Human Reproduction vol 11 no 10 pp 2251-2256, 1996
Changes in nuclear morphology in the human
endometrial glandular epithelium in women with
unexplained infertility
P.Dockery1-4, K.Pritchard 2 , M.A.Warren 2 , T.C.Li3
and I.D.Cooke3
'Department of Anatomy, University College, Cork, Ireland and
Departments of 2Biomedical Science and 3Obstetncs &
Gynaecology, University of Sheffield, UK
••To whom correspondence should be addressed
In the light and electron microscopical study reported
here, we document the structure of the nucleus, nucleolus
and nuclear channel system (NCS) in the uterine
glandular epithelium in both fertile and infertile women
during the early luteal phase. Nuclear volume was found
to be larger in the infertile group at day 5 after the
luteinizing hormone surge (LH + 5) compared to daymatched fertile subjects. A two-way analysis of variance
performed on nucleolar volume data from fertile and
infertile women biopsied on days LH + 4, + 5 and + 6
revealed a significant effect of condition but no effect of
day or interaction. Nucleolar volume decreased from day
LH + 2 to day LH + 6 in the fertile group, the sharpest
decrease occurring between day LH + 3 and day LH + 4.
The largest mean volume of the NCS was found at day
LH + 5 in the fertile group and day LH + 6 in the infertile
group. The results suggest a delay in the development
of this organelle in the infertile women. The present
study has documented alterations in the nuclei of uterine
glandular cells from infertile patients. In these infertile
women, there is also a delayed elaboration of the
secretory apparatus and this delay correlates well with
the delayed/reduced expression of a luteal-specific glycoprotein.
Key words: endometnum/infertility/morphometry/nuclear
channel system/nucleolus
Introduction
The human endometrium undergoes dramatic structural reorganization during the early luteal phase in preparation for
implantation (Comillie et al, 1985, Dockery and Rogers,
1989) In the glandular epithelium this is characterized by the
evolution of a secretory 'triad' of cellular features (Cornillie
et al, 1985) which includes the accumulation of glycogenrich material, the formation of giant mitochondria and the
development of nuclear channel systems (NCS) (Nilsson, 1962;
Clyman, 1963a,b; Terzakis, 1965; Wynn and Wooley, 1967;
More et al., 191 A, More and McSeveney, 1980, Comillie et al,
1985; Dockery et al, 1988a, 1993; Spornitz, 1992)
Dubrausky and Pohlmann (1960) were the first to describe
© European Society for Human Reproduction and Embrvoloev
this unusual organelle, then known as the nucleolar channel
system and subsequently renamed nuclear channel system due
to the inconsistent anatomical association with the nucleolus
(More et al, 1974; Cornillie et al, 1985). The work of More
and McSeveney (1980) provided a clear insight into the
organization of this structure, which exists as an elaboration
of the inner nuclear envelope and forms a spherical, or coneshaped, stack of interdigitating membranous tubules each
surrounded by a granular intranuclear matrix. The initial
confusion over the time of appearance and formation of this
organelle (Clyman, 1963a; Cavazos et al, 1967; Terzakis,
1965; Armstrong et al, 1973; More et al, 1974, 1975) was
in large part due to the inconsistencies in the measurement of
timing and subjective methods of analysing the endometnal
biopsies; careful sampling, measurement of luteinizing hormone (LH) with respect to the phase of the menstrual cycle
and use of reliable morphometnc methods have removed
many such inaccuracies (see Dockery and Rogers, 1989). The
development of this organelle seems to be stimulated by
progesterone (Kohorn et al, 1972). Gore and Gordon (1974)
suggested that the NCS may not be present in some cases of
infertility, and this is indeed the case if biopsies are taken at
day LH+4. However, by days LH+5 and +6 the organelle is
present and seems more elaborate than those seen in normal
fertile women (Dockery et al, 1993). While its function is
unknown, some reports link the NCS to the elaboration of the
cell's secretory apparatus (Clyman, 1963b; Lugibuhl, 1968,
Kohorn et al, 1970, 1972; Gordon et al, 1973); however, this
remains speculative (Dockery and Rogers, 1989). In earlier
studies (Li et al, 1990; Dockery et al, 1993), we reported a
delay in the development of glandular epithelial cells in women
classified as having unexplained infertility who had apparently
normal concentrations of circulating steroids, normal regular
menstrual cycles with no tubal or uterine pathology, evidence
of ovulation and partners with normal semen profiles as defined
by the World Health Organization (1980). These changes in
morphology were correlated with changes in the expression
of a luteal-specific glycoprotein (Graham et al, 1990).
In the present study, we have employed a number of new
and traditional stereological probes on diis same, well-defined
tissue to obtain further details of die nuclear events occurring
in human uterine glands. In particular, our aim was to obtain
reliable quantitative data on the relative and absolute volumetric
changes of glandular cell nuclei, nucleoli, NCS and the tubular
component of die NCS over time in well-defined groups of
fertile and infertile subjects. Such information is likely to
dirow light on the development of the NCS in normal and
infertile women and may lead to a better understanding of the
cellular basis of infertility in humans.
2251
P.Dockery et al
Table L Classification of endometrial biopsies from fertile and infertile
study groups
Chronological date of biopsy
Number of patients
Fertile
LH+2
LH+3
LH+4
LH+5
LH+6
Total
Infertile
4
4
4
4
.4
.4.a
4
ii
LH+ = day after luteinizing hormone surge was detected
Materials and methods
Subjects
Robertson (1984) demonstrated that biopsies taken from the isthmus
or lower portion of the uterus were less responsive to changing
hormonal concentrations. Therefore, in the present study biopsies
were obtained from the site of normal implantation (i e. fundus and
upper part of the body), cut up in a standardized manner, and
processed for light and electron microscopy (see below).
Endometrial biopsies were obtained, using a Sharmans' curette,
from the following groups of women (with informed consent and
appropriate Ethics Committee approval). (1) Normal fertile subjects:
endometrial biopsies were obtained from 20 women with proven
fertility on the following days after the LH surge: +2, 4, 5 and 6.
These women had apparently normal menstrual cycles of 25-35
days length, had not received any steroid hormones (including oral
contraceptives) or used an intrautenne contraceptive device for the
past 3 months. They were aged between 18 and 40 years and had no
known history of tubal disorder, and had one or more successful
pregnancy(ies). All of the samples were chronologically dated in
relation to the LH surge (Li et al, 1987). Table I shows the numbers
of patients in each group. (2) Infertile subjects: this group comprised
11 women whose infertility was still unexplained after extensive
investigation at the Jessop Hospital for Women in Sheffield, UK
(Table I). Subjects were included in this group when they had
infertility of > 1 year duration and no cause of the infertility could
be found on routine clinical examination. For this diagnosis, the
following conditions had to be fulfilled: (a) normal, regular sexual
intercourse; (b) absence of ejaculatory problems; (c) normal semen
analysis as defined by WHO (Belsey et al., 1980) which included
sperm density of at least 20xlO 6 /ml, progressive mouhty of at least
40%, at least 40% morphologically normal spermatozoa, ejaculatory
volume of 3*1.5 ml, no evidence of seminal infection, (d) normal
regular menstrual cycles of between 25 and 35 days; (e) presence of
one or more objective criteria of ovulation; (f) normal thyroid function
test and prolactin concentrations and (g) normal hysterosalpingogram,
laparoscopy, with chromotubation.
Histological procedures
In the present study all tissue was processed in a standardized manner
(see Dockery et al, 1988b). Both semi-thin (0.5 Jim) and thin (70 nm)
sections were cut from Epon blocks on a Reichert OMU4 microtome
using glass and diamond knives. The thin sections were picked up
on copper grids, stained with uranyl acetate and lead citrate for
electron microscopy and examined on a JEOL 100 microscope. The
semi-thin sections were stained with Toluidine blue and examined on
an Olympus light microscope fitted with a drawing tube.
2252
Figure 1. A nuclear channel system in the nucleus of an
endometrial glandular cell Bar = 0.5 (im.
Morphometric procedures
Light microscopy
The semi-thin sections obtained from three blocks per biopsy (selected
by lottery) were subjected to morphometnc analysis. The volumeweighted mean volume of the gland cell nuclei was obtained following
the procedure described by Dockery et al. (1988b) Briefly, this
consisted of taking intercept length measurements of nuclear profiles
using a gnd of regularly arranged points. Fields of view were
identified as 'local vertical windows' of glandular epithelial cells
sectioned through their longitudinal axis At least 50 nuclear profiles
were assessed from each block.
Electron microscopy
Two blocks of the same tissue were selected by lottery from each of
the 20 fertile and 11 infertile biopsies, further tnmmed down and
sectioned for electron microscopy (Figure 1) A systematic series of
micrographs of glandular epithelial cells were taken at an initial
magnification of X2000. From pilot studies based on analysis of
variance of different sampling levels using techniques published
previously (Gundersen and 0sterby, 1981), the optimum sampling
strategy was determined. This was 10 fields of view from each of
six micrographs per block (two blocks per individual), giving 120
fields for each of the 31 individuals examined. Measurements were
performed at X52 000 and all magnifications were determined using
a grating replica with 2160 squares per mm The volume fraction
(Vv) of the nucleolus, or NCS to the nucleus and the Vv of tubular
components to the total channel system were obtained by point
counting (Williams, 1977). As the fractional volumes of nuclear
features obtained from electron micrographs and light microscopic
nuclear volume data were obtained from the same blocks and subject
to identical processing, they were combined to produce absolute
volume estimates of nucleolus, NCS and tubular component of
the NCS.
Statistical analysis
Every feature was initially calculated on a per individual basis.
Individual values were pooled within groups and means and standard
deviations were calculated for each group. Data of absolute volumes
were tested directly while Vv data underwent logarithmic transformations to make the distributions suitable for statistical analysis by
parametric tests. Untransformed data are presented in the tables. Data
were analysed using a one-way analysis of variance, to examine
Nuclear morphology of endometrial glandular epithelium
Table n . Two-way analysis of vanance (F-values) of the effect of condition (fertile or infertile) and chronological date (LH + 2 to LH+6) on nuclear,
nucleolar and NCS volume
Source of variation
Condition
df = 1,18
Volume nucleus
Vv nucleolus nucleus
Volume nucleolus
Vv NCSnucleus
Volume NCS
Vv tubule NCS
Volume tubule
0 07
11 74
12 58
0 22
008
10 20
0001
Interaction
d f = 2,18
Day
df. = 2,18
080
<0 01
<0 01
064
0 78
<0.01
0 98
6.97
1.25
264
5 18
371
931
3 15
<0 01
0.32
0 10
<0 05
<0 05
<0.01
0.07
7.10
0.64
0 42
7.88
5.47
0 38
4 23
<0 01
054
066
<0 01
<0 05
0 69
<0 05
d f = degrees of freedom (numerator, denominator), Vv = volume fraction; NCS = nuclear channel system
differences with time in each group, and a Tukey's honestly significant
difference test was used to compare differences between groups. A
two-way analysis of vanance (ANOVA) was performed to assess the
effect of condition (fertile or infertile), day (of biopsy) and interaction
for data between days LH+4, 5 and 6. Data for days LH+2 and +3
were not included in this analysis as material from infertile patients
was not available The conventional probability of P < 0.05 was
taken as the limit of statistical significance All of the tests were
performed using SPSS software on an IBM compatible PC.
Results
Nuclear volume
A two-way ANOVA performed on data from days LH+4 to
LH+6 revealed no significant effect of condition (fertile or
infertile), but a significant effect of chronological date and a
significant interaction (Table H). The nuclear volume in the
fertile population was numerically largest at day LH+3 and
smallest at day LH+6, the largest nuclei in the infertile group
being found at day LH+5 and the smallest at day LH+6
(Table HI). The nuclei in the LH+5 infertile group were found
to be significantly larger than in the age-matched fertile group
(P < 0.05).
Vv nucleolus to nucleus
The two-way ANOVA performed on the data from fertile and
infertile groups between days LH+4 and LH+6 revealed a
significant effect of condition (fertile or infertile) but there
was no significant effect of chronological date of biopsy or
interaction (Table IT).
The mean Vv in the fertile group decreased from 5.88 ±
2.3% at LH + 2 to 1.98 ± 0.78% at LH+6; the reduction was
most dramatic between days LH+3 and LH+4 (Table DT).
Values in the infertile group at day LH+4 were virtually
identical to the fertile values at LH + 2, suggesting a delayed
response in the infertile group. At day LH+4, although the
mean value for the infertile group was numerically double that
of the fertile group (Table EH), the values were not statistically
significantly different, due to the large variation in the infertile
subjects. Such large variations may suggest that the Vv of
nucleolus to nucleus showed variable response in the individuals within this group.
Volume of nucleolus
A two-way ANOVA showed that there was a significant effect
of condition (fertile or infertile) but that there was no effect
of chronological date of biopsy and no significant interaction
(Table II).
There was a reduction in the volume of the nucleolus in the
fertile group during the period of study (Table HI). The sharpest
decrease took place between days LH+3 and LH+4. A
progressive decrease was also seen in the infertile group, with
a sharp decrease being observed in the infertile group between
days LH+5 and LH+6.
Vv nuclear channel system to nucleus
A two-way ANOVA revealed no significant effect of condition
(fertile or infertile) but there was a significant effect of
chronological date of biopsy (P < 0.05) and there was a
significant interaction (P < 0.01; Table H).
The proportion of the NCS in the nucleus peaked at day
LH+ 5 in the fertile women and appeared to reach a maximum
at day LH+6 in infertile women, again suggesting delayed
development in the infertile group (Table IV). However, the
peak value of the infertile group (3 10 ± 6.75) was almost
double that of the peak value in the fertile group (1.77 ±
1.43), possibly suggesting the infertile group showed both
delay and a compression of NCS development at this time.
Volume of nuclear channel system
A two-way ANOVA revealed a pattern similar to that of the
Vv of NCS to nucleus (Table H), with significant effects of
chronological date of biopsy (P < 0.05) and interaction
(P < 0.05)
The mean volume of the NCS was largest at day LH+5
[coefficient of variation (CV) = 84%] in the fertile group
(Table IV). In the infertile group the largest volume was
measured at day LH+6 (CV = 63%) However, although the
differences were large, they were not statistically significant
due to the high inter-individual variability. Again, the volume
of the NCS in the infertile group at day LH+6 was much
larger than the peak value (on day LH + 5) in the fertile group
(Table IV).
Vv tubular component: nuclear channel system
A characteristic of the NCS is the presence of a complex
tubular system embedded in a granular matrix. The Vv and
volume of the tubules were determined separately from the
whole NCS.
A two-way ANOVA revealed a significant effect of condition
2253
P.Dockery et at
Table HL Influence of condition (fertile or infertile) and chronological date
on volume of nucleus and nucleolus Values represent mean ± SD for each
group
Chronological date
LH+2
Volume of nucleus
Vv nucleolusinucleus
Volume of nucleolus
LH+3
Volume of nucleus
Vv nucleolus:nucleus
Volume of nucleolus
LH+4
Volume of nucleus
Vv nucleolus:nucleus
Volume of nucleolus
LH+5
Volume of nucleus
Vv nucleolusnucleus
Volume of nucleolus
LH+6
Volume of nucleus
Vv nucleolusmucleus
Volume of nucleolus
Fertile
Infertile
280 5 ± 31 8
5 88 ± 2 25
16.64 ± 6 79
319 5 ± 53 1
4 73 ± 1 17
15 36 ± 4 42
288 1 ± 32 9
2 60 ± 1 10
7 36 ± 2 79
265 3 ± 15.6
5 26 ± 2 46
14 14 ± 7.26
265 2 ± 1 8 9*
2 72 ± 0 %
7 21 ± 2 59
326 4 ± 30 1
3 81 ± 0 98
12.3 ± 2 89
260.8 ± 19 I 1
1.98 ± 0 78b
5 10 ± 1 79°
231 1 ± 39.3'
3 73 ± 1.48
8 52 ± 3 15
Vv = volume fraction; NCS = nuclear channel system.
"Significantly smaller than LH+5 infertile
Significantly smaller than LH+2 fertile
c
Significantly smaller than LH+2 fertile
Table IV. Influence of condition (fertile or infertile) and chronological date
on volume of nuclear channel system (NCS) Values represent mean ± SD
for each group
LH+2
Vv NCS:nucleus
Mean volume of NCS
Vv tube NCS
Volume tube
LH+3
Vv NCS.nucleus
Mean volume of NCS
Vv tube NCS
Volume tube
LH+4
Vv NCSnucleus
Mean volume of NCS
Vv tube:NCS
Volume tube
LH+5
Vv NCSrnucleus
Mean volume of NCS
Vv rube.NCS
Volume tube
LH+6
Vv NCSnucleus
Mean volume of NCS
Vv rube.NCS
Volume tube
Infertile
Fertile
0
0
0
0
—
0
0
0
0
—
0 45 ±
1 22 ±
12 00 ±
0 363 ±
0 53
1.34
8 29
0 317
1.77 ±
4 74 ±
18.80 ±
0 938 ±
143
4 03
4.59
0 931
0 38 ± 0.35
1 24 ± 1 06
10.90 ± 8 67
0 160 ± 0 127
0 78 +
2.10 ±
23 80 ±
0 428 ±
068
2.00
8.97
0 537
3 10 ±
771 ±
7.70 ±
1.380 ±
o o o o
Chronological date
6.75
4 85
2.69
0.0269
(fertile or infertile) and chronological date of biopsy but there
was no significant interaction (Table II).
The proportion of the NCS occupied by tubules was found
to increase with time in both groups, and was largest at day
2254
LH+6 in both groups. Although not statistically significantly
different, in numerical terms almost one quarter of the fertile
NCS was made up of tubules at this time, with only one sixth
of the infertile NCS similarly composed (Table IV).
Volume of tubular component
A two-way ANOVA revealed a borderline effect of chronological date of biopsy (P < 0.07), and a significant interaction
(P < 0.05) (Table H).
The mean volume of the tubular component of the NCS
was largest on LH+5 (CV = 99%) in the fertile group. The
largest volume in the infertile group occurred at LH+6 (CV =
64%), tubule volume in this group being numerically three
times greater than that of the fertile group on day LH+6.
However, although the differences were large, they were
not statistically significant due to the high inter-individual
variability.
Discussion
In the present study we have documented precisely in the
glandular epithelium from a well-defined group of fertile
women the nuclear changes occurring during the early luteal
phase, and have discovered a number of deviations from this
pattern in a group of women diagnosed as unexplained infertile.
Overall, these results suggest a delay in the nuclear development in women suffering from unexplained infertility compared
to fertile controls. This represents the first account of a
morphological difference in a group previously diagnosed as
'unexplained' and, although the mechanism by which delay
in NCS development can cause infertility remains unclear,
previous work suggests that it affects secretory function
(Graham et al, 1990). Roberts et al. (1989) reported that the
nucleolar area was maximal during the early secretory phase
and decreased by the mid secretory phase. They speculated
that the decrease in nucleolar area during this period may be
due to a decrease in oestrogen blood concentrations and/
or glandular oestrogen receptors. A decrease in oestrogen
concentration occurs at the same time as progesterone concentration increases and it seems likely that the relative amounts
of steroids affect transcription. A possible explanation for the
observed change in nuclear volume could be as follows:
ribosomes, produced by the nucleolus and assembled in the
cytoplasm, have been stimulated to multiply prior to their
cytoplasmic role in secretion Once the nucleolus has produced
the ribosomal precursors (early in the luteal phase) it declines
in volume and the nbosomes that it has produced begin to
translate secretory proteins. Such secretory changes have been
reported using morphometric techniques (Dockery et al, 1993).
These results generally agree with the findings of the present
study, although it should be noted that the area occupied by
an object in a single section may not truly reflect its 'size'
(Mayhew, 1991). However, the Vv of an object does reflect
its relative volume when sectioned randomly, although such a
ratio can be difficult to interpret (Williams, 1977).
In the present study we were able to document the formation
and structural reorganization of the NCS Recent studies on
well-timed endometrial biopsies from fertile women (Dockery
Nuclear morphology of endometrial glandular epithelium
et al, 1988b, 1993) have shown that the appearance of the
NCS is a rapid process. Thus, it is seldom seen at day LH + 3,
yet by day LH+4 it is well developed. Due to its appearance
in the early luteal phase its presence has been linked to
utenne glandular stimulation by high amounts of circulating
progesterone. Little is known about the function of the NCS
and, although it has been suggested that it in some way
facilitates the rapid massive transfer of RNA and ribosomal
precursors from nucleus to cytoplasm, there is little evidence
for this (Spornitz, 1992). By day LH+2 the cytoplasm is very
densely packed with ribosomes and the nucleolus which
produced them is in decline, while the NCS does not appear
until day LH+4. Clearly, the timing of the development of
this organelle and its alleged function are not consistent.
To assume a simple link between circulating steroid concentrations and a morphological feature may be tempting but is
naive with regard to the female reproductive system (Nikas
et al, 1995). It is much more likely that the relative levels of
several influences, as well as the priming by the immediately
preceding hormonal environment, will cause a response. Even
this is likely to be threshold dependent, with individuals
behaving differently. Indeed, it has been shown that there is
no simple relationship between peripheral hormone values and
morphometric indices (Johannisson et al, 1987; Rogers et al,
1989). It has been suggested that the endometrium must reach
a requisite level of development before it can respond in an
adequate way to progesterone (Johannisson et al., 1987).
The function of the NCS is not yet known. It has been
suggested that it is involved in the rapid transport of new
mRNA (More et al, 1974), but more recently Buchwalow
et al. (1985) localized ribonucleoproteins and nucleoside
phosphatases within the NCS. What messages these mRNA
may convey remains to be elucidated; furthermore, the mechanisms of transport via such a route remain obscure. Also, given
that the lumen of the tubules is continuous with that of the
nuclear envelope, which is itself linked to the cavity of the
endoplasmic reticulum, the role of mRNA within the space
remains puzzling. There remains the possibility that the NCS
somehow feeds back information concerning the progress
and products of transcription within the rough endoplasmic
reticulum to the nucleus and so refines its control at a time of
rapid and extensive secretory activity. It is also of interest to
note that Ca 2+ release channels have been reported to be
associated with the inner nuclear envelope (Gerasimenko et al,
1995), which may suggest an alternative role for the NCS in
the dynamic secretory activity of these cells.
Observations by Luginbuhl (1968) showed that proliferative
endometrium cultured in the absence of progesterone developed
limited secretory activity but did not develop NCS In addition,
secretory endometrium failed to develop NCS if cultured in
the absence of progesterone. Progesterone administration
in vitro and in vivo was reported to promote their development
(Clyman, 1963b; Kohorn et al., 1970). The pharmacology
of NCS formation seems relatively specific. For example,
compounds such as 19-nortestosterone and similar progestational steroids failed to induce NCS formation, although
the two other components of the 'functional triad', giant
mitochondrial formation and glycogen deposits, were stimu-
lated to develop. Kohorn et al. (1972) linked the 17P-position
on the D ring of the progestational steroids to channel
formation. They also noted that treatment with oestrogens
appeared to disrupt the channel systems. Dehou et al. (1987),
reporting on hormone replacement therapy in cases of premature menopause, commented that high concentrations of oestrogen in the early luteal phase prevented the appearance of the
NCS. Dockery et al. (1991), in a study on the effect of
hormone replacement therapy on the fine structure of the
glandular epithelium in women with premature ovarian failure,
also noted differences in the elaboration of the NCS in
women apparently hyperstimulated with oestrogen followed
by high amounts of progesterone. Recent (P.Dockery et al.,
unpublished) work suggests that the administration of RU486
prior to the formation of the NCS (on day LH+3) prevents
its formation, while administration of the drug later in the
cycle causes the disappearance of this organelle. Li et al.
(1991) suggested that i.m. injections of progesterone during
the luteal phase were not sufficient to promote the overelaboration of the secretory apparatus and, in particular, the
NCS.
In the present study we have documented the structure of
the NCS in fertile and infertile women during the early
luteal phase using reliable objective methods on well-defined
subjects. Although there was considerable variation in the
NCS, suggesting a variation in response of individuals with
each group, a basic pattern was seen. This showed a delay of
NCS formation and possible compression of development up
to day LH+6 in women who had previously been diagnosed
as 'unexplained' infertile compared with the fertile group.
The elaboration of this progesterone-dependent organelle is
associated with the development of the secretory apparatus in
the gland cell (Dockery et al., 1993); however, further studies
are necessary to elucidate the function of this unusual intranuclear organelle. Morphological changes have traditionally
been used to assess endometnal biopsies (Noyes et al., 1950),
and the use of morphometric techniques has improved the
objectivity of this assessment (Li etal, 1988) The introduction
of a variety of stage-specific molecular markers has added a
greater insight into the physiology of this dynamic tissue
(Aplin, 1989; Lessey et al., 1992; Hey et al, 1995).
It is of interest to note that, in the women in our study with
delayed NCS formation, there was also a delayed elaboration
in the secretory apparatus (Dockery et al, 1993) and that this
delay correlates well with the delayed/reduced expression of
a luteal-specific glycoprotein (Graham et al, 1990) However,
the developmental sequence of these cellular aberrations
remains to be established. Finally, perhaps the diagnosis of
'unexplained' infertility should be reconsidered, since this
group has now been shown to have a distinct recognizable
difference from ferule controls.
Acknowledgement
This work was partly funded by RGC grant HKU 263/92M to
PDockery while at the University of Hong Kong
References
Aplm, J D. (1989) Cellular biochemistry of the endometrium In Wynn, R M
and Jollie, WP (eds), Biology of the Uterus Plenum, New York and
London, pp 89-129
2255
PJDockery et aL
Armstrong, E M , More, I.A.R., McSeveney, D and Charfield, WR. (1973)
Reappraisal of the ultrastructure of the human endometnal glandular cell.
J Obstet. Gynaecol Br Commonw, 80, 446-460.
Belsey, M.A., Ehasson, R , Gallegos, A J . et al (1980) Laboratory manual
for the examination of human semen and semen cervical mucus interactions.
Press Concern, Singapore.
Buchwalow, IB., Belyaeva, L.A and Zavalshina, L.E (1985) Localisation
of nucleoside phosphatases (ATPase and 5'-nucleotidasc) and nuclear
nbonucleolar proteins in the human endometnal glandular cells during the
secretory phase Ada Histochem., 77, 205-208
Cavazos, F , Green, J.A., Hall, D.G and Lucas, FV (1967) Ultrastructure of
the human endometnal glandular cell Am. J. Obstet Gynecol, 99, 833-854.
Clyman, M J . (1963a) A new structure observed in the nucleolus of the human
endometrial epithelial cell. Am. J. Obstet. Gynecol, 86, 430-432
Clyman, M J . (1963b) Electron microscopic changes produced in the human
endometnum by norethindrone acetate with ethyl estradioL Feml Stenl.,
14, 352-364
Comilhe, FJ., Lauweryns, J.M. and Brosens, I A (1985) Normal human
endometrium. Gynecol. Obstet. Invest, 20, 113-129
Dehou, M F , Lejeune, B., Ainjis, C and Leroy, F (1987) Endometnal
morphology in stimulated in vitro fertilisation cycles and after steroid
replacement therapy in cases of primary ovarian failure. FertiL Stenl, 48,
995-1000
Dockery, P. and Rogers, A W (1989) The effects of steroids on the fine
structure of the endometnum Balheres Clin. Obstet Gynaecol, 3, 227-247
Dockery, P , Li, T.C., Rogers, A.W. et al (1988a) The ultrastructure of the
glandular epithelium in the timed endometnal biopsy. Hum. Reprod, 3,
826-834.
Dockery, P , Li, T C , Rogers, A W et al. (1988b) An examination of the
variation in the timed endometnal biopsy Hum. Reprod, 3, 715-720.
Dockery, P., Tidey, R., LJ, T C and Cooke, I D. (1991) A morphometnc study
of the utenne glandular epithelium m women with premature ovarian failure
undergoing hormone replacement therapy Hum. Reprod, 6, 1354-1364.
Dockery, P , Pntchard, K., Taylor, A et al (1993) The fine structure of the
human endometrial glandular epithelium in cases of unexplained infertility:
a morphometnc study Hum. Reprod., 8, 667-673.
Dubrauszky, V and Pohlmann, G (1960) Structurveranderungen am Nukeolus
von Korpusendometnumzellen wahrend der Sekreuonsphase Natunvissenschaften, 47, 523-524.
Gerasimenko, O V, Gerasimenko, J V, Tepkin, A.V. and Petersen, O.H (1995)
ATP dependent accumulation of inositol tnphosphate or cyclic ADP-nbose
mediated release of Ca 2 + from the nuclear envelope Cell, 80, 439-444.
Gordon, M., Kohom, E l , Gore, B Z and Rice, S.I (1973) Effects of
postovulatory oestrogens on the fine structure of the epithelial cells in the
human endometnum J Reprod. Fertil, 34, 375-378
Graham, R.A., Self, M.W, Aplm, J D et aL (1990) An endometnal factor in
unexplained infertility Br. Med J, 300, 1428-1431
Gundersen, H J G and 0sterby, R (1981) Optimising sampling efficiency of
stereologica] studies m biology or 'Do more less well' J Microsc, 121,
65-73
Hey, N A , U, T C , Devine, P.L el al (1995) MUC1 in secretory phase
endometnum: expression in precisely dated biopsies and flushings from
normal and recurrent rruscamage panents. Hum. Reprod., 10, 2655-2662
Johannisson, E., Landgren, B.-M., Rohr, H.P. and Dicsfalusy, E (1987)
Endometnal morphology and penpheral hormone levels in women with
regular menstrual cycles Fertil SlenL, 38, 564-571.
Kohom, E I , Rice, S I and Gordon, M. (1970) In vitro production of nucleolar
channel system by progesterone in the human endometnum. Nature, 228,
671-672
Kohom, E.I, Rice, S I., Hcmperiy, S. and Gordon, M. (1972) The relation
and structure of the progestational steroids to nucleolar differentiation in
the human endometnum. J. Clin. Endocnnol Metab., 34, 257-264
Lessey, B A., Damajanovich, L , Coutafans, C. et aL (1992) Integnn adhesion
molecules in the human endometnum. J CIm Invest, 90, 188-195.
Li, T.C , Rogers, A.W., Lenton, E.A et aL (1987) A comparison between two
methods of chronological dating of endometrial biopsies during the luteal
phase and their correlation with histological dating. FemL StenL, 48,
928-932.
Li, T C , Rogers, A.W, Dockery, P. et al (1988) A new method of histologic
daring of human endometrium in the luteal phase. FemL SteriL, 50, 52-60.
La, T.C, Dockery, P., Rogers, A.W and Cooke, I.D. (1990) A quantitative
study of endometrial development in the luteal phase: a comparison between
women with unexplained fertility and women with normal fertility Br
J. Obstet. GynaecoL, 97, 576-582.
2256
Li, T.C, Dockery, P and Cooke, LD (1991) Effects of exogenous progesterone
administration on the morphology of normally developing endometnum in
the pre-unplantation period Hum. Reprod., 6, 641-644.
Luginbuhl, WH (1968) Electron microscopic study of the effects of tissue
culture on the human endometnum Am. J Obstet Gynecol., 102, 192-201
Mayhew, TM. (1991) The new stereological methods for interpreting
functional morphology from slices of cells and organs. Exp. PhysioL, 76,
639-665
More, I A.R. and McSeveney, D. (1980) The three dimensional structure of
the nucleolar channel system in the endometnal glandular cell senal
sectioning and high voltage electron microscopic studies J. Anas, 130,
673-682
More, I A.R, Armstrong, EM., McSeveney, D and Chatfield, WR. (1974)
The morphogenesis and fate of the nucleolar channel system in the human
endometnal glandular cell. J UltrastrucL Res, 47, 74-85.
More, I A R , Armstrong, E M and McSeveney, D. (1975) Observations on
the three dimensional structure of the nucleolar channel system of the
human endometrial glandular cell J. Anat., 119, 163-167.
Nikas, G., Drakakas, P, Loutradis, D. et aL (1995) Utenne pinopodes as
markers of the mdation window in cycling women receiving exogenous
oestradiol and progesterone Hum. Reprod., 10, 1208—1213.
Ntlsson, O (1962) Electron microscopy of the glandular epithelium in the
human uterus II Early and late luteal phase. J Ultrastruct Res., 6,422-431
Noyes, R.W, Hertig, A T and Rock, A.T (1950) Dating the endometnal
biopsy Feml. Stenl., 1, 3-25
Roberts, D K., Lavia, L.A, Horbelt, D V and Walker, N J. (1989) Changes
m nuclear and nucleolar areas of endometnal glandular cells throughout
the menstrual cycle Int. J Gynecol PathoL, 8, 36-45.
Robertson, WB (1984) A reappraisal of the endometnum in infertility In
Fox, H (ed) Climes m Obstetncs and Gynaecology, 11, 209-226
Rogers, P., Murphy, C , Cameron, I et aL (1989) Uterine receptivity in
women receiving steroid replacement therapy for premature ovanan failure:
ultrastructural and endocnnological parameters Hum. Reprod., 4, 349-354
Spomitz, U (1992) The functional morphology of the human endometnum
and decidua. Adv. Anat. Embryol Cell BwL, Y2A, 99 pp
Terzakis, J.A (1965) The nucleolar channel system of the human endometrium.
J Cell BwL, 27, 293-304
World Health Organization (WHO) Task force on methods for the determination
of the fertile penod Temporal relationships between ovulation and defined
changes in concentration of plasma oestradiol-17, lutemising hormone
follicle stimulating hormone and progesterone Am J Obstet GynecoL,
138, 383-390
Williams, M A (1977) Quantitative methods in biology In Glauret, A M
(ed ), Practical Methods for Biological Microscopy, Vol. 6 North Holland,
Amsterdam, pp 5—84
Wynn, R M. and Woolley, R.S (1967) Ultrastructural cyclic changes in the
human endometnum II Normal postovulatory phase Fertil. Stenl, 18,
721-738
Received on January 22, 1996, accepted on June 26, 1996