| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Articles |
Departments of Obstetrics and Gynecology (M.I., R.N.) and Pathology (Y.M., A.O.), Wakayama Medical College, Wakayama 641-0012, Japan
Address all correspondence and requests for reprints to: Masaaki Iwahashi, M.D., Department of Obstetrics and Gynecology, Wakayama Medical College, 811-1 Kimiidera, Wakayama 641-0012, Japan. E-mail: masaaki{at}wakayama-med.ac.jp
Abstract
To investigate the characteristic structure of hydatidiform mole, various types of collagen expression were determined in human villous tissues obtained from normal pregnancies (n = 17) and complete hydatidiform moles (n = 10). Indirect immunofluorescent staining was performed to detect type I, III, and VI collagen with specific monoclonal antibodies. Collagens were also extracted from the villous tissues obtained from normal pregnancy and hydatidiform mole by the salt precipitation method. Immunohistochemical staining for type I, III, and VI collagen revealed weak staining of the villous stroma in hydatidiform mole compared with that in normal pregnancy. Both the ratios of type III to type I collagen and the ratios of type V to type I collagen in the villous tissues were significantly decreased (P < 0.05) in molar pregnancy compared with those in normal pregnancy. These results suggest that alterations in the distribution and composition of collagen might play an important role in determining the pathophysiology and structure of hydatidiform mole.
THE EXTRACELLULAR matrix (ECM) is considered to play an important role in the stability of tissues and in regulating the growth and differentiation of cells (1, 2). Synthesis, accumulation, and catabolism of the ECM occur during wound healing and during the initiation and progression of numerous diseases (3). The ECM of normal villous tissues has been studied (4, 5), but changes in the ECM in the villous tissues of hydatidiform mole are not fully understood.
The hydatidiform mole has such a characteristic macroscopic appearance that it may be diagnosed by the ultrasonographer, attendant at delivery, or the pathologist. Microscopic examination discloses diffuse, marked villous enlargement due to massive stromal edema. The edema displaces the mesenchymal stroma centrally, creating an acellular clear space called a central cistern. There is marked proliferation of cyto- and syncytiotrophoblast (6).
In the present study we investigated the distribution and composition of the various types of collagen, the major component of ECM, in hydatidiform moles by immunofluorescent staining and SDS-PAGE. The results were compared with those obtained during normal pregnancy.
Materials and Methods
This project was approved by the committee on investigations involving human subjects of Wakayama Medical College. Informed consent was obtained from each subject, after the purpose and nature of the study were fully explained.
Tissues
Normal villous tissues were obtained from 17 women, aged 20–38 yr, during early pregnancy (7–12 weeks) by dilatation and curettage for termination of pregnancy and were immediately frozen in liquid nitrogen. Molar tissues were obtained from 10 women, aged 22 to 41 yr, by dilatation and curettage at 7–13 weeks gestation. The pathological diagnosis of all 10 cases was complete hydatidiform mole. We excluded necrotic villous tissue from analysis by histological examination. Gestational age was determined from the date of the last menstrual period and by ultrasonographic measurements performed in early pregnancy.
Primary antibodies
Monoclonal antibodies (mAbs) against each 
1-chain
of human type I, III, and VI collagen were used. Preparation of the
antibodies was described previously (7). In brief, BALB/c
mice were immunized with each type of collagen after it had been
extracted from human placentas. The spleen cells of these mice were
then hybridized with myeloma cells. After
hypoxanthine-aminopterine-thymidine selection, positive hybrids were
identified by enzyme-linked immunosorbent assay. The specificity of
each antibody was determined by immunoblotting or by inhibition in an
enzyme-linked immunosorbent assay. No cross-reaction was recognized
among these antibodies.
Immunohistochemistry
Immunohistochemical analysis was performed by the standard indirect immunofluorescence method. In brief, 3-µm frozen sections were rehydrated in PBS at room temperature and then incubated with the primary antibody (diluted 1:100 in PBS) for 12 h at 4 C in a humidified chamber. After incubation, the sections were washed twice in PBS for 3 min each time. Each section was then incubated for 1 h at room temperature with human plasma-preabsorbed, fluorescein isothiocyanate-conjugated goat antibodies against mouse Igs diluted 1:100 in PBS (Organon Teknik Co., West Chester, PA). Subsequently, the sections were washed again in PBS, mounted in buffered glycerol, and examined under a fluorescence microscope (Olympus Corp., Tokyo, Japan).
SDS-PAGE of pepsin-solubilized collagens from the human villous tissues
Minced samples of human villous tissues were washed overnight in cold distilled water and freed of blood. Tissues were homogenized with a Polytron homogenizer (Brinkmann Instruments, Inc., Westbury, NY) in 50 vol 0.5 mol/L acetic acid that contained 1 mg/mL pepsin (Sigma, St. Louis, MO). Collagens were extracted with constant stirring for 24 h at 4 C. Each solution was centrifuged at 39,000 x g for 1 h at 4 C. Collagens were reextracted from the pellet under the same conditions as those described above for 48 h. The supernatants corresponding to individual samples were then combined, and collagens were precipitated by addition of 4.0 mol/L NaCl to a final concentration of 2.0 mol/L. Each precipitate was dissolved in 0.5 mol/L acetic acid, and the solution was dialyzed against 0.02 mol/L Na2HPO4. Precipitated collagens were redissolved in 0.5 mol/L acetic acid, dialyzed exhaustively against 0.05 mol/L acetic acid, and finally lyophilized.
The solubility of the tissue collagen from each villous sample
was estimated by comparing the hydroxyproline content of the initial
homogenate with that of the final solution of collagen
(8). Type V collagen was isolated by salt precipitation
from pepsin digests of human villous tissues by the methods described
previously (9, 10). The extracted type V collagen was also
lyophilized. Estimation of the relative abundance of the
1(III)-chain and
1(V)-chain was performed by interrupted gel
electrophoresis (11). Electrophoresis was performed in an
8% polyacrylamide slab gel (Sigma). The gel and electrode
buffers were 0.1 mol/L phosphate buffer, pH 7.2, containing 0.1% SDS
(Nacalai Tesque, Inc., Kyoto, Japan), as previously described
(12). Lyophilized samples of collagen and type V collagen
were dissolved at a concentration of 0.2 mg/mL and denatured by heating
in the gel buffer that contained 1% SDS at 60 C for 30 min. Aliquots
of 25 µL of each solution of denatured collagens and 5 µL denatured
type V collagen were applied to the gel and subjected to
electrophoresis at 80 mA. After 1.5 h the current was switched
off, sample wells were filled with a solution of 20%
ß-mercaptoethanol (Wako Chemical Co., Osaka, Japan) in gel buffer,
and the ß-mercaptoethanol was allowed to diffuse into the gel for
1 h to cleave the intramolecular disulfide bonds of type III
collagen, [1(III)]3.
Then electrophoresis was resumed and allowed to continue for another
1 h. Each collagen -chain was stained with Coomassie brilliant
blue (Sigma) and quantitated by densitometry. The relative
amounts of 1(III) or
1(V) chains were calculated by dividing the
intensities of bands areas under densitometric peaks of
1(III) and 1(V) by
that of 1(I). In this method, the -chains
derived from type VI collagen could not be evaluated because of
degradation of this collagen by pepsin.
Type I collagen content in the human villous tissues
Type I collagen was isolated by salt precipitation from human villous tissues (10 g wet weight) of normal and molar pregnancy by the methods described previously (9, 10). The extracted type I collagen was lyophilized, and we measured the weight of this collagen.
Statistical analysis
Densitometric data and type I collagen weights were expressed as the mean ± SEM. Mean values were compared by Student’s t test or ANOVA using a StatView software program (Abacus Concepts, Inc., Berkeley, CA) on a Macintosh computer (Cupertino, CA). Two-tailed P < 0.05 was considered statistically significant.
Results
Immunohistochemical analysis of the villous tissues
Control sections were stained with goat antibodies against mouse
IgG without prior application of the appropriate primary antibody
(Figs. 1A
, 2A, and 3A). When the mAbs
were first allowed to react with an excess of each specific type of
collagen, no immunostaining was observed.
|
|
|
) and type III
(Fig. 2B) collagen showed a fibrillary
pattern in the villous stroma in normal pregnancy. However,
immunostaining for these types of collagen showed a weak and fragmented
pattern in the villous stroma in molar pregnancy (Figs. 1C and 2C).
Immunostaining with the mAbs against type VI collagen showed a
microfibrillary pattern in the villous stroma in normal pregnancy (Fig. 3B). However, immunostaining for this
collagen showed a weak and fragmented pattern in the villous stroma in
molar pregnancy (Fig. 3C). The intensity of staining of the villous
tissues from normal pregnancy and complete mole by each of the
antibodies against various types of collagen was subjectively graded
from 1+ to 3+, and the results are summarized in Table 1.
|
Although the relative levels of 1(I) were
similar in the villous tissues obtained from normal pregnancy and
hydatidiform mole, those of 1(V) and those of
1(III) decreased in the villous tissues of
hydatidiform mole compared with those of normal pregnancy (Fig. 4). The data are summarized in Table 2. The mean ratio of the intensity of the
band of 1(III) to that of
1(I) in villous tissues of hydatidiform mole
was significantly lower than that in villous tissues of normal
pregnancy (P < 0.05). Moreover, the mean ratio of the
intensity of the band of 1(V) to that of
1(I) in villous tissues of hydatidiform mole
was significantly lower than that in villous tissues of normal
pregnancy (P < 0.01).
|
|
Type I collagen content in villous tissues of hydatidiform mole was significantly lower than that in villous tissues of normal pregnancy (8.9 ± 2.5 vs. 19.2 ± 4.3 mg; P < 0.01).
Discussion
In the present study we investigated changes in the distribution and composition of the ECM, including type I, III, V, and VI collagens, in the human villous tissues obtained normal pregnancy and hydatidiform mole. We were able to solubilize 70–85% of collagen in the human villous and molar tissues, as measured by reference to levels of hydroxyproline (data not shown). Therefore, it was postulated that the extracted collagen might accurately reflect the entire complement of collagen in the tissues.
The decreased expression of interstitial collagens, such as type I, III, and VI collagens, was confirmed immunohistochemically in the villi of hydatidiform mole compared with those of normal pregnancy. Moreover, the type I collagen content in the villous tissues of hydatidiform mole was significantly lower than that in villous tissues of normal pregnancy. This finding was consistent with the immunohistochemical results. However, the relative level of type I collagen in the total collagen extracted from the villous tissues was similar in normal and molar pregnancy.
Although type I and type III collagens are commonly found in combination, the ratio of type III or type V to type I collagen in the tissues of hydatidiform mole was significantly lower than that in normal pregnancy. Changes in the ratio of type III to type I collagen have been demonstrated in human skin (11) and in human atherosclerosis (13, 14). A possible cause might be an alteration in the density of cells in the human villous tissues. Recently, cell density-dependent effects have been reported in the various types of cell, such as mesangial cells (15, 16), endothelial cells (17, 18), vascular smooth muscle cells (19, 20), fibroblasts (21, 22), and primitive mesenchymal cells (23). It has been suggested that cell density might modulate biological behavior, with changes in signal transduction responses to hormonal stimulation, in growth, in the synthesis and composition of the ECM, and in the synthesis of specific proteins (15, 16). Wolthuis et al. (16) reported that mesangial cells synthesized relatively more type I collagen per cell at higher cell densities, whereas synthesis of type III and type IV collagens in each cell did not depend on cell density. Therefore, it is suggested that hydatidiform mole might show intense proliferation of the trophoblastic cells, and that at higher cell densities these cells might synthesize relatively more type I collagen than the other collagen.
Type V collagen was originally described as a component of chorionic
and amniotic membranes (24). It is thought to play a major
role in maintaining a barrier against pathogens and inflammatory cells
and in preventing the loss of amniotic fluid (25). Type V
collagen has the ability to bind to insulin (26) and
heparin/heparan sulfate (27) with apparently higher
affinity than collagen types I, II, III, IV, or VI; fibronectin; or
laminin. Yaoi et al. (26) indicated that
insulin bound to type V collagen retains mitogenic activity and that
heparin/heparan sulfate modulates the biological activities of vascular
endothelial cell growth factor (28) and basic fibroblast
growth factor (29, 30). These findings suggest that type V
collagen might be important in the compartmentalization, storage,
stabilization, and modulation of the activities of various growth
factors. As type V collagen also binds to thrombospondin
(31), the interactions of this collagen with
thrombospondin and heparan sulfate may be important in the assembly of
the ECM and in the regulation of its biological functions. In addition,
this collagen has been shown to mediate the behaviors of cells,
including their attachment, spread proliferation, and morphogenesis
(32, 33). Therefore, it is suggested that decreased
relative levels of 1(V) or type V collagen in
the villi might provide a biochemical basis for functional regulation
of the trophoblast cells of hydatidiform mole. It has also been
postulated that trophoblast invasion might be modulated by interactions
between the ECM receptors on the surface of trophoblast cells and the
corresponding ligands (34, 35). Thus, a reduction of type
V collagen in the villi might result in reduced attachment or increased
metastatic ability of trophoblast cells in hydatidiform moles.
In conclusion, decreased interstitial collagens, including type I, III, V, and VI collagens, were found in the stroma of the villi in hydatidiform moles. Therefore, these alterations in the distribution and the composition of collagens might result in the morphological and functional characteristics of hydatidiform mole.
This study provides some clues to understanding the pathophysiology of hydatidiform mole in terms of the ECM metabolism. Further work is needed to elucidate the mechanisms regulating the trophoblastic expression of genes for other types of collagen in normal pregnancy and hydatidiform mole.
Received April 4, 2000.
Revised October 4, 2000.
Revised March 20, 2001.
Accepted March 25, 2001.
References
B chain: increased
proportion in human atherosclerosis. Science. 213:666–668.
chains. Proc Natl Acad Sci USA. 73:2379–2383.1-6 integrin subunits
in human developing and term placenta. Lab Invest. 65:347–356.[Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
