News Physiol Sci 18: 65-70, 2003;
doi:10.1152/nips.01417.2002
1548-9213/03 $5.00
News in Physiological Sciences, Vol. 18, No. 2, 65-70,
April 2003
© 2003 Int. Union Physiol. Sci./Am. Physiol. Soc.
Angiogenesis and Vascular Remodeling by Intussusception: From Form to Function
Haymo Kurz1,
Peter H. Burri2 and
Valentin G. Djonov2
1 Institute of Anatomy II, University of Freiburg, D-79104 Freiburg, Germany; and
2 Institute of Anatomy, University of Bern, CH-3012 Bern, Switzerland
Address for reprint requests and other correspondence: V. G. Djonov, Institute of Anatomy, Buehlstrasse 26, University of Berne, CH-3012 Berne, Switzerland (E-mail djonov{at}ana.unibe.ch).
 |
Abstract
|
|---|
During most instances of angiogenesis, not only are the capillaries or terminal vessels generated and modified, but the supplying vascular system is subjected to remodeling as well. Intussusception, i.e., transluminal pillar formation, is one essential mechanism for growth, arborization, bifurcation remodeling, and pruning. Complex and efficient vascular beds can thus be generated by local interactions between vascular cells and hemodynamic conditions.
|
Introduction
|
|---|
In vertebrates, the cardiovascular system is the first to function for the same purpose throughout the entire lifespan of the individual. It generates pressure for the displacement of blood within it and of metabolites through its walls. However, initial steps of its formation are laid down "hard-wired" in molecular networks that work independently from blood flow and metabolic demands, and circulation commences even somewhat earlier than would be required for supporting embryonic growth (15). Yet as soon as the pulsating heart and the primordial vasculature form a closed circuit, the components of the vessel wall are exposed not only to pressure-induced tangential wall stress but also to flow-induced shear stress (15). Clearly, the latter relates most directly to internal friction of the system and hence to cardiac energy consumption, whereas the former relates to wall thickness and hence to its consumption of material and space. If we assume overall energetic and structural "costs" to be minimized during evolution, we expect the structure of the circulatory system to represent an optimized solution for this "cost minimization" problem.
It has long been recognized that, in fluid transport systems in which only one pressure source, the heart, provides the driving force for transport through billions of minuscule capillaries, one optimizing feature is a system of hierarchical bifurcations. However, it is not so easy to determine by which mechanisms such perfused systems are generated, how they are adapted in response to hemodynamics during development, and whether optimized structures and functions emerge (10). We will address these questions here using an embryonic system, because we are convinced that we can better understand and influence pathological conditions of the adult once we have understood how new blood vessels are made and remodeled in the embryo.
Although it is becoming more widely recognized that multiple mechanisms are involved in angiogenesis ("making vessels" in its widest sense), we want to focus here on intussusception, i.e., transluminal pillar formation, as one particular way of expanding and modifying vessels (6, 13). Interestingly, this mechanism of nonsprouting angiogenesis not only occurs in many organs and at different levels of the system but may be directly involved in structural remodeling for optimization of vessel form and function.
|
A model system for studying the development of vessel networks
|
|---|
To assess dynamic processes in a perfused system, in vivo observation of flow and of structural changes in the vessel wall is strongly desirable. Few organs exist that allow prolonged (several hours) videomicroscopy; among them is the chicken chorioallantoic membrane (CAM). The CAM serves as the major respiratory and calcium-resorbing organ in the avian egg that covers the eggshell from inside. It originates from fusion of the angiogenic mesodermal allantois to the ectodermal chorion and thus corresponds to the placenta of mammals. It not only produces its own blood islands and endothelial cells (ECs), but it also has a well-developed lymphatic drainage system. The vascular bed of the CAM expands at an extraordinary speed for ~1 wk and is transformed from a single layer of capillaries into a two-layered system: the gas-exchanging capillary plexus and the chorionic epithelium merge, and arteries and veins remain "underneath" in the mesenchyme, where they show a typical ramified bifurcation pattern. Most importantly, as in other respiratory organs, oxygenation is highest in the capillary bed and in veins, whereas arteries carry blood with low oxygen saturation. Three implications are obvious: EC proliferation and expansion of the capillary plexus are not driven by hypoxia (11), other metabolic influences (such as in muscle, brain, liver, etc.) play a much less important role, and the topography of the ramified convective system is not limited by the configuration of a parenchyma. The CAM has also been successfully used for studying hemodynamics and permeability of microvessels, for analysis of vascular cells and their interaction with those of transplanted tissues or tumors, and for assessing the effects of growth factors and of biomaterials.
|
Intussusception in the microvasculature
|
|---|
Nonsprouting angiogenesis by means of intussusception, i.e., "growth within itself," is an important mode of capillary formation (in addition to sprouting angiogenesis) and was termed intussusceptive microvascular growth (IMG). It was first observed in the rapidly expanding postnatal lung capillary bed and later in that of the CAM by Burri and coworkers (12). The presence of numerous tiny holes in vascular corrosion casts (studied with scanning electron microscopy) were shown, upon serial sectioning of tissue and subsequent transmission electron microscopy, to correspond to slender transcapillary (intraluminal) tissue pillars or posts (6). The same four consecutive steps in pillar formation (already described in the lung by Burri and coworkers in 1990) were observed in all investigated organs (Fig. 1
): 1) protrusion of opposing capillary walls into the lumen and the creation of a contact zone between the ECs; 2) reorganization of their intercellular junctions and central perforation of the endothelial bilayer; 3) formation of an interstitial pillar core by invading supporting cells (myofibroblasts, pericytes) and deposition of matrix, such pillars ranging in diameter from 1 to 2.5 µm; and 4) enlargement in girth of the pillars without additional qualitative alteration.
View larger version (36K):
[in this window]
[in a new window]
|
FIGURE 1. Basic steps of intussusception: protrusion of opposing capillary walls into the lumen (A and B) and creation of a contact zone between the endothelial cells (C). After central perforation of the cellular bilayer, the fused endothelial cells form a transluminal cuff, invaded later by myofibroblasts or pericytes (D).
|
|
Essentially the same sequence of pillar formation was observed during the initial remodeling of the capillary plexus into immediate pre- and postcapillary feeding vessels. This process was termed intussusceptive arborization (IAR) (4, 6) and provides a mechanism by which well-perfused capillary segments are transformed into terminal arterioles and veins (Fig. 2). The IAR is initiated by "vertical" pillar formation in rows, which demarcate future vessels. Narrow tissue septa formed by pillar reshaping and pillar fusions segregate the new vascular entity, and "horizontal" folds separate them from the capillary plexus. As a result of this process, a complex arterial and venous vascular tree arises from the primitive capillary plexus, forming a second layer.
View larger version (68K):
[in this window]
[in a new window]
|
FIGURE 2. Interplay of all 3 modes of intussusception: intussusceptive microvascular growth (IMG), intussusceptive arborization (IAR), and intussusceptive branching remodeling (IBR). Insertion of transluminal pillars (arrowheads) results in rapid expansion of the capillary plexus, i.e., IMG (A). IAR generates feed vessels from the capillary plexus by "vertical" pillar formation in rows (arrows in A), which demarcate future vessels. Narrow tissue septa (arrows in B and C) formed by pillar reshaping and pillar fusions segregate the new vascular entity. Formation of "horizontal" pillars and folds (arrowheads in D) separate the newborn feeding vessels from the capillary plexus. IBR finally adapts the deepness of branching angle and diameters of daughter vessels in the newly formed supplying and draining vessels by insertion of transluminary pillars at branching points (arrows in E, F, and G). Moreover, IBR leads to vascular pruning by repetitively eccentric formation, augmentation, and fusion of pillars (arrowheads in F, G, and H). Adapted from Ref. 6, with permission.
|
|
Although the morphological hallmarks of IMG and IAR were frequently observed in capillaries and terminal microvessels with diameters <25 µm, the formation of transluminal pillars and folds in arteries and veins with diameters of up to 110 µm was a surprise. Such pillars were closely associated with bifurcations and were shown to be involved in remodeling and pruning of larger microvessels (5). This process hence was termed intussusceptive branching remodeling (IBR) [Figs. 2
and 3; see also the supplemental film online (http://nips.physiology.org/cgi/content/full/18/2/65/DC1)]. Although the same four steps were observed in all three modes of intussusception, there is an important difference in the last mode. In contrast to IMG and IAR, whereby new vessel segments are generated, IBR does not produce new vessels. Rather, it may lead to 1) simple progression (upstream in arteries, downstream in veins) of a bifurcation point while maintaining its geometric proportions but diminishing the bifurcation angle, 2) remodeling of one or both branch diameters in relation to the stem and thus structural adaptation, or 3) drastic narrowing and subsequent regression of a branch, a process termed vascular pruning.
View larger version (141K):
[in this window]
[in a new window]
|
FIGURE 3. Vascular casts of the chick chorioallantoic membrane (CAM) prepared between days 8 and 10 of embryonic development, demonstrating the process of IBR. A: overview of the supplying artery and capillary plexus. Four of the five visible bifurcations embrace a transluminal pillar (arrowheads show holes in cast), each of which is at a different stage of tissue augmentation. B–D: during the initial stage of IBR (B), a pillar (hole) appears near the bifurcation angle, from which it is separated by a luminal gap (arrows). Subsequent tissue augmentation of the pillar causes the gap to narrow (C) and finally to retract (D). As a result of this process, the pillar merges with the connective tissue (*) in the bifurcation angle. Typical sites are also indicated, where diameter measurements were performed (arrows in D). Scale bars: A = 100 µm; B–D = 20 µm. Reprinted from Ref. 5, with permission.
|
|
Different aspects of angiogenesis and remodeling in perfused microvascular systems are thus accomplished via these three modes of intussusception. IMG can rapidly expand a capillary plexus and produce a large endothelial surface without altering capillary dimensions or identity. IAR can generate feed vessels from the capillary plexus by changing their dimensions, position, and identity and thus influence "endpoint density" of the supplying network. Finally, IBR can modulate the position of and diameters at branching points in larger microvessels and thus can influence pre- and postcapillary transport properties. Moreover, IBR can lead to removal of branches, and this process of vascular pruning is of considerable importance to achieve transport efficiency. Notably, none of the three modes requires immediate EC proliferation; rather, they preferentially occur by shape changes (thinning, spreading, etc.) of ECs when proliferation is decreasing or even absent. But how are these three modes of intussusception coordinated?
|
Hemodynamics and intussusception
|
|---|
One biomechanical prerequisite for the formation of any transluminal structure is the absence of too high distending or shear forces from the site of pillar or fold formation. in the capillary bed, this condition may be more easily fulfilled than in larger microvessels, and hence IMG and IAR are more frequently observed than IBR. For those who doubt that transluminal structures can be formed at all in the vasculature, recall that the millimeter-sized embryonic heart, while pulsating, is transformed from a simple tube into a four-chambered pump via multiple transluminal septation. Recent gene expression studies (8) strongly support that ECs may receive instructive signals from the hemodynamic alterations they are exposed to, and thus regions of locally altered shear or distending stress may allow or even trigger the growth of septa and ridges, pillars and folds. In the CAM, heterogeneity of EC proliferation had suggested (11) and in vivo videomicroscopy confirmed (4) that IMG and IAR predominantly occur in regions with accelerated blood flow. But since pressure and flow are tightly coupled, it is difficult to tell which information (shear or distending stress) was more important for vascular cells to perform intussusception. It is probably a combination of both that trigger IMG, IAR, or IBR. Such regions of locally altered hemodynamics may be distended transcapillary pillars (IMG), high-flow capillary segments (IAR), or the combination of low shear with high wall stresses at stream dividers (IBR). Direct experimental evidence for these proposals has not yet been provided. However, after clamping side branches in the CAM, flow velocity increased by at least 50% in downstream arteries, and at least twice as many bifurcations underwent IBR shortly after clamping (5). Remodeling began within 30 min after onset of enhanced pressure and flow and, by insertion of multiple pillars at stream dividers, produced more acute branching angles. The time needed to complete pillar formation ranged from 40 to 120 min. This rapid response indicates that the machinery needed for transluminal pillar formation is present and needs only to be switched on in endothelial and supporting cells, because the time course of altered gene expression after biomechanical activation of ECs is in the range of at least 120 min (8). But how are structural changes (via bifurcation remodeling) linked to function?
|
Bifurcation geometry: optimum form—and function?
|
|---|
The relationship between the diameters of the stem (d0) and branch (d1, d2) vessels is such that (on average) the combined cross-sectional area of the branches is greater than that of the stem. This relationship has been denoted by the area ratio ß = (d12 + d22)/d02. Since ß > 2 for most parts of the system, the velocity of blood flow is slowed down toward the periphery. The distribution of blood flow at each bifurcation also depends on the asymmetry ratio between the two branches [
= d1/d2 (with d1 
d2)]. The process of vascular pruning, i.e., nonperfusion and removal of branches at bifurcations, may thus be signified by << 1. The most-often-used parameter to characterize tree-like ramified networks, however, is the so-called bifurcation exponent 
in the equation d0 = d1 + d2, which combines both area and asymmetry ratio into a single number. Detailed considerations of these parameters have been published (1, 10). Here it will suffice to mention that most directly links form and function, e.g., the famous "Murray’s Law" aiming at minimum power consumption and constant shear stress throughout the system, which postulates = 3.0. However, the alternative "Kurz-Sandau Law" ( = 2.7) corresponds to minimum wall material and constant tensile wall stress, whereas the "West-Brown-Enquist Law" ( = 2.0–3.0, depending on bifurcation level) makes interesting predictions about the scaling properties of vascular systems. So we can rephrase our question concerning the link between form and function: which values can actually be observed during bifurcation remodeling?
We calculated geometric characteristics , ß, and from diameter measurements at arterial and venous sites of IBR (Fig. 3D and Table 1). Unexpectedly, values for arteries ( = 2.73–2.90) and veins ( = 2.93–3.75) were significantly different (5). Moreover, the values derived from diameters directly at the site of bifurcation (', ß', and ') were significantly different from those derived from measurements in the branches (, ß, and ) in that the harmonic means of all three parameters increased from the site of IBR to the downstream (arteries) or upstream (veins) vessel segment. Obviously, IBR may be accompanied by short-range deviations from a global optimum of vascular design, and it is often associated with more asymmetric bifurcations. Yet near-optimum bifurcation geometries, as predicted by theory, emerge via IBR but from "opposite sides" in arteries and veins. Although the minimum tissue optimum ( = 2.7) is closely met in small arteries, small veins ( = 3.75) deviate considerably from any optimum condition, but larger (developmentally older) bifurcations of either type are remodeled toward the minimum energy optimum ( 
3.0). Values of < 3.0 in arterial and > 3.0 in venous branches both reflect the same situation: shear stresses are lower upstream (stem in arteries, branches in veins) of IBR than downstream (stem in veins, branches in arteries). Thus IBR in venous and arterial bifurcations is compatible with similar fluid shear- and wall-stress-sensing mechanisms by endothelial and supporting cells, as illustrated in Fig. 4. Whereas it remains to be shown whether comparable mechanisms regulate capillary density during IMG or feeding vessel density during IAR, the concept of an instructive role of hemodynamics for blood vessel patterning has been implemented in computer simulations.
View larger version (25K):
[in this window]
[in a new window]
|
FIGURE 4. Possible mechanisms of IBR. A: at the bifurcation site, local diameters like d' are even larger than d0 in the stem vessel, resulting in locally reduced shear stress [ , transduced by endothelial cells (ECs)] and enhanced circumferential wall tension [T, transduced by pericytes (PCs) or vascular smooth muscle cells (VSMCs)]. Additionally, flow disturbance (shaded area) near the stream divider may promote pillar formation. B: ECs are thus instructed to form initial contacts in zone of altered flow. Pillar formation leads to d1' and d2', thus reducing T and enhancing locally, which may induce signaling between ECs and PCs/VSMCs. Additional signaling may also lead to dilation in branches (indicated by slightly larger diameters than d1 and d2). C: fusion of bifurcation site with the growing pillar leads to more acute flow divider where both T and may be compared with local "set points" generated by intrinsic EC and PC/VSMC machinery. Remodeling may either continue with further pillar formation and fusion (repeat A and B) or replace IBR by nonintussusceptive remodeling mechanisms. Finally, the combination of pressure- and flow-related signals stop remodeling as shown here, where a slightly larger d0 and optimized bifurcation geometry have been stabilized by additional layers of mural cells. Reprinted from Ref. 5, with permission.
|
|
|
Simulation of vascular growth and remodeling
|
|---|
In a recent model, local growth and transport processes lead to global functional and structural remodeling (Ref. 9; Fig. 5). This was the first angiogenesis model in which hemodynamic and transport characteristics were known at any individual vascular branch and which produced interdigitating arteriovenous patterns that closely resembled those of CAM microvessels. Most importantly, realistic morphogenesis was obtained only if regression of branches at bifurcations was included and if remodeling was regulated by shear stress, not by pressure. Clearly, congruence of real measurements with structural and functional output of a model cannot positively prove that indeed the same mechanisms were in effect in the model as in the biological system. But the observed realistic patterns and hemodynamic predictions strongly support the view that vascular pruning is a prerequisite for vascular pattern formation and that shear stress is a dominant player in vascular remodeling even during development. Moreover, the model shows how a combination of deterministic and random mechanisms that act locally can lead to self-organized, stable complexity on a global scale and thus can account for the coexistence of a nearly constant stem vessel pattern with individual variability of the microvasculature. Interestingly, a different approach combining in vivo data with computer simulations (14) emphasizes the necessity for combined action of hemodynamic with metabolic signals for stable microvascular remodeling in the adult and also questions the general validity of Murray’s Law.
View larger version (121K):
[in this window]
[in a new window]
|
FIGURE 5. Computer simulation of arterial and venous growth on a spheroidal surface. Note interdigitating pattern of branching arteries (red) and veins (blue). Such patterns that mimicked CAM bifurcations were obtained only if shear stress-regulated vessel growth and regression were included in the remodeling algorithm. Reprinted from Ref. 9, with permission.
|
|
|
Cells and molecules
|
|---|
The way we look at the circulatory system has considerably changed over the past two decades. Most importantly, is has become clear that multiple cellular mechanisms and an ever-increasing combination of genes and molecules are involved in building and adapting blood vessels throughout a lifetime (15). Although ECs had been identified early as key players in blood vessel formation and remodeling, interactions of ECs with macrophages, pericytes, and smooth muscle cells came into focus much later. Today, the importance of pericytes and smooth muscle cells for capillary stabilization has been recognized, and interactions of these supporting cells, together with macrophages, during vascular remodeling are intensely studied.
It is known that mechanical stress can induce biochemical cascades, beginning with the activation of ion channels within seconds and resulting in rearrangements of cytoskeletal components and adaptations in gap junction complexes within minutes or hours. Additional stimuli may originate from interactions between blood cells and the endothelium. The various signaling pathways involved in these processes are not fully understood, but in an analogy to the concept of arteriogenesis (2), intussusception may not only involve endothelial and contractile periendothelial cells but also macrophages, whose actions may be coordinated by shear stress. Recruitment of such periendothelial cells and their prolonged association with the larger pillars probably stabilizes pillar structure. Angiopoetins (Ang) and their Tie receptors, as well as ephrins and Eph-B receptors or macrophage chemotactic protein-1, are likely candidates for mediating such cell-cell interactions.
On the one hand, the vasculature of mice lacking Ang-1 and Tie-2 remains at a primitive stage of development and fails to undergo further remodeling. On the other hand, overexpression of Ang-1 or of Ang-1 in combination with vascular endothelial growth factor (VEGF) is characterized not only by the formation of large vessels but by the presence of abundant small holes at their bifurcations, a finding that is symptomatic of intussusception. Signaling through the platelet-derived growth factor (PDGF)-B and TGF-ß systems are further promising candidates for mediating between shear stress and intussusception. Notably, application of PDGF-B on the differentiated CAM led to an abundance of larger pre- and postcapillary microvessels but did not expand the capillary plexus, whereas exogenous VEGF-A affected the capillary plexus and terminal feeding vessels but not larger microvessels. Also, Notch4 signaling was shown to be involved in remodeling of the vascular system, and reported vascular alterations are strongly reminiscent of disturbed intussusceptive remodeling in the perineural plexus.
|
Conclusions and open questions
|
|---|
We feel that the importance of intussusception during angiogenesis and vascular remodeling may have been underestimated in the past. Most studies focus on sprouting angiogenesis, which of course is important in many regions, notably the central nervous system, where signs of IMG and IBR were rarely found (14). Yet vascular cells can apparently switch their behavior in various ways. On the one hand, high VEGF levels can induce ECs that undergo IMG to either sprout or fuse (7) or maybe even both. On the other hand, recent morphological work strongly supports a role for intussusception even during tumor angiogenesis and tissue repair (3). We hope that in the near future, screening for genes whose expression is changed by biomechanical stimuli will be combined with more detailed morphological studies and in situ registration of hemodynamic parameters in and around vascular bifurcations. Once we have understood how local mechanisms contribute to global self-regulation during vascular morphogenesis, we can hope to address more specifically local variations, like shunt and collateral formation or adaptation to temporally changing requirements, and thus the mutual influences of form and function.
|
Acknowledgments
|
|---|
We thank Dr. Konrad Sandau for furnishing us with a software tool. We are also grateful to Krystyna Sala, Barbara Krieger, Karl Babl, and Elisabeth de Peyer for their technical assistance.
This research was financially supported by the Swiss National Science Foundation (Grant no. 31-55895.98) and by the Bernese Cancer League.
|
References
|
|---|
- Bennett SH, Eldridge MW, Zaghi D, Zaghi SE, Milstein JM, and Goetzman BW. Form and function of fetal and neonatal pulmonary arterial bifurcations. Am J Physiol Heart Circ Physiol 279: H3047–H3057, 2000.
- Buschmann I and Schaper W. Arteriogenesis versus angiogenesis: two mechanisms of vessel growth. News Physiol Sci 14: 121–125, 1999.
- Djonov VG, Andres AC, and Ziemiecki A. Vascular remodeling during the normal and malignant life cycle of the mammary gland. Microsc Res Tech 52: 182–189, 2001.
- Djonov VG, Galli AB, and Burri PH. Intussusceptive arborization contributes to vascular tree formation in the chick chorio-allantoic membrane. Anat Embryol (Berl) 202: 347–357, 2000.
- Djonov VG, Kurz H, and Burri PH. Optimality in the developing vascular system: bifurcation remodeling by means of intussusception as an efficient adaptation mechanism. Dev Dyn 224: 391–402, 2002.
- Djonov V, Schmid M, Tschanz SA, and Burri PH. Intussusceptive angiogenesis: its role in embryonic vascular network formation. Circ Res 86: 286–292, 2000.
- Drake CJ and Little CD. VEGF and vascular fusion: implications for normal and pathological vessels. J Histochem Cytochem 74: 1351–1355, 1999.
- Garcia-Cardena G, Comander J, Anderson KR, Blackman BR, and Gimbrone MA. Biomechanical activation of vascular endothelium as a determinant of its functional phenotype. Proc Natl Acad Sci USA 98: 4478–4485, 2001.
- Gödde R and Kurz H. Structural and biophysical simulation of angiogenesis and vascular remodeling. Dev Dyn 220: 387–401, 2001.
- Kurz H. Physiology of angiogenesis. J Neurooncol 50: 17–35, 2000.
- Kurz H, Ambrosy S, Wilting J, Marmé D, and Christ B. Proliferation pattern of capillary endothelial cells in chorioallantoic membrane development indicates local growth control, which is counteracted by vascular endothelial growth factor application. Dev Dyn 203: 174–186, 1995.
- Patan S, Haenni B, and Burri PH. Evidence for intussusceptive capillary growth in the chicken chorio-allantoic membrane (CAM). Anat Embryol (Berl) 187:121–130, 1993.
- Patan S, Munn LL, Tanda S, Roberge S, Jain RK, and Jones RC. Vascular morphogenesis and remodeling in a model of tissue repair. Circ Res 89: 723–731, 2001.
- Pries AR, Reglin B, and Secomb TW. Structural adaptation of microvascular networks: functional roles of adaptive responses. Am J Physiol Heart Circ Physiol 281: H1015–H1025, 2001.
- Tomanek RJ, editor. Assembly of the Vasculature and Its Regulation. Berlin: Birkhäuser, 2002.
This article has been cited by other articles:
|
|
|
|
 
D. McLeod
A chronic grey matter penumbra, lateral microvascular intussusception and venous peduncular avulsion underlie diabetic vitreous haemorrhage
Br J Ophthalmol,
May 1, 2007;
91(5):
677 - 689.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. P. Borowicz, D. R. Arnold, M. L. Johnson, A. T. Grazul-Bilska, D. A. Redmer, and L. P. Reynolds
Placental Growth Throughout the Last Two Thirds of Pregnancy in Sheep: Vascular Development and Angiogenic Factor Expression
Biol Reprod,
February 1, 2007;
76(2):
259 - 267.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. Dome, M. J.C. Hendrix, S. Paku, J. Tovari, and J. Timar
Alternative Vascularization Mechanisms in Cancer: Pathology and Therapeutic Implications
Am. J. Pathol.,
January 1, 2007;
170(1):
1 - 15.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. C. W. Hsia
Quantitative morphology of compensatory lung growth
Eur. Respir. Rev.,
December 1, 2006;
15(101):
148 - 156.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. M. Prior, H. T. Yang, and R. L. Terjung
What makes vessels grow with exercise training?
J Appl Physiol,
September 1, 2004;
97(3):
1119 - 1128.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Mechanisms and Limits of Induced Postnatal Lung Growth
Am. J. Respir. Crit. Care Med.,
August 1, 2004;
170(3):
319 - 343.
[Full Text]
[PDF]
|
|
|
|
|
|
|
X. Yan, D. J. Bellotto, D. J. Foster, R. L. Johnson Jr., H. K. Hagler, A. S. Estrera, and C. C. W. Hsia
Retinoic acid induces nonuniform alveolar septal growth after right pneumonectomy
J Appl Physiol,
March 1, 2004;
96(3):
1080 - 1089.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Introduction
A model system for...
