Identification of a Core Set of 58 Gene Transcripts With Broad and Specific Expression in the Microvasculature
Objective— Pathological angiogenesis is an integral component of many diseases. Antiangiogenesis and vascular targeting are therefore promising new therapeutic principles. However, few endothelial-specific putative drug targets have been identified, and information is still limited about endothelial-specific molecular processes. Here we aimed at determining the endothelial cell-specific core transcriptome in vivo.
Methods and Results— Analysis of publicly available microarray data identified a mixed vascular/lung cluster of 132 genes that correlated with known endothelial markers. Filtering against kidney glomerular/nonglomerular and brain vascular/nonvascular microarray profiles separated contaminating lung markers, leaving 58 genes with broad and specific microvascular expression. More than half of these have not previously been linked to endothelial functions or studied in detail before. The endothelial cell-specific expression of a selected subset of these, Eltd1, Gpr116, Ramp2, Slc9a3r2, Slc43a3, Rasip1, and NM_023516, was confirmed by real-time quantitative polymerase chain reaction and/or immunohistochemistry.
Conclusions— We have used a combination of publicly available and own microarray data to identify 58 gene transcripts with broad yet specific expression in microvascular endothelium. Most of these have unknown functions, but many of them are predicted to be cell surface expressed or implicated in cell signaling processes and should therefore be explored as putative microvascular drug targets.
Angiogenesis is integral to many pathological processes and is currently targeted therapeutically by vascular endothelial growth factor (VEGF-A) inhibitors in cancer and age-dependent macula degeneration (AMD).1–3⇓⇓ However, VEGF-A plays pivotal roles during blood vessel formation and homeostasis,4,5⇓ and several side effects of anti-VEGF therapy have been reported.2,6,7⇓⇓ Some tumor types are also relatively insensitive to VEGF-A inhibition. Other antiangiogenic targets have been tested, but human trials have been negative so far despite promising preclinical data. Recently, Delta-like 4 (DLL4) was identified as a promising new antiangiogenic target.8–13⇓⇓⇓⇓⇓ Like VEGF-A, DLL4 has a critical and dose-dependent function during vascular development. Both VEGF-A and DLL4 elicit signaling cascades via receptors on endothelial cells. A number of other endothelial cell-specific proteins have been identified, but their usefulness as antiangiogenic targets remains to be determined. The VEGF-A and DLL4 examples argue that additional microvascular targets should be sought in signaling pathways that are specific to endothelial cells. Considering this, surprisingly few published attempts have been made to determine endothelial cell-specific transcriptomes. Thus far, published studies have analyzed the transcriptomes of cultured endothelial cells,14–16⇓⇓ or endothelial cells from normal or pathological tissues acquired by laser capture17,18⇓ or cell sorting,19–22⇓⇓⇓ but only 1 study investigated endothelial profiles in normal tissues in comparison with corresponding nonvascular cells.20
Here, we aimed at determining the mammalian microvascular core transcriptome in vivo. Using several sources of transcriptional profiling data obtained from organs, tissues, and tissue fragments with different microvascular content, we identify a core set of 58 gene transcripts with specific yet broad expression in the microvasculature. This set includes most known and currently used endothelial markers, but also a surprisingly large number of gene products that have not been studied in detail or linked to endothelial functions before.
An expanded methods section is provided as a data supplement (available online at http://atvb.ahajournals.org).
The GNF Symatlas microarray data set covering 60 normal mouse tissues was downloaded from the Novartis Foundation.23 Affymetrix target sequences were matched against ENSEMBL transcript and gene sequences using blast.24 In cases where multiple probe sets were annotated to the same gene signals were averaged. Correlation searches were performed using Kdr (Kinase insert Domain protein Receptor, VEGFR2, Flk-1), Cdh5 (Cadherin-5, VE-cadherin), and Pecam1 (Platelet Endothelial Cell Adhesion Molecule, CD31). Transcripts with Pearson correlation less than 0.75 to any of the baits were excluded. The resulting cluster of lung/microvascular transcripts was segregated using transcriptional profiles of kidney glomeruli.25 Briefly, differential expression in the glomerulus versus rest of the kidney (ROK) was assessed by paired t test. Glomeruli versus ROK expression ratios were determined by calculating the linear average of the paired ratios, which were transformed into log2 ratios. Probability values were obtained by paired t test. Transcripts with log2 ratio ≥2 and probability value ≤0.0001 were considered upregulated. Transcripts that were not upregulated in the glomeruli were removed, leaving 71 transcripts.
Isolation of Microvascular Fragments, RTQ-PCR, and Transcriptional Profiling
Microvascular fragments were isolated from platelet-derived growth factor-B (pdgfb) +/+ and −/−27 E18.5 embryos on NMRI background and from adult mice on C57Bl6/129sv mixed background as described elsewhere.26 Total RNA samples were obtained from individual adults or embryos. Real-time quantitative PCR (RTQ-PCR) was performed using a 7300 instrument (Applied Biosystems) and standard protocols and reagents. Three RNA samples from E18.5 brain microvascular fragments, 3 corresponding rest-of-brain samples, and 3 samples from adult brain microvascular fragments were individually hybridized to Affymetrix GeneChip Mouse Genome 430 2.0 microarrays. Array data was processed using the Affy and gcrma packages in the Bioconductor project (http://www.bioconductor.org). Student t test was used to evaluate differential expression. The false discovery rate method was used to perform multiple test correction for probability values.28 The log2 change was calculated as the log2 of the average expression difference between E18.5 microvessels and “rest-of-brain” tissue or adult brain microvessels. Our microarray data have been deposited in NCBIs Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/) and are accessible through GEO Series accession number GSE9202.
Human normal tissue microarrays (MaxArray Human Normal Tissue MicroArray Slides, Zymed/Invitrogen and Human Normal Organ Tissue Microarray, US Biomax/BioKat) were stained with rabbit-α-Slc9a3r2 polyclonal antibody (The Swedish Human Proteome Resource (HPR) Center) and fluorescein isothiocyanate (FITC)-conjugated mouse-α-ASMA (α-smooth muscle actin) monoclonal antibody (Abcam).
Correlation Searches With Kdr, Cdh5, and Pecam1 Identify a Cluster of Microvascular/Lung-Specific Gene Transcripts
We searched for transcripts with expression profiles similar to those of 3 well-established vascular endothelial-specific markers (baits) Kdr (Vegfr2), Cdh5 (VE-cadherin), and Pecam1 (Cd31) across a broad collection of Affymetrix data (Symatlas; see data supplement) representing different organs and cell types (Figure 1A through 1C). We found that transcripts for known broad vascular endothelial markers correlated with 1 or more of the 3 baits, with Pearson correlation values ≥0.75. Examples of such markers include Tie1, Robo4, Eng, Epas1, Notch4, Esam1, and EphB4 (red arrows in Figure 1A through 1C). The correlating transcripts also included several known markers for lung epithelium, such as Surfactant protein (Sftp) b, -c, and –d (blue arrows in Figure 1A through 1C). Known vascular markers, similar to known lung markers, showed a dominant peak of expression in lung (Figure 2A). Most likely, this result reflects a significantly higher proportion of endothelial cells in lung (approximately 50%) than in most other organs (5% or less). This is also illustrated by immunostanings of the endothelial proteins PECAM1, Claudin-5, Endoglin, and ICAM1 (supplemental Figure I). We therefore assumed that the transcripts with Pearson correlation ≥0.75 to any of the endothelial baits represented a mixed cluster containing novel markers for endothelium and lung epithelium (eg, Gpr116; Figure 2A). The lung/vascular cluster consisted of 132 gene transcripts of which the majority (104 transcripts) was common to the three baits (Figure 2B; supplemental Table III).
Segregation of the Lung/Microvascular Transcript Cluster
Overall, the renal vascular density is low compared to the lung. Accordingly, vascular markers showed modest peaks in whole kidney RNA (Figure 2A). Kidney glomeruli, however, have a similarly high density of endothelial cells as the lung (approximately 50%),29 but glomeruli represent less than 5% of the total kidney tissue. Glomeruli also contain a population of specialized epithelial cells—podocytes—that are distinct from the lung epithelium. We therefore took advantage of Affymetrix transcript profiles of glomerular and nonglomerular kidney tissues25 to segregate the lung/vascular cluster into putative vascular (glomerulus-enriched) and lung (not glomerulus-enriched) markers (Figure 3A). By applying a log2 ratio=2 threshold for glomerular enrichment, the lung/vascular cluster segregated into 2 parts: 1 containing all well-established vascular markers (log2 ratio ≥2), and the other (log2 ratio <2) containing all known lung epithelial markers (Figure 3B through 3D). The former contained 71 gene transcripts (supplemental Table II and supplemental Methods), of which 32 are known to be expressed by endothelial cells. For the remaining 39 we have not been able to find published literature on their vascular-specific expression or function, although for some of them expression in cultured endothelial cells, or circumstantial evidence for a role in vascular biology, may have been demonstrated.
Expression in Brain Microvessels and Developmental Regulation of the 71-Gene Cluster
We predicted that most of the genes in the 71-gene cluster should have broad but specific expression in microvessels. However, our data did not exclude that part of the cluster might represent (1) nonvascular transcripts common to lung and glomeruli, (2) vascular transcripts commonly specific to lung and glomeruli, or (3) vascular-specific transcripts that are adult-specific (because Symatlas represents mainly adult tissues). To distinguish between these possibilities we assessed the 71-gene cluster by Affymetrix analysis of microvessels isolated from a third tissue—embryonic brain. This analysis showed that 58 of the 71 genes were significantly overexpressed in E18.5 brain microvessels compared to “rest-of-brain”, and therefore likely represent broad, or general, microvascular markers (Figure 4A; supplemental Table III).
Among the 13 transcripts of the 71-cluster that were not validated in brain microvessels, Vegfa was the only one that showed significantly higher expression in “rest-of-brain” (Figure 4A; supplemental Table II). The presence of Vegfa within the 71-cluster likely reflects its strong expression in podocytes and lung epithelium, rather than specific expression in endothelial cells.5 Four additional transcripts with reported vascular function lacked brain microvessel enrichment: Numb, Itga3, Tcf21, and Hod (Figure 4A, left). Numb is expressed in neuronal cells.30 Itga3 and Tcf21 have roles in the kidney and respiratory system31,32⇓ and may be expressed in glomerular and lung epithelium, similar to Vegfa. Hod has been functionally implicated in developing heart and larger vessels,33 and may be blood vessel-specific outside of the brain.
To address potential developmental regulation, we compared the expression of the 71-cluster between E18.5 and adult brain microvascular fragments by Affymetrix analysis (Figure 4B). This revealed that 12 of the transcripts were overexpressed in embryonic vessels, whereas 10 were overexpressed in adult vessels. Two of the latter, Itga8 and Pscd3 (Figure 4), were not enriched in E18.5 brain microvessels compared to “rest-of-brain”. Itga8 is known to function in the kidney,34 so it may represent a nonvascular transcript common to lung and glomeruli. Pscd3, however, may be a candidate specific marker of mature microvessels.
In summary, analysis of the 71-cluster by brain microvessel profiling confirmed that 58 of them had a broad or general microvascular expression (the 58-gene microvascular cluster) and that 20 of the 58 (and an additional 2 in the 71-cluster) were developmentally regulated.
Validation by Real-Time Quantitative PCR
The large proportion of predicted novel microvessel markers (32/58) made it important to validate the microarray data. We therefore analyzed a subset of the 32 novel transcripts and 1 of the baits (Cdh5) by RTQ-PCR (Figure 5A). This analysis included genes that we found particularly interesting; the putative G protein-coupled receptors (GPCRs) EGF-latrophilin 7 transmembrane domain containing 1 (Eltd1; also called ETL or EGF-TM7-latrophilin-related protein) and GPCR 116 (Gpr116), the receptor activity modifying protein 2 (Ramp2), the putative transporter regulatory protein Slc9a3r2, the putative transporter Slc43a3, the Ras interacting protein 1 (Rasip1), and the hypoxia induced gene 2 (Hig2; NM_023516). The RTQ-PCR analysis showed that all of these were upregulated in E18.5 brain microvessels compared to rest-of-brain, most of them at fold ratios comparable to Cdh5 (Cdh5, 112-fold; Eltd1, 254-fold; Gpr116, 1187-fold; Ramp2, 114-fold; Slc9a3r2, 87-fold; Slc43a3, 15-fold; Rasip1, 53-fold; Hig2; NM_023516, 3-fold).
Expression in Pericyte-Deficient Microvessels
Microvessels contain endothelial cells and pericytes. However, we did not identify known pericyte markers within the 71-cluster, perhaps reflecting that such markers are not specific or uniformly expressed in pericytes.35 To get an indication of possible pericyte expression of the novel microvascular markers we analyzed brain microvessels from pericyte-deficient pdgfb null embryos.36,37⇓ We analyzed expression of Eltd1, Gpr116, Ramp2, Slc9a3r2, Slc43a3, Rasip1 and Hig2;NM_023516 in wild-type and pdgfb−/− microvessels by RTQ-PCR, expecting that putative pericyte markers would be significantly down-regulated in the latter.26,38⇓ As shown in Figure 5B, none of these genes behaved as a pericyte transcript. Hig2;NM_023516 was downregulated in pdgfb−/− microvessels (0.7 fold-change; log 2 ratio=−0.53) but at much lower than expected for a genuine pericyte marker26 Thus, Hig2;NM_023516 is either expressed in both endothelial cells and pericytes, or is positively regulated in endothelial cells by the presence of pericytes. Further experiments are required to distinguish between these 2 possibilities.
A polyclonal antiserum against 1 of the human candidate genes, Slc9a3r2, was used to stain human tissue arrays. The Slc9a3r2 antibodies generated a strong and specific endothelial expression pattern in all tissues tested (Figure 6 and data not shown).
Most widely used pan-vascular endothelial markers play pivotal roles in vascular development as demonstrated by gene targeting experiments. For example, knockouts of Kdr, Flt1, Tie1, Tie2, Cdh5, Cldn5, and Eng are all lethal during embryonic development.39–46⇓⇓⇓⇓⇓⇓⇓ These and several other genes and proteins with well-established roles in endothelial biology are present in the core cluster of 58 gene transcripts identified through this study. However, about 60% (32 transcripts) of the gene transcripts in this cluster have not been significantly (or at all) implicated in endothelial cell biology before. On average, the expression of these “novel” endothelial core transcripts was as “endothelial-specific” as the known endothelial markers, as judged by their ranking in the different datasets (Figure 4A and supplemental Tables I through III and supplemental Methods). The most important feature of our present study is therefore that it provides a significantly expanded view on the endothelial-specific core transcriptome, thereby opening new windows on endothelial biology. As an example, during the preparation of this manuscript the phenotype of Ramp2 null mice was reported, demonstrating an essential role for this coreceptor in the formation of the embryonic vasculature.47,48⇓ Further functional analysis of all 32 novel endothelial core transcripts will be a formidable task, but we have begun generating appropriate tools (eg, antibodies and knockout mice) for such work. The human proteome resource (HPR) systematically produces polyclonal antibodies and tissue expression patterns of all human proteins (http://www.proteinatlas.org/). Within its current version (3.1), the expression of several known (Pecam1, Claudin-5, Endoglin, Icam-1) and 1 of our novel proteins (Slc9a3r2) can be viewed. The staining patterns provided in the atlas together with the data reported here show broad endothelial-specific expression of these markers (Figure 6 and supplemental Figure I; data not shown; http://www.proteinatlas.org/). The data also show that Slc9a3r2 is strongly expressed in tumor endothelium (supplemental Figure II).
Our approach to the endothelial transcriptome is different from the one recently reported by Seaman et al,22 who compared endothelial gene expression in different tissues. The usefulness of such information for eg, tumor vessel-specific targeting, depends on whether the markers that distinguish different endothelial cells are indeed endothelial-specific or not. We aimed at finding transcripts that are common to all microvascular endothelial cells, but our data may be used to validate and further analyze the tissue-specific endothelial transcriptomes reported by others. Seaman et al22 describe 27 brain endothelial markers (BEMs). We found that 22 of these were significantly overexpressed in brain microvessels versus rest-of-brain (supplemental Figure IIIA), whereas 2 were overexpressed in rest-of-brain and 3 were equally expressed in the 2 tissues. Thus, most of the reported BEMs22 could be validated, but a few might be markers for brain rather than for brain endothelium. Similarly, of the 15 liver endothelial markers (LEMs),22 4 were overexpressed (at significant levels) in brain microvessels whereas 4 were overexpressed in rest-of-brain (supplemental Figure IIIB). Thus, some of the reported LEMs do not appear liver endothelial or even liver-specific. Of 12 reported angiogenic endothelial markers (AEMs),22 10 were upregulated in E18.5 brain microvessels (which are undergoing sprouting and remodeling) in comparison with adult brain vessels (which are quiescent). However, only one of the 12 AEMs was overexpressed in E18.5 microvessels whereas 4 were overexpressed in rest-of-brain (supplemental Figure IIIC and IIID). Thus, most of the AEMs are likely general rather than endothelium-specific markers of cell proliferation and tissue remodeling. Finally, among the 13 tumor endothelial markers (TEMs) identified,22 4 were overexpressed in E18.5 microvessels compared to rest-of-brain (supplemental Figure IIIE). These, together with an additional 3 TEMs, showed developmental regulation in brain microvessels. Thus, it appears that at least some of the reported TEMs22 mark developing brain vessels and may represent AEMs rather than TEMs (at least in brain).
Two of the novel endothelial markers identified here, Eltd1 and Gpr116, are predicted GPCRs for which currently no functional data are available. Eltd149 and Gpr11650 are both predicted to belong to the subfamily of adhesion-type GPCRs. Eltd1 has been suggested to be involved in cardiomyocyte differentiation and coronary angiogenesis based on expression in cardiomyocytes, bronchiolar smooth muscle cells (SMCs), and vascular SMCs in heart and lung.51 Intriguingly, we instead found Eltd1 mRNA to be a broad marker for vascular endothelial cells. Gpr116 (Q8BZJ9) has been reported to be expressed in lung, kidney, and placenta.52 Gpr116 mRNA was recently localized to endothelial or mesangial cells in developing mouse glomeruli.53 Our present data strongly suggest that Gpr116 is broadly expressed in vascular endothelium. Comments on the endothelial-specific expression of Ramp2, Slc9a3r2, Slc43a3, Rasip1, and NM_023516 are provided as an Expanded Discussion in the supplemental materials.
In summary, we have taken a new approach to determine a microvascular-specific core transcriptome in vivo, ie, transcripts that are highly endothelial (and possibly pericyte)-specific, but also broadly expressed within the microvasculature in different mammalian tissues. Using a combination of publicly available and our own transcript profiles, we identified 58 such transcripts, which include most known, and currently used, endothelial markers but also a large number of protein-coding transcripts for which information is currently rudimentary. Many of these are implicated in signal transduction, thereby opening new windows on endothelial-specific cell signaling and pinpointing novel putative targets for pharmacological intervention in endothelial cells.
Sources of Funding
This study was supported by grants from the Swedish Cancer Foundation, the Association for International Cancer Research (UK), the EU 6th framework integrated project LYMPHANGIOGENOMICS LSHG-CT-2004-503573, the Novo Nordisk, Strategic Research, Söderberg, Hedlund, IngaBritt and Arne Lundberg, and Knut and Alice Wallenberg Foundations.
Per Lindahl received support from the Swedish Cancer Foundation of greater than or equal to $10 000 and the EU 6th framework project in Lymphangiogenomics, also greater than or equal to $10 000.
Christer Betsholtz received support from the Swedish Cancer Foundation of greater than or equal to $10 000, the Association for International Cancer Research of greater than or equal to $10 000, and the EU 6th framework project in Lymphangiogenomics of greater than or equal to $10 000, and has ownership interest with regards to intellectual property of less than $10 000.
Original received March 5, 2008; final version accepted April 29, 2008.
- ↵Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim W, Berlin J, Baron A, Griffing S, Holmgren E, Ferrara N, Fyfe G, Rogers B, Ross R, Kabbinavar F. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med. 2004; 350: 2335–2342.
- ↵Ridgway J, Zhang G, Wu Y, Stawicki S, Liang WC, Chanthery Y, Kowalski J, Watts RJ, Callahan C, Kasman I, Singh M, Chien M, Tan C, Hongo JA, de Sauvage F, Plowman G, Yan M. Inhibition of Dll4 signalling inhibits tumour growth by deregulating angiogenesis. Nature. 2006; 444: 1083–1087.
- ↵Hellstrom M, Phng LK, Hofmann JJ, Wallgard E, Coultas L, Lindblom P, Alva J, Nilsson AK, Karlsson L, Gaiano N, Yoon K, Rossant J, Iruela-Arispe ML, Kalen M, Gerhardt H, Betsholtz C. Dll4 signalling through Notch1 regulates formation of tip cells during angiogenesis. Nature. 2007; 445: 776–780.
- ↵Leslie JD, Ariza-McNaughton L, Bermange AL, McAdow R, Johnson SL, Lewis J. Endothelial signalling by the Notch ligand Delta-like 4 restricts angiogenesis. Development. 2007; 134: 839–844.
- ↵Suchting S, Freitas C, le Noble F, Benedito R, Breant C, Duarte A, Eichmann A. The Notch ligand Delta-like 4 negatively regulates endothelial tip cell formation and vessel branching. Proc Natl Acad Sci USA. 2007; 104: 3225–3230.
- ↵Huminiecki L, Bicknell R. In silico cloning of novel endothelial-specific genes. Genome Res. 2000; 10: 1796–1806.
- ↵Peters DG, Zhang XC, Benos PV, Heidrich-O'Hare E, Ferrell RE. Genomic analysis of immediate/early response to shear stress in human coronary artery endothelial cells. Physiol Genom. 2002; 12: 25–33.
- ↵Favre CJ, Mancuso M, Maas K, McLean JW, Baluk P, McDonald DM. Expression of genes involved in vascular development and angiogenesis in endothelial cells of adult lung. Am J Physiol Heart Circ Physiol. 2003; 285: H1917–H1938.
- ↵St Croix B, Rago C, Velculescu V, Traverso G, Romans KE, Montgomery E, Lal A, Riggins GJ, Lengauer C, Vogelstein B, Kinzler KW. Genes expressed in human tumor endothelium. [see comment]. Science. 2000; 289: 1197–1202.
- ↵Su AI, Wiltshire T, Batalov S, Lapp H, Ching KA, Block D, Zhang J, Soden R, Hayakawa M, Kreiman G, Cooke MP, Walker JR, Hogenesch JB. A gene atlas of the mouse and human protein-encoding transcriptomes. Proc Natl Acad Sci USA. 2004; 101: 6062–6067.
- ↵He L, Sun Y, Takemoto M, Norlin J, Tryggvason K, Samuelsson T, Betsholtz C. The glomerular transcriptome and a predicted protein-protein interaction network. J Am Soc Nephrol. 2008; 19: 260–268.
- ↵Bondjers C, He L, Takemoto M, Norlin J, Asker N, Hellstrom M, Lindahl P, Betsholtz C. Microarray analysis of blood microvessels from PDGF-B and PDGF-Rbeta mutant mice identifies novel markers for brain pericytes. FASEB J. 2006; 20: 1703–1705.
- ↵Bostrom H, Willetts K, Pekny M, Leveen P, Lindahl P, Hedstrand H, Pekna M, Hellstrom M, Gebre-Medhin S, Schalling M, Nilsson M, Kurland S, Tornell J, Heath JK, Betsholtz C. PDGF-A signaling is a critical event in lung alveolar myofibroblast development and alveogenesis. Cell. 1996; 85: 863–873.
- ↵Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Royal Stat Soc. 1995; Ser B: 289–300.
- ↵Helmchen U, Kneissler U, Helmchen U, Bohle A. Three-dimensional electron microscopy of the structure of the normal mesangium of the rat glomerulus. Kidney Int. 1977: 215–216.
- ↵Huang EJ, Li H, Tang AA, Wiggins AK, Neve RL, Zhong W, Jan LY, Jan YN. Targeted deletion of numb and numblike in sensory neurons reveals their essential functions in axon arborization. Genes Dev. 2005; 19: 138–151.
- ↵Kreidberg JA, Donovan MJ, Goldstein SL, Rennke H, Shepherd K, Jones RC, Jaenisch R. Alpha 3 beta 1 integrin has a crucial role in kidney and lung organogenesis. Development. 1996; 122: 3537–3547.
- ↵Quaggin SE, Schwartz L, Cui S, Igarashi P, Deimling J, Post M, Rossant J. The basic-helix-loop-helix protein pod1 is critically important for kidney and lung organogenesis. Development. 1999; 126: 5771–5783.
- ↵Lindahl P, Johansson BR, Leveen P, Betsholtz C. Pericyte loss and microaneurysm formation in PDGF-B-deficient mice. Science. 1997; 277: 242–245.
- ↵Hellstrom M, Kalen M, Lindahl P, Abramsson A, Betsholtz C. Role of PDGF-B and PDGFR-beta in recruitment of vascular smooth muscle cells and pericytes during embryonic blood vessel formation in the mouse. Development. 1999; 126: 3047–3055.
- ↵Bondjers C, Kalen M, Hellstrom M, Scheidl SJ, Abramsson A, Renner O, Lindahl P, Cho H, Kehrl J, Betsholtz C. Transcription profiling of platelet-derived growth factor-B-deficient mouse embryos identifies RGS5 as a novel marker for pericytes and vascular smooth muscle cells. Am J Pathol. 2003; 162: 721–729.
- ↵Gory-Faure S, Prandini MH, Pointu H, Roullot V, Pignot-Paintrand I, Vernet M, Huber P. Role of vascular endothelial-cadherin in vascular morphogenesis. Development. 1999; 126: 2093–2102.
- ↵Carmeliet P, Lampugnani MG, Moons L, Breviario F, Compernolle V, Bono F, Balconi G, Spagnuolo R, Oostuyse B, Dewerchin M, Zanetti A, Angellilo A, Mattot V, Nuyens D, Lutgens E, Clotman F, de Ruiter MC, Gittenberger-de Groot A, Poelmann R, Lupu F, Herbert JM, Collen D, Dejana E. Targeted deficiency or cytosolic truncation of the VE-cadherin gene in mice impairs VEGF-mediated endothelial survival and angiogenesis. Cell. 1999; 98: 147–157.
- ↵Nitta T, Hata M, Gotoh S, Seo Y, Sasaki H, Hashimoto N, Furuse M, Tsukita S. Size-selective loosening of the blood-brain barrier in claudin-5-deficient mice. J Cell Biol. 2003; 161: 653–660.
- ↵Li DY, Sorensen LK, Brooke BS, Urness LD, Davis EC, Taylor DG, Boak BB, Wendel DP. Defective angiogenesis in mice lacking endoglin. Science. 1999; 284: 1534–1537.
- ↵Ichikawa-Shindo Y, Sakurai T, Kamiyoshi A, Kawate H, Iinuma N, Yoshizawa T, Koyama T, Fukuchi J, Iimuro S, Moriyama N, Kawakami H, Murata T, Kangawa K, Nagai R, Shindo T. The GPCR modulator protein RAMP2 is essential for angiogenesis and vascular integrity. J Clin Invest. 2008; 118: 29–39.
- ↵McKnight AJ, Gordon S. The EGF-TM7 family: unusual structures at the leukocyte surface. J Leukocyte Biol. 1998; 63: 271–280.
- ↵Fredriksson R, Lagerstrom MC, Lundin LG, Schioth HB. The G-protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. [see comment]. Mol Pharmacol. 2003; 63: 1256–1272.
- ↵Nechiporuk T, Urness LD, Keating MT. ETL, a novel seven-transmembrane receptor that is developmentally regulated in the heart. ETL is a member of the secretin family and belongs to the epidermal growth factor-seven-transmembrane subfamily. J Biol Chem. 2001; 276: 4150–4157.