Protease activity in rabbit uterine secretion and blastocyst between 5 days 18 hours and 7 days 18 hours after mating was studied histologically and electrophoretically by film gelatin lysis. At definite times after the beginning of implantation, the uterine secretion contains two and the blastocyst covering at least one protease, all with different electrophoretic mobility. Digestion of the blastocyst covering begins on the seventh day post coitum in the area of the embryonic disc. The function of the proteases, origin of the blastocyst protease, and possible activation mechanisms of these enzymes are discussed.

During the preimplantation period the rabbit uterus contains a secretion characterized by several pregnancy-specific proteins and glycoproteins (Schwick, 1965; Beier, 1968; Kirchner, 1969). These include the proteases secreted after the fifth day post coitum (p.c.). For lack of evidence to the contrary, we assume that they are involved in hydrolysis of the blastocyst covering (Kirchner, Hirschhäuser & Kionke, 1971). The term blastocyst covering designates the protective layer around the blastocyst, consisting of zona pellucida, mucoprotein layer, and any attached uterine secretion (gloeolemma, Böving, 1957). After 7 days p.c. the blastocyst itself exhibits proteolytic activity. Denker (1971 b) localized this activity in the trophoblast. The following studies show that before and after attachment of the blastocyst occurs, the uterine secretion and the blastocyst covering contain several proteases of differing electrophoretic mobility, while the trophoblast itself remains inactive.

For histological and electrophoretic demonstration of protease activity we used the gelatin layer of exposed, developed, fixed, washed and dried black- and-white film as a protein substrate. This allows more accurate topographic localization of the faintest activities than do home-made gelatin layers (Denker, 1971 a; Owers & Blandau, 1971). Agfa Isopan IFF 15 DIN proved to be best suited, better than the Adox KB 14 used previously because without pretreatment its gelatin remains comparatively rigid during incubation in a moist atmosphere at 37 °C. Colour negative film (Bergström, 1970) and slide film (Adams & Tuqan, 1961) were less sensitive.

Uterine tissue samples 1–2 cm in length, with and without blastocysts, were frozen in liquid, nitrogen-cooled isopentane. In addition, blastocysts obtained by flushing uteri were washed carefully in physiological NaCl solution and frozen immediately with CO2 on the dissecting plate. Tissue sections 10 μm thick were made in a cryostat microtome at –23 °C, thawed on film strips, and freeze-dried overnight in the cryostat. After warming to room temperature, the preparations were incubated three to ten hours at 37 °C in a moist Petri dish, after which they could be examined without further treatment. Photographic documentation was performed with the Aristophot (Leitz) and with a microscope equipped with an automatic camera (Orthomat, Leitz). For histological control we thawed tissue sections of the same series on microscope slides, fixed them with ethanol–chloroform–acetic acid (Carnoy mixture), and stained them with gallocyanin–chromalum. Electrophoretic studies of proteases in uterine secretion and tissue homogenate were made with slides covered with 1·5 % agar, such as are used in immunoelectrophoresis. After the electrophoretic run the agar layers were detached from the slides, applied to film strips, and incubated for 3–24 h at 37 °C in a moist Petri dish.

As described in an earlier paper (Kirchner et al. 1971), we found no protease activity in uterine secretion before the fifth day p.c. At 5 days 18 hours p.c. the secretion protease can be demonstrated clearly by histological methods. Fig. 1A shows a uterus with blastocyst at this time. Around the blastocyst the film layer is brighter than in other parts of the uterine lumen. Proteolytic activity in parts of the uterus lacking a blastocyst is distributed unevenly (Fig. 1B). Some areas in the main ducts, or even the entire main ducts, exhibit higher activity than the smaller ducts and crypts. The endometrium itself shows a negative reaction.

Fig. 1.

Five days 18 hours post coitum. Uterus with blastocyst: activity around the blastocyst exceeds that in other parts of uterine lumen. B. Uterus without blastocyst: areas of higher activity are found in the main ducts. C. Isolated blastocyst (overall): the blastocyst covering is inactive, except for a few spots (arrows). D. Isolated blastocyst (detail): part of blastocyst covering with spots of activity. 1 A, IB × 30, 1CX20, ID × 300.

Fig. 1.

Five days 18 hours post coitum. Uterus with blastocyst: activity around the blastocyst exceeds that in other parts of uterine lumen. B. Uterus without blastocyst: areas of higher activity are found in the main ducts. C. Isolated blastocyst (overall): the blastocyst covering is inactive, except for a few spots (arrows). D. Isolated blastocyst (detail): part of blastocyst covering with spots of activity. 1 A, IB × 30, 1CX20, ID × 300.

Sections of isolated and carefully washed blastocysts at 5 days 18 hours p.c. indicate that at this stage the blastocyst covering is still proteolytically inactive (Fig. 1C). A few sites at which we found low activity may be interpreted as traces of attached uterine secretion (Fig. 1C arrows, ID).

At 6 days 18 hours p.c. distribution of uterine and blastocyst activity is similar to that at 5 days 18 hours p.c. (Fig. 2A); mesometrially and anti-mesometrially the activities are equal, but generally higher than in other parts of the uterine lumen. This was the latest time at which we were consistently able to flush out intact blastocysts. The entire covering of such a blastocyst now shows high proteolytic activity (Fig. 2B). Simultaneously its firmness is decreased; using the cryostat technique it was impossible to section an isolated blastocyst without any damage. The trophoblast itself is inactive. The few small bright spots are due to traces of enzyme, washed from the blastocyst covering during thawing procedure in the cryostat. The blastocoelic fluid was negative.

Fig. 2.

Six days 18 hours post coitum. A. Uterus with blastocyst (overall): activity is equal mesometrially and antimesometrially. B. Isolated blastocyst (overall): high activity in the whole blastocyst covering. 2A, 2B × 10.

Fig. 2.

Six days 18 hours post coitum. A. Uterus with blastocyst (overall): activity is equal mesometrially and antimesometrially. B. Isolated blastocyst (overall): high activity in the whole blastocyst covering. 2A, 2B × 10.

On the seventh day p.c. the orientation of the blastocyst in the mesometric– antimesometrial axis is established. The blastocyst attaches to the endometrium, and a few hours later disintegration begins. Seven days 3 hours p.c. in the area of the blastocoelic disc (mesometrially) lysis of the blastocyst covering is in full progress (Fig. 3A), while antimesometrially the layer is still completely intact (Fig. 3B).

Fig. 3.

Seven days 3 hours post coitum. A. Mesometrial region of a uterus with blastocyst: digestion of the blastocyst covering has started in the area of the embryonic disc. B. Antimesometrial region of uterus with blastocyst: the blastocyst covering is still undamaged. 3 A, 3B × 30.

Fig. 3.

Seven days 3 hours post coitum. A. Mesometrial region of a uterus with blastocyst: digestion of the blastocyst covering has started in the area of the embryonic disc. B. Antimesometrial region of uterus with blastocyst: the blastocyst covering is still undamaged. 3 A, 3B × 30.

Seven days 18 hours p.c. (Fig. 4A) we found a protease pattern similar to that already described by Denker (1971 b); highest activity antimesometrially in the area of trophoblast, low activity mesometrially in the area of embryonic disc (Fig. 4B). Microscopic examination reveals that the endometrium and trophoblast are inactive in the antimesometrial region as well. The digesting blastocyst covering is responsible for gelatin lysis; its former structure is still recognizable. Some fragments of the blastocyst covering are carried into the surrounding endometrium, but may also be found in longitudinal sections (Fig. 4C, D) in front of and rather distant behind the blastocyst in the uterine ducts.

Fig. 4.

Seven days 18 hours post coitum. A. Uterus with blastocyst (overall): small activity, mesometrially, high activity antimesometrially. B. Mesometrial region of uterus with blastocyst: lysis of blastocyst covering has ceased. C, D. Longitudinal sections of uterus with blastocyst : fragments of blastocyst covering have been carried deep into the uterine lumen. E. Boundary between high and low activity (detail of 4 A). The arrows indicate the trophoblast cells. 4 A × 10, 4B × 30, 4C, D, E × 50.

Fig. 4.

Seven days 18 hours post coitum. A. Uterus with blastocyst (overall): small activity, mesometrially, high activity antimesometrially. B. Mesometrial region of uterus with blastocyst: lysis of blastocyst covering has ceased. C, D. Longitudinal sections of uterus with blastocyst : fragments of blastocyst covering have been carried deep into the uterine lumen. E. Boundary between high and low activity (detail of 4 A). The arrows indicate the trophoblast cells. 4 A × 10, 4B × 30, 4C, D, E × 50.

When all fragments were completely dissolved the activity also ceased. The sharp boundary of protease activity therefore indicates the point where lysis of the blastocyst covering is complete, and does not mark the border of the embryonic disc. The arrows point to the trophoblast.

The proteases in the uterine secretion can be easily obtained together with the remaining secretion by flushing the uterine horns. By gel and ion exchange chromatography as well as by disc and cellulose acetate electrophoresis it could not be separated from the β-glycoprotein fraction (Kirchner et al. 1971). Agar gel electrophoresis now showed the uterine secretion to contain two proteases of different electrophoretic mobility in agar gel, which appear after the fifth day p.c. (Fig. 5A). Since both move towards the cathode, they must therefore be assigned the β2-globulins. By cutting out the protease-containing agar and eluting the proteins with physiological NaCl solution we were able to separate the two proteases (Fig. 5B, C), enabling further immunological analysis.

Fig. 5.

Electrophoresis of uterine and blastocyst proteases. A. Proteases in uterine secretion six days 18 hours post coitum. B, C. Isolated uterine proteases. D. Protease in the antimesometrial part of blastocyst covering seven days 18 hours, post coitum.

Fig. 5.

Electrophoresis of uterine and blastocyst proteases. A. Proteases in uterine secretion six days 18 hours post coitum. B, C. Isolated uterine proteases. D. Protease in the antimesometrial part of blastocyst covering seven days 18 hours, post coitum.

Secretion protease which is quite different from the blastocyst protease could be localized in the blastocyst covering (Fig. 5D). It also seems to be composed of two molecules. It moves towards the anode, migrating with the α2-globulins. Material for analysis of the blastocyst protease was obtained from the anti-mesometrial side of an implanted blastocyst of the 7 days 18 hours p.c. From isolated blastocysts of 6 days 18 hours p.c. the protease could not be extracted in sufficient amounts.

Two results demand special attention: 1. The blastocyst covering itself contains an inherent protease. 2. The blastocyst protease is not identical with any of the uterine secretion proteases. Since furthermore the trophoblast is proteolytically inactive, two questions follow: (a) Where does the blastocyst protease originate? (b) What are the functions of the secretion proteases?

Beginning with the second question we assume that the secretion proteases are involved in the first implantation steps, perhaps by participation in digestion of the blastocyst covering after the seventh day p.c. But it is conceivable that they are primarily necessary for establishment of the mesometrial–antimesometrially orientated attachment of the blastocyst prior to the seventh day p.c. In this case there would be a temporarily fixed hydrolysis of the apical walls of the adjacent endometrial cells and (or) the outer layer of the blastocyst covering itself. This would require restricted localization of the activity.

Previously we were unable to determine the site of synthesis of the uterine secretion proteases in the endometrial cells with enzymic methods. The conspicuously high activity in the larger parts of the uterine lumen led us to infer that either the proteases are secreted only in these areas, or the proteases are activated there. Using immune histological methods we have been able to show that at the 5 days 18 hours p.c. the blastocyst covering is surrounded by uteroglobin-containing outer layer (Kirchner, 1972). We interpreted this as attached uterine secretion, perhaps identical with the gloeolemma (Boving, 1957). This layer might also contain high concentrations of secretion proteases, possible traces of which we found in the activity spots of washed blastocysts (Fig. 1 C, D). In both cases the result would be an accumulation of protease in the middle part of the uterus, where we assume it has its main function.

Digestion of the blastocyst covering requires trypsin-like enzymes (Bowman & McLaren, 1970). We localized enzymes of this kind in the blastocyst covering itself, but could not pin down their place of origin. Either during the sixth day p.c. or earlier, an inactive (and therefore with enzymic methods not detectable) protease penetrates into the blastocyst covering. Here it is activated during the sixth day p.c. by an intrinsic or extrinsic factor; or perhaps previously in one of the steps of formation of the blastocyst covering an inactive protease is incorporated and later activated by an extrinsic factor. Such a factor may be synthesized by the trophoblast, but could also come from the uterine secretion. In uterine secretion Schwick (1965) has found a neuraminidase which may participate in hydrolysis of the acid mucosubstances. This carbohydrate lysis is considered to facilitate the following proteolysis (Denker, 1970). A reaction of this kind could serve as a model for activation of an incorporated blastocyst protease. We shall pursue these questions.

This investigation was supported by the Deutsche Forchungsgemeinschaft.

Adams
,
C. W. M.
&
Tltqan
,
N. A.
(
1961
).
The histochemical demonstration of proteases by a gelatin-silver film substrate
.
J. Histochem. Cytochem
.
9
,
469
472
.
Beier
,
H. M.
(
1968
).
Biochemisch-entwicklungsphysiologische Untersuchungen am Proteinmilieu für die Blastozystenentwicklung des Kaninchens (Oryctolagus cuniculus)
.
Zool. Jb. (Anat
.)
85
,
72
190
.
Bergstrom
,
S.
(
1970
).
Estimation of proteolytic activity at mouse implantation sites by the gelatin digestion method
.
J. Reprod. Fert
.
23
,
481
485
.
Böving
,
B. G.
(
1957
).
Rabbit egg coverings
.
Anat. Rec
.
127
,
270
.
Bowman
,
P.
&
Mclaren
,
A.
(
1970
).
The reaction of the mouse blastocyst and its zona pellucida to enzymes in vitro
.
J. Embryol. exp. Morph
.
24
,
331
334
.
Denker
,
H. W.
(
1970
).
Topochemie hochmolekularer Kohlenhydratsubstanzen in Frühentwicklung und Implantation des Kaninchens. IL Beiträge zu entwicklungsphysiologischen Fragestellungen
.
Zool. Jb. (Physiol
.)
75
,
246
308
.
Denker
,
H. W.
(
1971a
).
Substratfilmtest für den Proteasennachweis
.
Acta Histochem. Suppl
.
10
,
303
305
.
Denker
,
H. W.
(
1971b
).
Enzym-Topochemie von Frühentwicklung und Implantation des Kanichens. III. Proteasen
.
Histochemie
25
,
344
360
.
Kirchner
,
C.
(
1969
).
Untersuchungen an uterusspezifischen Glykoproteinen während der frühen Gravidität des Kaninchens Oryctolagus cuniculus
.
Wilhelm Roux Arch. EntwMech. Org
.
164
,
97
133
.
Kirchner
,
C.
(
1972
).
Immune histologic studies on the synthesis of a uterine-specific protein in the rabbit and its passage through the blastocyst coverings
.
Fert. Steril
.
23
,
131
136
.
Kirchner
,
C.
,
Hirschhäuser
,
C.
&
Kionke
,
M.
(
1971
).
Protease activity in rabbit uterine secretion 24 hours before implantation
.
J. Reprod. Fert
.
27
,
259
260
.
Owers
,
N. O.
&
Blandau
,
R. J.
(
1971
).
Proteolytic activity of the rat and guinea pig blastocyst in vitro
.
In The Biology of the Blastocyst
(ed.
R. J.
Blandau
), pp.
207
223
.
The University of Chicago Press
.
Schwick
,
H. G.
(
1965
).
Chemisch-entwicklungsphysiologische Beziehungen von Uterus zu Blastocyste des Kaninchens Oryctolagus cuniculus
.
Wilhelm Roux Arch. EntwMech. Org
.
156
,
283
343
.