- Email: [email protected]
The placenta as a cell source in fetal tissue engineering
The Placenta as a Cell Source in Fetal Tissue Engineering By Amir Kaviani, Tjo¨rvi E. Perry, Carmen M. Barnes, Jung-Tak Oh, Moritz M. Ziegler, Steven J. Fishman, and Dario O. Fauza Boston, Massachusetts
Purpose: This study was aimed at determining whether fetal tissue constructs can be engineered from cells derived from the placenta. Methods: A subpopulation of morphologically distinct cells was isolated mechanically from specimens of human placenta (n ⫽ 6) and selectively expanded. The lineage of these cells was determined by immunofluorescent staining against multiple intermediate filaments and surface antigens. Cell proliferation rates were determined by oxidation assays and compared with those of immunocytochemically identical cells derived from human amniotic fluid samples (n ⫽ 6). Statistical analysis was by analysis of variance (ANOVA). After expansion, the cells were seeded onto a polyglycolic acid polymer/poly-4-hydroxybutyrate scaffold. The resulting construct was analyzed by both optical and scanning electron microscopy.
significantly different when compared with mesenchymal fetal cells isolated from human amniotic fluid; however, it was greater than previously reported rates for similar cells obtained from postnatal or adult tissues. Construct analysis showed dense layers of cells firmly attached to the scaffold without evidence of cell death. Conclusions: Subpopulations of nontrophoblastic, mesenchymal cells can be isolated consistently from the human placenta. These cells proliferate as rapidly as fetal mesenchymal amniocytes in vitro and attach firmly to polyglycolic acid scaffolds. The placenta can be a valuable and practical source of cells for the engineering of select fetal tissue constructs. J Pediatr Surg 37:995-999. Copyright 2002, Elsevier Science (USA). All rights reserved.
Results: The immunocytochemical profile of expanded placental cells was consistent with a nontrophoblastic, mesenchymal origin. Their proliferation rate in culture was not
INDEX WORDS: Fetal surgery, tissue engineering, placenta, chorionic villus sampling, congenital anomalies, birth defects, fetus, prenatal, neonate, transplantation.
S
experiments, that mesenchymal fetal cells can be isolated consistently from amniotic fluid, expanded in sufficient numbers in vitro, and seeded successfully onto biodegradable scaffolds for tissue engineering applications.11, 12 In the current study, we sought to determine whether comparable fetal cells could also be obtained from human placental samples, expanded, and processed in vitro for tissue engineering applications. Specifically, we sought to (1) characterize these cells, (2) compare their proliferation rates with that of similar fetal cells obtained from human amniotic fluid; and (3) analyze their attachment and proliferation profiles after seeding
URGICAL TREATMENT of many congenital anomalies is often hindered by scarce tissue availability, particularly at birth. In such instances, artificial prostheses or, occasionally, autologous grafting are the only options for reconstruction. Follow-up of these patients, however, has shown high morbidity rates.1-5 Complications include infection, patch disruption, and need for reoperation, commonly resulting in poor functional outcome.3-5 Fetal tissue engineering recently has emerged as a promising concept in surgical reconstruction of severe birth defects.6-8 Utilizing fetal cells obtained from a tissue biopsy performed in utero, an autologous bioprosthesis can be engineered in parallel to the remainder of gestation and become readily available for implantation in the neonatal period, if necessary. Until now, this concept has been used successfully in animal experiments for treatment of bladder, skin, diaphragm, and chest wall defects.6,7 Although these preliminary reports are encouraging, the harvest of fetal tissue as a cell source poses significant risks to both the mother and the fetus. Potential complications include induction of premature labor and direct injury to the fetus, which have been reported after both open and minimally invasive fetal tissue harvest.8-10 Recently we have shown, both in animal and human
Journal of Pediatric Surgery, Vol 37, No 7 (July), 2002: pp 995-999
From the Departments of Surgery, Children’s Hospital and Harvard Medical School, and Harvard Center for Minimally Invasive Surgery, Boston, MA. Presented at the 53rd Annual Meeting of the Section on Surgery of the American Academy of Pediatrics, San Francisco, California, October 19-21, 2001. Supported by grants from the United States Surgical Corporation and The Children’s Hospital Surgical Foundation. Address reprint requests to Dario O. Fauza, MD, Children’s Hospital, 300 Longwood Ave, Fegan 3, Boston, MA 02115. Copyright 2002, Elsevier Science (USA). All rights reserved. 0022-3468/02/3707-0011$35.00/0 doi:10.1053/jpsu.2002.33828
995
996
KAVIANI ET AL
unto an implantable, biodegradable scaffold maintained in a bioreactor. MATERIALS AND METHODS This study was approved and overseen by the institutional review boards of Harvard Medical School and The Brigham and Women’s Hospital, Boston, MA (protocol # 2000-P-002168/2-BWH).
Cell Harvest and Culture Specimens of human placenta (n ⫽ 6) were obtained after cesarean section performed at 33 to 35 weeks of gestation. The maternal decidua was separated away carefully from the mesenchymal core and discarded. The remaining mesenchymal core, which is composed only of fetal cells, was minced and digested with a collagenase/dispase mixture (10% Type II collagenase [Worthington Biochemical, Grand Rapids, MI], 4.0 U Dispase II [Boerhinger, Mannheim, Germany], 2.5 mmol/L CaCl2) for 30 minutes, in a 5% CO2 incubator at 37°C. After trituration and passage through a 100-m mesh (Fischer Scientific, Pittsburgh, PA), cells were centrifuged at 1800 rpm for 15 minutes, resuspended in 10 mL of serum-free Dulbecco’s modified Eagle Medium (GIBCO, Grand Island, NY), and counted. Samples of amniotic fluid (8 to 22 ml) were obtained after routine ultrasound-guided amniocentesis performed on pregnant women (n ⫽ 6) with a gestational age ranging from 16 to 21 weeks, and centrifuged at 1800 rpm for 15 minutes. The pellets were removed and resuspended in 8 mL of serum-free Dulbecco’s Modified Eagle Medium (Gibco, Grand Island, NY). All samples of both placental and amniotic fluid cells then were processed in the following manner. Cells were plated at an initial density of 3 million per 150 cm2 in culture dishes (Becton Dickinson, Franklin Lakes, NJ) covered with 5-mm2 slide cover slips (Becton Dickinson). They were fed daily with Dulbecco’s Modified Eagle Medium supplemented with 20% fetal bovine serum (Sigma Chemical, St Louis, MO), 5 ng/mL fibroblast growth factor (Promega, Madison, WI), glutamine, penicillin and fungizone (all the latter from Gibco), in a 95% humidified, 5% CO2 incubator at 37°C for 48 hours, after which they were inspected for cell attachment. Cover slips containing a subpopulation of morphologically distinct cells were isolated, removed, and placed in separate 30-cm2 wells (Becton Dickinson) containing the same culture medium described above. After achieving confluence, cell passages were at a 1:4 ratio.
Cell Identification After 3 to 5 passages, cells were plated on LabTek tissue culture chamber slides (Fischer Scientific, Pittsburgh, PA). They were fixed with 4% neutral buffered formaldehyde (Sigma Chemical) for 10 minutes, rinsed in Dulbecco’s Phosphate Buffered Saline (PBS; Sigma Chemical), and fixed with 100% methanol (Fischer Scientific) for 10 minutes. They were rinsed with a “blocking solution” consisting of 2% horse serum (Sigma Chemical) and 0.5% Triton (Sigma Chemical) in PBS and incubated in this solution for 30 minutes. Cells then were incubated with mouse monoclonal ascites primary antibodies, including: desmin (Sigma Chemical); vimentin (Sigma Chemical); smooth muscle actin (SMA; Dako, Carpenteria, CA); cluster of differentiation 31 (CD31; Sigma Chemical); von Willebrand’s factor (vWF; Dako); cytokeratins 7, 8, 18 (Dako); fibroblast surface protein (FSP; Sigma Chemical); and calponin (Sigma Chemical). Primary antibody dilutions were as follows: desmin (1:400); vimentin (1:500); SMA (1:500); CD31 (1:1000); vWF (1:000); cytokeratins 7, 8, and 18 (1:500); FSP (1:1500); and calponin (1:500). Primary antibody incubation lasted 45
minutes, after which the slides were rinsed 3 times with blocking solution and incubated in a goat antimouse FITC secondary antibody (Sigma Chemical) for 30 minutes. The slides then were rinsed 3 times in blocking solution, mounted in DAPI fluorescence medium (Vector Labs, Burlingame, CA), and viewed with a Zeiss fluorescence microscope (Carl Zeiss, Jena, Germany).
Proliferation Assays After 5 passages, cells derived from both sources (placenta and amniotic fluid) with identical immunocytochemical staining properties were placed in separate wells containing 5 mL of the same culture medium described above, at 20,000 cells per milliliter. An oxidationreduction assay was performed using AlamarBlue (BioSource International, Camarillo, CA), as previously described.13,14 AlamarBlue has exc ⫽ 530 ⫺ 560 nm and emits light at a peak of 590 nm. Alliquots (100 L) were removed and assayed from each culture plate at 2, 4, 8, 16, 24, 36, and 72 hours of cell culture. Utilizing a fluorescence microplate reader (Fmax Molecular Devices, Sunnyvale, CA) driven by SOFTmax PRO software (Molecular Devices, Sunnyvale, CA), the emitted fluorescence was tabulated and plotted. A sample consisting of 10% AlamarBlue solution unexposed to cells was used as a control.
Construct Assembly Scaffolds were composed of commercially available unwoven polyglycolic acid (PGA) at a density of 63 g/mL (Albany International, Mansfield, MA), pretreated with 10% poly-4-hydroxybutyrate (P4HB; TEPHA, Cambridge, MA) for additional tensile strength. Disks with a surface area of 10 mm2 and thickness of 1 mm were prepared from the PGA/P4HB composite. Confluent culture plates of mesenchymal placental cells were washed once with PBS and trypsinized with 0.25% Trypsin-EDTA (Sigma Chemical) for 5 minutes. Cells then were placed in a 50-mL disposable sterile centrifuge tube (Fisherbrand, Pittsburgh, PA) and centrifuged at 900 rpm for 10 minutes at room temperature. Next, cells were resuspended in cell culture medium. Thirty-six million cells were placed in a 30-mL disposable sterile centrifuge tube (Fisherbrand), to which a previously gas sterilized PGA-P4HB composite disk (described above) and 25 mL of cell culture medium were added. The disks were seeded dynamically in a 37°C rotating hybridization oven (Hyroller, Cambridge, MA) for 48 hours. Resulting constructs then were removed and prepared for analysis.
Scanning Electron Microscopy Constructs were fixed in 0.1% glutaraldehyde (Sigma Chemical) for 1 hour. Next, the specimens were immersed in a 0.1% formalin solution (Sigma Chemical) for 18 hours. After fixation was complete, constructs were dehydrated serially in a graded fashion with ethanol. Samples were allowed to dry overnight, after which they were mounted, sputtercoated with gold (Desk II, Denton Vacuum, Cherry Hill, NJ), and imaged with a scanning electron microscope (ISI-DS 130, Topcon, CA).
Statistical Analysis Statistical analysis was performed by analysis of variance for repeated measures (ANOVA) at 95% confidence limits. P values of less than .05 were considered significant.
RESULTS
Selected rapidly proliferating placental cells stained positively for SMA, desmin, CD-31, cytokeratins 8 and 18, and calponin and negatively for cytokeratin 7 and vimentin. This staining pattern is consistent with a non-
PLACENTAL-BASED FETAL TISSUE ENGINEERING
997
Fig 1. Immunocytochemical staining of fetal mesenchymal cells derived from human placenta for smooth muscle actin (green) and calponin (red). (A) Original magnification ⴛ30; (B) Original magnification ⴛ45.
trophoblastic, mesenchymal origin and is identical to the profile found on the selected amniocytes (Fig 1; Table 1). Their proliferation rate in culture was not significantly different when compared with that of mesenchymal fetal cells isolated from amniotic fluid (P ⫽ .49); however, it was greater than previously reported rates for cells obtained from postnatal or adult tissues (Fig 2).11 Scanning electron microscopy of mesenchymal placental constructs showed dense, confluent layers of cells surrounding the scaffold matrices and firm cell adhesion to PGA polymer, without evidence of cell death (Fig 3). DISCUSSION
With the exception of the decidual layers, the placenta is composed entirely of fetal tissue.15 Our data show that the placenta can serve as a source of nontrophoblastic, mesenchymal cells applicable in fetal tissue engineering.
Considering the mesenchymal nature of these cells, fetal constructs composed of them could be used in the surgical treatment of a variety of congenital anomalies. In the current study, specimens were harvested from third trimester placentas. Yet, routine diagnostic placental sampling usually is carried out much earlier in pregnancy. Chorionic villus sampling (CVS), for instance, is performed normally between 12 and 18 weeks of gestation. Whether fetal mesenchymal cells obtained from routine CVS may be processed as those currently studied is yet to be determined. However, previous reports point to a more prominent mesenchymal core in the placenta early in gestation, suggesting that CVS samples should yield even more mesenchymal cells than the specimens processed herein.15 A small additional sample of placen-
Table 1. Immunocytochemistry of Rapidly Proliferating Cells Isolated From Human Placenta and Amniocentesis Showing an Identical Profile From Both Sources, Consistent With a Mesenchymal Lineage
Surface antigen CD-31 Calponin Cytoskeletal marker Desmin Vimentin SMA Cytokeratin-7 Cytokeratin-8 Cytokeratin-18
Placental cells
Amniocytes
⫺ ⫹
⫺ ⫹
⫹ ⫺ ⫹ ⫺ ⫹ ⫹
⫹ ⫺ ⫹ ⫺ ⫹ ⫹
Fig 2. Results of AlamarBlue™ proliferation assays comparing mesenchymal cells derived from amniotic fluid and placenta. There were no statistically significant differences between them. However, these rates were significantly higher than those previously reported for comparable cells derived from tissues harvested postnatally.11
998
KAVIANI ET AL
Fig 3. Scanning electron micrograph of a PGA/P4-HB composite (A) before and (B) after dynamic seeding with mesenchymal cells derived from human placenta (original magnification ⴛ520).
tal tissue could easily be processed at the time of routine diagnostic placental sampling, providing cells available for a fetal tissue engineering application in the neonatal period or even later in life, if necessary. Further studies are necessary before clinical application of engineered fetal tissue derived from placental cells. In vivo animal experiments currently are underway to explore the long-term efficacy of such constructs in the treatment of certain congenital anomalies. Different bioprosthetic scaffolds, with varying cell attachment, mechanical, and degradation properties, must be investigated to determine the optimal engineered construct for each specific application. Multiple manipulations of the bioreactor environment, which influence the extent of differentiated function and proliferation of the seeded
cells, as well the final construct architecture, can be examined. Perhaps more importantly, the full differentiation potential of these mesenchymal cells is yet to be explored in its entirety. Finally, their pattern of histocompatibility antigen expression must be analyzed, aiming at potential heterologous tissue engineering applications. Although the abovementioned studies should be pursued before clinical application, the current data suggest that the placenta can be a valuable and practical source of cells for the engineering of select fetal tissue constructs. ACKNOWLEDGMENT The authors thank Jeffrey Pettit for his excellence in laboratory assistance.
REFERENCES 1. Towne BH, Peters G, Chang JH: The problem of “giant” omphalocele. J Pediatr Surg 15:543-548, 1980 2. Grady RW, Mitchell ME: Newborn exstrophy closure and epispadias repair. World J Urol 16:200-204, 1998 3. Pelizzo G, Dubois R, Laine X, et al: Surgical treatment of diaphragmatic agenesis by transposition of a muscle flap: Report on 15 cases. Eur J Pediatr Surg 10:8-11, 2000 4. Delarue A, Camboulives J, Bollini G, et al: Delayed cure of an omphalocele requiring abdominosternoplasty, right hepatectomy and partial splenectomy. Eur J Pediatr Surg 10:58-61, 2000 5. Moss RL, Chen CM, Harrison MR: Prosthetic patch durability in congenital diaphragmatic hernia: A long-term follow-up study. J Pediatr Surg 36:152-154, 2001 6. Fauza DO, Fishman SJ, Mehegan K, et al: Videofetoscopically assisted fetal tissue engineering: Bladder augmentation. J Pediatr Surg 33:7-12, 1998 7. Fauza DO, Fishman SJ, Mehegan K, et al: Videofetoscopically assisted fetal tissue engineering: Skin replacement. J Pediatr Surg 33:357-361, 1998 8. Fauza DO, Marler JJ, Koka R, et al: Fetal tissue engineering: Diaphragmatic replacement. J Pediatr Surg 36:146-151, 2001
9. Harrison MR: Fetal surgery. Am J Obstet Gynecol 174:12551264, 1996 10. Irwin BH, Vane DW: Complications of intrauterine intervention for treatment of fetal obstructive uropathy. Urology 55:774, 2000 11. Kaviani A, Perry TE, Dzakovic A, et al: The amniotic fluid as a source of cells for fetal tissue engineering. J Pediatr Surg 36:16621665, 2001 12. Kaviani A, Guleserian A, Perry TE, et al: The amniotic fluid as a source of cells in fetal tissue engineering: Initial human experience. Presented at the 87th Annual Meeting of the American College of Surgeons Surgical Forum, New Orleans, LA, October 7-12, 2001 13. Ahmed SA, Gogal RM, Walsh JE: A new rapid and simple non-radioactive assay to monitor and determine the proliferation of lymphocytes: an alternative to [3H]thymidine incorporation assay. J Immunol Methods 170:211-224, 1994 14. Nakayama GR: Assessment of the Alamar blue assay for cellular growth and viability in vitro. J Immunol Methods 204:205-208, 1997 15. Phillips OP, Velagaleti GV, Tharapel AT, et al: Discordant direct and culture results following chorionic villus sampling and the diagnosis of a third cell line in the fetus. Prenat Diagn 17:170-172, 1997
PLACENTAL-BASED FETAL TISSUE ENGINEERING
999
Discussion Dr Feltis: If you could imagine a scenario in which you had a prenatal diagnosis of a diaphragmatic hernia, harvesting some fetal cells and grow a diaphragm patch for another available time, at the time of birth, is that a reasonable scenario? A. Kaviani (response): Absolutely, these cells take little time to attach to the tissue culture plates but then once they start proliferating, they proliferate exponentially such that at a time-point approximately 3 or 4 weeks after they are harvested, one would have more than enough cells for a 10-cm2 construct for implanting. Dr LaQuaglia (New York, NY): What I was interested in is the placenta; did you look into whether there were
stem cells in your placental samples? For instance, mesenchymal stem cells from other tissues have been differentiated into fat, or muscle, et cetera; how rich a source or supply of stem cells could that be? A. Kaviani (response): That is something that we are actively investigating in our laboratory. It certainly makes sense telelogically that the placenta would be a rich source of stem cells, but the burden for proving that something is a stem cell is increasing. What we are working on is trying to differentiate these cells into more mature cell types and then studying the actual behavior of those cells, not only by immunohistochemical staining patterns but also by determining if they act like the cells they are supposed to differentiate into.
Purpose: This study was aimed at determining whether fetal tissue constructs can be engineered from cells derived from the placenta. Methods: A subpopulation of morphologically distinct cells was isolated mechanically from specimens of human placenta (n ⫽ 6) and selectively expanded. The lineage of these cells was determined by immunofluorescent staining against multiple intermediate filaments and surface antigens. Cell proliferation rates were determined by oxidation assays and compared with those of immunocytochemically identical cells derived from human amniotic fluid samples (n ⫽ 6). Statistical analysis was by analysis of variance (ANOVA). After expansion, the cells were seeded onto a polyglycolic acid polymer/poly-4-hydroxybutyrate scaffold. The resulting construct was analyzed by both optical and scanning electron microscopy.
significantly different when compared with mesenchymal fetal cells isolated from human amniotic fluid; however, it was greater than previously reported rates for similar cells obtained from postnatal or adult tissues. Construct analysis showed dense layers of cells firmly attached to the scaffold without evidence of cell death. Conclusions: Subpopulations of nontrophoblastic, mesenchymal cells can be isolated consistently from the human placenta. These cells proliferate as rapidly as fetal mesenchymal amniocytes in vitro and attach firmly to polyglycolic acid scaffolds. The placenta can be a valuable and practical source of cells for the engineering of select fetal tissue constructs. J Pediatr Surg 37:995-999. Copyright 2002, Elsevier Science (USA). All rights reserved.
Results: The immunocytochemical profile of expanded placental cells was consistent with a nontrophoblastic, mesenchymal origin. Their proliferation rate in culture was not
INDEX WORDS: Fetal surgery, tissue engineering, placenta, chorionic villus sampling, congenital anomalies, birth defects, fetus, prenatal, neonate, transplantation.
S
experiments, that mesenchymal fetal cells can be isolated consistently from amniotic fluid, expanded in sufficient numbers in vitro, and seeded successfully onto biodegradable scaffolds for tissue engineering applications.11, 12 In the current study, we sought to determine whether comparable fetal cells could also be obtained from human placental samples, expanded, and processed in vitro for tissue engineering applications. Specifically, we sought to (1) characterize these cells, (2) compare their proliferation rates with that of similar fetal cells obtained from human amniotic fluid; and (3) analyze their attachment and proliferation profiles after seeding
URGICAL TREATMENT of many congenital anomalies is often hindered by scarce tissue availability, particularly at birth. In such instances, artificial prostheses or, occasionally, autologous grafting are the only options for reconstruction. Follow-up of these patients, however, has shown high morbidity rates.1-5 Complications include infection, patch disruption, and need for reoperation, commonly resulting in poor functional outcome.3-5 Fetal tissue engineering recently has emerged as a promising concept in surgical reconstruction of severe birth defects.6-8 Utilizing fetal cells obtained from a tissue biopsy performed in utero, an autologous bioprosthesis can be engineered in parallel to the remainder of gestation and become readily available for implantation in the neonatal period, if necessary. Until now, this concept has been used successfully in animal experiments for treatment of bladder, skin, diaphragm, and chest wall defects.6,7 Although these preliminary reports are encouraging, the harvest of fetal tissue as a cell source poses significant risks to both the mother and the fetus. Potential complications include induction of premature labor and direct injury to the fetus, which have been reported after both open and minimally invasive fetal tissue harvest.8-10 Recently we have shown, both in animal and human
Journal of Pediatric Surgery, Vol 37, No 7 (July), 2002: pp 995-999
From the Departments of Surgery, Children’s Hospital and Harvard Medical School, and Harvard Center for Minimally Invasive Surgery, Boston, MA. Presented at the 53rd Annual Meeting of the Section on Surgery of the American Academy of Pediatrics, San Francisco, California, October 19-21, 2001. Supported by grants from the United States Surgical Corporation and The Children’s Hospital Surgical Foundation. Address reprint requests to Dario O. Fauza, MD, Children’s Hospital, 300 Longwood Ave, Fegan 3, Boston, MA 02115. Copyright 2002, Elsevier Science (USA). All rights reserved. 0022-3468/02/3707-0011$35.00/0 doi:10.1053/jpsu.2002.33828
995
996
KAVIANI ET AL
unto an implantable, biodegradable scaffold maintained in a bioreactor. MATERIALS AND METHODS This study was approved and overseen by the institutional review boards of Harvard Medical School and The Brigham and Women’s Hospital, Boston, MA (protocol # 2000-P-002168/2-BWH).
Cell Harvest and Culture Specimens of human placenta (n ⫽ 6) were obtained after cesarean section performed at 33 to 35 weeks of gestation. The maternal decidua was separated away carefully from the mesenchymal core and discarded. The remaining mesenchymal core, which is composed only of fetal cells, was minced and digested with a collagenase/dispase mixture (10% Type II collagenase [Worthington Biochemical, Grand Rapids, MI], 4.0 U Dispase II [Boerhinger, Mannheim, Germany], 2.5 mmol/L CaCl2) for 30 minutes, in a 5% CO2 incubator at 37°C. After trituration and passage through a 100-m mesh (Fischer Scientific, Pittsburgh, PA), cells were centrifuged at 1800 rpm for 15 minutes, resuspended in 10 mL of serum-free Dulbecco’s modified Eagle Medium (GIBCO, Grand Island, NY), and counted. Samples of amniotic fluid (8 to 22 ml) were obtained after routine ultrasound-guided amniocentesis performed on pregnant women (n ⫽ 6) with a gestational age ranging from 16 to 21 weeks, and centrifuged at 1800 rpm for 15 minutes. The pellets were removed and resuspended in 8 mL of serum-free Dulbecco’s Modified Eagle Medium (Gibco, Grand Island, NY). All samples of both placental and amniotic fluid cells then were processed in the following manner. Cells were plated at an initial density of 3 million per 150 cm2 in culture dishes (Becton Dickinson, Franklin Lakes, NJ) covered with 5-mm2 slide cover slips (Becton Dickinson). They were fed daily with Dulbecco’s Modified Eagle Medium supplemented with 20% fetal bovine serum (Sigma Chemical, St Louis, MO), 5 ng/mL fibroblast growth factor (Promega, Madison, WI), glutamine, penicillin and fungizone (all the latter from Gibco), in a 95% humidified, 5% CO2 incubator at 37°C for 48 hours, after which they were inspected for cell attachment. Cover slips containing a subpopulation of morphologically distinct cells were isolated, removed, and placed in separate 30-cm2 wells (Becton Dickinson) containing the same culture medium described above. After achieving confluence, cell passages were at a 1:4 ratio.
Cell Identification After 3 to 5 passages, cells were plated on LabTek tissue culture chamber slides (Fischer Scientific, Pittsburgh, PA). They were fixed with 4% neutral buffered formaldehyde (Sigma Chemical) for 10 minutes, rinsed in Dulbecco’s Phosphate Buffered Saline (PBS; Sigma Chemical), and fixed with 100% methanol (Fischer Scientific) for 10 minutes. They were rinsed with a “blocking solution” consisting of 2% horse serum (Sigma Chemical) and 0.5% Triton (Sigma Chemical) in PBS and incubated in this solution for 30 minutes. Cells then were incubated with mouse monoclonal ascites primary antibodies, including: desmin (Sigma Chemical); vimentin (Sigma Chemical); smooth muscle actin (SMA; Dako, Carpenteria, CA); cluster of differentiation 31 (CD31; Sigma Chemical); von Willebrand’s factor (vWF; Dako); cytokeratins 7, 8, 18 (Dako); fibroblast surface protein (FSP; Sigma Chemical); and calponin (Sigma Chemical). Primary antibody dilutions were as follows: desmin (1:400); vimentin (1:500); SMA (1:500); CD31 (1:1000); vWF (1:000); cytokeratins 7, 8, and 18 (1:500); FSP (1:1500); and calponin (1:500). Primary antibody incubation lasted 45
minutes, after which the slides were rinsed 3 times with blocking solution and incubated in a goat antimouse FITC secondary antibody (Sigma Chemical) for 30 minutes. The slides then were rinsed 3 times in blocking solution, mounted in DAPI fluorescence medium (Vector Labs, Burlingame, CA), and viewed with a Zeiss fluorescence microscope (Carl Zeiss, Jena, Germany).
Proliferation Assays After 5 passages, cells derived from both sources (placenta and amniotic fluid) with identical immunocytochemical staining properties were placed in separate wells containing 5 mL of the same culture medium described above, at 20,000 cells per milliliter. An oxidationreduction assay was performed using AlamarBlue (BioSource International, Camarillo, CA), as previously described.13,14 AlamarBlue has exc ⫽ 530 ⫺ 560 nm and emits light at a peak of 590 nm. Alliquots (100 L) were removed and assayed from each culture plate at 2, 4, 8, 16, 24, 36, and 72 hours of cell culture. Utilizing a fluorescence microplate reader (Fmax Molecular Devices, Sunnyvale, CA) driven by SOFTmax PRO software (Molecular Devices, Sunnyvale, CA), the emitted fluorescence was tabulated and plotted. A sample consisting of 10% AlamarBlue solution unexposed to cells was used as a control.
Construct Assembly Scaffolds were composed of commercially available unwoven polyglycolic acid (PGA) at a density of 63 g/mL (Albany International, Mansfield, MA), pretreated with 10% poly-4-hydroxybutyrate (P4HB; TEPHA, Cambridge, MA) for additional tensile strength. Disks with a surface area of 10 mm2 and thickness of 1 mm were prepared from the PGA/P4HB composite. Confluent culture plates of mesenchymal placental cells were washed once with PBS and trypsinized with 0.25% Trypsin-EDTA (Sigma Chemical) for 5 minutes. Cells then were placed in a 50-mL disposable sterile centrifuge tube (Fisherbrand, Pittsburgh, PA) and centrifuged at 900 rpm for 10 minutes at room temperature. Next, cells were resuspended in cell culture medium. Thirty-six million cells were placed in a 30-mL disposable sterile centrifuge tube (Fisherbrand), to which a previously gas sterilized PGA-P4HB composite disk (described above) and 25 mL of cell culture medium were added. The disks were seeded dynamically in a 37°C rotating hybridization oven (Hyroller, Cambridge, MA) for 48 hours. Resulting constructs then were removed and prepared for analysis.
Scanning Electron Microscopy Constructs were fixed in 0.1% glutaraldehyde (Sigma Chemical) for 1 hour. Next, the specimens were immersed in a 0.1% formalin solution (Sigma Chemical) for 18 hours. After fixation was complete, constructs were dehydrated serially in a graded fashion with ethanol. Samples were allowed to dry overnight, after which they were mounted, sputtercoated with gold (Desk II, Denton Vacuum, Cherry Hill, NJ), and imaged with a scanning electron microscope (ISI-DS 130, Topcon, CA).
Statistical Analysis Statistical analysis was performed by analysis of variance for repeated measures (ANOVA) at 95% confidence limits. P values of less than .05 were considered significant.
RESULTS
Selected rapidly proliferating placental cells stained positively for SMA, desmin, CD-31, cytokeratins 8 and 18, and calponin and negatively for cytokeratin 7 and vimentin. This staining pattern is consistent with a non-
PLACENTAL-BASED FETAL TISSUE ENGINEERING
997
Fig 1. Immunocytochemical staining of fetal mesenchymal cells derived from human placenta for smooth muscle actin (green) and calponin (red). (A) Original magnification ⴛ30; (B) Original magnification ⴛ45.
trophoblastic, mesenchymal origin and is identical to the profile found on the selected amniocytes (Fig 1; Table 1). Their proliferation rate in culture was not significantly different when compared with that of mesenchymal fetal cells isolated from amniotic fluid (P ⫽ .49); however, it was greater than previously reported rates for cells obtained from postnatal or adult tissues (Fig 2).11 Scanning electron microscopy of mesenchymal placental constructs showed dense, confluent layers of cells surrounding the scaffold matrices and firm cell adhesion to PGA polymer, without evidence of cell death (Fig 3). DISCUSSION
With the exception of the decidual layers, the placenta is composed entirely of fetal tissue.15 Our data show that the placenta can serve as a source of nontrophoblastic, mesenchymal cells applicable in fetal tissue engineering.
Considering the mesenchymal nature of these cells, fetal constructs composed of them could be used in the surgical treatment of a variety of congenital anomalies. In the current study, specimens were harvested from third trimester placentas. Yet, routine diagnostic placental sampling usually is carried out much earlier in pregnancy. Chorionic villus sampling (CVS), for instance, is performed normally between 12 and 18 weeks of gestation. Whether fetal mesenchymal cells obtained from routine CVS may be processed as those currently studied is yet to be determined. However, previous reports point to a more prominent mesenchymal core in the placenta early in gestation, suggesting that CVS samples should yield even more mesenchymal cells than the specimens processed herein.15 A small additional sample of placen-
Table 1. Immunocytochemistry of Rapidly Proliferating Cells Isolated From Human Placenta and Amniocentesis Showing an Identical Profile From Both Sources, Consistent With a Mesenchymal Lineage
Surface antigen CD-31 Calponin Cytoskeletal marker Desmin Vimentin SMA Cytokeratin-7 Cytokeratin-8 Cytokeratin-18
Placental cells
Amniocytes
⫺ ⫹
⫺ ⫹
⫹ ⫺ ⫹ ⫺ ⫹ ⫹
⫹ ⫺ ⫹ ⫺ ⫹ ⫹
Fig 2. Results of AlamarBlue™ proliferation assays comparing mesenchymal cells derived from amniotic fluid and placenta. There were no statistically significant differences between them. However, these rates were significantly higher than those previously reported for comparable cells derived from tissues harvested postnatally.11
998
KAVIANI ET AL
Fig 3. Scanning electron micrograph of a PGA/P4-HB composite (A) before and (B) after dynamic seeding with mesenchymal cells derived from human placenta (original magnification ⴛ520).
tal tissue could easily be processed at the time of routine diagnostic placental sampling, providing cells available for a fetal tissue engineering application in the neonatal period or even later in life, if necessary. Further studies are necessary before clinical application of engineered fetal tissue derived from placental cells. In vivo animal experiments currently are underway to explore the long-term efficacy of such constructs in the treatment of certain congenital anomalies. Different bioprosthetic scaffolds, with varying cell attachment, mechanical, and degradation properties, must be investigated to determine the optimal engineered construct for each specific application. Multiple manipulations of the bioreactor environment, which influence the extent of differentiated function and proliferation of the seeded
cells, as well the final construct architecture, can be examined. Perhaps more importantly, the full differentiation potential of these mesenchymal cells is yet to be explored in its entirety. Finally, their pattern of histocompatibility antigen expression must be analyzed, aiming at potential heterologous tissue engineering applications. Although the abovementioned studies should be pursued before clinical application, the current data suggest that the placenta can be a valuable and practical source of cells for the engineering of select fetal tissue constructs. ACKNOWLEDGMENT The authors thank Jeffrey Pettit for his excellence in laboratory assistance.
REFERENCES 1. Towne BH, Peters G, Chang JH: The problem of “giant” omphalocele. J Pediatr Surg 15:543-548, 1980 2. Grady RW, Mitchell ME: Newborn exstrophy closure and epispadias repair. World J Urol 16:200-204, 1998 3. Pelizzo G, Dubois R, Laine X, et al: Surgical treatment of diaphragmatic agenesis by transposition of a muscle flap: Report on 15 cases. Eur J Pediatr Surg 10:8-11, 2000 4. Delarue A, Camboulives J, Bollini G, et al: Delayed cure of an omphalocele requiring abdominosternoplasty, right hepatectomy and partial splenectomy. Eur J Pediatr Surg 10:58-61, 2000 5. Moss RL, Chen CM, Harrison MR: Prosthetic patch durability in congenital diaphragmatic hernia: A long-term follow-up study. J Pediatr Surg 36:152-154, 2001 6. Fauza DO, Fishman SJ, Mehegan K, et al: Videofetoscopically assisted fetal tissue engineering: Bladder augmentation. J Pediatr Surg 33:7-12, 1998 7. Fauza DO, Fishman SJ, Mehegan K, et al: Videofetoscopically assisted fetal tissue engineering: Skin replacement. J Pediatr Surg 33:357-361, 1998 8. Fauza DO, Marler JJ, Koka R, et al: Fetal tissue engineering: Diaphragmatic replacement. J Pediatr Surg 36:146-151, 2001
9. Harrison MR: Fetal surgery. Am J Obstet Gynecol 174:12551264, 1996 10. Irwin BH, Vane DW: Complications of intrauterine intervention for treatment of fetal obstructive uropathy. Urology 55:774, 2000 11. Kaviani A, Perry TE, Dzakovic A, et al: The amniotic fluid as a source of cells for fetal tissue engineering. J Pediatr Surg 36:16621665, 2001 12. Kaviani A, Guleserian A, Perry TE, et al: The amniotic fluid as a source of cells in fetal tissue engineering: Initial human experience. Presented at the 87th Annual Meeting of the American College of Surgeons Surgical Forum, New Orleans, LA, October 7-12, 2001 13. Ahmed SA, Gogal RM, Walsh JE: A new rapid and simple non-radioactive assay to monitor and determine the proliferation of lymphocytes: an alternative to [3H]thymidine incorporation assay. J Immunol Methods 170:211-224, 1994 14. Nakayama GR: Assessment of the Alamar blue assay for cellular growth and viability in vitro. J Immunol Methods 204:205-208, 1997 15. Phillips OP, Velagaleti GV, Tharapel AT, et al: Discordant direct and culture results following chorionic villus sampling and the diagnosis of a third cell line in the fetus. Prenat Diagn 17:170-172, 1997
PLACENTAL-BASED FETAL TISSUE ENGINEERING
999
Discussion Dr Feltis: If you could imagine a scenario in which you had a prenatal diagnosis of a diaphragmatic hernia, harvesting some fetal cells and grow a diaphragm patch for another available time, at the time of birth, is that a reasonable scenario? A. Kaviani (response): Absolutely, these cells take little time to attach to the tissue culture plates but then once they start proliferating, they proliferate exponentially such that at a time-point approximately 3 or 4 weeks after they are harvested, one would have more than enough cells for a 10-cm2 construct for implanting. Dr LaQuaglia (New York, NY): What I was interested in is the placenta; did you look into whether there were
stem cells in your placental samples? For instance, mesenchymal stem cells from other tissues have been differentiated into fat, or muscle, et cetera; how rich a source or supply of stem cells could that be? A. Kaviani (response): That is something that we are actively investigating in our laboratory. It certainly makes sense telelogically that the placenta would be a rich source of stem cells, but the burden for proving that something is a stem cell is increasing. What we are working on is trying to differentiate these cells into more mature cell types and then studying the actual behavior of those cells, not only by immunohistochemical staining patterns but also by determining if they act like the cells they are supposed to differentiate into.