J. Anim. Sci. 2002. 80:2671-2681
© 2002 American Society of Animal Science
Factors affecting cellular outgrowth from porcine inner cell masses in vitro1
K. R. Schilperoort-Haun and
A. R. Menino, Jr.2
Department of Animal Sciences, Oregon State University, Corvallis 97331-6702
2 Correspondence:
phone: 541-737-3011; fax: 541-737-4174; E-mail:
alfred.r.menino{at}orst.edu.
 |
Abstract
|
|---|
During early embryonic development, endodermal cells leave the inner cell mass (ICM) and migrate over an extracellular matrix located on the blastocoelic side of the trophectoderm to form extraembryonic endoderm. Two experiments were conducted to evaluate factors supporting porcine endodermal cell migration in vitro. In Exp. 1, porcine ICM were cultured on matrices of collagen IV, fibronectin, or laminin. Percentages of ICM generating cellular outgrowth on fibronectin (5/11; 45%) and laminin (4/10; 40%) were similar (P > 0.10); however, collagen IV (0/10; 0%) failed (P < 0.05) to support cellular outgrowth. Inner cell mass and outgrowth areas and numbers of cells in outgrowths were similar (P > 0.10) for fibronectin and laminin, and increased (P < 0.05) with time in culture. In Exp. 2, ICM were cultured on fibronectin or laminin in medium containing 0 or 500 µg/mL of the inhibitory tripeptide, arg-gly-asp (RGD), or on laminin in medium containing 0 or 10 µg/mL recombinant human tissue inhibitor of matrix metalloproteinases-2 (rhTIMP-2). Inner cell mass and outgrowth areas and numbers of cells in the outgrowths for ICM cultured on fibronectin did not differ (P > 0.10) due to the presence of RGD. Inner cell masses cultured on laminin in medium containing 500 µg/mL RGD had fewer cells in the outgrowths and slower rates of cell migration compared with 0 µg/mL (P < 0.05). No differences (P > 0.10) in ICM and outgrowth areas and numbers of cells in the outgrowths were observed for ICM cultured on laminin in medium containing 0 or 10 µg/mL rhTIMP-2. Both fibronectin and laminin supported porcine ICM outgrowth in vitro; however, because outgrowth on fibronectin was not inhibited by RGD, endodermal cells must express an integrin that recognizes an alternative sequence in fibronectin. Cell migration on laminin was inhibited by RGD, suggesting either RGD competes with laminin for binding sites on endodermal cells or binding RGD alters endodermal cell migration on laminin. Because rhTIMP-2 had no effect on cell outgrowth, porcine ICM do not appear to be responsive to the proliferative effects of rhTIMP-2.
Key Words: Embryos • Germ Line • Matrix • Proteins • Metalloproteinases • Pigs
|
Introduction
|
|---|
The early swine blastocyst is comprised of a single layer of trophectoderm surrounding a fluid-filled blastocoel and inner cell mass (ICM). Endodermal cells appear on the inner surface of the ICM and migrate on the blastocoelic side of the trophectoderm to form extraembryonic endoderm. The extracellular matrix (ECM) glycoprotein, fibronectin, is present on the blastocoelic surface of the trophectoderm and is proposed to support cell movement (Richoux et al., 1989). Laminin is also an ECM component and influences cell attachment and movement (Mecham, 1991). Cell-ECM interactions are mediated primarily by cell-surface receptors called integrins (Simon et al., 1996; Bowen and Hunt, 2000). A key structural element in fibronectin is the tripeptide sequence, arg-gly-asp (RGD), which functions as an integrin recognition site. From studies using RGD as a cell adhesion inhibitor, several integrins have been identified that recognize RGD (Hynes, 1992).
Extracellular proteinases, like the matrix metalloproteinase (MMP) family, also participate in cell migration through degradation of the ECM (Nagase and Woessner, 1999). Tissue inhibitors of MMP (TIMP) regulate MMP proteolysis and limit the amount of ECM degradation (Birkedal-Hansen et al., 1993). In addition to their properties as proteolytic inhibitors, TIMP can also provide growth-promoting effects (Satoh et al., 1994). The gelatinases, MMP-2 and -9, have been detected in porcine embryos during endodermal and mesodermal cell proliferation, but it is not known if these enzymes participate in cell migration events in the embryo (Chamberlin and Menino, 1995).
To elucidate the mechanism of porcine endodermal cell migration from the ICM, two experiments were conducted. In Experiment 1, cellular outgrowth from porcine ICM was evaluated on matrices of collagen IV, fibronectin, and laminin. In Experiment 2, the effects of RGD and TIMP-2 on ICM outgrowth on matrices of fibronectin and laminin were investigated.
|
Materials and Methods
|
|---|
Embryo Collection and Culture
Fourteen pubertal crossbred Yorkshire x Hampshire gilts were observed once daily for estrus and handmated with one of two boars at 0 and 24 h after detected estrus. Embryos were recovered at slaughter (n = 2) or surgically (n = 12) 5 to 6 d after estrus (d 0 = onset of estrus). Reproductive tracts recovered at slaughter were transported to the laboratory where embryos were collected by flushing the uteri with the alpha modification of Eagle’s Minimum Essential Medium (MEM; Sigma Chemical Co., St. Louis, MO). Surgical recovery of embryos was performed by induction of anesthesia in gilts by injection of 5 mL of a 1:1 mixture of xylazine (Rompum, Lloyd Laboratories, Shenandoah, IA) and ketamine HCl (Ketaset, Fort Dodge Laboratories, Fort Dodge IA) into the ear vein, and anesthesia was maintained during surgery via inhalation of oxygen and halothane (Fluothane; Fort Dodge Laboratories). The reproductive tract was exteriorized via midventral laparotomy and uteri flushed with 
MEM supplemented with 25 mM HEPES (Sigma). Morulae to hatched blastocysts were recovered from the uterine flushings by aspiration and transferred to screw-cap tissue culture tubes containing 10 mL HEPES-buffered MEM and 1.5% BSA at 39°C and transported to the laboratory. Embryos were evaluated for morphology using an inverted-stage phase-contrast microscope (100 to 200x). Embryos were washed three times and cultured overnight in microdrops containing MEM with 1.5% BSA (MEM + BSA) under paraffin oil (Fisher Scientific Co., Tustin, CA) in a humidified atmosphere of 5% CO2 in air at 39°C.
Inner Cell Mass Isolation
For embryos which failed to complete the hatching process during the overnight culture, zonae pellucidae were removed by brief exposure to acidified PBS (pH 2.0; Barlett and Menino, 1995). Inner cell masses were isolated from blastocysts using immunosurgical procedures as described by Solter and Knowles (1975) and Bartlett and Menino (1995). Briefly, embryos were incubated in microdrops of rabbit anti-porcine serum (Sigma; diluted 1 to 4 in MEM) for 1 h, washed in three changes of MEM + BSA, incubated in guinea pig complement (ICN Immunobiologicals, Lisle, IL; diluted 1 to 4 in MEM + BSA) for 1 h, washed three times, and placed in microdrops of 25 mM HEPES-buffered MEM + BSA for isolation of ICM. All incubations were conducted at 39°C in a humidified atmosphere of 5% CO2 in air, and embryo manipulations were conducted at room temperature. Inner cell masses were isolated by repeated aspiration through a finely drawn siliconized glass capillary pipette until the lysed trophectodermal cells were clearly removed from the ICM.
Experiment 1. Effects of ECM-Type on ICM Outgrowth
Inner cell masses were cultured on 60 x 15 mm plastic tissue culture dishes (Becton Dickinson and Company, Lincoln Park, NJ) in 25-µL microdrops coated with type IV collagen, fibronectin, or laminin. Culture dishes were prepared by aliquoting 25 µL of 10 µg/mL solutions of type IV collagen, fibronectin, or laminin prepared in sterile PBS with 1% antibiotic (Sigma) onto the surface of the dish, covering the drops with paraffin oil and incubating overnight. Solutions were removed the following morning, and each drop was rinsed three times with MEM + BSA and overlaid with MEM + BSA. Inner cell masses were cultured for 96 h in a humidified atmosphere of 5% CO2 in air at 39°C. At 24-h intervals, ICM were observed for attachment and cellular outgrowth with an inverted stage phase contrast microscope and photographed. Numbers of cells migrating away from the ICM core were counted, and length and width measurements of the ICM and respective outgrowth were determined with an ocular micrometer. Also at 24-h intervals, 15 µL of conditioned medium were recovered from each microdrop, replaced with fresh medium, and stored at -20°C until assayed for gelatinase activity. Inner cell mass and outgrowth areas were determined by tracing the photomicrograph with a compensating polar planimeter (model L-20-M; LASICO, Los Angeles, CA) and by computing the actual microscopic area.
Experiment 2. Effects of RGD on ICM Outgrowth on Fibronectin and Laminin and rhTIMP-2 on ICM Outgrowth on Laminin
Development of porcine ICM was evaluated on matrices of fibronectin or laminin in medium containing RGD (Sigma) or recombinant human TIMP-2 (rhTIMP-2; provided by Keith Langley of Amgen, Inc., Thousand Oaks, CA). Inner cell masses were harvested and processed as described above; however, preparation of the culture medium included the addition of 500 µg/mL RGD or 10 µg/mL rhTIMP-2. At 24-h intervals, ICM and the respective outgrowths were measured and photographed, and numbers of cells in the outgrowths were counted as described.
SDS-PAGE and Zymography
All reagents for SDS-PAGE and zymography were obtained from Sigma, unless otherwise stated. Electrophoresis was performed at 4°C in 10% SDS-polyacrylamide gels copolymerized with 1% gelatin under nonreducing conditions as described by Brenner et al. (1989). Conditioned medium was treated with double strength sample buffer containing 0.125 M Tris-HCl (pH 6.8), 20% glycerol, 5% SDS, and 0.025% Bromophenol Blue and loaded into one-dimensional slab gels. Low-range molecular mass markers (97.4 to 14.4 kDa; BioRad, Richmond, CA) were used as standards. After electrophoresis, gels were incubated in 2.5% Triton X-100 for 1 h and transferred to an incubation bath containing 50 mM Tris-HCl, 5 mM CaCl2, and 0.15 M NaCl (pH 8.4) and shaken at room temperature for 36 to 48 h. Gels were fixed and stained with 1 g/L amido black (BioRad) in 10:30:60 acetic acid:methanol:water for 1 h and destained in 10:30:60 acetic acid:methanol:water. The appearance of clear lytic zones or bands against a dark staining background was used to indicate the presence of gelatinase activity. In our laboratory, gelatin zymography enables detection of 1 ng of a collagenase (EC 3.4.24.3) standard (Schilperoort-Haun and Menino, 2002).
Statistical Analysis
Differences in the percentages of ICM undergoing attachment or outgrowth due to treatment were evaluated by chi-square procedures. Data for morphometric variables and cell numbers in the outgrowth were analyzed by repeated measures ANOVA. Sources of variation in the ANOVA for Exp. 1 and 2 were matrix (collagen IV, fibronectin and laminin), time and the matrix x time interaction and treatment (0 and 500 µg/mL RGD or 0 and 10 µg/mL rhTIMP-2), time and the treatment x time interaction, respectively. F-Ratios for matrix and treatment were computed using the within-subjects error as per the repeated measures model, whereas F-ratios for time and the interactions were computed using the residual error. If significant effects were observed in the ANOVA, differences between means were determined using Duncan’s multiple-comparison test. Relationships between cell proliferation and changes in outgrowth areas and time were evaluated by correlation-regression analysis. All analyses were performed using the Number Cruncher Statistical System software program (NCSS version 2000; NCSS, Kaysville, UT).
|
Results
|
|---|
Experiment 1. Effects of ECM-Type on ICM Outgrowth
A total of 73 embryos with normal morphology (22 morulae, 6 blastocysts, 24 expanded blastocysts, and 21 hatched blastocysts) were recovered from six gilts, and 58 ICM were successfully isolated. Inner cell masses recovered from embryos collected as or cultured to hatched blastocysts from each gilt were randomly distributed across all matrices. Incidences of attachment by ICM isolated from hatched blastocysts did not differ (P > 0.10) on collagen IV (3/10; 30%), fibronectin (8/11; 73%), and laminin (6/10; 60%). Incidences of cellular outgrowth were similar (P > 0.10) for fibronectin (5/11; 45%) and laminin (4/10; 40%); however, collagen IV (0/10; 0%) failed to support cellular outgrowth (P < 0.05). A representative sequence of cell outgrowth on laminin is presented in Figure 1
. Because of the poor development on collagen IV, ICM recovered from embryos cultured to blastocysts or expanded blastocysts from each gilt were randomly distributed on only fibronectin and laminin. Inner cell masses recovered from this group of embryos attached to fibronectin and laminin at a low frequency (1/13; 8% and 2/14; 14%, respectively) and failed to generate cellular outgrowths (0/13 and 0/14, respectively). All subsequent analyses for Experiments 1 and 2 were conducted on ICM recovered from hatched blastocysts.
View larger version (99K):
[in this window]
[in a new window]
|
Figure 1. Porcine inner cell masses at 0 h (A) and accompanying outgrowths on laminin at 24 (B), 48 (C), 72 (D), and 96 h (E) of culture. Bar = 50 µm.
|
|
Because no outgrowth was supported by collagen IV, only ICM areas were available for morphometric comparisons with fibronectin and laminin. No difference (P > 0.10) was observed in ICM and outgrowth areas due to matrix-type, and the matrix x time interaction was not significant (Table 1). Inner cell mass and outgrowth areas increased over time where areas at 0 and 24 h were less (P < 0.05) than at 96 h of culture (Table 1; Figure 2A). Regression analysis of ICM and outgrowth areas over time revealed that the rates of increase in area on fibronectin (563.8 µm2/h) and laminin (409.0 µm2/h) were similar (P > 0.10), and both were greater (P < 0.05) than collagen IV (-11.0 µm2/h).
View this table:
[in this window]
[in a new window]
|
Table 1. Inner cell masses (ICM) and outgrowth areas (µm2 x 103) and numbers of cells in the outgrowths for porcine ICM cultured on matrices of collagen IV, fibronectin, or laminin
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Figure 2. Changes in inner cell masses (ICM) and outgrowth areas (A) and numbers of cells in outgrowths (B) for porcine ICM cultured on collagen IV, fibronectin, or laminin.
|
|
Although the mean number of cells in outgrowths on laminin was more than twofold the number of cells on fibronectin, there was no difference (P > 0.10) due to matrix-type (Table 1
). The matrix x time interaction was also not significant. Numbers of cells in the outgrowths increased (P < 0.05) over time in culture, whereby cell numbers at 24 h were less (P < 0.05) than at 72 and 96 h (Table 1; Figure 2B). Despite the lack of difference between fibronectin and laminin treatments in numbers of cells in outgrowths, the rate of cellular migration on laminin (0.75 cell/h) was greater (P < 0.05) than on fibronectin (0.19 cell/h). Gelatinase activity was not detected in conditioned medium recovered over the 96-h culture period.
Experiment 2. Effects of RGD on ICM Outgrowth on Fibronectin and Laminin and rhTIMP-2 on ICM Outgrowth on Laminin
A total of 80 embryos with normal morphology (2 blastocysts, 15 expanded blastocysts and 63 hatched blastocysts) were recovered from eight gilts, and 67 ICM were successfully isolated. Inner cell masses recovered from hatched blastocysts from each gilt were randomly distributed within a matrix-type and paired treatment conditions. Addition of RGD to the culture medium did not reduce (P > 0.10) the incidences of ICM attachment or outgrowth on fibronectin or laminin (Table 2). No differences (P > 0.10) in ICM and outgrowth areas on fibronectin were observed due to RGD, and the treatment x time interaction was not significant (Table 3). ICM and outgrowth areas increased (P < 0.05) over time, whereby the area at 96 h was greater (P < 0.05) than at 0 through 72 h of culture (Table 3; Figure 3A). Rates of change in ICM and outgrowth areas on fibronectin were not different (P > 0.10) in cultures with 0 or 500 µg/mL RGD (9011.8 and 6717.2 µm2/h, respectively). Numbers of cells migrating away from the ICM on fibronectin also did not differ (P > 0.10) due to the addition of RGD, and the treatment x time interaction was not significant (Table 3). Numbers of cells in the outgrowths increased (P < 0.05) over time in culture where cell numbers at 96 h were greater (P < 0.05) than at 24 h of culture (Table 3; Figure 3B). Rates of cellular migration did not differ (P > 0.10) in medium containing 0 µg/mL RGD (0.55 cell/h) compared with 500 µg/mL RGD (0.33 cell/h).
View this table:
[in this window]
[in a new window]
|
Table 2. Incidences of attachment and cellular outgrowth for porcine inner cell masses (ICM) cultured for 96 h on fibronectin or laminin in medium containing 0 or 500 µg/mL arg-gly-asp (RGD)
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Inner cell masses (ICM) and outgrowth areas (µm2 x 103) and numbers of cells in the outgrowths for porcine ICM cultured on fibronectin or laminin in medium containing 0 or 500 µg/mL arg-gly-asp (RGD)
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Figure 3. Changes in inner cell masses (ICM) and outgrowth areas (A) and numbers of cells in outgrowths (B) for porcine ICM cultured on fibronectin in 0 (-) or 500 (+) µg/mL arg-gly-asp (RGD).
|
|
No differences (P > 0.10) in ICM and outgrowth areas on laminin were observed due to addition of RGD, and the treatment x time interaction was not significant (Table 3). Inner cell masses and outgrowth areas increased over time where areas at 0 h were less (P < 0.05) than at 48 through 96 h, and areas at 24 and 48 h were less (P < 0.05) than at 72 and 96 h of culture (Table 3; Figure 4A). Rates at which the cellular outgrowth areas enlarged were similar (P > 0.05) for ICM cultured in medium containing 0 (934.7 µm2/h) or 500 µg/mL RGD (656.3 µm2/h). Numbers of cells in outgrowths on laminin were less (P = 0.052) in cultures with 500 µg/mL compared with 0 µg/mL RGD (Table 3). Numbers of cells in outgrowths increased (P < 0.05) over time where cell numbers were lower (P < 0.05) at 24 than 96 h of culture (Table 3). The treatment x time interaction was also significant (P = 0.02). Further analysis revealed less (P < 0.05) cells in outgrowths at 24 h compared with 96 h of culture for ICM developing on laminin in 0 µg/mL RGD (Figure 4B). No differences (P > 0.10) were observed in cell numbers due to time in culture for ICM in 500 µg/mL RGD (Figure 4B). Furthermore, numbers of cells in outgrowths in 0 µg/mL RGD were greater (P < 0.05) than 500 µg/mL RGD at 72 and 96 h of culture (Figure 4B). Regression analysis revealed a decreased (P < 0.05) rate of migration by cells cultured with 500 µg/mL RGD (-0.07 cell/h) compared with 0 µg/mL RGD (0.52 cell/h). A representative culture of porcine ICM on laminin in 0 and 500 µg/mL RGD is depicted in Figures 5A and B, respectively. Inner cell masses cultured on laminin with 500 µg/mL RGD exhibited irregular cellular spacing and larger cells in the outgrowths.
View larger version (18K):
[in this window]
[in a new window]
|
Figure 4. Changes in inner cell masses (ICM) and outgrowth areas (A) and numbers of cells in outgrowths (B) for porcine ICM cultured on laminin in 0 (-) or 500 (+) µg/mL arg-gly-asp (RGD). For panel B, means within a treatment without common letters differ (P < 0.05) and means within an hour with an asterisk differ (P < 0.05) from + RGD.
|
|
View larger version (96K):
[in this window]
[in a new window]
|
Figure 5. Porcine inner cell masses cultured on laminin in 0 (A) and 500 µg/mL (B) arg-gly-asp at 72 h of culture. Bar = 50 µm.
|
|
No differences (P > 0.05) in the incidences of attachment (9/11; 82% vs 5/11; 45%) or cellular outgrowth (8/11; 73% vs 4/11; 36%) on laminin were observed in 0 vs 10 µg/mL rhTIMP-2, respectively. Inner cell mass and outgrowth areas did not differ (P > 0.10) due to the presence of rhTIMP-2, and the treatment x time interaction was not significant (Table 4). Inner cell mass and outgrowth areas increased (P < 0.05) over time in culture where areas at 0 through 48 h were less (P < 0.05) than at 72 and 96 h of culture (Table 4; Figure 6A). Rates of increase in ICM and outgrowth areas were similar (P > 0.10) for ICM cultured in 0 (3752.1 µm2/h) and 10 µg/mL rhTIMP-2 (9317.4 µm2/h). Results observed for cell numbers in the outgrowths were consistent with morphometric observations. No differences (P > 0.10) in numbers of cells in the outgrowths were observed due to rhTIMP-2, and the treatment x time interaction was not significant (Table 4). Numbers of cells in the outgrowths increased (P < 0.05) over time where cell numbers were less (P < 0.05) at 24 h vs 72 and 96 h of culture (Table 4; Figure 6B). Cell migration rates were similar (P > 0.05) in 0 (0.62 cell/h) compared with 10 µg/mL rhTIMP-2 (0.78 cell/h).
View this table:
[in this window]
[in a new window]
|
Table 4. Inner cell masses (ICM) and outgrowth areas (µm2 x 103) and numbers of cells in the outgrowths for porcine ICM cultured on laminin in medium containing 0 or 10 µg/mL rhTIMP-2
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Figure 6. Changes in inner cell masses (ICM) and outgrowth areas (A) and numbers of cells in outgrowths (B) for porcine ICM cultured on laminin in 0 (-) or 10 (+) µg/mL rhTIMP-2.
|
|
|
Discussion
|
|---|
In the present study, fibronectin and laminin supported outward cellular migration of endodermal cells from porcine ICM. Fibronectin is present on the blastocoelic surface of the trophectoderm in rodent and swine embryos and is proposed to support endodermal cell migration during genesis of the extraembryonic endoderm (Richoux et al., 1989; Carnegie, 1991). Although laminin supported cellular outgrowth to a similar extent as fibronectin, laminin has been proposed to stabilize cell-ECM interactions and, for some cells, even serve as a "stop" signal for migration (Coopman et al., 1991). In accord with this notion of cell-ECM stabilization, porcine endodermal cells secrete laminin following migration (Richoux et al., 1989), and rodent parietal endoderm cells secrete laminin during formation of Reichert’s membrane, the thin basement membrane located between the trophectoderm and parietal endoderm (Smith and Strickland, 1981; Fowler et al., 1990). In an earlier report from our laboratory (Bartlett and Menino, 1995), sheep ICM failed to generate outgrowth on laminin, whereas fibronectin and collagen IV were supportive. Cellular behavior on laminin is apparently complex because in addition to porcine ICM, mouse trophoblastic outgrowth, rat endodermal cell migration and, to a limited extent, bovine inner cell mass outgrowth, are supported in vitro by laminin (Armant et al., 1986; Carnegie and Cabaca, 1991; Schilperoort-Haun and Menino, 2002). Koshikawa et al. (2000) reported a dual function for laminin depending on whether the cells were in quiescent or regenerative tissues. Laminin would anchor and stabilize cells in quiescent tissues, whereas in regenerative processes or wound-healing laminin would stimulate cell movement. It is possible that under the culture conditions provided, endodermal cells behave as cells in regenerative tissues and migrate on laminin, whereas in vivo the conditions promote quiescent behavior on laminin.
Cellular outgrowth in the presence of RGD should be inhibited if endodermal cells utilize an integrin that recognizes the RGD sequence (Hynes, 1992). The present findings suggest that porcine endodermal cells migrate on fibronectin without requiring an RGD-recognizing integrin. The 5ß1 integrin is the principal cellular receptor for fibronectin and recognizes the RGD sequence. Because no decrease in attachment and outgrowth was observed with RGD, it is likely that endodermal cells can bind fibronectin through another integrin which recognizes an alternative sequence in fibronectin. The 4ß1 integrin binds fibronectin at the EILDV site, and both 4 and ß1 subunits have been detected in d 11 to 15 pig embryos (Bowen et al., 1996).
Porcine ICM cultured on laminin in the presence of RGD had fewer cells in outgrowths and slower rates of cell migration. These observations suggest either RGD competes with laminin for laminin-binding integrins on endodermal cells or RGD binding modifies endodermal cell behavior on laminin. The former explanation is less likely because the RGD sequence in laminin does not appear to be a recognition site for laminin-binding integrins (Kimber and Spanswick, 2000). Mouse trophoblast attachment and outgrowth on fibronectin were inhibited by hexapeptides containing the RGD sequence, but no inhibition was observed on laminin by RGD peptides (Armant et al., 1986). Furthermore, plates coated with E1-4, an elastase-digested fragment of laminin containing the only RGD sequence in the intact molecule, did not support mouse trophoblastic outgrowth (Armant, 1991). Porcine ICM cultured with RGD on laminin initially attached and generated cellular outgrowths similar to the control. However, after 24 h in RGD, cell numbers in the outgrowths stabilized despite, albeit slower than the control, a continued increase in outgrowth area. These observations suggest two deviations in cell migration on laminin occurred after exposure to RGD. First, cells stopped leaving the ICM and, second, those cells which had initially grown out of the ICM ceased proliferating yet continued to spread. This behavior may in effect reproduce in vivo events occurring in extraembyronic endoderm formation. Porcine endodermal cells may require exposure to the ECM in a specific sequence to replicate the in vivo process of migration and stabilization. In a crude manner, addition of RGD replicates this sequence of events in vivo, where endodermal cells initially interact with fibronectin, and subsequently develop fixed cellular interactions when laminin appears. This mechanism is supported by the results of the present study where progressive outgrowth on laminin occurs unless these cells are exposed to the RGD-sequence and the RGD-binding integrins are activated. Whatever physiologic conditions impact the cell to induce a particular response, the actual behavior on laminin has been partly attributed to the expression of specific integrins resulting from the cellular stimulus, e.g., laminin binding to the 6ß4 integrin induces static adhesion (Koshikawa et al., 2000). In this manner, fibronectin exposure may induce expression of 6ß4 in endodermal cells, thereby causing stabilization when laminin is encountered.
Porcine preimplantation embryos produce the gelatinases, MMP-2 and -9, during the period of endodermal cell migration (Chamberlin and Menino, 1995). Matrix metalloproteinases are involved in cellular migration and function to promote a limited and controlled amount of ECM degradation. However, gelatinase activity was not detected during the period of culture in the present study, suggesting a trophoblastic source for the gelatinases observed by Chamberlin and Menino (1995). Gelatinase inhibitors, e.g., TIMP, are secreted concomitantly with many proforms of MMP to provide a balance for MMP proteolysis and to participate in regulated ECM degradation during cellular migration (Matrisian, 1992; Birkedal-Hansen et al., 1993). Growth factor effects have also been reported for TIMP-1 and -2 in numerous cell types (Hayakawa et al., 1994). Satoh et al. (1994) observed enhanced in vitro bovine embryo development with TIMP-1 produced by granulosa and oviductal cells and concluded the stimulatory effect was directly attributed to greater cell proliferation and/or stabilization of the ECM via inhibition of MMP activity. In a system similar to the one used in this study for porcine ICM, Behrendtsen and Werb (1997) observed greater endodermal cell differentiation and outgrowth by mouse ICM cultured with TIMP-1. Outgrowth areas, numbers of cells in the outgrowths and plasminogen activator production were also stimulated by rhTIMP-2 in bovine ICM cultured on fibronectin (Schilperoort-Haun and Menino, 2002). Addition of rhTIMP-2 to the culture medium in this study did not result in any significant changes in porcine endodermal cell proliferation or migration.
The results of this study suggest that the ECM proteins, fibronectin and laminin, support migration of endodermal cells from isolated porcine ICM. An RGD-recognizing integrin is not necessary for endodermal cell migration on fibronectin, and it is likely that porcine endodermal cells express the 4ß1 integrin. Cell migration is altered on laminin when ICM are cultured with RGD, and this may be due to RGD exposure altering cellular receptors for laminin. Porcine ICM development in vitro was not stimulated by rhTIMP-2. To further elucidate the mechanisms supporting porcine endodermal cell migration and proliferation, subsequent studies should be directed at evaluating interactions between the ECM and their specific cellular receptors.
|
Implications
|
|---|
An early event in development of the fetal membranes in swine is the formation of extraembryonic endoderm. Extraembryonic endoderm results from the differentiation, proliferation, and migration of a population of endodermal cells from the embryo’s inner cell mass. Little is known about the factors regulating this process in the early swine embryo. Our research has identified extracellular matrices that support cellular outgrowth from the swine inner cell mass. The cell migration response on a specific matrix can also be modified by altering components of the culture conditions. This work has provided basic information regarding key processes in swine embryos that may provide insights in identifying aberrant developmental mechanisms leading to embryonic mortality.
|
Footnotes
|
|---|
1 The authors wish to thank Keith Langley of Amgen, Inc. (Thousand Oaks, CA) for his generous contribution of rhTIMP-2 and Nora Ross for her assistance in preparation of the manuscript. Technical Paper No. 11396; Oregon Agricultural Experiment Station. 
Received for publication November 19, 2001.
Accepted for publication May 3, 2002.
|
Literature Cited
|
|---|
Armant, D. R. 1991. Cell interactions with laminin and its proteolytic fragments during outgrowth of mouse primary trophoblast cells. Biol. Reprod. 45:664–672.[Abstract]
Armant, D. R., H. Kaplan, and W. Lennarz. 1986. Fibronectin and laminin promote in vitro attachment and outgrowth of mouse blastocyst. Dev. Biol. 116:519–523.[Medline]
Bartlett, S. E., and A. R. Menino, Jr. 1995. Evaluation of extracellular matrices and the plasminogen activator system in sheep inner cell mass and trophectodermal outgrowth in vitro. Biol. Reprod. 52:1436–1445.[Abstract]
Behrendtsen, O., and Z. Werb. 1997. Metalloproteinases regulate parietal endoderm differentiating and migrating in cultured mouse embryos. Dev. Dyn. 208:255–265.[Medline]
Birkedal-Hansen, H., W. G. I. Moore, M. K. Boden, L. J. Windsor, B. Birkedal-Hansen, A. DeCarlo, and J. A. Engler. 1993. Matrix metalloproteinases: a review. Crit. Rev. Oral Biol. Med. 4:197–250.[Abstract/Free Full Text]
Bowen, J. A., F. W. Bazer, and R. C. Burghardt. 1996. Spatial and temporal analyses of integrin and Muc-1 expression in porcine uterine epithelium and trophectoderm in vitro. Biol. Reprod. 55:1098–1106.[Abstract]
Bowen, J. A., and J. S. Hunt. 2000. The role of integrins in reproduction. Proc. Soc. Exp. Biol. Med. 223:331–343.[Abstract/Free Full Text]
Brenner, C., R. Adler, D. Rappolee, R. Pedersen, and Z. Werb. 1989. Genes for extracellular matrix-degrading metalloproteinases and their inhibitor, TIMP, are expressed during early mammalian development. Genes Dev. 3:848–859.[Abstract/Free Full Text]
Carnegie, J. A. 1991. Immunolocalization of fibronectin and laminin within rat blastocysts cultured under serum-free conditions. J. Reprod. Fertil. 91:423–434.[Abstract]
Carnegie, J. A., and O. Cabaca. 1991. The influence of extracellular matrix components on the proliferation and migration of inner cell mass-derived parietal endoderm cells. Biol. Reprod. 45:572–580.[Abstract]
Chamberlin, R., and A. R. Menino, Jr. 1995. Partial characterization of gelatinases produced by preimplantation porcine embryos. Biol. Reprod. 52 (Suppl.):179.[Abstract]
Coopman, P. J., M. E. Bracke, J. C. Lissitzky, G. K. DeBryune, F. M. VanRoy, J. M. Foidart, and M. M. Mareel. 1991. Influence of basement membrane molecules on directional migration of human breast cell lines in vitro. J. Cell Sci. 98:395–401.[Abstract/Free Full Text]
Fowler, K. J., K. Mitrangas, and M. Dziadek. 1990. In vitro production of Reichert’s membrane by mouse embryo-derived parietal endoderm cell lines. Exp. Cell Res. 191:194–203.[Medline]
Hayakawa, T., K. Yamashita, E. Ohuchi, and A. Shinagawa. 1994. Cell growth-promoting activity of tissue inhibitor of metalloproteinases-2 (TIMP-2). J. Cell Sci. 107:2373–2379.[Abstract]
Hynes, R. O. 1992. Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69:11–25.[Medline]
Kimber, S. J., and C. Spanswick. 2000. Blastocyst implantation: the adhesion cascade. Cell Dev. Biol. 11:77–92.
Koshikawa, N., G. Giannelli, V. Cirulli, K. Miyazaki, and V. Quaranta. 2000. Role of cell surface metalloprotease MT1-MMP in epithelial cell migration over laminin-5. J. Cell Biol. 148:615–624.[Abstract/Free Full Text]
Matrisian, L. 1992. The matrix-degrading metalloproteinases. BioEssays 14:455–463.[Medline]
Mecham, R. P. 1991. Laminin Receptors. Annu. Rev. Cell Biol. 7:71–91.[Medline]
Nagase, H., and J. F. Woessner. 1999. Matrix metalloproteinases. J. Biol. Chem. 274:21491–21494.[Free Full Text]
Richoux, V., T. Darribére, J. C. Boucaut, J. E. Fléchon, and J. P. Thiery. 1989. Distribution of fibronectins and laminin in the early pig embryo. Anat. Rec. 233:72–81.
Satoh, T., K. Kobayashi, M. Kikuchi, Y. Senda, and H. Hoshi. 1994. Tissue inhibitor of matrix metalloproteinases (TIMP-1) produced by granulosa and oviduct cells enhances in vitro development of bovine embryos. Biol. Reprod. 50:835–844.[Abstract]
Schilperoort-Haun, K. R., and A. R. Menino, Jr. 2002. Evaluation of extracellular matrix proteins and tissue inhibitor of matrix metalloproteinases-2 on bovine inner cell mass outgrowth in vitro. In Vitro Cell. Dev. Biol. Anim. 38:41–47.[Medline]
Simon, C., M. J. Gimeno, A. Mercader, A. Frances, J. G. Velasco, J. Remohi, M. L. Polan, and A. Pellicer. 1996. Cytokines-adhesion molecules-invasive proteinases. The missing paracrine/autocrine link to embryo implantation? Mol. Hum. Reprod. 2:405–424.[Abstract/Free Full Text]
Smith, K. K., and S. Strickland. 1981. Structural components and characteristics of Reichert’s Membrane, an extra-embryonic basement membrane. J. Biol. Chem. 256:4654–4661.[Abstract/Free Full Text]
Solter, D., and B. B. Knowles. 1975. Immunosurgery of mouse blastocyst. Proc. Natl. Acad. Sci. USA 72:5099–5102.[Abstract/Free Full Text]
Top
Abstract
Introduction
