HSPA12B Is Predominantly Expressed in Endothelial Cells and Required for Angiogenesis
Objective— HSPA12B is the newest member of HSP70 family of proteins and is enriched in atherosclerotic lesions. This study focused on HSPA12B expression in mice and its involvement in angiogenesis.
Methods and Results— The expression of HSPA12B in mice and cultured cells was studied by: (1) Northern blot; (2) in situ hybridization; (3) immunostaining with HSPA12B-specific antibodies; and (4) expressing Enhanced-Green-Fluorescent-Protein under the control of the HSPA12B promoter in mice. The function of HSPA12B was probed by an in vitro angiogenesis assay (Matrigel) and a migration assay. Interacting proteins were identified through a yeast two-hybrid screening. HSPA12B is predominantly expressed in vascular endothelium and induced during angiogenesis. In vitro angiogenesis and migration are inhibited in human umbilical vein endothelial cells in the presence of HSPA12B-neutralizing antibodies. HSPA12B interacts with multiple proteins in yeast 2-hybrid system.
Conclusions— We provide the first evidence to our knowledge that the HSPA12B is predominantly expressed in endothelial cells, required for angiogenesis, and interacts with known angiogenesis regulators. We postulate that HSPA12B provides a new mode of angiogenesis regulation and a novel therapeutic target for angiogenesis-related diseases.
Blood vessel development and formation are essential for organ growth and repair, wound healing, and reproduction cycle, and an imbalance of functional vessels contributes to diseases such cancer and ischemia.1 During development of the vascular system, vasculogenesis refers to the process in which endothelial progenitors differentiate, proliferate, multiply, and migrate to give rise to a primitive vascular network of arteries and veins; angiogenesis refers to the process of blood vessels expansion/remodeling from the existing endothelial cell (EC) network through proliferating, sprouting, pruning, and remodeling. Pericytes and smooth muscle cells are recruited to cover nascent endothelial channels, which provide strength and regulation of vessel perfusion, a process termed arteriogenesis.2 The formation and maintenance of functional blood vessels is a complex process involving the interplay of multiple genes. These genes include members of many signaling pathways such as vascular endothelial growth factors/ vascular endothelial growth factor receptors, angiopoietin/Tie families, platelet-derived growth factor, transforming growth factor-β, Notch pathways, certain integrins, neuronal axon guidance molecules such as ephrin, semaphorins, netrins, and robo, transcriptional factors, and many other genes.3 In adults, many of the embryonic and early pathways are reactivated in situations of neoangiogenesis.
Despite great progresses in finding key regulators in angiogenesis, characterizing new genes is still necessary and greatly beneficial for a full understanding of the process. The precise and delicate coordination, combination, and collaboration of these molecular players in the right time, space, and dose, so critical for the formation and maintenance of functional blood vessels would have been greatly aided by molecular chaperones. For example, an endothelial-specific molecular chaperone might explain the endothelial-specific effects of many transcriptional factors that are expressed rather broadly.4 But to date, no endothelial-specific molecular chaperone has been reported.
Heat shock proteins (HSP) are a group of proteins that are abundant in cells, highly conserved among species, and associated with stress responses. HSP70 is the largest and most conserved family of HSP.5 HSP70s function as molecular chaperones, assisting in protein synthesis, folding, assembly, trafficking between cellular compartments, and degradation.6 HSP70s participate in cellular stress response by binding and refolding misfolded proteins, or removing the consistently-misfolded proteins through the ubiquitin–proteasome system by interacting with CHIP (carboxyl terminus of HSP70-interacting protein), an HSP70-associated ubiquitin ligase.7,8 In addition to its chaperone functions, it has been increasingly recognized that HSP70s also act as signaling molecules9 and regulators in cellular processes such as apoptosis.10 Expression of HSP70 family members has been considered to be ubiquitous and not restricted to any specific cell type.
We have previously cloned a new subfamily of HSP70 proteins, the HSPA12 subfamily.11 The first member of this subfamily, HSPA12A, is a candidate gene for atherosclerosis susceptibility, based on evidence of genetic linkage and expression profiling.11,12 The second member of this family, HSPA12B was cloned through sequence homology (60% identical to HSPA12A) and is also enriched in atherosclerotic lesions. We set out to characterize the expression and function of this gene and the mechanism by which it influences angiogenesis.
Northern Blot and In Situ Hybridization
Mouse multiple-tissue blots and human heart-tissue blots were purchased from Clontech (Mountain View, Calif); the rat brain tissue blot and time-course blot were purchased from Seegene (Rockville, Md). Northern Blots and in situ hybridizations experiments were performed as previously described.11 Northern blot probes were generated by polymerase chain reaction and in situ hybridization probes were generated with a Dig RNA labeling kit from Roche (Indianapolis, Ind). Primers sequences are listed in supplemental Figure I (see http://atvb.ahajournal.org). Mice were housed in East Tennessee State University Animal Care Facility and fed normal chow diet. Twelve-week-old C57BL/6 were from Jackson Laboratories. HSPA12B-BAC-EGFP transgenic mice were generated in the Rockefeller University.
Endothelial Cell Cultures and HSPA12B Expression Constructs
Please see http://atvb.ahajournal.org.
Antibody Production and Western Blot Analysis
Two antibodies, the N-terminus-specific Ab4110 and the C-terminus-specific Ab4112, were generated by immunizing rabbits with peptides MLTVPEMGLQGLYISSC (mHSPA12B amino acid (AA)1 to 17) and CVDVSTNRSVRAAIDFLSN (mHSPA12B AA667–685), respectively (ProSci Inc, Poway, Calif). Enzyme-linked immunosorbent assay-positive sera were further purified by affinity column using the same peptides. The Western blots were performed using standard procedures. For membranes probed with primary antibodies Ab4110 or Ab4112, a goat-anti-rabbit IgG/horseradish peroxidase conjugate (Pierce, Rockford, Ill) was used as the secondary antibody and signals were developed using a chemiluminescent substrate (WesternDura; Pierce). As a control for loading and normalization, membranes were reprobed with an anti-GAPDH monoclonal antibody (MMS-580S; Covance, Berkeley, Calif). Chemiluminescence signal was scanned with a FLA-500 phosphorimager (Fujifilm, Tokyo, Japan) and quantified by ImageGauge software.
Mice were anesthetized with ketamine/xylazine and euthanized by cardiac perfusion with 4% paraformaldehyde. The organs were removed, post-fixed in 4% paraformaldehyde for 1 hour, incubated in 25% sucrose at 4°C for 96 hours, and stored in −80°C. Tissue sections were generated with a Leica CM1850 cryostat. Immunofluorescent stainings were performed by standard protocols with AlexaFluor-488 goat-anti-rabbit IgG (H+L) (green) and AlexaFluor-555 goat-anti-rat IgG (H+L) (red) (Molecular Probes, Carlsbad, Calif). Briefly, sections were blocked in 10% bovine serum albumin blocking solution for 1 hour, incubated with primary antibodies (1:200) for 4 hours at room temperature, washed 4 times with 1× phosphate-buffered saline for 15 minutes each, incubated with secondary antibodies-conjugates (1:200) for 1 hour, and washed 4 times with 1× phosphate-buffered saline for 15 minutes each before mounting. The anti-enhanced-Green-Fluorescent-Protein (EGFP) was purchased from Abcam (Ab290; Cambridge, Mass), rat-anti-CD31 (PECAM-1) from Pharmingen (San Diego, Calif).
HSPA12B-BAC-EGFP Transgenic Mice and Confocal Imaging
To generate the transgenic mice expressing EGFP under the promoter of mHSPA12B (HSPA12B-BAC-EGFP mice), a bacterial artificial chromosome (BAC) containing HSPA12B was engineered so that the EGFP replaced the HSPA12B coding region at the starting ATG. Transgenic mice were generated from the engineered BAC and were of FVB/N crossed with Swiss-Webster background.13 Two independent founder lines were studied and gave identical results. Tissue sections (20 um) from 12- to 16-week-old transgenic mice and their wild-type littermates were imaged using a Leica TCS SP2 confocal microscope system. Z-series were collected and maximum intensity projection images were created from the series.
Transfection of Endothelial Cells, Yeast 2-Hybrid Screening, and Migration Assay
Please see http://atvb.ahajournals.org.
Matrigel-Based In Vitro Angiogenesis Assays
Wells of a 24-well plate were coated with 250 μL Matrigel (BD Biosciences) and incubated at 37°C for 30 minutes; 5×104 human umbilical vein endothelial cells (HUVEC) in complete EC medium were added to Matrigel-coated wells and incubated at 37°C with 5% CO2 for 24 hours. When required, cells were recovered from Matrigel with Cell Recovery Solution (BD Biosciences) in the presence of 1× protease inhibitors cocktail (Sigma) for making cell extracts.
All results were expressed as the mean±SD. Significance of differences was determined with the Student t test with significance at P<0.05.
HSPA12B mRNA Is Expressed in Multiple Organs and in Blood Vessels Specifically
In mouse multi-tissue Northern blot, a single &3.3-kb HSPA12B mRNA transcript was detected at the highest level in heart, followed by lung (Figure 1A). Longer exposure revealed HSPA12B was expressed in all tissues examined, as confirmed by reverse-transcription polymerase chain reaction (not shown). HSPA12B mRNA of similar size was detected in human heart (supplement II) and rat brain (supplement II). In mouse brain, HSPA12B was present at 17.5 day postcoitum embryo, peaked at day 3, and decreased gradually (Figure 1B).
In situ hybridization experiments revealed that, compared with sense-probe control, the antisense probe recognized a distinctly vessel pattern, indicating HSPA12B was expressed specifically in blood vessels (Figure 1C).
HSPA12B Proteins Are Expressed in ECs
Two HSPA12B-specific antibodies, Ab4110 (N-terminus-specific) and Ab4112 (C-terminus-specific), recognized &76-kDa protein (predicted MW 76-kDa) in 293 cells overexpressing HSPA12B and in HUVEC cells, and &80-kDa protein in 293 cells overexpressing Flag-tagged-HSPA12B (Figure 2A). The interaction was specific because blocking peptides used to elicit the antibodies abolished the signals.
We then measured HSPA12B protein levels in multiple mouse organs by Western blots. The microvessel-rich lung, instead of heart, expressed the highest level, suggesting a post-transcriptional regulation mechanism (Figure 2B). In a panel of randomly selected cells, HSPA12B proteins were present in HUVEC cells at the highest level (Figure 2C), and the order of expression in human primary endothelial cell lines was HMVC≥HUVEC>HAEC (Figure 2C), and not detected in human primary fibroblasts and human peripheral blood mononuclear cells (not shown).
We next characterized HSPA12B cell-type specificity by immunohistochemistry. HSPA12B proteins were detected specifically in blood vessels in mouse heart and brains (Figure 3A), consistent with in situ hybridization. Double staining with anti-HSPA12B antibody Ab4110 and endothelial cell-specific anti-CD31 (PECAM-1) revealed complete colocalization (Figure 3B), indicating that HSPA12B was specifically expressed in endothelial cells. Immunostaining with a colorimetric reagent instead of fluorescent reagents produced the same endothelial cell-specific patterns (supplement III).
Finally, we tracked the HSPA12B expression in mice expressing EGFP under the control of the HSPA12B promoter (HSPA12B-BAC-EGFP mice). This method takes advantage of the observation that bacterial artificial chromosome (BAC) commonly preserves the expression pattern of the gene it encodes, presumably because of its length.13 Fluorescent EGFP signals were observed specifically in vessels of various calibers in brain and heart, adipose tissue capillaries, endothelial cells lining the aorta (Figure 3C), as well as lung alveolar capillaries, skeletal muscle capillaries, renal glomeruli and intertubular capillaries, and endothelial cells lining the aortic root and aortic valve epithelium (supplement IV). By contrast, inside liver, only the portal veins in the periportal area were fluorescent, whereas the fenestrated sinusoidal endothelial cells were consistently negative (supplement IV). Again, the fluorescent signals localized closely to CD31, as shown in fluorescent microscopy with anti-CD31 staining (supplement V).
Thus, evidence from in situ hybridization, Western blot, immunohistochemistry, and HSPA12B-BAC-EGFP transgenic mice is consistent with HSPA12B being predominantly expressed in endothelial cells lining blood vessels. These results suggest that HSPA12B may play a role in angiogenesis.
HSPA12B Expression Was Increased in Confluent Endothelial Cells When Cells Came Into Contact and Reduced in Quiescent Cells
Proteins functioning in angiogenesis often display upregulation during cell proliferation/alignment/movement. Therefore, we examined HSPA12B expression in actively growing versus quiescent endothelial cells. Subconfluent, exponentially growing HUVEC cells (Expo) were grown to confluence and were maintained in complete endothelial cell-specific medium for an additional 8 hours (Conf). Thereafter, the medium was replaced by M199 medium containing only 5% fetal bovine serum without endothelial cell-specific growth factors for 8 hours (M199–8 hour). The HSPA12B protein levels increased significantly in confluent, active HUVEC cells, and decreased with the withdrawal of endothelial-specific growth factors (Figure 4A).
Modulation of HSPA12B Protein Expression During In Vitro Angiogenesis
In vitro angiogenesis operates on the principle that endothelial cells form capillary structures when cultured on a supportive matrix derived from a murine tumor (Matrigel). This assay reproduces the angiogenic processes of migration, alignment, and cell differentiation, and has proved to be an important tool for studying the mechanisms of angiogenesis.14 A typical in vitro angiogenesis experiment is shown in supplement VI. Endothelial cells plated on Matrigel underwent alignment and elongation within 1 to 3 hours after seeding, establishing the pattern for further capillary networking. By 6 hours, formation of cords had begun and by 12 hours virtually all cells had fused into continuous cords. Stabilization and refinement of the cords progressed until 24 hours. We studied the kinetics of endogenous HSPA12B protein expression during in vitro angiogenesis of HUVEC cells, and found it increased within 6 hours and leveled off at 24 hours (Figure 4B). Thus HSPA12B expression is modulated during in vitro angiogenesis: upregulated during tubule formation.
N Terminus–Specific Anti-HSPA12B Antibodies Block Angiogenesis
A transient increase in HSPA12B levels might be functionally important in the regulation of tubule formation. Thus, we tested the effect of neutralizing HSPA12B on angiogenesis. A representative experiment is shown in Figure 5A. Delivery of the preimmune isotypic control caused no gross deformation in the tubule network formation. In contrast, the N-terminus-specific Ab4110 inhibited tubule formation, indicating HSPA12B was required for in vitro angiogenesis. The Ab4110 was capable of quantitatively immunoprecipitating HSPA12B proteins from cell lysates overexpressing HSPA12B (supplement VII), thus may serve as neutralizing reagents once delivered inside the cells. The transfection efficiency achieved was consistently >95% as indicated by the addition of a nonrelevant AlexaFluor-555-labeled antibody (Figure 5A, top panel).
Expression of EGFP-HSPA12B but not HSPA12B-EGFP Chimeras in HUVEC Cells Interferes With Angiogenesis
The previous experiment suggested that masking the N-terminus of HSPA12B with Ab4110 inhibited angiogenesis. To corroborate this, we tested the effect of overexpressing an N-terminus–tethered form of HSPA12B chimera (EGFP-HSPA12B) on angiogenesis, using EGFP as well as a C-terminus-tethered HSPA12B chimera (HSPA12B-EGFP) as controls. To do this, we transiently transfected HUVEC cells with: (1) EGFP; (2) the HSPA12B-EGFP fusion; and (3) the EGFP-HSPA12B fusion (Figure 5B, top panel). We tracked the numbers of transfected cells (green cells) inside tubule-like structures at the end of a 24-hour angiogenesis assay (Figure 5B, bottom panel). For both controls, transfected cells (green cells) were readily incorporated into the tubule-like structures, at the ratios close to transfection efficiency (Figure 5B, bottom panels, left and middle). In contrast, no green cells were observed in the tubule-like structures for HUVEC cells transfected with the EGFP-HSPA12B (Figure 5B, bottom panels, right). Thus, tagging an EGFP group to the N-terminus, but not the C-terminus, of the HSPA12B inhibited angiogenesis, suggesting an unblocked N-terminus of HSPA12B is needed for angiogenesis.
N Terminus-Specific Anti-HSPA12B Antibodies Reduce Migration
Migration is an essential early event of angiogenesis, thus we examined whether neutralizing HSPA12B would interfere with migration using a Boyden chamber assay. As shown in supplement VIII, migration was stimulated by the growth medium containing endothelial cell growth factors in HUVEC cells (2.7±0.4-fold over background). In contrast, HUVEC cells transfected with Ab4110 showed a significant reduction (P<0.01) in the number of cells transmigrating toward chemoattractant medium (1.5±0.3-fold over background). The difference between control and Ab4112 was not significant. Thus, HSPA12B is involved in the HUVEC cells chemotactic activities toward angiogenic factors as measured in this assay.
HSPA12B Interacts With Multiple Proteins
To further investigate the molecular mechanism underlying HSPA12B’s role on angiogenesis, we searched HSPA12B-interacting proteins on a yeast 2-hybrid system using a human bone marrow endothelial cell random-primed cDNA library. We obtained 22 unique DNA prey fragments from 232 positive yeast clones after screening 134 millions of interactions with HSPA12B. A ranking system, which took account of prey fragments’ redundancy, independence, frequency, distributions of reading frames and stop codons, as well as appearance in all previous screens, was used to assess the reliability of each interaction.15 The 2 top-ranked clones are AKAP12 (A-kinase-anchoring protein, also known as Gravin or SSeCKS) and hPODXL (human podocalyxin-like), both of which have been reported to mediate cell adhesion. Other clones include aryl hydrocarbon receptor nuclear translocator (ARNT), an angiogenesis regulator.16 Whether these interactions represent biologically relevant interactions that mediate HSPA12B function remains to be determined.
As the newest member of HSP70 family, little was known for the normal expression and function of HSPA12B. In this study, we report that HSPA12B is predominantly expressed in endothelial cells, as assayed by in situ hybridization, immunostaining, and EGFP expression in mice under the control of HSPA12B native regulatory elements. We also report that HSPA12B is required for angiogenesis and an unblocked N-terminus is needed for these angiogenic activities. We provide evidence that HSPA12B may exert its effects during cellular adhesion and cell migration. Our data suggest that HSPA12B may be an EC-specific chaperone.
A common perception is that HSP70 proteins are expressed ubiquitously. Here, we showed that the HSPA12 subfamily is highly tissue-specific. HSPA12B is predominantly expressed in endothelial cells, and HSPA12A is highly expressed in neuronal cells.17 Further studying of their roles in development and pathology should yield important insights into those processes. The HSPA12 subfamily is the least conserved subfamily of the HSP70 family, judging by primary sequence homology. Still, 3-dimensional modeling shows that the signature ATPase domain that covers approximately two-thirds of the HSP70 is well-conserved in HSPA12 (FUGUE program, Z-score of 22.18 for HSPA12B, and 28.39 for HSPA12A, when higher Z-score means higher confidence and a Z-score >6 means the prediction is almost certain).
In the vascular endothelia, HSPA12B expression is prevalent but not universal, demonstrating vessel heterogeneity. HSPA12B is found in adipose tissue, brain, heart, kidney, and lung, but not in liver sinusoidal endothelial cells, which are fenestrated and where blood pressure is low. In brain and heart, vessels of all calibers expresses HSPA12B, but in adipose tissue and lung, HSPA12B is present primarily in capillaries. The latter is consistent with the expression levels observed in cultured primary cells (microvessel ≥ vein > aorta).
Ab4110, once delivered into HUVEC cells, inhibited angiogenesis in Matrigel assay. This could be the result of either complete removal/precipitation of HSPA12B protein, or masking the HSPA12B N-terminus region, where the conserved ATPase domain resides. The second possibility is more likely, because the C-terminus-specific Ab4112 did not block angiogenesis, even though Ab4112 possessed similar levels of affinity toward HSPA12B in immunoprecipitation. Accordingly, overexpressing the N-terminus-masked HSPA12B (EGFP-HSPA12B), but not the C-terminus-masked HSPA12B (HSPA12B-EGFP), prevented otherwise normal cells from participating in angiogenesis. Thus, interfering with the N-terminus region of HSPA12B abrogated angiogenesis. The dominant-negative fashion of latter results was consistent with the hypothesis that HSPA12B has multiple domains and involves protein-protein interactions for its proper function. HSP70 proteins consist of two functional domains: the highly conserved ATPase domain of 44-kDa in the N-terminus and the substrate binding region of 25-kDa in the C-terminus.18 Also, it is thought that chaperones cooperate extensively, sometimes forming multi-chaperone systems that work sequentially or simultaneously to ensure the efficient biogenesis of cellular proteins in their respective cellular compartments.19 For example, HSC70 interacts with HSP40 and Hop through the C-terminal domain, and interacts with Hip through N-terminal ATPase domain, and BAG-1 competes with Hip in binding to the ATPase domain.20 This network of cooperating and competing cofactors regulates the chaperone activity of HSC70. It is conceivable that the presence of EGFP group in the N-terminus might either block HSPA12B activity or disrupt its normal interactions.
Recently, a zebrafish HSPA12B homologue has been characterized and found to be specifically expressed in zebrafish blood vessels. Knocking-down of HSPA12B by RNAi disrupted zebrafish blood vessel normal development, and inhibited in vitro angiogenesis and migration in HUVEC cells (VP Sukhatme, unpublished data, 2006). These results are consistent with and complementary to our findings.
The mechanism by which HSPA12B influence angiogenesis remains unclear, but our data provides evidence that HSPA12B may regulate endothelial cell adhesion and migration, critical steps during angiogenesis. Firstly, HSPA12B is induced in active, confluent endothelial cells and during the early stage of in vitro angiogenesis when cells make contact and migrate. Second, inhibiting HSPA12B reduces HUVEC cell migration in a migration assay. Finally, 2 top-ranked interacting proteins from a yeast 2-hybrid screening, AKAP12 and hPODXL, are known to regulate cell adhesions. AKAP12 has been reported to inhibit angiogenesis and promote tight-junction formation in blood-brain barrier.21 hPODXL encodes a sialomucin protein that regulates cell adhesion in podocyte22 and tumor,23 and is expressed in endothelial cells and bind to L-selectin.24 It has also been reported that AKAP12 is expressed strongly in subconfluent, and only weakly in confluent, endothelial cell cultures, a pattern opposite to that of HSPA12B.25 One obvious hypothesis will be that HSPA12B promotes angiogenesis by mediating the turnover of anti-angiogenic/pro-tight junction proteins such as AKAP12.
In this study, we also demonstrated that the HSPA12B-BAC-EGFP transgenic mice expressed EGFP strongly and specifically in endothelial cells. These mice will be a valuable tool for marking endothelial cells and tracking blood vessels during normal development or under pathological conditions.
In summary, we have shown that HSPA12B is expressed predominantly in vascular endothelium, required for angiogenesis, and may interact with known angiogenesis mediators. HSPA12B knockout mice should ultimately help to confirm the role of HSPA12B in angiogenesis and shed light on its mechanism. The characterization of HSPA12B provides new insight on angiogenesis and may lead to new target for disease intervention.
We thank Dr Nathaniel Heintz and the Gene Expression Nervous System Atlas (GENSAT) Project, NINDS Contract N01NS02331 to The Rockefeller University (New York, NY), for generating the HSPA12B-BAC-EGFP transgenic mice under our suggestion. We thank members of Dr Vikas Sukhatme’s laboratory (Harvard Medical School) for sharing their unpublished results on the phenotype of HSPA12B knockdown zebrafish. We thank Dr Youxing Qu (University of Minnesota), for FUGUE program analyses.
Source of Funding
This work was supported by a startup fund from East Tennessee State University and an Atorvastatin Research Award from Pfizer to Zhihua Han.
Original received April 21, 2006; final version accepted June 21, 2006.
Oettgen P. Transcriptional regulation of vascular development. Circ Res. 2001; 89: 380–388.
Gething M-J. Guidebook to molecular chaperones and protein-folding catalysts. Oxford University Press; 1997.
Hartl FU, Hayer-Hartl M. Molecular chaperones in the cytosol: from nascent chain to folded protein. Science. 2002; 295: 1852–1858.
Jiang J, Ballinger CA, Wu Y, Dai Q, Cyr DM, Hohfeld J, Patterson C. CHIP is a U-box-dependent E3 ubiquitin ligase: identification of Hsc70 as a target for ubiquitylation. J Biol Chem. 2001; 276: 42938–42944.
Xu Q. Role of heat shock proteins in atherosclerosis. Arterioscler Thromb Vasc Biol. 2002; 22: 1547–1559.
Han Z, Truong QA, Park S, Breslow JL. Two Hsp70 family members expressed in atherosclerotic lesions. Proc Natl Acad Sci U S A. 2003; 100: 1256–1261.
Dansky HM, Shu P, Donavan M, Montagno J, Nagle DL, Smutko JS, Roy N, Whiteing S, Barrios J, McBride TJ, Smith JD, Duyk G, Breslow JL, Moore KJ. A phenotype-sensitizing apoe-deficient genetic background reveals novel atherosclerosis predisposition Loci in the mouse. Genetics. 2002; 160: 1599–1608.
Nicosia RF, Ottinetti A. Modulation of microvascular growth and morphogenesis by reconstituted basement membrane gel in three-dimensional cultures of rat aorta: a comparative study of angiogenesis in matrigel, collagen, fibrin, and plasma clot. In Vitro Cell Dev Biol. 1990; 26: 119–128.
Demand J, Luders J, Hohfeld J. The carboxy-terminal domain of Hsc70 provides binding sites for a distinct set of chaperone cofactors. Mol Cell Biol. 1998; 18: 2023–2028.
Economou CG, Kitsiou PV, Tzinia AK, Panagopoulou E, Marinos E, Kershaw DB, Kerjaschki D, Tsilibary EC. Enhanced podocalyxin expression alters the structure of podocyte basal surface. J Cell Sci. 2004; 117: 3281–3294.
Somasiri A, Nielsen JS, Makretsov N, McCoy ML, Prentice L, Gilks CB, Chia SK, Gelmon KA, Kershaw DB, Huntsman DG, McNagny KM, Roskelley CD. Overexpression of the Anti-Adhesin Podocalyxin Is an Independent Predictor of Breast Cancer Progression. Cancer Res. 2004; 64: 5068–5073.
Sassetti C, Tangemann K, Singer MS, Kershaw DB, Rosen SD. Identification of Podocalyxin-like Protein as a High Endothelial Venule Ligand for L-selectin: Parallels to CD34. J Exp Med. 1998; 187: 1965–1975.