help button home button Am J Pathol Epitomics, Inc.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lovecchio, M.
Right arrow Articles by Perlino, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lovecchio, M.
Right arrow Articles by Perlino, E.
(American Journal of Pathology. 2003;163:2543-2553.)
© 2003 American Society for Investigative Pathology

ß1C Integrin Expression in Human Endometrial Proliferative Diseases

Mariarosaria Lovecchio*, Eugenio Maiorano{dagger}, Rosa A. Vacca*, Giuseppe Loverro{ddagger}, Margherita Fanelli§, Leonardo Resta{dagger}, Sergio Stefanelli{ddagger}, Luigi Selvaggi{ddagger}, Ersilia Marra* and Elda Perlino*

From the Institute of Biomembranes and Bioenergetics,* National Research Council, Bari; the Departments of Pathological Anatomy and Genetics,{dagger} Obstetrics and Gynecology,{ddagger} and Internal Medicine and Public Medicine,§ University of Bari School of Medicine, Bari, Italy


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Integrins are ubiquitous cell adhesion molecules that are involved in maintaining normal tissue morphology and have been implicated in the aggressive behavior of several malignancies. ß1C integrin is an alternatively spliced variant of the ß1A integrin subunit that, at variance with ß1A, inhibits epithelial cell proliferation. ß1C integrin is expressed in non-proliferative, benign prostatic epithelium and is down-regulated in prostatic adenocarcinoma. In the current study, we examined ß1C expression at mRNA and protein levels in 18 endometrial adenocarcinoma and in 20 endometrial hyperplastic tissues, using Northern and Western blotting analysis and immunohistochemistry. The pattern of integrin expression was compared to that of the endometrium of 14 normal cycling women. The results of this study document inhibited ß1C integrin expression in endometrial adenocarcinoma, both at the mRNA and protein levels, at variance with significantly up-regulated ß1C mRNA expression in endometrial hyperplasia, in comparison with normal proliferative endometria. Our data suggest a key role of the regulation of ß1C integrin expression in the pathogenesis of endometrial proliferative diseases: ß1C integrin may act as growth modulator in cancer cells, playing a role in downstream intracellular signaling.

The endometrium is a complex system, which involves interactions among different cell types, steroid hormones, cytokines, adhesion molecules, and growth factors and which probably relies on an integrated network of intercellular and intracellular signaling pathways in which all of the above molecules act in a paracrine, autocrine, or endocrine fashion.1 Moreover, the normal process of cell growth requires a delicate balance between hormones, growth factors, and adhesion molecules, including integrins, and apoptosis-related genes; an alteration of the synthesis and/or expression of these factors can contribute to the genesis of benign or malignant proliferative diseases. Among the above factors, integrins, which belong to a large superfamily of cell adhesion receptors,2 coordinate the cell-extracellular matrix adhesive recognition and undergo dynamic alterations during the menstrual cycle in the human endometrium.

Integrins are heterodimeric transmembrane complexes of {alpha} and ß subunits, in which the cytoplasmic tails modulate the receptor ligand binding affinity, distribution, surface expression, cell adhesion, and spreading.3 Moreover, they play a fundamental role in the regulation of cell proliferation and differentiation by their effects on different pathways of signal transduction in cooperation with other molecules, such as hormones, cytokines and growth factors. In fact, integrin expression is differentially regulated in tumors and altered expression in cancer cells, such as in prostatic carcinoma,4,5 has been correlated with changes in invasiveness, tumor progression, and metastatic potential.6

Integrins display dynamic temporal and spatial patterns of expression in the endometrium during the menstrual cycle and pregnancy.7 In addition, previous studies have documented an inverse correlation between integrin expression and tumor differentiation in adenocarcinoma cells.1

Among the ubiquitous ß1 integrin family, an alternatively spliced variant of the human ß1 integrin subunit, designated ß1C, containing a unique 48 amino acid cytoplasmic domain, has been identified.8 Recent studies have demonstrated a unique function of ß1C in the regulation of cell growth2,9 ; expression of recombinant ß1C in mammalian cells completely inhibited DNA synthesis and cell proliferation, while the wild-type ß1 had no effect.10 Furthermore, in vivo expression of ß1C is thought to correlate with a "non-proliferating cell" phenotype, since ß1C is expressed in normal prostatic cells, while it is down-regulated in prostatic adenocarcinoma.11,12

While in a recent study we have shown that ß1 and ß1C integrins13 are present in the human endometrium throughout the menstrual cycle, in the postmenopausal age, and pregnancy, no data are available on the expression of ß1C integrin in proliferative diseases of the human endometrium such as adenocarcinoma and hyperproliferative lesions.

In consideration of the supposed pivotal role of the adhesion molecules in the regulation of endometrial growth and differentiation, the study of their expression in hyperplastic and neoplastic conditions may allow a better understanding of endometrial biology and tumorigenesis.

Based on these premises, the aim of this study was to explore the expression of ß1C integrin, both at mRNA and protein levels by means of Northern and Western blotting and immunohistochemistry to determine its altered expression in hyperplastic and neoplastic endometrial tissues, in comparison with normal proliferative endometria.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Study Population

The study population included 18 patients with endometrial adenocarcinoma, 20 patients with endometrial (simple or complex) hyperplasia (Table 1) , and 14 patients with proliferative endometria (3 in the early-, 4 in the mid-, and 5 in the late-proliferative phase undergoing hysterectomy for subserosal leiomyoma; for two patients it was not possible to establish the date of the last menstrual cycle). For the latter group of patients, the dating was based both on patients history and on histological examination.


View this table:
[in this window]
[in a new window]
  		Table 1. Clinicopathological Features of 18 Patients with Endometrial Carcinoma and 20 Patients with Endometrial Hyperplasia

 
RNA Extraction and Northern Blot Analysis

Fresh endometrial samples were collected under sterile conditions from the above cohort of patients and snap-frozen in liquid nitrogen. The frozen tissue samples were pulverized to a fine powder and cellular RNA was extracted using the guanidinium isothiocyanate-cesium chloride procedure.14 Total RNA (25 µg) isolated from the tissues was electrophoresed through 1% denaturing agarose gel containing 660 mmol/L formaldehyde, and transferred15 to a nylon membrane (Hybond N+; Amersham, Milan, Italy). The filters were subsequently pre-hybridized overnight at 42°C with a buffer consisting of 50% formamide, 5X Denhardt’s solution (1% Ficoll 400, 1% polyvinylpirrolidone, 1% bovine serum albumin), 5X SSPE [3 mol/L NaCl, 200 mmol/L Na2H2PO4 (pH 7.0), 19 mmol/L ethylenediaminetetraacetic acid (EDTA)], 0.5% sodium dodecyl sulfate (SDS) and 100 µg/ml sonicated salmon sperm DNA. The filters were then hybridized for 20 hours at 42°C by adding 3 x 106 cpm of 32P-labeled probe per milliliter to the pre-hybridization solution. Hybridization was carried out as already described by Perlino et al.12 The filters were washed once with 2X SSPE, 0.1% SDS, for 10 minutes at room temperature, then with 1X SSPE, 0.1% SDS at 42°C, followed by several washes in 0.1X SSPE, 0.1% SDS at 65°C and finally exposed at -80°C overnight or longer to Kodak X-OMAT AR 5 film (Kodak, Rochester, NY). Radiolabeled probes were generated using the Megaprime DNA labeling kit (Amersham), 5 µl of {alpha}[32P]-dCTP (3000 Ci/mmol, Amersham)16 and 25 ng of double-stranded either 116-bp fragment specific for the ß1C integrin or a full-length human ß1 cDNA.8 The specific 116-bp ß1C fragment (nucleotides 2435–2550)8 was generated by PCR using pBluescript ß1C plasmid as template and the resulting fragment was subcloned in the pBluescript vector as previously described.12

mRNA levels were normalized using a cDNA probe17 corresponding to the ribosomal 28S RNA, a constitutively expressed gene, expressed at invariant levels. For this purpose, the blots were stripped in 0.1% boiling SDS and re-probed with the radiolabeled-32P 28S cDNA probe. Quantitative analysis was performed by densitometric scanning of the autoradiographs using a Bio-Rad GS-700 densitometer (Bio-Rad, Richmond, CA); multiple exposures of the same Northern blots in a linear range were performed. The ratio between the 4.3-kb long ß1C mRNA levels and the 28S rRNA levels was calculated for each sample to take into account for differences in RNA loading. The average of either ß1C or ß1 mRNA expression levels in control normal endometrium, derived from 12 of the 14 patients, was set at 100 (arbitrary units). The variability of ß1C or ß1 mRNA levels in the normal tissue specimens, measured by comparing the percentage of mRNA level from different normal endometrial tissues present on the same filter, was always less than 10% in the different measurements.

ß1C or ß1 mRNA levels in neoplastic and hyperplastic endometria were calculated as percentage of normal endometria mRNA levels hybridized on the same filter. For each specimen, the mean value (± SEM) of results obtained in at least three experiments was calculated.

Immunoblotting

Either normal (7 cases), tumoral (8 cases), and hyperplastic (9 cases) endometrial frozen tissue samples obtained from hysterectomy were homogenized in lysis buffer containing 0.1% SDS, 1% Nonidet P-40, 50 mmol/L Tris-HCl (pH 7.5), 150 mmol/L NaCl, 200 mmol/L LiCl, 5 mmol/L EDTA, 10% glycerol, 10 µg/ml aprotinin, 120 µg/ml leupeptin, and 170 µg/ml phenylmethylsulfonyl fluoride. The homogenate was sonicated for 20 seconds then centrifuged for 30 minutes at 14,000 x g at 4°C. 2-mercapto-ethanol (1%) was added to each lysate for 30 minutes at 4°C to further solubilize potentially cross-linked molecules; it was then centrifuged, and supernatants were collected. One hundred and fifty micrograms of tissue extracts were electrophoresed on 7.5% SDS-polyacrylamide gel under reducing conditions and immunoblotting was carried out as previously described13,18 using either 5 µg/ml rabbit polyclonal affinity-purified antibody to ß1C integrin (provided by Dr. Languino) or 1 µg/ml mouse monoclonal antibody to ß1 integrin (Transduction Laboratories, Lexington, KY) or 10 µg/ml antibody to ß-tubulin ({sigma}-Aldrich Italia, Milan, Italy). The blots were incubated with antibody for 16 hours at 4°C in TBS-T [20 mmol/L Tris (pH 7.5), 150 mmol/L NaCl, 0.2% Tween-20] containing 5% nonfat dry milk. The membrane was then washed three times in TBS-T and incubated with horseradish peroxidase-conjugate goat affinity-purified antibody to rabbit or mouse IgG (Amersham Pharmacia Biotech, Milano, Italy), for ß1C integrin or ß1 integrin, respectively, diluted 1:20,000 in TBS-T for 1 hour at room temperature. After three washes in TBS-T, the proteins were visualized using the Amersham enhanced chemiluminescence system according to the manufacturer’s instructions. Protein levels were normalized using the constitutively expressed ß-tubulin protein. For this purpose, the blots were stripped at 55°C for 20 minutes with stripping buffer [2% SDS, 10 mmol/L ß-mercaptoethanol, 6 mmol/L Tris-HCl, (pH 6.8)] and hybridized, as described above, with 10 µg/ml antibody to ß-tubulin. Densitometric values for immunoreactive bands were quantified using a GS-700 Imaging Densitometer (Bio-Rad). ß1C and ß1 protein levels were calculated as a percentage of the control (proliferative endometrium) taken as 100 in arbitrary units, on normalization using ß-tubulin as control for protein loading. The variability of ß1C or ß1 protein levels in the normal tissue specimens, measured by comparing the percentage of protein level from different normal endometrial tissues present on the same filter, was always less than 5% in the different measurements.

For each specimen, the mean value (± SEM) of the results obtained in at least three experiments was calculated.

Histopathology and Immunohistochemistry

The study material included all surgical samples from which tissue fragments for the molecular investigations had been taken, with the exception of one case of endometrial carcinoma for which tissue blocks were not available for the immunohistochemical procedures.

The surgical samples were fixed in 10% neutral buffered formalin for 12 to 24 hours, embedded in paraffin, cut, and stained with hematoxylin and eosin (H&E). The histological preparations were reviewed by two pathologists (E. M. and L. R.) to compare the histological dating with the clinical dating in women with normal cycles, to specify the histological subtype of endometrial carcinomas and define tumor grade.

A single paraffin block per case was selected for immunostaining based on good morphological preservation. Five-micrometer-thick sections were cut, collected on positively charged slides, dewaxed, and rehydrated. Following quenching of endogenous peroxidase with 3% H2O2 for 15 minutes at room temperature, the sections were immunostained for ß1 integrin and its spliced variant ß1C integrin using an avidin-biotin peroxidase technique (ABC) with an automated immunostainer (TecMate 500; Dako, Glostrup, Denmark). Before the staining procedure, the sections to be incubated with anti-ß1 integrin antibodies were immersed in 1% SDS in TBS for 10 minutes at room temperature. A mouse monoclonal antibody against ß1 integrin (clone: 18, dilution 1:20; Transduction Laboratories) and a rabbit antiserum against ß1C integrin (provided by Dr. L. Languino and used at a 1.7 µg/ml dilution, as previously reported)12 were used as primary antibodies with overnight incubations at 4°C.

Control sections for specificity included staining of positive controls (normal and carcinomatous breast) and of negative control sections, which were incubated with the immunoglobulin fraction of normal mouse or rabbit sera in place of the specific immunoreagent (Fig 11) .



View larger version (135K):
[in this window]
[in a new window]
  		Figure 11. Controls of immunostaining specificity. A: Negative control for ß1C integrin immunoreactivity in endometrial carcinoma obtained by substituting the primary antibody with non-immune serum. Both the epithelial and stromal cells do not show any immunoreactivity. (ABC method anti-ß1C, x160). B: Positive control for ß1C integrin immunoreactivity in normal breast. The ductules show consistent immunoreactivity that is mainly compartmentalized at the apical pole of the epithelial cells. (ABC method anti-ß1C, x200).

 
Evaluation of ß1 Integrin and ß1C Integrin Immunoreactivity

In all cases ß1 integrin and ß1C integrin immunoreactivity was independently evaluated by two pathologists (E. M. and L. R.) by separately counting the relative number of immunoreactive stromal and epithelial cells in 10 different microscopic fields, observed at x400 magnification; the extent of the immunoreactivity within each cell component (stromal versus epithelial), meant as the percentage of immunoreactive cells, was semiquantitatively scored as follows: 0, absence of immunoreactive stromal/epithelial cells; 1, 1% to 10% immunoreactive stromal/epithelial cells; 2, 11% to 25% immunoreactive stromal/epithelial cells; 3, 26% to 50% immunoreactive stromal/epithelial cells; 4, >50% immunoreactive stromal/epithelial cells.

HL60 Cells

Human leukemia HL60 cells were grown in RPMI 1640 (Life Technologies, Milan, Italy), with 50 µg/ml gentamicin, 2 mmol/L glutamine, and 15% inactivated fetal calf serum, at 37°C in presence of 5% CO2. Total RNA from differentiated cells was prepared 24 hours after incubation with 160 nmol/L TPA (phorbol-12-myristate-13-acetate; Sigma), as previously described.19 A sample of normal human liver, obtained during cholecystectomy, was also used to generate mRNA.

Statistical Analysis

The levels of ß1, ß1C, and 28 S mRNA among the different endometrial samples are expressed as means (± SEM) of normalized densities of cDNA bands. Regulation of ß1, ß1C, and 28 S mRNA expression with respect to controls in the separate measurements recorded for each patient was assessed by paired Student’s t-test. Differences between samples (endometrial carcinoma tissues and hyperplastic tissues) were evaluated by unpaired Student’s t-test comparing the mean differences respect controls. Semiquantitative evaluation of immunoreactivity in the different groups was performed by {chi}2 test; mean values of ß1 and ß1C immunoreactivity in epithelial and stromal cells, among the different groups, were compared by ANOVA followed by Duncan post hoc test.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Patients

The salient clinicopathological features of patients with hyperproliferative endometrial disease are illustrated in Table 1 . Overall, there were 11 endometrioid adenocarcinomas (6 well-differentiated, G1; 3 moderately, G2; and 2 poorly differentiated, G3), 4 serous adenocarcinomas (G1: 2 cases, G2 and G3: one case each), 2 adenosquamous (one G1 and one G3) and 1 clear cell carcinoma (G3) and, with the exception of 2 G1 carcinomas from premenopausal patients, all endometrial adenocarcinomas (K) occurred in postmenopausal women. The mean age of patients with histologically proven endometrial adenocarcinoma was 66 (± 9) (range, 50 to 85 years of age).

Among the endometrial hyperplasias, 13 cases of simple and 7 cases of complex hyperplasia with cytologic atypia were histologically identified.20 The mean age of the patients with simple or atypical hyperplasia was 55 (±13) (range, 35 to 83 years of age). None of the patients had received any adjuvant treatment (radiotherapy, chemotherapy, or hormone therapy) before surgery.

RNA Expression Analysis in Endometrial Carcinoma Tissues

Steady-state levels of ß1C mRNA were quantified by Northern blotting of RNA from 12 endometrial neoplastic and 12 non-neoplastic tissues. Because of the very low amount of RNA obtained from the tissue samples it was necessary to analyze total RNA rather than poly(A+)RNA.

Figure 1 shows the results of a typical Northern analysis of ß1C mRNA. A 4.3-kb-long transcript corresponding to the ß1C mRNA length, was found in all samples. Total RNA extracted from HL60 cells (lane 1) and human liver RNA (lane 2) were used as positive and negative controls of ß1C mRNA expression, respectively, since ß1C is produced by HL60 cells but not from normal human liver. As expected, ß1C mRNA levels in endometrial tissues was lower (37%) than the mRNA levels in HL60 cells (data not shown); moreover, a threefold increase of ß1C mRNA was documented in normal proliferative endometrial samples, in comparison with normal human liver (12% of HL60 mRNA level; data not shown).



View larger version (70K):
[in this window]
[in a new window]
  		Figure 1. ß1C mRNA expression is down-regulated in human endometrial carcinoma and up-regulated in endometrial hyperplastic tissues. Total RNA was isolated from 12 normal proliferative endometria, 12 endometrial carcinomas, and 17 endometrial hyperplasias. ß1C mRNA expression was evaluated by Northern blotting using the 116-b ß1C-specific probe. Twenty-five micrograms of total RNA were used for each sample. Lane 1: RNA from HL60 cells used as positive control. Lane 2: RNA from human liver used as negative control. Lanes 3 to 5: RNA from normal proliferative tissues. Lanes 6 to 10: RNA from neoplastic endometrial tissues. Lanes 11 to 12: RNA from hyperplastic tissues. As described in Material and Methods, after hybridization with the ß1C cDNA probe, the filter was stripped and hybridized again with the full-length ß1 probe. To normalize for the amounts of total RNA loaded for each sample, the blot was stripped and rehybridized once again using a 28S rRNA probe.

 
The autoradiographic analysis of Northern hybridization (Figure 1) showed variable expression of ß1C mRNA in the different neoplastic tissues (lanes 6 to 10), in comparison with the RNA from samples of proliferative endometrium (lanes 3 to 5).

To normalize the differences due to mRNA loading and transfer, the same blots were dehybridized and rehybridized again with a human 28S rRNA cDNA probe. The values of ß1C mRNA levels in all tissues were related to the 28S RNA levels, accounting for the RNA loading in each sample and calculated as a percentage of control (normal proliferative tissue). Finally, the mean values of at least three separate measurements for each patient were recorded and the average values ± SEM were reported in Figure 2 . Down-regulation of ß1C mRNA levels in comparison with the controls resulted in statistical significance (T = -3.03, P = 0.01). As illustrated in Figure 2 , the percentile expression of ß1C integrin in this subset of samples ranged from 38% to 137%, with an average value of 70% ± 9. A decrease greater than 10% of ß1C steady-state level mRNA expression (61% ± 9) was detected in 10 of 12 endometrial carcinomas (83%), as compared to proliferative endometrium. In particular, in eight of the carcinoma samples the difference, with respect to the control, was statistically significant (0.05 < P < 0.001); the remaining two carcinoma samples, though still showing down-regulated ß1C mRNA expression, were not considered statistically evaluable, because only a few measurements per case were available. Of the two carcinoma samples (K2 and K3) that did not show decreased ß1C mRNA levels, one (K2) demonstrated a statistically significant (P < 0.05) increased ß1C mRNA level (137% ± 13) in comparison with normal proliferative tissue, whereas the ß1C mRNA level of the remaining case (K3) was comparable (110% ± 12) to normal endometrial tissue (P > 0.05).



View larger version (17K):
[in this window]
[in a new window]
  		Figure 2. ß1C and ß1 mRNA expression in neoplastic endometria. ß1C and ß1 mRNA expression was analyzed as described in Figure 1 . The average of either ß1C or ß1 mRNA expression levels in 12 normal proliferative endometrial tissue samples (NP) was set at 100. ß1C and ß1 mRNA levels in neoplastic endometria (K1-K12) were calculated as percentage of NP. Mean values ± SEM from at least three different experiments are illustrated. Down-regulation of ß1C mRNA levels, in comparison with controls, was statistically significant (T = -3.03, P = 0.01). Down-regulation of ß1 mRNA levels in comparison with the controls was not statistically significant (T = -0.33, P = 0.74).

 
Since ß1C is a spliced variant of the ß1 integrin subfamily, alterations of its mRNA level could simply reflect an altered splicing mechanism occurring in endometrial cancer. To test such possibility, total ß1 mRNA level in all endometrial specimens was also measured using a full-length ß1 cDNA probe that hybridizes with all ß1 integrin variants.8 The 4.2-kb-long transcript corresponding to the ß1 mRNA length is illustrated in Figure 1 . The ß1 mRNA steady-state levels ranged between 34% and 150%, with an average value of 97% ± 10; the difference with the control group was not statistically significant (T = -0.33, P = 0.74).

In consideration of the relatively low number of endometrial carcinoma samples included in this study, a statistical evaluation of the relationships between ß1C mRNA expression and tumor histotype, grade, and stage was not feasible. Nevertheless, a trend toward a direct correlation between ß1C mRNA expression and tumor grade was apparent, as illustrated in Figure 3 , and may deserve further investigation in larger series of cases.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 3. ß1C mRNA levels and tumor grade and stage in endometrial carcinoma. ß1C mRNA expression was calculated as described in Figure 1 ; the average of ß1C mRNA expression levels in 12 normal proliferative endometrial tissues specimens (NP) was set at 100. ß1C mRNA levels in neoplastic tissues were calculated as percentage of NP. Mean values ± SEM from at least three different experiments are shown. A: ß1C mRNA levels in 6 well-differentiated (G1), 2 moderately differentiated (G2) carcinomas, and 4 poorly differentiated (G3) endometrial carcinomas. B: ß1C mRNA levels in 8 stage I, 1 stage II endometrial carcinoma, 1 stage III, and 2 stage IV endometrial carcinomas (according to International Federation of Gynaecology and Obstetrics).

 
RNA Expression Analysis in Endometrial Hyperplastic Tissues

ß1C mRNA levels were evaluated in a cohort of 17 endometrial hyperplastic samples, seven of which showed simple hyperplasia and the remaining being atypical hyperplasia based on histopathological examination. A statistically significant increase of ß1C steady-state mRNA levels with respect to the controls was detected (T = 3.72, P < 0.001). As illustrated in Figure 4 , the percentile increase of ß1C mRNA ranged from 102% ± 2 to 322% ± 38, with an average value of 157% ± 15. In 12 of the 17 endometrial hyperplastic samples examined, the increased ß1C mRNA levels detected in individual measurements of each case, in comparison with controls, were statistically significant (0.05 < P < 0.001). One of the samples was not suitable for statistical evaluation due to the limited number of measurements available, while the remaining four cases (H4, AH5, H8, and H9) showed decreased (77% ± 2) ß1C mRNA levels in comparison with controls, but the difference was not statistically significant.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 4. ß1C and ß1 mRNA expression in hyperplastic endometrial tissues. ß1C and ß1 mRNA expression was analyzed as described in Figure 1 . The average of either ß1C or ß1 mRNA expression levels in 12 normal proliferative endometrial tissues (NP) was set at 100. ß1C and ß1 mRNA levels in 17 hyperplastic endometrial tissues were calculated as percentage of NP. Mean values ± SEM from at least three different experiments are shown. Increase of ß1C steady-state mRNA level with respect to controls resulted in statistical significance (T = 3.72, P < 0.001). Increase of ß1 mRNA steady-state level with respect to the controls resulted in statistical significance (T = 4.43, P < 0.001).

 
Furthermore, as illustrated in Figure 5 , ß1C mRNA levels were higher in the 7 atypical hyperplastic (AH) samples (191% ± 29 of the controls) than in the 10 samples of simple hyperplasia (H) (134% ± 13 of the controls), (T = 2.05, P = 0.058).



View larger version (7K):
[in this window]
[in a new window]
  		Figure 5. ß1C mRNA level in endometrial hyperplasia (H: simple hyperplastic tissues, 10 cases; AH: atypical hyperplastic tissues, 7 cases) (T = 2.05, P = 0.058). ß1C mRNA expression was calculated as described in Figure 1 : the average of ß1C mRNA expression levels in 12 normal proliferative endometrial tissues specimens (NP) was set at 100. ß1C mRNA levels in hyperplastic tissues were calculated as percentage of NP. Mean values ± SEM from at least three different experiments are shown.

 
As for the carcinoma samples, the full-length ß1 integrin mRNA expression was also measured in the same 17 hyperplastic samples. The ß1 mRNA steady-state levels, with respect to controls, ranged between 71% and 362% with an average value of 179% ± 18; the difference was highly significant (T = 4.43, P < 0.001). As for ß1C, the mean value of ß1 mRNA expression was higher in the atypical hyperplasia subgroup (219% ± 34) than in the simple hyperplasia subgroup (151% ± 15) (T = 2.07, P = 0.055).

Quite obviously, the difference between ß1C and ß1 integrins mRNA expression in carcinomatous and hyperplastic tissues was highly statistically significant (ß1C: T = 4.77, P < 0.0001; ß1: T = 4.03, P < 0.001).

ß1 and ß1C Protein Expression in Normal and Diseased Endometrial Tissues

Seventeen endometrial samples harboring endometrial proliferative diseases were selected for immunoblotting analysis of ß1C and ß1 integrins, the results of which are shown in Figure 6 . Both ß1C and ß1 proteins were expressed in normal endometrial samples (Figure 6A , lanes 1 and 2), in 8 carcinoma samples (lanes 3 to 10), and in 9 hyperplastic tissues (lanes 11 to 19). The values of ß1C and ß1 protein expression were calculated as a percentage of proliferative endometrium, after normalization for ß-tubulin protein expression. The mean value of at least three separate measurements for each patient were recorded and the average ± SEM are reported in Figure 6B .



View larger version (37K):
[in this window]
[in a new window]
  		Figure 6. ß1C and ß1 protein expression in endometrial carcinoma and hyperplasia. A: Normal (lanes 1 and 2), tumor (lanes 3 to 10), and hyperplastic (lanes 11 to 19) tissue detergent extracts were electrophoresed on 7.5% SDS-polyacrylamide gel under reducing conditions, and immunostained using antibody either to ß1C integrin or to ß1 integrin or to ß-tubulin. B: The mean values ± SEM of either ß1C or ß1 levels in neoplastic and hyperplastic tissues, normalized using ß-tubulin, were calculated as a percentage of the levels detected in 7 normal proliferative tissues (NP). At least three separate measurements for each specimen were performed. Down-regulated ß1C protein levels, with respect to the normal endometria, were detected in all pathological samples of this set (tumoral tissues: T = -27.06, P < 0.0001; hyperplastic tissues: T = -10.6, P < 0.0001). Total ß1 protein content expressed in neoplastic endometrial tissues and in hyperplastic tissues with respect to controls were not statistically different (carcinoma tissues: T = -2.04, P = 0.09; hyperplastic tissues: T = -0.93, P = 0.38).

 
Drastically down-regulated ß1C protein levels, with respect to the normal endometria, were detected in all pathological samples of this set (tumoral tissues: T = -27.06, P < 0.0001; hyperplastic tissues: T = -10.6, P < 0.0001). In particular, the percentile expression, with respect to the control, ranged from 4% (±1) to 27% (±5) (mean value: 13% ± 3) in the 8 carcinomatous tissues and from 20% to 73% (mean value: 38% ± 6) in the 9 hyperplastic tissues examined.

In contrast, the total ß1 protein content expressed in neoplastic endometrial tissues and in hyperplastic tissues, with respect to controls, was not statistically different (carcinoma tissues: T = -2.04, P = 0.09; hyperplastic tissues: T = -0.93, P = 0.38).

ß1C Immunohistochemistry

The results of the semiquantitative evaluation of ß1 and ß1C immunoreactivity are schematically illustrated in Figure 7 , and show a strong association between percentage of immunoreactive epithelial cells and the type of lesion (ß1: {chi}2 = 22.61, P = 0.0009; ß1C: {chi}2 = 56.52, P < 0.001); mean values and ANOVA results are reported in Table 2 .



View larger version (22K):
[in this window]
[in a new window]
  		Figure 7. ß1C and ß1 immunoreactivity in endometrial tissues. The results of semiquantitative evaluation of ß1C (A: {chi}2 = 56.52, P < 0.001) and ß1 (B: {chi}2 = 22.61, P = 0.0009) immunoreactive epithelial cells are reported for the endometrial tissues from 14 patients in the proliferative phase (NP), 17 patients affected by endometrial carcinoma (K), and 20 patients affected by endometrial hyperplasia (H).

 

View this table:
[in this window]
[in a new window]
  		Table 2. ß1C and ß1 Immunoreactivity in Endometrial Tissues

 
Overall, ß1C immunoreactivity in epithelial cells was localized in the cytoplasm and showed staining reinforcement at the poles of the cells. Epithelial cells of normal proliferative endometria (Figure 8) demonstrated the highest percentages of ß1C immunoreactive cells in the whole series of cases and in these cases the percentage of ß1C-positive epithelial cells (mean, 74.57%, range = 64% to 90%) was comparable to that of ß1 immunoreactive cells (mean, 73.93%, range = 28% to 91%). Also, ß1C immunoreactivity was comparatively more intense and the number of immunoreactive cells higher in the epithelial compartment, in comparison with the stromal compartment of both normal and diseased endometria (Table 2) .



View larger version (163K):
[in this window]
[in a new window]
  		Figure 8. ß1C integrin immunoreactivity in proliferative endometrium. Most epithelial glandular cells exhibit moderate to strong immunostaining that is mainly localized at the basal and apical pole of the cells. Occasional stromal cells also display weak immunoreactivity. (ABC method anti-ß1C, x200).

 
Reduced ß1C immunoreactivity, in terms of number of immunoreactive epithelial cells, was detected both in endometrial hyperplasia (mean value = 37.5%, range = 21% to 65%) (Figure 9) and in carcinomas (mean value = 16.88%, range 8% to 28%) (Figure 10) and this difference was statistically significant (F = 136.3, P < 0.001); the post hoc test, in fact, detected a significant difference both between hyperplastic and carcinomatous samples and between each of these groups in comparison with normal proliferative endometria.



View larger version (163K):
[in this window]
[in a new window]
  		Figure 9. ß1C integrin immunoreactivity in atypical endometrial hyperplasia. Several epithelial cells demonstrate weak to moderate immunostaining, without apparent intracellular compartmentalization. (ABC method anti-ß1C, x100).

 


View larger version (179K):
[in this window]
[in a new window]
  		Figure 10. ß1C integrin immunoreactivity in endometrioid carcinoma. Groups of neoplastic epithelial cells show moderate to strong intracytoplasmic immunoreactivity. (ABC method anti-ß1C, x100).

 
Interestingly, the highest percentage of ß1C immunoreactive cells was detected in two well-differentiated carcinomas while the lowest percentages were demonstrated in two poorly differentiated cancer cells. Nevertheless, a correlation between ß1C immunoreactivity and tumor differentiation, tumor stage, or histotype, could not be statistically assessed due to the relatively low number of cases.

Stromal cells exhibited an increase of both ß1C and ß1 immunoreactive cells in endometrial carcinoma and hyperplasia while ß1C and ß1 immunoreactive stromal cells were almost undetectable in normal proliferative endometria, but the difference was not statistically significant (F = 0.62, P = 0.54 and F = 2.35, P = 0.11 for ß1C and ß1, respectively). Figure 11 shows positive and negative controls as reported in the Materials and Methods section.


   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
The results of the current study show for the first time that a down-regulation of ß1C integrin-spliced variant expression, both at mRNA and protein level, takes place in endometrial adenocarcinoma in comparison with normal endometrial samples. It should be noted that all endometrial adenocarcinomas analyzed occurred in postmenopausal women; thus, the down-regulation of ß1C integrin expression in these tissues is very significant considering that normally, ß1C integrin expression is maximal in atrophic endometrium compared with proliferative and secretive endometria.13 These data parallel those obtained by our group and by others documenting a down-regulation of ß1C expression in prostate, breast, and lung cancer.12,18,21,22

Interestingly, we documented a significantly up-regulated ß1C mRNA expression in hyperplastic endometrial tissues. These same tissue samples also demonstrated decreased ß1C protein levels, although this decrease was less intense (38% ± 6 of protein expression in comparison with the normal proliferative endometrial tissue) when compared to the carcinomatous tissues (13% ± 3 of protein expression). These results are further supported by those obtained by immunohistochemistry, documenting reduced percentages of immunoreactive epithelial cells in endometrial hyperplasia and even more reduced percentages in endometrial carcinomas, in comparison with normal proliferative tissues.

Since ß1C is one of the four known ß1 variants, representing a spliced variant of the full-length ß1 integrin molecule, one might argue that variations of ß1C protein expression might recapitulate modified expression of the full-length protein. Therefore, we also measured full-length ß1 integrin expression in the same series of tissues and found no significant alterations in neoplastic and hyperplastic samples, in comparison with controls.

Interestingly, we observed decreased ß1C integrin mRNA expression in the carcinoma tissues, indicating that the control of ß1C integrin expression might be regulated at a transcriptional and/or posttranscriptional level in such a fashion as to maintain selectively reduced mRNA and/or protein levels of ß1C. In contrast, ß1 protein levels, which are likely to reflect ß1A levels, although decreased at the mRNA level, may remain constant in normal and neoplastic tissues by compensatory mechanisms. In fact, considering that ß1B and ß1D variants are unlikely to be expressed in endometrial tissues, since ß1B expression is restricted to skin and liver tissues, whereas ß1D is detectable in striated muscle cells only,23,24 the contribution of ß1A variant is essential for the evaluation of the overall ß1 integrin expression.

The up-regulated ß1C mRNA expression, despite decreased ß1C immunoreactivity observed in endometrial hyperplasia, however, suggests that the synthesis of ß1C by epithelial cells might be regulated at a posttranscriptional level.

On the other hand, ß1C immunoreactivity may be more closely related to protein stability than to transcriptional activity; in fact, processing of heterologous nuclear RNA, modulation of the nucleo-cytoplasmic transport of mature mRNA and changes in mRNA stability may regulate the posttranscriptional control of mRNA,26 and raised steady-state protein levels can occur without parallel increase of transcriptional activity. Furthermore, prolonged protein storage may justify increased ß1C immunoreactivity notwithstanding decreased mRNA.

Steady-state expression of mRNA is controlled by both transcriptional and posttranscriptional processes; moreover, protein levels may not accurately reflect the expression of their respective mRNA. Altered mRNA translation can regulate the expression of heat-shock proteins, membrane immunoglobulins, and other proteins, as well as cellular proteins in vesicular stomatitis virus-infected cells. In such instances, either the existence of specific inhibitors of mRNA translation or the lack of components of the translational machinery has been implicated.25

On these premises and based on the results of the current study, we can postulate that ß1C integrin expression may be down-regulated at the level of protein synthesis, by reducing the translation efficiency, as supported by the detection of almost equal mRNA amounts in carcinoma and in normal endometria tissues.

The study of adhesion molecule expression in human tissues and cancer cell lines may be relevant for better understanding the role of these factors in cell cycle regulation and in tumor genesis and progression and, indeed, the adhesion pathway is very important in moderating neoplastic progression and aggressiveness.27 Alterations of adhesion molecules seem to commonly occur in tumors as they progress to malignancy, as demonstrated by reduced tumorigenicity and motility in several different cells overexpressing {alpha}5ß1.28 In addition, stepwise loss of integrins, such as {alpha}1, {alpha}4, and {alpha}vß3, with increasing tumor grade has been shown.29 In this regard, the results of the current study document progressively decreased ß1C expression, at protein level, in endometrial hyperplasia, atypical hyperplasia, and carcinomas, with the exception of increased ß1C mRNA levels in hyperplastic tissues. These differences, in comparison with normal proliferative endometria, suggest that the inhibitory effects of this integrin on cell proliferation are increasingly reduced or lost with the progression from low proliferative toward highly proliferative hyperplastic lesions and finally, to malignant lesions. Consequently, a pivotal role for ß1C can be postulated in the progression from hyperproliferative conditions, such as simple endometrial hyperplasia, to pre-cancerous lesions, such as atypical hyperplasia, and to invasive carcinoma. Possibly, the active role of ß1C might be exerted on tumor differentiation as well since we documented a trend for a direct correlation between ß1C expression and the grade of endometrial carcinoma. The relative rarity of atypical hyperplastic lesions and the necessity of examining both cryopreserved and formalin-fixed tissues did not allow the inclusion of a large number of these cases in the present study, thus making it unsuitable to perform adequate statistical evaluation. Nevertheless, these preliminary results may be an incentive to further investigate this issue in larger series of cases.

In consideration of the counteracting roles of ß1 and ß1C in cell cycle regulation, the results of the current study add further insight for the understanding of the molecular control of cell growth and the mechanisms of intracellular signal transduction. Although the regulatory mechanisms in endometrial tissues is a very complex scenario with multiple controllers, the present demonstration of a key role of the ß1 and ß1C integrin molecules may be critical to devise novel effective diagnostic and/or prognostic approaches.


   Acknowledgements
 
We thank Giuseppe Sgaramella for technical assistance, Vito Cataldo for photographic assistance, and Dr. S. J. Reshkin for helpful and stimulating discussion.


   Footnotes
 
Address reprint requests to Dr. Elda Perlino, Institute of Bioenergetics and Biomembranes, C.N.R., Bari, Via Amendola 165/A, I-70126, Bari, Italy. E-mail: perlino{at}area.ba.cnr.it

Supported in part by a grant from ASI (ARS-99–77 to E. P.).

Accepted for publication August 7, 2003.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Boyd J: Molecular genetics of human endometrial carcinoma. Khan SA Stancel GM eds. Protooncogenes and Growth Factors in Steroid Hormone Induced Growth and Differentiation. 1994:pp 193-205 CRC Press, Boca Raton
  2. Fornaro M, Languino LR: Alternatively spliced variants: a new view of integrin cytoplasmic domain. Matrix Biology 1997, 16:185-193
  3. Hemler ME, Weitzman JB, Pasqualini R, Kawaguchi S, Kassner PD, Berdichevsky FB: Structure, biochemical properties, and biological functions of integrin cytoplasmic domains. Takada Y eds. Integrins: The Biological Problems. 1995:pp 1-35 CRC Press Inc., Boca Raton
  4. Fornaro M, Tallini G, Zheng DQ, Flanagan WM, Manzotti M, Languino LR: p27Kip1 acts as a downstream effector of and is coexpressed with ß1C integrin in prostatic adenocarcinoma. J Clin Invest 1999, 103:321-329[Medline]
  5. Witkowski M, Rabinovitz I, Nagle RB, Affinito KD, Cress AE: Characterization of integrin subunits, cellular adhesion and tumorigenicity of four human prostate cell lines. Cancer Res Clin Oncol 1993, 119:637-644
  6. Albelda SM: Biology of disease: role of integrins and other cell adhesion molecules in tumor progression and metastasis. Lab Invest 1993, 68:4-17[Medline]
  7. Lessey BA, Castelbaum A, Buck CA, Ley Y, Yowell CW, Sun J: Further characterization of endometrial integrins during menstrual cycle and pregnancy. Fertil Steril 1994, 62:497-506[Medline]
  8. Languino LR, Ruoslahti E: An alternative form of the integrin ß1 subunit with a variant cytoplasmic domain. J Biol Chem 1992, 267:7116-7120[Abstract/Free Full Text]
  9. Fornaro M, Zheng DQ, Languino LR: The novel structural motif Gln795- Gln802 in the integrin ß1C cytoplasmic domain regulates cell proliferation. J Biol Chem 1995, 270:24666-24669[Abstract/Free Full Text]
  10. Meredith J, Jr, Takada Y, Fornaro M, Languino LR, Schwartz MA: Inhibition of cell cycle progression by the alternatively spliced integrin ß1C. Science 1995, 269:1570-1572[Abstract/Free Full Text]
  11. Fornaro M, Manzotti M, Tallini G, Slear AE, Bosari S, Ruoslahti E, Languino LR: ß1C integrin in epithelial cells correlates with a nonproliferative phenotype: forced expression of ß1C inhibits prostate epithelial cell proliferation. Am J Pathol 1998, 153:1079-1087[Abstract/Free Full Text]
  12. Perlino E, Lovecchio M, Vacca RA, Fornaro N, Moro L, Ditonno P, Battaglia M, Selvaggi FP, Mastropasqua MG, Bufo P, Languino LR: Regulation of mRNA and protein levels of ß1 integrin variants in human prostate carcinoma. Am J Pathol 2000, 157:1727-1734[Abstract/Free Full Text]
  13. Vacca RA, Marra E, Loverro G, Maiorano E, Napoli A, Lovecchio M, Selvaggi L: Differential expression of ß1C integrin messenger ribonucleic acid and protein levels in human endometrium and decidua during the menstrual cycle and pregnancy. J Clin Endocrinol Metab 2003, 88:720-729[Abstract/Free Full Text]
  14. Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ: Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 1979, 18:5294-5299[Medline]
  15. Sambrook J, Fristch EF, Maniatis T: Transfer of DNA to nylon membranes. Molecular Cloning: A Laboratory Manual. 1989:p 7.49 Cold Spring Harbor Laboratory Press, Cold Spring Harbor
  16. Feinberg AP, Vogelstein B: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 1983, 132:6-13[Medline]
  17. Bhatia P, Taylor WR, Greenberg AH, Wright JA: Comparison of glyceraldehyde-3-phosphate dehydrogenase and 28S-ribosomal RNA gene expression as RNA loading controls for Northern blot analysis of cell lines of varying malignant potential. Anal Biochem 1994, 216:223-226[Medline]
  18. Fornaro M, Tallini G, Bofetiado CJM, Bosari S, Languino LR: Down-regulation of ß1C integrin, an inhibitor of cell proliferation, in prostatic carcinoma. Am J Pathol 1996, 149:765-773[Abstract]
  19. Feuerstein N, Cooper HL: Studies of differentiation of promyelocitic cells by phorbol ester. II. A methylation inhibitor, 3-deazaadenosine, inhibits the induction of specific differentiation proteins: lack of effect on early and late phosphorylation events. Biochem Biophys Acta 1984, 781:247-256[Medline]
  20. Ronnett BM, Kurman RJ: Precursor lesions of endometrial carcinoma. Kurman RJ eds. Blaustein’s Pathology of the Female Genital Tract. 5th ed. 2001:pp 467-484 New York, NY, Springer-Verlag
  21. Manzotti M, Dell’Orto P, Maissoneuve P, Fornaro M, Languino LR, Viale G: Down-regulation of ß1C integrin in breast carcinomas correlates with high proliferative fraction, high histological grade, and large size. Am J Pathol 2000, 156:169-174[Abstract/Free Full Text]
  22. Patriarca C, Alfano RM, Sonnenberg A, Graziani D, Cassani B, de Melker A, Colombo P, Languino LR, Fornaro M, Warren WH, Coggi G, Gould VE: Integrin laminin receptor profile of pulmonary squamous cell and adenocarcinomas. Hum Pathol 1998, 29:1208-1215[Medline]
  23. Balzac F, Belkin AM, Koteliansky VE, Balabanov YV, Altruda F, Silengo L, Tarone G: Expression and functional analysis of a cytoplasmic domain variant of the ß1 integrin subunit. J Cell Biol 1993, 121:171-178[Abstract/Free Full Text]
  24. Belkin AM, Zhidkova NI, Balzac F, Altruda F, Tomatis D, Maier A, Tarone G, Kotelianski VE, Burridge K: ß1D integrin displaces the ß1A isoform in striated muscles: localization at junctional structures and signaling potential nonmuscle cells. J Cell Biol 1996, 132:211-226[Abstract/Free Full Text]
  25. Defilippi P, Truffa G, Stefanuto G, Altruda F, Silengo L, Tarone G: Tumor necrosis factor {alpha} and interferon {gamma} modulate the expression of vitronectin receptor (integrin ß3) in human endothelial cells. J Biol Chem 1991, 266:7638-7645[Abstract/Free Full Text]
  26. Kren BT, Kumar NM, Wang S, Gilula NB, Steer CJ: Differential regulation of multiple gap junction transcripts and protein during rat liver regeneration. J Cell Biol 1993, 123:707-718[Abstract/Free Full Text]
  27. Tisty TD: Cell-adhesion-dependent influences on genomic instability and carcinogenesis. Curr Opin Cell Biol 1998, 10:647-653[Medline]
  28. Saunders RJ, Mainiero F, Giancotti FG: The role of integrins in tumorigenesis and metastasis. Cancer Invest 1998, 16:329-344[Medline]
  29. Lessey BA, Albelda S, Buck CA, Castelbaum AJ, Yeh I, Kohler M, Berchuck A: Distribution of integrin cell adhesion molecules in endometrial cancer. Am J Pathol 1995, 146:717-726[Abstract]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
J. P. Venables
Aberrant and Alternative Splicing in Cancer
Cancer Res., November 1, 2004; 64(21): 7647 - 7654.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lovecchio, M.
Right arrow Articles by Perlino, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lovecchio, M.
Right arrow Articles by Perlino, E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS