[CANCER RESEARCH 30, 974-979,
April 19701
The Role of the Dermis in the Induction
Papilloma Virus'
of Neoplasia by Shope
CharlesBreedisand John W. Kreider2
Department
of Pathology,
School of Medicine,
University of Pennsylvania.
Philadelphia,
Pennsylvania
Pathology, College of Medicine. Pennsylvania State University, Hershey, Pennsylvania 1 7033 fJ. W. K./
19104
fC. B.J and Department
of
SUMMARY
MATERIALS AND METHODS
Interactions of mesenchyme with epithelial cells are known
to be responsible for the survival, proliferation,
and
differentiation of embryonic and adult epithelia of a variety
of phenotypes.
The importance
of the dermis in the
induction of neoplasia in rabbit epidermis infected with
Shope papilloma virus was studied. Skin fragments were
dissociated into epidermal and dermal components with
trypsin, infected with Shope papilloma virus, and grafted to
nondermal sites. The results were consistent with the view
that little or no dermis is required for the development of
Animals. Giant Checker and New Zealand White rabbits of
both sexes were housed in individual cages and fed Purina
typical
Shope
papillomas
from
Shope
papillomas
virus
infected epidermal cells.
INTRODUCTION
It is now well established that communicative interactions
between phenotypically different cells are responsible for the
elicitation of the capacity for continued proliferation and
differentiation (8). Specifically in the case of skin, it has
been shown that dermis can influence the spreading, orienta
tion, proliferation, and keratinization of epidermis (11). In
view of the importance of the dermis in the normal behavior
of epidermis,
it was of interest
to determine
whether
dermis
is required for the induction of neoplasia in autografts of
rabbit skin infected with SPy.3 For this purpose, chips of
skin taken
from
the ears of rabbits
were split into epidermal
and dermal components with trypsin. Various combinations
of epidermis (both in sheets and cell suspensions) and viable
or freeze-killed dermis were either SPV-infected or nonin
fected and grafted to the panniculus carnosus or the
lumbodorsal fascia. Observations on the papilomatous trans
formation of these grafts suggested that the presence of little
or no dermis is required for the induction of neoplasia in
this system.
chow supplemented
BSS
1 Supported
by
USPHS
Grants
CA-10320-01
and
1-ROl-CA1
1097-
2Recipient of USPHS Research Career Development Award
7-K4-CA38,809-O1A1.
3The abbreviations used are: SPY, Shope papilloma virus; BSS,
Hanks' balanced salt solution.
Received July 29, 1969; accepted September 19, 1969.
974
alone.
After
enzyme
digestion,
the
epidermis
was
mechanically detached from the dermis and either used as an
intact sheet or further separated into single cells by scraping
with a Teflon policeman during a brief soak in a solution of
buffered citrate-NaCl solution (1 part 4% sodium citrate in
0.89% NaCl solution to 4 parts calcium-magnesium-free
phosphate-buffered
NaCl solution).
Cell clumps were
eliminated by drawing the suspension through a sterile tuft
of cotton. Equal numbers of cells were used in the control
and SPV-infected grafts.
Infection with Virus. Fragments of undissociated skin,
epidermal sheets, and epidermal cell suspensions were washed
in several changes of BSS. Grafts to be infected were
incubated for 1 hr at 37° in a UT' dilution of virus
suspension in BSS. Control grafts were incubated in BSS
alone. The epidermal cell suspensions were centrifuged onto
sheets
O1A1.
with fresh kale.
Virus. An approximately
10% extract of cottontail
papilloma tissue (tumors obtained from Earl Johnson, Rago,
Kansas) was prepared. Two pools of virus (stored at —70°)
were used in the experiments. These preparations were
infectious at a dilution of UT4.
Preparation of Autogralts. All grafts were obtained from
dorsal ear skin. The skin was shaved with a razor blade,
washed several times with 70% ethanol, and air dried. After a
light coating of yellow petrolatum USP was applied, split
thickness sheets of skin approximately 0.5 x 5 x 7 mm were
sliced off with a stainless steel safety razor blade.
Enzyme Dissociations. The technique was a modification of
a previously described method (4). Split-thickness grafts
floating epidermis up were incubated for I hr at 37° in
0.25% trypsin (1/250, Difco Laboratories, Detroit, Mich.) in
BSS. Control grafts of undissociated skin were incubated in
of sterile
tea
bag paper
(6 x 6 mm)
to facilitate
subsequent grafting manipulations.
Grafting Procedure. Rabbits were anesthetized i.v. with
Nembutal and were given an i.m. injection of penicillin and
streptomycin. Grafts were placed upon either the panniculus
carnosus or the lumbodorsal fascia. The fragments were
pinned to the subjacent tissue at 2 points with small hooks
bent from fine stainless-steel wire. The grafts were kept from
contact with overlying skin and muscle by suturing a dish
CANCER
RESEARCH
VOL. 30
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1970 American Association for Cancer Research.
Dermis and, Induction
@
shaped plastic cover, about 3.2 cm in diameter and 0.3 cm
deep, over them (Fig. 1). These covers were cut from the
cups of Model FB-24 “Disposotrays―
purchased from Linbro
Chemical Co., New Haven, Conn.
After 3 weeks of growth, the animals were killed with
chloroform. The grafts were removed and fixed in Bouin's
solution, embedded in paraffin, sectioned, and stained with
hematoxylin and eosin.
by SPV
Table 2
Incidence of papilomatous and epidernial cysts
in grafts to the lumbodorsal fascia
Type of
total
cystsDermal
experiment and graftNo.of
papillomatous
graftsNo.of
epidermal
cystsNo.of
innormal
role
interactionEpidermis118Epidermis,dermis88Epidermisfrozen
RESULTS
Grafts on the Panniculus Carnosus. The incidence of
papillomatous
and epidermal cysts developing in grafts
placed on the panniculus carnosus is recorded in Table 1.
Epidermis alone formed keratinizing epidermal cysts in the
single case tested. In the presence of SPV, epidermis
developed
into
papillomatous
cysts in 5 of 6
alone
cases.
Epidermis in the presence of dermis also responded to SPV
infection with papilloma formation. The proximity of dermis
to the outer layer of the panniculus carnosus raised the
possibility that small numbers of dermal cells adherent to
the panniculus might participate in the SPV-epidermal inter
actions. For this purpose, an additional series of grafts were
placed on the lumbodorsal fascia. Further, a plastic cover
prevented
..
of Neoplasia
any part
of the graft
from
coming
into
contact
with the panniculus (Fig. 1).
dermis53Dermal
inneoplastic
role
interactionEpidermis,
SPV21154Epidermis,dermis,SPY15140Epidermis,frozen
dermis,SPY541DermalsusceptibilityDermis11Dermis,
SPY11Frozen
dermis,SPY410
Table 1
Incidence of papillornatous and epidermal cysts in grafts of
the panniculus
of total
Type of
of papilloma
graftsNo.
cystsEpidermis11Epidermis,
graftNo.
fragments
carnosus
of epidermal
cystsNo.
SPV65@-1Epidermis,dermis,
SPY110
Grafts
on
papillomatous
the
Lumbodorsal
and
epidermal
Fascia.
cysts
The
incidence
developing
in
of
grafts
placed on the lumbodorsal fascia is recorded in Table 2.
In the first group of experiments, the importance of dermis
in
the
development
of
keratinizing
epidermal
cysts
was
studied. No significant difference was noted in the ability of
epidermal cells to form epidermal cysts, whether alone or in
combination with viable or freeze-killed dermis (Figs. 2 and
4).
In the second group of experiments, analogous results were
obtained. SPV infection at the time of grafting resulted in a
high incidence of papillomatous transformation.
It made
little difference in tumor incidence whether the epidermal
cells were alone or in combination with dermis or freeze
killed dermis. However, tumors that developed in recom
binants
with
viable
dermis
were
larger
than
tumors
developing from epidermis alone or in combination with
freeze-killed dermis (Figs. 3, 5, 6, 7).
The third group of experiments were attempts to control
some phases of the previous studies. The observations that
epidermal and papillomatous cysts can form from dermal
APRIL
alone indicates
the presence
of viable epidermis
in
the hair follicles remaining in the dermis. The development
of a papilloma in 1 of 4 such dermal grafts which had been
frozen and thawed 5 times clearly demonstrates survival of
epidermal cells despite this drastic treatment. Freezing, there
fore, cannot be expected to be totally effective in destroying
dermal mesenchymal cells.
To determine whether a continuing but transitory effect of
the dermis on the “pure―
epidermal grafts was responsible
for their ability to form papillomas when SPV-infected,
we
performed
the following experiment.
In 11 rabbits,
epidermal cell suspensions were grafted to the lumbodorsal
fascia and permitted to develop in situ for 3 weeks. At the
end of this time, the epidermal grafts were surgically
exposed and infected with SPV by needle inoculation. The
incisions were closed and the grafts were permitted to grow
for an additional 3 weeks. At the conclusion of this period,
the animals were killed and the grafts were sectioned. Four
of the 11 grafts contained no surviving epidermis. Two of
the 1 1 grafts were typical
were very hyperplastic,
pletely
fulfilling
papillomas.
The S remaining
grafts
resembling papillomas but i@t com
the inflexible
criteria
used for this diagnosis.
The fact that 2 typical papillomas developed after SPV
infection of long-established epidermal grafts indicates that a
transitory, short-term effect of dermis on the epidermal cells
does not exist.
Susceptibility of Epidermal Sheets and Cell Suspensions to
Spy. The epidermal component in these experiments was
grafted both as an intact sheet and as a clump-free cell
suspension (5 to 15 million cells/graft). Approximately equal
1970
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1970 American Association for Cancer Research.
975
Charles Breedis and John W. Kreider
numbers of both types of grafts were used in all phases of
these studies. The development of both epidermal and
papillomatous cysts did not depend upon the state of the
epidermal graft. Sheets and cell suspensions were equivalent
in their. behavior. There was no reformation of hair follicles,
rete pegs, or sebaceous glands in any epidermal grafts
without dermis in heterotopic sites.
for the
In this study, transplants of epidermis infected with SPV
almost invariably became papillomatous.
There was no
dependence of this phenomenon on the physical presence of
dermis or on transitory dermal influences. The papillomas
which developed in transplants of epidermis alone were in
every way representative of the typical morphology of this
tumor and identical with the controls, with the exception
that tumors resulting from epidermal-dermal recombinants
were sometimes larger than the tumors which developed
from epidermis alone. This result may be explained by the
presence of the increment of epidermal cells within hair
follicles
follicles
of the dermal fragment.
That
alone can generate papillomas
these residual hair
was shown in the
results of Table 2.
It is possible that the “pure―
epidermal grafts used in this
study actually contained some dermal fibroblasts. However,
in our experience and that of Billingham, microscopic
sections through the epidermal sheets have never shown
adherent fibroblasts. Therefore, the dermal contaminant, if it
does exist, must be a very small percentage of the cells
grafted.
The results of the present study have direct relevance to
some basic questions in both the biology of skin differentia
tion and related problems in neoplasia. It is well known that
the differentiation of epithelium is the result of its inter
actions with mesenchyme. The degree of specificity of this
interaction is quite variable. For example, only salivary gland
mesenchyme can initiate morphogenesis of salivary gland
epithelium (7), but in the thymus (3) and pancreas (12)
nonspecific
mesenchymal
components
will
suffice.
For
epidermis, it has been shown that isolated rat epidermal cells
placed beneath the renal capsule or in skeletal muscle will
form nests of keratinizing epidermis. In the latter site, there
was even some formation of dermal papillae and hair follicles
(6). In the mouse, sheets of pure epidermis placed on the
lumbodorsal fascia form epidermal cysts ( 1). Hair follicles
and sebaceous glands are not regenerated. Clearly, the results
of both the present report and the previous studies support
the conclusion that mesenchyme elicited at a variety of
nondermal sites can interact with epidermal cells to permit
survival,
proliferation,
limited
formation
of skin appendages
and even neoplastic transformation.
In contrast, another study indicated that isolated normal
epidermis and the epithelium of basal cell carcinomas of
humans
would
not survive in vivo in the absence
of contact
with dermis (10). Other investigators (2, 5, 9) have reported
failure of tumors to develop in transplants of isolated mouse
epidermis which was previously treated with methylchol
anthrene. These latter experiments are difficult to interpret
976
technique
in this system
interferes
of tumors in carcinogen-treated
full
thicknessskin.
There has been only one other study on the influence of
epitheliomesenchymal
interactions on tumor production by
an oncogenic virus. It was concluded that polyoma virus
could not induce the neoplastic transformation of mouse
salivary
DISCUSSION
transplantation
with the development
gland
epithelium
in the
absence
of salivary
gland
mesenchyme (7). The apparent contradiction between this
result and the one reported here can be resolved when the
marked specificity of the normal epithelial-mesenchymal
interaction in the salivary gland is compared to the lack of
apparent specificity in skin. In the case of salivary gland,
both normal and neoplastic epithelium remains dependent
upon histotypic
mesenchyme for survival and growth.
However, in the case of skin, a variety of mesenchymal types
will support the survival and proliferation of both normal
and SPY-transformed epidermal cells.
The results of the current experiments indicate that the
conditions which are sufficient for implanted epidermal cells
to differentiate and form a keratinizing squamous epithelium
without epidermal appendages are also sufficient for their
transformation
and growth as papilloma cells. No more
complex or specific environment is required.
REFERENCES
1. Ashley, C. A. Epithelial and Stromal Roles in Carcinogenesis. I.
Histological Changes and Ultimate Fate of Internally Trans
planted Skin and Component Parts Separated by Tryptic
Digestion. Arch. Pathol., 72: 86—97,1961.
2. Ashley, C. A. Epithelial and Stromal Roles in Carcinogenesis. II.
Tumor Development in Internal Transplants of Skin and
Epithelium Treated with 3-Methylcholanthrene. J. Nat!. Cancer
Inst.,
26:1445—1459,
1961.
3. Auerbach, R. Morphogenetic Interactions in the Development of
the Mouse Thymus Gland., Develop. Biol., 2: 271—284,1960.
4. Billingham, R. E., and Medawar, P. B. The Technique of Free
Skin Grafting in Mammals. J. Exptl. Biol., 28: 385—402, 1951.
5. Billingham, R. E., On, J. W., and Woodhouse, D. L. Transplanta
tion of Skin Components during Chemical Carcinogenesis with
20-Methylcholanthrene. Brit. J. Cancer, 5: 417—432, 1951.
6. Bilingham, R. E., and Silvers, W. K. Studies on the Conservation
of Epidermal Specificities of Skin and Certain Mucosas in Adult
Mammals.J. Exptl. Med., 125: 429—447,1967.
7. Dawe, C. J., Morgan, W. D., and Slatick, M. S. Influence of
Epitheliomesenchymal
Interactions on Tumor Induction
Polyoma Yirus. lntern. J. Cancer, 1: 419—450, 1966.
By
8. Holfreter, J. Mesenchyme and Epithelia in Inductive and
Morphogenetic Processes. In: R. Fleischmajer and R. E. Billing
ham (eds.), Epithelial-Mesenchymal Interactions, pp. 1—30.
Baltimore: The Williams& WilkinsCo., 1968.
9. Marchant, J., and Orr, J. W. A Further Investigation of the Role
of Skin Components in Chemical Carcinogenesis. Brit. J. Cancer,
9: 128—136,1955.
10. Yan Scott, E. J., and Reinertson, R. P. The Modulating Influence
of Stromal Evnironment on Epithelial Cells Studied in Human
Autotransplants. J. Invest. Dermatol., 36: 109—131, 1961.
11. Wessels, N. K. Differentiation of Epidermis and Epidermal
Derivatives. New EngI. J. Med., 277: 21—33,1967.
12. Wessels, N. K., and Cohen, J. H. Early Pancreas Oncogenesis:
Morphogenesis, Tissue Interactions, and Mass Effects. Develop.
Biol., 15: 237—270,1967.
CANCER RESEARCH VOL. 30
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1970 American Association for Cancer Research.
Dermis and Induction
of Neoplasia
by SPV
Fig. 1. Schematic diagram illustrating anatomic relationships in placement of grafts on the lumbodorsal fascia.
Fig. 2. Portion of the wall of an epidermal cyst developing from a suspension of epidermal cells planted on the lumbodorsal fascia after
centrifugation onto a fragment of tea bag paper. Note the abundant proliferation of mesenchymal cells and the reformation of germinal, spinous,
keratohyalin, and keratinized layers. 5, strands of tea bag paper. H & E, X 150.
Fig. 3. Portion of the wall of a papilomatous cyst developing from an epidermal cell suspension, treated except for SPY infection, in the same
manner as the graft in Fig. 2. Materials in both Fig. 2 and Fig. 3 are from the same rabbit. A typical Shope papilloma is present. The epithelium is
hyperplastic and hyperkeratotic. There is a marked acanthosis with an exaggerated intrusion of dermal papilae into the epithelium. H & E, X 150.
Fig. 4. Epidermal cyst developing from epidermal cell suspension grafted to the lumbodorsal fascia. Identical with Fig. 2, except from a different
rabbit. H & E, X 150.
Fig. 5. Papillomatous cyst developing from an SPV-infected epidermal cell suspension grafter to the lumbodorsal fascia. Same rabbit as Fig. 4.
H&E,X 150.
Fig. 6. Papifiomatous cyst developing from an SPY-infected epidermal sheet grafter to the lumbodorsal fascia. F, lumbodorsal fascia.
H & E, X 14.
Fig. 7. Same rabbit as Fig. 6. Papillomatous cyst which developed from an SPY-infected epidermal sheet combined with dermis and grafted to the
lumbodorsal fascia. Note that the papilloma development is more extensive than in Fig. 6. F, lumbodorsal fascia. H & E, X 14.
APRIL
1970
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1970 American Association for Cancer Research.
977
@
@
@@-@--@. @J
@:
Charles Breedis and John
@
@@:-
@;@-
@!
W. Kreider
-
@_-@1
_@Th@
-
-@-
muscle
-ö
@
@
r@
@
@—@‘@:
.
‘.
,. . , .— . .
vi':
. -
. ,.,. : - @‘
‘
‘I@
r
-
@
.
@
::---@-@
@
..—
@
-@
—_:
::
—
—
e
@
@:,-
@
—@-----:.
.@—.
- -@ -.-..@- @.
—..@
——
*
@
.
—
_@
—
.r
-
@.@-&. @--@—.;@:--..
—@
——._@
—
—
--.:
———..
t
—
•:
_-
;-- -@:
—
‘5
3
—.@
-@.
@1@@s,;:?@: j
—
.-
978
CANCER
RESEARCH
VOL.30
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1970 American Association for Cancer Research.
@
@
. ‘:
\@
,@
.
@,
:‘
@p_ .
@@@7i;
‘ : ‘
Dermis and Induction
@
F
@
I @/\
-@.
@
.-.@:
@* @L ,,@
:
of Neoplasia
by SPV
(@
..
@--@
@.i
@:@:
,
I
I
@
,-..-@
- @.
;..
@
1
APRIL
:@;!Ii@0
F
—
@. ..
-@
,-.)@
.
-@-
1970
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1970 American Association for Cancer Research.
979
The Role of the Dermis in the Induction of Neoplasia by Shope
Papilloma Virus
Charles Breedis and John W. Kreider
Cancer Res 1970;30:974-979.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/30/4/974
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1970 American Association for Cancer Research.