[CANCER RESEARCH 44, 2991-2999,
July 1984]
Coexpression of Simple Epithelial Keratins and Vimentin by Human
Mesothelium and Mesothelioma in Vivo and in Culture
Paul J. LaRocca1 and James G. Rheinwald2
Division of Cell Growth and Regulation, Dana-Farber Cancer Institute and Department of Physiology and Biophysics, Harvard Medical School,
Boston, Massachusetts 02115
ABSTRACT
We have determined the intermediate filament proteins present
in normal and malignant mesothelium in vivo. Pure sheets of
normal lung mesothelium were obtained by transfer to agarcoated slides or by gentle scraping and cytocentrifugation. Cytoplasmic filament networks in the mesothelium were labeled via
indirect immunofluorescence both by anti-Mr 40,000 keratin and
anti-vimentin antisera. Two-dimensional gel electrophoresis of
the Triton:high-salt-insoluble proteins of normal lung mesothel
ium disclosed the presence of vimentin and all but the largest
(M, 55,000) of the four simple epithelial keratins synthesized by
mesothelial cells in culture. Samples of three peritoneal and three
pleural mesotheliomas were found to contain either all four simple
epithelial keratins or all but the M, 55,000 keratin. Notably, none
of the keratins characteristic of stratified and many glandular
epithelia and their malignant forms was present in these meso-,
theliomas. Two mesothelioma samples from which the tumor
cells could be obtained free of other cell types were found to
contain vimentin as well as simple epithelial keratins and to
synthesize these same proteins during short-term culture. None
of the mesotheliomas placed in culture grew progressively in
medium optimal for the growth of normal mesothelial cells. These
data demonstrate that malignant mesothelial cells preserve the
normal pattern of intermediate filament protein synthesis, includ
ing coexpression of simple epithelial keratins and vimentin, and
suggest the use of this characteristic as an aid in the identification
of cells of mesothelial origin.
INTRODUCTION
wanted to determine whether vimentin is present in the cells of
the intact mesothelium also, or whether it is only acquired after
cell detachment or cultivation, as has been suggested by others
(12, 21, 32, 39). The set of keratin proteins synthesized by
normal human peritoneal mesothelial cells in culture is different
from that of stratified squamous epithelial and many glandular
epithelial cell types (22, 27, 38, 43). We wished to determine
whether mesotheliomas express any keratins different from
those of normal mesothelium, inasmuch as squamous cell car
cinomas often synthesize keratins inappropriate for their cell type
of origin (23, 43, 44). We report here that normal mesothelial
cells in the intact tissue coexpress keratins and vimentin, that
the largest keratin M, 55,000 appears to be induced by conditions
of culture, and that mesotheliomas express vimentin and the
same set of keratins as does normal mesothelium.
MATERIALS
AND
METHODS
Preparation of Mesothelial Tissues. Sheets of normal lung mesothe
lium were obtained free of underlying connective tissue by a modification
of the imprinting method (42). A molten solution of 1% agarose (Sea
Kem) in water was applied to glass slides and dried overnight. The
surface of freshly excised human fetal lung was rinsed in isotonic PBS3
and exposed to the air. Just as the mesothelium began to appear dry,
an agarose-coated slide was pressed against the lung for several sec
onds. When the slide was pulled away, parts of the mesothelial cell layer
adhered to the agar and were peeled off. Slides were examined under
an inverted microscope with phase contrast optics to identify areas
containing large sheets of mesothelium. These were rinsed with PBS to
remove any blood cells. Slides were then either briefly air-dried and
immersed in cold (-20°) methanol to fix for immunofluorescence micros
copy, or the agarose and cell material were scraped into Triton:high-salt
The mesothelium is the simple squamous epithelium that lines
the pleural, pericardial, and peritoneal cavities and covers the
outer surfaces of the organs contained therein (6, 7). Unlike most
other epithelia, the mesothelium is of embryonic mesodermal
origin. Perhaps because of this the malignant form, mesothe
lioma, can adopt either sarcomatous or carcinomatous forms,
often making differential diagnosis difficult (9, 18, 37).
This laboratory has recently determined the cell type-specific
requirements of normal, diploid human mesothelial cells for rapid
growth in long-term, serial culture (4). Mesothelial cells are
extraction buffer and stored frozen for future electrophoretic analysis.
Tissues to be used as electrophoresis samples were collected on slides
that had been coated with electrophoretically purified agarose in order
to reduce the levels of contaminants which would be stained in the
subsequent silver staining procedure. As an alternate method for obtain
ing small sheets and clusters of mesothelial cells for ¡mmunofluores-
different from most other normal epithelial cell types in that they
express in culture high levels of 2 different types of intermediate
filament proteins: keratins and vimentin (42,44). Because normal
mesothelial cells that detach from the mesothelium and float in
ascites fluid also contain both vimentin and keratins (4), we
Surgical specimens of solid peritoneal and pleural mesotheliomas
(kindly provided by Dr. J. M. Corson, Brigham and Women's Hospital)
1 Recipient
of
National
Research
Service
Awards
T32
CA09361
and
IF32AG05303 ¡npartial support of this investigation.
1 Recipient of grants from the National Cancer Institute and the National Institute
on Aging.
Received July 25,1983;
cence, the lung surface was gently scraped with a rubber policeman into
PBS. The cell suspension was then diluted 1:1 with serum and applied
to a glass slide by spinning in a cytocentrifuge. The slides were then air
dried and fixed in cold methanol. For these experiments, lungs were
obtained from 18 to 22-week-old human fetuses or from adult rats.
were collected within 1 h after resection and minced into 1 cu mm pieces.
Some fragments from each tumor were placed in culture, and some were
frozen in Triton:high-salt buffer until extracted for electrophoresis. Me
sothelioma cells were also obtained from a pleural effusion of mesothe
lioma (MS-7). This effusion was found by cytopathological evaluation to
consist of >90%
mesothelioma
cells. The cells were concentrated
3The abbreviation used is: PBS, phosphate-buffered
accepted March 30,1984.
KCI:16 mw HajrtPQt:2muKHJPO*,
by
saline (137 mm NaCI:3 HIM
pH 7.3).
JULY 1984
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1984 American Association for Cancer Research.
2991
P. J. La/tocca and J. G. Rheinwald
centrifugation from 200 ml of effusion. One fourth of the cell pellet was
extracted for electrophoretic analysis, and the rest was used to initiate
cultures.
Frozen ampuls of cells concentrated from the ascites fluid of a patient
with an ovarian surface epithelial neoplasia were thawed and applied to
slides by cytocentrifugation for examination by immunofluorescence. This
was the same ascites fluid from which the normal mesothelial strain LP9 have been derived (4, 43).
Indirect Immunofluorescence
Microscopy.
Anti-M, 40,000 keratin
antiserum, raised in rabbits, is specific for this particular keratin (43, 44).
Anti-stratum corneum keratins antiserum, raised in rabbits, reacts
strongly with epidermal keratins and cross-reacts to various degrees
with all the "mesothelial" or "simple epithelial" keratins (43, 44). An antihuman fibroblast
vimentin
antiserum
was raised in rabbits.
An anti-
vimentin antiserum raised in guinea pigs was kindly provided by Dr.
Werner Franke, German Cancer Research Center, Heidelberg, West
Germany. Mouse monoclonal antibody OC125, which recognizes a sur
face antigen of ovarian carcinoma cells (1), was obtained from Dr. Robert
Bast of this institute. Rhodamine- and fluorescein-conjugated
secondary
antibodies were obtained from Miles and Cappel Laboratories. The
procedures used for indirect immunofluorescence microscopy and pho
tography has been described (4,44).
Cell Labeling, Extraction, and Electrophoresis. Methods were sim
ilar to those previously described (4, 43, 44). Proteins synthesized by
cultured cells were labeled by a 4-hr incubation with [^SJmethionine (50
ßd/ml) (specific activity, ~800 Ci/mmol, MEN Chemicals) in otherwise
methionine-free minimal Eagle's medium (Flow Laboratories) supple
mented with 20% dialyzed fetal calf serum. Labeled cultured cells,
agarose slide-collected mesothelium, and minced tumor tissue were
rinsed with PBS, Dounce homogenized in Triton:high-salt extraction
buffer (20 mw Tris-HCI, pH 7.4:0.6 M KCI:1% Triton X-100:1 ITIMphenylmethylsulfonylfluoride)
at 0°, and then sonicated for 45 sec. Lysates
were then centrifuged at 10,000 x g for 45 min at 4°.The insoluble
pellets (containing the keratins and vimentin) were then dissolved in
O'Farrell 2-dimensional lysis buffer. The proteins were separated in the
first dimension by isoelectric focusing (29) and in the second dimension
by sodium dodecyl sulfate:polyacrylamide gel electrophoresis (10% acrylamide) (20). High-purity agarose (Sea Kern) was used to hold the
isoelectric focusing tube gel in place at the top of the sodium dodecyl
sulfate:polyacrylamide slab gel. The possible presence of basic (isoelec
tric points 7 to 8) keratins was examined by separating the proteins by
nonequilibrium pH gradient electrophoresis for 3.5 hr in the first dimen
sion, according to the method of O'Farrell ef al. (30). Labeled proteins
were detected by scintillation fluor-enhanced (Enhance; MEN Chemicals)
autoradiography.
Protein Detection in Gels by Silver Staining. An ultrasensitive silver
method (25) was used with the following modifications. After fixation in
10% glutaraldehyde, the gel was rinsed for at least 48 hr to more
completely remove glutaraldehyde and consequently reduce the back
ground staining. Ethanolamine (5 ¿J/ml)was added to the citric acid
fixative to prevent the formaldehyde from overdeveloping the silver (10).
Wet gels were then photographed over a light box with Kodak Technical
Pan film (ASA 50), which was developed in a 1:36 dilution of HC-110 for
8 min. Gels were placed in a solution of 0.02% sodium carbonate and
stored in sealed plastic bags for future examination.
Cell Culture Methods. A population of normal, diploid, human pleural
mesothelial cells (strain HPM-2) was cultured from pleural effusion fluid
withdrawn from an adult female. The normal, diploid, human peritoneal
mesothelial strain LP-9 has been described previously (4, 43). LP-9 and
HPM-2 were grown as described (4), in either M199 or M199/
MCDB202:15% fetal calf serum (Sterile Systems):hydrocortisone
(0.4
Mg/ml; Calbiochem):epidermal
growth factor (10 ng/ml; Collaborative
Research). In order to maximize keratin synthesis (Fig. 3, b to d), cells
were grown to confluence during one passage in the absence of epider
mal growth factor.
Cells from a pleural effusion mesothelioma (MS-7) and from a solid
peritoneal mesothelioma (MS-10) which could be dislodged as clusters
2992
of pure tumor cells were placed into culture in the growth medium
described above.
Keratin Nomenclature. Keratins 40KM, 44KM, 52KM and 55KM, the
"mesothelial keratins" described in Wu ef a/. (43), are keratins 19,18, 8,
and 7, respectively, independently identified as "simple epithelial keratins"
by Moll ef a/. (22).
RESULTS
M, 40,000 Keratin Immunofluorescence of Mesothelia. Nor
mal mesothelium was transferred from the outer surface of a
fetal human lung to an agarose-coated slide by an imprinting
technique described by Whitaker et al. (42). Cell shape, nuclear
position, and intermediate filament content was examined by
phase-contrast microscopy (Fig. 1a) and indirect immunofluor
escence with anti-40KM keratin. Favorable areas of the trans
ferred mesothelium were found to contain this keratin in a fibrous
network (Fig. 1b) similar in appearance to that of cultured epithe
lial cells. Heterogeneity of staining presumably resulted from
variability in the amount of cytoskeleton transferred to the slide.
In these preparations, the apical portion of the plasma membrane
adheres to the agarose, and the cells apparently fracture such
that variably amounts of the cytoplasmic contents are removed
to the slide, even though the nucleus (or its outline) can be seen
in each transferred cell (Fig. 1a).
Small clusters to large sheets of mesothelial cells could also
be dislodged by scraping the lung surface with a rubber police
man and depositing the PBS:serum suspension of these cells on
a slide with the aid of a cytocentrifuge (Fig. 1c). Clusters of easily
identified mesothelial cells stained positively for the 40KM keratin
and also revealed details of the cytoplasmic filament network
(Fig. 1d). This was consistent with the results of our preliminary
examination of paraffin sections of mesothelium by immunoperoxidase staining (44). The mesothelium was also stained strongly
by anti-total stratum corneum keratins, an antiserum which
cross-reacts with most of the keratins (43, 44) (data not shown),
as had been reported previously (5, 36).
The M, 40,000 keratin was also detected by immunofluores
cence in secondary cultures of mesotheliomas MS-7 (Fig. 1ft)
and MS-10 (data not shown). The mesothelioma cells had a
morphology in culture (Fig. 1g) distinct from that of normal
mesothelial cells (Fig. 1e). MS-7 and MS-10 could not grow
progressively in the culture medium permissive for long-term
growth of normal pleural and peritoneal mesothelial cells and
ceased dividing in secondary culture. We initiated 5 other cultures
of solid pleural and peritoneal mesotheliomas, but the only cells
that grew had morphologies and growth requirements of either
normal mesothelial cells or of normal connective tissue fibroblasts and were not tumorigenic in nude mice. Apparently the
mesothelioma cells present in those 5 tumor specimens could
not grow at all under the conditions we used.
Detection of Vimentin in Normal Mesothelial Cells by Im
munofluorescence. This laboratory previously had reported that
normal mesothelial cells present among the cells in an ascites
fluid specimen, which had been allowed to attach to a coverslip,
contained both keratin and vimentin (4). We wished to confirm
this coexpression in ascites fluid mesothelial cells that never had
been exposed, even briefly, to culture conditions, and also to
determine whether the cells of intact mesothelium contain vimen
tin. We applied human ascites fluid cells and intact rat lung
mesothelium (dislodged by scraping with a rubber policeman) to
CANCER
RESEARCH
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1984 American Association for Cancer Research.
VOL. 44
Intermediate Filament Proteins of Mesothelium and Mesothelioma
Table 1
Keratin types and vimentin coexpression in normal and malignant mesothelia
Keratins
(40KM/19)
Normal mesothelium8
Pleural
Pleural (HPM-2)
Peritoneal (LP-9)
v6
c
c
(44KH/18)
+°
+
+
+
++/-++
+
(46K>,)
(52IW8)
+/-
(55KM/7)
++
++
Vimentin
+
+
+
Malignant mesothelium
Pleural (MS-2)
Pleural (MS-11)
Pleural (MS-7)
Peritoneal(MS-4)
Peritoneal (MS-5)
Peritoneal(MS-10)
" Data summarized from Fig. 3 and Refs. 4 and 43.
v, in vivo',c, in culture.
0 +, moderate amounts; ++, relatively high amounts; +/-, variable amounts among experiments; -, none
present.
" Data summarized from Figs. 3 and 4.
e Amount in mesotheliomacells indeterminate because of connective tissue present in tumor specimen.
' Relative amounts unknown due to partial proteolysis.
slides by cytocentrifugation and double labeled them with antivimentin and anti-keratin antisera. As shown in Fig. 2, a to c, all
mesothelial cells in rat lung mesothelium coexpressed vimentin
and keratin. Consistent with the previous report from this labo
ratory (4), some of the ascites fluid cells coexpressed vimentin
and keratin (Fig. 2, d to f). Vimentin-positive, keratin-negative
cells in the ascites fluid were assumed to be macrophages (3,
11); these cells attached and spread under culture conditions
but did not divide (4). The keratin-positive, vimentin-negative
cells were presumed to be neoplastic cells of ovarian surface
epithelial origin because cells in the sample which stained with
the ovarian surface-specific monoclonal antibody OC125 (1) were
vimentin negative (Fig. 2, g to /) and keratin positive (data not
shown). Thus, keratin and vimentin coexpression is a normal
characteristic of mesothelial cells and is retained by cells that
become detached from the mesothelium.
Electrophoretic Analyses of Mesothelial and Mesothelioma
Cytoskeletal Proteins. The Triton:high-salt-insoluble protein
fraction extracted from normal and malignant mesothelial sam
ples were analyzed by 2-dimensional gel electrophoresis to
determine precisely the intermediate filament proteins present
(Table 1). Fetal human pleural mesothelium collected by stripping
with an agar-coated slide, precluding contamination with under
lying connective tissue, contained vimentin, keratins 40KM and
44Km, and, in greatest abundance, keratin 52KM (Fig. 3a). Keratin
55KM, synthesized by adult peritoneal mesothelial cells in culture
(4, 43) (see Fig. 3, c and d) was not detectable. Normal pleural
mesothelial cells (HPM-2) synthesized vimentin and all 4 simple
epithelial keratins in culture, negating the possibility that cells of
pleural origin are unable to synthesize keratin 55KM. A rapidly
turning over, basic M, 46,000 keratin which is expressed at
variable levels by mesothelial cells growing in culture (4, 43) (see
Fig. 3cO,also was not detected in the in vivo mesothelium.
A relatively pure cell suspension of pleural mesothelioma (MS-
synthesized these same proteins in secondary culture (Fig. 3, e
and g). MS-10 contained vimentin, keratins 40KM and 44KM, a
small amount of keratin 52KM, and what appeared to be proteolytic digestion products of 52KM (Fig. 3f). Cultured MS-10 cells
synthesized vimentin and all 4 simple epithelial keratins (Fig. 3h).
Thus, malignant mesothelial cells may also increase their expres
sion of 55KM in culture.
Samples of 3 other solid peritoneal mesotheliomas and 4 other
pleural mesotheliomas were extracted and analyzed by 2-dimen
sional gel electrophoresis. Three samples gave no interpretable
spots due to failures of isoelectric focusing or to paucity of
mesothelioma cells in a predominantly fibrous tumor. Four sam
ples could be scored (Fig. 4). Keratins 40KM, 44KM, and 52KM
were present in all samples, with keratin 55KM present in all but
one sample (Fig. 4o). One tumor (Fig. 4d) also contained a
protein migrating in the position of keratin 46KM. Vimentin was
detected in 3 of the samples, but since these contained connec
tive tissue, the vimentin content of the tumor cells themselves
was indeterminate.
Proteolysis of vimentin (14, 28) and of keratin 52KM (35) was
extensive in some tumor extracts (Fig. 3f; Fig. 4, a, b, and d),
precluding quantitative estimation of these proteins when it
occurred. Evaluation of gels made from extracts of small tissue
samples was also complicated because the silver staining
method used to detect very low levels of protein also stained
minor impurities in the buffers and in the agarose used to
immobilize the tube gel during the second dimension of electro
phoresis. These impurities were most common in the M, 56,000
to 70,000 range (eg., cf. Fig. 3 c and d, and see Fig. 3 a and Fig.
4 b, c, and d). Despite the staining background, no proteins
migrating at the position of any known keratin (22,43) other than
the simple epithelial keratins were identified in any of the me
sothelioma extracts.
7) and clusters of cells teased from a solid peritoneal mesothe
lioma (MS-10) were extracted directly and their intermediate
DISCUSSION
filament proteins compared with those synthesized by cells from
these same tumors after 1 to 2 weeks in culture. MS-7 contained
vimentin and keratins 40KM, 44KM, and 52KM in vivo and also
We previously had identified vimentin in the normal mesothelial
cells present in ascites fluid and rising to higher levels in these
cells during serial culture (4). Nevertheless, it was necessary to
JULY 1984
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1984 American Association for Cancer Research.
2993
P. J. LaRocca and J. G. Rheinwald
examine mesothelial cells in situ, since it has been proposed that
vimentin may become expressed as a consequence of loss of
cell-cell contact after detachment from an epithelial sheet (21,
32). It is also known that some epithelial cell types (12, 39) but
not others (4, 34) abnormally synthesize vimentin under certain
conditions of culture.
We have found by indirect immunofluorescence microscopy of
scraped mesothelium and by 2-dimensional gel electrophoretic
analysis of pure, agar slide-transfer preparations of mesothelium
that this tissue contains both keratins and vimentin in vivo.
Previous reports of an absence of detectable vimentin in crosssections of chick (36) and mammalian (33) mesothelium by
immunofluorescence microscopy may have been due to the
relatively low vimentin content in mesothelial cells compared to
the high levels present in fibroblasts and endothelial cells and
because the mesothelium is very thin and easily damaged during
handling and fixation. Mesothelial cells may express keratins and
vimentin because they are derived from the embryonic mesoderm (this subject is reviewed in Refs. 9, 18, and 37), inasmuch
as most other cell types of mesodermal origin, including mesenchyme-derived cells and cells of some hematopoietic lineages,
express vimentin (2, 8, 12, 13, 15, 31, 36).
The types of keratins in normal mesothelial tissue are restricted
to the simple epithelial keratins, as we had predicted from the
previous results of this laboratory on peritoneal mesothelial cells
in culture (4, 43). Samples of human pleural mesothelium con
tained the 40 KM, 44 KM, and 52 KM simple epithelial keratins,
but very little or no keratin 55 KM. This lack of keratin 55 KMwas
not simply a regional difference between pleural and peritoneal
mesothelial cells, however, because pleural mesothelial cells
expressed the keratin 55 KM in culture. We do not know the
mechanism of this specific regulation of keratin 55 KMbut it could
be the result of very rapid cell division or the abnormal culture
substratum. We had found previously that synthesis and content
of all 4 keratins are controlled coordinately in mesothelial cells in
culture as a function of growth rate (4, 34).
Solid and effusion mesotheliomas and 2 short-term cultured
mesothelioma populations contained 3 or 4 of the simple epithe
lial keratins and vimentin. Six of 9 surgical samples were suc
cessfully analyzed by 2-dimensional gel electrophoretic separa
tion and silver staining. Qualitative determination of the inter
mediate filament proteins was possible by this method, but
quantitative estimation of the M, 52,000 simple epithelial keratin
and vimentin was often prevented by substantial amounts of
proteolysis. These 2 intermediate filament proteins are particu
larly susceptible to digestion by an endogenous protease in
some tissue types and cell lines (14, 28, 35). Furthermore, the
presence of vimentin in the mesothelioma cells of most samples
could not be confirmed by electrophoretic analysis because of
the presence of connective tissue, including vimentin-containing
fibroblasts and endothelial cells. Immunofluorescence or immunoperoxidase staining with antivimentin will undoubtedly be a
more generally useful method for screening tumors for vimentin
expression.
We tried to culture mesothelioma cells in order to compare
their growth requirements with those of normal mesothelial cells
and to identify selective markers for malignant transformation of
this cell type. Only 2 of 7 tumors placed in culture yielded a
viable population of cells other than normal fibroblasts and
mesothelial cells, and these did not grow progressively in a
2994
medium permissive for rapid, long-term growth of normal me
sothelial cells. This suggests that mesothelioma cells have sig
nificantly altered growth requirements which are difficult to sat
isfy in culture. We suspect that this is the reason that we have
been unable to find any published reports of mesothelioma cell
lines. Normal mesothelial cells are anchorage independent, form
ing large colonies in semisolid medium.4 Thus, there is no basis
at present for culturing mesothelioma cells or selectively identi
fying them.
In addition to the coexpression of keratins and vimentin in
mesothelioma, the specific subset of keratins expressed by
mesotheliomas could be used to aid in the differential diagnosis
of tumors in the pleural and peritoneal cavities. Tumors of
stratified epithelial and mammary epithelial origin contain 2 or
more keratins from a set of more basic (M, 58,000 and 56,000)
and more acidic (M, 52,000, 50,000, and 46,000) keratins (ker
atins 5 and 6 and keratins 13, 14, and 17, respectively) (22-24,
26, 44). Carcinomas of the colon, stomach, and ovary would not
be distinguishable from mesothelioma by the keratin pattern
alone because these tumors also are restricted to simple epithe
lial keratins (22-24). Nevertheless, mesotheliomas are carcinoembryonic antigen negative or very weakly positive (5,19, 40,
41 ), while adenocarcinomas of the colon are strongly carcinoembryonic antigen-positive (17). Ovarian epithelial neoplasms can
be recognized by a specific antibody against a surface antigen
(1). Our results indicate that vimentin is likely to be present in
mesotheliomas, in contrast to its absence in tumors of all other
epithelial cell types (31). Vimentin has also been found recently
in several renal carcinomas (16) which originate from another
mesoderm-derived epithelial cell type. The data we have pre
sented here suggest that coexpression of simple epithelial ker
atins and vimentin is a useful marker for distinguishing normal
and malignant cells of mesothelial origin from those of most other
cell types.
ACKNOWLEDGMENTS
We are very grateful to Dr. J. M. Corson, Brigham and Women's Hospital, for
providing mesothelioma samples and pathology reports and for helpful comments
on the manuscript. Susan Rehwoldt and Terese O'Connell provided excellent
technical assistance with cell culture and preparation of vimentin antiserum. We
thank Dr. Oswaldo Alberti, Brigham and Women's Hospital, for providing fetal
human lung samples, Dr. Werner Franke, German Cancer Research Institute,
Heidelberg, West Germany, for his gift of guinea pig anti-vimentin, and Dr. Robert
Bast of this institute for a sample of OC125 antibody. We appreciate Lynne Dillon's
skillful processing of the manuscript.
REFERENCES
1. Bast, R. C., Feeney, M., Lazarus, H., Nadler, L. M., Colvin, R. B., and Knapp,
R. C. Reactivity of a monoclonal antibody with human ovarian carcinoma. J.
Clin. Invest., 68.- 1331-1337, 1981.
2. Bennett, G. S., Fellini, S. A., Croop, J. M., Otto, J. J., Bryan, J., and Holtzer,
S. Differences among 100-A filament subunits from different cell types. Proc.
Nati. Acad. Sci. USA, 75: 4364-4368,1978.
3. Cain, H., Kraus, B., Krauspe, R., Osbom, M., and Weber, K. Vimentin filaments
in peritoneal macrophages at various stages of differentiation and with altered
function. Virchows Arch. B Cell Pathol., 42: 65-81, 1983.
4. Connell, N. D., and Rheinwald, J. G. Regulation of cytoskeletal composition in
mesothelial cells: reversible loss of keratin and increase in vimentin during
rapid growth in culture. Cell, 34: 245-253.1983.
5. Corson, J. M., and Pinkus, G. S. Mesothelioma: profile of keratin proteins and
carcinoembryonic antigen: an immunoperoxidase study of 20 cases and com
parison with pulmonary adenocarcinomas. Am. J. Pathol., 108: 80-88,1982.
6. Cunningham, R. S. On the origin of the free cells of serous exúdales. Am. J.
* P. J. LaRocca and J. G. Rheinwald, manuscript in preparation.
CANCER
RESEARCH
VOL. 44
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1984 American Association for Cancer Research.
Intermediate Filament Proteins of Mesothelium and Mesothelioma
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
Physiol., 59: 1-36,1922.
Cunningham, R. S. The physiology of the serous membranes. Physio). Rev.,
6: 242-280, 1926.
Dellagi, K., Vainchenker, W., Vinci, G., Paulin, D., and Brauet, J. C. Alteration
of vimentin intermediate filament expression during differentiation of human
hemopoietic cells. EMBO J., 2: 1509-1515, 1983.
Donna, A., and Betta, P. G. Mesotheliomas: a new embryological approach to
primary tumors of coelomic surfaces. Histopathology (Oxf.), 5: 31-44, 1981.
Eschenbruch, M., and Burk, R. R. Experimentally improved reliability of ultra
sensitive silver staining of proteins in polyacrylamide gels. Anal. Biochem.,
r25: 96-99, 1982.
Franke, W. W., Schmid, E. A., Breitkreutz, D., Luder, M., Boukamp, P.,
Fusenig, N. E., Osbom, M., and Weber, K. Simultaneous expression of two
different types of intermediate sized filaments in mouse keratinocytes prolif
erating m vitro. Differentiation, 14: 35-50,1979.
Franke, W. W., Schmid, E., Osbom, M., and Weber, K. Different intermediatesized filaments distinguished by immunofluorescence microscopy. Proc. Nati.
Acad. Sci. USA, 75: 5034-5038, 1978.
Franke, W. W., Schmid, E., Winter, S., Osbom, M., and Weber, K. Widespread
occurrence of intermediate-sized filaments of the vimentin-type in cultured
cells from diverse vertebrates. Exp. Cell Res., 723: 25-46,1979.
Granger, B. L, and Lazarides, E. Desmin and vimentin coexist at the periphery
of the myofibril Z disc. Cell, 78: 1053-1063, 1979.
Granger, B. L., Repasky, E. A., and Lazarides, E. Synemin and vimentin are
components of intermediate filaments in avian erythrocytes. J. Cell Bid., 92:
299-312,1982.
Herman, C., Moesker, 0., Kant, A., Huysmans, A., Vooijs, G. P., and Ramaekers, F. C. S. Is renal cell (Grawitz) tumor a carcinosarcoma? Evidence from
analysis of intermediate filament types. Virchows Arch. B Cell Pathd., 44: 7383, 1983.
Heyderman, E., and Neville, A. M. A shorter immunoperoxidase technique for
the demonstration of carcinoembryonic antigen and other cell products. J. Clin.
Pathol. (Lond.), 30: 138-140, 1977.
Klemperer, P., and Rabin, C. B. Primary neoplasms of the pleura. Arch. Pathol.,
77:385-412,1931.
Kwee, W. S., Veldhuizen, R. W., Golding, R. P., Mullink, H., Stan, J., Donner,
R., and Boon, M. E. Histologie distinction between malignant mesothelioma,
benign lesion and carcinoma metastasis. Evaluation of the application of
morphometry combined with histochemistry and immunostaining. Virchows
Arch. A Pathol. Anat. Histol., 397: 287-299, 1982.
Laemmli, U. K. Cleavage of structural proteins during the assembly of the head
of bacteriophage T4. Nature (Lond.), 227: 680-685, 1970.
Lane, E. B., Hogan, B. L. M., Kurkinen, M., and Garrels, J. I. Coexpression of
vimentin and cytokeratins in parietal endoderm cells of early mouse embryo.
Nature (Lond.), 303: 701-704,1983.
Moll, R., Franke, W. W., Schiller, D. L., Geiger, B., and Krepier, R. The catalog
of human cytokeratins: patterns of expression in normal epithelia. Cell, 37:
11-24,1982.
Moll, R., Krepier, R., and Franke, W. W. Complex cytokeratin polypeptide
patterns observed in certain human carcinomas. Differentiation 23: 256-269,
1983.
Moll, R., Levy, R., Czemobilsky, B., Hohlweg-Majert, P., Dallenback-Hellweg,
G., and Franke, W. W. Cytokeratins of normal epithelia and some neoplasms
of the female genital tract. Lab. Invest., 49: 599-610, 1983.
Morrisey, J. H. Silver stain for proteins in polyacrylamide gels: a modified
procedure with enhanced uniform sensitivity. Anal. Biochem., 777: 307-310,
1981.
Nelson, W. G., Battifora, H., Santana, H., and Sun, T-T. Specific keratins as
molecular markers of neoplasms with a stratified epithelial origin. Cancer Res.,
44: 1600-1603, 1984.
JULY
27. Nelson, W. G., and Sun, T-T. The 50- and 58-kdalton keratin classes as
molecular markers for stratified squamous epithelial cell culture studies. J. Cell
Biol., 97:244-251,1983.
28. Nelson, W. J., and Traub, P. Properties of a Ca2*-activated protease specific
for the intermediate-sized filament protein vimentin in Ehrtich-ascites tumour
cells. Eur. J. Biochem., 776: 51-57, 1981.
29. O'Farrell, P. H. High resolution two-dimensional electrophoresis of proteins.
Anal. Biochem., 250: 4007-4021, 1975.
30. O'Farrell, P. Z., Goodman, H. M., and O'Farrell, P. H. High resolution two-
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
dimensional electrophoresis of basic as well as acidic proteins. Cell, 72:11331142, 1977.
Osborn, M., and Weber, K. Tumor diagnosis by intermediate filament typing:
a novel tool for surgical pathology. Lab. Invest. 48: 372-394, 1983.
Ramaekers, F. C. S., Hagg, D., Kant, A., Moesker, O., Jap, P. H. K., and
Vooijs, G. P. Coexpression of keratin- and vimentin-type intermediate filaments
in human metastatic carcinoma cells. Proc. Nati. Acad. Sci. USA, 80: 26182622, 1983.
Ramaekers, F. C. S., Puts, J. G., Kant, A., Moesker, O., Jap, P. H. K., and
Vooijs, G. P. Use of antibodies to intermediate filaments in the characterization
of human tumors. In'. Cold Spring Harbor Symposia on Quantitative Biology,
Vol. 46, pp. 331 -340. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory,
1981.
Rheinwald, J. G., O'Connell, T. M., Connell, N. D., Rybak, S. M., AllenHoffmann, B. L., LaRocca, P. J., Wu, Y.^J., and Rehwoldt, S. M. Expression
of specific keratin subsets and vimentin in normal human epithelial cells—a
function of cell type and condition of growth during serial culture. In: Cancer
Cells 1/The Transformed Phenotype, pp. 217-228. Cold Spring Harbor, NY:
Cold Spring Harbor Laboratory, 1984.
Schiller, D. L., and Franke, W. W. Limited proteolysis of cytokeratin A by an
endogenous protease: removal of positively charged terminal sequences. Cell
Biol. Int. Rep., 7: 3,1983.
Schmid, E. A., Tappscott, S., Bennet, G. S., Croop, J., Fellini, S. A., Hoitzer,
H., and Franke, W. W. Differential location of different types of intermediatesized filaments in various tissues of the chicken embryo. Differentiation, 75:
27-40,1979.
Stout, A. P., and Murray, M. R. Localized pleural mesothelioma. Arch. Pathol.,
34:951-962,
1942.
Tseng, S. C. G., Jarvinen, M. J., Nelson, W. G., Huang, J.-W., WoodcockMitchell, J., and Sun, T.-T. Correlation of specific keratins with different types
of epithelial differentiation: monoclonal antibody studies. Cell, 30: 361-372,
1982.
Virtanen, I., Lehto, V. P., Lehtonen, E., Vartio, T., Kurki, P., Wager, O., Small,
J. V., Dahl, D., and Badley, R. A. Expression of intermediate filaments in
cultured cells. J. Cell Sci., 50: 45-63,1981.
Walts, A. A., Said, J. W., and Banks-Schlegel, S. Keratin and carcinoembryonic
antigen in exfoliated mesothelial and malignant cells: an immunoperoxidase
study. Am. J. Clin. Pathol., 80: 671-676, 1983.
Wang, N. S., Huang, S. N., and Gold, P. Absence of carcinoembryonic antigenlike material in mesothelioma: an immunohistochemical differentiation from
other cancers. Cancer (Phila.), 44: 937-943,1979.
Whitaker, D., Papadimitriou, J. M., and Walters, M. N.-l. The mesothelium:
techniques for investigating the origin, nature and behaviour of mesothelial
cells. J. Pathol., 732: 263-271, 1980.
Wu, Y.>J., Parker, L. M., Binder, N. E., Beckett, M. A., Sinard, J. H., Griffiths,
C. T., and Rheinwald, J. G. The mesothelial keratins: a new family of cytoskeletal proteins identified in cultured mesothelial cells and nonkeratinizing
epithelia. Cell, 37: 693-703, 1982.
Wu, Y.-J. and Rheinwald, J. G. A new small (40kd) keratin filament protein
made by some cultured human squamous cell carcinomas. Cell, 25: 627-635,
1981.
1984
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1984 American Association for Cancer Research.
2995
••>>
' ^v^
, . >-v% ; f, ^'.
1¿v / ?'.' -"i * •»il
,
•#M
•
ä
r4 ' f<4V
Fig. 1. Morphology and M, 40,000 keratin immunofluoresœnce of human mesothelium in vivo and of mesothelioma cells in culture, a, fetal lung mesothelium transferred
to an agarose-coated slide; b, immunofluorescence of same field as a; c, scraped and cytocentrifuged fetal lung mesothelium; cf, immunofluorescence of same field as c;
e, normal peritoneal mesothelial strain LP-9 in secondary culture; I, ¡mmunofluorescence of a different LP-9 culture; g, pleural mesothelioma strain MS-7 in secondary
culture; h, immunofluorescence of a different culture of MS-7. a, c, e, and g are phase contrast; b, d, f, and h are anti-M, 40,000 keratin immunofluorescence. Bar in a is
50 urn and is the same magnification for 6; bar in c is 50 »mand is the same magnification for d; bar in e is 200 urn and is the same magnification for g; bar in f is 50 ^m
and is the same magnification for h.
2996
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1984 American Association for Cancer Research.
Intermediate Filament Proteins of Mesothelium and Mesothelioma
Fig. 2. Vimentin and keratin immunofluorescence of intact lung mesothelium and of mesothelial cells In ascites fluid, a to c, intact mesothelium scraped from rat lung.
Double labeling with guinea pig antivimentin and fluorescein (b) and with rabbit anti-M, 40,000 keratin and rhodamine (c). Small, retractile cells in a are erythrocytes. d to
f. human ascites fluid cells from patient with ovarian epithelial neoplasia, double labeled with guinea pig antivimentin and fluorescein (e) and with rabbit anti-stratum
comeum keratins and rhodamine (f). g to i, human ascites fluid cells from the same patient, double labeled with rabbit antivimentin and rhodamine (h) and with mouse
monoclonal anti-ovarian carcinoma cell surface and fluorescein (/). Bar in a is 50 pm and is the same magnification for b to i. In d, arrows indicate mesothelial cells. In d
and g, arrowheads indicate neoplastic ovarian epithelial cells. Vimentin-positive, keratin-negative cells in d to t are presumably macrophages.
JULY
1984
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1984 American Association for Cancer Research.
2997
bt
t
a
•
':v'<ïr--.'.¡'.'i
.:••
•.-,
?'•;."•'"'
v
'T^
v
C
b
b
2998
b'
^
1a
CANCER
RESEARCH
VOL. 44
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1984 American Association for Cancer Research.
Intermediate Filament Proteins of Mesothelium and Mesothelioma
K
—y
»
a^~
b
cFig. 4. Two-dimensional gel electrophoretic separation of mesothelioma cytoskeletal proteins. Triton:high-salt-insoluble proteins of solid tumor samples separated
horizontally by charge (loll, basic; right, acidic) and vertically by sodium dodecyl sulfate:polyacrylamide (10%) gel electrophoresis. Gels a and c were focused isoelectrically
between pH 6.5 and 5.0. Gels b and d were separated by nonequilibrium pH gradient gel electrophoresis to permit detection of more basic proteins, a, peritoneal
mesothelioma MS-4; £>,
pleural mesothelioma MS-2; c, peritoneal mesothelioma MS-5; d, pleural mesothelioma MS-11. All gels are silver stained. Lower case letters and
arrows within each frame, position of keratins, vimentin, and actin, as in the legend of Fig. 3; arrowheads, proteolytic breakdown products of vimentin and the 52K«
keratin. * in frame a, prominent species of size and charge different from that of any known keratin, v, vimentin.
Fig. 3. Two-dimensional gel electrophoretic separation of the cytoskeletal proteins of normal human mesothelium and mesotheliomas. Triton:high-salt-insolubte protein
fractions of tissues, tumors, and cultured cells separated horizontally by their charge (left, basic; right, acidic) and vertically by sodium dodecyl sulfate;polyacrylamide
(10%) gel electrophoresis. Gels a, c, e, and g were isoelectrically focused between pH 6.5 and 5.0 in the horizontal dimension. Gels b, d, f, and h were separated by
nonequilibrium pH gradient gel electrophoresis in the horizontal dimension in order to permit detection of more basic proteins, a, human fetal lung mesothelium collected
by slide transfer (spots marked with * are agarose contaminants detected by silver staining); b, cultured adult lung mesothelial strain HPM-2; c and d. cultured adult
peritoneal mesothelial cell strain LP-9; e, pleural effusion mesothelioma MS-7; f, peritoneal mesothelioma MS-10; g, secondary culture of MS-7; h, primary culture of MS10. Proteins in a, d, and e were revealed by silver staining, in frame f by Coomassie blue staining, and in b, c, g, and h by autoradiography of [^Slmethionine-proteins.
Lower case letters within each frame, positions of keratins and vimentin: a, 40«M/19;b, 44KM/18; c, 52K«/8;d, 55KM/7; e, 46KM. Arrows, actin. c' is presumed to be a
proteolytic digestion product of keratin 52KM/8. v, vimentin.
JULY
1984
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1984 American Association for Cancer Research.
2999
Coexpression of Simple Epithelial Keratins and Vimentin by
Human Mesothelium and Mesothelioma in Vivo and in Culture
Paul J. LaRocca and James G. Rheinwald
Cancer Res 1984;44:2991-2999.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/44/7/2991
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 17, 2017. © 1984 American Association for Cancer Research.