by Amber N. Stratman, Michael J. Davis, and George E. Davis VEGF and FGF prime vascular tube morphogenesis and sprouting directed by hematopoietic stem cell cytokines by Amber N. Stratman, Michael J. Davis, and George E. Davis Blood Volume 117(14):3709-3719 April 7, 2011 ©2011 by American Society of Hematology

SCF, IL-3 and SDF-1α synergistically act to promote EC tube morphogenesis in the presence or absence of pericytes and function downstream of VEGF-A/FGF-2 EC priming events. SCF, IL-3 and SDF-1α synergistically act to promote EC tube morphogenesis in the presence or absence of pericytes and function downstream of VEGF-A/FGF-2 EC priming events. (A) SCF, IL-3, and SDF-1α were each provided as morphogenic stimuli for ECs individually or in different combinations and compared with control conditions in 3D collagen matrices. Total EC tube area was quantitated for each condition after 72 hours of culture (n ≥ 10; P ≤ .01). (B) Representative images of control versus SCF/IL-3/SDF-1α–stimulated cultures are shown (Bar equals 100 μm, left) as well as an electron micrograph of EC tube assembly (10 μm, right; n ≥ 10; P ≤ .01). (C) EC-pericyte morphogenic coculture assays were established in the presence of the hematopoietic cytokines individually or in different combinations and compared with control conditions in 3D collagen matrices. EC tube areas were quantified after both 72 and 120 hours. (D) Representative images of the cocultures under control versus SCF/IL-3/SDF-1α–stimulated conditions are shown (bar equals 50 μm), along with an electron micrograph of these cultures demonstrating the relationship between the ECs and pericytes (bar equals 2 μm). (E) EC-only morphogenic assays were established in the presence of each hematopoietic cytokine alone or the 3 factors together (All) versus controls. Cell lysates were collected at day 3 and Western blot analysis performed to assess kinase signaling and protein expression changes that correlate with the ability of hematopoietic cytokines to stimulate tube morphogenesis, and determine synergistically acting promorphogenic signaling pathways. (F) A schematic depicting a 2-step process of EC priming versus morphogenic cues. (G) E6 Quail vitelline vein explants primed with VEGF-A/FGF-2 and placed into morphogenic assays containing either VEGF-A or the combined hematopoietic cytokines. *Significance over control. Fixed cultures were immunostained with QH-1 antibodies. Arrows indicate the vessel explant border (n ≥ 10; P ≤ .01). Amber N. Stratman et al. Blood 2011;117:3709-3719 ©2011 by American Society of Hematology

SCF can be replaced by either Flt-3L or M-CSF to act in conjunction with IL-3 and SDF-1α to stimulate EC tube formation and hematopoietic cytokines facilitate pericyte recruitment to EC tubes and vascular basement membrane matrix assembly in 3D collagen mat... SCF can be replaced by either Flt-3L or M-CSF to act in conjunction with IL-3 and SDF-1α to stimulate EC tube formation and hematopoietic cytokines facilitate pericyte recruitment to EC tubes and vascular basement membrane matrix assembly in 3D collagen matrices. (A) M-CSF and Flt3L were added to replace SCF and were compared with SCF/IL-3/SDF-1α by adding them to IL-3/SDF-1α. In addition, ECs were either primed with FGF alone, VEGF-A/FGF-2, or control and then cultured for 72 hours. After fixation, total EC tube area was determined (n ≥ 10; P ≤ .01). *Denotes significance over control ECs; +denotes significance over the FGF-2–only primed ECs. (B) ECs were left untreated or were primed with VEGF-A/FGF-2 and then incorporated with GFP pericytes and hematopoietic cytokines (SCF/IL-3/SDF-1α) into 3D collagen matrices. Assays were allowed to develop over 5 days and average EC vessel area measured (n ≥ 10; P ≤ .01). *Denotes significance over control. (C,D) Basement membrane matrix deposition was assessed in the presence or absence of EC priming with VEGF-A/FGF-2. Cultures were fixed and stained with antibodies to laminin, collagen type IV, and CD31. Fluorescent images of this staining were acquired, quantitated (C), or overlaid with images of GFP pericytes (D). Bar equals 20 μm. Amber N. Stratman et al. Blood 2011;117:3709-3719 ©2011 by American Society of Hematology

VEGF-A and FGF-2 prime EC tube morphogenic responses to SCF, IL-3, and SDF-1α by up-regulating hematopoietic cytokine receptors. VEGF-A and FGF-2 prime EC tube morphogenic responses to SCF, IL-3, and SDF-1α by up-regulating hematopoietic cytokine receptors. (A,B) ECs were primed for 16 hours with either control, VEGF/FGF, or SCF/IL-3/SDF-1α treatments and then were suspended in 3D collagen matrices with either control, VEGF, or SCF/IL-3/SDF-1α additions. Priming cues are listed first, followed by the morphogenic stimuli. Cultures were fixed after 72 hours, photographed, and quantitated for total EC tube area (n ≥ 10; P ≤ .01). Bar equals 100 μm. (C) Quail vitelline vessels were isolated and the vessels primed overnight with VEGF-A/FGF-2 or control conditions. Explants were placed into collagen gels and were treated with either VEGF-A, hematopoietic cytokines, or control conditions. After 5 days, the vessels were stained with QH-1 to visualize quail ECs and representative images are shown. Bar equals 75 μm. (D) Quantification of EC tube formation from vitelline vessel explants is shown (n ≥ 6; P ≤ .01). (E) Reverse transcription polymerase chain reaction analysis of EC mRNA primed with VEGF-A, FGF-2, or the combination of VEGF-A/FGF-2 versus control treated cells for 16 hours. (F) Western blot analysis of ECs primed with VEGF-A, FGF-2, or the combination of VEGF-A/FGF-2 versus control cells for 16 hours. (G) Western blot analysis of embryonic day 6 quail vitelline vessel explants primed with VEGF-A/FGF-2 versus nonprimed control vessels for 16 hours. *Denotes significance over control-control condition; +denotes significance over the control-factors condition. Amber N. Stratman et al. Blood 2011;117:3709-3719 ©2011 by American Society of Hematology

VEGF-A/FGF-2 priming events are functionally separable and show distinct signaling requirements for hematopoietic cytokine-induced EC tube morphogenesis in 3D collagen matrices. VEGF-A/FGF-2 priming events are functionally separable and show distinct signaling requirements for hematopoietic cytokine-induced EC tube morphogenesis in 3D collagen matrices. (A) ECs were primed with the combination of VEGF-A/FGF-2 for the indicated times to identify the timing requirements for the priming event. After priming, ECs were placed in culture with the hematopoietic factors and fixed after 72 hours (n ≥ 10; P ≤ .01). *Denotes significance over the 0-hour time point; +significance over the 4-hour time point. (B) The indicated VEGF isoforms were tested for their ability to prime ECs for 16 hours. ECs were then placed in culture with the hematopoietic factors and fixed after 72 hours (n ≥ 10; P ≤ .01). *Denotes significance over control; +denotes significance over the FGF-2 alone condition. (C) Inhibitors of p38 MAP kinase (SB203580, 10μM), protein kinase A (Rp-cAMPS, 10μM), protein kinase G (cGMPS, 10μM), Rho kinase (Y27632, 10μM) and MMPs (GM6001, 5μM) were added during the priming step or separately during the tube morphogenesis step to assess whether signaling differences exist between the 2 steps. Cultures were fixed after 72 hours, and total tube area was quantitated (n ≥ 10; P ≤ .01). *Denotes significance from control. Representative images of cultures treated with the indicated inhibitors during the priming versus morphogenesis are shown. Bar equals 100 μm. Amber N. Stratman et al. Blood 2011;117:3709-3719 ©2011 by American Society of Hematology

Hematopoietic cytokines stimulate EC angiogenic sprouting in conjunction with VEGF-A or BMP-4, and the EC transcription factor, Runx1, controls hematopoietic cytokine receptor expression that is necessary for EC morphogenic responses in 3D collagen matrices. Hematopoietic cytokines stimulate EC angiogenic sprouting in conjunction with VEGF-A or BMP-4, and the EC transcription factor, Runx1, controls hematopoietic cytokine receptor expression that is necessary for EC morphogenic responses in 3D collagen matrices. (A,B) The indicated factors were added to the collagen matrix and ECs were seeded on the collagen gel surface and EC sprouting was quantitated (A) after 24 hours (n ≥ 10; P ≤ .01) and photographed (B) from the side. Arrowheads indicate the monolayer surface. (C) ECs were primed with VEGF-A, FGF-2, VEGF-A/FGF-2, or control and then were seeded on the surface of gels that contained the indicated individual or combined factors (n ≥ 10; P ≤ .01). *Denotes significance over control. (D) SiRNA suppression of the transcription factor Runx1 in ECs with the use of 2 independent siRNAs (Smartpool-SP or Single-SN) directed to Runx1 was tested during EC tube formation (top) or EC sprouting (bottom; n ≥ 10; P ≤ .01). *Denotes significant blockade from Luciferase controls. (E) Reverse transcription polymerase chain reaction analysis was performed on ECs treated with siRNAs to Runx1 versus luciferase controls to determine mRNA expression for each of the hematopoietic cytokine receptors as well as VEGFR2 and controls. (F) Western blot analysis was performed to determine protein expression for each of the hematopoietic cytokine receptors as well as VEGFR2 after siRNA suppression of Runx1 in ECs. Amber N. Stratman et al. Blood 2011;117:3709-3719 ©2011 by American Society of Hematology

Role for hematopoietic stem cell cytokines and their receptors during quail developmental vascularization events. Role for hematopoietic stem cell cytokines and their receptors during quail developmental vascularization events. (A) Quail CAM tissue was isolated at day 6 of development, fixed, and double stained for the quail EC-specific marker, QH1, versus c-Kit, IL-3Rα, and CXCR4. Arrows indicate vessel wall borders; arrowheads indicate circulating blood cells. (B) Western blot analysis of embryonic 6-day quail vitelline vessels reveals the presence of hematopoietic cytokine receptors in the vasculature. (C) The chemical inhibitors AMD3100, imatinib, and ISCK03 were injected into quail eggs at day 3 of development (100nM) individually or in combination as well as blocking antibodies to SCF and IL-3 (20 μg/mL) versus controls. Embryos were allowed to develop until day 6. Embryo images reveal cranial and abdominal hemorrhage phenotypes (arrows) in embryos treated with hematopoietic cytokine or receptor antagonists, but not in controls. (D) Histologic analysis of developing quail tissue reveals marked hemorrhage in treated embryos. (E) Data are presented showing the frequency and location of hemorrhage visualized in treated versus control embryos. In addition, survival data as well as tissue mass are indicated (n ≥ 7; P ≤ .01). (F) Quail CAM tissue was collected from control versus imatinib/AMD3100 or α-IL-3/SCF–treated embryos at day 6 of development and immunostained for the quail EC-specific marker, QH1. The number of vessel branch points is quantified (top), as well as the area of nonvascularized tissue space (bottom; n ≥ 10; P ≤ .01). *Significance from control. (G) Left panels are representative sections from control versus α-IL-3/SCF–treated embryos at 6 days of development, where reduced sprouting is observed in forebrain parenchyma. Bar equals 100 μm. Representative QH1 stained images from 6-day quail CAM are shown from the indicated conditions showing marked vascular remodeling defects in the imatinib/AMD3100 or α-IL-3/SCF–treated embryos compared with control. Bar equals 100 μm. Amber N. Stratman et al. Blood 2011;117:3709-3719 ©2011 by American Society of Hematology

A 2-step model of vascular tube morphogenesis and sprouting: VEGF and FGF prime ECs to respond to hematopoietic stem cell cytokines to form tubes and sprout in 3D matrices. A 2-step model of vascular tube morphogenesis and sprouting: VEGF and FGF prime ECs to respond to hematopoietic stem cell cytokines to form tubes and sprout in 3D matrices. (A) A schematic is shown depicting the 2-step process of EC priming followed by tube morphogenesis and associated downstream effects connected to each step, including up-regulation of the hematopoietic cytokine receptors after VEGF-A/FGF-2 priming events, hematopoietic cytokine-induced morphogenesis, and vessel stabilization through the recruitment of pericytes and basement membrane matrix deposition. A Venn diagram compartmentalizes each growth factor on the basis of its primary function as either a priming or promorphogenic molecule. FGF-2 has strong EC priming activity but does facilitate the action of promorphogenic hematopoietic cytokines. VEGF has strong EC priming activity and also facilitates promorphogenic hematopoietic cytokines during EC sprouting events. (B) Images of VEGF/FGF-primed EC cultures stimulated with VEGF-A versus the hematopoietic factors as morphogenic cues (top). Bar equals 20 μm. VEGF/FGF primed vitelline vessel explants stimulated with VEGF-A versus the hematopoietic factors as morphogenic cues (bottom). Bar equals 100 μm. Amber N. Stratman et al. Blood 2011;117:3709-3719 ©2011 by American Society of Hematology