Blood vessels are critical for the delivery of oxygen and nutrients to all cells in the body. To properly function, blood vessels and their primary components must develop and mature into a healthy network, capable of dynamic alterations to meet new needs of the body. The early genetic and molecular programs that "push" the vasculature to develop are the same programs that reactivate when there are normal changes to the body such as injury, muscle growth or decline, or aging; and when pathologies arise like cancer, stroke, and diabetes. Therefore, it is crucial to understand how the vasculature develops into a healthy system by studying all components as they mature. Endothelial cells that comprise the vessels themselves are joined by specialized partner cells called pericytes that help guide and mature vessel growth. Pericytes lie elongated along endothelial cells and have multiple points of contact with the endothelium. In this position, pericytes assist in cell-cell communication and even blood flow regulation in the microvasculature. To study the relationship between endothelial cells and pericytes during development, we observed vascular morphology in three and four dimensions, as well as the genetic and molecular mechanisms underlying how these cells are recruited and interact in several experimental models. Thus, to thoroughly analyze the morphology of these vessels, we developed a rigorous methodology using a MATLAB program to determine the colocalization and coverage of pericytes associated with vessels in large image sets. After developing analytical methods to investigate all the components of the blood vessel wall, we expanded our investigation of how pericytes and other aspects of microvasculature develop in animal models, specifically a more commonly used murine model for vascular development and for treatment of human diseases. Our findings of vascular development in mice suggest that there are important differences in how human and mouse brain blood vessels form. Therefore, studies using mice must be carefully designed to account for these discrepancies. Additionally, research into why human and mouse neurovascular development and maturation are different can aid in the development of improved experimental models to better treat human pathologies. / Doctor of Philosophy / Blood vessels have the crucial job of delivering oxygen and nutrients to all the cells in the body. To perform this duty, blood vessels- and the components that make them- must develop and mature into a healthy network, capable of altering itself to meet new needs of the body. The early programs that “push” the vessel system to develop are the same programs that reactivate when there are normal changes to the body such as injury, muscle growth or decline, or aging; and when abnormal diseases arise like cancer, stroke, and diabetes. Therefore, it is critical to understand how blood vessels develop into healthy systems by studying all of their components as they mature. Endothelial cells that comprise the vessels themselves are joined by specialized partner cells called pericytes that help guide and mature vessel growth. Pericytes lie elongated along endothelial cells and have multiple points of contact with the endothelium. In this position, pericytes assist in cell-cell communication and even blood flow regulation in smaller vessels called capillaries. To study the relationship between endothelial cells and pericytes during development, we observed vascular anatomy in three and four dimensions, as well as mechanisms underlying how these cells come together and interact in several experimental models. Thus, to thoroughly analyze the morphology of these vessels, we developed a rigorous methodology using a MATLAB program to determine the colocalization and coverage of pericytes associated with vessels in large image sets. After developing analytical method to investigate all the components of the blood vessel wall, we expanded our investigation of how pericytes and other aspects of blood vessels develop in animal models, specifically a more commonly used animal model for vascular development and for treatment of human diseases. Our findings of vascular development in mice suggest that there are important differences in how human and mouse brain blood vessels form. Therefore, studies using mice must be carefully designed to account for these discrepancies. Additionally, research into why human and mouse neurovascular development and maturation are different can aid in the development of improved experimental models to better treat human illness and injury.

Parthenocarpy, or seedless fruit development, has an agronomic importance in many horticultural crops. In most fruit, fertilization or seed set usually determines whether fruit growth is sustained. Naturally occurring parthenocarpy results from a genetic lesion that permits fruit to develop in the absence of fertilization and seed development. Parthenocarpy can also be induced artificially with cytokinin, gibberellin or auxin plant growth regulators applied to anthesis pistils. This thesis describes genetic research using Arabidopsis as a model plant to identify integral mechanisms that control parthenocarpy and the initiation of fruit development. The growth and structure of the Arabidopsis pistil was determined post-fertilization. Experiments were designed to understand how plant growth regulators induce Arabidopsis silique (fruit) development in emasculated anthesis stage pistils. Exogenous gibberellin (GA3) induced growth and cellular differentiation most comparable to pollinated pistils. Dependencies on gibberellins during silique development were examined in mutants defective for gibberellin biosynthesis (ga1, ga4-1, ga5-1) or perception (spy-4, gai-1). Although exogenous GAs are effective at inducing parthenocarpy, mutant studies concluded that GAs are not the sole cue for fruit development in Arabidopsis. Mutants blocked in GA perception could develop siliques in response to pollination, auxin, cytokinin but not to exogenously applied gibberellins. Silique structure in pollinated gai-1 and ga5-1 provided strong evidence for a model supporting evidence of an auxin-like signal regulating structural development and that GAs limit anticlinal cellular division. A specialized function for GAI and related GRAS family members in controlling cellular division during fruit development was uncovered. A mutant that forms parthenocarpic siliques without fertilization (fwf), was also characterized. The presence of surrounding floral whorls reduced the extent of parthenocarpic silique formation in fwf. Silique growth in the fwf background was examined when hormone perception, ovule and carpel identity functions were removed genetically. This established that FWF functions independent of GAI-mediated GA perception. Carpel identity conferred by FUL was critical for parthenocarpic silique elongation and ovule development beyond integument initiation, nucellar specification and subsequent morphogenesis, was essential for parthenocarpic silique development in fwf. Silique elongation occurs over a four-day period post-pollination or post-anthesis. This coincides with a similar time period in which fwf ovules remained receptive to fertilization. These observations are congruent with the hypothesis that FWF potentially represses a signal transduction process initiated within the ovule that mediates subsequent transition from carpel to silique development. Further analysis revealed that aberrant testa shape (ats) a mutant defective in integument formation enhanced parthenocarpic development in fwf, indicating that an ovule located repressor other than fwf can function to affect silique formation. Other studies have shown that ethylene can modulate auxin-dependent growth in both aerial and root tissues by altering both polar and lateral auxin transport. The contribution of ethylene perception to signal transduction between ovule and carpel was also genetically assessed. Constitutive ethylene responses, conferred by ctr1-1, enhanced cellular expansion in fwf and also the autonomous silique development in fis-2, which develops autonomous endosperm. ats ctr1-1 and ino ctr1-1 double mutants were also found to be parthenocarpic. This indicates that ethylene perception and integumentary structure play an important role in autonomous silique development, conceivably by changing the polar and lateral movement of an auxin-like signal within the integumentary tissues of the ovule. fwf and ats were fine mapped on chromosome 5 of Arabidopsis. Candidate genes were identified corresponding to both mutations but only the identity of FWF was established. Auxin Response Factor 8 (ARF8) was cloned and sequenced from the fwf mutant background. The gene encodes a protein with a amino-terminal DNA binding domain and a carboxy-terminal protein binding domain which homo- and hetero- dimerizes with other ARF or Aux / IAA class proteins. ARF8 sequence from fwf mutants encoded a mutation in the translation start site. Complementation of fwf plants by the transformation of wild type copies of ARF8 into fwf plants was hampered by reduced transformation efficiency. However wild type L.er and No.O plants transformed with mutant copies of ARF8 were obtained in higher frequency, and these formed parthenocarpic siliques when primary transformants were emasculated. This indicated that an interfering protein is produced from the mutated ARF8 gene that has altered regulatory activity. Sequence analysis indicated this and found that interference resulted from functional activity of the Q-rich and carboxy-terminal domains of the ARF8 protein. This inference is consistent with other published molecular data, which has demonstrated that the carboxy-terminal domain, together with the Q-rich region of selected ARF members, can activate auxin-responses. Thus the FWF / ARF8 protein may have a dual role, repressing carpel growth development through the DNA binding domain and then ensuring activation of silique development through the carboxy-terminal domain. The combined molecular and genetic data has been used to construct models concerning the genetic control of silique development. The first model considers the role of plant hormones and how signals from floral whorls surrounding the carpel and from within the ovule control silique growth. A model is also presented for the control of adaxial growth and development of the outer integument by the INNER NO OUTER gene. Finally the role of FWF and SPY in controlling floral tissue identity and boundary tissue specification is considered in a third model. Modification of the FWF / ARF8 gene could be used as a tool to improve fruit set and retention in horticultural crops, in addition to creating seedless parthenocarpic fruit. / Thesis (Ph.D.)--Agriculture and Wine, 2001.

京都大学 / 0048 / 新制・論文博士 / 博士(医学) / 乙第13213号 / 論医博第2163号 / 京都大学大学院医学研究科脳統御医科学系専攻 / (主査)教授 山下 潤, 教授 髙橋 良輔, 教授 木村 剛 / 学位規則第4条第2項該当 / Doctor of Medical Science / Kyoto University / DFAM

Krüppel-like factor 2 (KLF2) is expressed in endothelial cells in the developing heart, particularly in areas of high shear stress, such as the atrioventricular (AV) canal. KLF2 ablation leads to myocardial thinning, high output cardiac failure and death by mouse embryonic day 14.5 (E14.5) in a mixed genetic background. This work identifies an earlier and more fundamental role for KLF2 in mouse cardiac development in FVB/N mice. FVB/N KLF2-/- embryos die earlier, by E11.5. E9.5 FVB/N KLF2-/- hearts have multiple, disorganized cell layers lining the AV cushions, the primordia of the AV valves, rather than the normal single layer. By E10.5, traditional and endothelial-specific FVB/N KLF2-/- AV cushions are hypocellular, suggesting that the cells accumulating at the AV canal have a defect in endothelial to mesenchymal transformation (EMT). E10.5 FVB/N KLF2-/- hearts have reduced glycosaminoglycans in the cardiac jelly, correlating with the reduced EMT. However, the number of mesenchymal cells migrating from FVB/N KLF2-/- AV explants into a collagen matrix is reduced considerably compared to wild-type, suggesting that the EMT defect is not due solely to abnormal cardiac jelly. Echocardiography of E10.5 FVB/N KLF2-/- embryos indicates that they have abnormal heart function compared to wild-type. E10.5 C57BL/6 KLF2-/- hearts have largely normal AV cushions. However, E10.5 FVB/N and C57BL/6 KLF2-/- embryos have a delay in the formation of the atrial septum that is not observed in a defined mixed background. KLF2 ablation results in reduced Sox9, UDP-glucose dehydrogenase (UGDH), Gata4 and Tbx5 mRNA in FVB/N AV canals. KLF2 binds to the Gata4, Tbx5 and UGDH promoters in chromatin immunoprecipitation assays, indicating that KLF2 could directly regulate these genes. Thus KLF2 plays a role in EMT, through its regulation of important cardiovascular genes. E10.5 FVB/N KLF2-/- embryos show gaps in the endothelial lining at the dorsal aorta and a number of blood cells localized outside the aorta suggesting either hemorrhaging or inability of the hematopoietic progenitors to reach the aortic endothelium and enter circulation. This is not observed in KLF2-/- embryos in a mixed genetic background. In conclusion, KLF2-/- cardiovascular phenotypes are genetic background-dependent. KLF4 is another member of the Krüppel-like transcription factor family phylogenetically close to KLF2. It is known to play an important role in vascular regulation. Our studies show that in vascular development KLF4 plays a complementary role to KLF2, indicated by cranial hemorrhaging in E9.5 KLF2-/-KLF4-/- embryos in an undefined mixed background. This phenotype is absent in either of the single knockouts. The role of KLF2 and KLF4 in vascular development has not been studied as much as adult vascular regulation. This study begins to define the roles of these two transcription factors in development of blood vessels. Congenital heart and valve defects are a common cause of infant mortality. KLF2 has never been studied in this context. Thus this work is important for a better understanding of the biology of valve development.

Les complications vasculaires rétiniennes constituent des évènements qui peuvent être observés au cours de rétinopathies pouvant être à l’origine d’une cécité à tous les stades de la vie. Ces complications concernent particulièrement la rétinopathie du prématuré, la rétinopathie diabétique et la dégénérescence maculaire liée à l’âge. Les lipides offrent de nombreuses possibilités pour prévenir et éventuellement freiner le développement de ces rétinopathies. Parmi eux, la classe des plasmalogènes est particulièrement riche en acides gras poly-insaturés (AGPI), qui sont libérés par une phospholipase indépendante du calcium (iPLA2) et qui sont précurseurs de métabolites biologiquement actifs. Certains de ces métabolites sont connus pour être impliqués dans la modulation de l’angiogenèse rétinienne. L’objectif de ce travail de thèse a été d’évaluer l’implication des plasmalogènes dans le développement vasculaire rétinien par l’intermédiaire de la libération des AGPI par la iPLA2. Pour vérifier cette hypothèse, nous avons caractérisé les évènements cellulaires et moléculaires du développement vasculaire rétinien postnatal chez un modèle animal d’inhibition de la iPLA2 rétinienne que nous avons préalablement développé, ceci de manière comparative avec un modèle de déficience totale en plasmalogènes. Nous avons également tenté de mettre en évidence de potentielles altérations du métabolisme des plasmalogènes chez au cours d’une rétinopathie à composante vasculaire chez l’homme, la rétinopathie diabétique. Nos résultats ont suggéré que les plasmalogènes sont indispensables pour le développement physiologique des vaisseaux rétiniens. Ils seraient impliqués dans le contrôle de la formation de la trame astrocytaire et la mise en place du réseau endothélial par l’intermédiaire des AGPI libérés par la iPLA2. Les mécanismes moléculaires impliqueraient la voie des Angiopoïétines-Tie sans affecter celle du VEGF. Chez l’homme, nous avons noté une réduction des AGPI circulants, en particulier l’acide docosaexanéïque et l’acide arachidonique, sur les phosphatidyl-éthanolamines chez tous les patients diabétiques avec ou sans rétinopathie diabétique, sans implication des formes plasmalogènes. Nos résultats suggèrent une implication du métabolisme des plasmalogènes dans le contrôle du développement vasculaire en période péri-natale mais pas au cours de la rétinopathie diabétique. Ce contrôle serait exercé par l’intermédiaire des AGPI libérés par la iPLA2. / Retinal vascular complications are secondary events of several retinopathies that result in blindness at all ages. Such complications can be observed in retinopathy of prematurity, diabetic retinopathy and age-related macular degeneration. Lipids, and particularly polyunsaturated fatty acids (PUFAs), display beneficial properties in the prevention of such retinopathies. Among the different lipid classes, the plasmalogen subclass is particularly interesting since it is known to be rich in PUFAs. These PUFAs are known to be released by a calcium-independent phospholipase (iPLA2) and further converted into biologically active metabolites. Some of these metabolites are known to be involved in the modulation of retinal angiogenesis. The aim of this work was to evaluate the involvement of plasmalogens in retinal vascular development through PUFA release by iPLA2. To check this hypothesis, we have comparatively characterized cellular and molecular mechanisms of postnatal retinal vascular development in an animal model of retinal iPLA2 inhibition as well as in a model of plasmalogens deficiency. On the other hand, we have attempted to identify potential alterations in plasmalogen metabolism in diabetic retinopathy. Our results suggest that plasmalogens are essential for the physiological development of retinal vessels. They are involved in the control of astrocyte template formation and the development of the primary vascular network through PUFA released by iPLA2. Molecular mechanisms by which PUFAs from plasmalogens control retinal vascular development involve Angiopoietin-Tie pathways, without affecting those involving VEGF. In the human study, we have observed a decrease in the bioavailability of circulating PUFAs, and especially docosaexaneic acid and arachidonic acid binded to phosphatidyl-ethanolamine in all diabetic patients with or without diabetic retinopathy. Plasmalogens were not involved in these modifications. Our results suggest that plasmalogen metabolism is involved in the control of primary vascular growth during retinal development but not in diabetic retinopathy. Plasmalogens may control early steps of retinal vascular development through the release of PUFAs by iPLA2.

Plants reduce inorganic carbon to synthesize biomass that is comprised of mostly polysaccharides and lignin. Growth is intricately regulated by external cues such as light, temperature, and water availability and internal cues including those generated by the circadian clock. While many aspects of polymer biosynthesis are known, their regulation and distribution within the stem are poorly understood. Plant biomass is perhaps the most abundant organic substance on Earth and can be used as feedstock for energy production. Various grass species are under development as energy crops yet several of their attributes make them challenging research subjects. Brachypodium distachyon has emerged as a grass model for food and energy crop research. I studied rhythmic growth, a phenomenon important to understanding how plant biomass accumulates through time, and vascular system development, which has biofuel feedstock conversion efficiency and yield. Growth rate changes within the course of a day in a sinusoidal fashion with a period of approximately 24 hours, a phenomenon known as rhythmic growth. Light and temperature cycles, and the circadian clock determine growth rate and the timing of rate changes. I examined the influences of these factors on growth patterns in B. distachyon using time-lapse photography. Temperature and, to a lesser extent, light influenced growth rate while the circadian clock had no noticeable effect. The vascular system transports important materials throughout the plant and consists of phloem, which conducts photosynthates, and xylem, which conducts water and nutrients. The cell walls of xylem elements and ground tissue sclerenchyma fibers are comprised of cellulose, hemicelluloses, and lignin. These components are important to alternative energy research since cellulose and hemicellulose can be converted to liquid fuel, but lignin is a significant inhibitor of this process. I investigated vascular development of B. distachyon by applying various histological stains to stems from three key developmental. My results described in detail internal stem anatomy and demonstrated that lignification continues after crystalline cellulose deposition ceases. A better understanding of growth cues and various anatomical and cell wall construction features of B. distachyon will further our understanding of plant biomass accumulation processes.

Große Arterien spalten sich wiederholt und bilden so arterielle Bäume. Die Menge des Blutflusses bestimmt, welche Teilgefäße stabilisiert oder entfernt werden, weshalb arterielle Bäume normalerweise physisch voneinander getrennt sind. Manchmal formen sich aber kleine Gefäßverbindungen zwischen arteriellen Bäumen, Kollaterale, und bilden eine bedeutende Ausnahme vom Konzept der blutflussvermittelten Netzwerkoptimisierung. Die Funktion von Kollateralen und Mechanismen, die sie stabilisieren sind bisher nicht erforscht. Kollaterale sind von medizinischem Belang, da sie im Falle einer Gefäßverstopfung als natürlicher Bypass fungieren und so eine Ischämie effizient mildern. Ziel der vorliegenden Arbeit war, das grundlegende Verständnis über die Bildung und Stabilisierung von Kollateralen zu verbessern, indem die generelle Rolle von rabep2 während der Blutgefäßentwicklung untersucht wurde, einem Gen welches kürzlich mit unterschiedlicher Kollateraldichte und Schwere von Schlaganfällen bei Mäusen assoziiert wurde. In meinen Studien untersuchte ich auch RABEP1, ein Protein welches hohe strukturelle Ähnlichkeit mit RABEP2 aufweist, um zu verstehen ob beide Proteine ähnliche Funktionen in der Blutgefäßentwicklung haben. Ich nutzte Knock-downs von rabep1 und rabep2 im Zebrafisch und in HUVEC, um die Rollen der Gene in der Entwicklung und Stabilisierung vorhandener Gefäße zu untersuchen. Dabei fand ich heraus, dass beide Proteine für die ordnungsgemäße Bildung und den Erhalt von Blutgefäßen im Zebrafisch essenziell sind. Mithilfe des knock-down in HUVEC analysierte ich, welche zellulären Mechanismen des Endothels durch RABEP1 und RABEP2 kontrolliert werden und so die schweren vaskulären Defekte im Zebrafisch bedingen. Zusammenfassend zeigt meine Arbeit, dass ein penibles Gleichgewicht zwischen RABEP1 und RABEP2 Expression notwendig ist, um eine ordnungsgemäße Funktion der Endothelzellen und eine korrekte Entwicklung und Erhaltung des Blutgefäßsystems zu gewährleisten. / Major arteries form individual arterial trees by branching repeatedly. Vascular adaption through blood flow-mediated network remodelling removes vessel segments with poor flow, leading to physical separation of individual arterial trees. Sometimes small vessel connections between arterial trees, called collaterals, are formed and stabilized. Collaterals normally receive low levels of blood flow and therefore represent notable exceptions to the concept of blood-flow mediated network optimisation. The function of collaterals, and mechanisms that form and stabilize them, are not yet understood. Collaterals are of major clinical importance, as they can rapidly enlarge its diameter and act as natural bypass in case of occlusion, thereby efficiently temper the severity of ischemia. The present work aimed to advance the fundamental understanding of how collateral vessels are formed and stabilized, by investigating the role of rabep2 on the developing vasculature, a gene that has recently been associated with differences in collateral density and stroke severity in mice. In my approaches, I included RABEP1, which shares high structural similarity to RABEP2, to investigate whether both proteins share similar functions. I used knock-downs of rabep1 and rabep2, both in zebrafish and Human Umbilical Vein Endothelial Cell (HUVEC), to investigate their specific role during development and maintenance of blood vessels. With these approaches, I discovered that both proteins are essential for proper establishment and maintenance of the zebrafish vasculature. The knock-down in HUVEC helped me to dissect cellular mechanisms of endothelial cells in which RABEP1 and RABEP2 play crucial roles, in order to gain a deeper understanding of the mechanisms causing the severe defects in zebrafish vasculature. Together, I showed that a tight equilibrium of RABEP1 and RABEP2 expression is needed for proper function of endothelial cells and proper development and maintenance of the vasculature.

Advancements in clinical care have led to a growing cohort of preterm-born individuals now entering adulthood. Before birth, such adults were often exposed to a suboptimal intrauterine environment, and after delivery, key developmental stages that would normally occur in utero during the third trimester had to take place under ex utero physiological conditions. Through detailed cardiovascular phenotyping, this thesis investigates the cardiovascular changes in preterm-born young adults, utilising a cohort of individuals with data collection since recruitment at birth. The detailed perinatal information was first used to design nested case-control studies to investigate the effects of early lipid and glucocorticoid exposure on long-term cardiovascular physiology in individuals born preterm. It was demonstrated that intravenous lipid administration leads to an artificial elevation of total cholesterol levels in immediate postnatal life, which is associated with long-term changes in aortic and left ventricular function proportional to the degree of cholesterol elevation. Additionally, exposure to antenatal glucocorticoids relates to a regional increase in aortic arch stiffness in young adulthood, as well as changes in glucose metabolism. It was then shown that young adults born preterm have increased left ventricular mass, out of proportion to blood pressure, and a unique three-dimensional left ventricular geometry, with reduced systolic and diastolic function compared to term-born controls. Similarly, they also show distinct differences in the right ventricle, with increased right ventricular mass and a proportion having clinically impaired right ventricular systolic function. Finally, it was demonstrated that preterm-born individuals have increased circulating levels of antiangiogenic factors in young adulthood, which relate to capillary rarefaction and blood pressure elevation. These findings are of considerable public health relevance given that nearly 10% of births are now preterm. Understanding whether modification of these variations in cardiovascular structure and function prevent the development of cardiovascular disease in this growing subgroup of the population will be of future interest.

During embryogenesis and post-embryonic patterning, active transport of the phytohormone auxin, reflected in the expression of the Arabidopsis PIN family of auxin efflux mediators, generates local auxin distributions that are crucial for correct organ and tissue specification. Polar auxin transport routes have also long been postulated to regulate vein formation in the leaf. The molecular identification of PIN proteins has made it possible to investigate this hypothesis further by visualizing auxin transport routes in developing leaves. In Arabidopsis leaf primordia, PIN1 is expressed before the earliest known markers of vascular identity, in domains that are gradually restricted to sites of vein formation. PIN1 polarity indicates that auxin is directed towards distinct “convergence points” (CPs) in the marginal epidermis, from which it defines the sites of major vein formation. Within incipient veins, PIN1 polarity indicates drainage of auxin into preexisting veins, such that veins connected at both ends display two divergent polarities. Local auxin application triggers the formation of ectopic CPs and new veins, demonstrating the sufficiency of auxin as a vein-specifying signal. However, not all PIN1-labeled auxin transport routes differentiate as veins: Minor veins are initially unstable, suggesting local competition for auxin transport. Expression of ATHB8, a marker of vascular cell selection, correlates with enhanced PIN1 expression domain (PED) stability and vascular differentiation. Auxin application and auxin transport inhibition reveal that both CP formation in the epidermis and subepidermal PED dynamics are auxin-dependent and self-organizing. Furthermore, normal auxin perception through the ARF-Aux/IAA signaling pathway is required for the restriction of PIN1-mediated auxin transport to narrow subepidermal domains. ARF-Aux/IAA signaling is known to control auxin transport through the regulation of PIN1 dynamics, but the mechanism of this regulation is unclear. It is here shown that two redundantly acting AUXIN RESPONSE FACTOR (ARF) transcription factors, ARF5/MONOPTEROS (MP) and ARF7/NPH4, jointly regulate both PIN1 expression and localization during lateral root patterning in Arabidopsis, in part through the direct transcriptional activation of PIN1 by MP. Taken together, these results indicate that feedback between PIN-mediated auxin transport and ARF-Aux/IAA signaling regulates the patterning of root and shoot organs.

During embryogenesis and post-embryonic patterning, active transport of the phytohormone auxin, reflected in the expression of the Arabidopsis PIN family of auxin efflux mediators, generates local auxin distributions that are crucial for correct organ and tissue specification. Polar auxin transport routes have also long been postulated to regulate vein formation in the leaf. The molecular identification of PIN proteins has made it possible to investigate this hypothesis further by visualizing auxin transport routes in developing leaves. In Arabidopsis leaf primordia, PIN1 is expressed before the earliest known markers of vascular identity, in domains that are gradually restricted to sites of vein formation. PIN1 polarity indicates that auxin is directed towards distinct “convergence points” (CPs) in the marginal epidermis, from which it defines the sites of major vein formation. Within incipient veins, PIN1 polarity indicates drainage of auxin into preexisting veins, such that veins connected at both ends display two divergent polarities. Local auxin application triggers the formation of ectopic CPs and new veins, demonstrating the sufficiency of auxin as a vein-specifying signal. However, not all PIN1-labeled auxin transport routes differentiate as veins: Minor veins are initially unstable, suggesting local competition for auxin transport. Expression of ATHB8, a marker of vascular cell selection, correlates with enhanced PIN1 expression domain (PED) stability and vascular differentiation. Auxin application and auxin transport inhibition reveal that both CP formation in the epidermis and subepidermal PED dynamics are auxin-dependent and self-organizing. Furthermore, normal auxin perception through the ARF-Aux/IAA signaling pathway is required for the restriction of PIN1-mediated auxin transport to narrow subepidermal domains. ARF-Aux/IAA signaling is known to control auxin transport through the regulation of PIN1 dynamics, but the mechanism of this regulation is unclear. It is here shown that two redundantly acting AUXIN RESPONSE FACTOR (ARF) transcription factors, ARF5/MONOPTEROS (MP) and ARF7/NPH4, jointly regulate both PIN1 expression and localization during lateral root patterning in Arabidopsis, in part through the direct transcriptional activation of PIN1 by MP. Taken together, these results indicate that feedback between PIN-mediated auxin transport and ARF-Aux/IAA signaling regulates the patterning of root and shoot organs.