tissue were also collected, postfixed and cut at 35 µm thickness using a vibratome (Leica). After that, sections were blocked with 5% normal donkey serum (Vector Laboratories) and 0.1% Triton-X in 0.01M PBS and incubated with primary antibodies diluted in blocking solution overnight at 4ºC. The primary antibody information is as following: Goat anti-mouse aminopeptidase N/ANPEP (CD13; R&D systems; AF2335; 1:100), ZIC3 polyclonal antibody (Invitrogen; PA5-29073; 1:100), Rat anti-mouse vascular adhesion molecule (VCAM1; MilliporeSigma; CBL1300; 1:200), Mouse anti-α-smooth muscle actin (SMA, MilliporeSigma; A5228, 1:200), Rabbit anti-mouse ionized calcium binding adaptor molecule 1 (Iba-1; Wako, 019-19741; 1:200), Rabbit anti-Glial Fibrillary Acidic Protein (GFAP; Dako, z0334; 1:500), Rabbit anti-mouse NeuN (Millipore, ABN78, 1:500). To visualize brain microvessels, sections were incubated with Dylight 488 or 647-conjugated L. esculentum lectin as we have described previously⁴⁴. After incubation with primary antibodies, sections were washed with PBS for three times and incubated with fluorophore-conjugated secondary antibodies. Sections were imaged with a Nikon AIR MP+ confocal/ multiphoton microscope (Nikon). Z-stack projections and pseudo-coloring were performed using Nikon NIS-Elements Software. Image post analysis was performed using ImageJ software.

Molecular Biology Related Procedures

DNA isolation and genotyping

Mouse genomic DNA was isolated from tail biopsies (2 - 5 mm) and following overnight digestion at 56  into 100 μL of tail digestion buffer containing 10 mM Tris-HCl (pH 9.0), 50 mM KCl, 0.1% Triton X-100 and 0.4 mg/mL Proteinase K. Next, the tail will be incubated at 98 for 13 minutes to denature the Proteinase K. After centrifugation at 12000 rpm for 15 min, the supernatants were collected for PCR. The primers details are listed in the KEY RESOURCES TABLE. The PCR conditions were as follows: 1) 94 °C for 3 min; 2) 35 cycles at 94 °C for 30 sec, 60 °C for 30 sec, and 72 °C for 35 sec; 3) 72 °C for 5 min. PCR products were separated on 2% agarose gel.

Quantification and statistical analysis

Sample sizes were calculated using nQUERY, assuming a two-side alpha-level of 0.05, 80% power and homogeneous variances for the 2 samples to be compared, with the means and SEM for different parameters predicted from pilot study. All the data are presented as mean ± SEM as indicated in the figure legends and were analyzed by GraphPad Prism 8. For multiple comparisons, Bartlett’s test for equal variances was used to determine the variances between the multiple groups and one-way analysis of variance (ANOVA) followed by Tukey test was used to test statistical significance, using GraphPad Prism 8 software. A P value of less than 0.05 was considered statistically significant.

No competing interests reported.

ExtendedFigures.docx

New genetic tools for central and peripheral vascular endothelia

Status:

Version 1

Abstract

Figures

Main

Results

Discussion

Declarations

References

Methods

Additional Declarations

Supplementary Files

Status:

Version 1