Indian J Physiol Pharmacol 1999; 43 (3): 315-322

A SIMULTANEOUS EX VIVO MODEL OF EMBRYOGENESISII. VASCULOGENESIS

NANDINI DEY* AND BHANU IYENGAR

Institute of Pathology (lCMR),Safdarjung Hospital Campus,New Delhi - 110 029

( Received 00 April 10, 1999)

Abstract: Vasculogenesis was simultaneously studied with embryogenesisin in ovo chick embryo culture, whi h was harvested at 40 hours.Endodermal cells and vascular endothelial cells were studied using a newcombination of stains, immunohistochemistry (for nuclei and basementmembrane) and NADPH-diaphorase activity in whole-mounts, paraffinsections and etched semithin sections. The model can be used for thestudy of developmental process of blood vessels as well as embryonicphysiology of blood vessels vis-a-vis organogenesis in response to differentangiogenic agents, drug trials, cancer therapy by angiostatic chemicals/radiations and toxins. Considering that vasculogenesis/angiogenesis as oneof the fundamental phenomena in physiology, pathophysiology, toxicologyand pharmacology of developmental sciences, the model in developingembryo is presented.

Key words: in ovo culture plastic sections NADPH-diaphorase

whole mounts IHC endothelial cells

INTRODUCTION

The phenomenon of capillary formationhas been considered as a basic event inreproduction, development, angiogenicdiseases, tumour and repair. Embryonicmicrovasculature is so far studied in foetusor in neonates of chick and mammals forthe study of neural transplant (1-5),Embryonic microvasculature in general hasalso been studied for various aspects indifferent animals in central nervous system(6-12). Literature survey reveals that thestudies of embryonic microvasculaturedevelopment were either carried out for theneural transplants/grafts or the late stage

*Corresponding Author

embryonic study of specific brain regions/blood-brain-barrier. Hence in all thesestudies, development of vasculature wasstudied at the particular stage, when theinitial capillary formation has alreadyoccurred. In fact, in neither of these studies,early development of capillaries (earliestpossible stage along with organogenesis) hasbeen studied. The present study wasconducted to use chick embryo as a modelfor the development of the blood vessels vis­a-vis organogenesis. In an earlier articlescopes and avenues of the organogenesismodel have been presented (5a). The presentarticle puts forward the morphological dataof a (simultaneous) model for the earliest

316 Dey and Iyengar

vasculogenesis, which can be utilized for awide variety of studies. Positive technicalaspects of the model are that it is easy tostandardize, highly reproducible and allowsa good range of experimental manipulations.It also resolves the ethical problems ofhandling big laboratory animals. Thus forthe initial information regarding thephenomenon of early vasculogenesis vis-a­vis organogenesis the model can be readilyutilized. The report suggests the use of themodel at the preliminary stages of drug­trials, which can subsequently be tested III

the mammalian system.

METHODS

Blood vessels harvested from 40 hoursold in DUO cultured chick (Leg horn; Gallusgallus) embryos (the total number ofembryos used for the standardization andexperiment varied from 120-150). Time andconditions of cultures are as described inthe preceding article. Blood vessels are fixedin situ using formalglutaraldehyde(4°C). Insome experiments thicker vessels weremicrodissected out and perfused withfixatives for better results (especially forplastic section). In addition to routine H &E, a new combination of stains was tried onthe extraembryonic blood vessels (asdescribed in preceding article). For theparaffin sections, the blood vessels weremounted in a flat condition as well as rolledto get a consistency (whichever wassuitable). Haematoxylin with either eosin,pyronin or PAS were used for staining.Semithin sections for light microscopy weredone at the ll-l thickness in spurr or araldite(EM Sciences). Etching of plastic sectionsand subsequent staining were done asmentioned in the preceding article. Whole

Indian J Physial Pharmacal 1999; 43(3)

mount immunohistochemistry of postperfused and fixed blood vessels were donefor laminin (polyclonaD and Ki 67(monoclonal) (Dill: 100; DAKO USA) usingABC kit (Vector Laboratories USA) withdiaminobenzedine (DAB) (0.5 mg/rol of DABand 0.01% HzO z) as chromogen. Details ofthe IRC procedure is mentioned in thepreceding article (5a). Haematoxylin wasused as counter stain. Whole mountNADPR·diaphorase staining was performedas described in the preceding article. Sincezona opaca and zona pellucida are ellipticalin shape, their area were determined fromlesser and greater diameter's (from cameralucida drawings) using Olympus microscope(camera lucida factor 9.9). Parametersdetermined are zona opaca area, zonapeHucida area, zona opaca index (obtainedfrom the ratio of zona opaca area and bodylength), zona pellucida index (obtained fromthe ratio of zona pellucida area and bodylength). Mean ± S.E.M. were used for thecomputation of data. Students 't' test wasapplied to get significance levels of thedifferences between means. To reviewthe vascular map, photographs of theembryo were taken in cultured conditionsafter injecting 0.2% trypan bluesubblastodermaly. DAB, NBT, NADPH,trypan blue and other AR grade laboratoryreagents were procured from Sigma USA.Routine histological stains were obtainedfrom the local sources. Molarity, percentageand the sources of other chemic?ls werementioned in the preceding article.

RESULTS

Su bblastodermal inj ections of 0.2%trypan blue to in DUO embryos show thecirculation map at 40 hours of development

Indian J Physiol Pharmacol 1999; 43(3) Model of Embryogenesis: Vasculogenesis 317

Fig. 1 : Photomicrograph of paraffin (A, B, C) and semithin plastic (D) sections of 40 hours old embryos stainedusing H & E (A), pyronin and H (B, Dl, and PAS and H (C). Photograph A (40 x101 and B [100 x 10[ showzona pellucida (inset A: embryo in in ouo); similarly C [100 x 101: zona opaca (arrow at empty fat droplets,open arrow at dark granular inclusions); D [100 x 101: endothelium (arrow at the presumptive basementmembrane).

(Fig. lA inset). In the whole mounts of theblood vessels from 40 hour-old embryos twodistinct regions are identified: (i) thin layersof vessels with serosal covering thatimmediately encircles the growing embryo.This region is zona pellucida. (ii) outsidezona pellucid a lies the thick-endodermal­cell-rich opaque zone of blood vessels calledzona opaca. Zonae are elliptical inshape covering the zona vascularis, whichem bodies sin us terminalis. Theextraembryonic tissue is made up of (i) innerendodermal cell layer immediate to yolk (ii)outer splanchnic mesoderm whichdifferentiates into blood vessels, bloodislands and mesenchymal cells(extraembryonic coelom lies betweensplanchnic and somatic mesoderm) and (iii)outer serosal layer.

With the new combination of stainsthe zona pellucid a, zona opaca and theirtransitional zone are well-visualized (Fig.IC and IE, Ref. 1). The thinner vesselsof zona pellucida are stained moredifferentially by indigocarmine whilezona opaca blood vessels are moreprominently lined by dark brown to redstaining endodermal cells (Fig. IE [arrow],Ref. 5a). Blood cells in clumps tend to takePAS positive colour. A typical transitionalzone is shown in Fig. lC (5a). Table I showsthe parameters of the extraembryonicregions.

Blood vessels

The mesoderm-derived blood vesselsaround 40 hours of incubation contain

318 Dey and Iyengar Indian J Physiol Pharmacol 1999; 43(3)

Fig. 2: Photomicrograph of whole mounts (A, B, C, D) of the embryos stained for NADPH-diaphorase activity(A), laminin antigen (B,D) and Ki 67 antigen (C). Photograph A [6.3 x 6.31 shows both zona opaca (aITow)and pellucida (open arrow); similarly B [100 x'10]: zona pellucid a (arrow at the presumptive basementmembrane deposition, inset [20 x 101; C [40 x 10J:. zona pellucid a (arrow at Ki 67 positive nuclei, inset[100 x 10]); D [6.3 x 6.3J: zona opaca (arrow atlaminin positive endothelium, inset [10 x 101).

endothelial linings. In paraffin sections (Fig.lA, IB and lC), blood cells are observedonly inside the endothelial encasements. Insemithin sections the spindle shapedendothelial nucleus and its thin film ofcytoplasm is observed (Fig. ID). Part of theserosal layer is seen on the other side ofblood vessels. Presumptive basementmembrane can be poorly identified at onearea (Fig. ID, arrow). However, the vesselis seen incompletely closed.Immunohistochemical staining for Ki 67 inwhole-mounts shows endothelial cell nucleias dots in low magnification (Fig. 2 C,arrow). In high magnification (inset) theylook like pearls (spindle shaped) in thestring. In contrast, endothelium apears(when stained for laminin) like a continuous

line in low magnification (Fig. 2D, arrow)as well as in high magnification (Fig. 2D,inset). However, in oil immersion objective(lOOx) laminin is shown to stain thepresumptive basal lamina (Fig. 2B).Interestingly, the staining appearsdiscontinuous (Fig. 2B, arrow). Inset of thesame picture shows laminin staining inpositive controls from different area of thesame specimen.

Blood vessels associated endodermal cells

In zona pellucid a, blood vessels appearcharacteristically thinner with cuboidalendodermal cells rich in PAS positivity. Theendodermal cells have eccentric nucleusand histochemically highly reactive

Indian J Physiol Pharmacol 1999; 43(3) Model of Embryogenesis: Vasculogenesis 319

TABLE I : Quantification of Extraembryonic regions from animals harvested at forty-hours of in DUO culture

Diagramaticrepresentation

EXTRA EMBRYONIC

AREAS

Items Description ValuesMean ±S.E.M.

Zona opaca Elliptical 82.64 mm 2 ±9.85 (25)(Zo) Area

Zona Pellucid a Elliptical 17.38 mm 2 ±1.58 (25)(Zp) Area

Zona opaca [Zo \B11 13.29 mm±l.Ol (25)Index

Zona Pellucid a [Zp\Bll 2.785 mm±O.23 (25)Index

homogeneous cytoplasmic substances (thatreact with PAS, alcian blue, safranin, DAB,picric acid, toluidin blue, eosin, pyronin andBiebrich scarlet acid fuchsin). In paraffinsections, they appear strongly positive formost of the stains (results not shown).Figure lA shows blood vessels of this areawith adjacent endodermal cells containingthe yolk drops. The characteristic featureof these cuboidal endodermal cells is thatthese are loosely arranged aroundblood vessels as compared to zona opaca(compare Fig. IB & lC) although theyappear intensely dark stained. On the otherhand, endodermal cells of zona opaca arecolumnar in structure (Fig. lC) which arearranged tightly along the blood vessels.They contain fatty substances that giveempty appearance in paraffin sections (Fig.lC, arrow). The other featues of thecytosolic material of these cells are thepresence of pleomorphic granular drops. Inparaffin sections, conspicuous membranebound dark granular structures arealso seen in zona opaca endodermal cells(Fig. lC, open arrow).

When both zonae are stained for theNADPH-diaphorase activity, they showcertain histochemical differentiations. Thediaphorase activity is seen in the opaca

vessels as well as adjacent endodermal cells(Fig. 2A, arrow) in contrast to zona pellucidawhere vessels only are positive (Fig. 2A, openarrow).

DISCUSSION

Development of endothelium duringembryogenesis occurs via bothvasculogenesis (endothetial cells originatingfrom the progenitor cell types) andangiogenesis (new capillary is born fromexisting vessels) in contrast to adults whereangiogenesis is the rule (13). Mobbs andMcMillan (14) found that in chick, themesodermal layer differentiates into bloodvessels when studied between stage 11 and15 of Hamburger and Hamilton (15). Sincestage 11 corresponds to 34 hours and stage15 corresponds to 48 hours of incubation,and our study is restricted to the initial 40hours of development, we considervasculogenesis to be under operation (16).At this hour the primary circulation (extraembryonic circulation to yolk sac) isestablished (14, 17). However, like otherswe also found it difficult to identify arteriesor veins except by the direction of the bloodflow following the subblastodermal injectionof dye (14).

320 Dey and Iyengar

The study of vasculogenesis has beendone either on CAM or in shell-less cultures(18-22). Recently Hirashima et al studiedthe proliferation, differentiation 'and cell­cell adhesion of endothelial cell progenitorsin vitro (23), In this study, the cellularevents that occur during vasculogenesishave been studied using vascularendothelial-cadherin, pIafelet-endothelialcell adhesion molecule-l and vascularendothelial growth factor.

Here we report the results of the studyat 40 hours of development (between stage11 and 15) which is earlier as compared toboth CAM study (mostly starts from day 6)and shell-less cultures (starts from stage 15­23 corresponding to 2-3 days of incubation).For the zonae of vasculature, indices werechosen to be a better representation of bodylength of the embryos on which the areasof the zonae are dependent. The ratio ofzonal areas (zona opaca area: zona pellucid aarea) is 4,75 which is comparable to theratio of zonal indices (4.77).

'The differences in the staining patternof the zonae (using the new combination ofstains) is_attributed to the endodermal cellsassociated with the vasculogenesis. Thehigh-stained character of zona opacaendodermal cells can be explained by theirhigh content of cytoplasmic materials. Sincethe endoderm of avian yolk sac abutsdirectly on the yolk mass (nutrient store)and is primarily involved in transport ofyolk to extraembryonic (as well asembryonic) circulation, the study of theendodermal cells associated to vasculogenicareas is justified (14). These cells containboth lipoid and protein materials (14). Yolk­granules and cells of zona opaca have low

Indian J Physiol Pharmacol 1999; 43(3)

density lipid, phosvitin, ex, P-lipovitellin,calcium, iron, phospholipid, phosphoprotein,polysaccharides which can explain andmatch their highly staining nature(including PAS positivity) with theirfunction (actively absorbing region of yolksac in contrast to zona pellucida (see 14,16). Presence of empty appearance(vaculated) of endodermal cells in paraffinsection in our study confirms the inclusionof lipid materials as observed by others (16).The respective cuboidal and columnarstructure of zona pellucida and zona opacaendodermal cells with their pleomorphic/granular drops in this study are also inconfirmation with another study thatcorrelates their function (absorption) withmorphological features (high endocytoticactivity) (14). Since zonae are metabolicallyhighly active areas, we studied NADPH­diaphorase activity in these regions. To ourknowledge, we are the first to reportdiaphorase activity in early blood vesselsand associated endodermal cells (zonaopaca). It is interesting to note that theenzyme activity matches the metabolicstatus of zona opaca endodermal cells.

Proliferation is one of the characteristicfeatures of blood vessel development (see13), In order to know the status of theendothelial cells, blood vessels in whole­mount have been stained for Ki 67 whichshows the proliferative nature of endothelialnucleus. Although the blood vessels inwhole-mounts and paraffin sections mostlyappear continuous, in a few areas ofsemi thin sections, the vascular endotheliumis discontinuous. However, we have notfound such a discontinuity as often as Mobbsand McMillian (14). Since basal lamina isthe important component of basement

Indian J Physiol Pharmacol 1999; 43(3)

membrane formation In capillaryendothelium, we stained the whole-mountof vessels for laminin. Antisera for Ki 67and laminin were used in chick because oftheir highly conserved nature as mentionedby several authors (24-26). Lamininantibody has already been used forimmunoreactivity in chick embryos byRibatti et al (27). In addition, both positiveand negative controls were run duringstandardization and experiments (28-29).

Negative controls were run using buffer inplace of primary antibody in positive controlslides and the test samples.

In oil immersion objectives, the lamininstaining appears incomplete anddiscontinuous as observed by others (14).In fact, the basal lamina is better observedin laminin stain than semithin sections.

Model of Embryogenesis: Vasculogenesis 321

Allen and Wilson ultrastructurally identifiedthe lack of basement membrane in 3 dayold area vasculosa vessels, a capillarycharacter similar to the tumour associatedangiogenesis (22). Our results of incompleteand discontinuous laminin deposition as wellas its rare and poor morphologicalappearance in the presumptive basementmembrane area (in semithin sections)explains the similar lack of basementmembrane encouraging such a model foruse In study of tumour associatedangiogenesis.

ACKNOWLEDGEMENTS

Dr. N andini Dey is RA (l CMR).Photographs were taken by Mr. R Joshi(MET Division, lOP). Technical assistancewas provided by Mrs. M. B. Jagannath.

REFERENCES

1. Broadwell RD, Charlton HM, Ganong WF, Salem anM, Sofroniew M. Allografts of CNS tissue possessa blood-brain barrier. I grafts of medial preopticarea in hypogonadal mice. Exp Neural 1989; 105 :135-151.

2. Rosestein JM. Neocortical transplants in themammalian brain lack a blood-brain barrier tomacromolecules. Science 1987; 235: 772-774.

3. Swenson RS, Shaw P, Alones V, Kozlowski G,Zimmer J, Castro AJ. Neocortical transplantgrafted into the new born rat brain demonstrate ablood-brain-barrier to macromolecules. NeurasciLett 1989; 100: 25-39.

4. Krum JM, Rosenstein JM. Patterns of angiogenesisin neural transplant model. I autonomictissue transplants. J Camp Neural 1987; 258: 420­434.

5. Krum JM, Rosenstein JM. Patterns of angiogenesisin neural transplant model. II fetal neocorticaltransplants. J Camp Neural 1988; 271: 331-345.

(5a) Dey N, Mukherjee A, Iyengar B. Asimultaneous ex vivo model of embryogenesis: I.Organogenesis. Indian J Physiol Pharmacal 1999;43 :

6. Risua W, Gautschi-Sova P, Bohlen P. Endothelialcell growth factors in embryonic and adult chickbrain are related to human acidic fibroblast growthfactor. EMBO J 1988; 7: 959-962.

7. Bauer HC, Steiner M, Bauer H. Embryonicdevelopment of the CNS microvasculature in themouse: new insights into the structuralmechanisms of early angiogenesis. In steiner R,Weisz PB, Langer R, ed. Angiogenesis BirkhauserVerlag 1992; 64-68.

8. Yoshida Y, Yamada M, Wakabayashi K, Ijuta F.Endothelial fenestrae in the rat fetal cerebrum.Dev Brain Res 1988; 44: 211-216.

9. Marin-Padilla M. Early vascularization of theembryonic cerebral cortex: golgi and electronmicroscopic studies. J Camp Neural 1985; 241:237-249.

10. Braun LD, Cornford EM, Oldendrof WHo Newbornrabbit blood-brain-barrier is selectively permeableand differs substantially from the adult. JNeurochem 1980; 34: 147-152.

11. Wakai S, Hirokawa N. Development of the blood­brain-barrier in the chick embryo. Cell Tiss Res1978; 195: 195-203.

322 Dey and Iyengar

12. Herken R, Goetz W, Wattjes KH. Initialdevelopment of capillaries in the neuroendotheliumof the mouse. J Anat 1989; 164: 85-92.

13. Hanahan D, Folkman J. Patterns and emergingmechanisms of the angiogenic switch duringtumorogenesis. Cell 1996; 86: 353-364.

14. Mobbs IG, McMillan DB. Structure of theendodermal epithelium of the chick yolk sac duringearly stages of development. Am J Anal 1979; 155 :287-310.

15. Hamburger V, Hamilton HL. A series of normalstages in the development of the chick embryo. JMorph 1951; 88: 48-96.

16. Bodemer CWo Modern embryology. Holt Rinehartand Winston, Inc. 1968; 193-213.

17. Cirotto C, Arangi 1. How do avian embryosbreathe? Oxygen transport in the blood of earlychick embryos. Camp Biochem Physiol 1989; 94:607-613.

18. Iruela-Arisepe ML, Lane TF, Redmond D, ReillyM, Bodender RP, Kavanagh TJ, Sage EH.Expression of SPARC during development of thechicken chorioallantoic membrane: evidence forregulated proteolysis in vivo. Mol Bioi Cell 1995;6: 327-343.

19. Henry CB, Defluw DO. Differential lectin bindingto microvascular endothelial glycoconjugatesduring normal angiogenesis in the chickchorioallantoic membrane. Microuasc Res 1995; 49 :201-211.

20. Rizzo V, Steinfeld R, Kyriakides C, Defouw DO.The microvascular unit of the 6 day chickchorioallantoic membrane: a fluorescent confocalmicroscopic and ultrastructural morphometricanalysis of endothelial selectivity. Microuasc Res1993; 46: 320-332.

Indian J Physiol Pharmacol 1999; 43(3)

21. Stewart R, Nelson J, Wilson DJ. Growth of thechick area vasculosa in ouo and in shell-lessculture. J Anal 1990; 172 : 81-87.

22. Allen WE, Wilson DJ. Early embryonicangiogenesis in the chick area vasculosa. J Anal1993; 183: 579-585.

23. Hirashma M, Kataoka H, Nishikawa S, MatsuyoshiN, Nishikawa S. Maturation of embryonicstem cells into endothelial cells in vitromodel of vasculogenesis. Blood 1999; 93: 1253­1263.

24. Sawhney N, Hall PA. Ki 67-Structure, functionand new antibodies. J Palhol 1992; 168 : 161-162.

25. Key G, Becker MH, Baron B, Duchrow M, SchueterC, Flat HD. New Ki 67 equivalent murinemonoclonal antibodies (MIB 1-3) generated againtbacterially expressed parts of the Ki 67 cDNAcontaining 62 base pairs repetitive elementsencoding for the Ki 67 epitope. Lab Invesl 1993;68 : 629-636.

26. Engel J. Laminins and other strange proteins.Biochem 1992; 31: 10643-10651.

27. Ribatti D, Bertossi M, Nico D, Vacca A,Ria R, Roncali L and Presta M. Role of basicfibroblast growth factor in the formationof the capillary plexus in the chick embryochorioallantoic membrane. An in situhybridization, immunohistochemical andultrastructural study. J Submicrosc Cytol Palhol1998; 30: 127-136.

28. Rose DSC, Maddox PH, Brown DC.Which proliferation markers for routineimmunohistology? A comparison of five antibodies.J Clin Pathol 1994; 47: 1010-1014.

29. Balaskas C and Gabella G. Lamininimmunoreactivity in enteric ganglia of the chickembryo. Cell Tissue Res 1997; 289: 243-251.