- Email: [email protected]
Morphometric analysis of the branching of the vascular tree in the chick embryo area vasculosa
Microvascular Research 128 (2020) 103935
Contents lists available at ScienceDirect
Microvascular Research journal homepage: www.elsevier.com/locate/ymvre
Morphometric analysis of the branching of the vascular tree in the chick embryo area vasculosa
T
Diego Guidolina, Roberto Tammab, Cinzia Tortorellaa, Tiziana Anneseb, Simona Ruggierib, ⁎ Andrea Marzulloc, Domenico Ribattib, a
Department of Neuroscience, Section of Anatomy, University of Padova, Padova, Italy Department of Basic Medical Sciences, Neurosciences and Sensory Organs, University of Bari Medical School, Bari, Italy c Department of Emergency and Organ Transplantation, Pathology Section, University of Bari Medical School, Bari, Italy b
A R T I C LE I N FO
A B S T R A C T
Keywords: Angiogenesis Area vasculosa Branching Chick embryo Vascular tree Vasculogenesis
The chick embryo includes the area vasculosa is subdivided into 2 concentric zones, the inner transparent area pellucida vasculosa and the surrounding less transparent area opaca vasculosa, peripherally limited by the sinus terminalis. In this study, we have analyzed by a modern morphometric approach the total length of the vascular network, the number of vascular branches, of the branching points density, the modality of vessel ramification, and spatial arrangement of the vascular network in four consecutive stages of development of the area vasculosa. The results have shown that there is a significant 15% increase in the total length of the vascular network associated with a progressive increase of the number of vascular branches and of the branching points density. Moreover, the results indicated that vascular spatial disorder significantly decreased during development in area vasculosa, suggesting a more uniform occupancy of the tissue by the vascular pattern. Finally, a more regular pattern of branching was observed, as indicated by the significant decrease of topological disorder of the vascular tree.
Introduction The vascularized membranes of the chick embryo include the area vasculosa and the chorioallantoic membrane (CAM). The former is subdivided into 2 concentric zones, the inner transparent area pellucida vasculosa and the surrounding less transparent area opaca vasculosa, that is peripherally limited by the sinus terminalis (Augustine 1970; Mayer and Packard 1978). The vessels of the area vasculosa are involved in the absorption and mobilization of nutrients from the yolk. Histologically, the area vasculosa is formed by the ectoderm, the intermediate mesoderm, and the endoderm. The mesoderm splits into somatopleura and splanchnopleura; the somatopleura lies tightly under the ectoderm, while the splanchnopleura lying on the endoderm, contains blood vessels. Most of the difficulty related to the study of the dynamics of vascular growth in the chick area vasculosa has resulted from the absence of readily quantifiable test systems. The expansion of the area vasculosa is generally constant in all developmental phases examined, but may be influenced by yolk hydration from the albumen (Romanoff 1960). As suggested by Stewart et al. (1990), the area vasculosa is a relatively inelastic structure so that
the increase in the volume of yolk may be accommodated by the growth of the more lateral area vitellina alone. These data may be used for a more analytical interpretation of the ways by which the growth of the vascular system in the area vasculosa occurs. Two main mechanisms seem to be implicated: the release of collaterals from vessels measuring 1.01–2 mm and the lengthening of 2.01–7 mm vessels. These mechanisms may be correlated with models for the generation of sequence of structures in the space during morphogenesis, as proposed by Meinhardt and Gierer 1980). In this way, the growth of blood vessels in the area vasculosa appears to be similar to the growth of other tubular structures during embryogenesis, such as the salivary glands and the lung (Bernfield et al. 1984), occurring through the release of collaterals and the lengthening of the primary tubular formations. In this study, we have analyzed by a modern morphometric approach aimed at characterizing, in four consecutive stages of development of the chick embryo area vasculosa, not only the overall structure of the vascular network (total length, number of vascular branches and branching points density), but also the modality of vessel ramification, and the spatial arrangement of the growing vascular tree.
⁎ Corresponding author at: Department of Basic Medical Sciences, Neurosciences and Sensory Organs, University of Bari Medical School, Policlinico - Piazza G. Cesare, 11, 70124 Bari, Italy. E-mail address: [email protected] (D. Ribatti).
https://doi.org/10.1016/j.mvr.2019.103935 Received 10 June 2019; Received in revised form 22 August 2019; Accepted 11 October 2019 0026-2862/ © 2019 Published by Elsevier Inc.
Microvascular Research 128 (2020) 103935
D. Guidolin, et al.
Fig. 1. Main steps of the image analysis procedure. A. Image of the area vasculosa at stage 1. B. Binary image of the vascular profiles. C. Branching points and branches connecting them. In red are shown examples of the circular regions centered on branching points and used to estimate the branching order associated to each branching point. D. Sholl analysis, involving the evaluation of the number of intersections of the vascular pattern with concentric circles of increasing radii around the central root point of the vascular tree. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Materials and methods
exclusive of all background an adaptive discrimination procedure was applied 5 (Russ 1995). This method operated with a local threshold: the mean grey value of a neighboring region was calculated for every pixel; this value plus an offset threshold constant defined the local threshold for that pixel. A 3 × 3 median filter was then applied to clean the resulting binary image (Fig. 1 B) and the area fraction covered by vessels was measured. By using binary thinning procedures, the skeleton of the binary image was finally derived and pruned to remove eventually present small artefactual branches (Fig. 1 C). From this image the total length of the vascular pattern was calculated, and branching points were identified as previously described (Guidolin et al. 2004b). The number of branching points and the number of branches forming the vascular tree were then evaluated. In addition, from the set of positions formed by the branching points parameters characterizing the spatial arrangement of the vascular tree (spatial disorder, SD) and the regularity of branching along the structure (topological disorder, TD) were estimated as previously described (Guidolin et al. 2004a). Both these parameters range from 0 to 1, lower values indicating a more uniform spatial distribution or branching occurrence respectively. Branching mode. The branching pattern is determined by the division
Forty fertilized White Leghorn chick eggs, staged according to Hamburger & Hamilton (HH) (Hamburger and Hamilton 1951) stages, were incubated from the start of their embryogenesis in an incubator under conditions of constant humidity at a temperature of 37 °C. At HH stage 13, a square window was opened in the egg shell after removal of 2–3 ml of albumen so as to detach the developing chorioallantoic membrane from the shell. The opening was closed with cellophane tape and the incubation continued. The area vitelline was photographed in ovo at 4 different stages of development accordingly to HH stages: HH 13 (stage 1), HH 20 (stage 2); HH 24 (stage 3); HH 27 (stage 4). Morphometry. Grey level images of the different stages of development of the area vasculosa were saved as TIFF files, processed and analyzed by using the ImageJ software (Schneider et al. 2012), freely available at http://rsb. info.nih.gov/ij/, and macro routines specifically developed by the authors. The main steps of the image analysis procedure are illustrated in Fig. 1. Global morphometric features. To select vessel profiles virtually 2
Microvascular Research 128 (2020) 103935
D. Guidolin, et al.
of a main trunk into smaller branches. Bifurcation and trifurcation are the most commonly described modes, although other types have been observed (Bradac 2011; Kopylova et al. 2017). As illustrated in Fig. 1 C, to characterize this feature a circular region (14 pixels diameter) centered on each branching point was explored to count the number n of branches in it. The number n-1 was then considered as the index of the branching mode occurring at the considered branching point. Sholl analysis. Sholl analysis (Sholl and Uttley 1953) was performed to quantify the variations in vascular distribution during vascular tree development. Previously used to quantify characteristics of dendritic processes branching off neuronal cell bodies, Sholl analysis examines the number of branches that intersect concentric circles of increasing radii around the central root point of the vascular tree (Fig. 1 D). To perform the analysis on the binary images of the vascular patterns a specific plugin for Image J was used (Ferreira et al. 2014) providing the number of intersections as a function of radius. From this distribution a metric named branching index (BI), that compares the difference in the number of intersections made in consecutive pairs of circles relative to the distance from the central focal point (Garcia-Segura and PerezMarquez 2014) for a definition and a discussion), was also derived. It gives a synthetic quantitative measure of vessel ramification.
Fig. 2. Changes in the number of vascular branches and in the branching points density (expressed as number of branching points per unit length of the vascular tree, BPL) during the observed period of development. * = p < .05 vs. the stage 1.
in the overall complexity of the pattern were also detected during its development. As illustrated in Fig. 2, a progressive increase of the number of vascular branches and of the branching points density was observed in the course of the tested period. These parameters exhibited a significant 51% and 14% increase respectively at stage 4 as compared to stage 1. Furthermore, the spatial arrangement of the vascular network significantly changed. Spatial disorder (SD) significantly decreased from stage 1 to stage 4, indicating a more uniform spatial occupancy of the vascular pattern, and a more regular pattern of branching was observed, as indicated by the significant decrease of topological disorder (TD). The results of the Sholl analysis were consistent with these findings but provided some additional information. BI values confirmed the above-mentioned increase in the number of branches with time. Moreover, as shown in Fig. 3 A, the data suggested that vessels initially ramified mainly at the periphery and subsequently in the regions around the central root point. As far as the mode of branching is concerned, vessel ramification was found to occur almost exclusively by bifurcation. In all samples, from > 80% of the branching points two branches emerged, while trifurcation or higher branching modes were associated to < 20% of the branching points (Fig. 3 B).
Statistics Differences between the four groups of samples were statistically tested by one-way analysis of variance followed by Bonferroni's test for multiple comparisons. The GraphPad Prism 3.0 statistical package (GraphPad Software Inc., San Diego CA, USA) was used for the analysis, and p < .05 was considered as the limit for statistical significance. Results At the HH stages 13 and 20 the area vasculosa surrounding the embryo was approximately round in shape. The trunk of the right vitelline artery ran in a nearly straight line for a short distance towards the lateral border of the area vasculosa. It then made a concave loop medially to run in the anterior, lateral and posterior parts of the area vasculosa where it terminated as a capillary network which continued peripherally with the sinus terminalis. From this latter structure, the blood was taken to the heart through the anterior and posterior vitelline veins. Blood was pumped away from the embryo through the omphalomesenteric arteries towards the sinus terminalis and returned through the anterior and posterior vitelline veins. At the same time, the intraembryonic vasculature had been established with dorsal aorta, branchial arch and cephalic arteries. At HH stages 24 and 27 the area vasculosa continued to expand and became denser; the formation of the posterior vein was completed and the lateral vitelline vein began to form. Morphometric data are summarized in Table 1. As shown, a significant 15% increase in the total length of the vascular network was observed at stage 4 as compared to the more primitive stage 1. Changes
Discussion At HH 18–20 stage the primary vascular network in the embryo and yolk sac is already formed through vasculogenesis and further blood vessel development proceeds through sprouting angiogenesis (Poole and Coffin 1989). Vascular endothelial growth factor (VEGF) and its receptors 1 and 2 (VEGFR-1 and VEGFR-2) are implicated in vasculogenesis and angiogenesis (Risau, 1997). We have previously demonstrated that VEGF is expressed in the endodermal cells of the chick embryo area vasculosa immediately adjacent to the mesodermal endothelial cells which, in turn, expressed VEGFR-2 (Nico et al. 2001). Estimation of a network complexity by quantification of the number
Table 1 Morphometric analysis of the area vasculosa (mean ± sem, N = 40). Stage
Area (mm2)
Length (mm)
BPL (n/mm)
Branches (n)
SD
TD
Branching order
BI
1
65.7 ± 2
330.9 ± 10
0.29 ± 0.009
170 ± 5
0.59 ± 0.02
0.69 ± 0.02
2.06 ± 0.06
28.1 ± 0.9
2 3 4
66.7 ± 2 67.6 ± 2 71.7 ± 2
373.0 ± 12 376.9 ± 12 378.0 ± 12
0.32 ± 0.010 0.33 ± 0.010 0.34 ± 0.010
219 ± 7 214 ± 7 257 ± 8
0.59 ± 0.02 0.56 ± 0.02 0.49 ± 0.01
0.67 ± 0.02 0.65 ± 0.02 0.060 ± 0.02
2.09 ± 0.07 2.01 ± 0.06 2.18 ± 0.07
33.6 ± 1 34.1 ± 1 35.2 ± 1
* * *
*
* * *
BP, branching point. SD, spatial disorder. TD, topological disorder. BI, branching index. ⁎ p < .05 vs. stage 1. 3
*
* *
* * *
Microvascular Research 128 (2020) 103935
D. Guidolin, et al.
Fig. 3. A. Values of the number of intersections of vessels with the consecutive concentric circles of the Sholl analysis at the different time steps considered. B. Typical distribution of the branching orders observed at the level of the branching points. As shown, the vast majority of branching points correspond to bifurcations.
adult (Maione and Giraudo 2015) tissues. A structural feature of particular interest in order to get a better understanding of tissue perfusion is the way the pattern of growing vessels fills the available space. In this respect, the results of the present study indicated that vascular spatial disorder significantly decreased during development in area vasculosa, suggesting a progressively more uniform occupancy of the tissue by the vascular pattern. Furthermore, a more regular pattern of branching was observed, as indicated by the significant decrease of topological disorder of the vascular tree. Consistent with this finding are the data provided by the Sholl analysis, suggesting that branching starts at the periphery of the growing vascular network and is subsequently extended to the regions near the root point. As mentioned before, an additional interesting aspect of vascular plasticity concerns the way the growing vascular network branches, since different patterns can be generated at each branching point. Knowledge of these variations are of particular importance when adult vascular networks are concerned, being helpful to surgeons and clinicians. In this respect, studies on cerebral vascularization suggested that bifurcation and trifurcation branching types are the most frequently observed (Cilliers and Page 2016). The results on the area vasculosa here presented show consistency with these findings. The vascular tree mainly grows by subsequent bifurcations, with only a minority of trifurcations. Branching patterns of higher order, likely representing crossover points (Bhuiyan et al. 2012), were seldom observed. A morphometric approach quite similar to the one used in the present study was also applied to analyze the embryonic lung branching morphogenesis (Carraro et al. 2009, 2014) and some morphological similarity can be identified between the pattern of development of the two structures in terms of branching complexity (i.e. branching point density and order of branching). This is not completely surprising, since in most of the biological tubular systems the branching pattern is the result of two signaling pathways, one promoting and another restricting branching (Horowitz and Simons 2008) and their balance is critical for producing a mature system that is neither suffused by ectopic branches, nor too sparse. The nature of these signals, however, differs between systems, as indicated by the branching promoting role fulfilled by FGF in the airway development but by VEGF in the area vasculosa (Nico et al. 2001).
of branching points and related measures (such as number of branches and length) are well recognized morphometric parameters to quantitatively describe a vascular network and the changes it undergoes during sprouting angiogenesis (Guidolin et al. 2004a, 2004b; Zudaire et al. 2011; Belle et al. 2014). They are also the main output provided by presently available integrated software packages for the image analysis of vascular networks, such as ‘AutoTube’ (Montoya-Zegarra et al., 2018) and ‘AngioQuant’ (Niemistö et al. 2005). In the present paper, starting from these primary parameters (basically characterizing the overall structure of the network), the morphometric analysis has been further expanded to characterize two additional morphological features of a vascular tree. The first concerns the way the pattern of growing vessels fills the available space. This feature is of interest to get a better understanding of tissue perfusion and has been characterized by parameters capturing the level of spatial order/disorder exhibited by the vascular network. These parameters increase as far as the spatial distribution of the branching points in the tissue (spatial disorder) or the occurrence of branching points along the tree structure (topological disorder) become less uniform. A second question of interest is how the network grows during the angiogenic process. In this respect, two aspects deserve consideration from the morphological standpoint, namely how a main trunk divides into smaller branches, and how the branching occurs at different distances from the root point of the vascular tree. These issues were here addressed by automatic image analysis procedures estimating the branching order at each branching point and exploiting the so-called Sholl analysis (i.e. the enumeration of vessels at varying distances from anchor points) respectively. Thus, in view of the increasing importance (as pointed out by the available literature) of angiogenesis and vasculogenesis under normal and pathological conditions the possibility to expand the information pattern obtainable following a morphometric approach could be of particular interest to the field. Some of these morphometric strategies were previously explored to characterize the regional and developmental variations of vascular network in the chick embryo CAM (Guidolin et al. 2004b; Reizis, 2005; Belle et al. 2014) or have been applied to CT images of the coronary arterial tree as a time-efficient approach to obtain data useful to quantitatively assess hypotheses on the coronary circulation (Wischgoll et al. 2009). Their combined use, however, could provide a more complete and quantitative picture of the morphological features of a vascular network. The present study was focused on the chick embryo area vasculosa and reports a first, at least in our knowledge, quantitative analysis of its morphological features during development. We have demonstrated that in the course of the development of the chick embryo area vasculosa there is a significant 15% increase in the total length of the vascular network associated with a progressive increase of the number of vascular branches and of the branching points density. From the functional standpoint, the increase of these parameters characterizing the vascular network has been correlated with an optimization of gas exchange and metabolic support in both embryonic (Reizis, 2005) and
Declaration of interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. References Augustine, J.M., 1970. Expansion of the area vasculosa of the chick after removal of the ectoderm. J Embryol Exp Morphol 24, 95–108. Belle, J., et al., 2014. Stretch-induced intussuceptive and sprouting angiogenesis in the chick chorioallantoic membrane. Microvasc. Res. 95, 60–67.
4
Microvascular Research 128 (2020) 103935
D. Guidolin, et al.
Mayer, B.W., Packard, D.S., 1978. A study of the expansion of the chick area vasculosa. Dev. Biol. 63, 335–351. Meinhardt, H., Gierer, A., 1980. Generation and regeneration of sequence of structures during morphogenesis. J. Theor. Biol. 85, 429–450. Montoya-Zegarra, J.A., et al., 2018. AutoTube: a novel software for the automated morphometric analysis of vascular networks in tissues. Angiogenesis 22, 223–236. Nico, B., et al., 2001. Vascular endothelial growth factor and vascular endothelial growth factor receptor-2 expression in the chick embryo area vasculosa. Histochem. J. 33, 283–286. Niemistö, A., Dunmire, V., Yli-Harja, O., Zhang, W., Shmulevich, I., 2005. Robust quantification of in vitro angiogenesis through image analysis. IEEE Trans. Med. Imaging 24, 549–553. Poole, T.J., Coffin, J.D., 1989. Vasculogenesis and angiogenesis: two distinct morphogenetic mechanisms establish embryonic vascular pattern. J. Exp. Zool. 251, 224–231. Reizis, A., 2005. Regional and developmental variations of blood vessel morphometry in the chick embryo chorioallantoic membrane. J. Exp. Biol. 208, 2483–2488. Risau, W., 1997. Mechanisms of angiogenesis. Nature 386, 671–674. Romanoff, A., 1960. The extraembryonic membranes. In: The Avian Embryo: Structural and Functional Development. MacMillan, New York, pp. 1039–1141. Russ, J., 1995. The Image Processing Handbook. CRC Press, Boca Raton, FL. Schneider, C.A., et al., 2012. NIH image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675. Sholl, D.A., Uttley, A.M., 1953. Pattern discrimination and the visual cortex. Nature 171, 387–388. Stewart, R., et al., 1990. Growth of the chick area vasculosa in ovo and in shell-less culture. J. Anat. 172, 81–87. Wischgoll, T., et al., 2009. Extraction of morphometry and branching angles of porcine coronary arterial tree from CT images. Am. J. Phys. Heart Circ. Phys. 297, H1949–H1955. Zudaire, E., et al., 2011. A computational tool for quantitative analysis of vascular networks. PLoS One 6, e27385.
Bernfield, M., et al., 1984. Remodelling of the basement membrane as a mechanism of morphogenetic tissue interaction. In: Trelstad, R.L. (Ed.), The Role of Extracellular Matrix in Development. Alan R. Liss, New York, pp. 542–572. Bhuiyan, A., et al., 2012. Detection and classification of bifurcation and branch points on retinal vascular network. In: 2012 International Conference on Digital Image Computing Techniques and Applications (DICTA). IEEE. Bradac, G.B., 2011. Vascular Territories. Cerebral Angiography. Springer Berlin Heidelberg, pp. 87–89. Carraro, G., et al., 2009. miR-17 family of microRNAs controls FGF10-mediated embryonic lung epithelial branching morphogenesis through MAPK14 and STAT3 regulation of E-cadherin distribution. Dev. Biol. 333, 238–250. Carraro, G., et al., 2014. miR-142-3p balances proliferation and differentiation of mesenchymal cells during lung development. Development 141, 1272–1281. Cilliers, K., Page, B.J., 2016. Review on the anatomy of the middle cerebral artery; cortical branches, branching pattern and anomalies. Turkish Neurosurgery 26, 653–661. Ferreira, T.A., et al., 2014. Neuronal morphometry directly from bitmap images. Nat. Methods 11, 982–984. Garcia-Segura, L.M., Perez-Marquez, J., 2014. A new mathematical function to evaluate neuronal morphology using the Sholl analysis. J. Neurosci. Methods 226, 103–109. Guidolin, D., et al., 2004a. Order and disorder in the vascular network. Leukemia 18, 1745–1750. Guidolin, D., et al., 2004b. A new image analysis method based on topological and fractal parameters to evaluate the angiostatic activity of docetaxel by using the Matrigel assay in vitro. Microvasc. Res. 67, 117–124. Hamburger, V., Hamilton, H.L., 1951. A series of normal stages in the development of the chick embryo. J. Morphol. 88, 49–92. Horowitz, A., Simons, M., 2008. Branching morphogenesis. Circ. Res. 103, 784–795. Kopylova, V.S., et al., 2017. Fundamental principles of vascular network topology. Biochem. Soc. Trans. 45, 839–844. Maione, F., Giraudo, E., 2015. Tumor angiogenesis: Methods to analyze tumor vasculature and vessel normalization in mouse models of cancer. In: Methods in Molecular Biology. Springer New York, pp. 349–365.
5
Contents lists available at ScienceDirect
Microvascular Research journal homepage: www.elsevier.com/locate/ymvre
Morphometric analysis of the branching of the vascular tree in the chick embryo area vasculosa
T
Diego Guidolina, Roberto Tammab, Cinzia Tortorellaa, Tiziana Anneseb, Simona Ruggierib, ⁎ Andrea Marzulloc, Domenico Ribattib, a
Department of Neuroscience, Section of Anatomy, University of Padova, Padova, Italy Department of Basic Medical Sciences, Neurosciences and Sensory Organs, University of Bari Medical School, Bari, Italy c Department of Emergency and Organ Transplantation, Pathology Section, University of Bari Medical School, Bari, Italy b
A R T I C LE I N FO
A B S T R A C T
Keywords: Angiogenesis Area vasculosa Branching Chick embryo Vascular tree Vasculogenesis
The chick embryo includes the area vasculosa is subdivided into 2 concentric zones, the inner transparent area pellucida vasculosa and the surrounding less transparent area opaca vasculosa, peripherally limited by the sinus terminalis. In this study, we have analyzed by a modern morphometric approach the total length of the vascular network, the number of vascular branches, of the branching points density, the modality of vessel ramification, and spatial arrangement of the vascular network in four consecutive stages of development of the area vasculosa. The results have shown that there is a significant 15% increase in the total length of the vascular network associated with a progressive increase of the number of vascular branches and of the branching points density. Moreover, the results indicated that vascular spatial disorder significantly decreased during development in area vasculosa, suggesting a more uniform occupancy of the tissue by the vascular pattern. Finally, a more regular pattern of branching was observed, as indicated by the significant decrease of topological disorder of the vascular tree.
Introduction The vascularized membranes of the chick embryo include the area vasculosa and the chorioallantoic membrane (CAM). The former is subdivided into 2 concentric zones, the inner transparent area pellucida vasculosa and the surrounding less transparent area opaca vasculosa, that is peripherally limited by the sinus terminalis (Augustine 1970; Mayer and Packard 1978). The vessels of the area vasculosa are involved in the absorption and mobilization of nutrients from the yolk. Histologically, the area vasculosa is formed by the ectoderm, the intermediate mesoderm, and the endoderm. The mesoderm splits into somatopleura and splanchnopleura; the somatopleura lies tightly under the ectoderm, while the splanchnopleura lying on the endoderm, contains blood vessels. Most of the difficulty related to the study of the dynamics of vascular growth in the chick area vasculosa has resulted from the absence of readily quantifiable test systems. The expansion of the area vasculosa is generally constant in all developmental phases examined, but may be influenced by yolk hydration from the albumen (Romanoff 1960). As suggested by Stewart et al. (1990), the area vasculosa is a relatively inelastic structure so that
the increase in the volume of yolk may be accommodated by the growth of the more lateral area vitellina alone. These data may be used for a more analytical interpretation of the ways by which the growth of the vascular system in the area vasculosa occurs. Two main mechanisms seem to be implicated: the release of collaterals from vessels measuring 1.01–2 mm and the lengthening of 2.01–7 mm vessels. These mechanisms may be correlated with models for the generation of sequence of structures in the space during morphogenesis, as proposed by Meinhardt and Gierer 1980). In this way, the growth of blood vessels in the area vasculosa appears to be similar to the growth of other tubular structures during embryogenesis, such as the salivary glands and the lung (Bernfield et al. 1984), occurring through the release of collaterals and the lengthening of the primary tubular formations. In this study, we have analyzed by a modern morphometric approach aimed at characterizing, in four consecutive stages of development of the chick embryo area vasculosa, not only the overall structure of the vascular network (total length, number of vascular branches and branching points density), but also the modality of vessel ramification, and the spatial arrangement of the growing vascular tree.
⁎ Corresponding author at: Department of Basic Medical Sciences, Neurosciences and Sensory Organs, University of Bari Medical School, Policlinico - Piazza G. Cesare, 11, 70124 Bari, Italy. E-mail address: [email protected] (D. Ribatti).
https://doi.org/10.1016/j.mvr.2019.103935 Received 10 June 2019; Received in revised form 22 August 2019; Accepted 11 October 2019 0026-2862/ © 2019 Published by Elsevier Inc.
Microvascular Research 128 (2020) 103935
D. Guidolin, et al.
Fig. 1. Main steps of the image analysis procedure. A. Image of the area vasculosa at stage 1. B. Binary image of the vascular profiles. C. Branching points and branches connecting them. In red are shown examples of the circular regions centered on branching points and used to estimate the branching order associated to each branching point. D. Sholl analysis, involving the evaluation of the number of intersections of the vascular pattern with concentric circles of increasing radii around the central root point of the vascular tree. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Materials and methods
exclusive of all background an adaptive discrimination procedure was applied 5 (Russ 1995). This method operated with a local threshold: the mean grey value of a neighboring region was calculated for every pixel; this value plus an offset threshold constant defined the local threshold for that pixel. A 3 × 3 median filter was then applied to clean the resulting binary image (Fig. 1 B) and the area fraction covered by vessels was measured. By using binary thinning procedures, the skeleton of the binary image was finally derived and pruned to remove eventually present small artefactual branches (Fig. 1 C). From this image the total length of the vascular pattern was calculated, and branching points were identified as previously described (Guidolin et al. 2004b). The number of branching points and the number of branches forming the vascular tree were then evaluated. In addition, from the set of positions formed by the branching points parameters characterizing the spatial arrangement of the vascular tree (spatial disorder, SD) and the regularity of branching along the structure (topological disorder, TD) were estimated as previously described (Guidolin et al. 2004a). Both these parameters range from 0 to 1, lower values indicating a more uniform spatial distribution or branching occurrence respectively. Branching mode. The branching pattern is determined by the division
Forty fertilized White Leghorn chick eggs, staged according to Hamburger & Hamilton (HH) (Hamburger and Hamilton 1951) stages, were incubated from the start of their embryogenesis in an incubator under conditions of constant humidity at a temperature of 37 °C. At HH stage 13, a square window was opened in the egg shell after removal of 2–3 ml of albumen so as to detach the developing chorioallantoic membrane from the shell. The opening was closed with cellophane tape and the incubation continued. The area vitelline was photographed in ovo at 4 different stages of development accordingly to HH stages: HH 13 (stage 1), HH 20 (stage 2); HH 24 (stage 3); HH 27 (stage 4). Morphometry. Grey level images of the different stages of development of the area vasculosa were saved as TIFF files, processed and analyzed by using the ImageJ software (Schneider et al. 2012), freely available at http://rsb. info.nih.gov/ij/, and macro routines specifically developed by the authors. The main steps of the image analysis procedure are illustrated in Fig. 1. Global morphometric features. To select vessel profiles virtually 2
Microvascular Research 128 (2020) 103935
D. Guidolin, et al.
of a main trunk into smaller branches. Bifurcation and trifurcation are the most commonly described modes, although other types have been observed (Bradac 2011; Kopylova et al. 2017). As illustrated in Fig. 1 C, to characterize this feature a circular region (14 pixels diameter) centered on each branching point was explored to count the number n of branches in it. The number n-1 was then considered as the index of the branching mode occurring at the considered branching point. Sholl analysis. Sholl analysis (Sholl and Uttley 1953) was performed to quantify the variations in vascular distribution during vascular tree development. Previously used to quantify characteristics of dendritic processes branching off neuronal cell bodies, Sholl analysis examines the number of branches that intersect concentric circles of increasing radii around the central root point of the vascular tree (Fig. 1 D). To perform the analysis on the binary images of the vascular patterns a specific plugin for Image J was used (Ferreira et al. 2014) providing the number of intersections as a function of radius. From this distribution a metric named branching index (BI), that compares the difference in the number of intersections made in consecutive pairs of circles relative to the distance from the central focal point (Garcia-Segura and PerezMarquez 2014) for a definition and a discussion), was also derived. It gives a synthetic quantitative measure of vessel ramification.
Fig. 2. Changes in the number of vascular branches and in the branching points density (expressed as number of branching points per unit length of the vascular tree, BPL) during the observed period of development. * = p < .05 vs. the stage 1.
in the overall complexity of the pattern were also detected during its development. As illustrated in Fig. 2, a progressive increase of the number of vascular branches and of the branching points density was observed in the course of the tested period. These parameters exhibited a significant 51% and 14% increase respectively at stage 4 as compared to stage 1. Furthermore, the spatial arrangement of the vascular network significantly changed. Spatial disorder (SD) significantly decreased from stage 1 to stage 4, indicating a more uniform spatial occupancy of the vascular pattern, and a more regular pattern of branching was observed, as indicated by the significant decrease of topological disorder (TD). The results of the Sholl analysis were consistent with these findings but provided some additional information. BI values confirmed the above-mentioned increase in the number of branches with time. Moreover, as shown in Fig. 3 A, the data suggested that vessels initially ramified mainly at the periphery and subsequently in the regions around the central root point. As far as the mode of branching is concerned, vessel ramification was found to occur almost exclusively by bifurcation. In all samples, from > 80% of the branching points two branches emerged, while trifurcation or higher branching modes were associated to < 20% of the branching points (Fig. 3 B).
Statistics Differences between the four groups of samples were statistically tested by one-way analysis of variance followed by Bonferroni's test for multiple comparisons. The GraphPad Prism 3.0 statistical package (GraphPad Software Inc., San Diego CA, USA) was used for the analysis, and p < .05 was considered as the limit for statistical significance. Results At the HH stages 13 and 20 the area vasculosa surrounding the embryo was approximately round in shape. The trunk of the right vitelline artery ran in a nearly straight line for a short distance towards the lateral border of the area vasculosa. It then made a concave loop medially to run in the anterior, lateral and posterior parts of the area vasculosa where it terminated as a capillary network which continued peripherally with the sinus terminalis. From this latter structure, the blood was taken to the heart through the anterior and posterior vitelline veins. Blood was pumped away from the embryo through the omphalomesenteric arteries towards the sinus terminalis and returned through the anterior and posterior vitelline veins. At the same time, the intraembryonic vasculature had been established with dorsal aorta, branchial arch and cephalic arteries. At HH stages 24 and 27 the area vasculosa continued to expand and became denser; the formation of the posterior vein was completed and the lateral vitelline vein began to form. Morphometric data are summarized in Table 1. As shown, a significant 15% increase in the total length of the vascular network was observed at stage 4 as compared to the more primitive stage 1. Changes
Discussion At HH 18–20 stage the primary vascular network in the embryo and yolk sac is already formed through vasculogenesis and further blood vessel development proceeds through sprouting angiogenesis (Poole and Coffin 1989). Vascular endothelial growth factor (VEGF) and its receptors 1 and 2 (VEGFR-1 and VEGFR-2) are implicated in vasculogenesis and angiogenesis (Risau, 1997). We have previously demonstrated that VEGF is expressed in the endodermal cells of the chick embryo area vasculosa immediately adjacent to the mesodermal endothelial cells which, in turn, expressed VEGFR-2 (Nico et al. 2001). Estimation of a network complexity by quantification of the number
Table 1 Morphometric analysis of the area vasculosa (mean ± sem, N = 40). Stage
Area (mm2)
Length (mm)
BPL (n/mm)
Branches (n)
SD
TD
Branching order
BI
1
65.7 ± 2
330.9 ± 10
0.29 ± 0.009
170 ± 5
0.59 ± 0.02
0.69 ± 0.02
2.06 ± 0.06
28.1 ± 0.9
2 3 4
66.7 ± 2 67.6 ± 2 71.7 ± 2
373.0 ± 12 376.9 ± 12 378.0 ± 12
0.32 ± 0.010 0.33 ± 0.010 0.34 ± 0.010
219 ± 7 214 ± 7 257 ± 8
0.59 ± 0.02 0.56 ± 0.02 0.49 ± 0.01
0.67 ± 0.02 0.65 ± 0.02 0.060 ± 0.02
2.09 ± 0.07 2.01 ± 0.06 2.18 ± 0.07
33.6 ± 1 34.1 ± 1 35.2 ± 1
* * *
*
* * *
BP, branching point. SD, spatial disorder. TD, topological disorder. BI, branching index. ⁎ p < .05 vs. stage 1. 3
*
* *
* * *
Microvascular Research 128 (2020) 103935
D. Guidolin, et al.
Fig. 3. A. Values of the number of intersections of vessels with the consecutive concentric circles of the Sholl analysis at the different time steps considered. B. Typical distribution of the branching orders observed at the level of the branching points. As shown, the vast majority of branching points correspond to bifurcations.
adult (Maione and Giraudo 2015) tissues. A structural feature of particular interest in order to get a better understanding of tissue perfusion is the way the pattern of growing vessels fills the available space. In this respect, the results of the present study indicated that vascular spatial disorder significantly decreased during development in area vasculosa, suggesting a progressively more uniform occupancy of the tissue by the vascular pattern. Furthermore, a more regular pattern of branching was observed, as indicated by the significant decrease of topological disorder of the vascular tree. Consistent with this finding are the data provided by the Sholl analysis, suggesting that branching starts at the periphery of the growing vascular network and is subsequently extended to the regions near the root point. As mentioned before, an additional interesting aspect of vascular plasticity concerns the way the growing vascular network branches, since different patterns can be generated at each branching point. Knowledge of these variations are of particular importance when adult vascular networks are concerned, being helpful to surgeons and clinicians. In this respect, studies on cerebral vascularization suggested that bifurcation and trifurcation branching types are the most frequently observed (Cilliers and Page 2016). The results on the area vasculosa here presented show consistency with these findings. The vascular tree mainly grows by subsequent bifurcations, with only a minority of trifurcations. Branching patterns of higher order, likely representing crossover points (Bhuiyan et al. 2012), were seldom observed. A morphometric approach quite similar to the one used in the present study was also applied to analyze the embryonic lung branching morphogenesis (Carraro et al. 2009, 2014) and some morphological similarity can be identified between the pattern of development of the two structures in terms of branching complexity (i.e. branching point density and order of branching). This is not completely surprising, since in most of the biological tubular systems the branching pattern is the result of two signaling pathways, one promoting and another restricting branching (Horowitz and Simons 2008) and their balance is critical for producing a mature system that is neither suffused by ectopic branches, nor too sparse. The nature of these signals, however, differs between systems, as indicated by the branching promoting role fulfilled by FGF in the airway development but by VEGF in the area vasculosa (Nico et al. 2001).
of branching points and related measures (such as number of branches and length) are well recognized morphometric parameters to quantitatively describe a vascular network and the changes it undergoes during sprouting angiogenesis (Guidolin et al. 2004a, 2004b; Zudaire et al. 2011; Belle et al. 2014). They are also the main output provided by presently available integrated software packages for the image analysis of vascular networks, such as ‘AutoTube’ (Montoya-Zegarra et al., 2018) and ‘AngioQuant’ (Niemistö et al. 2005). In the present paper, starting from these primary parameters (basically characterizing the overall structure of the network), the morphometric analysis has been further expanded to characterize two additional morphological features of a vascular tree. The first concerns the way the pattern of growing vessels fills the available space. This feature is of interest to get a better understanding of tissue perfusion and has been characterized by parameters capturing the level of spatial order/disorder exhibited by the vascular network. These parameters increase as far as the spatial distribution of the branching points in the tissue (spatial disorder) or the occurrence of branching points along the tree structure (topological disorder) become less uniform. A second question of interest is how the network grows during the angiogenic process. In this respect, two aspects deserve consideration from the morphological standpoint, namely how a main trunk divides into smaller branches, and how the branching occurs at different distances from the root point of the vascular tree. These issues were here addressed by automatic image analysis procedures estimating the branching order at each branching point and exploiting the so-called Sholl analysis (i.e. the enumeration of vessels at varying distances from anchor points) respectively. Thus, in view of the increasing importance (as pointed out by the available literature) of angiogenesis and vasculogenesis under normal and pathological conditions the possibility to expand the information pattern obtainable following a morphometric approach could be of particular interest to the field. Some of these morphometric strategies were previously explored to characterize the regional and developmental variations of vascular network in the chick embryo CAM (Guidolin et al. 2004b; Reizis, 2005; Belle et al. 2014) or have been applied to CT images of the coronary arterial tree as a time-efficient approach to obtain data useful to quantitatively assess hypotheses on the coronary circulation (Wischgoll et al. 2009). Their combined use, however, could provide a more complete and quantitative picture of the morphological features of a vascular network. The present study was focused on the chick embryo area vasculosa and reports a first, at least in our knowledge, quantitative analysis of its morphological features during development. We have demonstrated that in the course of the development of the chick embryo area vasculosa there is a significant 15% increase in the total length of the vascular network associated with a progressive increase of the number of vascular branches and of the branching points density. From the functional standpoint, the increase of these parameters characterizing the vascular network has been correlated with an optimization of gas exchange and metabolic support in both embryonic (Reizis, 2005) and
Declaration of interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. References Augustine, J.M., 1970. Expansion of the area vasculosa of the chick after removal of the ectoderm. J Embryol Exp Morphol 24, 95–108. Belle, J., et al., 2014. Stretch-induced intussuceptive and sprouting angiogenesis in the chick chorioallantoic membrane. Microvasc. Res. 95, 60–67.
4
Microvascular Research 128 (2020) 103935
D. Guidolin, et al.
Mayer, B.W., Packard, D.S., 1978. A study of the expansion of the chick area vasculosa. Dev. Biol. 63, 335–351. Meinhardt, H., Gierer, A., 1980. Generation and regeneration of sequence of structures during morphogenesis. J. Theor. Biol. 85, 429–450. Montoya-Zegarra, J.A., et al., 2018. AutoTube: a novel software for the automated morphometric analysis of vascular networks in tissues. Angiogenesis 22, 223–236. Nico, B., et al., 2001. Vascular endothelial growth factor and vascular endothelial growth factor receptor-2 expression in the chick embryo area vasculosa. Histochem. J. 33, 283–286. Niemistö, A., Dunmire, V., Yli-Harja, O., Zhang, W., Shmulevich, I., 2005. Robust quantification of in vitro angiogenesis through image analysis. IEEE Trans. Med. Imaging 24, 549–553. Poole, T.J., Coffin, J.D., 1989. Vasculogenesis and angiogenesis: two distinct morphogenetic mechanisms establish embryonic vascular pattern. J. Exp. Zool. 251, 224–231. Reizis, A., 2005. Regional and developmental variations of blood vessel morphometry in the chick embryo chorioallantoic membrane. J. Exp. Biol. 208, 2483–2488. Risau, W., 1997. Mechanisms of angiogenesis. Nature 386, 671–674. Romanoff, A., 1960. The extraembryonic membranes. In: The Avian Embryo: Structural and Functional Development. MacMillan, New York, pp. 1039–1141. Russ, J., 1995. The Image Processing Handbook. CRC Press, Boca Raton, FL. Schneider, C.A., et al., 2012. NIH image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675. Sholl, D.A., Uttley, A.M., 1953. Pattern discrimination and the visual cortex. Nature 171, 387–388. Stewart, R., et al., 1990. Growth of the chick area vasculosa in ovo and in shell-less culture. J. Anat. 172, 81–87. Wischgoll, T., et al., 2009. Extraction of morphometry and branching angles of porcine coronary arterial tree from CT images. Am. J. Phys. Heart Circ. Phys. 297, H1949–H1955. Zudaire, E., et al., 2011. A computational tool for quantitative analysis of vascular networks. PLoS One 6, e27385.
Bernfield, M., et al., 1984. Remodelling of the basement membrane as a mechanism of morphogenetic tissue interaction. In: Trelstad, R.L. (Ed.), The Role of Extracellular Matrix in Development. Alan R. Liss, New York, pp. 542–572. Bhuiyan, A., et al., 2012. Detection and classification of bifurcation and branch points on retinal vascular network. In: 2012 International Conference on Digital Image Computing Techniques and Applications (DICTA). IEEE. Bradac, G.B., 2011. Vascular Territories. Cerebral Angiography. Springer Berlin Heidelberg, pp. 87–89. Carraro, G., et al., 2009. miR-17 family of microRNAs controls FGF10-mediated embryonic lung epithelial branching morphogenesis through MAPK14 and STAT3 regulation of E-cadherin distribution. Dev. Biol. 333, 238–250. Carraro, G., et al., 2014. miR-142-3p balances proliferation and differentiation of mesenchymal cells during lung development. Development 141, 1272–1281. Cilliers, K., Page, B.J., 2016. Review on the anatomy of the middle cerebral artery; cortical branches, branching pattern and anomalies. Turkish Neurosurgery 26, 653–661. Ferreira, T.A., et al., 2014. Neuronal morphometry directly from bitmap images. Nat. Methods 11, 982–984. Garcia-Segura, L.M., Perez-Marquez, J., 2014. A new mathematical function to evaluate neuronal morphology using the Sholl analysis. J. Neurosci. Methods 226, 103–109. Guidolin, D., et al., 2004a. Order and disorder in the vascular network. Leukemia 18, 1745–1750. Guidolin, D., et al., 2004b. A new image analysis method based on topological and fractal parameters to evaluate the angiostatic activity of docetaxel by using the Matrigel assay in vitro. Microvasc. Res. 67, 117–124. Hamburger, V., Hamilton, H.L., 1951. A series of normal stages in the development of the chick embryo. J. Morphol. 88, 49–92. Horowitz, A., Simons, M., 2008. Branching morphogenesis. Circ. Res. 103, 784–795. Kopylova, V.S., et al., 2017. Fundamental principles of vascular network topology. Biochem. Soc. Trans. 45, 839–844. Maione, F., Giraudo, E., 2015. Tumor angiogenesis: Methods to analyze tumor vasculature and vessel normalization in mouse models of cancer. In: Methods in Molecular Biology. Springer New York, pp. 349–365.
5