Uploaded by User3977


Path. Res. Pract. 192, 1068-1076 (1996)
The Chick Embryo Chorioallantoic Membrane as an in vivo
Wound Healing Model
D. Ribatti\ A. Vacca 2 , G. Ranieri 2 , S. Sorino1 and L. Roncalj1
11nstitute of Human Anatomy, Histology and Embryology,
2Department of Biomedical Sciences and Human Oncology, University of Bari
Medical School, Bari, Italy
The chick embryo chorioallantoic membrane (CAM) was used as an in vivo wound
healing model. A full excision of a 1 mm 2 CAM area was filled by a granulation tissue
after 96-120 h, which eventually formed a scar in 75% of the cases. In the remaining
25%, a solution of continuity was left which, however, was smaller in size than the
One observed immediately after the excision. Under the microscope, the CAM area
involved in the repair process showed: i. hyperplasia of the chorionic epithelium; ii.
about three times as many microvessels and fibroblasts in the mesenchyme as in the
normal adjacent control regions; iii. an inflammatory infiltrate mostly consisting of
macrophages; and iv. a strong positivity for fibronectin in the extracellular matrix.
The validity of this experimental model appears to be confirmed by the fact that
we were able to reproduce all the critical events controlling the wound healing process, such as re-epithelization, angiogenesis, formation of an inflammatory infiltrate
and deposition of one of the main constituents of the extracellular matrix, such as
Wound healing (WH) is characterized by the formation of a granulation tissue consisting of inflammatory
cells, newly-formed blood vessels and fibroblasts embedded in a loose collagenous extracellular matrix.
Re-epithelization, angiogenesis and matrix deposition
are critical events controlling this process 8•
Angiogenesis is confined to the wound site 1 and
plays a pivotal role for successful WHo Revascularization is required to furnish the new tissue and metabolities, and to dispose of the waste products of
metabolism. Angiogenesis occurs as a higher regulated
process which is rapidly stimulated after injury and
ceases when WH is complete l2 •
Extracellular matrix proteins are known to be important in many phases of WH. Among these proteins, fibronectin, a glycoprotein secreted by
proliferating endothelial cells ,21, forms the provisional extracellular matrix of wounds 14,19 and promotes migration of endothelial cells over the wound 27•
A frequently employed method to stimulate WH in
vivo is the subcutaneous implantation of sponges 9 • In
this work, we examined the modes of WH development
in an experimentally induced lesion in the chick embryo chorioallantoic membrane (CAM). Specifically,
we examined WH with respect to the amount of the
associated vasoproliferative processes, the characteristics of the inflammatory infiltrate and of the extracellular matrix adjacent to the site of the lesion. Our
© 1996 by Gustav Fischer Verlag, Stuttgart
The CAM as an in vivo Wound Healing Model· 1069
Table 1. Quantitation of microvessels in the CAM's intermediate mesenchyme
Under the wound
Distant from the wound (Control)
Microvessel density (%)a
No. of intersection points b
54.5±5.2 C
a For specification see text, b Mean ± SD, C Statistically significant difference p < 0.001 vs number of intersection points distant
from the wound.
Table 2. Quantitation of fibroblasts in the CAM's intermediate mesenchyme
Under the wound
Distant from the wound (Control)
Fibroblast density(%)a
No. of intersection points b
80.4±12.5 C
a For specification see text, b Mean+SD, C Statistically significant difference p < 0.001 vs number of intersection points distant
from the wound.
findings suggest that CAM can be regarded as an in
vivo WH model.
Macroscopic Examination
The CAM region surrounding the continuity solution showed swift vasoproliferative processes as early
as 24 h after excision (HH stage 36) (Fig.1a). In
75% of the cases, the solution of continuity was filled
by a granulation tissue within 48 to 72 h (HH stages
37-3 8). The scar developed upon this tissue during
the next 24-48 h (HH stages 39-40) (Fig.1b). In
the remaining 25% of the specimens, a solution of continuity smaller than the initial one was still present.
Microscopic Examination
Where the macroscopic examination showed that
the solution of continuity had been filled by granulation tissue, the histologic examination confirmed that
CAM continuity had been fully restored. Conversely,
when a macroscopic analysis revealed that the continuity solution had only been filled by the scar tissue, an
optical empty space between the two juxtaposed CAM
stumps (Fig.2a) could be observed under the microscope. The chorionic epithelium of the stumps, normally consisting of 1 or 2 cell layers, was here
formed by 2 to 6 layers (Fig.2 a-c). Within 96120 h following the lesion, microvessels and fibroblasts density in the underlying mesenchyme was
about three times as high as in the adjacent normal control regions (Fig. 2 b, c Tables 1, 2}. The mesenchyme
also showed the presence of an inflammatory infiltrate, mostly distributed in a perivascular position
(Fig. 2 d} and consisting prevailingly of macrophages
that reacted positively to naphthol-AS-D-chloroace-
tate esterase activity. Under normal conditions these
cells are located almost exclusively inside the vessels
(Fig.3 a); instead, where the lesion had been performed experimentally, their location was mostly interstitial (Fig. 3 b). Lastly, the extracellular matrix reacted
with fibronectin in the mesenchyme beneath the lesion.
Fibronectin deposition started 24 h after the lesion had
been produced and reached its maximum extension
after 24-48 h (Fig.4).
In this work, we have followed the time-course of the
repair process of an experimentally induced wound in
the chick embryo CAM. In 75% of the cases, a 1 mm 2
resection area forms a scar some 96 -120 h later. The
scar itself consists of a granulation tissue formed by
newly-formed vessels, fibroblasts and inflammatory infiltrate essentially composed of macrophages.
The process of WH depends upon a variety of interactions between cells and the extracellular matrix components, and wound repair is in part mediated by
multiple growth factors released by inflammatory cells
and from their storage sites of the extracellular matrix.
In the CAM mesenchyme beneath the lesion we demonstrated that the number of microvessels was three
times higher than in the normal adjacent regions which
were considered as the control. WH angiogenesis has
been extensively studied 1,22 and is characterized by disruption of the integrity of endothelial cell monolayers
in pre-existing microvessels, basement membrane lysis,
endothelial migration and proliferation, lumen and
capillary loop formation. Angiogenesis is an absolute
requirement for the development of the granulation tissue during WH. For example, under the category of
wound repair the healing of peptic ulcers and of myocardial infarction is dependent upon angiogenesis 11.
Healing is accelerated by the use of exogenous angio-
1070 . D. Ribatti et al.
Fig. 1 a - b a: 1O-day old CAM (HH stage 36), 24 h after a full thickness excision of a small CAM's area (asterisk). The margins
of the lesion show an intense vasoproliferative response - b: 14-day old CAM (HH stage 40), 120 h after the excision. The
solution of continuity has been filled by a scar tissue (arrows). Original magnifications: a, x30; b, x20.
The CAM as an in vivo Wound Healing Model· 1071
Fig. 2 a-d. Light micrographs showing in (a) an optically empty space (arrow) between the two juxtaposed CAM stumps of an
experimental case in which continuity solution had been partially filled by the scar tissue. Note in (b, c) the two stumps photographed at higher magnification showing the hyperplasia of the chorionic epithelium (C) with the increase of vessels and
fibroblasts in the underlying mesenchyme (M), and in (d) inflammatory cells located in perivascular position. Original magnifications: a, x 100; b, d, x400; c, x250.
1072 . D. Ribatti et al.
Fig. 2c, d.
The CAM as an in vivo Wound Healing Model· 1073
• f', <;"
to .•
... ',,'
Fig. 3 a-b. Naphtol-AS-D-chloroacetate esterase positive cells in a control (a) and an experimental (b) case. In (a) the cells are
prevalently located inside the vessels; in (b) they are outside, in the CAM's mesenchyme. Original magnifications: a, b, x2S0.
1074 . D. Ribatti et al.
Fig. 4. CAM's immunoperoxidase staining using a mAB to the fibronectin, in an experimental case in which CAM's continuity
solution had been completely filled by the scar tissue. Note, on the left, in the mesenchyme (M) underlying the chorionic
epithelium (C), a immunoreactivity (arrowhead) condensed in some areas of the epithelium-connective interface. Original
magnification: x150.
genic factors 4, 5, 23. In contrast, when neovascularization is suppressed or delayed, WH itself is retarded 2 ,25.
Simultaneously with the increase in vascular density,
we also demonstrated in the CAM mesenchyme about
three times as many fibroblasts as in the contiguous
control regions. Hence, fibroblasts participate in large
numbers in the formation of the granulation tissue during the WH process. There is possibly a close relationship between the increase in vascular density and the
number of fibroblasts. Recently, a transformation of
fibroblasts into endothelial cells during angiogenesis
has been demonstrated 18. Moreover, co-cultures of endothelial cells and fibroblasts in collagen gels elicited
an angiogenic response as demonstrated by the reorganization of endothelial cells into a capillary-like network with lumens 2o •
An adequate stimulus for fibroblasts multiplication
could be represented by the basic fibroblast growth factor (bFGF) released by the macrophages 22 ,30. In our
work, these cells represent the main cell population
in the inflammatory infiltrate, which in turn participates in the composition of the granulation tissue
and shows an interstitial arrangement instead of the
intravascular distribution observed under normal conditions. Macrophages are another source of angiogenic
activity in a WHo They migrate in the depth of a wound
where oxygen tension is very low l6 , a condition stimulating these cells to release angiogenic factors, including bFGF, transforming growth factors (TGF-a and -Pl,
granulocyte-macrophage colony stimulating factor
(GM-CSF), vascular endothelial growth factor
(VEGF), interleukin-8 (IL-8) and tumor necrosis factor alpha (TNF-a)22,30.
lt is of note that WH is delayed when transformation
of monocyte into macrophage is prevented by corticosteroids 13.
Angiogenic factors, such as bFGF, could be mobilized from the CAM extracellular matrix strores 29 by
the proteases released by the endothelial cells.
We also demonstrated a marked positivity for fibronectin in the extracellular matrix underlying the lesion
during the granulation tissue formation process. Cells
forming granulation tissue, such as macrophages, produce fibronectin deposited into the tissue matrix l9 • Proliferating endothelial cells also synthesize and secrete
fibronectin in vivo 3, 21 and increased fibronectin appears in dermal blood vessels exhibiting endothelial
cell proliferation during delayed type hypersensitivity
reactions 6 • In addition, fibronectin derived from extravased plasma increases markedlr after injury and supports endothelial cell migration . In turn, fibronectinrich matrix constitutes an adequate substrate for fibro-
The CAM as an in vivo Wound Healing Model· 1075
blasts recruitment, adhesion and differentiation, while
the mobilization of fibroblasts is more likely related to
tenascin expression 24 •
In conclusion, this preliminary study shows that
CAM, usually employed as an in vivo model to study
angiogenesis 17 can also be utilized as an in vivo WH
model. In this respect, then, CAM could allow a very
detailed study of angiogenetic kinetics, of the relationship between the latter and the fibroblast intervention,
of the different stages in the formation of the inflammatory infiltrate and how it becomes enriched in
macrophages, and of the formation of the extracellular
matrix. Concerning this latter point, CAM may represent a model for investigating matrix healing, i. e. for
neomodeling of collagenic (basement membrane and
interstitial) and elastic networks which are probably
involved in CAM permeability, strength and plasticity.
These studies could be extended to cover the functional activity of the cell populations participating in
WH for a better understanding of its physiopathology.
Finally, this experimental model may be applied to
study neoangiogenesis and competent fibroblasts recruitment modulation in some pathological conditions, i. e. abnormal scarring, tumor associated
angiogenesis and fibrosis.
Material and Methods
The experimental procedures for this study followed the
published Guiding Principles in the Care and Use of Animals
approved by the Council of the American Physiological Society. Fifty fertilized White Leghorn chick eggs, staged according to Hamburger and Hamilton (HH) 15 were incubated from
the start of their embryogenesis in an incubator under of 60%
relative humidity at 37 DC. At the HH stage 13, a square window was opened in the egg shell after removing 2-3 ml of
albumen in order to detach the developing CAM from the
shell. The opening was closed with a glass tape and the incubation continued.
In vivo CAM Wound Assay
At the HH stage 35, a full thickness excision of a small
(1 mm 2 ) area of the CAM devoid of large blood vessels
was performed with the aid of a micro knife under a Zeiss
SR stereo microscope (Zeiss, Oberkochen, Germany)
equipped with the Camera System MC 63. Following removal, the embryos were returned to the incubator. All procedures were performed under sterile conditions.
In vivo Stereomicroscopy
The stereomicroscope was used to evaluate the CAM vasoproliferative response for up to 96-120 hours after excision. After 96-120 hours (HH stages 39-40) the embryos
and their membranes were fixed in vivo in Bouin's fluid. After
fixation, the chorioallantoic sacs were isolated from the embryos and dissected in strips, then embedded in paraffin, and
serially sectioned at 7 Jlm for microscopy.
Assessment of Angiogenic Response
Toluidine blue stained sections were used. The angiogenic
response of the area close to the excision was evaluated and
compared to the angiogenesis of a distant area in the developing CAM, as the normal control. A slightly-modified planimetric point count method 10, 28 was used to evaluate the
number of small vessels, i. e. tubes with a single layer of endothelial cells and a diameter of 3-10 Jlm. Briefly, 4 to 6 x
160 fields of every 3rd section (covering almost the whole
section) within 30 serial sections from each specimen were
analyzed with a 484-intersection point square mesh inserted
in the eyepiece of a Leitz Dialux 20 photomicroscope (Leitz,
Wetzlar, Germany). The number of vessels was calculated as
the number of mesh intersection points which were occupied
by the transversally cut vessels, expressed again as a percentage of the total number of intersection points.
Fibronectin deposition into the wounds was assessed by
using a three-step avidin-biotin-immunoperoxidase as previously described 31 • Briefly, deparaffinized and rehydrated
sections exhausted in their endogenous peroxidase with
7.5% H 20 2 were sequentially incubated with: i. an anti-fibronectin murine monoclonal antibody (F3648, Sigma Chemical
Co., St. Louis, Mo); ii. a biotin-labeled horse anti-mouse Ig
(Vector Inc., Burlingame, CAl, and iii. an avidin-horseradish-peroxidase conjugate (Vector Inc.), red-stained with a
3-amino-9 ethylcarbazole (Sigma) solution, counterstained
with Gill's hematoxylin n. 2 (Polyscience Inc., Washington,
PAl and mounted in buffered glycerin.
Naphthol-AS-D Chloroacetate Esterase Activity
After deparaffinization and rehydration, adjacent serial
sections were stained for naphthol-AS-D-chloroacetate activity 26 to highlight macrophages.
Quantitation of Fibroblasts
Fibroblasts were counted in the CAM mesenchyme on 4 to
6 randomly selected x 160 fields of every 3rd section within
30 serial sections from each specimen. The 484-point mesh at
x 160 was used, and the number of fibroblasts was calculated
as the number of intersection points occupied in the area.
Statistical Analysis
Means ± 1 standard deviation (SD) were determined for all
variables. The statistical significance of the differences between mean values of the intersection points was determined
by the Student t-test for unpaired data.
This study was supported in part by grants from the National Research Council, Rome (Progetto Coordinato n.
95.02983.CT14), the Associazione Italiana per la Ricerca
sui Cancro (A.I.R.C.), Milan, and the Ministero dell'Universita e della Ricerca Scientifica (60%), Rome, Italy.
1076 . D. Ribatti et al.
1 Banda WJ, Knighton DR, Hunt TK, Werb Z (1982)
Isolation of a non mitogenic angiogenic factor from wound
fluid. Proc Nat! Acad Sci USA 79: 7773-7777
2 Broadley KN, Aquino AM, Woodward SC, BuckleySturrock A, Sato Y, Rifkin DB, Davidson JM (1989) Monospecific antibodies implicate basic fibroblast growth factor in
normal wound repair. Lab Invest 61: 571-575
3 Broadwell CR, Gospodarowicz D, Nicolson GL (1978)
Identification, localization and role of fibronectin in cultured
endothelial cells. Proc Nat! Sci USA 75: 3273-3277
4 Buckley A, Davidson JM, Kamerath CD, Wolt TB,
Woodward SC (1985) Sustained release of epidermal growth
factor accelerates wound repair. Proc Nat! Acad Sci USA 82:
5 Buntrock P, Jentzsch KD, Heder G (1982) Stimulation
of wound healing using brain extract with fibroblast growth
factor activity. Exp Pathol 21: 46-53
6 Clark RAF, Dvorak HF, Colvin RB(1981) Fibronectin
in delayed-type hypersensitivity skin reactions: associations
with vessel permeability and endothelial cell activation.J Immunol16: 787-793
7 Clark RA, Della Pelle P, Mansesu E, Lanigan JM, Dvorak HB, Colvin RB (1982) Blood vessel fibronectin increases in
conjunction with endothelial cell proliferation and capillary
ingrowth during wound healing. J Invest Dermatol 79:
8 Cohn IK, Diegelman RF, Lindblad WJ (Eds) (1992)
Wound Healing: Biochemical and Clinical Aspects W.B. Saunders, Philadelphia-London
9 Edwards LC, Pernokas LN, Dunphy JE(1957) The use
of a plastic sponge to sample regeneration in healing wounds.
Sur§ Gynecol Obstet 105: 303-309
1 Elias H, Hyde DM(1983) Stereological measurements
of isotropic structures. In: Elias H, Hyde DM (Eds) A Guide
to Practical Stereology, pp 25 -44. Karger, Basel
11 Folkman J, Brem H(1992) Angiogenesis and inflammation. In: Gallin JI, Goldstein 1M, Snyderman R(Eds) Inflammation, Basic Principles and Clinical Correlates, pp 821839. Raven Press, New York
12 Folkman J, Shing Y (1992) Angiogenesis. J Bioi Chern
267: 10931-10934
13 Gabbiani G, Montandon D (1977) Reparative processes
in mammalian wound healing: the role of contractile phenomena. Int Rev Cytol48: 187-219
14 Grinnel F, Billingham RE, Burgess L(1981) Distribution
of fibronectin during wound healing in vivo. J Invest Dermatol 76: 181-189
15 Hamburger V, Hamilton HL (1951) A series of normal
stages in development of the chick embryo. J Morphol 88:
16 Knighton DR, Hunt TA, Schenenstuhl H, Halliday BJ,
Werb Z, Banda MJ (1983) Oxygen tension regulates the ex-
pression of angiogenesis factor by macrophages. Science 221:
17 Knighton DR, Fiegel VD, Philipps GD (1991)The assays
of angiogenesis. In: Clinical and Experimental Approaches to
Dermal and Epidermal Repair: Normal and Chronic Wounds,
pp 291-294. Wiley Liss Inc, New York
18 Kon K, Fujiwara T(1994) Transformation of fibroblasts
into endothelial cells during angiogenesis. Cell Tissue Res
278: 625 -628
19 Kurkinen M, Vaheri A, Roberts PJ, Stenman S (1980)
Sequential appearance of fibronectin and collagen in experimental granulation tissue. Lab Invest 43: 47-51
20 Kuzuya M, Kinsella JL(1994) Induction of endothelial
cell differentiation in vitro by fibroblasts-derived soluble factors. Exp Cell Res 215: 310-318
21 Jaffe EA, Mosher DF (1978) Synthesis of fibronectin by
cultured human endothelial cells. J Exp Med 147: 17791791
22 Leibovich SJ, Wiseman DM (1988) Macrophages,
wound repair and angiogenesis. In: Growth Factors and
Other Aspects of Wound Healing: Biological and Clinical Implications, pp 131-145. Alan R Liss Inc, New York
23 Linch SE, Colvin RB, Antoniades HN (1989) Growth
factors in wound healing: single and synergistic effects on partial thickness porcine wounds. J Clin Invest 84: 640-646
24 Mackie EJ, Halfter W, Liverani D (1988) Induction of
tenascin in healing wounds J Cell Bioi 107: 2757-2767
25 Mc Grath MH, Emery JMI (1985) The effect of inhibition of angiogenesis in granulation tissue on wound healing
and the fibroblast. Ann Plast Surg 15: 106-119
26 Moloney WC, Mc Pherson K, Fliegelman L (1960) Esterase activity in leucocytes demonstrated by the use of
naphthol AS-D-chloroacetate substrate. J Histochem Cytochern 8: 200-207
27 Murphy-Ulrich JE, Mosher DF(1986) Fibronectin and
disease processes. In: Vitto J, Perejda AJ(Eds) Connective Tissue Disease: Molecular Pathology of the Extracellular Matrix,
pp. 455-473. Marcel Dekker Inc, New York
28 Ribatti D, Vacca A, Bertossi M, de Benedictis G, Roncali
L, Dammacco F (1990) Angiogenesis induced by B-cell nonHodgkin's lymphomas. Lack of correlation with tumor malignancy and immunologic phenotype. Anticancer Res 10: 401406
29 Ribatti D, Urbinati C, Nico B, Rusnati M, Roncali L,
Presta M (1995) Endogenous basic fibroblast growth factor
is implicated in the vascularization of the chick embryo
chorioallantoic membrane. Dev Bioi 170: 39-49
30 Sunderkotter C, Steinbrink K, Goebeler M, Bhardwai R,
Sorg C (1994) Macrophages and angiogenesis. J Leukoc Bioi
55: 410-422
31 Vacca A, Ribatti D, Roncali L, Ranieri G, Serio G, Silvestris F, Dammacco F(1994) Bone marrow angiogenesis and
progression in multiple myeloma. Brit J Haematol 87: 503508
Received August 11, 1995. Accepted in revised form April 19, 1996
Key words: Chorioallantoic membrane - Chick embryo - Wound healing.;.. Angiogenesis
Ribatti, MI?, ~n~titute of Human Anatomy, Histology and Embryology, University of Bari Medical School, Piazza
GUlllO Cesare, 11, Pohchmco, 1-70124 Bari, Italy, Telephone: 0039.80.5473569, Telefax. 0039.80.5478309