Laminin deposition is dispensable for

0 downloads 0 Views 2MB Size Report
Lars Jakobsson,* Anna Domogatskaya,† Karl Tryggvason,† David Edgar,‡ and ...... X. F., Breitman, M. L., and Schuh, A. C. (1995) Failure of blood-island ...

The FASEB Journal • Research Communication

Laminin deposition is dispensable for vasculogenesis but regulates blood vessel diameter independent of flow Lars Jakobsson,* Anna Domogatskaya,† Karl Tryggvason,† David Edgar,‡ and Lena Claesson-Welsh*,1 *Department of Genetics and Pathology, Uppsala University, Uppsala, Sweden; †Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden; and ‡Department of Human Anatomy, University of Liverpool, Liverpool, UK Basement membranes (BMs) consisting of laminins, collagens, and heparan sulfate proteoglycans (HSPGs) are vital for proper endothelial cell function, but many aspects of their role in vascular development remain unknown. Here, we demonstrate that vascular structures within differentiating embryoid bodies are wrapped in a BM composed of ␣4- and ␣5-chain laminins, fibronectin, collagen IV, and HSPGs. In sprouting angiogenesis, laminins were produced by stalk cells, as well as the leading tip cell, and deposited along the sprout length, including tip cell filopodia. In embryonic stem cells deficient in laminins, due to lamc1 (laminin ␥1) deletion, vascular development and organization were largely unaffected. However, the frequency of vessels with wide lumens was increased 4-fold. Laminin-deficient vessels were moreover characterized by increased fibronectin levels and enhanced endothelial cell proliferation. We conclude that laminins are dispensable for vascular development but that they regulate lumen formation in the absence of flow and vascular tone.—Jakobsson, L., Domogatskaya, A., Tryggvason, K., Edgar, D., ClaessonWelsh, L. Laminin deposition is dispensable for vasculogenesis but regulates blood vessel diameter independent of flow. FASEB J. 22, 1530 –1539 (2008)


Key Words: angiogenesis 䡠 sprouting 䡠 basement membrane 䡠 VEGF

Blood vessels are composed of polarized endothelial cells (ECs) that rest on a basement membrane (BM) that consists of a three-dimensional (3-D) network of interconnected matrix proteins. Smooth muscle cells (pericytes) support and surround the vascular tube to different extents depending on the vessel type. Strict regulation of blood vessel formation (vasculogenesis and angiogenesis) is a prerequisite for vertebrate development and physiology, and deregulation of these processes is associated with several pathological conditions such as diabetic retinopathy, rheumatoid arthritis, and cancer. Angiogenesis is initiated by vascular endothelial 1530

growth factor (VEGF), which activates vascular endothelial growth factor receptor 2 (VEGFR-2) on the EC (1). VEGFR-2 activation triggers the release of proteases that degrade the BM, accompanied by detachment of smooth muscle cells. As a result, ECs participate in a program of active angiogenesis, proliferating and migrating into the surrounding tissue. Although breakdown of the BM is considered vital for generation of new blood vessels, little is known about the deposition and assembly of BM components during blood vessel formation. Moreover, a number of BM-derived peptides (for a review, see ref. 2) are known to possess antiangiogenic properties in vitro as well as in vivo, but their relevance for vessel stability and remodeling is unknown. Maturation of the newly formed vessel is thought to involve thickening of the BM and recruitment of supporting cells (pericytes) by the secretion of platelet-derived growth factor (3). Most molecules residing within the BM, such as laminins (LMs), collagens, fibronectin, and heparan sulfate proteoglycans (HSPGs), have been associated with diverse processes such as cell differentiation, attachment, migration, polarization, guidance, and survival (4). LMs constitute a family of heterotrimeric glycoproteins, with each family member containing one ␣-, ␤-, and ␥-chain. To date 5 ␣ (␣1–5), 3 ␤ (␤1–3), and 3 ␥ (␥1–3) chains are known to combine into at least 16 mammalian isoforms (including the major splice variants), which show developmental stage-specific and cell type-specific expression patterns. The LM ␣-chains are highly bioactive with binding sites for integrins, dystroglycans, and HSPGs. In addition, the ␣-chains are more restricted in their expression than the ␤- and ␥-chains. The only LM ␣-chains present in the vascular BM are the ␣4- and ␣5-chains, which are expressed and deposited during mouse development, at least from embryonic day (E) 11–13 and 1 Correspondence: Department of Genetics and Pathology, Uppsala University, Rudbeck Laboratory, Dag Hammarskjo¨ldsv 20, SE-75185, Uppsala, Sweden. E-mail: [email protected] doi: 10.1096/fj.07-9617com

0892-6638/08/0022-1530 © FASEB

onwards (5, 6); for a review, see ref. 7. Deletion of LM-␣5 results in embryonic lethality at 17.5 days postcoitum with brain and limb defects, enlarged and disorganized placental vessels, and aberrant renal glomeruli vascularization (8, 9). The LM-␣4 null mice are viable and fertile but exhibit prenatal vascular bleedings, a phenotype that is gradually lost during development (10). Many of the LM-chain specific knockouts result in embryonic lethality (gene targeting of ␣1 results in death at E7, ␤1 at E5.5, and ␥1 at E5.5) before onset of vasculogenesis (11, 12). To overcome problems with implantation and early embryonic lethality, we have studied embryonic stem cells with targeted deletions of lamc1 (the gene encoding LM-␥1). By in vitro differentiation of wild-type and mutant stem cells we have been able to monitor the differential expression of the LM chains and isoforms and to study the role of LMs in vasculogenic and angiogenic processes.

MATERIALS AND METHODS Embryonic stem cells Murine 129 SvJ, R1 wild-type and vegfr-2⫺/⫺ ES cells (13), were kind gifts from Dr. Andras Nagy and Dr. Janet Rossant (Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Toronto, Canada), respectively. The lamc1⫺/⫺ ES cells were established as described by Smyth et al. (12). Embryoid bodies Culturing of ES cell clones and generation of embryoid bodies were as described previously (14). At day 4, the embryoid bodies were placed in 8-well glass culture slides (BD Falcon; BD, Franklin Lakes, NJ, USA) in the presence or absence of LM111, LM411 (15), or LM511 (16) at a concentration of 20 ␮g/ml. Alternatively, embryoid bodies were placed in a polymerized collagen I gel, as described previously (14), with the addition of 32.4 ␮g/ml LM111, LM411, or LM511. Medium with or without growth factors was changed at day 8. Proliferation Bromodeoxyuridine (BrdU) was added to the cultures at day 14, and cultures were processed for immunostaining 24 h later. Proliferative cells were visualized using the anti-BrdU nuclease kit (Amersham Pharmacia Biotech, Uppsala, Sweden) in combination with Alexa 488 goat anti-mouse highly cross-absorbed IgG (Molecular Probes, Eugene, OR, USA). Immunostaining of embryoid bodies Embryoid bodies on glass slides and in collagen I gels were processed for whole-mount immunohistochemical staining as described previously (14). The following primary antibodies were used for detection: rat anti-mouse CD31 antibody (PharMingen, San Diego, CA, USA); fluorescein isothiocyanate (FITC) -conjugated mouse monoclonal anti-␣-smooth muscle actin antibody, goat anti-mouse VEGFR-2 (R&D, Minneapolis, MN, USA); rabbit anti-NG2, goat anti-collagen IV (Chemicon, Temecula, CA, USA); IgM mouse anti-mouse LAMININS IN VASCULAR DEVELOPMENT

HepSS-1 antibody (Seikagaku, Tokyo, Japan); rabbit antiserum to LM-␣4 (clone377) (kindly provided by Dr. Lydia Sorokin, Universita¨tsklinikum Mu¨nster, Mu¨nster, Germany; ref. 17); rabbit antiserum to LM-␣5 (kindly provided by Dr. Jeffrey H. Miner, Washington University School of Medicine, St. Louis, MO, USA; ref. 18); rabbit anti-pan LM (Sigma, St. Louis, MO, USA); rabbit anti-fibronectin (Dako, Copenhagen, Denmark). Secondary antibodies: Alexa 568 goat anti-rat IgG, Alexa 488 goat anti-rat IgG, Alexa 555 donkey anti-goat IgG, Alexa 488 donkey anti-rabbit, Alexa 555 goat anti-mouse IgM (Molecular Probes); Cy5 donkey anti-rabbit IgG (Jackson ImmunoResearch, West Grove, PA, USA); biotinylated goat anti-rat IgG antibody, horseradish peroxidase-conjugated streptavidin (Vector, Burlingame, CA, USA).

RESULTS Formation of a vascular BM in the embryoid body model To study the role of LM deposition and BM assembly in development of the vascular system, we used the embryoid body model, i.e., clusters of differentiating embryonic stem cells. As described by us and others (14, 19), embryonic stem cells differentiate into ECs that form capillary-like structures within the embryoid body, faithfully mimicking different stages in vasculogenesis and angiogenesis (Fig. 1A) (for a review, see ref. 20). We analyzed the presence and composition of the endothelial BM in this model. Immunostaining of day 10 embryoid bodies showed LM-␣4 (a component of LM411 and LM421) deposition on the surface of the endothelium (Fig. 1B). This agrees with previous observations of murine LM-␣4 in the vascular BM at E13. Immunoblotting showed expression of LM-␣4 as early as day 6 of differentiation in embryoid bodies (Supplemental Fig. 1). LM-␣4 preferentially colocalized with the EC protein CD31/PECAM (platelet-EC adhesion molecule) -1. LM-␣5 (a component of LM511 and LM521), which is present in the mature vascular BM, outlined the vascular structures but was also widely distributed throughout the embryoid body. In accordance, staining with the general LM-pan antibody recognizing the ␣1-, ␤1-, and ␥1chains, and thus 11 out of 16 known isoforms, showed abundant and extensive expression of LMs. Collagen IV, another major component of the vascular BM, displayed an expression pattern similar to that of LM-␣4 but with a slightly wider distribution. Fibronectin, which is known to bind and activate integrin ␣v␤3, thereby affecting angiogenesis, was detected in the EC BM, but also elsewhere (Fig. 1B). To examine the composition of the vascular BM during sprouting angiogenesis, embryoid bodies were cultured in a 3-D matrix of collagen I. Treatment with VEGF induced invasion of ECs into the surrounding matrix, as described previously (14) (Fig. 2). At day 12, embryoid bodies showed intense sprouting of capillarylike structures composed of CD31-positive ECs, surrounded by NG2 and/or ␣SMA-positive pericytes (Fig. 2). Fibronectin, collagen IV, and LM were all present in 1531

Figure 1. Vascular structures in embryoid bodies are embedded in a basement membrane. Wild-type embryoid bodies immunostained (whole mount, day 10) to visualize formation of vessel structures. A) Embryoid body stained for CD31. Box in the left panel indicates magnified area shown in the right panel. Scale bars ⫽ 300 ␮m (left); 50 ␮m (right). B) CD31 immunostaining of embryoid bodies, in combination with immunostaining for LM-␣4, LM-␣5, LM-pan, collagen IV, or fibronectin, as indicated. LM-␣4 (green) expression was restricted to the vasculature (arrowheads, upper left). Collagen IV (red) associated with the vasculature (arrowheads, lower left) and was detected in nonendothelial regions (arrow, lower left). LM-␣5, LM-pan, and fibronectin colocalized with CD31 (arrowheads) but were also abundantly deposited throughout the embryoid body (arrows). Scale bars ⫽ 20 ␮m (bottom right panel); 50 ␮m (all other panels).

the vascular BM stretching from the stalk to the tip cells, embedding both ECs and pericytes (Figs. 2 and 5 and data not shown). Immunostaining to detect heparan sulfate (HS) revealed the presence of HSPGs in the vascular BM. We conclude that vascular structures in embryoid bodies are embedded in an in vivo-like vascular BM, thereby constituting a relevant model for the study of the contribution of LMs to EC development and organization. Vascular development proceeds in the absence of LM deposition The LM-␥1 chain has been detected previously in embryoid bodies differentiated for 48 h (12). Inactivation of the lamc1 gene encoding the LM-␥1 chain in the mouse leads to embryonic death at E 5.5 (21). Embryoid bodies formed from lamc1⫺/⫺ ES cells still sup-

ported formation of ECs and their organization into vascular structures. The resulting vascular plexus were less branched than seen in wild-type cultures, and ECs were more spread out (Fig. 3A, compared with Figs. 1A and 6A, B). As expected, no reactivity against LM-␥1 (Supplemental Fig. 2A) was found, whereas diffuse, predominantly intracellular patches of other LM chains were detected (Fig. 3B). Immunostaining with antibodies for LM-␣4 and LM-␣5 and using the LM-pan antibody consistently demonstrated a complete absence of deposited LMs in embryoid bodies lacking LM-␥1 expression (Fig. 3B). Also, polymerized collagen IV was severely reduced in the vascular BM and instead found evenly dispersed over the culture (Fig. 3B, compare confocal images of lamc1 ⫺/⫺ cultures with wild-type in Fig. 3C). However, increased levels of fibronectin were detected in the absence of LM-␥1 expression (Fig. 3B, C and Supplemental Fig. 3).

Figure 2. BM deposition and pericyte recruitment to ECs in angiogenic sprouts. Angiogenic sprouting of embryoid bodies in a 3-D collagen I in response to VEGF-A165 (day 12). Left: overview of an embryoid body whole-mount stained for CD31 (red) and ␣SMA (green). Right: confocal microscopy analysis of staining for CD31 (red or green as indicated) and for BM components shows deposition of fibronectin (green), collagen IV (red), LM (LM-pan, green) and HS (Hepss1, red). HS was also abundantly expressed on NG2-positive pericytes (white, bottom panel). Arrowheads indicate BM components. Scale bars ⫽ 300 ␮m (left); 20 ␮m (right).


Vol. 22

May 2008

The FASEB Journal


Figure 3. LMs are dispensable for EC differentiation (vasculogenesis). LM-␥1-deficient embryoid bodies immunostained (whole mount, day 10) to visualize vessel structures. A) Embryoid body stained for CD31 illustrates slender structures but also EC sheets consisting of ECs with spread-out morphology (for comparison with wild type, see Fig. 1). Box in the left panel indicates magnified area shown in the right panel. Scale bars ⫽ 300 ␮m (left); 50 ␮m (right). B) Embryoid bodies immunostained for CD31 together with LM-␣4, LM-␣5, LM-pan, collagen IV, and fibronectin, respectively. None of the LMs (LM-␣4, LM-␣5, LM-pan; green) were deposited in the absence of LM-␥1 (arrow in bottom right panel), although diffuse reactivity, most likely representing intracellular LM, was seen (arrowheads in bottom right panel). Collagen IV (red) was lacking in the BM of lamc1⫺/⫺ ECs, although polymerized collagen IV was widely distributed throughout the embryoid body. Fibronectin colocalized with CD31 (arrowheads) but was also generally distributed over the embryoid body. Scale bars ⫽ 20 ␮m (bottom right panel); 50 ␮m (all other panels). C) Confocal analyses of wild-type and lamc1⫺/⫺ embryoid bodies show reduced levels of collagen IV but increased levels of fibronectin in EC BM in the absence of LM-␥1. Arrowheads indicate deposition of LM-␣4 (green), collagen IV (red), and fibronectin (green) to the EC BM. Arrows indicate nonendothelial depositions. In the LM-␥1-deficient embryoid bodies, LM-␣4 staining was markedly absent, collagen IV expression was reduced in the perivascular region, whereas fibronectin was detected at increased levels in the vascular BM. Nuclei are shown in blue. Scale bar ⫽ 20 ␮m.

We next asked whether LM-␥1 deficiency would affect angiogenic sprouting (Fig. 4). Although there was a close to complete lack of LM expression and deposition in angiogenic sprouts from LM-␥1-deficient embryoid bodies, the number and length of sprouts were indistinguishable from those of wild-type embryoid bodies (Fig. 4). Furthermore, the extent of pericyte coating was similar in wild-type and lamc1⫺/⫺ angiogenic sprouts. Collagen IV was expressed, but its perivascular deposition was absent in the sprouts of LM-␥1-deficient embryoid bodies. In contrast, perivascular deposition of fibronectin was clearly seen and vascular BM also contained HSPGs, as detected by HepSS1 immunostaining. We conclude that, despite the dramatic effects on the BM composition with both LM and collagen IV deficiency in the absence of LM-␥1, LAMININS IN VASCULAR DEVELOPMENT

no obvious vascular phenotype was recorded with regard to pericyte investment and ability of EC sprouting. LMs are produced and deposited from the stalk to the tip To investigate which part of the angiogenic sprout is responsible for the deposition of LMs within the vascular BM, we analyzed sprouting tips by high-resolution confocal microscopy. Angiogenic sprouts formed during development are known to be composed of proliferating stalk ECs, headed by a growth-arrested tip cell, which guides the path of the sprout (22). Immunostaining of wild-type embryoid bodies revealed the presence of LMs spanning from the stalk cells to the extending 1533

Figure 4. VEGF-induced invasive angiogenesis is not affected by depletion of BM-associated LMs. LM-␥1-deficient embryoid bodies form angiogenic sprouts in 3-D collagen I in response to VEGF-A165 (day 12). Left: overview of a LM-␥1deficient embryoid body whole-mount stained for CD31 (red) and ␣SMA (green). Right: staining for CD31 (red or green as indicated) and for BM components followed by confocal detection shows deposition of fibronectin (green), HS (Hepss1, red) but reduced levels of collagen IV (red) and absence of LM (green). HS is also abundantly expressed on NG2-positive pericytes (white, bottom panel). Arrowheads indicate BM components. Scale bars ⫽ 300 ␮m (left); 20 ␮m (right).

filopodia of the leading tip cell (Fig. 5 and Supplemental Movie 1). This was true also when staining for LM-␣4 (Supplemental Fig. 4). In the LM-␥1-deficient cells, no extracellular LM deposition was observed, while intracellular reactivity was evident in stalk cells as well as the tip cells indicative of LM production in both cell populations (Fig. 5 and Supplemental Movie 2). This implies that the ECs throughout the sprout produce the vascular BM.

EC differentiation and organization in the presence of exogenous LMs Addition of purified LMs has been shown to rescue or modulate different processes in the embryoid body model (23–26). Figure 6A shows that addition of purified LM111, LM411, or LM511 had no appreciable effect on vascular development in differentiating wildtype embryoid bodies. In contrast, addition of purified, recombinant LMs to the LM-␥1-chain-deficient cultures enhanced EC differentiation, as shown by the additional branch points and slender vessel formation (Fig. 6B). The most prominent effect was seen with LM111 and slightly less in LM411- and LM511-treated cultures, which is consistent with the extent of the BM rescue as determined by LM deposition (Supplemental Fig. 2B). As judged from immunostaining using the LM-pan antibody, none of the treatments resulted in complete restoration of LM deposition. Vascular stability in the absence of a LM-containing vascular BM

Figure 5. LMs are produced and deposited from the stalk to the tip cell filopodia. VEGF-induced sprouting endothelial tips extending from embryoid bodies cultured in 3-D collagen I (day 20). Confocal Z-stack of angiogenic sprouts wholemount stained for CD31 (red) and ␣SMA (green), LM-pan (white), and Hoechst (blue). In the wild type (top panel), LMs embedded the entire sprout from stalk cells to tip cell filopodia (arrowheads). LM-␥1-deficient cultures lacked LM deposition, but intracellular LM was detected (arrows). Asterisks indicate tip cell nuclei. Confocal sections are 5 ␮m in depth. The locations of sections are shown as dotted lines. Scale bars ⫽ 15 ␮m. 1534

Vol. 22

May 2008

LMs have been suggested to influence vascular stability and to serve as a path finding track both during developmental angiogenesis in the zebra fish and during vessel regeneration in the kidney (27, 28). To investigate the effects on vessel stability, survival, and regeneration, we compared angiogenic sprouting in wild-type and LM-␥1-deficient embryoid bodies over time. VEGF was added to the cultures for the first 10 days, followed by withdrawal of VEGF between days 10 and 14, and reintroduction of VEGF again at day 14 until day 18 (Fig. 7). At day 10, sprouts of ECs of wild-type as well as in LM-␥1-null embryoid bodies invaded the gel. Two days after withdrawal of VEGF, signs of disassembly of the sprouts were evident (data not shown). At day 14, the vascular sprouts were almost completely dissolved (one or two short sprouts re-

The FASEB Journal


Figure 6. Effect of exogenous LMs on vessel morphology. Wild-type or LM-␥1-deficient embryoid bodies, untreated or treated with either LM111, LM511, or LM411 on culture slides, were immunostained for CD31 (whole mount, day 10) to visualize the vasculature. A) Differentiation of wild-type ES cells into slender vessel structures was unaffected by addition of LMs. Arrows indicate slender vessels present in each treatment. B) Untreated LM-␥1-deficient embryoid bodies display flattened EC structures (arrowheads). EC morphology was rescued by addition of LM111 and to a lesser extent by addition of LM411 and LM511 (arrows indicate rescued phenotype). Scale bars ⫽ 1 mm (top rows); 50 ␮m (bottom rows, magnification of the central vascular plexus).

mained per embryoid body; Fig. 7B). The kinetics of the retraction of the vessels was similar in wild-type and lamc1⫺/⫺ cultures, suggesting that LM does not enhance EC survival in this setting. In conjunction with EC dissociation, the pericyte coat was also dissolved to a large extent. The kinetics of this process followed that of the EC degradation, i.e., with no apparent difference between wild-type and LM-␥1-deficient embryoid bodies. Reintroduction of VEGF at day 15 promoted restoration of vascular structures (Fig. 7). These structures failed to invade the collagen I matrix and were found close to the embryoid body “core.” We did not observe signs of regrowth of ECs in vascular BM sleeves, as has been shown to occur after interruption of therapeutic VEGF neutralization (28). It has been previously shown LAMININS IN VASCULAR DEVELOPMENT

that tumor vessels lacking mural cell coverage are more sensitive to VEGF depletion than vessels of normal composition (29). Consistent with this report, the few remaining EC sprouts after VEGF withdrawal were completely covered with ␣SMA-positive cells, to an extent never observed with continuous VEGF treatment (Fig. 7B). LMs control the vascular lumen diameter The assembly of a proper BM with deposited LMs has been shown to be a key event in polarization of several cell types, and apical-basal polarization of ECs is a prerequisite for lumen formation. To evaluate the importance of LM deposition for proper lumen 1535

Figure 7. Stability and survival of angiogenic sprouts is not affected by LM coating. Wild-type and lamc1⫺/⫺ embryoid bodies were cultured in collagen I in the presence and absence of VEGF-A165 for three different time periods: days 4 –10, 10 –14, and 14 –18, followed by CD31 immunostaining. A) Continuous treatment with VEGF over the course of all three time periods tested (⫹⫹⫹, i.e., days 4 –18; left panels). Treatment with VEGF during the first time period (days 4 –10), followed by VEGF withdrawal over the following two periods, days 10 –14 (⫹⫺) and 10 –18 (⫹⫺⫺) (middle panels). Treatment with VEGF, days 4 –10, followed by VEGF withdrawal, days 10 –14, and reintroduction of VEGF, days 14 –18 (⫹⫺⫹; right panels). B) Remaining sprouts after depletion of VEGF from days 10 –14 were densely covered with ␣SMA-positive cells (green) in wild-type as well as lamc1⫺/⫺ embryoid bodies. Scale bar ⫽ 50 ␮m.

formation we analyzed angiogenic sprouts formed from wild-type and lamc1⫺/⫺ embryoid bodies by confocal microscopy. Lumen-containing vessels were observed in wild-type as well as in lamc1⫺/⫺ embryoid bodies (Fig. 8A). The number of vessels with a diameter exceeding 36 ␮m was approximately 4⫻ higher in LM-␥1-null embryoid bodies compared to wild-type from day 14 and onwards (Fig. 8B). This finding is consistent with the larger vascular lumens observed in both LM-␣4 and LM-␣5 knockout animals (8, 10). Addition of purified, recombinant LMs did not affect sprout features in wild-type or lamc1⫺/⫺ embryoid bodies (Supplemental Fig. 5 and data not shown). Therefore, in order to analyze rescue of lumen diameter, we used chimeric embryoid bodies to allow directed delivery of endogenous LMs. In embryoid bodies composed of a mixture of lamc1⫺/⫺ and vegfr-2⫺/⫺ ES cells, a partial normalization of the lumen diameters was observed (Fig. 8B). VEGFR-2 is a prerequisite for the expansion, differen1536

Vol. 22

May 2008

tiation and migration of ECs. Thus, ECs in this chimeric setting are derived from the lamc1⫺/⫺ ES cells and lack LM expression, whereas non-ECs (i.e., pericytes) are either of vegfr-2⫺/⫺ origin (producing LMs) or lamc1⫺/⫺ origin. In cultures initially containing 90% lamc1⫺/⫺ cells and 10% vegfr-2⫺/⫺, a patchy pattern of extracellular LMs was observed in the embryoid body sprouts, although at considerably lower levels than in wild type (Fig. 8A). To examine whether the lumen enlargement was caused by differences in cell morphology and/or proliferation, the number of ECs of sprouts with certain diameter from both conditions was counted. No difference in the number of ECs/sprout was recorded when comparing vessels of similar size derived from wild-type and lamc1⫺/⫺ embryoid bodies. However, a correlation was found between increased vessel width and increased number of ECs/ sprout length (Supplemental Fig. 6). We, therefore, investigated possible differences in EC proliferation in wild-type and lamc1⫺/⫺ embryoid bodies during

The FASEB Journal


Figure 8. ECs polarize to form enlarged vascular lumens in the absence of extracellular LMs. VEGF-treated wild-type, LM-␥1-null, and chimeric (lamc1⫺/⫺:flk1⫺/⫺, 9:1) embryoid bodies in a 3-D collagen I matrix (day 18) formed lumenized vessel sprouts. A) Confocal stacks of sprouts stained for CD31 (red) to visualize ECs, NG2 (green) for pericytes, LMs (LM-pan, white), and nuclei (blue) revealed widened lumens in the lamc1-deficient embryoid bodies. Vessels of wild-type origin were covered with LM, whereas the LM-␥1-deficient cells lacked LM coating. Chimeric cultures allowed partial rescue of LM coating. Confocal sections are 5 ␮m in depth. The locations of sections are shown as dotted lines. Scale bars ⫽ 15 ␮m. B) Quantification of vessels with an outer diameter exceeding 36 ␮m at specific time points in the presence of VEGF. C) Confocal sections of sprouts stained for CD31 (red), NG2 (white), nuclei (blue), and BrdU-positive cells (green). Proliferation of both ECs (arrows) and pericytes (arrowheads) is observed. Scale bar ⫽ 50 ␮m. D) Quantification of the number of proliferating ECs/sprout.

the initial steps of lumen formation. Quantification of cells that had incorporated BrdU over a 24 h period revealed increased EC proliferation in the absence of LMs (Figs. 8C, D). Despite large variation within each group, a clear difference in the mean value was noted. This finding suggests that the tendency to increased proliferation in absence of LMs paralleled by increased fibronectin retention may be decisive for the luminal expansion. The increase in lumen size in the absence of LM-␣4 in gene targeted mice has been inferred to be due to a weakening of the vessel wall and consequent widening of the lumen. Our data, however, suggest that LMs either provide or restrict signals to control vascular lumen LAMININS IN VASCULAR DEVELOPMENT

formation in the absence of blood pressure and circulation.

DISCUSSION Here, we show that differentiation and maturation of endothelial precursors (i.e., vasculogenesis) proceed largely undisturbed under conditions of complete lack of LM deposition. Formation of angiogenic structures was not affected with regard to number and length of sprouts, but LM-deficient cultures displayed an increased diameter of the vascular tube. In chimeric embryoid bodies, endogenous LMs produced by non1537

endothelial sources restricted the vessel diameter. These data confirm and extend previous knockout studies in LM-␣4- and LM-␣5-deficient mice. We conclude that the increase in lumen diameter is independent of blood flow, which is lacking in embryoid bodies. Rather, control of lumen diameter by LM deposition in the BM may involve integrin ligation and integrindependent signal transduction in ECs, similar to that suggested for epithelial cells (30). That EC integrinmatrix interactions indeed may be altered in the absence of LM deposition was indicated by increased EC proliferation and the more spread-out morphology of ECs in the lamc1⫺/⫺ embryoid bodies (see Figs. 6 and 8). The presence of LM-␣4 and LM-␣5 chains in the vascular BM of embryoid bodies at all stages of differentiation implies an important role for these LMs in EC biology. Previously, the LM-␣5 chain has not been detected in vascular BM until E13 and has, therefore, not been considered vital for vasculogenesis. The increase of the LM-␣5 chain over time in the vascular BMs suggests a function in stabilization rather than early differentiation of ECs. LM-␣4 is the first ␣-chain detected in the vascular BM during development, and gene inactivation results in a relatively mild embryonic vascular phenotype with enlarged vascular lumens and bleedings. The vascular phenotype decreases after birth, and animals are fertile and unaffected under nonpathological conditions. This surprisingly mild phenotype might be explained by compensatory activity of other LM chains during early development. Notably, no ␣-chains were detected in vascular BMs at birth, but increasing levels of the LM-␣5 chain was evident at later stages (10). Gene inactivation of LM-␥1, which is a component of 10 of the 16 known LMs, leads to embryonic death at E5.5 due to defects in Reichert’s membrane and disturbed implantation (12, 21). To study developmental processes at later stages, Li and co-workers employed ES cells with targeted deletion of the lamc1 gene (24). None of the known LMs were deposited in epithelial BMs in the LM-␥1⫺/⫺ embryoid bodies. Furthermore, collagen IV failed to assemble due to lack of retention to the BM. In this study, however, both LM and collagen IV assembly was rescued by exogenous LM111. Overall, these data agree with our results on the features of vascular BMs in lamc1⫺/⫺ embryoid bodies. Our observations deviate slightly in that polymerization of collagen IV, although considerably reduced, was still evident in the absence of the LM-␥1 chain. Furthermore, increased levels of fibronectin in the LM-␥1deficient embryoid bodies were found. This increase might be a result of compensatory up-regulation of fibronectin, a mechanism not previously described, or more likely due to increased retention of fibronectin due to lack of competition with the LM-␣5 chain for binding to integrins ␣v␤3. The critical role of fibronectin in mammalian biology is demonstrated by the lethality of fibronectin-null mice at E11 with defective 1538

Vol. 22

May 2008

notochord and somite formation as well as vascular malfunctions (31). An elegant study by Parsons et al. (27) showed that inactivation of the LM-␥1 gene in zebra fish leads to a pronounced vascular phenotype. The dorsal aorta in lamc1⫺/⫺ zebra fish formed normally, but no sprouting was seen from intersomitic vessels, a defect possibly related to the defective perinotochordal BM organization. This finding clearly demonstrates the importance of extracellular LMs in patterning of the vasculature in this model. Nevertheless, due to potential contribution of maternally derived LMs at early embryonic development, it did not address the role for LMs in early EC differentiation. Although deficiency in LM deposition affects the morphology of early ECs in the embryoid body model (Fig. 6), we conclude that EC differentiation still occurs. This report provides the first evidence that LMs are dispensable for vasculogenesis in the mammalian system. Fine-tuning of vessel morphogenesis appears to involve LMs, as both LM111 and LM511 noticeably rescued the LM-␥1-deficient phenotype in the embryoid body cultures (see Fig. 6) and resulted in more branch points and slender vessels. Furthermore, collagen IV seems not to be required for early vascular development since it was essentially absent in the BM of LM-␥1-deficient embryoid bodies. In agreement with this observation, the collagen IV knockout mouse shows normal composition of the BM until E9.5 but is lethal around E11 and has a mild vascular phenotype with irregular protrusion of capillaries into the neural layer (32). The degree of BM and pericyte coating has been correlated with vessel stability. Hence, tumor vessels often show a deficient perivascular coat that has been associated with the higher vessel turnover in this condition. In this study, however, even a markedly compromised vascular BM lacking LMs and collagen IV did not affect vessel stability and EC survival. Furthermore, withdrawal of VEGF led to vessel disintegration with similar kinetics in wild-type and lamc1⫺/⫺ embryoid bodies (Fig. 7). It is interesting that angiogenic sprouts remaining after several days of VEGF deficiency were densely coated with pericytes, indicating that pericytes rather than the BM itself are critical in EC survival and vessel stability (14, 28, 33). Alternatively, it is also possible that fibronectin becomes more efficiently retained in the vascular BM in the absence of LM-␥1 and may compensate by mediating survival signals to ECs (34). In conclusion, although our data strongly indicate that LMs are not required for vasculogenesis per se, we concur that fine-tuning of the 3-D aspect of the vascular tube involves LM-dependent signal transduction and that in vivo conditions of flow and mechanical forces would further compromise vascular function under conditions of LM-deficiency. We thank Dr. Jeffrey Miner and Dr. Lydia Sorokin for kindly providing the LM-␣5 and LM-␣4 antiserum, respec-

The FASEB Journal


tively. This study was supported by grants to L.C.W. from the Swedish Science Council (project K2005-32X1255208A) and the Swedish Cancer Foundation (project 3820B05-10XBC).


4. 5.













Olsson, A. K., Dimberg, A., Kreuger, J., and Claesson-Welsh, L. (2006) VEGF receptor signalling—in control of vascular function. Nat. Rev. Mol. Cell. Biol. 7, 359 –371 Nyberg, P., Xie, L., and Kalluri, R. (2005) Endogenous inhibitors of angiogenesis. Cancer Res. 65, 3967–3979 Hellstrom, M., Kalen, M., Lindahl, P., Abramsson, A., and Betsholtz, C. (1999) 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 126, 3047–3055 Sottile, J. (2004) Regulation of angiogenesis by extracellular matrix. Biochim. Biophys. Acta 1654, 13–22 Patton, B. L., Miner, J. H., Chiu, A. Y., and Sanes, J. R. (1997) Distribution and function of laminins in the neuromuscular system of developing, adult, and mutant mice. J. Cell Biol. 139, 1507–1521 Sorokin, L. M., Pausch, F., Frieser, M., Kroger, S., Ohage, E., and Deutzmann, R. (1997) Developmental regulation of the laminin alpha5 chain suggests a role in epithelial and endothelial cell maturation. Dev. Biol. 189, 285–300 Hallmann, R., Horn, N., Selg, M., Wendler, O., Pausch, F., and Sorokin, L. M. (2005) Expression and function of laminins in the embryonic and mature vasculature. Physiol. Rev. 85, 979 – 1000 Miner, J. H., Cunningham, J., and Sanes, J. R. (1998) Roles for laminin in embryogenesis: exencephaly, syndactyly, and placentopathy in mice lacking the laminin alpha5 chain. J. Cell Biol. 143, 1713–1723 Miner, J. H., and Li, C. (2000) Defective glomerulogenesis in the absence of laminin alpha5 demonstrates a developmental role for the kidney glomerular basement membrane. Dev. Biol. 217, 278 –289 Thyboll, J., Kortesmaa, J., Cao, R., Soininen, R., Wang, L., Iivanainen, A., Sorokin, L., Risling, M., Cao, Y., and Tryggvason, K. (2002) Deletion of the laminin alpha4 chain leads to impaired microvessel maturation. Mol. Cell. Biol. 22, 1194 –1202 Miner, J. H., Li, C., Mudd, J. L., Go, G., and Sutherland, A. E. (2004) Compositional and structural requirements for laminin and basement membranes during mouse embryo implantation and gastrulation. Development 131, 2247–2256 Smyth, N., Vatansever, H. S., Murray, P., Meyer, M., Frie, C., Paulsson, M., and Edgar, D. (1999) Absence of basement membranes after targeting the LAMC1 gene results in embryonic lethality due to failure of endoderm differentiation. J. Cell Biol. 144, 151–160 Shalaby, F., Rossant, J., Yamaguchi, T. P., Gertsenstein, M., Wu, X. F., Breitman, M. L., and Schuh, A. C. (1995) Failure of blood-island formation and vasculogenesis in Flk-1-deficient mice. Nature 376, 62– 66 Jakobsson, L., Kreuger, J., Holmborn, K., Lundin, L., Eriksson, I., Kjellen, L., and Claesson-Welsh, L. (2006) Heparan sulfate in trans potentiates VEGFR-mediated angiogenesis. Dev. Cell 10, 625– 634 Kortesmaa, J., Yurchenco, P., and Tryggvason, K. (2000) Recombinant laminin-8 (alpha(4)beta(1)gamma(1)). Production, purification, and interactions with integrins. J. Biol. Chem. 275, 14853–14859 Doi, M., Thyboll, J., Kortesmaa, J., Jansson, K., Iivanainen, A., Parvardeh, M., Timpl, R., Hedin, U., Swedenborg, J., and Tryggvason, K. (2002) Recombinant human laminin-10 (alpha5beta1gamma1). Production, purification, and migration-promoting activity on vascular endothelial cells. J. Biol. Chem. 277, 12741–12748 Ljubimova, J. Y., Lakhter, A. J., Loksh, A., Yong, W. H., Riedinger, M. S., Miner, J. H., Sorokin, L. M., Ljubimov, A. V., and Black, K. L. (2001) Overexpression of alpha4 chain-con-




20. 21. 22.

23. 24.











taining laminins in human glial tumors identified by gene microarray analysis. Cancer Res. 61, 5601–5610 Miner, J. H., Patton, B. L., Lentz, S. I., Gilbert, D. J., Snider, W. D., Jenkins, N. A., Copeland, N. G., and Sanes, J. R. (1997) The laminin alpha chains: expression, developmental transitions, and chromosomal locations of alpha1–5, identification of heterotrimeric laminins 8 –11, and cloning of a novel alpha3 isoform. J. Cell Biol. 137, 685–701 Vittet, D., Prandini, M. H., Berthier, R., Schweitzer, A., MartinSisteron, H., Uzan, G., and Dejana, E. (1996) Embryonic stem cells differentiate in vitro to endothelial cells through successive maturation steps. Blood 88, 3424 –3431 Jakobsson, L., Kreuger, J., and Claesson-Welsh, L. (2007) Building blood vessels–stem cell models in vascular biology. J. Cell Biol. 177, 751–755 Smyth, N., Vatansever, H. S., Meyer, M., Frie, C., Paulsson, M., and Edgar, D. (1998) The targeted deletion of the LAMC1 gene. Ann. N. Y. Acad. Sci. 857, 283–286 Gerhardt, H., Golding, M., Fruttiger, M., Ruhrberg, C., Lundkvist, A., Abramsson, A., Jeltsch, M., Mitchell, C., Alitalo, K., Shima, D., and Betsholtz, C. (2003) VEGF guides angiogenic sprouting utilizing endothelial tip cell filopodia. J. Cell Biol. 161, 1163–1177 Murray, P., and Edgar, D. (2000) Regulation of programmed cell death by basement membranes in embryonic development. J. Cell Biol. 150, 1215–1221 Li, S., Harrison, D., Carbonetto, S., Fassler, R., Smyth, N., Edgar, D., and Yurchenco, P. D. (2002) Matrix assembly, regulation, and survival functions of laminin and its receptors in embryonic stem cell differentiation. J. Cell Biol. 157, 1279 –1290 Li, X., Chen, Y., Scheele, S., Arman, E., Haffner-Krausz, R., Ekblom, P., and Lonai, P. (2001) Fibroblast growth factor signaling and basement membrane assembly are connected during epithelial morphogenesis of the embryoid body. J. Cell Biol. 153, 811– 822 Dixelius, J., Jakobsson, L., Genersch, E., Bohman, S., Ekblom, P., and Claesson-Welsh, L. (2004) Laminin-1 promotes angiogenesis in synergy with fibroblast growth factor by distinct regulation of the gene and protein expression profile in endothelial cells. J. Biol. Chem. 279, 23766 –23772 Parsons, M. J., Pollard, S. M., Saude, L., Feldman, B., Coutinho, P., Hirst, E. M., and Stemple, D. L. (2002) Zebrafish mutants identify an essential role for laminins in notochord formation. Development 129, 3137–3146 Mancuso, M. R., Davis, R., Norberg, S. M., O’Brien, S., Sennino, B., Nakahara, T., Yao, V. J., Inai, T., Brooks, P., Freimark, B., Shalinsky, D. R., Hu-Lowe, D. D., and McDonald, D. M. (2006) Rapid vascular regrowth in tumors after reversal of VEGF inhibition. J. Clin. Invest. 116, 2610 –2621 Benjamin, L. E., Golijanin, D., Itin, A., Pode, D., and Keshet, E. (1999) Selective ablation of immature blood vessels in established human tumors follows vascular endothelial growth factor withdrawal. J. Clin. Invest. 103, 159 –165 Yu, W., Datta, A., Leroy, P., O’Brien, L. E., Mak, G., Jou, T. S., Matlin, K. S., Mostov, K. E., and Zegers, M. M. (2005) Beta1integrin orients epithelial polarity via Rac1 and laminin. Mol. Biol. Cell 16, 433– 445 George, E. L., Baldwin, H. S., and Hynes, R. O. (1997) Fibronectins are essential for heart and blood vessel morphogenesis but are dispensable for initial specification of precursor cells. Blood 90, 3073–3081 Poschl, E., Schlotzer-Schrehardt, U., Brachvogel, B., Saito, K., Ninomiya, Y., and Mayer, U. (2004) Collagen IV is essential for basement membrane stability but dispensable for initiation of its assembly during early development. Development 131, 1619 –1628 Darland, D. C., Massingham, L. J., Smith, S. R., Piek, E., Saint-Geniez, M., and D’Amore, P. A. (2003) Pericyte production of cell-associated VEGF is differentiation-dependent and is associated with endothelial survival. Dev. Biol. 264, 275–288 Wang, J., and Milner, R. (2006) Fibronectin promotes brain capillary endothelial cell survival and proliferation through alpha5beta1 and alphavbeta3 integrins via MAP kinase signalling. J. Neurochem. 96, 148 –159 Received for publication September 9, 2007. Accepted for publication November 8, 2007.


Suggest Documents