Portal de Periódicos da CAPES

In utero transplantation of wild-type fetal liver cells rescues factor X-deficient mice from fatal neonatal bleeding diatheses

2003; Elsevier BV; Volume: 1; Issue: 1 Linguagem: Inglês

10.1046/j.1538-7836.2003.00030.x

1538-7933

Elliot D. Rosen, Ivo Cornelissen, Zhong Liang, Amy L. Zollman, Michelle Casad, Julie Roahrig, Mark A. Suckow, F.J. Castellino,

Neurogenetic and Muscular Disorders Research

SummaryFactor X (FX)-deficient embryos suffer partial embryonic lethality with approximately 30% of the embryos arresting at midgestation. The remaining animals survive to term but die perinatally mainly from abdominal or intracranial hemorrhage. We have rescued FX-deficient mice by transplanting fetal liver cells from FX+/+, Rosa26 fetuses into midgestation embryos derived from FX+/− heterozygous crosses. FX−/− embryos were born at the expected frequency and approximately 50% of the FX−/− neonates survived longer than 4 months. FX−/− embryos receiving saline injections that survived to term died perinatally similar to untreated FX-deficient mice. The plasma levels of FX in the rescued 16-week-old FX−/− mice were approximately 1–6% of wild-type levels. β-Galactosidase-staining cells derived from the donor Rosa26 fetal liver cells were detected in 47% of the livers of adult mice. In addition, donor-derived cells were also recovered in the bone marrow, spleen, lung, and occasionally in the brain and testis. These results suggest that in utero cell transplantation could be an effective therapeutic strategy to treat pathologies resulting from the deficiency of hepatic-expressed factors. Factor X (FX)-deficient embryos suffer partial embryonic lethality with approximately 30% of the embryos arresting at midgestation. The remaining animals survive to term but die perinatally mainly from abdominal or intracranial hemorrhage. We have rescued FX-deficient mice by transplanting fetal liver cells from FX+/+, Rosa26 fetuses into midgestation embryos derived from FX+/− heterozygous crosses. FX−/− embryos were born at the expected frequency and approximately 50% of the FX−/− neonates survived longer than 4 months. FX−/− embryos receiving saline injections that survived to term died perinatally similar to untreated FX-deficient mice. The plasma levels of FX in the rescued 16-week-old FX−/− mice were approximately 1–6% of wild-type levels. β-Galactosidase-staining cells derived from the donor Rosa26 fetal liver cells were detected in 47% of the livers of adult mice. In addition, donor-derived cells were also recovered in the bone marrow, spleen, lung, and occasionally in the brain and testis. These results suggest that in utero cell transplantation could be an effective therapeutic strategy to treat pathologies resulting from the deficiency of hepatic-expressed factors. Maintaining the hemostatic balance between coagulation and anticoagulant factors is necessary to sustain the proper flow of blood through the vasculature. Insufficient expression of coagulation factors, such as factor (F)II, FV, FVII, FVIII, FIX, FX, and FXI result in hemorrhagic disorders, while decreased expression of anticoagulants protein C, protein S, and antithrombin III can lead to increased risk of thrombosis. As many hemostatic factors are primarily synthesized by hepatocytes, it is useful to consider low-level expression of these factors as a liver pathology. Thus, therapeutic strategies to treat hemophilia A or B include increasing hepatic expression of FVIII or FIX, respectively [1High K.A. Gene transfer as an approach to treating hemophilia.Circ Res. 2001; 88: 137-44Crossref PubMed Scopus (44) Google Scholar]. Moreover, bleeding diatheses resulting from FVII deficiencies have been treated recently by orthotopic liver transplantation [2Levi D. Pefkarou A. Fort J.A. DeFaria W. Tzaki A.G. Liver transplantation for factor VII deficiency.Transplantation. 2001; 72: 1836-7Crossref PubMed Scopus (15) Google Scholar]. Given the complexity and invasiveness of this procedure, the paucity of donor organs, and the need to match donors and recipients, organ transplantation is an unlikely therapeutic strategy for hemostatic disorders. However, since the liver is one of the few organs that regenerates after injury, transplantation of hepatocyte progenitors into a regenerating liver has been demonstrated in murine and rat models [3Moscioni A.D. Rozga J. Chen S. Naim A. Scott H.S. Demetriou A.A. Long-term correction of albumin levels in the Nagase analbuminemic rat: repopulation of the liver by transplanted normal hepatocytes under a regeneration response.Cell Transplant. 1996; 5: 499-503Crossref PubMed Scopus (51) Google Scholar, 4Markus P.M. Krause P. Fayyazi A. Honnicke K. Becker H. Allogeneic hepatocyte transplantation using FK 506.Cell Transplant. 1997; 6: 77-83Crossref PubMed Scopus (9) Google Scholar, 5Gupta S. Rajvanshi P. Aragona E. Lee C.D. Yerneni P.R. Burk R.D. Transplanted hepatocytes proliferate differently after CCl4 treatment and hepatocyte growth factor infusion.Am J Physiol. 1999; 276: G629-38PubMed Google Scholar, 6Kobayashi N. Miyazaki M. Fukaya K. et al.Transplantation of highly differentiated immortalized human hepatocytes to treat acute liver failure.Transplantation. 2000; 69: 202-7Crossref PubMed Scopus (74) Google Scholar, 7Sakamoto T. Ezure T. Lunz J. et al.Concanavalin A simultaneously primes liver hematopoietic and epithelial progenitor cells for parallel expansion during liver regeneration after partial hepatectomy in mice.Hepatology. 2000; 32: 256-67Crossref PubMed Scopus (28) Google Scholar, 8Sigal S.H. Rajvanshi P. Gorla G.R. et al.Partial hepatectomy-induced polyploidy attenuates hepatocyte replication and activates cell aging events.Am J Physiol. 1999; 276: G1260-72PubMed Google Scholar, 9Sokhi R.P. Rajvanshi P. Gupta S. Transplanted reporter cells help in defining onset of hepatocyte proliferation during the life of F344 rats.Am J Physiol Gastrointest Liver Physiol. 2000; 279: G631-40Crossref PubMed Google Scholar, 10Susick R. Moss N. Kubota H. et al.Hepatic progenitors and strategies for liver cell therapies.Ann N Y Acad Sci. 2001; 944: 398-419Crossref PubMed Scopus (57) Google Scholar, 11Braun K.M. Sandgren E.P. Cellular origin of regenerating parenchyma in a mouse model of severe hepatic injury.Am J Pathol. 2000; 157: 561-9Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 12Alison M.R. Poulsom R. Forbes S.J. Update on hepatic stem cells.Liver. 2001; 21: 367-73Crossref PubMed Scopus (82) Google Scholar, 13Fujino M. Li X.K. Kitazawa Y. et al.Selective repopulation of mice liver after Fas-resistant hepatocyte transplantation.Cell Transplant. 2001; 10: 353-61Crossref PubMed Scopus (15) Google Scholar, 14Allain J.E. Dagher I. Mahieu-Caputo D. et al.Immortalization of a primate bipotent epithelial liver stem cell.Proc Natl Acad Sci USA. 2002; 99: 3639-44Crossref PubMed Scopus (64) Google Scholar]. Following hepatectomy or chemical-induced liver damage, donor hepatocytes delivered via the portal vein or injected in the spleen colonize the liver of the recipient. Thus it is conceivable to repopulate the liver of individuals with coagulation deficiencies with donor hepatocytes expressing the appropriate factors. Cellular transplantation into immunocompetent animals still presents the immunological complications associated with organ transplantation. To overcome this problem, it is possible to transplant cells in utero into embryos before the development of immune competence [15Andreoletti M. LePercq J. Loux N. et al.In utero allotransplantation of retrovirally transduced fetal hepatocytes in primates: feasibility and short-term follow-up.J Matern Fetal Med. 1998; 7: 296-303Crossref PubMed Google Scholar]. Cells introduced before immunologic determination of ‘self’ will be incorporated into the developing fetus and should not elicit an immunologic response after the maturation of the immune system. Donor hematopoietic cells transplanted in utero into midgestation murine fetuses successfully colonize the bone marrow and spleen. This approach has been used to correct hematopoietic disorders in rodents and goats, and lysosomal storage disease in mice [16Mackenzie T.C. Flake A.W. Human mesenchymal stem cells persist, demonstrate site-specific multipotential differentiation, and are present in sites of wound healing and tissue regeneration after transplantation into fetal sheep.Blood Cells Mol Dis. 2001; 27: 601-4Crossref PubMed Scopus (135) Google Scholar, 17Hayashi S. Flake A.W. In utero hematopoietic stem cell therapy.Yonsei Med J. 2001; 42: 615-29Crossref PubMed Scopus (11) Google Scholar, 18Zanjani E.D. Pallavicini M.G. Ascensao J.L. et al.Engraftment and long-term expression of human fetal hemopoietic stem cells in sheep following transplantation in utero.J Clin Invest. 1992; 89: 1178-88Crossref PubMed Scopus (229) Google Scholar, 19Zanjani E.D. Flake A.W. Rice H. Hedrick M. Tavassoli M. Long-term repopulating ability of xenogeneic transplanted human fetal liver hematopoietic stem cells in sheep.J Clin Invest. 1994; 93: 1051-5Crossref PubMed Scopus (149) Google Scholar, 20Barker J.E. Deveau S. Lessard M. Hamblen N. Vogler C. Levy B. In utero fetal liver cell transplantation without toxic irradiation alleviates lysosomal storage in mice with mucopolysaccharidosis type VII.Blood Cells Mol Dis. 2001; 27: 861-73Crossref PubMed Scopus (20) Google Scholar, 21Casal M.L. Wolfe J.H. In utero transplantation of fetal liver cells in the mucopolysaccharidosis type VII mouse results in low-level chimerism, but overexpression of beta-glucuronidase can delay onset of clinical signs.Blood. 2001; 97: 1625-34Crossref PubMed Scopus (38) Google Scholar]. Similar methods have been applied clinically in humans to treat heritable hematopoietic insufficiency [22Flake A.W. Zanjani E.D. In utero hematopoietic stem cell transplantation. A status report.JAMA. 1997; 278: 932-7Crossref PubMed Google Scholar, 23Flake A.W. Zanjani E.D. Cellular therapy.Obstet Gynecol Clin North Am. 1997; 24: 159-77Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar, 24Shaaban A.F. Flake A.W. Fetal hematopoietic stem cell transplantation.Semin Perinatol. 1999; 23: 515-23Crossref PubMed Scopus (14) Google Scholar]. While the donor cells successfully colonize in the host, the procedure is complicated by graft vs. host disease since the donor hematopoietic cells derive from immunocompetent donors [25Flake A.W. In utero stem cell transplantation for the treatment of genetic diseases.Schweiz Med Wochenschr. 1999; 129: 1733-9PubMed Google Scholar]. Previously, we generated FX-deficient mice [26Dewerchin M. Liang Z. Moons L. et al.Blood coagulation factor X deficiency causes partial embryonic lethality and fatal neonatal bleeding in mice.Thromb Haemost. 2000; 83: 185-90Crossref PubMed Scopus (138) Google Scholar] by homologous recombination in embryonic stem cells. The FX-deficient embryos suffered partial embryonic lethality with approximately 30% of the embryos arresting at midgestation. The remaining animals survived to term but died perinatally, mainly from abdominal or intracranial hemorrhage. As the pathologies seen in these mice resemble symptoms observed in humans with severe FVII or FX coagulation deficiency, we explored the possibility of rescuing these animals by in utero hepatocyte transplantation. This rationale not only provided a therapeutic strategy for the treatment of hemophilia or other disorders resulting from insufficient expression of liver-derived proteins, but it also provided the opportunity to develop mice with conditional deficiencies. The availability of such animals would permit the study of the effects of the particular coagulation factor insufficiencies in experimental protocols that can only be performed on adult animals. Housing and procedures involving experimental animals were approved by the Institutional Animal Care and Use Committee of the University of Notre Dame, Notre Dame, Indiana. Embryonic age 12.5 (E12.5dpc) Rosa26 [27Friedrich G. Soriano P. Promoter traps in embryonic stem cells: a genetic screen to identify and mutate developmental genes in mice.Genes Dev. 1991; 5: 1513-23Crossref PubMed Scopus (1199) Google Scholar] embryos were used as liver cell donors. Pregnant Rosa26 donors were euthanized by cervical dislocation. Following abdominal sterilization with 70% ethanol, the peritoneal cavity was exposed with sterile instruments and the uterus excised and opened to expose the embryos within their yolk sac. The embryos were placed in a drop of sterile saline and the yolk sac and umbilicus removed to allow exsanguination. Each embryo was then placed into a fresh drop of sterile saline and the embryonic abdominal cavity was opened. The livers were removed and extraneous, non-hepatic tissue was dissected away. The livers were rinsed several times in sterile saline followed by trypsinization for 15 min at 37 °C in 5 mL of 0.05% solution. The hepatocytes were resuspended by gentle pipetting, and the trypsin quenched upon the addition of 10 mmol l−1 cold Dulbecco's modified Eagle's medium (DMEM)/F12 (Sigma; St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS) (Mediatech; Herndon, VA, USA). The cell suspension was then filtered through a 100-µm filter and the filter rinsed with an additional 25 mL of media. After centrifugation at 125 g for 10 min at 4 °C, the pelleted cells were resuspended in 15 mL of DMEM/F12 + 10% FBS, followed by a second filtration through a 40-µm filter and centrifugation at 125 g for 5 min at 4 °C. The cells were rinsed in 15 mL of cold phosphate-buffered saline (PBS), pelleted, resuspended in 1 mL of PBS, and the cell concentration determined using a hemocytometer. The cell concentration was adjusted to 105 cells per µL. Pregnant recipient mice were anesthetized using a Vetamac isoflurane vaporizer (Vetamac Inc.; Rossville, IN, USA), which delivered approximately 2% isoflurane in oxygen at a rate of 1 L min−1. The animals were placed in dorsal recumbancy and gently secured to a heating pad. The abdominal regions were shaved and sterilized with 70% ethanol, followed by betadine. A midline incision was made through the peritoneal walls and the uterine horns, containing the E12.5 embryos were gently externalized using cotton-tip applicators soaked in sterile saline. The uteri were kept constantly moist during the remainder of the procedure. Each embryo, visualized through the uterine wall, was punctured into the peritoneal cavity through the uterus with a 50-µm diameter-beveled glass pipette. A minimum of 1.5 × 105 fetal liver cells (FLCs) in a 1.5-µL volume was injected between the hind leg and the liver (Fig. 1). Once all the embryos were injected, the uterus was reinserted into the peritoneal cavity with the cotton tip applicators. The peritoneal wall was sutured with 5–0 vicryl and 1 mL of preheated saline injected into the peritoneal cavity. A 5–0 vicryl suture was then used to close the abdominal skin incision. The incision was swabbed with betadine solution and the animal given oxygen (2 L min−1) to enhance recovery. The pregnant mothers recovered from the surgery and delivered litters at the expected time. Bone marrow was flushed from the femurs of mice with 1 mL of normal saline and centrifuged at 220 g for 5 min. Cells were resuspended in 100 µL saline, counted, and 105 and 106 cells were pelleted using a Cytospin 3 centrifuge (Shandon; Pittsburgh, PA, USA) at 600 g for 8 min. Organs were fixed in 4% paraformaldehyde in PBS for 2 h, followed by three rinses with PBS and three 30-min rinses in X-gal rinse buffer (2 mmol L−1 MgCl2, 0.01% sodium deoxycholate, 0.02% NP-40 in PBS). Organs were then incubated in X-gal stain (5 mmol L−1 potassium ferrocyanide, 5 mmol L−1 potassium ferriccyanide, 1 mg mL−1 X-gal, 2 mmol L−1 MgCl2, 0.01% sodium deoxycholate, 0.02% NP-40 in PBS) for up to 48 h at 37 °C. After the development of stain, organs were rinsed three times for 30 min with PBS and stored in 70% ethanol. Mice were anesthetized with an intraperitoneal injection of rodent cocktail (0.015 mg xylazine/0.075 mg ketamine/0.0025 mg acepromazine/g weight of animal). A 1-cm section was cut from the tip of the tail with a scalpel and the tail was immediately immersed in 14 mL of saline at 37 °C in a clear 15 mL conical tube. Bleeding was followed visually and time to cessation was recorded. The amount of blood loss was also measured by determining the hemoglobin content of the blood collected in saline as described [28Sambrano G.R. Weiss E.J. Zheng Y.W. Huang W. Coughlin S.R. Role of thrombin signalling in platelets in haemostasis and thrombosis.Nature. 2001; 413: 74-8Crossref PubMed Scopus (463) Google Scholar]. Collection tubes were centrifuged and the erythrocyte pellet resuspended in 6 mL of red blood cell lysis buffer. The absorbency of the sample was measured at 575 nm. Test samples demonstrated the A575nm was proportional to the volume of collected blood. Attempts to colonize the fetal liver in order to generate adult animals with chimeric livers have employed mature hepatocytes, oval cells from regenerating livers, or late stage embryonic fetal liver cells transplanted into late stage fetuses [15Andreoletti M. LePercq J. Loux N. et al.In utero allotransplantation of retrovirally transduced fetal hepatocytes in primates: feasibility and short-term follow-up.J Matern Fetal Med. 1998; 7: 296-303Crossref PubMed Google Scholar, 29Allain J.E. Mahieu-Caputo D. Loux N. et al.Allotransplantation in utero and immortalization of primate fetal hepatocytes.J Soc Biol. 2001; 195: 57-63Crossref PubMed Scopus (2) Google Scholar, 30Andreoletti M. Pages J.C. Mahieu D. et al.Preclinical studies for cell transplantation: isolation of primate fetal hepatocytes, their cryopreservation, and efficient retroviral transduction.Hum Gene Ther. 1997; 8: 267-74Crossref PubMed Scopus (14) Google Scholar]. As these approaches have met with little success, we explored the strategy of injecting midgestation fetal liver cells into midgestation fetuses. We rationalized that the fetal liver is undergoing significant functional and developmental changes developing from an early hematopoietic organ into a neonatal liver. Therefore, the developing liver may be unlike the regenerating adult organ that readily incorporates donor cells. Rather, there may be discreet windows of opportunity to introduce cells of the appropriate developmental stage to colonize the uninjured but developing liver. Using injection protocols similar to those used for hematopoietic cells [31Kim H.B. Shaaban A.F. Yang E.Y. Liechty K.W. Flake A.W. Microchimerism and tolerance after in utero bone marrow transplantation in mice.J Surg Res. 1998; 77: 1-5Abstract Full Text PDF PubMed Scopus (82) Google Scholar, 32Shaaban A.F. Kim H.B. Milner R. Flake A.W. A kinetic model for the homing and migration of prenatally transplanted marrow.Blood. 1999; 94: 3251-7Crossref PubMed Google Scholar], we first examined how the age of recipient embryos affected the receptiveness of donor hepatocytes. C57Bl6 embryos, age E11.5 to E15.5, were injected with Rosa26 adult liver cells prepared by collagenase perfusion of the donor liver. Examination of their livers following birth indicated that occasional colonies of X-gal stained cells were observed most frequently in neonates that were injected as embryos at E12.5. The survival rate of embryos injected at E11.5 was low, which probably reflects inadvertent injury to the very small developing embryo by the injection needle. As the E12.5 embryos appeared to be most receptive to hepatic colonization, we compared colonization by adult liver cells to that by E12.5 FLCs. Preliminary results suggested that the FLCs colonized more efficiently than mature hepatocytes. Therefore, FLCs were used for subsequent experiments described in this paper. Timed matings of FX+/− heterozygotes were initiated to generate FX+/+, FX+/− and FX−/− embryos for use as recipients. Homozygous Rosa26 FX+/+, lacZ+/+ males were mated to wild-type C57Bl6 females to generate FX+/+, lacZ+/0 donor embryos. At E12.5, the pregnant C57Bl6 mothers were killed and the livers from the donor FX+/+, lacZ+/0 embryos harvested. Liver cells were prepared by trypsinizing the fetal livers, dispersing the cells by gentle pipetting of the liver cell suspension, filtering the cells through a 40-µm nylon mesh to eliminate cell clumps, and washing the cells by mild centrifugation. The cells were resuspended in PBS for intrauterine injection into the E12.5 recipient embryos. A total of 177 viable embryos from 22 litters were injected with FLC from Rosa26 E12.5 embryos. As controls, 134 viable embryos from 16 pregnant mothers were injected with saline. Interestingly, 17 of the E12.5 recipient embryos (for the Rosa26 cell injections) were already dead at various stages of resorbtion and thus were not manipulated. The fraction of resorbed embryos (17 of 194 total embryos) corresponds roughly to the expected number of FX−/− null embryos that one would expect to arrest at E10.5–E11.5 [26Dewerchin M. Liang Z. Moons L. et al.Blood coagulation factor X deficiency causes partial embryonic lethality and fatal neonatal bleeding in mice.Thromb Haemost. 2000; 83: 185-90Crossref PubMed Scopus (138) Google Scholar]. Approximately 10% of the embryos that developed to term were born dead either with no intestines or with intestines protruding from the abdomen. These non-viable pups showed no obvious signs of bleeding and were present at a similar ratio among all genotypes. We hypothesize that the intrauterine injections compromised the integrity of the abdominal wall in a small percentage of embryos, and the resulting herniation and externalization of the intestines led to neonatal lethality. In addition, all the pups from three litters died from maternal neglect. These neonates were removed from subsequent analysis. Thus, of the 177 embryos generated in FX heterozygous matings that were injected with Rosa26 FLCs, 117 viable neonates were born of which 32 were genotyped as wild-type, 73 heterozygotes, and 14 nulls. This distribution, with an under-representation of FX null neonates, is similar to that observed for the control group that received saline injections. The genotypic ratios are consistent with the partial embryonic lethality of FX-deficient embryos, with approximately 30–50% of FX−/− embryos dying at E10.5−E11.5. Of particular interest is the comparison of the survival of the FX−/− mice receiving in utero injections of either FLCs or saline shown in Fig. 2. All of the FX−/− neonates that received saline injections died within 3 days of birth, similar to that observed for untreated FX nulls [26Dewerchin M. Liang Z. Moons L. et al.Blood coagulation factor X deficiency causes partial embryonic lethality and fatal neonatal bleeding in mice.Thromb Haemost. 2000; 83: 185-90Crossref PubMed Scopus (138) Google Scholar]. In contrast, seven of 13 of the FX nulls that received FLCs survived longer than 4 weeks. One died spontaneously at 33 days, one was killed at 46 days, and five survived long-term (greater than 12 weeks) before they were killed for study. Of the FX nulls that received Rosa26 cells that died earlier than 4 weeks of age, the cause of death was similar to that observed with untreated or saline treated FX-deficient mice. Most of those dying within the first day showed signs of intraperitoneal bleeding, while those dying slightly later manifested bleeding at other sites, particularly intracranial bleeding with signs of hydrocephaly. Thus, approximately 50% of neonates receiving interuterine injections of FLCs were rescued from the neonatal lethality suffered by untreated or saline-treated FX−/− null mice. As the colonizing Rosa 26 FLCs contained a constitutively active lacZ gene, cells in the adult mice that derive from the donor population express β-galactosidase and stained blue when incubated with the appropriate chromogenic substrates. We prepared whole mounts of brains, hearts, livers, lungs, spleens, kidneys, intestines, bone marrows, thymuses, and ovaries or testes from 45 adult animals born from five litters, in which all the embryos were injected with Rosa26 FLC. A total of 47% of the mice showed evidence of Rosa26 colonization in the liver (Table 1) by macroscopic inspection of the whole mounts. In those livers displaying stained colonies, the extent of colonization (as estimated by the percentage of surface area that stains for β-gal) varied from ∼ 1–5% (Fig. 3).Table 1Extent of chimerism in organs of adult mice injected with Rosa26 FLCs at E12.5. Chimerism was established by the presence of X-gal staining regions in whole-mount preparations of isolated organs, PCR detection of lacZ DNA in DNA samples from total DNA preparation of isolated organs, or X-gal histologic staining of thin sections prepared from isolated organs. Organs were prepared from 45 mice for whole-mounts, nine mice for PCR analysis and 12 mice for histologic thin sections. Bone marrow was prepared from 21 animals. The entries indicate the fraction of organs for which there was detectable colonization by Rosa26 cellsOrganWhole mountPCRHistologic sectionsNumber positive/total%Number positive/total%Number positive/total%Heart0/4502/922ndndLiver21/45474/9449/1275Kidneyndnd3/9330/120Brain3/4571/9111/128Thymus1/4523/9332/1217Spleenndnd6/9676/1250Lung5/45117/9788/1267Bone marrowndnd2/7228/2138Intestinendnd6/967ndndOvaryndnd0/301/520Testis1/2150/402/728nd, not done. Open table in a new tab nd, not done. As the fetal liver is a hematopoietic organ, it is not surprising that significant colonization was also observed in the bone marrow and spleen. Bone marrow smears from 8/21 animals exhibited the unambiguous presence of Rosa26-derived cells (Table 1). However, the degree of colonization was low with the maximum fraction of donor-derived bone marrow cells in any animals being less than 1%. Donor colonies were also detected macroscopically in the spleen although estimating the degree of colonization was problematic since it is difficult to see the X-gal stain in contrast to the dark-red color of the spleen. Furthermore, spleens from control animals that did not receive donor Rosa26 cells displayed weak background staining with X-gal. In addition to the liver and hematopoietic organs, colonization, detected by macroscopic inspection of whole mounts, was also found in 5/45 lungs, 1/45 brains, and the testis in one of the 21 males. Because kidneys and intestines stained blue in control animals that did not receive Rosa26 cells, it was not possible to estimate the degree of colonization in those organs. A quantitative PCR strategy was employed to obtain a more accurate measure of the degree of colonization in each organ. The quantity of lacZ and protein C DNA was determined by PCR on identically sized samples of total DNA from each organ of nine animals. Since the protein C gene is found in both donor and recipient-derived cells while the lac Z gene is only found in the donor-derived cells, the ratio of the level of the lacZ gene to that of the protein C gene provided a measure of the fraction of donor cells in the sample organ. The strategy was validated by making defined DNA mixes of known proportions of wild-type and Rosa26 DNA and determining the ratio of the level of lacZ to PC DNA (data not shown). Heart, liver, kidney, brain, thymus, spleen, lung, bone marrow, intestine, and testis or ovaries were harvested from nine mice that received in utero injections of Rosa FLCs. At least trace amounts of lacZ DNA were found in at least one organ of 8/9 animals. LacZ DNA was detected in 4/9 livers, 7/9 lungs, and 6/9 spleens (Table 1). Intestines that were not completely rinsed of luminal contents showed very high LacZ levels, which probably reflects bacterial contamination contributing contaminating lacZ DNA. This PCR result is consistent with the high degree of β-gal staining seen in the intestine. Histologic sections of livers and lungs were examined to determine the morphology of the donor cells. The Rosa26-derived cells in the liver assumed normal hepatic structures indistinguishable from regions formed by the recipient animal (Fig. 4a,b). Most of the donor cells appeared in contiguous clusters, suggesting the donor hepatocytes formed colonies. Similarly, the β-gal-staining cells in the spleen formed donor clusters primarily in the white pulp and consisted of leukocytes in various stages of maturation, which were indistinguishable form recipient cells (Fig. 4c). Interestingly, a β-gal staining nodule was found on the surface of 1/21 testes (Fig. 3h). The cells in these nodules were rhomboid and contained large single or double nuclei with extensive granular cytoplasm and resembled hepatocytes (data not shown). Thus, it is conceivable these represented ectopic hepatocyte-like nodules on the outside of the testes. Interestingly, no donor cells were found within seminiferous tubules and spermatogonia. PCR analyses indicated that approximately 80% of the lungs contained donor cells, suggesting it was a frequent target organ. The β-gal-staining cells in the lung were of three types. Most were leukocytes in normal lymphoid structures adjacent to air passages and their presence is consistent with the FLC donor population containing a significant percentage of hematopoietic cells (Fig. 4d). In addition, columnar epithelial cells lining larger bronchioles also stained for β-gal (Fig. 4e). The source of these cells is not known, and suggests the donor cell population contained a heterogeneous collection of many cell types with a variety of developmental potentials. In addition, two liver-like nodules were found in two of the 45 lungs. These nodules contained cells that appeared hepatocyte-like (Fig. 4f). Moreover, 1/45 adult animals receiving in utero injection of FLCs exhibited an X-gal staining colony in the brain (Fig. 3f). Furthermore, lymph nodes in adipose tissue surrounding the thymus also exhibited macroscopic colonies that stained with X-gal (Fig. 3g). Most of the Rosa-derived cells surrounding the thymus were lymphoid although in one thymus a hepatocyte-like nodule was detected (data not shown). In addition, histologic examination of 212 thymuses demonstrated the presence of a few scattered X-gal staining cells that appeared indistinguishable from the surrounding cells of the host (Table 1). Factor X levels were determined in the five rescued null mice that survived longer than 3 weeks. We were unable to collect samples from the mice that died spontaneously at earlier ages. The plasma levels of FX in nulls ranged from 0.6 to 6.2% of levels seen in wild-type littermates that received parallel in utero injections (Fig. 5). Interestingly, the mouse with the highest degree of colonization (determined by the percentage of surface area that stained for X-gal) had the highest FX plasma level. This suggests that the degree of colonization correlated with the level of FX production. Heterozygous littermates had approximately 50% of FX activity found in wild-type mice While 1–5% of wild-type levels of FX rescued mice from the spontaneous lethal hemorrhage resulting from total FX deficiency, these low levels did not protect animals from traumatic injury. Bleeding time of surviving null littermates that received in utero FLC injections were compared to those of wild-type and heterozygous littermates (Fig. 6). A total of 7/8 heterozygous and 12/17 wild-type mice ceased bleeding within 300 s of excision of the distal 1 cm of the tail. Three additional wild-types stopped bleeding before 900 s when the experiment was terminated. In contrast, the rescued nulls bled significantly longer; 5/6 bled at least 15 min (at which point the experiment was terminated), and one stopped bleeding at 13 min. Thus the 1–5% levels of FX did not protect mice from hemorrhage following surgical injury. The results described in this paper suggest that in utero transplantation of hepatocyte progenitor cells could be an effective strategy to ameliorate and possibly prevent pathologies caused by genetic deficiency of liver expressed proteins. Injection of E12.5 FLCs into the peritoneal cavity of E12.5 day embryos in utero resulted in long-term incorporation of the donor cells into the recipient livers in approximately 50% of the cases. The resulting mosaic livers incorporated approximately 1–5% donor cells, which provided sufficient FX to rescue FX-deficient mice from perinatal lethality associated with a deficiency of this protein. Analysis of chromogenic β-gal stains of whole mounts of excised livers of from 45 adult mice that were injected with FLCs as fetuses in utero suggests that donor cells are detected macroscopically in 47% of the livers of the recipient mice. Interestingly, of the FX null embryos that were born following in utero injection of FLCs, approximately 50% survived significantly longer than untreated FX nulls or FX nulls derived from embryos injected with saline. These results are consistent with the contention that hepatic colonization by FLCs (to the extent detectable by whole-mount chromogenic staining) rescues FX-deficient mice from neonatal lethality. The results described in Fig. 3 suggest that 1–5% of wild-type levels of FX are sufficient to mediate rescue from spontaneous lethal neonatal bleeding diatheses. These results are similar to the levels of FVIII or FIX necessary to convert severe hemophilia to more manageable moderate or mild disease. However, these protein levels did not protect animals from hemorrhage following surgical injury. Interestingly, we analyzed DNA from entire organs from 12 different mice that received in utero injections of Rosa26 FLCs. The results indicated a larger percentage of Rosa26 donor cells than detected by macroscopic examination of whole mounts in several organs including lung and spleen. Therefore, we examined serial histologic sections and observed lacZ-positive cells in 8/12 lungs. The donor cells appeared as normal leukocytes in regular lymphoid structures scattered throughout the lung and their presence is consistent with the large number of hematopoietic cells in the donor cell population. This is not surprising since the E12.5 liver is primarily a hematopoietic organ. Donor-derived Rosa26 columnar epithelial cells were found in the epithelium lining bronchioles. The origin of these cells is unclear. In addition, a small nodule of hepatocyte-like cells was found in the lung corresponding to one of the colonies detected in the lung by macroscopic inspection of whole mounts. As expected, the FLCs also colonize hematopoietic organs as spleen, bone marrow, and lymphoid structures in lung and thymus. The spleens of 6/9 and the bone marrow of 8/21 mice that received in utero injections of FLCs stained positive for β-gal. In addition, Rosa26-derived cells were observed by macroscopic inspection of whole mounts stained with Xgal in one of 45 brains. The in utero transplantation of hepatocyte progenitors offers a potentially novel therapeutic strategy to treat pathologies associated with deficiency of liver expressed enzymes. In particular, this strategy could be useful for the treatment of hemophilia A and B. If similar results could be achieved for the expression of FVIII or FIX, severe hemophilia A or B might be converted to moderate or mild disease. In both cases, a significant percentage of hemophiliacs are born to known carriers. Therefore, in utero genetic screening of carrier mothers could readily identify a significant number of fetuses that will develop hemophilia. As the penetrance of FVIII and FIX deficiencies is high, it is known that fetuses with particular FVIII or FIX mutations will very likely develop clinical symptoms. Since prophylactic treatment of young children with FVIII or FIX deficiency before the onset of major symptoms is accepted practice, it is unlikely that intervention in utero would raise bioethical objections. As expected, the donor-derived cells persisted into adulthood and produced significant levels of FX. Because the cells were introduced into the mouse before the maturation of the immune system, the donor cells should be considered ‘self’ and the FX generated by the donor cells should not elicit an immune response. This strategy thus offers an advantage for the replacement of missing factors in adults where a significant percentage of patients develop inhibiting antibodies to the therapeutic proteins. The level of expression of FX achieved in the adult recipients is approximately 1–6% of wild-type levels. Preliminary studies indicated that the FLCs can be cultured in vitro and that early passage cells will mediate rescue of FX deficient animals. As these cells can be transformed with exogenous DNA or transfected with recombinant viruses, it is realistic to consider genetically modified cells with enhanced properties for in utero transplantation. In particular, it should be possible to engineer cells producing higher levels of the therapeutic protein or cells that have enhanced potential for colonization. Furthermore, the in vitro differentiation of murine embryonic stem cells into cells expressing hepatocyte specific genes has been reported [33Hamazaki T. Iiboshi Y. Oka M. et al.Hepatic maturation in differentiating embryonic stem cells in vitro.FEBS Lett. 2001; 497: 15-9Crossref PubMed Scopus (379) Google Scholar]. Thus it might be possible to develop donor cell populations from ES cells eliminating the need for early passage primary cells. In addition, multipotent cells have been derived from adult bone marrow [34Korbling M. Anderlini P. Peripheral blood stem cell versus bone marrow allotransplantation: does the source of hematopoietic stem cells matter?.Blood. 2001; 98: 2900-8Crossref PubMed Scopus (291) Google Scholar, 35Bonilla S. Alarcon P. Villaverde R. Aparicio P. Silva A. Martinez S. Haematopoietic progenitor cells from adult bone marrow differentiate into cells that express oligodendroglial antigens in the neonatal mouse brain.Eur J Neurosci. 2002; 15: 575-82Crossref PubMed Scopus (62) Google Scholar, 36Reyes M. Verfaillie C.M. Characterization of multipotent adult progenitor cells, a subpopulation of mesenchymal stem cells.Ann N Y Acad Sci. 2001; 938: 231-3Crossref PubMed Scopus (330) Google Scholar]. Another potential use of this procedure is in xenotransplantation since donor cells are introduced prior to the development of the immune system. Thus, the murine system provides a potential bioassay for the development of human hepatocyte progenitor cells. The authors thank Dr Alan Flake and Dr Ross Milner for providing training with techniques involved with in utero transplantation and Mayra Sandoval-Cooper for assistance with histological analysis. This work was supported in part by grants 19982 from N.I.H and Kleider-Pezold endowed professorship (to FJC) and grants HL65241 and HL60081 from N.I.H. (to EDR).