- Email: [email protected]
Effects of fibronectin and its peptide fragment on preimplantation mouse embryo
BASIC SCIENCE SECTION Effects of fibronectin and its peptide fragment on preimplantation mouse embryo Akira Tsuiki, MD, Janet Preyer, BSc, and Terry T. Hung, MD, PhD Miami, Florida To investigate the involvement of fibronectin in the early development of mouse embryo, embryos at the two-cell stage were cultured in serum-free Ham F-10 medium with or without fibronectin-related substances. When antibody raised against fibronectin was added at concentrations of 0, 10- 3 , 10- 2 , and 2 x 10- 2 (vol :vol), the percentages of cells in the blastocyst stage at 72 hours were 88.5%, 67.2%, 48.1%, and 0%, respectively. No inhibitory effects were observed with nonspecific antibody. A low concentration of exogenous fibronectin (up to 0.2 IJ.moI/L) slightly enhanced mouse embryo development after 48 hours. However, a higher concentration (>0.4 IJ.moI/L) of fibronectin inhibited mouse embryo development. A competitive blocker of cell membrane fibronectin receptors, oligopeptide (Gly-Arg-Gly-Asp-Ser), inhibited mouse embryo development in a dose-dependent manner. Meanwhile, control peptide (Gly-Arg-Gly-Glu-Ser-Pro), which does not contain the crucial cell binding sequence, did not show any inhibitory effects. These findings suggest the necessity of endogenous fibronectin-blastomere interaction in early mouse embryo development. (AM J OBSTET GVNECOL 1989;160:724-8.)
Key words: Fibronectin, synthetic peptide, antibody, mouse embryo, differentiation Fibronectin is a cellular surface and extracellular matrix glycoprotein implicated in the mediation of cell adhesion to extracellular matrix. Fibronectin has been found to promote and guide cell migration and differentiation in several parts of late embryonic development, including avian gastrula, neural crest, area vasculosa, and primordial germ cells.'-' The differentiation process in embryonic development is thought to start with compaction, when the blastomeres flatten and increase their contact with each other and develop distinct apical and basal membrane and cytoplasmic domains (polarization)! The changes associated with compaction clearly point to alterations in both the surface properties of the cells and the organization of the cytoskeleton! The question arises whether fibronectin is involved in the process of early embryonic development. Although immunocytochemical methods have been used to investigate the earliest appearance of fibronectin, its presence in the early embryo has remained equivocal because its detection appears to vary with different antibodies and methods of fixation.5 This study was designed to investigate the From the Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, Umvemty ofMiami School ofM ed/cme. Received for publication November 18, 1987.. revISed September 14, 1988; accepted October 14,1988. Repnnt requests: Terry T. Hung, MD, PhD. Division of Reproductive Endocnnology. Department of Obstetrics and Gynecology (D-5) , P. O. Box 016960, Mtami, FL 33101.
724
fibronectin-cell interaction in an in vitro culture system of mouse embryos from the two-cell stage up to the late blastocyst. Antibody raised against fibronectin, fibronectin itself, and two synthetic peptide probes were added to the culture medium to evaluate their effects on the development of the early mouse embryo.
Material and methods Collection of embryos. Virgin female mice (C57BLl6J x CBA, 5 to 9 weeks old) were made to superovulate by injection of 5 IU of pregnant mare's serum gonadotropin (Sigma Chemical Company, St. Louis, Mo.) followed 48 hours later by 5 IU of human chorionic gonadotropin (Sigma Chemical Co.). Each female mouse was then mated with a fertile male mouse (CD-I, 3 to 10 months old). The following morning the female mice were checked for the presence of a vaginal plug. Pregnant female mice were killed by cervical dislocation 40 to 44 hours after human chorionic gonadotropin injection. Embryos at the two-cell stage were recovered by flushing the oviducts with Dulbecco's phosphate buffered saline solution (pH = 7.4, 280 mOsm/kg; GIBCO Laboratories, Grand Island, N.Y.). Any embryos appearing abnormal or degenerated at this point were excluded from further culture. Culture media. The basic culture medium was a modified Ham F-I0 medium6 supplemented with 1% bovine serum albumin (free of fatty acids and globulin; Sigma Chemical Co.). Water used for the preparation of the media was supplied from a Milli-Q R04 reagent
Fibronectin and mouse embryo
Volume 160 Number 3
., 0
100
>-
80
UI
.,a.
60
OJ
.,>
40
.?: n;
20
..Q E "0 0
725
c
E
0
z
;1!
0
24 HOURS
ilill: CONTROL 11111
Ab (1 :50)
48 HOURS INCUBA TION TIME Ab (1 :1000)
72 HOURS
III Ab
(1 :100)
;;; NON - SPECIFIC ANTIBODY (1 :50)
Fig. 1. Effects of antibody (Ab) raised against fibronectin and nonspecific antibody on mouse embryo development over 72 hours. Error bars represent SEM. AsterISk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with control group.
water system equipped with ion exchangers and charcoal filters (Millipore Co., Bedford, Mass.). The substances to be examined were goat antiserum raised against rat fibronectin (anti-FN ; Calibiochem, La Jolla, Calif.), rat fibronectin (Sigma Chemical Co.). and the synthetic oligopeptides Gly-Arg-Gly-Asp-Ser (GRGDS; Peninsula Laboratories, Belmont, Calif.) and Gly-ArgGly-Glu-Ser-Pro (GRGESP; gift from Dr. Erkki Ruoslahti, La Jolla Cancer Research Foundation, La Jolla, Calif.). Reactions to anti-FN were examined at the concentrations of 0, 10- 3, 10- 2 , and 2 X 10- 2 (vol: vol). Reactions to fibronectin were examined at the concentrations of 0.04, 0.1, 0.2, 0.4, l.l, and 2.2 fJ.mol/ L. A competitive inhibitor of cell binding sequence of fibronectin, GRGDS was studied at the levels of 2, 10, 20, 40, 200, 400, and 1000 fJ.mollL. Another peptide, GRGESP, which is closely related to GRGDS structurally but does not contain the crucial cell binding sequence, was added at the same molar levels as those of GRGDS. Synthetic peptide solution was prepared according to a previous report. 7 The test and control media were adjusted to the range of pH 7.3 to 7.4 and of osmolarity (mean ± SD = 280 ± 5 mOsm/kg. These media were transferred to organ culture dishes (No. 3037; Falcon, Oxnard, Calif.) and preincubated in 5% CO 2 in air at 37° C overnight before the culture of mouse embryos. Culture of embryos. Normal embryos at the two-cell stage were washed with culture medium, then transferred to organ culture dishes containing I ml of the appropriate culture medium. These embryos were incubated under 5% CO 2 in air at 37° C in a humidified atmosphere for 72 hours. Embryos were examined by use of an inverted-phase contrast microscope every 24 hours. All embryos were classified as degenerated, two-
cell, three- or four-cell, five- or seven-cell, eight cell, compaction stage, morula, early blastocyst, and expanded blastocyst. In the control group (no fibronectinrelated substances added to medium), more than 70% of the embryos had developed to the blastocyst stage after 72 hours. In each tested medium, more than four replicates of 15 to 20 embryos at the two-cell stage were cultured to test the ability of the medium to enhance or inhibit embryo development. Data analysis. The control range for normal growth rate was defined as five to eight cells at 24 hours, com paction stage or greater at 48 hours, and early blastocyst or greater at 72 hours after culture. Because it is not appropriate to apply standard parametric statistical procedures to percentage data, the data were treated by arcsine transformation before statistical evaluation. Statistical analysis was by one-way analysis of variance followed by the Newman-Keuls multiple comparison test. Dose dependency was examined by the Spearman rank correlation test. Results When antibody raised against fibronectin was added to culture medium, it showed an inhibitory effect on mouse embryo development in a dose-dependent manner (r = 0.94 at 24 hours; r = -0.91 at 48 hours ; and r = - 0.98 at 72 hours). This inhibitory effect was observed from as early as 24 hours and started at a very low concentration of antibody (I ; 1000, vol: vol) . At a concentration of I ; 50 (vol; vol), no embryos had been able to develop to the blastocyst stage. When nonspecific antibody (antibody developed against l3-endorphin; produced and validated by one of the authors [T. T. H]) was added to the culture medium at the concentration of I: 50 (vol: vol), no inhibitory
726 Tsuiki, Preyer, and Hung
March 1989 Am J Obslel Gynecol
100 III
0
>.Q
80
LU
60
E
."
'"0
Q.
Qj
> 0
'"
~
Iii
40 20
E
(; Z ;J!
0
24 HOURS
48 HOURS INCUBA TlON TIME
liili CONTROL 11111
FN 0.21JM
72 HOURS
FN 0.041JM III FN O.1IJM ;;; FN 0.41JM • FN 1.11JM •
~
FN 2.21JM
Fig. 2. Effects of various concentrations of exogenous fibronectin on mouse embryo development over 72 hours. Error bars represent SEM. Asterisk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with control group.
100 VI
0
>
80
a:~
ClIVI
:::Ea:
w=>
...JO ...J:I:
W ..
u~
60 40
CD I
ItO
;J!
20 0
2
10
:!!i!' GRGDS
20
40
200 400 GRGESP
1000 CONT
PEP TIDES CONCENTRATION (101M) Fig. 3. Comparison of embryo growth at 24 hours with various concentrations of oligopeptides. Error bars represent SEM. Asterisk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with each control (CONT).
effect was observed at any stage of embryo development (Fig. 1). Fig. 2 shows the effects of exogenous fibronectin on embryo development over a 72-hour period of observation. In terms of percentage of cells at the compaction stage at 48 hours and at the blastocyst stage at 72 hours, up to 0.2 j.l.mollL of exogenous fibronectin slightly enhanced embryo development; however, higher concentrations (>0.4 j.l.moI/L) of fibronectin inhibited mouse embryo development (bimodal effects of fibronectin; enhancement at lower concentration and inhibition at higher concentration). These bimodal effects of fibronectin were not observed during the first 24 hours. Figs. 3 to 5 compare embryo growth with various concentrations of GRGDS and GRGESP at 24, 48, and 72 hours, respectively. The addition of GRGDS at increasing concentrations from 2 to 1000 j.l.moll L showed
inhibitory effects on embryo development. These effects were dose dependent (r = - 0.84 at 24 hours; r = 0.88 at 48 hours; and r == - 0.92 at 72 hours) and were observed at a concentration as low as 10 j.l.mollL after 24 hours. Complete inhibition occurred at a concentration of 1000 j.l.mol after 24 hours. In contrast, the addition of GRGESP did not show any inhibitory effects within the same concentration range as GRGDS.
Comment The extracellular matrix is considered important in organizing the cells by maintaining order in their environment and providing physical support for the cells. In addition, the order imposed by the extracellular matrix provides a hospitable substratum for the movement and migration of cells. 5 One of the major components of the extracellular matrix, fibronectin, reportedly first
Fibronectin and mouse embryo
Volume 160 I\umber 3
727
100 w
!.-'
80
«
~
(f)
~
(f)
zlI:
o=> -0
t;
::t
«co ..
Q.
60 40
:::IE~
0
u
iF
20 0
200 400 GRGESP
1000 CONT
PEPTIDES CONCENTRATION (J,JM) Fig. 4. Comparison of embryo growth at 48 hours with various concentrations of oligopeptides.
Error bars represent SEM. Asternk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with each control (CaNT).
100 80 ~
(f)~
>(f)
u II: o=>
60
~o
(f)::t
«N 40 ...J .... ID~
iF
20 0
2
10
20
~m GRGDS
40
200 400 GRGESP
1000
PEPTIDES CONCENTRA nON (J,JM) Fig. 5. Comparison of embryo growth at 72 hours with various concentrations of oligopeptides. A~tertsk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with each control (CaNT).
Error bars represent SEM.
appears at the blastocyst stage in mouse embryo.s Its detection appears to vary with the use of different monoclonal antibodies. Furthermore, the presence of other components of the extracellular matrix can actually mask fibronectin, rendering it undetectable by light microscope.' It has therefore been suggested that fibronectin may be even more widespread than previously thought.' Our results showed that the addition of antibody to fibronectin had an inhibitory effect on mouse embryo development; no inhibitory effect was observed after the addition of nonspecific antibody. This finding suggests that the inhibitory effect of antiFN is more likely due to the neutralization of endogenous fibronectin than to the toxicity of other components in the antibody. This finding could be considered the first indirect evidence suggesting the presence of endogenous fibronectin in a very early stage of the mouse embryo, because the culture medium used in
this study is serum free and any contamination of exogenous fibronectin in bovine serum albumin has been shown to be negligible. Based on the available data, the ability of the early embryo to synthesize fibronectin remains to be determined, because the presence of fibronectin can be endogenous. Although we cannot eliminate this ability, Darribere et al. 9 have used radiolabeled amino acid incorporation and twodimensional gel electrophoresis to show that fibronectin synthesis in Pleurodeles waltlii could be detected in oocytes, fertilized eggs, and cleaving embryos, and that stored maternal messenger ribonucleic acid coding for fibronectin contributed to this synthesis during early embryonic development. If endogenous fibronectin is required for normal embryo development, the question raised at this point is whether the addition of exogenous fibronectin would enhance embryo growth. The addition of exogenous
728 Tsuiki, Preyer, and Hung
fibronectin showed a bimodal effect (slight enhancement at lower concentration and inhibition at higher concentration) at 48 and 72 hours of incubation. This bimodal effect had not been found at the first 24-hour observation. Such a difference may be due to different responsiveness depending on the embryo stage. Perhaps exogenous fibronectin may not be essential for early embryo development (up to the eight-cell stage), as the presence of exogenous fibronectin in embryo development after the compaction stage remains to be determined. It is well known that fibronectin behaves like growth factor as a preferred substrate for the culture of many cells in serum-free, defined media at a concentration of 1 to 5 fLg / ml by increasing the plating efficiency." 10 This may be related to the enhancing effect of exogenous fibronectin at a lower concentration. Autoinhibition by a high concentration of exogenous fibronectin can be explained by the fact that a sufficiently high level of fibronectin would occupy most of their receptors on the blastomere, resulting in an inability to bind to endogenous (cellular) fibronectin, which mediates cell adhesion." A theoretic down regulation of fibronectin receptor sites on the blastomere membrane could be an alternative explanation for the bimodal effects of exogenous fibronectin. Cell attachment activity of fibronectin can be accounted for by small synthetic fragments of the cell attachment domain of fibronectin.'2 This sequence has been identified as Arg-Glu-Asp, a key structural element that constitutes a cellular recognition determinant. 13 Different oligopeptides containing this sequence have been synthesized, making it possible to inhibit the cell attachment function of fibronectin specifically in a non cytotoxic manner.' 7 II-II Our results utilizing one of these peptides, GRGDS, showed a significant dosedependent inhibitory effect on embryo development with increasing levels. In contrast, the control peptide, GRGESP, which contains no cell binding sequence, did not show any inhibitory effect when added at the same molar levels as GRGDS. These findings indicate that the inhibitory effects of GRGDS are specific and likely to involve a competitive block of receptors for the cell binding domain for fibronectin. Previous studies often required millimolar concentrations of peptide for full inhibition. s, 7. II·'S These findings also agree with our results, which indicate that complete inhibition of embryo development occurred at a concentration of 1 mmol/L of GRGDS. The conclusion drawn from these results is that endogenous fibronectin-blastomere interaction is crucial for early mouse embryo development. For embryo to develop normally, its blastomeres may have to be interconnected to each other, either via cell binding fibronectin alone or plus another component of the extracellular matrix. To clarify the possible role of fibronectin in the implantation process, in vivo admin-
March 1989
Am J Obstet Gynecol
istration of synthetic peptides to mated female mice and subsequent evaluation of their ability to conceive is under study in our laboratories. These studies will not only help in understanding the role of fibronectin in implantation, but may also provide potential clinical applications, such as the development of a new postcoital contraceptive agent or new diagnostic or therapeutic tools in the field of infertility. . We are grateful to Dr. E. R. Ruoslahti (La Jolla Cancer Research Foundation, La Jolla, Calif.) for his generous gifts of synthetic peptides. We also thank Dr. William J. LeMaire for his helpful discussion and support. REFERENCES I. Hynes RO. Fibronectin and its relation to cellular structure and behavior. In: Hay ED, ed. Cell biology of extracellular matrix. New York: Plenum Press, 1981:295-334. 2. Greenberg ]H, Seppa S, Seppa H, Hewitt AT. Role of collagen and fibronectin in neural crest cell adhesion and migration. Dev Bioi 1981;87:259-66. 3. Boucaut]C, Darribere T, Poole T], Aoyama H, Yamada KM, Thiery ]P. Biologically active synthetic peptides as probes of embryonic development: a competitive peptide inhibitor of fibronectin function inhibits gastrulation in amphibian embryos and neural crest cell migration in avian embryos.] Cell Bioi 1984;99:1822-30. 4. Hogan B, Costantini F, Lacy E. Compaction and the formation of the blastocyst: the first differentiation events. In: Hogan B, Costantini F, Lacy E, eds. Manupulating the mouse embryo. New York: Cold Spring Harbor Laboratory, 1986:41. 5. Sanders EJ. Non-collagenous glycoproteins. In: Sanders E], ed. Cytochemistry ofthe cell surface and extracellular matrix during early embryonic development. Stuttgart: Gustav Fischer Verlag, 1986:7. 6. Sokoloski ]E, Wolf DP. Laboratory details in an in vitro fertilization and embryo transfer program. In: Wolf DP, Quigley MM, eds. Human in vitro fertilization and embryo transfer. New York: Plenum Press, 1984:275-96. 7. Humphries M], Olden K, Yamada KM. A synthetic peptide from fibronectin inhibits experimental metastasis of murine melanoma cells. Science 1986;233:467-70. 8. Wartiovaara], Leivo I, Vaheri A. Expression of the cell surface-associated glycoprotein, fibronectin, in the early mouse embryo. Dev BioI 1979;69:247-57. 9. Darribere T, Boucher D, Lacroix ]C, Boucaut ]C. Fibronectin synthesis during oogenesis and early development of the amphibian Pleurodeles waltlii. Cell Differentiation 1984;14:171-7. 10. Orly ], Sato G. Fibronectin mediates cytokinesis and growth of rat follicular cells in serum-free medium. Cell 1979; 17:295-305. II. Yamada KM. Kennedy DW. Dualistic nature of adhesion protein function: fibronectin can autoinhibit fibronectin function. ] Cell Bioi 1984;99:29-36. 12. Pierschbacher MD, Ruoslahti E. Cell attachment activity of fibronectin can be duplicated by small synthetic fragments of the molecule. Nature 1984;309:30-4. 13. Pierschbacher MD, Ruoslahti E. Variants of the cell recognition site of fibronectin that retain attachmentpromoting activity. Proc Nat! Acad Sci USA 1984;81: 5985-8. 14. Armant R, Kaplan HA, Mover H, Lennarz W]. The effect of hexapeptides on attachment and outgrowth of mouse blastocysts cultured in vitro: evidence for the involvement of the cell recognition tripeptide Arg-Gly-Asp. Proc Nat! Acad Sci USA 1986;83:6751-5.
Key words: Fibronectin, synthetic peptide, antibody, mouse embryo, differentiation Fibronectin is a cellular surface and extracellular matrix glycoprotein implicated in the mediation of cell adhesion to extracellular matrix. Fibronectin has been found to promote and guide cell migration and differentiation in several parts of late embryonic development, including avian gastrula, neural crest, area vasculosa, and primordial germ cells.'-' The differentiation process in embryonic development is thought to start with compaction, when the blastomeres flatten and increase their contact with each other and develop distinct apical and basal membrane and cytoplasmic domains (polarization)! The changes associated with compaction clearly point to alterations in both the surface properties of the cells and the organization of the cytoskeleton! The question arises whether fibronectin is involved in the process of early embryonic development. Although immunocytochemical methods have been used to investigate the earliest appearance of fibronectin, its presence in the early embryo has remained equivocal because its detection appears to vary with different antibodies and methods of fixation.5 This study was designed to investigate the From the Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, Umvemty ofMiami School ofM ed/cme. Received for publication November 18, 1987.. revISed September 14, 1988; accepted October 14,1988. Repnnt requests: Terry T. Hung, MD, PhD. Division of Reproductive Endocnnology. Department of Obstetrics and Gynecology (D-5) , P. O. Box 016960, Mtami, FL 33101.
724
fibronectin-cell interaction in an in vitro culture system of mouse embryos from the two-cell stage up to the late blastocyst. Antibody raised against fibronectin, fibronectin itself, and two synthetic peptide probes were added to the culture medium to evaluate their effects on the development of the early mouse embryo.
Material and methods Collection of embryos. Virgin female mice (C57BLl6J x CBA, 5 to 9 weeks old) were made to superovulate by injection of 5 IU of pregnant mare's serum gonadotropin (Sigma Chemical Company, St. Louis, Mo.) followed 48 hours later by 5 IU of human chorionic gonadotropin (Sigma Chemical Co.). Each female mouse was then mated with a fertile male mouse (CD-I, 3 to 10 months old). The following morning the female mice were checked for the presence of a vaginal plug. Pregnant female mice were killed by cervical dislocation 40 to 44 hours after human chorionic gonadotropin injection. Embryos at the two-cell stage were recovered by flushing the oviducts with Dulbecco's phosphate buffered saline solution (pH = 7.4, 280 mOsm/kg; GIBCO Laboratories, Grand Island, N.Y.). Any embryos appearing abnormal or degenerated at this point were excluded from further culture. Culture media. The basic culture medium was a modified Ham F-I0 medium6 supplemented with 1% bovine serum albumin (free of fatty acids and globulin; Sigma Chemical Co.). Water used for the preparation of the media was supplied from a Milli-Q R04 reagent
Fibronectin and mouse embryo
Volume 160 Number 3
., 0
100
>-
80
UI
.,a.
60
OJ
.,>
40
.?: n;
20
..Q E "0 0
725
c
E
0
z
;1!
0
24 HOURS
ilill: CONTROL 11111
Ab (1 :50)
48 HOURS INCUBA TION TIME Ab (1 :1000)
72 HOURS
III Ab
(1 :100)
;;; NON - SPECIFIC ANTIBODY (1 :50)
Fig. 1. Effects of antibody (Ab) raised against fibronectin and nonspecific antibody on mouse embryo development over 72 hours. Error bars represent SEM. AsterISk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with control group.
water system equipped with ion exchangers and charcoal filters (Millipore Co., Bedford, Mass.). The substances to be examined were goat antiserum raised against rat fibronectin (anti-FN ; Calibiochem, La Jolla, Calif.), rat fibronectin (Sigma Chemical Co.). and the synthetic oligopeptides Gly-Arg-Gly-Asp-Ser (GRGDS; Peninsula Laboratories, Belmont, Calif.) and Gly-ArgGly-Glu-Ser-Pro (GRGESP; gift from Dr. Erkki Ruoslahti, La Jolla Cancer Research Foundation, La Jolla, Calif.). Reactions to anti-FN were examined at the concentrations of 0, 10- 3, 10- 2 , and 2 X 10- 2 (vol: vol). Reactions to fibronectin were examined at the concentrations of 0.04, 0.1, 0.2, 0.4, l.l, and 2.2 fJ.mol/ L. A competitive inhibitor of cell binding sequence of fibronectin, GRGDS was studied at the levels of 2, 10, 20, 40, 200, 400, and 1000 fJ.mollL. Another peptide, GRGESP, which is closely related to GRGDS structurally but does not contain the crucial cell binding sequence, was added at the same molar levels as those of GRGDS. Synthetic peptide solution was prepared according to a previous report. 7 The test and control media were adjusted to the range of pH 7.3 to 7.4 and of osmolarity (mean ± SD = 280 ± 5 mOsm/kg. These media were transferred to organ culture dishes (No. 3037; Falcon, Oxnard, Calif.) and preincubated in 5% CO 2 in air at 37° C overnight before the culture of mouse embryos. Culture of embryos. Normal embryos at the two-cell stage were washed with culture medium, then transferred to organ culture dishes containing I ml of the appropriate culture medium. These embryos were incubated under 5% CO 2 in air at 37° C in a humidified atmosphere for 72 hours. Embryos were examined by use of an inverted-phase contrast microscope every 24 hours. All embryos were classified as degenerated, two-
cell, three- or four-cell, five- or seven-cell, eight cell, compaction stage, morula, early blastocyst, and expanded blastocyst. In the control group (no fibronectinrelated substances added to medium), more than 70% of the embryos had developed to the blastocyst stage after 72 hours. In each tested medium, more than four replicates of 15 to 20 embryos at the two-cell stage were cultured to test the ability of the medium to enhance or inhibit embryo development. Data analysis. The control range for normal growth rate was defined as five to eight cells at 24 hours, com paction stage or greater at 48 hours, and early blastocyst or greater at 72 hours after culture. Because it is not appropriate to apply standard parametric statistical procedures to percentage data, the data were treated by arcsine transformation before statistical evaluation. Statistical analysis was by one-way analysis of variance followed by the Newman-Keuls multiple comparison test. Dose dependency was examined by the Spearman rank correlation test. Results When antibody raised against fibronectin was added to culture medium, it showed an inhibitory effect on mouse embryo development in a dose-dependent manner (r = 0.94 at 24 hours; r = -0.91 at 48 hours ; and r = - 0.98 at 72 hours). This inhibitory effect was observed from as early as 24 hours and started at a very low concentration of antibody (I ; 1000, vol: vol) . At a concentration of I ; 50 (vol; vol), no embryos had been able to develop to the blastocyst stage. When nonspecific antibody (antibody developed against l3-endorphin; produced and validated by one of the authors [T. T. H]) was added to the culture medium at the concentration of I: 50 (vol: vol), no inhibitory
726 Tsuiki, Preyer, and Hung
March 1989 Am J Obslel Gynecol
100 III
0
>.Q
80
LU
60
E
."
'"0
Q.
Qj
> 0
'"
~
Iii
40 20
E
(; Z ;J!
0
24 HOURS
48 HOURS INCUBA TlON TIME
liili CONTROL 11111
FN 0.21JM
72 HOURS
FN 0.041JM III FN O.1IJM ;;; FN 0.41JM • FN 1.11JM •
~
FN 2.21JM
Fig. 2. Effects of various concentrations of exogenous fibronectin on mouse embryo development over 72 hours. Error bars represent SEM. Asterisk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with control group.
100 VI
0
>
80
a:~
ClIVI
:::Ea:
w=>
...JO ...J:I:
W ..
u~
60 40
CD I
ItO
;J!
20 0
2
10
:!!i!' GRGDS
20
40
200 400 GRGESP
1000 CONT
PEP TIDES CONCENTRATION (101M) Fig. 3. Comparison of embryo growth at 24 hours with various concentrations of oligopeptides. Error bars represent SEM. Asterisk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with each control (CONT).
effect was observed at any stage of embryo development (Fig. 1). Fig. 2 shows the effects of exogenous fibronectin on embryo development over a 72-hour period of observation. In terms of percentage of cells at the compaction stage at 48 hours and at the blastocyst stage at 72 hours, up to 0.2 j.l.mollL of exogenous fibronectin slightly enhanced embryo development; however, higher concentrations (>0.4 j.l.moI/L) of fibronectin inhibited mouse embryo development (bimodal effects of fibronectin; enhancement at lower concentration and inhibition at higher concentration). These bimodal effects of fibronectin were not observed during the first 24 hours. Figs. 3 to 5 compare embryo growth with various concentrations of GRGDS and GRGESP at 24, 48, and 72 hours, respectively. The addition of GRGDS at increasing concentrations from 2 to 1000 j.l.moll L showed
inhibitory effects on embryo development. These effects were dose dependent (r = - 0.84 at 24 hours; r = 0.88 at 48 hours; and r == - 0.92 at 72 hours) and were observed at a concentration as low as 10 j.l.mollL after 24 hours. Complete inhibition occurred at a concentration of 1000 j.l.mol after 24 hours. In contrast, the addition of GRGESP did not show any inhibitory effects within the same concentration range as GRGDS.
Comment The extracellular matrix is considered important in organizing the cells by maintaining order in their environment and providing physical support for the cells. In addition, the order imposed by the extracellular matrix provides a hospitable substratum for the movement and migration of cells. 5 One of the major components of the extracellular matrix, fibronectin, reportedly first
Fibronectin and mouse embryo
Volume 160 I\umber 3
727
100 w
!.-'
80
«
~
(f)
~
(f)
zlI:
o=> -0
t;
::t
«co ..
Q.
60 40
:::IE~
0
u
iF
20 0
200 400 GRGESP
1000 CONT
PEPTIDES CONCENTRATION (J,JM) Fig. 4. Comparison of embryo growth at 48 hours with various concentrations of oligopeptides.
Error bars represent SEM. Asternk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with each control (CaNT).
100 80 ~
(f)~
>(f)
u II: o=>
60
~o
(f)::t
«N 40 ...J .... ID~
iF
20 0
2
10
20
~m GRGDS
40
200 400 GRGESP
1000
PEPTIDES CONCENTRA nON (J,JM) Fig. 5. Comparison of embryo growth at 72 hours with various concentrations of oligopeptides. A~tertsk denotes significantly lower (p < 0.05) percentage of normally developed embryos as compared with each control (CaNT).
Error bars represent SEM.
appears at the blastocyst stage in mouse embryo.s Its detection appears to vary with the use of different monoclonal antibodies. Furthermore, the presence of other components of the extracellular matrix can actually mask fibronectin, rendering it undetectable by light microscope.' It has therefore been suggested that fibronectin may be even more widespread than previously thought.' Our results showed that the addition of antibody to fibronectin had an inhibitory effect on mouse embryo development; no inhibitory effect was observed after the addition of nonspecific antibody. This finding suggests that the inhibitory effect of antiFN is more likely due to the neutralization of endogenous fibronectin than to the toxicity of other components in the antibody. This finding could be considered the first indirect evidence suggesting the presence of endogenous fibronectin in a very early stage of the mouse embryo, because the culture medium used in
this study is serum free and any contamination of exogenous fibronectin in bovine serum albumin has been shown to be negligible. Based on the available data, the ability of the early embryo to synthesize fibronectin remains to be determined, because the presence of fibronectin can be endogenous. Although we cannot eliminate this ability, Darribere et al. 9 have used radiolabeled amino acid incorporation and twodimensional gel electrophoresis to show that fibronectin synthesis in Pleurodeles waltlii could be detected in oocytes, fertilized eggs, and cleaving embryos, and that stored maternal messenger ribonucleic acid coding for fibronectin contributed to this synthesis during early embryonic development. If endogenous fibronectin is required for normal embryo development, the question raised at this point is whether the addition of exogenous fibronectin would enhance embryo growth. The addition of exogenous
728 Tsuiki, Preyer, and Hung
fibronectin showed a bimodal effect (slight enhancement at lower concentration and inhibition at higher concentration) at 48 and 72 hours of incubation. This bimodal effect had not been found at the first 24-hour observation. Such a difference may be due to different responsiveness depending on the embryo stage. Perhaps exogenous fibronectin may not be essential for early embryo development (up to the eight-cell stage), as the presence of exogenous fibronectin in embryo development after the compaction stage remains to be determined. It is well known that fibronectin behaves like growth factor as a preferred substrate for the culture of many cells in serum-free, defined media at a concentration of 1 to 5 fLg / ml by increasing the plating efficiency." 10 This may be related to the enhancing effect of exogenous fibronectin at a lower concentration. Autoinhibition by a high concentration of exogenous fibronectin can be explained by the fact that a sufficiently high level of fibronectin would occupy most of their receptors on the blastomere, resulting in an inability to bind to endogenous (cellular) fibronectin, which mediates cell adhesion." A theoretic down regulation of fibronectin receptor sites on the blastomere membrane could be an alternative explanation for the bimodal effects of exogenous fibronectin. Cell attachment activity of fibronectin can be accounted for by small synthetic fragments of the cell attachment domain of fibronectin.'2 This sequence has been identified as Arg-Glu-Asp, a key structural element that constitutes a cellular recognition determinant. 13 Different oligopeptides containing this sequence have been synthesized, making it possible to inhibit the cell attachment function of fibronectin specifically in a non cytotoxic manner.' 7 II-II Our results utilizing one of these peptides, GRGDS, showed a significant dosedependent inhibitory effect on embryo development with increasing levels. In contrast, the control peptide, GRGESP, which contains no cell binding sequence, did not show any inhibitory effect when added at the same molar levels as GRGDS. These findings indicate that the inhibitory effects of GRGDS are specific and likely to involve a competitive block of receptors for the cell binding domain for fibronectin. Previous studies often required millimolar concentrations of peptide for full inhibition. s, 7. II·'S These findings also agree with our results, which indicate that complete inhibition of embryo development occurred at a concentration of 1 mmol/L of GRGDS. The conclusion drawn from these results is that endogenous fibronectin-blastomere interaction is crucial for early mouse embryo development. For embryo to develop normally, its blastomeres may have to be interconnected to each other, either via cell binding fibronectin alone or plus another component of the extracellular matrix. To clarify the possible role of fibronectin in the implantation process, in vivo admin-
March 1989
Am J Obstet Gynecol
istration of synthetic peptides to mated female mice and subsequent evaluation of their ability to conceive is under study in our laboratories. These studies will not only help in understanding the role of fibronectin in implantation, but may also provide potential clinical applications, such as the development of a new postcoital contraceptive agent or new diagnostic or therapeutic tools in the field of infertility. . We are grateful to Dr. E. R. Ruoslahti (La Jolla Cancer Research Foundation, La Jolla, Calif.) for his generous gifts of synthetic peptides. We also thank Dr. William J. LeMaire for his helpful discussion and support. REFERENCES I. Hynes RO. Fibronectin and its relation to cellular structure and behavior. In: Hay ED, ed. Cell biology of extracellular matrix. New York: Plenum Press, 1981:295-334. 2. Greenberg ]H, Seppa S, Seppa H, Hewitt AT. Role of collagen and fibronectin in neural crest cell adhesion and migration. Dev Bioi 1981;87:259-66. 3. Boucaut]C, Darribere T, Poole T], Aoyama H, Yamada KM, Thiery ]P. Biologically active synthetic peptides as probes of embryonic development: a competitive peptide inhibitor of fibronectin function inhibits gastrulation in amphibian embryos and neural crest cell migration in avian embryos.] Cell Bioi 1984;99:1822-30. 4. Hogan B, Costantini F, Lacy E. Compaction and the formation of the blastocyst: the first differentiation events. In: Hogan B, Costantini F, Lacy E, eds. Manupulating the mouse embryo. New York: Cold Spring Harbor Laboratory, 1986:41. 5. Sanders EJ. Non-collagenous glycoproteins. In: Sanders E], ed. Cytochemistry ofthe cell surface and extracellular matrix during early embryonic development. Stuttgart: Gustav Fischer Verlag, 1986:7. 6. Sokoloski ]E, Wolf DP. Laboratory details in an in vitro fertilization and embryo transfer program. In: Wolf DP, Quigley MM, eds. Human in vitro fertilization and embryo transfer. New York: Plenum Press, 1984:275-96. 7. Humphries M], Olden K, Yamada KM. A synthetic peptide from fibronectin inhibits experimental metastasis of murine melanoma cells. Science 1986;233:467-70. 8. Wartiovaara], Leivo I, Vaheri A. Expression of the cell surface-associated glycoprotein, fibronectin, in the early mouse embryo. Dev BioI 1979;69:247-57. 9. Darribere T, Boucher D, Lacroix ]C, Boucaut ]C. Fibronectin synthesis during oogenesis and early development of the amphibian Pleurodeles waltlii. Cell Differentiation 1984;14:171-7. 10. Orly ], Sato G. Fibronectin mediates cytokinesis and growth of rat follicular cells in serum-free medium. Cell 1979; 17:295-305. II. Yamada KM. Kennedy DW. Dualistic nature of adhesion protein function: fibronectin can autoinhibit fibronectin function. ] Cell Bioi 1984;99:29-36. 12. Pierschbacher MD, Ruoslahti E. Cell attachment activity of fibronectin can be duplicated by small synthetic fragments of the molecule. Nature 1984;309:30-4. 13. Pierschbacher MD, Ruoslahti E. Variants of the cell recognition site of fibronectin that retain attachmentpromoting activity. Proc Nat! Acad Sci USA 1984;81: 5985-8. 14. Armant R, Kaplan HA, Mover H, Lennarz W]. The effect of hexapeptides on attachment and outgrowth of mouse blastocysts cultured in vitro: evidence for the involvement of the cell recognition tripeptide Arg-Gly-Asp. Proc Nat! Acad Sci USA 1986;83:6751-5.