GHENT UNIVERSITY
Remodeling of the canine endometrium during the estrous cycle Steven VAN CRUCHTEN
Thesis to obtain the academic degree of Doctor of Veterinary Science (PhD) Faculty of Veterinary Medicine, Ghent University 2004
Promoter: Prof. Dr. W. Van den Broeck Copromoter: Prof. Dr. P. Simoens Department of Morphology Faculty of Veterinary Medicine, Ghent University Salisburylaan 133, B-9820 Merelbeke
ISBN 90-5864-055-8
CONTENTS List of abbreviations GENERAL INTRODUCTION
1
AIMS OF THE STUDY
19
CHAPTER 1
ELECTRON MICROSCOPIC CHANGES OF THE CYCLIC CANINE 23 ENDOMETRIUM
CHAPTER 2 2.1
Summary
25
Introduction
26
Material & Methods
26
Results
30
Discussion
36
References
38
CELL DEATH IN THE CYCLIC CANINE ENDOMETRIUM
41
Review: Morphological and biochemical aspects of apoptosis, oncosis, necrosis and autophagic cell death
43
Summary
45
Introduction
46
History
46
Morphology
47
Mechanisms
51
Apoptosis and the Cell Cycle
58
Detection Methods
60
Concluding Remarks
63
References
64
2.2
2.3
CHAPTER 3
Apoptosis in the cyclic canine endometrium
71
Summary
73
Introduction
74
Material & Methods
75
Results
78
Discussion
82
References
86
Cell death in the canine endometrial surface epithelium: an electron microscopic study
91
Summary
93
Introduction
94
Material & Methods
94
Results
96
Discussion
98
References
102
PROLIFERATION PATTERNS IN THE CYCLIC CANINE
105
ENDOMETRIUM
CHAPTER 4
Summary
107
Introduction
108
Material & Methods
109
Results
112
Discussion
116
References
120
EXPRESSION OF MMP-1 AND TIMP-1 IN THE CYCLIC CANINE 125 ENDOMETRIUM
Summary
127
Introduction
128
Material & Methods
129
Results
132
Discussion
137
References
141
GENERAL DISCUSSION
145
SUMMARY
163
SAMENVATTING
169
CURRICULUM VITAE
177
BIBLIOGRAPHY
179
DANKWOORD
183
LIST OF ABBREVIATIONS AEC
aminoethylcarbazole
mM
millimolar
ANOVA
analysis of variance
MMP-1
matrix metalloproteinase-1
APES
3-aminopropyl-triethoxysilane
MPT
membrane permeability transition
BSA
bovine serum albumin
P4
progesterone
CEH
cystic endometrial hyperplasia
PBS
phosphate buffered saline
DAB
diaminobenzidine
PR
progesterone receptor
DAPI
SD
standard deviation
DNA
4’,6-diamidino-2-phenylindole dihydrochloride deoxyribonucleic acid
sem
standard error of the mean
E2
estradiol-17β
SEM
scanning electron microscopy
E. coli
Escherichia coli
TBS
tris buffered saline
EDTA
ethylenediaminotetraacetic acid
TEM
transmission electron microscopy
EM
electron microscopy
TIMP-1
tissue inhibitor of metalloproteinase-1
ER
estrogen receptor
TUNEL
FGF
fibroblast growth factor
v/v
terminal deoxynucleotidyl transferase mediated dUTP nick end labeling volume/volume
IgG1
immunoglobulin G1
w/v
weight/volume
IP3
inositol (1,4,5) triphosphate
im
intramuscular
iv
intravenous
kD
kilodalton
LH
luteinizing hormone
M
molar
GENERAL INTRODUCTION
General Introduction
3
Canine reproduction and reproductive disorders have been studied elaborately since many years. The canine estrous cycle shows some particular differences compared to other species. Special features are the pre-ovulatory luteinization of the tertiary ovarian follicles (England, 1998) and the long lifetime of the corpora lutea in the non-pregnant animal (Arthur et al., 1983).
The estrous cycle of the bitch The estrous cycle of the bitch is mono-estrous, literally meaning that only a single estrus occurs in each breading season (Pineda, 1989b). The cycle can be divided into proestrus, estrus, early metestrus, late metestrus and anestrus according to morphological and serological changes and alterations of behavior (Vermeirsch et al., 2001). Proestrus is the beginning of the period of sexual activity and lasts approximately 9 days. Proestrus starts when a serohemorrhagic vaginal discharge can be noticed (Pineda, 1989a; Schaefers-Okkens and Kooistra, 1996) and ends on the day when the bitch becomes sexually receptive for the male. During this cycle stage serum estrogen levels increase and reach maximum levels at the end of this stage (Pineda, 1989a; Schaefers-Okkens and Kooistra, 1996). One or two days after the maximal estrogen level there is a surge of luteinizing hormone (LH) in the blood (Concannon, 1986; Pineda, 1989a; England, 1998). Progesterone levels are low but start increasing before the peak of serum estrogens, due to the luteinizing granulosa cells of ripening follicles (Concannon, 1986; England, 1998). During estrus, which lasts about 9 days, the bitch is sexually receptive for the male and ovulations occur. During this period serum estradiol-17β levels decrease, progesterone levels increase and the LH peak occurs (Concannon, 1986). This LH peak is the trigger for ovulation which occurs 1 to 3 days later. After ovulation, follicles do not collapse but are marked by further proliferation of the luteinized tissue leading to functional corpora lutea before the end of estrus (England, 1998). The day on which the bitch no longer accepts the male for copulation is frequently considered the first day of metestrus (Pineda, 1989a; Schaefers-Okkens and Kooistra, 1996; England, 1998). Metestrus, literally meaning the period after estrus, has an average length of 2 to 3 months and can be further divided in early and late metestrus due to the large differences in serum progesterone levels and in ovarian and uterine histology between the beginning and end of this period (Vermeirsch et al., 2001). The term diestrus is not used in the dog since it depicts the period of luteal activity in between two estrous cycles in polyestrous animals (Chastain and Ganjam,
General Introduction
4
1986). Anestrus is the inactive stage in which the animal is sexually at rest (Pineda, 1989a). The average duration of this stage is 3 to 4 months. During anestrus corpora lutea are completely regressed and estradiol-17β and progesterone levels are below the basal values, viz. 2 to 10 pg/ml and 0.5 ng/ml, respectively (Arthur et al., 1983).
Cyclic endometrial remodeling in the bitch Canine endometrial remodeling during the estrous cycle has been studied previously (Keller, 1909; Arenas and Sammartino, 1939; Mulligan, 1942, Barrau et al., 1975b, SpanelBorowski et al., 1984). The different endometrial structures are shown during the various cycle stages in Figures 1 to 5. A schematic review of the literature is given in Figure 6. In the earliest study performed by Keller in 1909, the cyclic changes of the canine endometrium were investigated elaborately using haematoxylin-eosin stained paraffin sections. In this study, the different regions of the endometrium, viz. the surface epithelium, the superficial part of the stroma (stratum cellulare) and the deep part of the stroma (stratum reticulare) were investigated. Furthermore, the endometrial glands were divided into basal uterine glands and uterine crypts. These latter structures were described as undeep glands of 0.1 - 0.2 mm length lying in the superficial part of the stroma between the ducts of the “real” uterine glands (basal glands) that were localized in the deep part of the stroma. The epithelium of both structures showed no differences. The estrous cycle was divided into 4 cycle stages, viz. estrus, stage of glandular proliferation, stage of regression and a rest stage (Keller, 1909). The estrous stage was further subdivided into a first and a second phase corresponding to proestrus and early estrus, respectively. During the first phase of estrus numerous mitotic figures were observed in the neck region of the crypts, but mitoses were also present in the basal glands (Fig. 6). Only rarely mitotic cells were found in the surface epithelium at that stage. For the stroma, no differentiation between the stratum cellulare and stratum reticulare was possible and only rare mitotic stromal cells were noticed. During the second phase of estrus, the number of mitotic cells remained small in the crypts and the basal glands. In the superficial stroma brown pigment, viz. hemosiderin, was found as a result of digestion of blood which was expelled during the first phase of estrus. Collagen fibres, which formed a compact network during
General Introduction
5
anestrus, were spreading and formed a web-like pattern. In the stage of glandular proliferation, which corresponds with late estrus and the beginning of early metestrus, the corpora lutea were still developing and glandular proliferation reached its maximum (Fig. 6). At that time, mitoses were not observed in the surface epithelium and the crypts, but they were very numerous in the basal glands. However, in some dogs already pyknotic nuclei surrounded by some cytoplasm were observed in the glandular lumina of the basal glands (Fig. 6). Due to the large number of glandular cells, the stroma showed a compact arrangement. It was concluded that at the end of estrus no regression occurred but prominent proliferation was still present. The secretion process, which started during estrus, also continued during this cycle stage. During the next cycle stage, i.e. the stage of regression which corresponds to late metestrus, prominent cellular debris with pyknotic nuclei was observed in the lumina of the crypts and the basal glands. No such features were observed in the surface epithelium, in which numerous lipid droplets were present at that time. SE C
C
L
C
C
C
C C
L C C
C SE
C
S
S
G G G
G
1 C C SE
G G
2
G
C
C
C
G
G
G
G G
SE L C
C
C
L
C
G
G
C
SE
G
G
C
Figs. 1-5. Histological sections of the lumen (L), the surface epithelium (SE), the crypts (C), the stroma (S) and the basal glands (G) of the canine endometrium during proestrus (1), estrus (2), early metestrus (3), late metestrus (4) and anestrus (5). Morphological differences are most prominent between early metestrus and anestrus. During early metestrus the basal glands are most numerous and the height of all epithelia is at its maximal level. In contrast, smallest numbers of basal glands and lowest epithelia are found during late metestrus and anestrus. Bar = 100 µm. C
C
C
SE
S
L
C
C
C
C
S
S G
G
3
G
G
G
G
G
G G
G G
G
G
G
G
4
G
G
G
5
G G
G
G
G
General Introduction
Fig. 6. Schematic graph of the different published data on canine endometrial remodeling during the estrous cycle. All epithelia are colored in pink: the surface epithelium which borders the uterine lumen, the crypts which are lying in the superficial stroma underneath the surface epithelium, and the basal glands situated in the deep stroma. During proestrus, serum estradiol-17β (E2) levels rise and cellular height increases in both the surface and glandular epithelium (Barrau et al., 1975b). High ER and PR levels are present in all cell groups (Vermeirsch et al., 1999; 2000). Proliferation remains enigmatic: Spanel-Borowski et al. (1984) observed high numbers of H3 thymidine labeled cells (grey ellipses) in the surface epithelium, the crypts and the basal glands, whereas Keller (1909) and Barrau et al. (1975b) found numerous mitotic figures (brown lines) in the surface epithelium and crypts but not in the basal glands at that time. During estrus, serum progesterone (P4) levels start increasing and PR levels remain high in all cell groups (Vermeirsch et al., 2000). Prominent proliferation has been observed by Keller (1909) and Barrau et al. (1975b) in the basal glands but not by Spanel-Borowski et al. (1984). During early metestrus, serum P4 concentrations reach their maximum, whereas ER and PR levels are low in all cell groups except in the basal glands, in which the PR levels remain relatively high (Vermeirsch et al., 1999; 2000). According to Keller (1909) and Barrau et al. (1975b) proliferation of the basal glands peaks, but declines towards the end of this cycle stage and even cellular debris with pyknotic nuclei (grey dots) has been observed in some dogs (Keller, 1909). During late metestrus, serum P4 levels and cellular height of both the surface and glandular epithelium decrease, but the basal glands are still numerous (Barrau et al., 1975b). ER levels are relatively high, whereas PR levels are low in these cell groups, except for the stroma (Vermeirsch et al., 1999; 2000). Prominent cellular debris with pyknotic nuclei is present in the lumina of the crypts and the basal glands, whereas the surface epithelium shows fat accumulation with deformation of the nuclei (Keller, 1909). According to Arenas and Sammartino (1939) and Mulligan (1942), these lipid-loaded cells are desquamated (dotted pink line) during late metestrus and the beginning of anestrus. During anestrus, serum E2 and P4 levels are below the basal values and the number of crypts and basal glands decreases further (Barrau et al., 1975b). ER and PR levels are relatively high in all cell groups, except for the basal glands in which the PR reach their lowest level (Vermeirsch et al., 1999; 2000). Using zymography, high MMP-2 levels in canine endometrial extracts have also been reported at that time (Chu et al., 2002).
6
General Introduction
7
PROESTRUS E2 ↑ ER ++++ PR ++++
?
ER +++ PR ++++
MMP-1 ? ER ++++ PR ++++
MMP-2 ↑
ANESTRUS
ER +++ PR +++
ER +++ PR +
•
?
ER +++ PR +++
•
ER +++ PR ++
•
•
• •
LATE METESTRUS
ESTRUS
ER ++ ER + PR +++ ER ++++ PR +++ ER +++ PR +++ PR +++ ER ++ ER +++ PR ++ • PR +
MMP-1 ?
•
proliferation ?
ER +++ PR ++++
•
apoptosis ?
P4↓
P4 ↑
ER ++++ PR ++++
•
?
MMP-1 ?
?
ER +++ PR +++ apoptosis ?
• ••
•
•
• ER +++ PR ++
ER ++ PR ++
ER + PR ++
P4 ↑
MMP-1 ? ?
•
• •
MMP-1 ?
• • ? P4↓
•
•
ER ++ PR +++
EARLY METESTRUS
General Introduction
8
The nuclei of these cells were deformed due to fat accumulation. Keller (1909) suggested that these cells transport the fat towards the lumen. Some cells, though small in number, might undergo cell death. During the rest stage, corresponding to anestrus, the fat further disappeared from the surface epithelium which showed a cuboidal appearance at that time. Crypts were present and basal glands were less convoluted and had a lower epithelium. Some secretion was still present. The superficial stroma was rich in cells, whereas the stroma of the deeper part showed a more reticular appearance. No cilia were observed in any cycle stage. In further studies by Arenas and Sammartino (1939) and Mulligan (1942) similar results were found. However, in these studies it was suggested that the lipid-loaded cells during late metestrus and the beginning of anestrus are desquamated (Fig. 6) and replaced by new epithelium that regenerated from the crypts. Barrau et al. also investigated canine endometrial remodeling during the estrous cycle (1975b) and during pregnancy (1975a). These authors found two distinct phases of proliferation, of which the first occurred during proestrus and was characterized by growth of the crypts, whereas the second occurred halfway of estrus and showed proliferation of the basal glands (Fig. 6). During the third week of metestrus the uterus started to involute and by late metestrus much debris had accumulated in the glandular lumina (Fig. 6). By the end of metestrus the glands had returned to the anestrous stage condition. In the crypts and the surface epithelium, numerous lipid droplets and small pleomorphic nuclei were noticed. However, little information relative to the mechanisms involving regression was revealed in these studies. Furthermore, data on proliferation remain controversial as Spanel-Borowski et al. (1984) found only one proliferation peak, viz. during proestrus, for all endometrial cell groups. Only recently, some of the regulatory processes which might be involved in the cyclic remodeling of the canine endometrium have been investigated, i.e. the presence and distribution of steroid hormone receptors (Vermeirsch et al., 1999; 2000) and the expression of matrix metalloproteinases or MMPs (Chu et al., 2002). As described by Brenner and West (1975), steroid hormones are preferentially concentrated in target tissues by soluble proteins called receptors. These receptors act as both signal transducers and transcription factors which appear to regulate many different aspects of development, differentiation or cellular function
General Introduction
9
(Vermeirsch, 2001). The estrogen receptors are encoded by two genes, viz. alpha and beta, which have different or even opposite biological actions (Gustafsson, 1999). Vermeirsch et al. (1999, 2000) verified ER and PR expression in the cyclic canine endometrium and found highest staining scores for ERα and PR during proestrus and lowest scores for both receptors during early metestrus. Except for the PR in the epithelium of the basal glands, the staining scores were also high during anestrus (Fig. 6). No conclusive immunohistochemical results were obtained for ERβ. Furthermore, ERα immunostaining was negatively correlated with the serum progesterone levels, whereas the PR expression was positively correlated with the estradiol-17β : progesterone ratio (Vermeirsch, 2001). Concerning MMP expression, Chu et al. (2002) investigated MMP-2, MMP-7 and MMP-9 activities in the cyclic canine endometrium using zymography. However, using this technique no localization of the investigated MMPs was possible and no data on tissue inhibitors of metalloproteinases or TIMPs were included. In conclusion, the possible function of MMPs in canine endometrial remodeling remains unclear.
Cyclic endometrial remodeling in other species In humans, rodents, ruminants, pigs and horses endometrial remodeling and the basic mechanisms involved have been studied more elaborately. Endometrial remodeling in these species is controlled by several mechanisms, including apoptosis (Hopwood and Levison, 1976; Sato et al., 1997; Dharma et al., 2001; Wasowska et al., 2001), proliferation (Leroy et al., 1969; Conti et al., 1981; Salmi et al., 1998; Gerstenberg et al., 1999; Lai et al., 2000) and matrix changes regulated by MMPs (Rudolph-Owen et al., 1997; Hurst and Palmay, 1999; Salamonsen et al., 2000; Goffin et al., 2003). These mechanisms are influenced by steroid hormones and the presence of steroid hormone receptors (Brenner et al., 1990; Li et al., 1992; Watson et al., 1992; Boos et al., 1996). Furthermore, prominent cyclic changes of the endometrial surface epithelium have been observed in these species using electron microscopy (Ferreira-Dias et al., 1994; Garris, 1998; Spornitz et al., 1999; Nikas et al., 2000; Wick and Kress, 2002). These different aspects will be concisely overviewed in the following paragraphs.
General Introduction
10
Humans In women, high estrogen levels concomitant with high uterine ER levels are present during the follicular phase and have been linked to proliferation in both in vivo and in vitro studies (Brenner and West, 1975). In the human endometrium, proliferation is most prominent in the surface epithelial cells, glandular cells and stromal cells of the functionalis layer during the follicular phase (Salmi et al., 1998). During the early and mid-luteal phase, when estrogen and ER levels are low, proliferation diminished in the surface and glandular epithelium. Using scanning electron microscopy prominent morphological changes of the endometrial surface epithelium, involving pinopode formation, are observed at that time (Nikas et al., 2000). During the late luteal phase, small fissures and some surface defects are present (Nikas et al., 2000) and proliferation is completely absent in the functionalis layer of the endometrium (Salmi et al., 1998). In the basalis layer hardly any proliferation has been found throughout the estrous cycle (Salmi et al., 1998), although ER expression is relatively high during the entire cycle (Press et al., 1984). In vitro studies have shown that progesterone inhibits the estradiol-17β stimulated proliferation of endometrial epithelium after ovulation due to down regulation of the ER levels, and induces differentiation to secretory cells (Dahmoun et al., 1999). Withdrawal of ovarian steroids in the late luteal phase induces a chain of events resulting in menstruation, viz. the shedding of the functionalis layer of the endometrium (Dahmoun et al., 1999). On the second day of menstruation, a prominent increase of apoptosis has been observed in all cells of the functionalis layer of the human endometrium. By contrast, cells of the basal glands do not show any evidence of apoptosis throughout the cycle (Hopwood and Levison, 1976; Kokawa et al., 1996). This lack of apoptosis in the basalis layer has been linked to the persistent expression of ER in this region and is believed to play an important role in the regeneration of the functionalis layer after menstruation (Press et al., 1984). There is now strong evidence that MMPs play a pivotal role in the tissue breakdown at menstruation. The various MMPs have the capacity to degrade all the components of both interstitial matrix and basement membranes. In the human endometrium, MMPs are characteristically found at foci coincident with areas of tissue destruction (Salamonsen et al., 1999). The activity of MMPs in the human endometrium is regulated by steroid hormones but also by the presence of their inhibitors, i.e. TIMPs (Curry and Osteen, 2001). TIMPs-1, -2 and -3 are present throughout the cycle and their concentration seems to be cycle-independent. The increased production of MMPs without a concomitant increase in TIMPs in the endometrium at menstruation is therefore permissive for tissue destruction (Salamonsen et al.,
General Introduction
11
1999). Furthermore, MMP expression seems to be mainly regulated by progesterone as the withdrawal of this hormone results in a massive increase of these proteinases (Marbaix et al., 1992). Rodents In rats and mice, the estrous cycle has an average length of 4 days and can be divided into proestrus, estrus, metestrus and diestrus (Spornitz et al., 1999). Peak levels of estradiol-17β are reached during proestrus just before the LH peak. At the time of this peak, ERα concentration is two times higher than in metestrus (Wang et al., 2000). Progesterone levels increase during proestrus after the LH peak, they are low during estrus and increase slightly in metestrus, are low in diestrus but gradually rise during late diestrus to the peak on the day of proestrus (Brenner and West, 1975). The essential role of ER and PR in mediating uterine responses to estradiol-17β and progesterone has recently been confirmed by deletion of these genes in mice. ER knock-out mice display the inability to respond to the proliferative stimuli of estrogen, while mice lacking PR display estrogen-dependent hyperplasia of the uterine epithelium and stromal hypocellularity (Tibbetts et al., 1998). It has been suggested that a reduction of estradiol-17β levels during estrus and metestrus leads to apoptosis of the rat endometrial surface epithelium (Sato et al., 1997), as has been described in the hamster (Sandow et al., 1979). Furthermore, the differentiation and dedifferentiation of the rat endometrial surface epithelial cells have also been correlated to estradiol-17β, progesterone and their respective receptors (Spornitz et al., 1999). Concerning MMP activity, highest mRNA levels of MMP-3, MMP-7 and MMP-11 were observed during proestrus and estrus, whereas lowest levels were noticed during diestrus. These findings suggest a role for these MMPs during periods of endometrial growth (Rudolph-Owen et al., 1997). Ruminants In cows and ewes, plasma estrogens peak 3 times during each estrous cycle. The first surge triggers a LH release, the second occurs on days 1 to 4 of the cycle, and the third peak occurs on day 13 to 16 in the cow whereas on days 7 to10 in the ewe. Plasma progesterone levels follow a similar pattern in both species. In the cow and the ewe progesterone levels are low during the first few days of the cycle, rise rapidly as the corpus luteum becomes functional
General Introduction
12
and then decline quickly 3 or 4 days before the next estrus (Brenner and West, 1975). The findings of Boos et al. (1996) indicate that estrogens induce uterine ER and PR synthesis, whereas progesterone rather down regulates these receptors. However, exceptions to this rule were also noticed. In vitro studies have shown the production of MMP-1 and MMP-2 by ovine endometrial cells (Salamonsen et al., 1993). These MMPs might play a role in the remodeling of the endometrium in the ewe, but their precise function remains to be elucidated. Pigs In the sow, only one peak of plasma estrogen occurs just before estrus and before the LH peak. Progesterone concentrations rise earlier and reach higher levels during the cycle than observed in the cow and the ewe (Brenner and West, 1975). In this species, massive apoptosis has been noticed in the endometrial surface epithelium and in the endometrial glands during luteolysis, when progesterone and estradiol-17β levels were low. However, apoptosis was still prominent in the surface epithelium during the early follicular phase, when estradiol-17β levels were already high. The porcine endometrial surface epithelium has also been investigated using scanning and transmission electron microscopy (Leiser et al., 1988). Mitoses in these cells were numerous during proestrus and estrus and were still present during early metestrus, whereas practically no more mitotic figures were noticed during diestrus. No data on MMPs in pigs are available. Horses In the mare, estradiol-17β levels rise during pre-estrus and induce the synchronous expression of ER and PR and the proliferation of stromal cells (Aupperle et al., 2000). Low estradiol-17β levels and maximum progesterone values during early diestrus are associated with the highest proliferation rate and hormone receptor expression in epithelial cells. However, regional differences in proliferation in the cyclic equine endometrium have been described by Gerstenberg et al. (1999). Cells of the superficial strata of the endometrium, including the surface epithelium, the gland neck epithelium and the stroma were found to proliferate mainly during estrus, when the estrogen levels were high. In contrast, proliferation of the deeper secretory portions of the endometrial glands was restricted to a brief phase
General Introduction
13
during early diestrus, when progesterone concentrations were increasing. However, proliferation of the secretory gland epithelium is supposed to be a delayed response to estrogens rather than a direct response to progesterone (Gerstenberg et al., 1999). Cyclic morphological changes of the equine endometrial surface epithelium have been evaluated using SEM. In a study by Samuel et al. (1979) high numbers of ciliated cells were found during estrus, whereas these numbers decreased during diestrus. In contrast, no significant differences were observed by Ferreira-Dias et al. (1994), but specific alterations might have been missed as the micrographs were taken at random from a pool of endometrial specimens from mares in estrus and diestrus.
Rationale of the study In numerous species, the interactions of steroid hormones, steroid hormone receptors, apoptosis, proliferation and matrix metalloproteinases play a pivotal role in cyclic endometrial remodeling. These coordinated changes of body tissue, also called homeorhesis (Bauman, 2000), are far less understood in the dog. Recently, the presence and distribution of steroid hormone receptors (Vermeirsch et al., 1999; 2000) and MMP expression (Chu et al., 2002) have also been verified in the cyclic endometrium of dogs. However, the role of MMPs in canine endometrial remodeling still remains enigmatic, as no data on the distribution of MMPs and TIMP expression were included in the latter study. Furthermore, data on endometrial apoptosis in the dog and the ultrastructural changes of the canine endometrial surface epithelium are lacking, whereas it is unclear whether the endometrial proliferation is a biphasic or single event during the estrous cycle (Barrau et al., 1975b, Spanel-Borowski et al., 1984). A detailed knowledge of these parameters is a prerequisite for a better understanding of the physiological changes during pregnancy and for unraveling the pathogenesis of pathological conditions such as the pyometra-cystic endometrial hyperplasia complex.
General Introduction
14
References
Arenas N, Sammartino R. Le cycle sexuel de la chienne. Etude histologique. Bulletin d’Histologie Appliquée à la Physiologie et à la Pathologie 1939; 16: 229-259.
Arthur GH, Noakes DE, Pearson H. The estrous cycle and its control. In: Veterinary Reproduction and Obstetrics (Theriogenology). Ballière Tindall, London 1983: 3-45.
Aupperle H, Özgen S, Schoon H-A, Schoon D, Hoppen HO, Sieme H, Tannapfel A. Cyclic endometrial steroid hormone receptor expression and proliferation intensity in the mare. Equine Veterinary Journal 2000; 32: 228-232.
Barrau MD, Abel JH Jr, Torbit CA, Tietz WJ Jr. Development of the implantation chamber in the pregnant bitch. American Journal of Anatomy 1975a; 143: 115-130.
Barrau MD, Abel JH Jr, Verhage HG, Tietz WJ Jr. Development of the endometrium during the estrous cycle in the bitch. American Journal of Anatomy 1975b; 142: 47-66.
Bauman DE. Regulation of nutrient partitioning during lactation: homeostasis and homeorhesis revisited. In: Ruminant Physiology: Digestion, Metabolism, Growth, Reproduction. Garnsworthy PC, Wiseman J (Eds.), Nottingham University Press, Nottingham 2000: pp. 221-250
Boos A, Meyer W, Schwarz R, Grunert E. Immunohistochemical assessment of oestrogen receptor and progesterone receptor distribution in biopsy samples of the bovine endometrium collected throughout the oestrous cycle. Animal Reproduction Science 1996; 44: 11-21.
Brenner RM, West NB. Hormonal regulation of the reproductive tract in female mammals. Annual Review of Physiology 1975; 37: 273-302.
Brenner RM, West NB, McClellan MC. Estrogen and progestin receptors in the reproductive tract of male and female primates. Biology of Reproduction 1990; 42: 11-19.
Chastain CB, Ganjam VK. Clinical Endocrinology of Companion Animals. Lea and Febiger, Philadelphia 1986; 26-29: 454-471.
Chu PY, Salamonsen LA, Lee CS, Wright PJ. Matrix metalloproteinases (MMPs) in the endometrium of bitches. Reproduction 2002; 123: 467-477.
Concannon PW. Canine physiology of reproduction. In: Burke TJ (Ed.). Small animal reproduction and fertility. A clinical approach to diagnosis and treatment. Lea and Febiger, Philadelphia 1986: 23-77.
General Introduction
15
Conti CJ, Gimenez-Conti IB, Zerbe GO, Gerschenson LE. Differential effects of estradiol-17β and progesterone on the proliferation of glandular and luminal cells of rabbit uterine epithelium. Biology of Reproduction 1981; 24: 643-648.
Curry TE Jr, Osteen KG. Cyclic changes in the matrix metalloproteinase system in the ovary and uterus. Biology of Reproduction 2001; 64: 1285-1296.
Dahmoun M, Boman K, Cajander S, Westin P, Bäckström T. Apoptosis, proliferation, and sex hormone receptors in superficial parts of human endometrium at the end of the secretory phase. Journal of Clinical Endocrinology and Metabolism 1999; 84: 1737-1743.
Dharma SJ, Kholkute SD, Nandedkar TD. Apoptosis in endometrium of mouse during estrous cycle. Indian Journal of Experimental Biology 2001; 39: 218-222.
England GCW. Allen’s fertility and obstetrics in the dog. Sutton JB, Swift ST (Eds.), Blackwell Science Ltd., Oxford 1998: pp. 1-34.
Ferreira-Dias G, Nequin LG, King SS. Morphologic characteristics of equine endometrium classified as Kenney categories I, II, and III, using light and scanning electron microscopy. American Journal of Veterinary Research 1994; 55: 1060-1065.
Garris DR. Scanning electron microscopic and morphometric analysis of the guinea pig uterine luminal surface: cyclic and ovarian steroid-induced modifications. The Anatomical Record 1998; 252: 205-214.
Gerstenberg C, Allen WR, Stewart F. Cell proliferation patterns in the equine endometrium throughout the non-pregnant reproductive cycle. Journal of Reproduction and Fertility 1999; 116: 167-175.
Goffin F, Munaut C, Frankenne F, Perrier D’Hautrive S, Beliard A, Fridman V, Nervo P, Colige A, Foidart JM. Expression pattern of metalloproteinases and tissue inhibitors of matrix-metalloproteinases in cycling human endometrium. Biology of Reproduction 2003; 69: 976-984.
Gustafsson J-A. Estrogen receptor-β - a new dimension in estrogen mechanism of action. Journal of Endocrinology 1999; 163: 379-383.
Hopwood D, Levison DA. Atrophy and apoptosis in the cyclical human endometrium. Journal of Pathology 1976; 119: 159-166.
Hurst PR, Palmay RD. Matrix metalloproteinases and their endogenous inhibitors during the implantation period in the rat uterus. Reproduction Fertility and Development 1999; 11: 395-402.
General Introduction
16
Keller K. Über den Bau des Endometriums beim Hunde mit besonderer Berücksichtigung der cyklischen Veränderungen an der Uterindrüsen. Anatomische Hefte 1909; 118: 312-391.
Kokawa K, Shikone T, Nakano R. Apoptosis in the human endometrium during the menstrual cycle. Journal of Clinical Endocrinology and Metabolism 1996; 81: 4144-4147.
Lai M-D, Lee L-R, Cheng K-S, Wing L-Y C. Expression of proliferating cell nuclear antigen in luminal epithelium during the growth and regression of rat uterus. Journal of Endocrinology 2000; 166: 87-93.
Leiser R, Zimmermann W, Sidler X, Christen A. Normal-zyklische Erscheinungen im Endometrium und am Ovar des Schweines. Tierärztliche Praxis 1988; 16: 261-280.
Leroy F, Galand P, Chrétien J. The mitotic action of ovarian hormones on the uterine and the vaginal epithelium during the oestrous cycle in the rat: a radioautographic study. Journal of Endocrinology 1969; 45: 441-447.
Li W, Boomsma RA, Verhage HG. Immunocytochemical analysis of estrogen and progestin receptors in uteri of steroid treated and pregnant cats. Biology of Reproduction 1992; 47: 1073-1081.
Marbaix E, Donnez J, Courtoy PJ, Eeckhout Y. Progesterone regulates the activity of collagenases and related gelatinases A and B in human endometrial explants. Proceedings of the National Academy of Sciences USA 1992; 89: 11789-11793.
Mulligan RM. Histological studies on the canine female genital tract. Journal of Morphology 1942; 71: 431448.
Nikas G, Makrigiannakis A, Hovatta O, Jones HW Jr. Surface morphology of the human endometrium. Basic and clinical aspects. Annals of the New York Academy of Sciences 2000; 900: 316-324.
Pineda MH. Female reproductive system. In: Veterinary Endocrinology and Reproduction. McDonald LE (Ed.), Lea and Febiger, Philadelphia 1989a: pp. 303-354.
Pineda MH. Reproductive patterns of dogs. In: Veterinary Endocrinology and Reproduction. McDonald LE (Ed.), Lea and Febiger, Philadelphia 1989b: pp. 460-486.
Press MF, Nousek-Goebl N, King WJ, Herbst AL, Greene GL. Immunohistochemical assessment of estrogen receptor distribution in the human endometrium throughout the menstrual cycle. Laboratory Investigation 1984; 51: 495-503.
General Introduction
17
Rudolph-Owen LA, Hulboy DL, Wilson CL, Mudgett J, Matrisian LM. Coordinate expression of matrix metalloproteinase family members in the uterus of normal, matrilysin-deficient, and stromelysin-1 deficient mice. Endocrinology 1997; 138: 4902-4911.
Salmi A, Heikkilä P, Lintula S, Rutanen E-M. Cellular localization of c-Jun messenger ribonucleic acid and protein and their relation to the proliferation marker Ki-67 in the human endometrium. The Journal of Clinical Endocrinology and Metabolism 1998; 83: 1788-96.
Salamonsen LA, Kovacs GT, Findlay JK. Current concepts of the mechanisms of menstruation. Baillière’s Clinical Obstetrics and Gynaecology 1999; 13: 161-179.
Salamonsen LA, Nagase H, Suzuki R, Woolley DE. Production of matrix metalloproteinase 1 (interstitial collagenase) and matrix metalloproteinase 2 (gelatinase A : 72 kDa gelatinase) by ovine endometrial cells in vitro: different regulation and preferential expression by stromal fibroblasts. Journal of Reproduction and Fertility 1993; 98: 583-589.
Salamonsen LA, Zhang J, Hampton A, Lathbury L. Regulation of matrix metalloproteinases in human endometrium. Human Reproduction 2000; 15: 112-119.
Samuel CA, Ricketts SW, Rossdale PD, Steven DH, Thurley KW. Scanning electron microscope studies of the endometrium of the cyclic mare. Journal of Reproduction and Fertility 1979; 27: 287-292.
Sandow BA, West NB, Norman RL, Brenner RM. Hormonal control of apoptosis in hamster uterine luminal epithelium. American Journal of Anatomy 1979; 156: 15-36.
Sato T, Fukazawa Y, Kojima H, Enari M, Iguchi T, Ohta Y. Apoptotic cell death during the estrous cycle in the rat uterus and vagina. The Anatomical Record 1997; 248: 76-83.
Schaefers-Okkens AC, Kooistra HS. Voortplantingsproblematiek van de teef en de poes. Het Diergeneeskundig Memorandum 1996; 43: 52.
Spanel-Borowski K, Schmalz V, Thor-Wiedemann S, Pilgrim C. Cell proliferation in the principal target organs of the dog (Beagle) ovary during various periods of the estrous cycle. Acta Anatomica 1984; 120: 207-213.
Spornitz UM, Socin CD, Dravid AA. Estrous stage determination in rats by means of scanning electron microscopic images of uterine surface epithelium. The Anatomical Record 1999; 254: 116-126.
General Introduction
18
Tibbetts TA, Mendoza-Meneses M, O’Malley BW, Conneely OM. Mutual and intercompartimental regulation of estrogen receptor and progesterone receptor expression in the mouse uterus. Biology of Reproduction 1998; 59: 1143-1152.
Vermeirsch H. Immunohistochemical determination of receptors for sex steroid hormones in the genital tract of the female dog. PhD thesis, Ghent University 2001.
Vermeirsch H, Simoens P, Lauwers H, Coryn M. Immunohistochemical detection of estrogen receptors in the canine uterus and their relation to sex steroid hormone levels. Theriogenology 1999; 51: 729-743.
Vermeirsch H, Simoens P, Hellemans A, Coryn M, Lauwers H. Immunohistochemical detection of progesterone receptors in the canine uterus and their relation to sex steroid hormone levels. Theriogenology 2000; 53: 773-788.
Wang H, Eriksson H, Sahlin L. Estrogen receptors α and β in the female reproductive tract during the estrous cycle. Biology of Reproduction 2000; 63: 1331-1340.
Wasowska B, Ludkiewicz B, Stefanczyk-Krzymowska S, Grzegorzewski W, Skipor J. Apoptotic cell death in the porcine endometrium during the estrous cycle. Acta Veterinaria Hungarica 2001; 49: 71-79.
Watson ED, Skolnik SB, Zaneckosky HG. Progesterone and estrogen receptor distribution in the endometrium of the mare. Theriogenology 1992; 38: 575-580.
Wick R, Kress A. Ultrastructural changes in the uterine luminal and glandular epithelium during the oestrous cycle of the marsupial Monodelphis domestica (grey short-tailed opossum). Cells Tissues Organs 2002; 170: 111-131.
AIMS OF THE STUDY
Aims of the Study
21
Endometrial remodeling in humans, horses, cows and pigs is regulated by interactions of several parameters such as steroid hormones, steroid hormone receptors, apoptosis, cell proliferation and matrix metalloproteinases. Furthermore, prominent cyclic changes of the endometrial surface epithelium have also been observed in these species using electron microscopy. Information on these coordinated changes in the endometrium of the dog is limited. Therefore, the primary aim of this study was to obtain a better insight in the functional morphology of the canine endometrium during the estrous cycle by verifying:
The changes of the canine endometrial surface morphology during different cycle stages using electron microscopy
Cell
death
in
the
canine
endometrium
during
the
estrous
cycle
using
immunohistochemical detection of active caspase-3, TUNEL assay and transmission electron microscopy
Proliferation in the canine endometrium during the estrous cycle by means of immunohistochemical detection of Ki-67 expression and counting of mitotic figures
The changes in connective tissue using Van Gieson staining and immunohistochemical detection of collagenase-1 (MMP-1), which plays a pivotal role in human endometrial remodeling, and of its inhibitor TIMP-1 in the canine endometrium during different cycle stages
Possible correlations between each of these previous parameters and the serum estradiol17β and progesterone levels
CHAPTER
1
ELECTRON MICROSCOPIC CHANGES OF THE CYCLIC CANINE ENDOMETRIUM
Modified from: SCANNING ELECTRON MICROSCOPIC CHANGES OF THE CANINE UTERINE LUMINAL SURFACE DURING ESTRUS AND LATE METESTRUS
Van Cruchten S, Van den Broeck W, Simoens P, Lauwers H
Reproduction in Domestic Animals, 2002; 37: 121-126 CYCLIC CHANGES OF THE CANINE ENDOMETRIUM: AN ELECTRON MICROSCOPIC STUDY Van Cruchten S, Van den Broeck W, Roels F, Simoens P
Cells Tissues Organs, 2003; 173: 46-53
Chapter 1
25
Summary The endometrial surface morphology of 38 dogs during different stages of the estrous cycle was investigated with scanning electron microscopy. The cell surface altered from convex in proestrus and estrus to very variable in early metestrus, flattened in late metestrus and became completely plane in anestrus. Microvilli were numerous and long in proestrus and in estrus, became short and variable in number in early metestrus, decreased further in length in late metestrus and became very short and rare in anestrus. The variable appearance in early metestrus was not influenced by changing the osmolarity of the fixative and might be a physiological process. The number of glandular openings showed little variability throughout the estrous cycle. Ciliated cells were rare but present in all cycle stages except in late metestrus. However, in the latter cycle stage and in anestrus rare single strands were noticed. Transmission electron microscopy was used to determine the inner structure of these strands. Microtubuli were detected in transversal and longitudinal sections but without the 9 + 2 arrangement which is characteristic for cilia. The nature and function of these structures remain unclear.
Chapter 1
26
Introduction The luminal surface morphology of the uterus of different species including human has been studied in different cycle stages with the scanning electron microscope (SEM) (FerreiraDias et al., 1994; Garris, 1998; Spornitz et al., 1999; Nikas et al., 2000; Wick and Kress, 2002). In contrast, the endometrial surface morphology of dogs has not yet been investigated during the estrous cycle with SEM. Therefore this study was framed to verify the canine endometrial surface morphology during the estrous cycle and also to assess the presence of cilia as this item remains unclear (Keller, 1909; Ellenberger and Von Schumacher, 1914). Changes in the endometrial morphology may play an important role in the implantation process in the dog as has been shown in rats and humans. In these latter species, epithelial cells lining the uterine cavity lose their microvilli during the period of receptivity and develop membrane projections or so-called pinopodes which are linked to succesfull implantation (Psychoyos and Mandon, 1971; Enders and Nelson, 1973; Bentin-Ley et al., 1999; Nikas et al., 2000; Nardo et al., 2002). A precise knowledge of the endometrial ultrastructure is also prerequisited for understanding uterine diseases such as pyometra and cystic endometrial hyperplasia (CEH) which form an important pathological entity in the dog and occur mainly in metestrus (Hardy, 1980). SEM has already been performed on uterine samples of dogs suffering from the pyometra-CEH complex but apparently no samples of healthy control dogs in early and late metestrus were included in this study (Schoon et al., 1992). However, morphological changes during these cycle stages may reflect physiological alterations of endometrial cells resulting from the pathogenesis of the disease (Sandholm et al., 1975).
Materials and Methods Animals Uterine samples and serum samples were taken from 31 healthy adult dogs of different breeds that were presented for euthanasia or ovariohysterectomy at the Faculty of Veterinary Medicine in Ghent, Belgium and at 3 veterinary practices. For all dogs data concerning anamnesis, weight, litters and last proestrous bleeding were recorded. The dogs varied in age from 8 months to 7 years. Two dogs were in proestrus, three in estrus, five in early metestrus, 10 in late metestrus and 11 in anestrus.
Chapter 1
27
As sufficient specimens of dogs in proestrus, estrus and early metestrus are difficult to obtain from ovariohysterectomies in practices, we also sampled seven dogs which were acquired at the age of six months with approval of the local ethical committee. These dogs were 2 mongrels, 2 German Shepherds, 1 Boxer, 1 Labrador and 1 Cocker Spaniel. They were held in group in a pen connected with an outdoor run. They had water ad libitum and were fed once a day with a commercial food. The cycle stages of these dogs were followed using serology (Concannon et al., 1975) and vaginal cytology (Thrall and Olson, 1999). The serum estradiol-17β and progesterone levels were determined every two days from the first day of bloody vaginal discharge until the day of the desired cycle stage was reached. On that day the dogs were sedated with medetomidine (Domitor, Smithkline Beecham A.H., Louvain-laNeuve, Belgium) 80 µg/kg intravenously (i.v.) and subsequently euthanized with T61 (Hoechst Animal Health Benelux) 0.3 mL/kg i.v. After euthanasia the serological and cytological findings were confirmed by histologic examination of the uterus and both ovaria (Vermeirsch et al., 1999, 2000). Two dogs were in proestrus, two in estrus and two in early metestrus. The seventh dog was euthanized in late metestrus and served as a control for the other specimens in that cycle stage. The stage of the estrous cycle of each dog was determined using histologic and serological parameters as described by Vermeirsch et al. (1999, 2000) with slight modification. The animals were first sorted by histologic examination of the ovaries and the uterus, combined with the macroscopic evaluation of the female genital tract including both ovaries. After this preliminary morphological classification the animals were further classified according to their serum progesterone levels. Animals with uterine tissue at a proliferative stage and a progesterone level lower than 1 ng/mL were classified as being in proestrus. Animals with proliferative uterine tissue, developing corpora lutea and progesterone levels between 1 ng/mL and 15 ng/mL were classified as being in estrus. Dogs were classified in early metestrus when uterine tissue was proliferative and progesterone levels were above 15 ng/mL in the presence of growing or fully developed corpora lutea, or above 10 ng/mL when regressing corpora lutea were present. Animals with uterine tissue at a secretory stage and progesterone levels lower than 10 ng/mL and higher than 0.5 ng/mL were classified as being in late metestrus. When uterine tissue was at rest and progesterone levels were basal (≤ 0.5 ng/mL) the dogs were in anestrus.
Chapter 1
28
Serum samples taken immediately before surgery or euthanasia were stored at -20°C until assayed for the serum progesterone and/or estradiol-17β levels using a radioimmunoassay technique (Henry et al., 1987). Scanning electron microscopy (SEM) Directly after excision, samples of the cranial, middle and caudal parts of one uterine horn and a sample of the uterine body were flushed with PBS and subsequently fixed in a HEPES buffered 2% paraformaldehyde-2.5% glutaraldehyde solution (pH 7.2; 1100 mOsmol) for 24h to several days. Specimens of the antimesometrial side of the different parts of the uterine horn and the sample of the uterine body were postfixed in an unbuffered 1% osmium tetroxide solution for 2 hours followed by dehydration in ascending grades of alcohol. Subsequently they were critical point dried with CO2, mounted on a metal stub, platinumcoated and examined by a JEOL JSM 5600 LV scanning electron microscope. The morphological characteristics of each specimen are described in Table 2 using the terminology of Garris (1998). For each dog, the number of microvilli was counted in a randomly placed rectangular area of 2 by 2 µm at a magnification of 10,000x. The length of the microvilli was measured in the same manner. The number of openings of the glandular ducts and of the crypts or undeep glands (Barrau et al., 1975) was assessed in a randomly placed rectangular area of 10,000 µm2 at a magnification of 500x. The evaluation of these three parameters was performed twice and the mean is represented in Table 2. Photographs were taken digitally with a JEOL JSM 5600 LV SEM (Figs. 1, 2) and with a JEOL JSM6340F SEM when a higher resolution was required (Figs. 3, 4, 5, 6, 9, 10, 13, 14). Labeling of the figures was performed with Photoimpact (Ulead Systems Inc., Torrance CA, USA). For evaluation of the variable morphology observed in early metestrus samples of six dogs in that stage were fixed in different solutions with varying osmolarity (Table 1).
Chapter 1
29
Table 1. Different solutions with varying osmolarity in which uterine samples of six dogs in early metestrus were fixed. fixative
buffer
NaCl
osmolarity
2.5% glutaraldehyde 2% paraformaldehyde
0.1 M HEPES
0.4%
1100 mOsmol/L
2% glutaraldehyde 2% paraformaldehyde
5 mM HEPES
0.2%
770 mOsmol/L
1.5% glutaraldehyde 2% paraformaldehyde
1 mM HEPES
0.1%
620 mOsmol/L
2.5% glutaraldehyde 2% paraformaldehyde
0.1 M HEPES
1%
1380 mOsmol/L
2.5% glutaraldehyde 2% paraformaldehyde
0.1 M Na-cacodylate
0%
1000 mOsmol/L
2% glutaraldehyde
0.1 M Na-cacodylate
0.5%
450 mOsmol/L
1.5% glutaraldehyde
0.1 M Na-cacodylate
0%
275 mOsmol/L
Transmission electron microscopy (TEM) Directly after excision a uterine sample of all dogs in late metestrus and anestrus and the ampulla of a dog in estrus which was used as a positive control for cilia, were fixed for 3 hours in a cacodylate buffered 2% glutaraldehyde solution at room temperature. These specimens were rinsed in distilled water and then postfixed during 24h at 4°C in an osmiumpotassiumferrocyanide solution (1 mL 4% OsO4, 3 mL Na-cacodylate buffer 0.134 mol/L (pH 7.4) and 66 mg K3Fe(CN)6) (Roels et al., 1995). Subsequently the specimens were rinsed in distilled water, dehydrated in different grades of alcohol and finally embedded in epoxy resin at 60°C. Ultrathin sections were counterstained with uranyl acetate and lead citrate to be examined with a JEOL JEM-100B transmission electron microscope.
Chapter 1
30
Results Proestrus (Figs. 1-3) The endometrial surface had a cobblestone appearance (Fig. 1). All cells were convex and clearly separated from each other by cell borders that were situated on a deeper level than the cell surface (Fig. 2). The cells were covered by relatively long microvilli ranging from 0.5 to 1 µm high and very short microvilli of approximately 0.1 µm high (Table 2). The number of microvilli on the cells varied from 17 to 20 per µm2, while the number of glandular openings was approximately 5 per 10,000 µm2 in all dogs. No morphological differences were detected between the different parts of the uterus except for the presence of cilia. In one dog three ciliated cells on a total of approximately 10,000 cells were noticed in the cranial part of the uterine horn (Fig. 3), whereas no cilia were detected in the other parts and nor in the other dogs. Estrus (Fig. 4) The endometrial surface appearance was similar to that in proestrus. Most cells were convex and covered by long and numerous microvilli (Table 2). The number of glandular openings varied between 4 and 6 per 10,000 µm2 in all dogs except for one dog at the end of estrus in which approximately 10 glandular openings were counted. No differences in morphology were noticed in the various parts of the uterus except for the presence of cilia. Ciliated cells were found in three dogs in estrus but as in proestrus they were not present in each part of the uterus and they were very rare, ranging from 4 to 7 ciliated cells per 10,000 observed cells. Early metestrus (Figs. 5-8) In one dog at the beginning of early metestrus, the same morphological appearance was observed as in dogs in late estrus, but the number of microvilli on the endometrial cells varied strongly as some cells had numerous microvilli and others practically none (Table 2). The other dogs in early metestrus had a very variable aspect of the endometrial surface epithelium. In most cases, the endometrial appearance was smooth and the cell borders were prominent (Fig. 5). The cell surfaces varied strongly ranging from convex in some cells to flat in the
Chapter 1
31
neighboring cells (Fig. 6). Some cells were burst or ruffled. The microvilli were very short (0.1 - 0.2 µm high) and their number ranged from 2 to 30 per µm2. Due to the irregular morphology of the endometrial cells from early metestrus dogs, the effect of varying the osmolarity of the fixative on the endometrial morphology was examined (Table 1). However, the irregular morphology could not be attributed to osmolarity. The glandular openings varied from 4 to 10 per 10,000 µm². Ciliated cells were found in four of the seven dogs in early metestrus. As in proestrus and estrus they were not found in every part of the uterus of these dogs and their number was very small except for one dog in which some 100 ciliated cells on 10,000 observed endometrial cells were found. The presence and nature of these cilia was confirmed by TEM and compared with the ultrastructure of typical cilia in the canine uterine tube (Fig. 8). The 9 + 2 microtubular pattern characteristic for cilia (Fig. 7) was noticed in some transversal sections but several other atypical patterns were also observed (Fig. 7). Late metestrus (Figs. 9-12) The endometrial appearance varied from a mosaic pattern (Fig. 9) and convex cells with relatively long microvilli at the beginning of late metestrus to a smooth pattern and flat cells with short microvilli at the end of metestrus (Table 2). The number of glandular openings was very variable ranging from 3 to 16 per 10,000 µm2. No morphological differences were detected between the various parts of the uterus and in contrast to other cycle stages no ciliated cells were detected in late metestrus. However, long and slender strands were noticed on cells in and surrounding the glandular openings. These strands were variable in length and thickness, bent in different directions and not organized in groups (Fig. 10). TEM was used to determine the inner skeleton of these structures (Fig. 12). The transversal sections showed either several single microtubuli (Fig. 11) or microtubular doublets that were not strictly organized and that were variable in number. None of the observed strands showed the characteristic 9 + 2 microtubular pattern of cilia (Fig. 7).
Table 2. SEM endometrial morphology and mean hormone profiles ± s.d. (P4: progesterone (ng/ml) and E2: estradiol-17β (pg/ml)) of 4 dogs in proestrus, 5 in estrus, 7 in early metestrus, 11 in late metestrus and 11 in anestrus. Cycle stage
Serum levels
General
Cell
Cell surface Microvilli Microvilli
Glandular
endometrial boundaries appearance
length
number
openings
morphology
(µm)
(per µm²)
number (per 10,000 µm²)
P4 Proestrus 0.31 ± 0.13
E2 33.50 ± 12.02
Cobblestone
prominent
convex
< 0.1 - 1
17 - 20
5-6
Estrus
7.75 ± 4.11
19 ± 21.53
Cobblestone
variable
convex
0.5 - 2
20 - 35
4 - 11
Early
16.44 ± 5.19
-
Variable
prominent
variable
0.1 - 0.2
variable
3 - 10
1.79 ± 1.24
-
mosaic to
prominent
convex - flat
< 0.1 - 1
5 - 35
3 - 16
prominent
usually flat
< 0.1
5 - 30
4 - 13
Metestrus Late
smooth
Metestrus Anestrus
0.32 ± 0.16
-
mosaic to smooth
-, not analyzed
Chapter 1
33
Anestrus (Figs. 13-14) In anestrus the endometrial appearance was smooth in most cases (Fig. 13). The cell surfaces were flat and the number of microvilli was very small ranging from 5 to 9 per µm2 (Table 2). However, in three dogs the cells were rather convex and in four dogs the number of microvilli ranged from 19 to 30 per µm2. As in early and late metestrus the number of glandular openings was very variable ranging from 4 to 15 per 10,000 µm2. Cilia were found in two dogs in anestrus. They were rare, in one of these dogs only one ciliated cell was found along with single long and slender strands that were found around the glandular openings (Fig. 14). These strands were similar to those observed in late metestrus.
Figs. 1-3. SEM of the canine endometrium in proestrus. 1 The general endometrial surface is arranged in a cobblestone pattern. Notice the ciliated cell (arrow). Bar = 10 µm. 2 Close-up view of the endometrial surface. The cells are convex with numerous long (0.5-1 µm) and short (0.1 µm) microvilli. Bar = 1 µm. 3 Close-up view of a ciliated cell. Notice the short microvilli on the cells surrounding the ciliated cell. Bar = 1 µm. Fig. 4. SEM of the canine endometrium in estrus. Close-up view of the endometrial surface. The cell boundaries are masked by the numerous and long microvilli (2 - 3 µm long). Bar = 2 µm. Figs. 5-6. SEM of the canine endometrium in early metestrus. 5 The general endometrial surface is variable. Notice the two ciliated cells. Bar = 10 µm. 6 Higher magnification of the endometrial surface. Some cells are convex, some are flat. Bar = 3 µm. Figs. 7-8. TEM of cilia in the canine endometrium and in the canine uterine tube. 7 Transversal sections of cilia in the canine endometrium. A high variability of the inner ciliary structure can be noticed. Two sections show the 9 + 2 microtubular pattern which is characteristic for cilia (large arrows), the other sections have atypical features. In some cilia the two central single microtubuli are absent (large arrowhead), some show one single central microtubule (small arrow) and in the center of some cilia one or more doublets of microtubuli are present (small arrowheads). Bar = 100 nm. 8 Transversal sections of cilia in the canine uterine tube. Notice the characteristic 9 + 2 microtubular pattern present in all cilia. Bar = 100 nm.
Chapter 1
34
11
2
3
4
5
6
7
8
Chapter 1
35
9
10
11
12
13
14
Figs. 9-10. SEM of the canine endometrium in late metestrus. 9 The general endometrial appearance is arranged in a mosaic pattern. Cell boundaries of most cells are situated at a deeper level than the endometrial surface but some are protruding into the uterine lumen. Bar = 10 µm. 10 Higher magnification of the endometrial surface. The cell surfaces are covered by very short microvilli (0.1 µm). Long and slender strands are noticed on cells surrounding the endometrial crypts (arrows). Bar = 2 µm. Figs. 11-12. TEM of long and slender strands in late metestrus. 11 Transversal section of a strand. Notice the 6 + 1 single microtubular pattern. No doublets of microtubuli are present. Bar = 50 nm. 12 Longitudinal sections of strands. No concentric arrangement of microtubuli can be noticed. Bar = 500 nm. Figs. 13-14. SEM of the canine endometrium in anestrus. 13 The general endometrial appearance is smooth. The cell boundaries are prominent and protruding into the uterine lumen. Notice the ciliated cell (box) and the long and slender strands (arrows). Bar = 10 µm. 14 Close-up view of a ciliated cell (box). Long and slender strands (arrows) are present around this cell. Bar = 2 µm.
Chapter 1
36
Discussion In several species, such as the rat, the mouse and the guinea pig, morphological and physiological differences between the mesometrial and antimesometrial side of the uterus have been discussed (Jayatilak et al., 1989; Gu et al., 1994; Makker et al., 1994; Das et al., 1997). As SEM studies in these species (Garris, 1998; Spornitz et al., 1999) focus mainly on the antimesometrial side because of its role in implantation, we also investigated that part of the uterine wall of the dog to be able to compare our findings with those in other species. The results of the present study are similar to findings described in the rat and the guinea pig in which the cyclic changes of the endometrial surface are under ovarian steroid hormone control (Anderson et al., 1975; Garris, 1998; Spornitz et al., 1999). This suggests that the cyclic changes of the endometrial surface in dogs are influenced by progesterone and estradiol-17β as well. Morphological changes were most pronounced in early metestrus when a very variable aspect of the endometrium was noticed. This variability could not be explained by a different osmolarity of the cells at that stage, since no morphological differences were observed when different fixatives with varying osmolarity were used. The variable appearance of the canine endometrium coincides with the period of receptivity for the blastocyst. In this period the endometrial epithelial cells of rats and humans lose their microvilli and develop membrane projections or so-called pinopodes that are linked to successful implantation (Psychoyos and Mandon, 1971; Enders and Nelson, 1973; Bentin-Ley et al., 1999; Nikas et al., 2000; Nardo et al., 2002). In the dog no such structures were detected, but the number and length of microvilli had strongly decreased and the cell surfaces were irregular. In that period morphological changes of the apical and basolateral plasma membrane of human endometrial surface epithelial cells have been linked to a reorganization of their cytoskeleton with an increase in adhesion molecules (Denker, 1993; Thie et al., 1995, 1997). If this mechanism is also present in canine endometrial cells during metestrus, the increase in adhesion molecules might also play a role in the pyometra-CEH complex as Escherichia coli, which is a major pathogen in this disease complex, has a higher binding to the surface epithelium in metestrus than in other cycle stages. An alternative explanation of the irregular surface of the endometrial cells in early metestrus might be the presence of apoptosis. Apoptosis of the luminal surface epithelium at that period has been described in rats (Sato et al., 1997; Spornitz et al., 1999). In our study some cells were ruffled and resembled apoptotic cells described in a
Chapter 1
37
SEM study of the human endometrium (Nikas et al., 2000). This item should be elucidated by using immunohistochemical markers for apoptosis (Gavrieli et al., 1992; Saunders et al., 2000). In our study also some other parameters such as the number of glandular openings and the presence of endometrial cilia were studied. In contrast to other species (Garris, 1998; Spornitz et al., 1999) we found practically no difference in number of glandular openings in estrus and early metestrus as compared to the other cycle stages. This might reflect a species-specific characteristic or might be due to a higher folding of the endometrial surface which might have masked the openings. The ciliary pattern also showed no cyclic variation. As endometrial cilia are rare and apparently not cycle-dependent, it is doubtful that they are involved in creating a sperm reservoir in the canine uterus as has been suggested for the canine uterine tube (Pacey et al., 2000). The small number of ciliated cells also explains why even extensive histologic studies previously failed to detect cilia in the canine endometrium (Keller, 1909). Finally, a remarkable observation was the presence of rare and single strands in and around the glandular openings of several dogs in late metestrus and anestrus. These strands were more flexed than cilia, were variable in length and not organized in groups. The inner structure of these strands corresponds with the ultrastructure of so-called primary cilia which have been described in different organs of several species. The function of these primary cilia remains unknown (Bertelli and Regoli, 1994). Acknowledgements The authors wish to thank Prof. Dr. H. Lauwers for scientific advice, Dr. A. Boone, Dr. P. Herbots and Dr. I. Christiaens for their supply of specimens, and Dr. B. Hertsens, B. De Pauw and D. Jacobus for their excellent technical assistance with scanning and transmission electron microscopy. We are also very grateful to the staff of the Centre for Materials Advice and Analysis of the VITO for their cooperation in supplying the JEOL JSM-6340F scanning electron microscope.
Chapter 1
38
References
Anderson WA, Kang Y, DeSombre ER. Estrogen and antagonist-induced changes in endometrial topography of immature and cycling rats. The Journal of Cell Biology 1975; 64: 692-703.
Barrau MD, Abel JH Jr, Verhage HG, Tietz WJ Jr. Development of the endometrium during the estrous cycle in the bitch. American Journal of Anatomy 1975; 142: 47-66.
Bentin-Ley U, Sjögren A, Nilsson L, Hamberger L, Larsen JF, Horn T. Presence of uterine pinopodes at the embryo-endometrial interface during human implantation in vitro. Human Reproduction 1999; 14: 515-520.
Bertelli E, Regoli M. A morphological study of the primary cilia in the rat pancreatic ductal system: ultrastructural features and variability. Acta Anatomica 1994; 151: 194-197.
Concannon PW, Hansel W, Visek WJ. The ovarian cycle of the bitch: plasma estrogen, LH and progesterone. Biology of Reproduction 1975; 13: 112-121.
Das SK, Yano S, Wang J, Edwards DR, Nagase H, Dey SK. Expression of matrix metalloproteinases and tissue inhibitors of metalloproteinases in the mouse uterus during the peri-implantation period. Developmental Genetics 1997; 21: 44-54.
Denker H-W. Implantation: a cell biological paradox. Journal of Experimental Zoology 1993; 266: 541-558.
Ellenberger W, Von Schumacher S. Grundriss der vergleichenden Histologie der Haussäugetiere. Verlagsverhandlung, Paul Parey IV, Berlin 1914: pp. 269-273.
Enders AC, Nelson D. Pinocytotic activity of the uterus of the rat. American Journal of Anatomy 1973; 138: 277-300.
Ferreira-Dias G, Nequin LG, King SS. Morphologic characteristics of equine endometrium classified as Kenney categories I, II, and III, using light and scanning electron microscopy. American Journal of Veterinary Research 1994; 55: 1060-1065.
Garris DR. Scanning electron microscopic and morphometric analysis of the guinea pig uterine luminal surface: cyclic and ovarian steroid-induced modifications. The Anatomical Record 1998; 252: 205-214.
Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. The Journal of Cell Biology 1992; 119: 493-501.
Chapter 1
39
Gu Y, Jow GM, Moulton BC, Lee C, Sensibar JA, Park-Sarge OK, Chen TJ, Gibori G. Apoptosis in decidual tissue regression and reorganization. Endocrinology 1994; 135: 1272-1279.
Hardy RM. Cystic endometrial hyperplasia-pyometra complex. In: Current Therapy in Theriogenology: diagnosis, treatment and prevention of reproductive diseases in animals. Schille VM (Ed.), W.B. Saunders, Philadelphia, 1980: 624-630.
Henry M, Figueiredo AEF, Palhares MS, Coryn M. Clinical and endocrine aspects of the oestrous cycle in donkeys (Equus asinus). Journal of Reproduction and Fertility Supplement 1987; 35: 297-303.
Jayatilak PG, Puryear TK, Herz Z, Fazleabas A, Gibori G. Protein secretion by mesometrial and antimesometrial rat decidual tissue: evidence for differential gene expression. Endocrinology 1989; 125: 659-666.
Keller K. Über den Bau des Endometriums beim Hunde mit besonderer Berücksichtigung der cyklischen Veränderungen an der Uterindrüsen. Anatomische Hefte 1909; 118: 312-391.
Makker A, Singh MM, Chowdhury SR, Maitra SC, Kamboj VP. Uterine estradiol and progesterone receptor concentration in relation to circulating hormone levels and histoarchitecture during high endometrial sensitivity and induced decidualization in guinea pigs. Journal of Steroid Biochemistry and Molecular Biology 1994; 48: 535-543.
Nardo LG, Sabatini L, Rai R, Nardo F. Pinopode expression during human implantation. European Journal of Obstetrics and Gynecology and Reproductive Biology 2002; 101: 104-108.
Nikas G, Makrigiannakis A, Hovatta O, Jones HW Jr. Surface morphology of the human endometrium. Basic and clinical aspects. Annals of the New York Academy of Sciences 2000; 900: 316-324.
Pacey AA, Freeman SL, England GC. Contact of dog spermatozoa with homologous uterine tube epithelium prolongs flagellar activity in relation to the stage of the estrous cycle. Theriogenology 2000; 54: 109-118.
Psychoyos A, Mandon P. Scanning electron microscopy of the surface of the rat uterine epithelium during delayed implantation. Journal of Reproduction and Fertility 1971; 26: 137-138.
Roels F, De Prest B, De Pestel G. Liver and chorion cytochemistry. Journal of Inherited Metabolic Disease 1995; 18 Suppl. 1: 155-171.
Sandholm M, Vasenius H, Kivistö AK. Pathogenesis of canine pyometra. Journal of the American Veterinary Medical Association 1975; 167: 1006-1010.
Chapter 1
40
Sato T, Fukazawa Y, Kojima H, Enari M, Iguchi T, Ohta Y. Apoptotic cell death during the estrous cycle in the rat uterus and vagina. The Anatomical Record 1997; 248: 76-83.
Saunders PA, Cooper JA, Roodell MM, Schroeder A, Borchert CJ, Isaacson AL, Schendel MJ, Godfrey G, Chahill DR, Walz AM, Loegering RT, Gaylord H, Woyno IJ, Kaluyzhny AE, Krzyzek RA, Mortari F, Tsang M, Roff CF. Quantification of active caspase-3 in apoptotic cells. Analytical Biochemistry 2000; 284: 114-124.
Schoon H-A, Schoon D, Nolte I. Untersuchungen zur Pathogenese des “Endometritis-PyometraKomplexes” der Hündin. Journal of Veterinary Medicine A 1992; 39: 43-56.
Spornitz UM, Socin CD, Dravid AA. Estrous stage determination in rats by means of scanning electron microscopic images of uterine surface epithelium. The Anatomical Record 1999; 254: 116-126.
Thie M, Harrach-Ruprecht B, Sauer H, Fuchs P, Albers A, Denker HW. Cell adhesion to the apical pole of epithelium. European Journal of Cell Biology 1995; 66: 180-191.
Thie M., Herter P, Pommerenke H, Dürr F, Sieckmann F, Nebe B, Rychly J, Denker HW. Adhesiveness of the free surface of a human endometrial monolayer for trophoblast as related to actin skeleton. Molecular Human Reproduction 1997; 3: 275-283.
Thrall MA, Olson PN. Chapter 20: The Vagina. In: Diagnostic Cytology and Hematology of the Dog and Cat (2nd ed.). Mosby Inc., St. Louis, Missouri 1999: pp. 240-248.
Vermeirsch H, Simoens P, Hellemans A, Coryn M, Lauwers H. Immunohistochemical detection of progesterone receptors in the canine uterus and their relation to sex steroid hormone levels. Theriogenology 2000; 53: 773-788.
Vermeirsch H, Simoens P, Lauwers H, Coryn M. Immunohistochemical detection of estrogen receptors in the canine uterus and their relation to sex steroid hormone levels. Theriogenology 1999; 51: 729-743.
Wick R, Kress A. Ultrastructural changes in the uterine luminal and glandular epithelium during the oestrous cycle of the marsupial Monodelphis domestica (grey short-tailed opossum). Cells Tissues Organs 2002; 170: 111-131.
CHAPTER
2
CELL DEATH IN THE CYCLIC CANINE ENDOMETRIUM
2.1 Morphological and biochemical aspects of apoptosis, oncosis and necrosis
Modified from: MORPHOLOGICAL AND BIOCHEMICAL ASPECTS OF APOPTOSIS, ONCOSIS AND NECROSIS Van Cruchten S, Van den Broeck W Anatomia Histologia Embryologia, 2002; 31: 214-223
Chapter 2.1
45
Summary Recent investigations have demonstrated the need for a precise differentiation of various forms of cell death such as apoptosis, oncosis, necrosis and programmed cell death. Apoptosis is marked by cellular shrinking, condensation and margination of the chromatin and ruffling of the plasma membrane with eventually breaking up of the cell in apoptotic bodies. Cell death marked by cellular swelling should be called oncosis, whereas the term necrosis refers to the morphological alterations appearing after cell death. Apoptosis and oncosis are therefore premortal processes, while necrosis is a postmortal condition. The term programmed cell death refers to the “fixed” pathway followed by dying cells, whether or not with the characteristic morphology of apoptosis. Three mechanisms are actually known to be involved in the apoptotic process: a receptor-ligand mediated mechanism, a mitochondrial pathway and a mechanism in which the endoplasmic reticulum plays a central role. All three mechanisms activate caspases which are responsible for the characteristic morphological changes observed during apoptosis. A review of the different methods used for detecting apoptotic cells demonstrates that most of these techniques are not entirely specific.
Chapter 2.1
46
Introduction Cell death research has fascinated scientists of different disciplines for more than a century and was enhanced in 1972 when the term apoptosis was launched (Kerr, 1972) (Fig. 1). The increased insight in the different mechanisms of the apoptotic process clearly demonstrated its complexity and the difficulty to differentiate the various forms of cell death. Apoptosis and necrosis were previously considered as two different forms of cell death, but nowadays the distinction between both forms of cell death is less clear, as stated by Farber (1994): ”There is no field of basic cell biology and cell pathology that is more confusing and more unintelligible than is the area of apoptosis versus necrosis”. The present paper presents a short overview of the known apoptotic mechanisms and the various morphological forms of cell death, together with a critical survey of the different detection methods. 13389
14000 12000
10659
10000 number
8000 6000 4000 2000 0
2745 3
8
26
1972
1980
1985
114 1990
1995
2000
2003
year of publication
Fig. 1. Number of papers on apoptosis quoted in Medline public domain (search 2004, key word: apopto*).
History Recent apoptosis research is an elaboration of the studies by Kerr (1971) who induced atrophy of the liver in a rat by ligating a major branch of the portal vein. This resulted in hepathocyte death characterized by peculiar morphological features including decreased cell volume, ruffled cell membranes, condensed chromatin beneath the nuclear envelope, and eventually cell segregation with the forming of numerous vesicles containing intact organelles. This phenomenon was first called shrinking necrosis. A year later this name was
Chapter 2.1
47
altered in apoptosis, the Greek word for falling of leaves (Kerr, 1972). A next important step in apoptotic research was the discovery of DNA fragmentation in induced cell death. These fragments showed a typical ladder structure in gel electrophoresis, which suggested that the fragments were multiples of nucleosomes. The link between this ladder pattern and apoptosis was made by Wyllie and coworkers (1980). A third major event in apoptosis research were the findings of Ellis and Horvitz (1986) in the nematode Caenorhabditis elegans which will be further described below.
Morphology Pyknosis, karyorhexis, karyolysis, chromatolysis Cell death was first described by Virchow in 1859. At that time the terms degeneration, mortification and necrosis only referred to macroscopic observations. The first microscopic denominations for cell death appeared in 1879 with the introduction of the terms karyorhexis indicating the disintegration of the nucleus and karyolysis describing the disappearance of the nucleus. Ten years later the terms pyknosis and chromatin margination were introduced (Arnheim, 1890). In 1885 the term chromatolysis was introduced by Flemming who studied ovarian follicles of mammals and described disintegrating and eventually disappearing nuclei in epithelial cells of atretic follicles. Flemming (1885) called this process chromatolysis because he noticed that these nuclei eventually vanished. His detailed drawing of this process clearly depicts pyknotic chromatin and apoptotic bodies in the follicular lumen. The word chromatolysis was used later on by Gräper (1914) as an opposite to mitosis for describing cell elimination in the course of which cell debris is phagocyted by neighboring epithelial cells. This important viewpoint has been disregarded until the introduction of the new term apoptosis in 1972 (Kerr, 1972). Apoptosis, necrosis and oncosis (Figures 2 and 3) Apoptosis is marked by cellular shrinking, condensation and margination of the chromatin and ruffling of the plasma membrane, called budding. Eventually the cell becomes divided in apoptotic bodies which consist of cell organelles and/or nuclear material surrounded by an intact plasma membrane. These bodies are mostly engulfed by neighboring cells and in
Chapter 2.1
48
particular by macrophages (Kerr, 1972). The macrophage recognizes the apoptotic cell fragments by their expression of phosphatidylserine on the outside of the plasma membrane (Fadok et al., 1992). Other mechanisms of phagocytosis are mediated by vitronectin receptors or certain carbohydrates (Duvall et al., 1985; Savill et al., 1990). However, apoptotic cells are not always recognized by phagocytes and then they may undergo so-called secondary (Sanders and Wride, 1995) or apoptotic necrosis (Majno and Joris, 1995). This terminology seems strange, because necrosis and apoptosis are often considered as two opposite forms of cell death (Wyllie et al., 1980). However, according to Majno and Joris (1995) necrosis is not a form of cell death but the end stage of any cell death process. In this viewpoint saying that a cell dies by necrosis is the same as equalizing clinical death with postmortal decay (Majno and Joris, 1995). Necrosis is marked by cellular swelling, often accompanied by chromatin condensation and eventually leading to cellular and nuclear lysis with subsequent inflammation (Wyllie et al., 1980) (Fig. 2). Apparently, apoptotic cells that are not phagocyted show several of these necrotic features, except for an inflammation process. This may be due to the fact that the few apoptotic cells or bodies which are not recognized by phagocytes produce a concentration of chemo-attractive molecules that is too low to mobilize inflammatory cells (Majno and Joris, 1995).
APOPTOSIS
ONCOSIS
APOPTOSIS
NECROSIS blebbing
budding
NECROSIS
PHAGOCYTOSIS, INFLAMMATION
Fig. 2. Illustration of apoptosis and necrosis according to Kerr et al. (1995).
PHAGOCYTOSIS BY MACROPHAGES OR NEARBY CELLS
Fig. 3. Illustration of apoptosis, oncosis and necrosis according to Majno and Joris (1995).
Chapter 2.1
49
Another important descriptive term is oncosis, the Greek word for swelling, introduced almost one century ago by von Recklinghausen (1910), who observed a manifest cellular swelling of osteocytes in osteomalacia. Majno and Joris (1995) proposed to use the term oncosis for designating any cell death characterized by a marked cell swelling, while the term necrosis refers only to the features which appear after the cell has died (Fig. 3). This concept of the term oncosis has found increasing acceptance in recent studies (Park et al., 2000), and will also be used further in this paper. In descriptions of apoptosis and oncosis the ancient terms pyknosis, karyorhexis and karyolysis are still significant. Pyknosis and karyorhexis are common features of both processes (Sanders and Wride, 1995) whereas karyolysis seems more specific for oncosis, although apoptotic cells eventually also undergo nuclear lysis when they are phagocyted. These three terms describing nuclear changes are therefore not specific for either apoptosis or oncosis. Apoptosis and programmed cell death Apoptosis and programmed cell death are often used as synonyms (Savill et al., 1990), and in contrast to necrosis they have never been associated with inflammation phenomena. On the other hand many cytotoxic agents can also induce apoptotic changes that are evidently not planned by cellular programmation. Both terms are only synonyms in so far the denomination “programmed” is used as a reference to a fixed pathway followed by dying cells (Schwartz et al., 1990). Cell death occurring during organogenesis does not always present the clear morphological characteristics of apoptosis (Sanders and Wride, 1995; Schwartz et al., 1993). Programmed cell death without the typical morphological features of apoptosis is particularly well-known in invertebrates (Clarke, 1990; Schwartz et al., 1993), but has also been described in mammals and birds (Chu-Wang et al., 1990; Martin and Johnson, 1991). For this reason Clarke (1990) has proposed a different classification of cell death describing three morphological entities namely apoptosis, autophage cell death with numerous autolysosomes, and cytoplasmic cell death. The latter is very similar to oncosis and can be further divided in
Table 1. Classification of cell death according to Clarke (1990).
Various designations
Nucleus
Cell membrane
Cytoplasm
Heterophage elimination
Type 1
Apoptosis; shrinkage necrosis; precocious pyknosis; nuclear type of cell death
Nuclear condensation, clumping of chromatin leading to pronounced pyknosis
Convoluted, forming blebs
Loss of ribosomes from Prominent and important RER and polysomes; cytoplasm reduced in volume and becoming electron-dense
Type 2
Autophage cell death
Pyknosis in some cases, nuclear segments may bleb or segregate
Endocytosis, blebbing possible
Abundant autophage vacuoles; ER and mitochondria sometimes dilated; Golgi apparatus often enlarged
Occasional and late
Type 3A
Non-lysosomal disintegration
Late vacuolization followed by disintegration
Breaks
General disintegration; dilation of organelles, forming empty spaces that fuse with each other and with the extracellular space
Absent
Type 3B
Cytoplasmic type
Late increase of chromatin granularity
Cell swelling
Vacuolization caused by dilation of the ER, nuclear envelope, Golgi apparatus and sometimes mitochondria
Present
Chapter 2.1
51
two subtypes (Table 1). Transient forms of cell death have also been described, including an increase of autophage activity during ontogenesis (Lockshin and Zakeri, 1991). Some authors suggest that these different morphological forms of programmed cell death are due to the cell type rather than to any basic cellular mechanism (Lockshin and Zakeri, 1991). Others suggest that the variability in apoptotic appearance is dependent on the provoking stimulus (Schwartz et al., 1993). From the foregoing it can be concluded that programmed cell death is sometimes and enigmatically devoid of the characteristic morphology of apoptosis.
Mechanisms As mentioned earlier, the work of Ellis and Horvitz (1986) was an important breakthrough in the unwiring of the apoptotic process. By quantitative analysis of cell death in the nematode Caenorhabditis elegans, these researchers discovered that during embryogenesis the precise number of 131 out of 1090 somatic cells died. Subsequent genetic studies showed that 10 ced (cell death defective) genes are involved in this process, of which ced-3, ced-4 and ced-9 play an essential role in apoptosis. Ced-3 and ced-4 induce apoptosis, whereas ced-9 inhibits apoptosis. The proteins coded by ced-3 and ced-4 are very homologous to the cysteindependent protease “Interleukin 1β converting enzyme” (ICE) in vertebrates (Martin and Green, 1995). Subsequently, various proteases related to ICE were detected and designated by numerous synonyms, including the term caspases (cystein aspartate). All cells contain caspases that are present as inactive zymogens. They are further classified into initiator or upstream caspases (caspase-1, -2, -4, -5, -8, -9 and -10) and effector or downstream caspases (caspase-3, -6, -7) depending on their role after activation in the so-called caspase cascade. Both subtypes show a different pattern of activation. In contrast to effector caspases, which are activated by proteolytic cleavage, initiator caspases can also be activated without cleavage, viz. by dimerization (Donepudi and Grütter, 2002). Different studies have shown the link between these initiator and effector caspases. Inhibitors of caspase-1 and caspase-3 can suppress apoptosis (Enari et al., 1995). Because apoptosis induced by caspase-1 is not only inhibited by caspase-1 inhibitors but also by caspase-3 inhibitors, it has been suggested that caspase-1 and caspase-3 are linked to each other (Enari et al., 1995). Caspase-3 is known to splice various proteins including poly(ADP)-ribose polymerase (PARP) and is essential for the typical characteristics of apoptosis (Shiokawa et al., 1997).
Chapter 2.1
52
Fas ligand
INTERSTITIUM binding of Fas ligand on Fas
Fas RECEPTOR
cell membrane
activation of the death domains
Fas death domains
binding
CYTOPLASM
activation of FADD
FADD
procaspase-8 binds FADD
procaspase-8
caspase-8
activation of procaspase-8
activation
procaspase-3
caspase-3
breaking up of cytoplasmic and nuclear proteins + DNA
Fig. 4A. Schematic representation of caspase-3 activation by Fas ligand - Fas binding.
TNFα
INTERSTITIUM binding of TNFα on TNFR1
TNFR1 RECEPTOR
cell membrane activation of the death domains
TNFR1 death domains
binding
TRADD
activation of TRADD
binding
CYTOPLASM FADD
activation of FADD
procaspase-8 binds FADD
activation of procaspase-8
procaspase-3
caspase-8 activation caspase-3
procaspase-8
procaspase-3
Fig. 4B. Schematic representation of caspase-3 activation by binding of TNFα on TNFR1.
caspase-3
breaking up of cytoplasmic and nuclear proteins + DNA
Chapter 2.1
53
The important role of caspases in the apoptotic process is also evidenced by the effect of caspase-6 (Takahashi et al., 1996) which causes disintegration of the nucleus by splicing the protein lamine, an important structural protein of the nuclear membrane. Actual data indicate that three major mechanisms can lead to caspase activation: a receptor-ligand binding with activation of caspase-8, a mitochondrial mechanism with activation of caspase-9, and a process involving the endoplasmic reticulum with activation of caspase-12. These three mechanisms are briefly elucidated in the next paragraphs. Receptor-ligand mediated mechanism Apoptosis can be induced by binding of a ligand on a specific receptor which is located on the cell membrane (Fig. 4A, 4B and 4C). Procaspases are converted into active caspases after the receptor is activated by ligand binding. Well-known ligands are the Fas ligand and Tumor Necrosis Factor α (TNFα)(Saikumar et al., 1999). The Fas ligand has Fas as receptor, while TNFα binds to the TNFR1. Both receptors contain an analogous cytoplasmic domain responsible for the signal-transduction in apoptosis. This domain is called the “death domain” (DD)(Boldin et al., 1995). Experiments with yeast cells have revealed that two molecules are linked to the death domain: the Fas associated death domain (FADD) for Fas (Chinnaiyan et al., 1996), and the TNFR1 associated death domain (TRADD) for TNFR1 (Hsu et al., 1995). TRADD mediates apoptosis by binding FADD, so eventually both Fas and TNFR1 use FADD to transduce the death signal. Further signal transduction happens for both receptors via caspase-8, which was originally called FADD-like Interleukin-1β-converting enzyme (FLICE) (Enari et al., 1995). Procaspase-8 binds with its death domain to FADD and becomes activated to caspase-8. TRADD activates caspase-8 indirectly by association with FADD. Eventually, activated caspase-8 results in the conversion of procaspase-3 into the effector caspase-3 which seems to be essential for the typical morphology of apoptosis (Shiokawa et al., 1997). A highly specialized form of Fas mediated apoptosis is caused by cytotoxic T-lymphocytes. After Fas-Fas ligand binding the target cell membrane becomes perforated with perforins, and subsequently the protease granzyme B is released in the cytoplasm. Granzyme B is an enzyme that splices proteins after the amino acid aspartate in the same way as caspases do, resulting in a caspase-cascade with the characteristic morphological changes (Martin and Green, 1995).
Chapter 2.1
54
Other receptor-ligand mediated forms of apoptosis are known, but their exact mechanism is still unclear. DR3 (also known as Apo-3 or LARD) is a death receptor (DR) that induces activation of caspase-8 by binding to the ligand Apo-3L (Chinnaiyan et al., 1995). TRAILR2, DR4 and DR5 have TRAIL (TNF-Related Apoptosis Inducing Ligand or Apo-2L) as ligand (Walczak et al., 1997). Other death receptors such as DR6, CD27, CD134, CD137, CD30 and CD40 are known, but their ligands and their working mechanisms still have to be revealed (Ashkenazi and Dixit, 1999). Receptor-ligand mediated cell death does not always involve apoptosis but can also result in oncosis, depending on the molecular mechanisms involved (Denecker et al., 2001). Furthermore, possible switches between these different forms of cell death have been reported (Van den Berghe et al., 2003). In this latter study, inhibition of the 90-kDa heat shock protein (HSP90) shifted the response of L929 fibrosarcoma cells to TNF from oncosis to apoptosis. Other studies showed higher sensitivity of these cells to TNF induced oncosis by inhibition of caspases (Vercammen et al., 1998). Recently, it has also been demonstrated that binding of Fas or TRAIL to certain receptors does not induce cell death. Therefore, these receptors are called decoy receptors (DcR). These receptors bind the ligand but do not transduce the signal. DcR1 and DcR2 have TRAIL as a ligand, DcR3 has a high affinity for the Fas ligand. These decoy receptors are present in normal tissue and might protect cells against cell death (Ashkenazi and Dixit, 1999). Mitochondrial mechanism The receptor-ligand mediated apoptotic pathway is not the only mechanism in the apoptotic process. Certain cytotoxic agents, such as nitrogen monoxide and radiation, cause apoptosis in another manner involving the mitochondria and more specifically the mitochondrial protein cytochrome c (Boscá and Hortelano, 1999; Mathieu et al., 1999) (Fig. 5A). Cytochrome c is localized on the outside of the inner mitochondrial membrane and in the intermembrane space (Hirsch et al., 1997). It has an important function in the intracellular electron transport chain reaction for the production of ATP. During the apoptotic process cytochrome c is released in the cytosol (Hirsch et al., 1997). There it binds to the Apoptosis protease activating factor (Apaf-1) which is present in mammalian cells and is analogous to ced-4 in C. elegans. Cytochrome c and Apaf-1 form a complex together with dATP, which
Chapter 2.1
55
subsequently activates procaspase-9 to caspase-9 (Li et al., 1997). This results finally in the activation of procaspase-3 into caspase-3 which leads to the known characteristic morphological consequences.
Apaf-1 Apaf-1
DNA damage NUCLEUS
activation of Apaf-1
procaspase-3
DNA damage inducing p53 gene expression
activation
increase bax transcripts
caspase-9
breaking up of cytoplasmic and nuclear proteins+ DNA
p53 bax
activation of procaspase-9
procaspase-9 binds to active Apaf-1
procaspase-9
caspase-3
bax transport to the outer mitochondrial membrane
bax bcl-2/bcl-xL
MITOCHONDRION
MPT
more bax than bcl-2 induces opening or formation of the megachannel (MPT)
bcl-2/bcl-xL MPT bax
cytochrome c binds to Apaf-1
bax
bax bcl-2/bcl-xL
ion influx => volume increase
cytochrome c
the volume increase induces breaks in the outer mitochondrial membrane with subsequent release of cytochrome c
Fig. 5A. Schematic representation of the mitochondrial apoptotic mechanism induced by DNA damage.
Fas ligand binding on Fas and TNFα binding on TNFR1 induces activation of caspase-8
activation of Apaf-1
procaspase-3 inactive BID activation
activation
caspase-9
breaking up of cytoplasmic and nuclear proteins+ DNA
caspase-8
activation of procaspase-9
procaspase-9 binds to active Apaf-1
Apaf-1
procaspase-9
caspase-3
BID
transport to the outer mitochondrial membrane
MPT
bax bcl-2/bcl-xL
MPT
bax
bax bcl-2/bcl-xL
BID
BID induces formation or opening of the megachannel (MPT)
cytochrome c
ion influx => volume increase
the volume increase induces breaks in the outer mitochondrial membrane with subsequent release of cytochrome c
MITOCHONDRION
Fig. 5B. Schematic presentation of the mitochondrial mechanism induced by caspase-8.
bax bcl-2/bcl-xL
binding of cytochrome c to Apaf-1
Chapter 2.1
56
Most authors agree that this release of cytochrome c into the cytosol is regulated by proteins coded by genes of the bcl-2 (B-cell lymphoma) family. Bcl-2, bcl-xL, bcl-xS, bax and bad are members of this family. Bcl-2 is an apoptosis inhibitor which is found in vertebrates and represents the sequence homologue of ced-9 in C. elegans. Bcl-x codes for two different proteins namely bcl-xL(ong) and bcl-xS(hort). Bcl-xL is analogous to bcl-2 in length and function, and also suppresses apoptosis. Bcl-xS is 63 amino acids shorter and has lost its anti-apoptotic function. This protein is a competitor for bcl-2 and bcl-xL (Dietrich, 1997). The proteins of the bcl-2 family are mainly localized in the outer mitochondrial membrane. The ratio of the bax transcripts to the bcl-2/bcl-xL transcripts will determine whether cytochrome c is released or not. Rossé and colleagues (1998) noticed that the release of cytochrome c induced by bax could not be inhibited even in the presence of high concentrations of bcl-2. However, as no apoptosis was detected, it was concluded that bcl-2 inhibits the working of cytochrome c rather than its release. Until now it is still unclear how bax releases cytochrome c. Some authors suggest that bax and bcl-2/bcl-xL regulate the mitochondrial permeability via a megachannel which is also called mitochondrial permeability transition pore (MPT)(Green and Kroemer, 1998; Marzo et al., 1998). This channel is built up by proteins of the inner and outer mitochondrial membrane and by intermembrane proteins (Shimizu et al., 1999). The opening of this channel results in an influx of ions with subsequently swelling of the mitochondrion. This swelling causes breaks in the outer membrane while the inner membrane remains intact because it has a larger surface (Fig. 5A). Cytochrome c escapes through the outer membrane breaks and enters in the cytosol (Green and Kroemer, 1998). The ion influx explains the decrease of the transmembrane potential of the inner mitochondrial membrane noticed in apoptotic cells (Marzo et al., 1998). The formation of a megachannel also occurs in oncosis/necrosis, but in these latter processes the cellular metabolism is already damaged to the extent that activation of caspases is no longer possible and the cell has no control of its own cell death mechanism (Hirsch et al., 1997). However, the presence of megachannel formation contrasts with transmission electron microscopic studies showing that the mitochondria and organelles remain intact during the apoptotic process (Kerr, 1971). This megachannel formation may be a phenomenon which is either inconstant or occurs only in the later stages of the apoptotic process (Saikumar et al., 1999), while the bax-induced release of cytochrome c may be dependent on the Mg2+ concentration rather than on the mitochondrial permeability transition
Chapter 2.1
57
pore (Eskes et al., 1998). It appears that bax and bcl-2 play undoubtedly an important role in the release of cytochrome c, but their exact working mechanism remains unclear. Recent studies suggest that the mitochondrial mechanism and the receptor-ligand mechanism do not act completely independent of each other (Fig. 5B). Caspase-8 which is characteristic for the receptor-ligand mediated apoptosis can activate the protein bid, a member of the bcl-2 family which is localized in the outer mitochondrial membrane. Activation of bid leads to the release of cytochrome c with subsequent activation of caspase-9 (Li et al., 1998). Bid might also interact with bax (Desagher et al., 1999), but further research is necessary to clarify these mechanisms. Endoplasmic reticulum mediated mechanism Research in transgenic mice revealed a third possible mechanism in the apoptotic process. Agents that stress the endoplasmic reticulum such as tunicamycin and thapsigargin cause much apoptosis in embryonic fibroblasts of normal (heterozygous) mice but not in caspase12-/- knock-out mice (Nakagawa et al., 2000). No differences between the normal mice and the knock-out mice were observed when “non endoplasmic reticulum stressing agents” were administered such as Fas-specific antibodies or dexamethasone which activate the receptorligand mechanism and the mitochondrial mechanism, respectively. For this reason it has been suggested that the ER-mediated mechanism is independent of both former pathways (Mehmet, 2000; Nakagawa et al., 2000). However, the ER-mediated apoptotic pathway has recently been linked to the mitochondrial mechanism, in which Ca2+ plays a key role, in different cell types such as thymocytes (Orrenius et al., 2003), fibroblasts (Distelhorst and Roderick, 2003) and neurons (Verkhratsky and Toescu, 2003). High bax/bcl-2 ratios induce Ca2+ release from the ER into the cytosol (Fig. 6). This results in a Ca2+-mediated mitochondrial permeability transition and sequentially in cytochrome c release, caspase-9 and caspase-3 activation. Released cytochrome c also translocates to the ER and binds to the inositol (1,4,5) triphosphate receptor (IP3), resulting in sustained oscillatory cytosolic Ca2+ releases, which in turn lead to further cytochrome c release (Boehning et al., 2003). Furthermore, caspase-12, which becomes activated during ER stress, can also activate caspase-3 (Jayanthi et al., 2004) (Fig. 6).
Chapter 2.1
58
bax
bcl-2/bcl-xL
-
+
Ca2+
Apaf-1
cytochrome c
HOT SPOT
MPT 2+
Ca mediated mitochondrial membrane transition (MPT) resulting in cytochrome c release
Ca2+ release from ER
Cytochrome c binds to the IP3 receptor on the ER, resulting in sustained Ca2+ release bax
+ Tunicamycin Thapsigargin
ER stress
cytochrome c release
caspase-12 activation
caspase-3 activation
caspase-9 activation
MITOCHONDRION
Fig. 6. Schematic representation of the links between the endoplasmic reticulum (ER) and mitochondrion mediated apoptotic pathways.
Apoptosis and the cell cycle The cell cycle can be subdivided in 4 phases: the G1 phase with protein synthesis, the S phase with DNA replication, the G2 phase from the end of the replication to the eventual cell division and finally the M phase with the mitosis. Cells can also enter from the G1 phase in a rest phase, called the G0 phase. The cell cycle is regulated by different cyclins (A, B, D, E) and by cyclin dependent kinases (cdks). Cyclins are proteins that bind cdks resulting in the formation of active ”cyclin cdk complexes”. When these “cyclin cdk complexes” are present in sufficiently high concentrations, certain checkpoints are passed and the cell enters the next phase (Fig. 7A). The cell cycle can be controlled by changing the concentration and the activity of these complexes. Blocking these complexes can stop the cell cycle and even induce apoptosis (Fig. 7B). The p53 tumor suppressor gene codes for the p53 protein and plays a central role in the control of the cell cycle. This gene is mainly expressed when DNA damage occurs by γradiation or chemotherapeutics (Kamesaki, 1998; Mathieu et al., 1999). The subsequently produced p53 protein induces the expression of the p21 gene, which in turn blocks the “cyclin cdk complexes” and stops the cell cycle. High amounts of p53 also increase the number of
Chapter 2.1
59
bax transcripts and subsequently induce apoptosis (Kamesaki, 1998)(Fig. 7B). Cell division is also regulated by the proto-oncogene c-myc. C-myc overexpression causes p53 dependent apoptosis in murine embryonic fibroblasts. Cell death but not p53 expression could be blocked in embryonic fibroblasts deficient for genes coding for Apaf-1 and caspase-9 (Soengas et al., 1999). This indicates that there is a possible link between p53, c-myc and the mitochondrial mechanism. The exact function and working mechanism of c-myc is however still unclear. Apparently it can induce both apoptosis and mitosis, dependent on the cell type and the external stimulus (Evan et al., 1995). The retinoblastoma gene (Rb) codes for the Rb protein which plays a central role in the cell cycle (Fig. 7A). It inhibits the DNA replication by its high affinity for certain transcription factors. The Rb protein is phosphorylated by “cyclin cdk complexes” during certain phases of the cell cycle resulting in the release of Rb from the transcription factors with subsequently transcription of the DNA (Evan et al., 1995)(Fig. 7B).
M
M cyclin A
cyclin A
p
CDK 1
cyclin D
cyclin B p
CDK 2,4,5,6
cyclin B
Rb
cyclin E
p
cyclin A
p
CDK 1
p21
CDK 1
CDK 1
G2
cyclin D
CDK 1
CDK 2,4,5,6
transcription factors
CDK 2 transcription factors
Rb
G1
G2
cyclin E CDK 2 cyclin A
Rb
transcription factors
transcription factors
CDK 1 transcription factors
p21
DNAdamage
Rb p
Rb transcription factors
Rb
Rb
p
p53
p
transcription factors
Rb
Rb
p21 cyclin A
p
cyclin A CDK 2 CDK 2
CDK 2
S
S
Fig. 7A. The normal cell cycle.
Fig. 7B. Apoptosis in the cell cycle.
cytotoxic agents
G1
Chapter 2.1
60
Detection methods In this chapter different techniques for the detection of apoptosis are reviewed, and their advantages and limitations are discussed. Morphology Different morphological techniques have been developed for detecting apoptosis. However, these techniques are not always efficient to distinguish apoptosis from oncosis. The morphological changes in apoptosis are best seen with electron microscopy. Apoptotic cells can also be detected light microscopically by using stains which bind nucleic acids such as haematoxylin (Fig. 8A) and the Schiff reagent used in the Feulgen reaction (Fig. 8B). However, some experience is required to distinguish apoptotic from oncotic cells (McCarthy and Evan, 1998). Apoptosis can also be detected by fluorescence microscopy. Acridin orange, Hoechst and DAPI are fluorochromes used to demonstrate condensed DNA in apoptotic cells. The DNA in the acridin orange staining is yellow green in normal cells, while apoptotic DNA colors brightly green and is very condensed (McCarthy and Evan, 1998). However, these stainings have to be carefully interpreted because oncotic cells can also have condensed chromatin (Columbano, 1995). Gel electrophoresis Caspase activity can be demonstrated by Western blotting using specific antibodies against caspase substrates such as other caspases, lamines and PARP. This technique has the disadvantage to measure enzyme activity in a cell population and not in individual cells (McCarthy and Evan, 1998). Gel electrophoretic analysis of DNA has demonstrated that apoptosis is linked to internucleosomal DNA fragmentation. However, not all apoptotic cells have this characteristic DNA ladder pattern and some oncotic cells do also show this feature (Dong et al., 1997; Saikumar et al., 1999). The specificity of this ladder structure for the detection of apoptotic cells is therefore lower than usually claimed. Histochemistry DNA fragmentation can also be demonstrated in situ by the use of Terminal deoxynucleotidyl transferase mediated UTP Nick End Labeling (TUNEL)(Fig. 7C).
Chapter 2.1
61
A
B
C
D
Figs. 8A, 8B, 8C and 8D. Detection methods for apoptotic cells. All figures show follicular cells of an atretic ovarian follicle. The follicular lumen is situated on the upper side of the figures. Apoptotic bodies are marked by arrows and chromatin margination by arrowheads. Clusters of apoptotic bodies are surrounded by a dotted line (Bars = 10 µm). Fig. 8A. H.E. staining of a bovine ovary. Fig. 8B. Feulgen staining of a bovine ovary. Fig. 8C. TUNEL-technique of a canine ovary. Fig. 8D. Immunohistochemical staining of a bovine ovary with monoclonal antibodies against single stranded DNA. Notice the strong staining of the apparently intact nuclei (diamond-shaped arrows).
Chapter 2.1
62
In this technique labeled dUTPs are enzymatically coupled to the 3’-OH endings of the DNA fragments and are detected immunohistochemically (Gavrieli et al., 1992). This technique can be adjusted for fluorescence detection by microscopy or flow cytometry. As mentioned before, DNA fragmentation also occurs in oncotic cells and therefore the specificity of this method is doubtful (Saikumar et al., 1999). A more recently developed method is the immunohistochemical detection of single stranded DNA (ssDNA) (Fig. 7D). The formation of ssDNA can be induced in apoptotic cells and not in normal cells by heating up in formamide. Monoclonal antibodies directed against a specific nucleotide sequence (polydeoxycytidin or polydeoxythymidin) can only bind to ssDNA when this strand is at least 25-30 bases long. This technique is claimed to be very specific (Frankfurt et al., 1996). Another new immunohistochemical technique is the detection of caspase-3 (Saunders et al., 2000). As mentioned before, caspase-3 plays a central role in the apoptotic process and has so far not been detected in oncosis, which makes this technique very specific (Guo and Hay, 1999). However, for many investigators the TUNEL technique has become the standard method for identifying DNA fragmentation in apoptotic cells despite the problems of differentiating apoptosis from oncosis. For this reason double staining methods using TUNEL followed by immunohistochemical detection of caspase-3 have been developed (Urase et al., 1999). Flow cytometry Flow cytometry is a technique that can be used for detecting diverse parameters of the apoptotic process. DNA-analysis can be performed by labeling DNA with a fluorochrome (e.g. propidium iodide). The intensity of the signal is representative of the DNA contents of each analyzed cell and apoptotic bodies have less DNA than normal cells (McCarthy and Evan, 1998). Flow cytometry is especially used for the detection of phosphatidylserine located on the cell membrane of apoptotic cells. The protein annexin V binds phosphatidylserine that is exposed on the outside of the membrane of apoptotic cells (Vermes et al., 1995). However, this technique is not entirely specific for the detection of apoptotic cells, since annexin V can also bind phosphatidylserine on the inside of the damaged cell membranes of oncotic cells (Saikumar et al., 1999). Distinguishing apoptotic from oncotic
Chapter 2.1
63
cells is possible by performing a double staining technique with annexin V and propidium iodide, as the latter does not penetrate cells with an intact cell membrane. By this technique apoptotic cells are only colored by annexin V, whereas oncotic cells are stained with both annexin V and propidium iodide (Brush, 2000). Detection of mitochondrial membrane potential One of the newer techniques in apoptosis research is measuring the mitochondrial transmembrane potential by using fluorescent lipophilic molecules such as the J-aggregate forming 5,5’, 6,6’-tetrachloro-1,1’, 3,3’-tetraethylbenzimidazolcarbocyanine iodide (JC-1). In healthy cells JC-1 aggregates in the mitochondria by forming a polymer with a brightly red fluorescence. In apoptotic cells the transmembrane potential has decreased and the lipophilic molecules remain within the cytoplasm in their monomeric form with a green fluorescence (Bedner et al., 1999; Brush, 2000). This fluorescence can be detected by flow cytometry and fluorescence microscopy. However, the specificity of this technique is doubtful because the transmembrane potential also decreases in oncotic cells.
Concluding remarks Recent findings indicate that the existing opinions on cell death must be modified. The distinction between apoptosis and necrosis has become less clear because both phenomena have common features. The reintroduction of the term oncosis is therefore recommended. In contrast to necrosis, which is a postmortal event, apoptosis and oncosis are premortal processes. Many techniques are available for the detection of apoptosis but their specificity is usually incomplete. By combining two or more techniques it is often possible to distinguish apoptotic from oncotic cells but the development of more specific detection methods is recommended.
Chapter 2.1
64
References
Arnheim G. Coagulationsnekrose und Kernschwund. Virchows Archiv A: Pathologische Anatomie 1890; 120: 367-383.
Ashkenazi A, Dixit VM. Apoptosis control by death and decoy receptors. Current Opinion in Cell Biology 1999; 11: 255-260.
Bedner E, Li X, Gorczyca W, Melamed MR, Darzynkiewicz Z. Analysis of apoptosis by laser scanning cytometry. Cytometry 1999; 35: 181-195.
Boehning D, Patterson RL, Sedaghat L, Glebova NO, Kurosaki T, Snyder SH. Cytochrome c binds to inositol (1,4,5) triphosphate receptors, amplifying calcium-dependent apoptosis. Nature Cell Biology 2003; 5: 1051-1061.
Boldin MP, Varfolomeev EE, Pancer Z, Mett IL, Camonis JH, Wallach D. A novel protein that interacts with the death domain of Fas/APO1 contains a sequence motif related to the death domain. Journal of Biological Chemistry 1995; 270: 7795-7798.
Boscá L, Hortelano S. Mechanisms of nitric oxide-dependent apoptosis: involvement of mitochondrial mediators. Cellular Signalling 1999; 11: 239-244.
Brush MD. Recourse to death. A bevy of new products harnesses the power of flow cytometry for detecting apoptosis. The Scientist 2000; 14: 25-29.
Chinnaiyan AM, O’Rourke K, Yu GL, Lyons RH, Garg M, Duan DR, Xing L, Gentz R, Ni J, Dixit VM. Signal transduction by DR3, a death domain-containing receptor related to TNFR-1 and CD95. Science 1996; 274: 990-992.
Chinnaiyan AM, O´Rourke K, Tewari M, Dixit VM. FADD, a novel death domain-containing protein, interacts with the death domain of Fas and initiates apoptosis. Cell 1995; 81: 505-512.
Chu-Wang I-W, Oppenheim RW. Cell death of motoneurons in the chick embryo spinal cord. I. A light and electron microscopic study of naturally occurring and induced cell loss during development. The Journal of Comparative Neurology 1978; 177: 33-58.
Clarke PGH. Developmental cell death: morphological diversity and multiple mechanisms. Anatomy and Embryology 1990; 181: 195-213.
Columbano A. Cell death: current difficulties in discriminating apoptosis from necrosis in the context of pathological processes in vivo. Journal of Cellular Biochemistry. 1995; 58: 181-190.
Chapter 2.1
65
Denecker G, Vercammen D, Declercq W, Vandenabeele P. Apoptotic and necrotic cell death induced by death domain receptors. Cellular and Molecular Life Sciences 2001, 58: 356-370.
Desagher S, Osen-Sand A, Nichols A, Eskes R, Montessuit S, Lauper S, Maundrell K, Antonsson B, Martinou J-C. Bid-induced conformational change of Bax is responsible for mitochondrial cytochrome c release during apoptosis. The Journal of Cell Biology 1999; 144: 891-901.
Dietrich JB. Apoptose et gènes anti-apoptotiques de la famille Bcl-2. Archives of Physiology and Biochemistry 1997; 105: 125-135.
Distelhorst CW, Roderick HL. Ins(1,4,5)P3-mediated calcium signals and apoptosis: is there a role for Bcl2? Biochemical Society Transactions 2003; 31: 958-959.
Donepudi M, Grütter MG. Structure and zymogen activation of caspases. Biophysical Chemistry 2002; 101102: 145-153.
Dong Z., Saikumar P, Weinberg JM Ventachalam MA. Internucleosomal DNA cleavage triggered by plasma membrane damage during necrotic cell death. The American Journal of Pathology 1997; 151: 12051213.
Duvall E., Wyllie AH, Morris RG. Macrophage recognition of cells undergoing programmed cell death (apoptosis). Immunology 1985; 56: 351-358.
Ellis HM, Horvitz HR. Genetic control of programmed cell death in the nematode C. elegans. Cell 1986; 44: 817-829.
Enari M, Hug H, Nagata S. Involvement of an ICE-like protease in Fas-mediated apoptosis. Nature 1995; 375: 78-81.
Eskes R., Antonsson B, Osen-Sand A, Montessuit S, Richter C, Sadoul R, Mazzei G, Nichols A, Martinou J-C. Bax-induced cytochrome c release from mitochondria is independent of the permeability transition pore but highly dependent on Mg+ ions. The Journal of Cell Biology 1998; 143: 217-224.
Evan GI, Brown L, Whyte M, Harrington E. Apoptosis and the cell cycle. Current Opinion in Cell Biology 1995; 7: 825-834.
Fadok VA, Voelker DR, Campbell PA, Cohen JJ, Bratton DL, Henson PM. Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. The Journal of Immunology 1992; 148: 2207-2216.
Chapter 2.1
66
Farber E. Ideas in pathology: programmed cell death: necrosis versus apoptosis. Modern Pathology 1994; 7: 605-609.
Flemming W. Über die Bildung von Richtungsfiguren in Säugethiereiern beim Untergang Graaf’scher Follikel. Archiv der Anatomische Entwicklungs Geschichte, 1885: pp. 221-244.
Frankfurt OS, Robb JA, Sugarbaker EV, Villa L. Monoclonal antibody to single-stranded DNA is a specific and sensitive cellular marker of apoptosis. Experimental Cell Research 1996; 226: 387-397.
Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. The Journal of Cell Biology 1992; 119: 493-501.
Gräper L. Eine neue Anschauung über Physiologische Zellausschaltung. Archiv der Zellforschung 1914; 12: 373-394.
Green D, Kroemer G. The central executioners of apoptosis: caspases or mitochondria? Cell Biology 1998; 8: 267-271.
Guo M, Hay BA. Cell proliferation and apoptosis. Current Opinion in Cell Biology 1999; 11: 745-752.
Hirsch T, Marchetti P, Susin SA, Dallaporta B, Zamzami N, Marzo I, Geuskens M, Kroemer G. The apoptosis-necrosis paradox. Apoptogenic proteases activated after mitochondrial permeability transition determine the mode of cell death. Oncogene 1997; 15: 1573-1581.
Hsu H, Xiong J, Goeddel DV. The TNF receptor 1- associated protein TRADD signals cell death and NFκB activation. Cell 1995; 81: 495-504.
Jayanthi S, Deng X, Noailles P-A H, Ladenheim B, Cadet JL. Methamphetamine induces neuronal apoptosis via cross-talks between endoplasmic reticulum and mitochondria-dependent death cascades. The FASEB Journal 2004; 18: 238-251.
Kamesaki H. Mechanisms involved in chemotherapy-induced apoptosis and their implications in cancer chemotherapy. International Journal of Hematology 1998; 68: 29-43.
Kerr JFR. Shrinkage necrosis: a distinct mode of cell death. The Journal of Pathology 1971; 105: 13-20.
Kerr JFR. Apoptosis: a basic biological phenomenon with wide ranging implications in tissue kinetics. The British Journal of Cancer 1972; 26: 239-257.
Chapter 2.1
67
Kerr JFR, Gobe GC, Winterford CM, Harmon BV. Anatomical methods in cell death. Methods in Cell Biology 1995; 46: 1-27.
Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X. Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 1997; 91: 479-489.
Li H, Zhu H, Xu C-J, Yuan J. Cleavage of BID by caspase-8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 1998; 94: 491-501.
Lockshin RA, Zakeri Z. Programmed cell death and apoptosis. In: Apoptosis: The Molecular Basis of Cell Death. Current Communications in Cell and Molecular Biology, vol. 3. Tomei DL, Cope FO (Eds.), Cold Spring Harbor Laboratory Press, New York 1991: pp. 47-60.
Majno G, Joris I. Apoptosis, oncosis and necrosis. An overview of cell death. The American Journal of Pathology 1995; 146: 3-15.
Martin SJ, Green DR. Protease activation during apoptosis: death by a thousand cuts? Cell 1995; 82: 349352.
Martin DP, Johnson EM Jr. Programmed cell death in the peripheral nervous system. In: Apoptosis: The Molecular Basis of Cell Death. Current Communications in Cell and Molecular Biology, vol. 3. Tomei DL, Cope FO (Eds.), Cold Spring Harbor Laboratory Press, New York, 1991: pp. 247-261.
Marzo I, Brenner C, Kroemer G. The central role of the mitochondrial megachannel in apoptosis: evidence obtained with intact cells, isolated mitochondria and purified protein complexes. Biomedicine & Pharmacotherapy 1998; 52: 248- 251.
Mathieu J, Richard S, Ballester B, Chancerelle Y, Multon E. Apoptose et rayonnement gamma. Annales Pharmaceutiques Françaises 1999; 57: 314-323.
McCarthy NJ, Evan GI. Methods for detecting and quantifying apoptosis. Current Topics in Developmental Biology 1998; 36: 259-278.
Mehmet H. Caspases find a new place to hide. Nature 2000; 403: 29-30.
Nagata S. Apoptosis by death factor. Cell 1997; 88: 355-365.
Nakagawa T, Zhu H, Morishima N, Li E, Xu J, Yankner BA, Yuan J. Caspase-12 mediates endoplasmicreticulum-specific apoptosis and cytotoxicity by amyloid-β. Nature 2000; 403: 98-103.
Chapter 2.1
68
Orrenius S, Zhivotovsky B, Nicotera P. Regulation of cell death: the calcium-apoptosis link. Nature Reviews Molecular Cell Biology 2003; 4: 552-565.
Park BS, Kim GC, Back SJ, Kim ND, Kim YS, Kim SK, Jeong MH, Lim YJ, Yoo YH. Murine bone marrow-derived mast cells exhibit evidence of both apoptosis and oncosis after IL-3. Immunological Investigations 2000; 29: 51-60.
Rossé T, Olivier R, Monney L, Rager M, Conus S, Fellay I, Jansen B, Borner C. Bcl-2 prolongs cell survival after Bax-induced release of cytochrome c. Nature 1998; 391: 496-499.
Saikumar P, Dong Z, Mikhailov V, Denton M, Weinberg JM, Ventachalam MA. Apoptosis: definition, mechanisms and relevance to disease. American Journal of Medicine 1999; 107: 489-506.
Sanders EJ, Wride MA. Programmed cell death in development. International Review of Cytology 1995; 163: 105-173.
Saunders PA, Cooper JA, Roodell MM, Schroeder DA, Borchert CJ, Isaacson AL, Schendel MJ, Godfrey KG, Cahill DR, Walz AM, Loegering RT, Gaylord H, Woyno IJ, Kaluyzhny AE, Krzyzek RA, Mortari F, Tsang M, Roff CF. Quantification of active caspase 3 in apoptotic cells. Analytical Biochemistry 2000; 284: 114-124.
Savill J, Dransfield I, Hogg N, Haslett C. Vitronectin receptor-mediated phagocytosis of cells undergoing apoptosis. Nature 1990; 343: 170-173.
Schmaus H, Albrecht E. Zur Frage der Coagulationsnekrose. Deutsche Medizinische Wochenschrift 1899; 6: 89-114.
Schwartz LM, Smith SW, Jones MEE, Osborne BA. Do all programmed cell deaths occur via apoptosis? Proceedings of the National Academy of Sciences of the USA 1993; 90: 980-984.
Shimizu S, Narita M, Tsujimoto Y. Bcl-2 family proteins regulate the release of apoptogenic cytochrome c by the mitochondrial channel VDAC. Nature 1999; 399: 483-487.
Shiokawa D, Maruta H, Tanuma S. Inhibitors of poly(ADP-ribose) polymerase suppress nuclear fragmentation and apoptotic-body formation during apoptosis in HL-60 cells. FEBS letters 1997; 413: 99103.
Soengas MS, Alarcon RM, Yoshida H, Giaccia AJ, Hakem R, Mak TW, Lowe SW. Apaf-1 and caspase-9 in p53-dependent apoptosis and tumor inhibition. Science 1999; 284: 156-159.
Chapter 2.1
69
Takahashi A, Alnemri ES, Lazebnik YA, Fernandes-Alnemri T, Litwack G, Litwack RD, Moir RD, Goldman RD, Poirier GG, Kaufman SH, Earnshaw WC. Cleavage of lamin A by Mch2α but not CPP32: Multiple interleukin 1β-converting enzyme-related proteases with distinct substrate recognition properties are active in apoptosis. Proceedings of the National Academy of Sciences of the USA 1996; 93: 8395-8400.
Urase K, Momoi T, Fujita E, Isahara K, Uchiyama Y, Tokunaga A, Nakayama K, Motoyama N. Bcl-xL is a negative regulator of caspase-3 activation in immature neurons during development. Brain Research. Developmental Brain Research 1999; 116: 69-78.
Van den Berghe T, Kalai M, van Loo G, Declercq W, Vandenabeele P. Disruption of HSP90 function reverts tumor necrosis factor-induced necrosis to apoptosis. The Journal of Biological Chemistry 2003; 278: 5622-5629.
Vercammen D, Beyaert R, Denecker G, Goossens V, van Loo G, Declercq W, Grooten J, Fiers W, Vandenabeele P. Inhibition of caspases increases the sensitivity of L929 cells to necrosis mediated by tumor necrosis factor. Journal of Experimental Medicine 1998; 187: 1477-1485.
Verkhratsky A, Toescu EC. Endoplasmic reticulum Ca2+ homeostasis and neuronal death. Journal of Cellular and Molecular Medicine 2003; 7: 351-361.
Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C. A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V. Journal of Immunological Methods 1995; 184: 39-51.
Virchow R. Cellular pathology as based upon physiological and pathological histology (2nd ed.). Cover Publications, New York 1971: pp. 356-382.
von Recklinghausen F. Untersuchungen über Rachitis und Osteomalacie. Jena, Verlag Gustav Fischer 1910.
Walczak H, Degli-Esposti MA, Johnson RS. TRAIL-R2: a novel apoptosis-mediating receptor for TRAIL. The Embo Journal 1997; 16: 5386-5397.
Wyllie AH, Kerr JFR, Currie AR. Cell death: the significance of apoptosis. International Review of Cytology 1980; 68: 251-306.
2.2 Apoptosis in the canine endometrium during the estrous cycle
Modified from: APOPTOSIS IN THE CANINE ENDOMETRIUM DURING THE ESTROUS CYCLE Van Cruchten S, Van den Broeck W, Duchateau L, Simoens P Theriogenology, 2003; 60: 1595-1608
Chapter 2.2
73
Summary Apoptotic cell death in the endometria of 58 female dogs in different stages of the estrous cycle was assessed by using the terminal deoxynucleotidyl transferase mediated deoxyuridine triphosphate nick end labeling (TUNEL) assay and by immunohistochemical detection of caspase-3 activity on formalin fixed paraffin embedded sections. For both techniques the apoptotic index was determined in the surface epithelium, the stroma, the crypts and the basal glands by counting the percentage of stained cells in a total of 500 cells in each category. In the surface epithelium and the stroma TUNEL- and caspase-3 positive cells were rare (apoptotic index < 1) throughout the estrous cycle. However, for the stroma caspase-3 detection showed a significant increase of the apoptotic index in anestrus as well as an increase for both stroma and surface epithelium in late metestrus. The apoptotic index increased during late metestrus and anestrus in the crypts and in the basal glands, but in the crypts this increase was only significant when caspase-3 detection was used, whereas in the basal glands significant differences were found for both the TUNEL and caspase-3 reactions. It can be concluded that apoptosis is present in the different canine endometrial cells during the estrous cycle, but caspase-3 detection showed more significant differences than the TUNEL assay. Furthermore, a high apoptotic index suggestive of endometrial desquamation was not observed in the surface epithelium and no correlation was found between the apoptotic index and the serum progesterone levels in any cell group.
Chapter 2.2
74
Introduction Endometrial apoptosis during the estrous cycle has been studied in a number of species (Hopwood and Levison, 1976; Sato et al., 1997; Dharma et al., 2001; Wasowska et al., 2001) and is related to ovarian steroid hormone levels (Sandow et al., 1979; Rotello et al., 1992; Jo et al., 1993; Dahmoun et al., 1999; Salamonsen et al., 1999). Many papers have been published on the cyclic changes of the endometrium in dogs (Keller, 1909; Arenas and Sammartino, 1939; Mulligan, 1942; Barrau et al., 1975) but we have found no data on apoptosis and its possible relation to steroid hormone levels. In a previous scanning electron microscopic (SEM) study of the cyclic changes of the canine endometrial surface epithelium (Van Cruchten et al., 2003), we detected a highly variable aspect of surface epithelial cells in early metestrus. Some cells resembled apoptotic cells described in the human endometrium (Nikas et al., 2000), but their number was small and convincing proof for apoptosis was missing as only the surface morphology was observed. The present study was designed to verify the presence of apoptosis in the canine endometrium and its possible relation to serum progesterone levels. We therefore used two techniques, an in situ DNA 3'-end labeling method (TUNEL) and immunohistochemical detection of activated caspase-3, as the specificity of the detection methods for apoptosis is usually incomplete and apoptotic cell death may be more correctly estimated by combining two or more techniques (Van Cruchten and Van den Broeck, 2002). This study might form a basis for further reproductive and pathological research in this species. Apoptosis might be essential for maintaining homeostasis of the cell number in the canine endometrium during the estrous cycle and for tissue remodeling during blastocyst implantation and placental development as has been shown in other species (Akcali et al., 1996 ; Galán et al., 2000). Moreover, knowledge of apoptosis in the normal canine endometrium could also be important as a basis for further insight into the pathogenesis of pyometra and cystic endometrial hyperplasia (CEH) which constitute an important pathological entity in the dog (Noakes et al., 2001).
Chapter 2.2
75
Materials and methods Animals Samples of both ovaries and uterine horns were obtained from 58 healthy adult dogs presented for ovariohysterectomy or euthanasia at the Faculty of Veterinary Medicine, Ghent University, and at four veterinary clinics (Table 1). These dogs included 35 of the 38 dogs used in our previous SEM study of the cyclic changes of the canine endometrial surface (Van Cruchten et al., 2003) and 23 dogs used in a study on estrogen and progesterone receptors by Vermeirsch et al. (1999; 2000). For each dog data concerning the anamnesis, age, breed, litters and last proestrous bleeding were recorded. The dogs varied in age from 1 to 8.5 years and included 20 mongrels, 7 German Shepherds, 3 Boxers, 3 Labrador Retrievers, 3 Yorkshire terriers, 3 Maltese Dogs, 2 Cocker Spaniels, 2 West Highland White terriers, 2 Rottweilers, 1 Poodle, 1 Cairn terrier, 1 Shih-Tzu, 1 Newfoundlander, 1 Saint Bernard, 1 Jack Russell, 1 Belgian Shepherd, 1 Dobermann, 1 Munsterlander, 1 Fox terrier, 1 Golden Retriever, 1 Teckel and 1 Beagle. Tissue processing Immediately after excision samples were fixed in a phosphate buffered 3.5 % formaldehyde solution (pH 6.7) for 6 h (Van Cruchten et al., 2003) or for 48 h (Vermeirsch et al., 1999; 2000). All tissue samples were embedded in paraffin, and 5 µm thin sections were cut, mounted on 3-aminopropyl-triethoxysilane-coated slides (APES, Sigma, St. Louis, MO, USA), dried for 1 h at 60 °C on a hot plate and further dried overnight at 37 °C. Serology Serum samples, taken immediately before surgery or euthanasia, were stored at -20 °C until assayed for determination of the serum levels of the sex steroids using a RIA technique. Concentrations of progesterone and estradiol-17β were measured as described previously (Henry et al., 1987). Determination of cycle stage
Chapter 2.2
76
The stage of the estrous cycle was determined as in our previous studies (Van Cruchten et al., 2002; 2003) using histologic and serological parameters as described by Vermeirsch et al. (1999; 2000) with slight modification. The animals were first classified by macroscopic evaluation of the genital tract combined with histologic examination of the ovaries and uterus. After this preliminary morphological classification the animals were further classified according to the serum progesterone levels. Animals with uterine tissue at a proliferative stage and a progesterone level lower than 1 ng/mL were classified as being in proestrus. Animals with proliferative uterine tissue, developing corpora lutea and progesterone levels between 1 ng/mL and 15 ng/mL were classified as being in estrus. Bitches were classified in early metestrus when uterine tissue was proliferative and progesterone levels were above 15 ng/mL in the presence of growing or fully developed corpora lutea, or above 10 ng/mL when regressing corpora lutea were present. Animals with uterine tissue at a secretory stage and progesterone levels lower than 10 ng/mL and higher than 0.5 ng/mL were classified as being in late metestrus. When uterine tissue was at rest and progesterone levels were basal (≤ 0.5 ng/mL) the dogs were considered to be in anestrus. Eight dogs were in proestrus, 10 in estrus, 9 in early metestrus, 15 in late metestrus and 16 in anestrus. Apoptosis detection Terminal deoxynucleotidyl transferase (TdT)-mediated deoxyuridine triphosphate (dUTP) nick end labeling (TUNEL) methodology was used for the detection of DNA fragmentation. The TACSTM TdT kit (R&D Systems Europe Ltd, Abingdon, UK) was used and the protocol recommended by the manufacturer was followed. Briefly, deparaffinized and hydrated slides were treated with 20 µg/mL proteinase K for 15 min at room temperature (RT). The slides were rinsed in distilled water and then incubated for 5 min at RT in 50 mL of a 3 % H2O2 solution to block endogenous peroxidase activity. The samples were rinsed in PBS and incubated in a TdT labeling buffer for 5 min at RT and immediately incubated in a Labeling Reaction Mix which contained the biotinylated TdT-dNTP nucleotides, Co+ and the TdT enzyme in a humidified chamber at 37 °C for 1 hr. This was followed by incubation of the slides in a Stop buffer for 5 min at RT to stop the enzymatic reaction. They were then rinsed in PBS and incubated with 100 µL of 30 % normal goat serum for 30 min at 37 °C, in order to prevent nonspecific reactions. After rinsing in PBS the sections were incubated with 50 µL of Streptavidin-HRP detection solution for 10 min at RT. Finally, after rinsing in PBS, 50 µL DAB chromogene substrate (Liquid DAB+, DAKO, Glostrup, Denmark) was administered for
Chapter 2.2
77
5 min at RT. The reaction of DAB with peroxidase results in a brown staining. Mayer's hematoxylin was applied during 30 sec as a nuclear counterstain. In every staining procedure positive and negative controls were included. The positive control was a canine ovary section with atretic follicles and a uterine tissue section treated with DNase before the Labeling Reaction Mix was administered. The negative controls were a canine uterine tissue section without the Labeling Reaction Mix incubation and a canine uterine tissue section without the Labeling Reaction Mix incubation and streptavidin-HRP incubation. For the detection of caspase-3 activity, a polyclonal rabbit-anti-mouse antibody and the Cell and Tissue Staining Kit (R&D Systems Europe Ltd, Abingdon, UK) were used. After rehydration the sections were pretreated in an Antigen Retrieval Citra Solution (BioGenex, San Ramon, USA). This pretreatment consisted of microwaving the slides for 2 min at 700 Watts and then again for 3, 5 and 5 min at 200 Watts with 5 min rest in between. After cooling for 30 min at 4 °C and rinsing in distilled water, the slides were incubated for 5 min with 50 µL of Peroxidase Blocking Reagens. All incubations were carried out in a humidified environment. Slides were then rinsed in PBS and incubated consecutively with 50 µL of Serum Blocking Reagens G, Avidin Blocking Reagens and Biotin Blocking Reagens each for 15 min at RT. These incubations were carried out to prevent nonspecific reactions. The actual immunohistochemical detection of active caspase-3 was performed using a polyclonal rabbit anti-human/mouse caspase-3 antibody, which cross-reacts with the canine caspase-3. All sections were incubated with 50 µL of a 1 µg/mL concentrated primary antibody in PBS. After rinsing in PBS the sections were incubated for 45 min at RT with 50 µL of a ready-touse secondary biotinylated anti-rabbit antibody. The specimens were rinsed in PBS and incubated for 30 min at RT with streptavidine-HRP. Finally, after rinsing in PBS, 50 µL of DAB chromogen substrate was administered for 5 min. Mayer's hematoxylin was applied during 30 sec as a nuclear counterstain. In every staining procedure positive and negative controls were included. The positive control was a canine ovary section with atretic follicles known to contain many apoptotic cells. The negative controls were a uterine tissue section incubated with PBS instead of the primary antibody and a uterine tissue section incubated without the primary and secondary antibodies. For both techniques apoptosis was evaluated in the surface epithelium, the stroma, the crypts and the basal glands by counting the number of stained apoptotic cells in 500 cells of each category with a Leitz Diaplan light microscope at magnification x 400 by 2 independent
Chapter 2.2
78
observers. The percentage of TUNEL and caspase-3 positive cells is listed in Table 1 as the apoptotic index. For TUNEL, cells with brown staining of the nucleus, apoptotic bodies or granules in the cytoplasm together with aberrant nuclear morphology were counted as positive. For caspase-3, cells with brown granules in the cytoplasm or brown luminal debris were counted as positive. Photographs were taken with an Ilford PANF plus 50 ASA black and white film at magnification x 1000 with immersion oil. Hemosiderin staining To differentiate caspase-3 positivity from hemosiderin, a hemosiderin counterstaining was used. First, the slides were stained for caspase-3 as described, but they were not counterstained in haematoxylin. They were then incubated in a mixture of HCl 20 % and potassium ferrocyanide 10 % for 20 minutes. After rinsing in distilled water, the sections were counterstained in a neutral red solution for a few seconds and finally dehydrated and mounted. In this staining protocol caspase-3 positive cells stained brown, while hemosiderin stained blue and the background was pale pink. A photograph was taken digitally with a Leica DP50 camera at magnification x 400. Statistics Pairwise comparison of cycle stages in each cell category by the Wilcoxon rank sum test was used with fixation time (6 and 48 hours) as stratification factor at a significance level equal to 0.006 for each individual comparison (Tukey's multiple comparisons method with family confidence coefficient = 0.95). The Spearman rank correlation coefficient was used to assess whether apoptotic index correlated with serum progesterone levels within each cell category and at each cycle stage.
Results In the surface epithelium and the stroma of some dogs in anestrus and proestrus, brown staining due to hemosiderin was present. This did not complicate scoring of the slides stained for TUNEL as hemosiderin is distributed throughout the cytoplasm of the cells rather than confined to the nucleus. However, for detection of caspase-3, which is also localized in the
Chapter 2.2
79
cytoplasm, a hemosiderin specific counterstaining was used to avoid false positive results (Fig. 1). Fig. 1. Hemosiderin counterstaining of a caspase-3 stained uterine section of a dog in s
anestrus. Notice the uterine lumen (l), the surface epithelium (se), the crypts (c) and the
c
stroma (s). Blue stained hemosiderin (arrows) l
se
1
c
is present in the stroma (s) whereas brown stained caspase-3 positive cells (arrowheads) are noticed in the surface epithelium (se) and in the stroma (s). Bar = 20 µm.
Results for TUNEL assay and caspase-3 detection are shown in Figures 2-5, A and B. In the surface epithelium and the stroma (Figs. 6 and 7) only a small number of TUNEL- and caspase-3 positive cells (apoptotic index
