[CANCER RESEARCH 42, 1826-1837. May 1982]
0008-5472/82/0042-OOOOS02.00
Transepithelial Invasion and Intramesenchymal Infiltration of the Chick
Embryo Chorioallantois by Tumor Cell Lines1
Peter B. Armstrong,2 James P. Quigley,3 and Eric Sidebottom
Department of Zoology, University of California, Davis, California 95616 ¡P.B A.]: Department of Microbiology and Immunology, State University of New York,
Downstate Medical Center. Brooklyn. New York 11203 ¡J. P. Q.¡; and Sir William Dunn School of Pathology, Oxford University, Oxford, OX ; 3RE,
United Kingdom ¡E.S.¡
ABSTRACT
The ability of cultured tumor cell lines to invade across
epithelia was studied by placing 10" to 106 dispersed cells on
the chorionic epithelium (CE) of the chorioallantoic membrane
of the 10-day chick embryo. Tumor lines included Walker
carcinosarcoma, F87 cl 6T2 and B16-BL6 melanomas, and
KiSV-NIH, 3B77SC4, and HT 1080 sarcomas. The CE is a
bilayer of cells with a superficial periderm overlying a basal
layer. Invasion across an intact CE was very weak (limited to
the formation of "microtumors" by a small fraction of the
inoculated cells in 5 to 50% of the embryos) but was massive
(most or all of the inoculated cells invaded in over 99% of the
embryos) if the chorioallantoic membrane was traumatized in
a fashion which disrupted the periderm but left the basal layer
intact. Normal fibroblasts also invaded across the traumatized
CE. The histological picture of invasion suggests that cells
inoculated on the traumatized CE induced large-scale active
retraction of the basal layer, resulting in the formation of large
gaps in its continuity. Migration into the subjacent mesoderm
occurred through these gaps. The nodules formed by both
tumorigenic and normal cells became extensively vascularized
within 3 days.
INTRODUCTION
Intercellular invasion can be defined as the penetration of
cells of one type into the interiors of other tissues (1, 4-6). The
invasive activity of the cells of malignant tumors is one of the
principal means of tumor dissemination. Invasion results in the
local spread of tumor cells into surrounding normal tissues and
also plays an important part in the process of metastasis,
whereby tumor cells colonize distant organs (12, 22, 65, 70,
71 ). Contrasting with the invasive behavior of tumor cells is that
of cells of normal coherent tissues which are noninvasive in
the sense that they do not venture into adjacent tissues but
instead remain confined to the parent tissue for the life of the
organism (cf. Refs. 3-6).
The CAM" of the 9- to 11-day chick embryo has been used
as a convenient substrate for the study of malignant invasion.
The CAM, an extraembryonic organ, is organized as an exten
sively vascularized mesodermal layer sandwiched between
2 epithelial tissues, an external CE and an internal AE (Fig. 1).
For experimental purposes, the CAM is exposed by cutting a
1 Supported by funds of the Sir William Dunn School of Pathology and National
Science Foundation Grants PCM 77-18047 and PCM 80-24181.
2 To whom requests for reprints should be addressed.
3 Supported by a Fogarty Fellowship.
4 The abbreviations are: CAM. chorioallantoic membrane; CE. chorionic epi
thelium; AE, allantóle epithelium; PBS, phosphate-buffered saline [137 rriM NaCI.
2.7 mM KCI. 8.1 rriM Na2HPO... 1.5 mm KHjPO«(pH 7.3)].
Received October 26. 1981; accepted February 10, 1982.
window through the egg shell and underlying acellular shell
membrane. The CE lies immediately below the shell membrane.
Dispersed tumor cells obtained from cell culture (38, 58, 59,
66, 73) or expiants of solid tumors obtained from biopsies of
tumors growing in vivo (37) are placed on the surface of the
exposed CE, and invasion is considered to have occurred when
tumor cells are observed in the mesodermal layer beneath the
CE.
The present study is an investigation of the factors involved
in the ability of cells to traverse the barrier presented by the
CE and to infiltrate the mesoderm. At 9 to 11 days of embryonic
life, the CE is a continuous layer 2 cells thick with a superficial
layer of flat peridermal cells and a deep-lying basal cell layer
which rests on a thin basement membrane directly above the
mesoderm (Fig. 2). Our observations are consistent with the
notion that, in the absence of damage to the periderm, the
CAM is a formidable barrier to tumor invasion, but following
damage to the periderm, even in the absence of obvious
compromise of the underlying basal cell layer, both malignant
and some nonmalignant cells are able to gain access to the
underlying mesoderm. We suggest that, in this case, penetra
tion of the CE involves its retraction from areas of the underlying
mesoderm, permitting direct contact between grafted cells and
CAM mesoderm.
MATERIALS AND METHODS
Cells and Embryos. Normal fibroblasts were prepared by culturing
cell suspensions from trypsin-dispersed whole 15-day mouse embryos
and decapitated eviscerated 10-day chick embryos. Tumor lines used
were LLC-WRC 256 Walker carcinosarcoma (29), B16-BL6, a murine
melanoma line selected for its aggressive invasiveness (64), F87 cl
6T2, a metastatic murine melanoma,5 F87 cl 4T1, a nonmetastatic
melanoma,5 a nonproducer Kirsten sarcoma virus-transformed deriva
tive of the NIH-3T3 cell (KiSV-NIH; Ref. 73), the human sarcoma HT
1080 (68), and a nonproducer derivative of the normal rat kidney cell
transformed by the B77 avian sarcoma virus (3B77SC4; Ref. 45). Cells
were passaged and cultured using standard procedures. All lines were
screened for Mycop/asma contamination (11 ). Cultures were discarded
after 6 to 12 passages, and fresh cultures were initiated from frozen
stocks to reduce problems of genetic drift in culture. Embryonated
chick eggs were obtained from the following suppliers: A. E. Binning,
Great Park Farm, Abingdon, United Kingdom; Orchards Farm, Great
Missingdon, United Kingdom; Wickham Laboratories Ltd., Wickham,
Hants, United Kingdom; and Spafas, Inc., Norwich, Conn.
Preparation of the CAM. The CAM was exposed on the eighth day
of incubation using the artificial air sac method of Burnet and Barnard
(9). The embryos were candled, the shells were wiped with 70%
ethanol, the air sac was punctured, and a 1- x 1.5-cm window was
carefully scored in the shell over the CAM using a 0.3-mm-thick x 22-
5 E. Sidebottom and S. Clark. The contribution of cell fusion studies to the
analysis of metastasis, manuscript in preparation.
1826 CANCER RESEARCH VOL. 42
Research.
on September 12, 2021. © 1982 American Association for Cancer cancerres.aacrjournals.org Downloaded from