mhrep$0405
Molecular Human Reproduction vol.3 no.4 pp. 333–342, 1997
The progressive rise in the expression of α crystallin B chain in
human endometrium is initiated during the implantation window:
modulation of gene expression by steroid hormones
M.Gruidl1, A.Buyuksal1, A.Babaknia2, A.T.Fazleabas3, S.Sivarajah4, P.G.Satyaswaroop4 and
S.Tabibzadeh1,5
1Department of Pathology, Moffitt Cancer Center and the University
FL 33612, 2Department of OB/GYN, Women’s Health Institute, Costa
of South Florida, 12902 Magnolia Drive, Tampa,
Mesa, CA, 3Department of Obstetrics and
Gynecology, University of Illinois, College of Medicine at Chicago, Chicago, IL 60680, and 4Department of OB/GYN
Hershey Medical Center, Hershey, PA 17033, USA
5To
whom correspondence should be addressed
Human endometrium undergoes sequential changes during the menstrual cycle and becomes receptive to
implantation during a defined period in the secretory phase. We attempted to identify the genes expressed
during this period by representational difference analysis (RDA). When the cDNAs of a proliferative endometrium were used as the driver and the cDNAs of a post-ovulatory day 5 endometrium were used as the
tester, a number of bands were identified by RDA. DNA of the cloned RDA products revealed that the majority
of the clones contained a fragment of a cDNA identical to that of a crystallin B chain. Northern blot analysis
showed that the expression of the α crystallin B chain mRNA was absent during the proliferative phase. The
expression of the mRNA of α crystallin B chain first appeared in the secretory phase, progressively increased
during this phase and peaked in the late secretory endometria. The pattern of expression of α crystallin B
chain mRNA in the endometrium of mature cycling baboons (Papio anubis) was similar to that seen in human
endometrium. As revealed by Western blot analysis, the expression of the α crystallin B chain protein in
human endometrium followed a pattern of expression similar to its mRNA. At the cellular level, the
immunoreactive protein first appeared in the surface epithelial cells of human endometrium within the
implantation window without significant immunoreactivity in the underlying glandular cells. During the midand late secretory phases, the intensity of staining in the epithelial cells was enhanced and an intense
immunoreactivity was developed in the glandular epithelium. α crystallin B chain was virtually an epithelial
product and no immunoreactivity for this protein was detectable in the stromal cells, endothelial cells or
lymphoid cells. The expression of α crystallin B chain could be regulated, by medroxy progesterone acetate
as well as by oestrogen withdrawal, in human endometrial carcinoma cells (EnCa-101), transplanted to nude
mice. Based on the data presented here, the known function of α crystallin B chain and its distinct pattern
of expression in human endometrium, we suggest that this protein is an important factor within the molecular
repertoire that makes endometrium receptive to implantation.
Key words: crystallin/endometrium/implantation/oestrogen/progesterone
Introduction
Human endometrium undergoes characteristic phases of
proliferation and secretion. Highly coordinated and exquisite
mechanisms are involved in the processes that drive endometrium through these phases and prepare this tissue for
implantation (for a recent review, see Tabibzadeh and
Babaknia, 1995). The presence of endometrial receptivity
was first established in the rat and later validated in other
species (Psychoyos, 1973a,b, 1976, 1986, 1993). These
studies show that, with respect to implantation, endometrium
can be maintained in various states which include neutral,
receptive and non-receptive or refractory phases. It is
postulated that, in humans, an ‘implantation window’ also
exists, during which the endometrium becomes receptive
to implantation by the blastocyst (for a recent review see
Tabibzadeh and Babaknia, 1995). This phase is followed by a
‘non-receptive’ phase when the endometrium becomes refrac© European Society for Human Reproduction and Embryology
tory to the implantation process (Strauss and Gurpide, 1991;
Psychoyos, 1993). There is no general agreement as to the
dates and duration of such an ‘implantation window’. For
example, it was suggested that the implantation window is
confined to the post-ovulatory days (POD) 5–7 of the normal
menstrual cycle (Psychoyos, 1993). Rogers and Murphy (1989),
however, concluded that the human implantation window must
be at least 3.5 days, whereas Formigli et al. (1987) suggested
that the period of endometrial receptivity may be as long as 7
days. Based on the available data, it can be concluded
that the endometrial ‘implantation window’ in humans opens
several days after ovulation and closes several days prior to
menstruation.
So far, little information exists in humans about the genes
whose expression in endometrium is essential in the dialogue
established between the endometrium and the blastocyst. Most
of our current knowledge regarding the expression of such
333
M.Gruidl et al.
genes has been derived from animal studies. Based on animal
and human studies, the molecular repertoire that participates
in the process of implantation seems to consist of diverse
molecules including cytokines, hormones, and adhesion
molecules (Baker et al., 1992; Stewart et al., 1992; Tabibzadeh
et al., 1992, 1995a; Lessey et al., 1992; 1994; De et al., 1993;
Das et al., 1994; Julian et al., 1994; Ding et al., 1994; Das
et al., 1995; Fukuda et al., 1995). Among these in the mouse
are: interleukin (IL)-1, IL-6 and tumour necrosis factor
(TNF)-α (De et al., 1993); leukaemia inhibitory factor (LIF)
(Stewart et al., 1992); heparin binding-epidermal growth factor
(HB-EGF) (Das et al., 1994); tenascin (Julian et al., 1994);
amphiregulin (Das et al., 1995); Fos (Baker et al., 1992); and
calcitonin (Ding et al., 1994). We may not infer from animal
studies, however, that in humans, the same set of endometrial
genes are implicated in the endometrial receptivity and
implantation. For example, a transient surge in the expression
of amphiregulin mRNA occurred throughout the uterine
epithelium on day 4 of pregnancy in the mouse. With the
onset of mouse blastocyst attachment late on day 4, the
mRNA of amphiregulin accumulated in the luminal epithelium
exclusively at the site of blastocyst attachment (Das et al.,
1995). On the other hand, throughout the entire pregnancy in
humans, no immunoreactivity for amphiregulin was detectable
in the gestational endometria (Lysiak et al., 1995). The animal
studies, however, suggest that implantation is a dynamic
process requiring the expression of specific genes in endometrium whose presence may be essential or at least contributory to the implantation process. Given that putative
‘endometrial receptivity genes’ should be expressed within
implantation window, we attempted to unravel their identity
by comparing and identifying the mRNAs that are preferentially expressed in human endometrium during such period.
Our search identified several genes whose expression was
differentially regulated during the menstrual cycle. A major
gene identified was the α-crystallin B chain. Additional studies
are reported here which validated that the mRNA for the αcrystallin B chain and its product is expressed in the epithelial
cells of human endometrium solely in the secretory phase of
the menstrual cycle. The expression of the α-crystallin B
chain was found to be regulated by medroxy-progesterone
acetate (MPA).
Materials and methods
Materials
A polyclonal antibody raised against a synthetic peptide–haemocyanin
conjugate containing the amino acids 1–10 from the amino terminal
region of α crystallin B chain was obtained from Novacastra Laboratories (Newcastle, UK). A second polyclonal rabbit antibody raised
against the C-terminal domain of α crystallin B chain which was
used as an anti-serum or as affinity purified antibody was obtained
as a gift from Dr J.Horwitz (UCLA School of Medicine, CA, USA).
The avidin–biotin–complex (ABC) kit and a kit containing nitroblue
tetrazolium salt (NBT) and 5-bromo-4-chloro-3-indolyl phosphate,
toluidinium salt (BCIP) in dimethyl formamide were obtained from
Vector Laboratories (Burlingame, CA, USA). Except where indicated,
all other chemicals were obtained from Sigma Chemical Co. (St
Louis, MO, USA) or Fisher Scientific (Pittsburgh, PA, USA).
334
Processing of endometria
Endometrial tissues were obtained by biopsy or by curettage or from
hysterectomy specimens of normal fertile women, aged 25–45 years,
who underwent these procedures during investigation of nonendometrial abnormalities such as ovarian or cervical lesions. Women
were healthy and had no prior history of disease. Hysterectomy
specimens and endometrial biopsy samples were processed rapidly.
The date of endometrium was determined from the morphological
evaluation of endometrial sections using established criteria of Noyes
and Hertig (1950). Endometria were removed from mature cycling
baboons (Papio anubis) during the early proliferative, midproliferative, and mid-secretory phases and on the first day of
menstruation as described (Fazleabas et al., 1988). Each endometrial
sample was aliquoted as required. However, most tissues were used
as follows. Approximately 10% of the sample was processed for
paraffin sectioning and morphological examination. A further 70%
was flash frozen in a dry ice/ethanol bath for isolation of RNA or
protein and the remaining 20% was frozen in OCT mounting medium
(Tissue-Tek II, Miles Laboratories, Naperville, IL, USA) for cryostat
sectioning.
Regulation of α crystallin B chain in EnCa-101 tumour
grown in nude mice
Athymic Balb/C, nu/nu mice (Harlan Sprague–Dawley Inc.,
Indianapolis, USA) were castrated 1 week before tumour transplantation, and maintained in separate barrier facilities used exclusively for these animals. EnCa-101, a histologically well-differentiated
and oestrogen receptor-positive human endometrial carcinoma grown
s.c. in nude mice, was used in these studies. EnCa-101 exhibits
characteristics similar to the normal endometrial epithelium with
regard to its responses to steroids, oestrogen and progesterone.
Oestrogen enhances growth and the concentration of progesterone
receptors, and progestin inhibits growth, of these epithelial tumour
cells (Satyaswaroop et al., 1983; Clarke et al., 1987). Tumour tissue
(~20 mg) was transplanted s.c. at each of four sites (infrascapular
and lumbar regions of each flank) in castrated nude mice. Animals
were implanted s.c. with 17β-oestradiol pellets (Innovative Research
of America, Sarasota, FL, USA). These pellets are designed to
maintain the serum concentration of oestrogen at 200 pg/ ml for 60
days. Pellets were replaced at 55–60 day intervals. When the geometric
mean diameter of tumours reached 100 mm, 1, 2 or 5 mg medroxyprogesterone acetate (MPA) was administered to various sets of animals.
The serum MPA concentrations were determined by radioimmunoassay. One tumour was excised under ketamine–xylazine anaesthesia
before administration of MPA (time 0). Other tumours were removed
at various times as indicated. Tumours were frozen, pulverized and
stored individually in liquid nitrogen for Northern and Western blot
analysis of α crystallin B chain mRNA and protein. At each time
point, blood was removed from each animal by heart puncture for
determination of serum concentrations of MPA.
Representational difference analysis (RDA): cloning,
sequencing, and analysis of DNA sequences
RDA is a relatively new technique which utilizes subtractive hybridization and polymerase chain reaction (PCR) to amplify the differences
between two similar but slightly different cDNAs. The procedure
was performed as described (Hubank and Schatz, 1994). Briefly, two
different sets of cDNAs were prepared. After digestion with a
restriction enzyme (DpnII), cDNA fragments were ligated to linkers
to form the primer annealing sites for a subsequent round of PCR
amplification. The restriction enzyme-digested PCR-amplified cDNA
fragments are called ‘representations’. One pool of these representations from a mid-proliferative endometrium was used in a large molar
Expression of α crystallin B chain in human endometrium
excess (driver) relative to the other cDNA pool from a post-ovulatory
day 5 endometrium (tester). Linkers were removed by a restriction
enzyme digestion from both driver and tester representations and a
new set of linkers were ligated only to the tester representation. These
were used in a hybridization procedure that allowed sequences
common in both pools to hybridize. After hybridization, linkers at
both ends of the cDNA fragment(s) provided a site for primer
annealing in a subsequent PCR. The cDNAs from the tester, bound
to the driver cDNAs, were linearly and asymmetrically amplified.
The cDNAs unique to the driver whose linkers were removed by the
restriction enzyme digestion were not amplifiable. After the PCR
protocol, products which were linearly and asymmetrically amplified
were removed by mung bean nuclease digestion. Several rounds of
hybridization–subtraction and amplification with different linkers
ligated to representation of the tester were performed to remove
cDNAs which were present in both pools. Each successive round was
done with decreasing amounts of tester which effectively increased
the stringency of the subtraction. This allowed the differentially
expressed genes in the tester mRNAs to be identified. After the final
round of subtractive hybridization and amplification, the PCR products
were digested to remove the linkers. These products were size
fractionated by agarose gel electrophoresis. The differentially
expressed bands identified from RDA were isolated from agarose gel
slices (Qiaex II, Qiagen, Chatsworth, CA, USA). These were cloned
into a general cloning/sequencing vector after digestion with BamH1
(pBluescript II, Stratagene, La Jolla, CA, USA). The products
were sequenced with Sequenase ver. 2.0 (Amersham, Life Sciences,
Arlington, IL, USA) using the dideoxy chain termination method
(Sanger et al., 1977). For identification of the isolated clones, the
DNA sequences were compared with those deposited in the nucleotide
databases (GenBank and EMBL).
Isolation of RNA and Northern blotting
The RNA was extracted using the acid guanidinium thiocyanate–
phenol–chloroform extraction method as described by Chomczynski
and Sacchi (1987). The amount of RNA was assessed spectrophotometrically and the quality of RNA was assessed by evaluating the
integrity of ribosomal RNA by electrophoresis of 20 µg of total RNA
in 1% formaldehyde–agarose gel in the presence of ethidium bromide.
Poly-A RNA was purified by affinity chromatography using oligo
(dT) cellulose. Northern blotting was performed as described by
Sambrook et al. (1989) and Pampfer et al. (1991). Briefly, 20 µg of
total RNA was electrophoresed in a 1.4% agarose [3-[N-morpholino]propane sulphonic acid (MOPS)–formaldehyde gel (Sambrook et al.,
1989), transferred to nylon membrane (Hybond N; Amersham) by
standard capillary transfer in 103 sodium chloride/sodium citrate
(SSC). The RNA was fixed to the membrane by UV crosslinking in
UV Stratalinker™ (Stratagene). [32P]-labelled α crystallin B chain
(266 bp) cDNA fragment isolated by RDA, glyceraldehyde 3 phosphate dehydrogenase (GAPDH) (1.2 kb EcoRI fragment from ATCC
clone HHCMC23) full-length cDNA and clone 1a cDNA (1.9 kb)
fragment, labelled to a high specific activity, were prepared by random
primer labelling and purified by gel filtration. Prehybridization,
hybridization and post-hybridization washes were performed as
described by Church and Gilbert (1984). The membranes were
subjected to autoradiography at –70°C with intensifying screens. For
reprobing the membrane for GAPDH or clone 1a, the α crystallin B
chain probe was removed by incubating the membrane at 50°C in
25 ml of 75% formamide containing 0.13 SSC and 0.2% sodium
dodecyl sulphate (SDS). The relative abundance of α crystallin B
chain mRNA compared with the housekeeping genes GAPDH and
clone 1a was determined by normalizing the optical densities of α
crystallin B chain bands as follows: the autoradiograms were
Figure 1. Northern blot analysis of α crystallin B chain mRNA in
human endometrium throughout the menstrual cycle. 20 µg of total
RNA isolated from each of endometria dated to mid-proliferative
(MP), late proliferative (LP), and post-ovulatory days (P) 5, 9, and
14 were subjected to the Northern blot analysis. The blot was
probed for α crystallin B chain mRNA and then for glyceraldehyde
3 phosphate dehydrogenase (GAPDH) mRNA. Upper panel:
autoradiograms. Lower panel: mean relative optical density ratio
calculated as described in the Materials and methods section. The
size of α-crystallin B chain mRNA is 950 bp.
scanned by a laser scanning densitometer (Envisions, DynamicPro
30, Burlingame, CA, USA). Then, the optical densities of the bands
were quantified using SigmaGel (Jandel Scientific, San Rafael, CA,
USA). Three independent measurements were made from each band
and the mean values were calculated. The mean optical density values
of the α crystallin B chain were then divided by the mean optical
density values of the GAPDH (Tso et al., 1985; Zentella et al., 1991)
or clone 1a (Hsu et al., 1988) from the same sample. The largest
value was regarded as 100% (Figures 1, 2 and 6) and all other values
were calculated as percentages of this value, giving a mean relative
optical density ratio.
Western blot analysis
For solubilizing proteins, tissues were added to an equal volume of
23 SDS lysis buffer (6% SDS, 0.14 M Tris, pH 6.8, 22.4%
glycerol) and the chromosomal DNA was sheared by repeatedly
passing the sample through a 20 gauge needle and then through a 26
gauge needle. The samples were spun at 10 000 3 g for 10 min and
the amount of protein in the supernatants was determined using the
biconchononic acid (BCA) assay kit (Pierce, Rockford, IL, USA).
Then, mercaptoethanol and bromophenol blue were added to a final
concentration of 5 and 0.5% respectively, and the samples were
boiled for 5 min. Tissue lysates were subjected to SDS–PAGE and
separated proteins were transferred to nitrocellulose membranes
(NitroBind, MSI, Westboro, MA, USA). Blots were preblocked by
incubation in TBST (10 mM Tris, pH 8.0; 150 mM NaCl; 0.05%
Tween-20) containing 3% bovine serum albumin (BSA) at 25°C for
2 h. After washing in TBST (34), blots were stained using the
avidin–biotin–peroxidase complex (ABC) procedure (Hsu et al.,
1981). This was performed by sequential incubation of the blot, with
TBST containing 1% BSA and primary antibody diluted in TBST
335
M.Gruidl et al.
Figure 2. Northern blot analysis of α crystallin B chain mRNA in
baboon endometrium throughout the menstrual cycle. 20 µg of total
RNA isolated from each of endometria dated to early proliferative
(EP), mid-proliferative (MP), mid-secretory (MS) phases and day 1
of menstruation (MENS) were subjected to Northern blot analysis.
The blot was probed for α crystallin B chain mRNA and then for
glyceraldehyde 3 phosphate dehydrogenase (GAPDH) mRNA.
Upper panel: autoradiograms. Lower panel: mean relative optical
density ratio calculated as described in the Materials and methods
section. The size of α-crystallin B chain mRNA is 950 bp.
containing 1% BSA (2–12 h), and then with secondary antibody
diluted in TBST containing 1% BSA (2 h), and finally with ABC
(2 h). Each incubation was at room temperature and was followed
by two washes in TBST. The immunoreactive band was revealed
by incubation of the blots with a mixture of 3,39 diaminobenzidine tetrahydrochloride (DAB)–H2O2. In some experiments, the
immunoreactive protien was detected by using alkaline phosphataselabelled secondary antibody followed by staining in a mixture of
NBT and BCIP. As controls, primary antibody, secondary antibody
or ABC were omitted from the staining reaction. Laser densitometric
analysis was performed as described for the Northern blot analysis.
Immunohistochemical staining
Frozen sections were fixed in 10% buffered formalin for 5 min and
then washed in 0.1 M phosphate-buffered saline (PBS). Immunostaining was performed according to the ABC procedure (Hsu et al.,
1981) and as described above using the same controls. Sections were
viewed at the light microscopic level without counterstain.
Results
To identify the mRNAs that are preferentially expressed in the
secretory endometrium during the implantation window, we
used RDA. The cDNAs synthesized from mRNA isolated from
a mid-proliferative endometrium were used as the driver and
the cDNAs synthesized from mRNAs isolated from a postovulatory day 5 endometrium were used as the tester. A major
RDA product was cloned and 5 independent clones were used
for sequencing. The sequence of a 266 bp clone showed
identity to the published sequence of α crystallin B chain. To
336
verify that the α crystallin B chain is differentially expressed
in endometrium, the RNAs isolated from normal human
endometria were subjected to Northern blot analysis. The
mRNA of the α crystallin B chain was not detected in any of
the proliferative endometria. On the other hand, α crystallin
B chain mRNA expression appeared in the early secretory
endometria and peaked in the late secretory endometria (Figure
1). Probing three additional panels of endometria showed
similar results and confirmed these findings (data not shown).
The possibility existed that such a differential gene expression
could either be unique to human endometrium or could possibly
be present in cycling endometria of other species. Therefore,
we examined the expression of the mRNA of α crystallin B
chain in endometria of mature baboons (Papio anubis) whose
endometrium undergoes sequential phases of proliferation,
secretion and menstrual shedding very much similar to human
endometria (Fazleabas et al., 1988). The expression of α
crystallin B chain was absent in the baboon endometria in the
early and mid-proliferative phase with a progressive increased
expression in the mid-secretory and menstrual phases.
(Figure 2).
To examine whether the α crystallin B chain protein is
present within endometrium and whether its expression follows
the pattern of its mRNA expression, Western blot analysis was
performed on endometrial proteins. Probing the blots from a
panel of endometria with the protein-specific antibody showed
lack of detectable immunoreactive proteins during the proliferative phase (Figure 3A). On the other hand, an immunoreactive
protein with a molecular weight consistent with that reported
for α crystallin B chain appeared in the endometrium during
the early secretory phase (Figure 3A). The relative abundance
of the immunoreactive band progressively increased during
the secretory phase and peaked in the late secretory endometria
(Figure 3A). These findings were confirmed in two additional
panels of endometria. Furthermore, a different affinity-purified
antibody to α crystallin B chain gave a similar result (Figure
3B).
To localize the α crystallin B chain at the cellular level
in human endometrium, we used two different polyclonal
antibodies for immunohistochemical staining. The data
obtained from both antibodies were similar. There was no
detectable immunoreactive product in the proliferative endometria (Figure 4A,B). On the other hand, immunoreactive α
crystallin B chain appeared in the early secretory endometria.
An intense amount of immunoreactivity appeared first in the
surface epithelium (Figure 5C,D). The glandular epithelium
immediately beneath the surface epithelium showed immunoreactivity in glandular cells at this phase of the menstrual
cycle (Figure 5B). Within these glands, only some glandular
cells exhibited the staining whereas other cells remained nonreactive (Figure 5B). On the other hand, some glands in the
functionalis layer and all the glands in the basalis layer were
devoid of any detectable staining (Figure 5A). The intensity
of staining in the glandular epithelial cells progressively
increased toward the late secretory phase (Figure 5E,F). In
this phase of the menstrual cycle, virtually all the glands in
the upper functionalis as well as the surface epithelium
exhibited intense staining for α crystallin B chain protein
Expression of α crystallin B chain in human endometrium
Figure 3. Western blot analysis of the α crystallin B chain in human endometrium throughout the menstrual cycle. 20 mg of total
endometrial proteins isolated from each of the endometria dated to mid-proliferative (MP), late proliferative (LP), post-ovulatory days
(POD) 2, 3, 4, 5, 6, 7, 11, 13, 14 and HL60 (positive control) were subjected to Western blot analysis. The blots were probed with (A)
polyclonal (Novacastra) and (B) purified polyclonal (Dr Horwitz) antibodies to α crystallin B chain as described in the text. Upper half of
each panel: Western blot of human endometrial proteins probed for α crystallin B chain. Lower half of each panel: mean relative optical
densities of the Western blot bands. The polyclonal antibody (A) revealed a major band with a molecular weight of ~22 kDa. The major
band detectable with the purified polyclonal antibody (B) to the α crystallin B chain also had an approximate molecular weight of 22 kDa.
Figure 4. Immunohistochemical staining of α crystallin B chain in human endometrium in the mid-proliferative phase. The cryostat sections
of a mid-proliferative human endometrium were stained immunohistochemically, as described in the text using the purified polyclonal
antibody to α crystallin B chain. (A) Original magnification (370); (B) original magnification (3280).
(Figure 5E,F). In the same endometrium, however, the basalis
glands were not stained. During the menstrual phase, the
staining was confined to the glands in the menstrual endometrium and was seen in the luminal border of the epithelial
cells (Figure 5G,H). Throughout the entire secretory phase,
the staining for α crystallin B chain was present only in the
cytoplasm and nuclear staining was not detected. In addition,
immunoreactivity for the α crystallin B chain protein was
purely epithelial and other cells in the stroma including the
stromal cells, lymphoid cells, and endothelial cells, failed to
exhibit a detectable amount of immunoreactivity of this protein.
However, the smooth muscle cells around the small and
mid-size arteries showed a weak immunoreactivity in the
proliferative phase. The number of these cells progressively
increased during the secretory phase (data not shown).
Omitting the primary antibody in the immunostaining procedure produced no staining in the tissue sections (Figure 5I,J).
Expression of α crystallin B chain during the secretory
phase suggested that progesterone may be the prime signal
for induction of this expression. We tested this hypothesis in
an in-vivo steroid-sensitive experimental model of human
endometrial carcinoma, EnCa-101 (Satyaswaroop et al., 1983;
Clarke et al., 1987). Similar to normal endometrium, these
tumours exhibit responsiveness to oestrogen and progesterone. Oestrogen stimulates both growth and an increase in
progesterone receptor concentrations whereas progesterone
inhibits growth and induces secretory differentiation
Satyaswaroop et al., 1983; Clarke et al., 1987). Tumours
grown in the presence of oestrogen expressed α crystallin B
chain mRNA at a low level (Figure 6). A single administration
337
M.Gruidl et al.
Figure 5. Immunohistochemical staining of α crystallin B chain in human endometrium in the secretory phase. The cryostat sections of
human endometria were stained immunohistochemically, as described in the text using the purified polyclonal antibody to α crystallin B
chain. (A, B) post-ovulatory day 3 endometrium; (C, D) post-ovulatory day 5 endometrium; (E, F) post-ovulatory day 13/14 endometrium;
(G, H) menstrual endometrium; (I, J) post-ovulatory day 12; negative control (primary antibody omitted). Original magnification A, C, E,
G, H 370; B, D, E, H, J 3280.
338
Expression of α crystallin B chain in human endometrium
Figure 6. Northern blot analysis of α crystallin B chain mRNA
expression induced by medroxy-progesterone acetate (MPA), in
EnCa-101 tumours. Endometrial tumours were grown in the
presence of a serum concentration of 200 pg/ml 17-β oestradiol.
Two animals each received a single i.m. injection of 5 mg MPA.
Tumours were removed just prior to (0) and at indicated days after
MPA administration. Each lane was loaded with 5 µg poly-A RNA,
fractionated in 1.0% agarose gel with 0.66 M formaldehyde,
transferred to nylon membrane and hybridized first for α crystallin
B chain cDNA probe and then for glyceraldehyde 3 phosphate
dehydrogenase (GAPDH). Upper panel: autoradiograms. Lower
panel: normalized relative optical density ratio calculated as
described in the Materials and methods section. White boxes relate
to the first and the black boxes relate to the second animal.
of 5 mg of MPA significantly enhanced the expression of the
α crystallin B chain mRNA. This enhancement was seen as
early as 3 days after administration of MPA and persisted for
long periods (Figure 6). The kinetics of enhanced expression
of α crystallin B chain after administration of 1–2 mg of MPA
was similar to that seen after administration of 5 mg MPA
(data not shown). Western blot analysis showed the presence
of a protein with a molecular weight of ~22 kDa whose
expression closely followed the pattern of MPA-induced α
crystallin B chain mRNA expression and whose level of
expression increased several-fold by treatment with MPA
(Figure 7).
The progressive enhancement of expression of α crystallin
B chain continued during the entire secretory phase and was
observed even immediately prior to menstruation when the
concentration of serum oestrogen was falling. Therefore, we
tested the hypothesis that the expression of α crystallin B
chain mRNA might also be susceptible to regulation by
oestrogen withdrawal. The epithelial cells of the EnCa101
tumour are susceptible to oestrogen withdrawal. These tumours
grow in the presence of oestrogen, and oestrogen withdrawal
leads to their shrinkage (Satyaswaroop et al., 1983). Consistent
with this hypothesis, oestrogen withdrawal led to a significant
increase in the concentration of a crystallin B chain mRNA
(Figure 8). The kinetics of this enhancement were not identical
Figure 7. Western blot analysis of α crystallin B chain in the tissue
extracts of EnCa-101 grown in the presence of a serum
concentration of 200 pg/ml 17-β oestradiol and after a single i.m.
administration of 5 mg MPA. Tumours were removed and analysed
before medroxy-progesterone acetate (MPA) administration (lane 1)
and on other days as indicated (lanes 2–8). MPA-specific
radioimmunoassay showed the serum MPA concentrations to be
62 ng/ml on day 1, 98 ng/ml on day 3, 19 ng/ml on day 42 and
15 ng/ml on day 63. Upper panel: Western blot. Lower panel:
mean relative optical density calculated as described in the
Materials and methods section.
in tumours of all tumour-bearing mice (Figure 8). In tumours
removed from some animals, the enhancement of expression
was dramatic, whereas in tumours removed from other animals,
only a minimal increase in the mRNA expression could be
observed (Figure 8). A separate experiment performed in three
additional animals yielded similar results (data not shown).
Discussion
Human implantation requires interaction of an intrusive blastocyst with a receptive endometrium. The blastocyst first comes
into contact with the surface epithelium. Therefore, some of the
molecules that make endometrium susceptible to implantation
through interaction with the blastocyst undoubtedly reside in
the specialized surface epithelial cells. The molecular cues
that confer receptivity to endometrium and permit the surface
epithelial cells to interact with the blastocyst have not yet been
identified in humans. In this report, by using RDA, we
identified the α crystallin B protein as one of the proteins,
which is absent in the non-receptive proliferative endometrium,
and which appears in the surface epithelium within the
implantation window. Both the mRNA and protein of the α
crystallin B chain exhibited a similar pattern of expression.
The relative abundance of α crystallin B chain mRNA and
protein progressively increased during the secretory phase.
Immunohistochemical staining showed that, in human endometria, this expression was virtually confined to epithelium
and that its progressive increase in the secretory phase was
largely attributable to its increase in glandular epithelium.
α crystallin consists of two types of highly homologous
339
M.Gruidl et al.
Figure 8. Northern blot analysis of α crystallin B chain mRNA
expression following withdrawal of 17-β oestradiol. EnCa-101
tumours were grown s.c. in nude mice in the presence of oestrogen
pellets implanted s.c. The pellets were designed to maintain the
serum oestrogen concentration at ~200 pg/ml for 60 days. The
tumours were excised following removal of oestrogen pellets on the
indicated days and processed for Northern blot analysis. Serum
oestrogen concentrations dropped to ,20 pg/ml within a day after
pellet removal. Each sample was loaded with 5 µg poly-A RNA
fractionated in 1.0% agarose gel with 0.66 M formaldehyde,
transferred to nylon membrane and hybridized first with the α
crystallin B chain cDNA probe and then with cDNA probe to the
clone 1a. Upper panel: autoradiograms. Lower panel: mean
relative optical density ratio calculated as described in the Materials
and methods section.
subunits, αA and αB, of molecular weight of 20 and 22 kDa
(Quax-Jeuken et al., 1985; Bhat et al., 1991). The A and B
chains non-covalently self-associate to form a large macromolecular complex of ~40 subnits (Quax-Jeuken et al., 1985;
Bhat et al., 1991). Although, originally, it was believed that
the expression of the α crystallin was strictly confined to the
lens, this protein has subsequently been found in a variety of
normal tissues (Klemenz et al., 1993). More recently, evidence
has accumulated that allowed the classification of these proteins
as small heat shock proteins (Klemenz et al., 1991) and
identified them as molecular chaperones (Horwitz, 1992; Jakob
et al., 1993). α crystallin B chain is not the only member of
the heat shock protein family whose expression may be
essential during the secretory phase of the endometrium.
HSP60, HSP70 and HSP90 are also expressed in human
endometrium, and the expression of a small heat shock protein,
340
HSP27, rapidly increases after ovulation (Tabibzadeh et al.,
1996). Although at this point, no specific function can be
assigned to the α crystallin B chain in the surface epithelium
of endometrium, the function of this protein may be to protect
the native forms of other protein(s) essential for implantation.
It has been suggested that the constitutive expression of human
HSP27 and human α crystallin B chain confers resistance
against the cytotoxicity induced both by TNF-α and by
oxidative stress (Mehlen et al., 1995). We showed that human
endometrium expresses TNF-α mRNA and protein (Tabibzadeh, 1991; Hunt et al., 1992) and releases this protein in a
menstrual cycle-dependent fashion (Tabibzadeh et al., 1995b).
The amount of endometrial TNF-α progressively increases
during the secretory phase. TNF-α is cytotoxic and leads to
the activation of the phospholipase A2, the generation of lipid
mediators of inflammation and a rapid rise in the concentration
of mitochondrial oxygen free radicals (Jacquire-Sarlin et al.,
1994). Heat shock proteins intervene with the DNA strand
breaks and lipid peroxidation imposed by the reactive oxygen
species, and protect mitochondrial structure and function
(Jacquire-Sarlin et al., 1994). Thus, the function of the heat
shock proteins in human endometrium may be to counteract
these effects of TNF-α and to suppress the non-specific
aggregation of damaged proteins within cells during the
secretory phase.
Among the physiological signals, both systemic and local
factors may be implicated in the regulation of α crystallin B
chain in the epithelial cells. Among the steroid hormones,
oestrogen regulates the production of a small 24 kDa heat
shock protein (Fuqua et al., 1989). Expression of the mRNA
of this protein was significantly induced in the MCF7 breast
carcinoma cells by oestrogen (Fuqua et al., 1989). We tested
the hypothesis that the expression of α crystallin B chain is
also regulated by steroid hormones. The progressive rise in
the amount of endometrial α crystallin B chain during the
secretory phase may be induced by the progressive rise
in the systemic concentration of progesterone followed by
oestrogen withdrawal. We showed that the expression of α
crystallin B chain mRNA is increased by administration
of MPA and by oestrogen withdrawal. However, the same
treatments did not change the level of expression of HSP27
mRNA (data not shown). Progesterone is known to induce 17β oestradiol dehydrogenase in the endometrial epithelium, the
enzyme that inactivates oestrogen by converting it to oestrone
(Satyaswaroop et al., 1982). Furthermore, progestins downregulate oestrogen receptors (Tseng et al., 1975). These actions
of progesterone are, therefore, reminiscent of oestrogen withdrawal. A recent report on the α crystallin B chain promoter
region in humans did not indicate the presence of progesteroneresponse elements; however, it showed the presence of several
cis-acting sequence elements, including multiple half-site oestrogen-response elements (Frederikse et al., 1994). Such findings suggest that the expression of α crystallin B chain mRNA
may be directly regulated by oestrogen. Immunocytochemical
studies using monoclonal antibodies to oestrogen receptors
have shown that the oestrogen receptor disappears from glandular epithelial cells in the mid-secretory phase of the menstrual
cycle (Bayard et al., 1978). The inhibitory influence of oestro-
Expression of α crystallin B chain in human endometrium
gen and the stimulatory effects of MPA on α crystallin B
chain in the in-vivo experimental system are consistent with
these findings. Local factors may also be implicated in the α
crystallin B chain mRNA expression. For example, it has been
shown that, in glial cells, exposure to TNF-α resulted in the
accumulation of mRNA of α crytallin B chain (Mehlen et al.,
1995). Therefore, it is conceivable that the expression of the
α crystallin B chain may also be regulated, at the local level,
by TNF-α.
In conclusion, by using RDA, we identified the expression
of α crystallin B gene in human endometrium during the
secretory phase of the normal menstrual cycle. Northern blot
and Western blot analysis showed that α crystallin B chain
mRNA and protein are differentially expressed in endometrium.
Immunohistochemical analyses using specific antibodies to α
crystallin B chain further confirmed that the expression of this
protein was initiated during the implantation window and
progressively increased during the secretory phase in the
surface and underlying glandular epithelium of endometrium.
This expression was confined to the endometrial epithelial
cells, and other cell constituents of endometrium including the
stromal, lymphoid and stromal cells did not exhibit immunoreactivity for α crystallin B chain. The regulation of α crystallin
B by oestrogen and progesterone was further demonstrated in
an in-vivo model of human endometrial carcinoma, EnCa-101,
which behaves similarly to normal endometrium with regard
to its steroid hormone responses. Both oestrogen withdrawal
and progestin administration led to the induction of α crystallin
B mRNA as well as protein. Based on these results, we suggest
that this protein is an important factor within the molecular
repertoire that makes endometrium receptive to implantation.
Acknowledgements
This work was supported by grants CA46866 to ST, CA62211 to
PGS and HD29964 to ATF from NIH.
References
Baker, D.J., Nagy, F. and Nieder, G.L. (1992) Localization of c-fos-like
proteins in the mouse endometrium during the peri-implantation period.
Biol. Reprod., 47, 492–501.
Bayard, F., Damilano, S., Robel, P. and Baulieu, E.E. (1992) Cytoplasmic and
nuclear estradiol and progesterone receptors in human endometrium. J.
Clin. Endocrinol. Metab., 72, 635–639.
Bhat, S.P., Horwitz, J., Srinivasan, A. and Ding, L. (1991) Alpha B-crystallin
exists as an independent protein in the heart and in the lens. Eur. J.
Biochem., 202, 775–781.
Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation
by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal.
Biochem., 162, 156–159.
Church, G.M. and Gilbert, W. (1984) Genomic sequencing. Proc. Natl. Acad.
Sci. USA, 81, 1991–1994.
Clarke, C.L., Feil, P.D. and Satyaswaroop, P.G. (1987) Progesterone receptor
regulation by 17-β estadiol in human endometrial carcinoma grown in nude
mice. Endocrinology, 121, 1642–1648.
Das, S.K., Chakraborty, I., Paria, B.C. et al. (1995) Amphiregulin is an
implantation specific progesterone regulated gene in the mouse uterus. Mol.
Endocrinol., 9, 691–705.
Das, S.K., Wang, X.-N., Paria, B.C. et al. (1994) Heparin-binding EGF-like
growth factor gene is induced in the mouse uterus temporally by the
blastocyst solely at the site of its apposition: a possible ligand for
interaction with blastocyst EGF-receptor in implantation. Development,
120, 1071–1083.
De, M., Sanford, T.R. and Wood, G.W. (1993) Expression of interleukin 1,
interleukin 6 and tumour necrosis factor α in mouse uterus during the periimplantation period of pregnancy. J. Reprod. Fertil., 97, 83–89.
Ding, Y.-Q., Zhu, L.-J., Bagchi, M.K. and Gagchi, I.C. (1994) Progesterone
stimulates calcitonin gene expression in the uterus during implantation.
Endocrinology, 135, 2265–2274.
Fazleabas, A.T., Miller, J.B. and Verhage, H.G. (1988) Synthesis and release
of estrogen- and progesterone proteins by the baboon (Papio anubis) uterine
endometrium. Biol. Reprod., 39, 729–736.
Formigli, L., Formigli, G. and Roccio, C. (1987) Donation of fertilized uterine
ova to infertile women. Fertil. Steril., 47, 162–165.
Frederikse, P.A., Dubin, R.A., Haynes, J.I. and Piatigorsky, J. (1994) Structure
and alternate tissue-preferred transcription initiation of the mouse α Bcrystallin/small heat shock protein gene. Nucleic Acids Res., 22, 5686–5694.
Fukuda, M.N., Sato, T., Nakayama, J. et al. (1995) Torphinin and tastin, a
novel cell adhesion molecule complex with potential involvement in embryo
implantation. Genes Dev., 9, 1199–1210.
Fuqua, S.A., Salingaros, M.B. and McGuire, W.L. (1989) Induction of
the estrogen-regulated ‘24K’ protein by heat shock. Cancer Res., 49,
4126–4129.
Horwitz, J. (1992) Alpha-crystallin can function as a molecular chaperone.
Proc. Natl. Acad. Sci. USA, 89, 10449–10453.
Hsu, S.M., Raine, L. and Fanger, H. (1981) Use of avidin–biotin–peroxidase
complex (ABC) in immunoperoxidase techniques: a comparison between
ABC and unlabeled antibody (PAP) procedures. J. Histochem. Cytochem.,
29, 577–580.
Hsu, C.J., Komm, B.S., Lyttle, C.R. and Frnakel, F. (1988) Cloning of estrogen
regulated messenger ribonucleic acids from rat uterus. Endocrinology, 122,
631–639.
Hubank, M. and Schatz, D.G. (1994) Identifying differences in mRNA
expression by representational difference analysis of cDNA. Nucleic Acids
Res., 22, 5640–5648
Hunt, J.S., Chen, H.-L., Hu, X.-L. and Tabibzadeh, S. (1992) Tumor necrosis
factor-α mRNA and protein in human endometrium. Hum. Reprod., 47,
141–147.
Jacquire-Sarlin, M.R., Fuller, K., Dinh-Xuan, A.T. et al. (1994) Protective
effects of hsp70 in inflammation. Experientia, 50, 1031–1038.
Jakob, U., Gaestel, M., Engel, K. and Buchner, J. (1993) Small heat shock
proteins are molecular chaperones. J. Biol. Chem., 268, 1517–1520.
Julian, J., Chiquet-Ehrishmann, R., Erickson, H.P. and Carson, D.D. (1994)
Tenascin is induced at implantation site in the mouse uterus and interferes
with epithelial cell adhesion. Development, 120, 661–671.
Klemenz, R., Frohli, E., Steiger, R.H. et al. (1991) Alpha B-crystallin is a
small heat shock protein. Proc. Natl. Acad. Sci. USA, 88, 3652–3656.
Klemenz, R., Andres, A.C., Frohli, E. et al. (1993) Expression of the murine
small heat shock proteins hsp 25 and alpha B crystallin in the absence of
stress. J. Cell Biol., 120, 639–645.
Lessey, B.A., Damjanovich, L., Coutfaris, C. et al. (1992) Integrin adhesion
molecules in the human endometrium. Correlation with the normal and
abnormal menstrual cycle. J. Clin. Invest., 90, 188–195.
Lessey, B.A., Castelbaum, A.J., Buck, C.A. et al. (1994) Further
characterization of endometrial integrins during the menstrual cycle and in
pregnancy. Fertil. Steril., 62, 497–506.
Lysiak, J.J., Johnson, G.R. and Lala, P.K. (1995) Localization of amphiregulin
in the human placenta and decidua throughout gestation: role in trophoblast
growth. Placenta, 16, 359–366.
Mehlen, P., Mehlen, A., Guillet, D. et al. (1995) Tumor necrosis factorα induces changes, in the phosphorylation, cellular localization, and
oligomerization of human hsp27, a stress protein that confers cellular
resistance to this cytokine. J. Cell. Biochem., 58, 248–259.
Noyes, R.W. and Hertig, A.T (1950) Dating the endometrial biopsy. Fertil.
Steril., 1, 3–25.
Pampfer, S., Tabibzadeh, S., Chuan, F.-C. and Pollard, J.W. (1991) Molecular
cloning of a novel transcript for colony stimulating factor-1 from human
endometrial glands. Production of a transmembrane form of the protein.
J. Mol. Endocrinol., 5, 1931–1938.
Psychoyos, A. (1973a) Endocrine control of egg implantation. In Green, R.O.
(ed.), Handbook of Physiology, Section 7 (Endocrinology). Vol 2 (Female
Reproductive System), Part 2. American Physiological Society, Washington
DC, pp. 187–215.
Psychoyos, A. (1973b) Hormonal control of ovoimplantation. Vitam. Horm.,
31, 201–256.
Psychoyos, A. (1976) Hormonal control of uterine receptivity for nidation. J.
Reprod. Fertil., 25 (Suppl.), 17–28.
341
M.Gruidl et al.
Psychoyos, A. (1986) Uterine receptivity for nidation. Ann. N.Y. Acad. Sci.,
476, 36–42.
Psychoyos, A. (1993) The implantation window: basic and clinical aspects.
In Mori, T., Aono, T., Tominaga, T. and Hiroi, M. (eds), Perspectives on
Assisted Reproduction. Ares Serono Symposia no. 4, Rome, pp. 57–62.
Quax-Jeuken, Y., Quax, W., van Rens, G. et al. (1985) Complete structure of
the alpha B-crystallin gene: conservation of the exon-intron distribution in
the two nonlinked alpha-crystallin genes. Proc. Natl. Acad. Sci. USA, 82,
5819–5823.
Rogers, P.A.W. and Murphy, C.R. (1989) Uterine receptivity for implantation:
human studies in blastocyst implantation. In Yoshinaga, K. (ed.), Blastocyst
Implantation. Adams Publishing Group Ltd, pp. 231–246.
Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning. A
Laboratory Manual. 2nd edn. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, NY, USA.
Sanger, F., Nickeln, S. and Coulson, A.R. (1977) DNA sequencing with chainterminating inhibitors. Proc. Natl. Acad. Sci. USA, 74, 5463–5467.
Satyaswaroop, P.G. and Mortel, R. (1983) Human endometrial adenocarcinoma
transplanted into nude mice. Growth regulation by estradiol. Science, 219,
58–60.
Satyaswaroop, P.G., Wartell, D.J. and Mortel, R. (1982) Distribution of
progesterone receptor, estradiol dehydrogenase, and 20α-dihydroprogeserone dehydrogenase activities in human endometrial glands and
stroma: progestin induction of steroid dehydrogenase activities in vitro is
restricted to the glandular epithelium. Endocrinology, 111, 743–749.
Stewart, C.L., Kaspar, P., Brunet, L.J. et al. (1992) Blastocyst implantation
depends on maternal expression of leukaemia inhibitory factor. Nature,
359, 76–79.
Strauss, J.F. and Gurpide, E. (1991) The endometrium: regulation and
dysfunction. In Yen, S. and Jaffe, R. (eds), Reproductive Endocrinology.
W.B.Saunders Co./Harcourt, Brace and Janovich Inc., Philadelphia,
pp. 309–356.
Tabibzadeh, S. (1991) Ubiquitous expression of TNF-α/cachectin in human
endometrium. Am. J. Reprod. Immunol., 26, 1–5.
Tabibzadeh, S. (1992) Patterns of expression of integrin molecules in human
endometrium throughout the menstrual cycle. Hum. Reprod., 7, 876–882.
Tabibzadeh, S. and Babaknia, A. (1995) The signals and molecular pathways
involved in implantation, a symbiotic interaction between blastocyst and
endometrium involving adhesion and tissue invasion. Mol. Hum. Reprod.,
1, see Hum. Reprod., 10, 1579–1602.
Tabibzadeh, S., Kong, Q.F., Babaknia, A. and May, L.T. (1995a) Progressive
rise in the expression of IL-6 in human endometrium during menstrual
cycle is initiated during the implantation window. Mol. Hum. Reprod., 1,
see Hum. Reprod., 10, 2793–2799.
Tabibzadeh, S., Zupi, E., Babaknia, A. et al. (1995b) Site and menstrual
cycle-dependent expression of proteins of tumour necrosis factor (TNF)
receptor family, BCL-2 oncoprotein and phase-specific production of TNFα
in human endometrium. Hum. Reprod., 10, 227–286.
Tabibzadeh, S., Kong, Q.F., Satyaswaroop, P.G. and Babaknia, A. (1996) Heat
shock proteins in human endometrium throughout the menstrual cycle.
Hum. Reprod., 11, 633–640.
Tseng, L. and Gurpide, E. (1975) Effect of progestins on estradiol receptor
levels in human endometrium. J. Clin. Endocrinol. Metab., 41, 402–404.
Tso, J.Y., Sun, X.-H., Kao, T.-H. et al. (1985) Isolation and characterization
of rat and human glyceraldehyde-3-phosphate dehydrogenase cDNAs:
genomic complexity and molecular evolution of the gene. Nucleic Acids
Res., 13, 2485–2502.
Zentella, A., Weis, F.M.B., Ralph, D.A. et al. (1991) Early gene responses to
transforming growth factor-β in cells lacking growth-suppressive RB
function. Mol. Cell. Biol., 11, 4952–4958.
Received on October 25, 1996; accepted on January 22, 1997
342