The artificial womb

1 Ann. Acad. Sci. ISSN 0077-8923 ANNALS OF THE NEW YORK ACADEMY OF SCIENCESI ssue:Reproductive ScienceThe artificial wombCarlo Bulletti,1 Antonio Palagiano,2 Caterina Pace,2 Angelica Cerni,3 Andrea Borini,4and Dominique de Ziegler51 Physiopathology of Reproduction Unit, Cervesi s General Hospital in Cattolica, Cattolica, of Naples, Naples, of Bologna, Bologna, Procreazione, Forl` , Baudeloque, Groupe HopitalierCochien, Paris, FranceAddress for correspondence: Carlo Bulletti, MD, Fisiopatologia della Riproduzione, Ospedale Cervesi di Cattolica (Rimini), ViaBeethoven, 2-47841, Cattolica, Italy. Email: availability of computer-controlled artificial hearts, kidneys, and lungs, as well as the possibility of implantinghuman embryos inex vivouterus models or an artificial endometrium, presents new perspectives for creating anartificial uterus.

2 Survival rates have also improved, with fetuses surviving from as early as 24 weeks of gestation. Theseadvances bring new opportunities for complete or partial ectogenesis through the creation of an artificial womb , onethat could sustain the growth and development of fetuses outside of the human : womb ; artificial uterus; ectogenesis; embryo implantation; fetal surgery; assisted reproductive technologyIntroductionThe mastery ofin vitrofertilization and improve-ments in the survival of premature infants haveopened new opportunities forectogenesis, the im-plantation and full development of fetusesin , ectogenesis denotes the availability of an arti-ficial uterus capable of fostering the development ofembryos to artificial uterus could assist women with dam-aged or diseased uteri by allowing them to conceiveand carry infants, despite an inability to do so ontheir own.

3 A second possibility is that an artificialuterus could serve as an incubator for preterm ba-bies, specifically those who are delivered before ap-proximately 24 weeks of gestation the minimumfor viability with current incubators. The develop-ment of such an incubator could provide a break-through for reducing fetal mortality and morbiditythat stems from prematurity. Artificial wombs, as they are currently con-ceived, would function by connecting to an extra-corporeal supply of maternal blood or replacementfluids. The artificial womb would supply nutrientsand oxygen to an incubated fetus and would be ca-pable of disposing of waste materials. This would,therefore, necessitate an artificial placenta for medi-ating the necessary exchanges between fetal circula-tion and the system that would replace the first attempts to support the implantation ofhuman embryos outside of the human body wereconducted in 1982 in Bologna, Italy, and continuedat Mount Sinai Hospital in New York City in 1983.

4 The first results were published in pro-gram ofex vivohuman uteri maintained throughextracorporeal perfusion led to different sets of ex-periments targeted at clarifying uterine physiologyand 6 The report of the first hu-man embryo implantation in anex vivo, isolated,extracorporeally perfused uterus occurred in 1989,stirring comments and bringing ethical concernsto attention, including those voiced by the editorof the journalFertility and ,the experimental program conducted in Italy wasstopped because of the ethical issues raised and thestrong and vociferous opposition from the polit-ical studies were conductedusing theex vivomodel, however, not for study-ing embryo implantation. Rather, these studies in-vestigated various pharmacological effects on uter-ine physiology. These included, for example, a setdoi: Acad.

5 Sci. 1221 (2011) 124 128c 2011 New York Academy of artificial endometrium obtainedin vitrofromepithelial cells placed on a bed of studies documenting the existence of a directvagina-to-uterus transport called the first uterinepass effect. This work details some important char-acteristics of the uterus, including the unexpectedlyhigh levels of vaginally delivered hormones such ,6 The technology ofex vivoperfusion in humanuteri was helpful in learning to imitate mechanismsgoverning human uterine receptivity and embryoimplantation. Today, however, more advanced tech-nologies are available and make possible the notionthat an artificial uterus could support fetal devel-opment, at least for part of the pregnancy. Onceperfected, however, an artificial womb would al-low for the possibility to continue or initiate fetaldevelopment outside of the mother s artificial endometriumHung-Ching Liu, a researcher at Cornell Univer-sity in New York, prepared a co-culture system thatcombined epithelial and stromal cells a model re-ported as being the first step toward an , these experiments were notextended beyond the 6-day limit for embryo re-search enacted in the United States.

6 As illustrated inFigure 1, several studies have employed artificial en-dometrium models and reported their responses tosteroid hormones in various 12In theseco-culture systems, epithelial and stromal cells werelayered on matrigel supports. This permits epithe-lial cells to exhibit spontaneous orientation andpromising viability, allowing the development ofnew models for studying maternal embryo inter-actions. Relatedly, in Japan in 1996, Kuwabara9re-ported attempts at preserving a developing goat forthree weeks in an incubator that reproduced theuterus and placenta, together with amniotic fluidand blood supply. Similar attempts at developing anartificial womb took place artificial placentaPlacenta preservation through extracorporeal per-fusion has been attempted in the past to study hor-mone metabolism and the respective fetal and ma-ternal contributions to various example, it has been demonstrated thatthe human placenta can still supply nutrients anddispose of waste products when connected to an ar-tificial 19 However several attempts at uter-ine transplantation have taken place, but all ,47 Placentation, or the passage of nu-trients via the placenta, under these conditions re-mains an unresolved issue, in particular because ofimmunosuppressive therapies.

7 It is possible that theplacenta may offer immune protection to the fe-tus by passing antibodies such as immunoglobulin(IgG).17 Following the premature removal of the fetusfrom the maternal uterus, the three umbilical cordvessels could remain open by inhibiting their phys-iological occlusion, for example, through heparinflushing, placement of a stent, or by creating anarterial bypass that sustains exchanges between ma-ternal and fetal 23 Areservoirofmater-nal blood could therefore support placental phys-iology, provide nutrients for the fetus, and assurethe removal of waste. Yet another option, wherethe artificial uterus and placenta would be directlyconnected to the maternal circulation, has ,25 Notably, the use of artificial suppliers and dis-posers has the advantage of allowing the fetus todevelop in a separate environment where it is ex-cluded from the possible deleterious effects of ma-ternal disease, or from exposure to pollutants, al-cohol, or ,27 29 Furthermore, independentsystems could not bear the risk of an immune reac-tion in the maternal system against the fetus becauseof insufficient gestational immune , waste disposal could also beachieved by dialysis.

8 Oxygenation of the embryo orfetus has been achieved with extracorporeal mem-brane oxygenation (ECMO), a validated techniquethat has successfully allowed the development of agoat fetus for up to 237 hours in an amniotic Acad. Sci. 1221 (2011) 124 128c 2011 New York Academy of techniques are problematic, however, forproviding proper nutrition. Total parenteral nu-trition, as studied in infants suffering from severeshort bowel syndrome, offers a 5-year survival ofapproximately 20%.30 32 Another potential application of ectogenesis isthe possibility of xenopregnancy. Interspecific preg-nancies (sometimes called interspecies pregnanciesor xenopregnancy) involve an embryo or fetus anda uterine carrier belonging to a different speaking, letting a fetus developin a carrier of a different species is the equivalentof xenografts compared to allografts.

9 Such an ap-proach would put a higher demand on the gesta-tional immune tolerance system to avoid ,interspecificges-tation would encounter issues linked to the differenttypes of placentation that characterize each example, the most invasive hemochorial placen-tation that characterizes humans implies a profounddown-regulation of the maternal immune , the endotheliochorial type of placen-tation characteristic of cats and dogs, and thosecharacteristic of pigs, ruminants, and whales, doesnot allow for true contact between maternal bloodand the fetal chorion. Hence, interspecific pregnan-cies could bear the risk of generating inappropriateinteractions between the fetal trophoblast and theendometrium of the the case of an embryo from a Bactrian cameltransferred to the uterus of a Dromedary, pregnan-cies can be carried to term with no other interven-tion beyond the embryo ability ofone species to survive inside the uterus of anotheris often unidirectional pregnancies are not neces-sarily successful in the inverse situation.

10 For exam-ple, horse embryos survive in the donkey uterus, butdonkey embryos perish in the uterus of an mouse embryos survive in the uterusof the white-footed mouse, but reciprocal transfersinvariably fail. Ethical concerns are not consideredhere but would certainly arise if the development ofhuman embryos were to be attempted in the uteriof other incubatorAlthough no technology currently exists for sup-porting the development of embryos from concep-tion to viability, the importance of developing suchtechnologies is now recognized. For example, newtechnologies have allowed our group to propose thecomputer-controlled sequence of two artificial kid-neys: one provides the balanced fluid content ofsynthetic amniotic sacs, while the other permits theproper perfusion of the umbilical vessel with filteredblood coming from a reservoir of maternal artificial lung and heart and a temperature-controlled cabinet complete the 6To re-place placental function, an artificial placenta,or other filtering system capable of supportingnutrition, would be 2005, the French biologist Henri Atlan predictedin a book,L Ut erus Artificiel, that within 100 years,artificial reproductive technology would master thein vitrodevelopment of the human fetus.