In vitro development of bovine nuclear transfer embryos from transgenic clonal lines of adult and fetal fibroblast cells of the same genotype

This study examined bovine cloning strategies that may be used for gene targeting in animals of known phenotypic traits. Fibroblast cells derived from an adult and a fetus of the same genotype were transfected with a plasmid (pEGFP-N1) containing the enhanced green fluorescence protein and neomycin-resistant genes. After transfecting 2 x 10(5) cells, 49 adult and 35 fetal cell colonies were obtained. Green fluorescence expression was observed in 35 out of 49 (71.4%) adult clones and in 30 out of 35 (85.7%) fetal clones. Developmental rates to the blastocyst stage following nuclear transfer (NT) did not differ among nontransfected cell lines (adult, 20.0%; NT fetal, 18.3%), whereas developmental rates were significantly lower for adult and fetal cell lines expressing enhanced green fluorescent protein (EGFP; 11.3% and 6.4%, respectively, P < 0.05). However, there was no decrease in NT developmental rates (19.8%) when donor nuclei from EGFP-transfected cell lines not expressing EGFP but retaining neomycin-resistant gene expression were used as donor nuclei. NT embryos from adult and fetal cell lines had similar morphology, cell number, and ploidy. The results indicated that adult and NT fetal cells (identical genotype) can complete clonal propagation, including transfection and selection, and can be used to produce transgenic NT embryos; however, a possible deleterious effect of EGFP on embryo development should be considered in future gene targeting studies.     

Bovine nuclear transfer embryo development using cells derived from a cloned fetus.

Many different cell types have been used to generate nuclear transfer embryos and fetuses. However, little is known about the potential of fibroblasts derived from a nuclear transfer fetus as donor cells for nuclear transfer. The ability of cloned fetuses or animals to be cloned themselves is of great interest in determining whether successive generations of clones remain normal or accumulate genetic or phenotypic abnormalities. We generated a bovine fibroblast cell line from a cloned fetus, that continued to divide beyond 120 days (94 doublings,18 passages) in continuous culture. As long-term survival of cells in culture is a desirable characteristic for use in transgenic cell production, passage 2 and 18 cells were compared as donor cells for nuclear transfer (NT). When cells from passage 2 (2 weeks in culture) and passage 18 (4 months in culture) were used for nuclear transfer, there was no significant difference in development rate to blastocyst (35.4 versus 44.6%, P=0.07). A greater proportion of late passage cells were in G0/G1 whether under serum-fed (64 versus 56%, P<0.01) or serum-starved (95 versus 88%, P<0.01) culture conditions. Following embryo transfer, equivalent day 30 pregnancy rates were observed for each group (P 2: 2/19 versus P 18: 2/13). A slightly retarded fetus was surgically removed at day 56 and the remaining three fetuses died in utero by day 60 of gestation. Our results show that fibroblast cells derived from regenerated cloned fetuses are capable of both in vitro and in vivo development. The longevity of this regenerated cell line would allow more time for genetic manipulations and then to identify stable transfected cells prior to their use as NT donor cells. Although no live fetuses were produced in this study the results provide encouraging data to show that a cloned fetus can itself be recloned to produce another identical cloned fetus. Further studies on this and other recloned fetuses are necessary to determine whether the failure to produce live offspring was a result of inadequate sample size or due to the cell type selected.     

Transmission of mitochondrial DNA in pigs and progeny derived from nuclear transfer of Meishan pig fibroblast cells

In embryos derived by nuclear transfer (NT), fusion, or injection of donor cells with recipient oocytes caused mitochondrial heteroplasmy. Previous studies have reported varying patterns of mitochondrial DNA (mtDNA) transmission in cloned calves. Here, we examined the transmission of mtDNA from NT pigs to their progeny. NT pigs were created by microinjection of Meishan pig fetal fibroblast nuclei into enucleated oocytes (maternal Landrace background). Transmission of donor cell (Meishan) mtDNA was analyzed using 4 NT pigs and 25 of their progeny by PCR-mediated single-strand conformation polymorphism (PCR-SSCP) analysis, PCR-RFLP, and a specific PCR to detect Meishan mtDNA single nucleotide polymorphisms (SNP-PCR). In the blood and hair root of NT pigs, donor mtDNAs were not detected by PCR-SSCP and PCR-RFLP, but detected by SNP-PCR. These results indicated that donor mtDNAs comprised between 0.1% and 1% of total mtDNA. Only one of the progeny exhibited heteroplasmy with donor cell mtDNA populations, ranging from 0% to 44% in selected tissues. Additionally, other progeny of the same heteroplasmic founder pig were analyzed, and 89% (16/18) harbored donor cell mtDNA populations. The proportion of donor mtDNA was significantly higher in liver (12.9 +/- 8.3%) than in spleen (5.0 +/- 3.9%), ear (6.7 +/- 5.3%), and blood (5.8 +/- 3.7%) (P < 0.01). These results demonstrated that donor mtDNAs in NT pigs could be transmitted to progeny. Moreover, once heteroplasmy was transmitted to progeny of NT-derived pigs, it appears that the introduced mitochondrial populations become fixed and maternally-derived heteroplasmy was more readily maintained in subsequent generations.     

Production of transgenic blastocyst by nuclear transfer from different types of somatic cells in cattle

The present study examined the effects of genetic manipulation to the donor cell and different types of transgenic donor cells on developmental potential of bovine nuclear transfer (NT) embryos. Four types of bovine somatic cells, including granulosa cells, fetal fibroblasts, fetal oviduct epithelial cells and fetal ovary epithelial cells, were transfected with a plasmid (pCE-EGFP-Ires-Neo-dNdB) containing the enhanced green fluorescent protein (EGFP) and neomycin-resistant (Neor) genes by electroporation. After 14 days selection with 800 μg/mL G418, transgenic cell lines from each type of somatic cells were obtained. Nontransgenic granulosa cells and all 4 types of transgenic somatic cells were used as nuclear donor to produce transgenic embryos by NT. There was no significant difference in development rates to the blastocyst stage for NT embryos from transgenic and nontransgenic granulosa cells (44.6% and 42.8%, respectively), and transfer of NT embryos derived from transgenic and nontransgenic granulosa cells to recipients resulted in similar pregnancy rates on day 90 (19% and 25%, respectively). The development rates to the blastocyst stage of NT embryos were significantly different among different types of transgenic donor cells (P<0.05). Blastocyst rates from fetal oviduct epithelial cell and granulosa cell (49.1% and 44.6%, respectively) were higher than those from fetal fibroblast (32.7%) and fetal ovary epithelial cell (22.5%). These results suggest that (i) genetic manipulation to donor cells has no negative effect on in vitro and early in vivo developmental competence of bovine NT embryos and (ii) granulosa and fetal oviduct epithelial cells can be used to produce transgenic bovine NT embryos more efficiently. In addition, GFP can be used to select transgenic NT embryos as a non-invasive selective marker.     

Production of transgenic blastocyst by nuclear transfer from different types of somatic cells in cattle

Genetically modified animals have many poten-tial applications in basic research, human medicine and agriculture. Pronuclear DNA microinjection has been almost the only practical means of producing transgenic animals during the last 20 years, but the low efficiency (1%—5%)[1] of this method has actu-ally been the obstacle that hampered its further appli-cation in animal biotechnology. The birth of Dolly[2], the first somatically cloned animal, made it possible to produce transgenic animals b…     

Establishment and characterization of fetal fibroblast cell lines for generating human lysozyme transgenic goats by somatic cell nuclear transfer

This study was performed to qualify goat fetal fibroblast (GFF) cell lines for genetic modification and somatic cell nuclear transfer (SCNT) to produce human lysozyme (hLYZ) transgenic goats. Nine GFF cell lines were established from different fetuses, and the proliferative lifespan and chromosomal stability were analyzed. The results suggested that cell lines with a longer lifespan had stable chromosomes compared with those of cells lines with a shorter lifespan. According to the proliferative lifespan, we divided GFF cell lines into two groups: cell lines with a long lifespan (GFF1/2/7/8/9; group L) and cell lines with a short lifespan (GFF3/4/5/6; group S). Next, a hLYZ expression vector was introduced into these cell lines by electroporation. The efficiencies of colony formation, expansion in culture, and the quality of transgenic clonal cell lines were significant higher in group L than those in group S. The mean fusion rate and blastocyst rate in group L were higher than those in group S (80.3 ± 1.7 vs. 65.1 ± 4.2 % and 19.5 ± 0.6 vs. 15.1 ± 1.1 %, respectively, P < 0.05). After transferring cloned embryos into the oviducts of recipient goats, three live kids were born. PCR and Southern blot analyses confirmed integration of the transgene in cloned goats. In conclusion, the lifespan of GFF cell lines has a major effect on the efficiency to produce transgenic cloned goats. Therefore, the proliferative lifespan of primary cells may be used as a criterion to characterize the quality of cell lines for genetic modification and SCNT.     

Development and expression of the green fluorescent protein in porcine embryos derived from nuclear transfer of transgenic granulosa-derived cells

Nuclear transfer (NT) techniques have advanced in the last few years, and cloned animals have been produced from somatic cells in several species including pig. In this study we examined the feasibility of using granulosa-derived cells (GCs) as donor cells combined with a microinjection procedure to transfer those nuclei. In vitro matured oocytes were enucleated by aspirating the first polar body and adjacent cytoplasm. Mural GCs infected with an enhanced green fluorescence protein (EGFP) gene were serum-starved (0.5% serum, 7 days), injected directly into cytoplasm of enucleated oocytes and the oocytes were electrically activated. The reconstructed embryos were cultured for 7 days and stained with Hoechst 33342 to determine the number of nuclei. Non-manipulated oocytes were electrically activated and cultured as controls. At 9 h post-activation, the pronuclear formation rates were 78.7+/-3.7% in NT and 97.4+/-4.4% in controls at 9 h post-activation. After 7 days culture, the cleavage rates were 24.5+/-7.2% in NT and 79.3+/-5.6% in controls. The blastocysts formation rates were 4.9+/-2.4% in NT and 26.8+/-3.8% in controls. To examine the effect of activation time on development of NT embryos, oocytes were activated at 0-0.5, 1-2, or 3-4 h post-injection. At 18 h post-activation the pronuclear formation rates were higher (62.5+/-7.3%) in the 3-4 h group as compared to the 0-0.5 h (22.0+/-12.5%) or 1-2h (44.5+/-6.3%) groups (P<0.05). However, the cleavage rates (9.6+/-4.6 to 10.7+/-4.2%) and the blastocysts formation rates (1.2+/-2.4 to 4.9+/-3.7%) were not different among treatments (P>0.05). The mean cell number of blastocysts was 15.7+/-5.7 in NT and 25.3+/-24.7 in controls. Green fluorescence was observed in roughly half of the embryos from the one-cell to the blastocyst stage. These results indicate that granulosa-derived cell nuclei can be remodeled in the cytoplasm of porcine oocytes, and that the reconstructed embryos can develop to the blastocyst stage. In addition, EGFP can be used as a marker for gene expression of donor nuclei.     

克隆猪技术及其在转基因猪研究中的应用

哺乳动物核移植技术是一种可以获得基因组遗传信息完全相同的后代的生物技术。猪体细胞核移植技术包括以下几个环节：卵母细胞的体外成熟、供体细胞的分离和处理、体细胞的核转移、重构胚胎的人工激活、胚胎体外培养和胚胎移植。由于该技术在最近几年的迅速发展,很多实验室已通过该技术成功获得了克隆猪后代。核移植克隆猪技术的出现为生产转基因猪提供了一种有效的方法,并且是目前生产基因打靶猪的惟一方法。至今利用克隆猪技术已经成功获得了一系列的转基因猪和基因敲除猪。以核移植技术产生基因修饰猪目前正处于从基础研究走向应用的过渡阶段。尽管猪体细胞核移植克隆的效率（出生克隆猪数占所用卵数的比例）还不高,但是由于通过该技术能够对猪基因组进行特定的修饰,确保生产的克隆动物100％为转基因动物,从而大大提高了转基因猪的制作效率,可以预料猪核移植技术将会对医药业和农业产生重大的影响。     

Characteristics of primary and immortalized fibroblast cells derived from the miniature and domestic pigs

Background

The pig, Sus scrofa domestica includes both the miniature and commercial domestic breed. These animals have influenced the human life and economies and have been studied throughout history. Although the miniature breeds are more recent and have increasingly been used in a variety of biomedical studies, their cell lines have rarely been established. Therefore, we sought to establish primary and immortal cell lines derived from both the miniature and domestic pig to better enable insight into possible in vivo growth differences.

Results

The in vitro lifespan of primary domestic pig fibroblast (PF) and miniature pig fibroblast (MPF) cells using a standard 3T3 protocol was determined. Both of the primary PF and MPF cells were shown to have a two-step replicative senescence barrier. Primary MPF cells exhibited a relatively shorter lifespan and slower proliferation rate compared to those of primary PF cells. Beyond senescence barriers, lifespan-extended PF and MPF cells were eventually established and indicated spontaneous cellular immortalization. In contrast to the immortalized PF cells, immortal MPF cells showed a transformed phenotype and possessed more frequent chromosomal abnormalities and loss of p53 regulatory function. The lifespan of primary MPF and PF cells was extended by inactivation of the p53 function using transduction by SV40LT without any detectable senescent phenotype.

Conclusion

These results suggest that p53 signaling might be a major determinant for the replicative senescence in the MPF cells that have the shorter lifespan and slower growth rate compared to PF cells in vitro.     

Effect of activation time on the nuclear remodeling and in vitro development of nuclear transfer embryos derived from bovine somatic cells

This study was conducted to investigate the effect of recipient activation time on the chromatin structure and development of bovine nuclear transfer embryos. Serum-starved skin cells were electrofused to enucleated oocytes, activated 1-5 hr after fusion, and cultured in vitro. Some fused eggs were fixed at each time point after fusion without activation, or 3 or 7 hr after activation. Some nocodazole treated zygotes were fixed to analyze their chromosome constitutions. The proportion of eggs with a morphologically normal premature chromosome condensation (PCC) state increased 1-2 hr after fusion. Whereas eggs with elongated chromosome plate increased as activation time was prolonged to 3 hr, and 5 hr after fusion, 58.1% of eggs showed more than two scattered chromosome sets. The proportion of eggs with a single chromatin mass (40.6-56.7%) significantly increased when eggs were activated within 2.5 hr after fusion (P < 0.05). Only 23.3% of reconstituted embryos activated 5 hr after fusion formed one pronucleus-like structure (PN), whereas, 64.5-78.3% of embryos activated 1-2.5 hr after fusion formed one PN. The proportion of embryos with normal chromosome constitutions decreased as activation time was prolonged. Development rates to the blastocyst stage were higher in eggs activated within 2 hr after fusion (17.3-21.7%) compared to those of others (0-8.6%, P < 0.05). The result of the present study suggests that activation time can affect the chromatin structure and in vitro development of bovine nuclear transfer embryos.     

Differential development of rabbit embryos derived from parthenogenesis and nuclear transfer

Parthenogenetic development (PA) is often used as a model to investigate activation protocols for nuclear transfer (NT) embryos. The objective of this study was to compare the development, as well as the dynamics of the nuclear materials and microtubules of PA and NT embryos following similar activation treatment. Our results demonstrate that, during parthenogenesis, activation through either electrical pulses or chemical stimulation alone resulted in low cleavage rates and compromised development. A combination of two sets of electrical pulses and a 2-h-exposure to chemical activation medium (5 microg/ml cycloheximide (CHX) and 2 mM 6-dimethylaminopurine (6-DMAP) in KSOM+0.1% BSA) could effectively activate rabbit oocytes, and resulted in a 99% (n = 73) cleavage rate with greater than 60% (n = 73) developing to blastocysts at day 4. However, the same activation protocol following NT resulted in only 65-72% of oocytes cleaved (depending on donor cell type), with less than 20% developing to the blastocyst stage. The differences observed between NT and PA embryos subjected to the same activation protocol were also evident in terms of the time required for their development to the blastocyst stage, as well as the cell numbers present in blastocysts at day 6. Furthermore, laser confocal microscopy revealed that pronuclear formation in the NT embryos was delayed by comparison to that in the parthenotes. In conclusion, our study suggests that an effective protocol for parthenogenesis cannot promise a comparable outcome for NT embryos.     

Establishment of an efficient somatic cell nuclear transfer system for production of transgenic pigs

Handmade cloning (HMC) is now an established procedure used in several species for somatic cell nuclear transfer, but only applied in two related laboratories for pigs. The aim of this review is to facilitate widespread application by summarizing the process of establishment and explaining the background of the incorporated special approaches. Optimized steps of traditional cloning in pigs (in vitro maturation, activation, embryo culture) were merged with those of the micromanipulation-free HMC that has been modified according to the specific needs of sensitive porcine oocytes (partial zona digestion before enucleation, two-step zona-free fusion with the somatic cell; initiation of activation with the second fusion). The zona-free approach required embryo culture to the blastocyst stage before surgical transfer of embryos to the uterine horns of recipient sows in the proper phase of an unstimulated cycle. Eventually a competitive, inexpensive and reliable alternative to traditional porcine nuclear transfer cloning techniques evolved that is also suitable to produce transgenic offspring containing various genetic modifications to establish models for several human diseases with genetic background. Further improvements and involvement of additional techniques to increase the overall efficiency and facilitate practical applications are expected in the foreseeable future.     

Epithelial and fibroblast cell lines derived from a spontaneous mammary carcinoma in a MMTV/neu transgenic mouse

Female murine mammary tumor virus (MMTV)/neu transgenic mice, expressing a wild-type rat neu oncogene driven by an MMTV promoter, develop focal mammary adenocarcinomas that are pathologically very similar to human breast tumors. Two new cell lines were established from a mammary tumor that arose in a female MMTV/neu transgenic mouse. One of these lines, mammary carcinoma from Neu transgenic mouse A (MCNeuA), has an epithelial morphology, is cytokeratin positive, and expresses high levels of the neu transgene. Karyotyping and comparative genomic hybridization analyses demonstrated genomic alterations in the MCNeuA cell line. The other line, N202Fb3, has a fibroblast morphology, is cytokeratin negative, and expresses the neu transgene at a very low level. This cell line also expresses smooth muscle alpha-actin, suggesting that it is a myofibroblast line. The MCNeuA cell line is tumorigenic when injected into syngeneic MMTV/neu transgenic mice, with an in vivo doubling time of about 14 d. The rationale for establishing this tumor cell line was to provide a tumor transplantation system for rapidly assessing immunotherapeutic interventions before testing in the more cumbersome model of spontaneous tumor development in the MMTV/neu transgenic mice. Mice immunized with a Neu extracellular domain protein vaccine were protected against a subsequent inoculation of MCNeuA cells, indicating that this cell line will be useful for evaluating cancer vaccine strategies. This tumor cell line may also prove useful in studying the biological properties of the neu oncogene and its role in the malignant process. In addition, the tumor-derived fibroblast line may be useful for studying tumor-stromal cell interactions.     

Effects of the time interval between fusion and activation on in vitro rabbit nuclear transfer efficiency when nuclear donor cells are derived from older adults

Cloning older adult rabbits can serve as a model in animal breeding, biodiversity preservation and in human therapeutic cloning. To establish the required exposure time of fibroblasts from these kind of animals to reprogramming factors, in the present study three different time intervals between fusion and activation were tested (30 min, 30-ADF group; 60 min, 60-ADF group; and 90 min, 90-ADF group). Vitrified epithelial fibroblasts derived from four older adult rabbit females (D1, D2, D3 and D4) and cultured from passages 0 to 4 were used as nuclear donors. Nuclear status of reconstructed embryos was not evaluated. No differences were observed in blastocyst rate (30-ADF 21% vs 60-ADF 19% vs 90-ADF 18%). Differences in hatching rates did not reach significance (30-ADF 11% vs 60-ADF 18% vs 90-ADF 18%). However, in the 60- and 90-ADF groups, embryos reached the blastocyst stage earlier than in the 30-ADF group (day 4: 40% and 50% vs 8%; p > 0.05). Moreover, the quality of blastocysts (good vs poor) was lower in the 30-ADF group (good: 30-ADF 38% vs 60-ADF 90% vs 90-ADF 90%; p > 0.05). Overall, these results suggest an unfavourable effect of the shortest exposure time tested (30 min). Differences between specimen origins were detected (blastocyst and hatching rates: D2 (26%; 25%) and D4 (25%; 27%) vs D1 (10%; 11%) and D3 (12%; 12%)), but significance were not reached. Effect of culture passage was not detected in any parameter studied.     

In vitro development of nuclear transfer embryos derived from porcine embryonic germ cells and their descendent neural precursor cells

Undifferentiated stem cells may support a greater development of cloned embryos compared with differentiated cell types due to their ease of reprogramming during the nuclear transfer (NT) process. Hence, stem cells may be more suitable as nuclear donor cells for NT procedures than are somatic cells. Embryonic germ (EG) cells are undifferentiated stem cells that are isolated from cultured primordial germ cells (PGC) and can differentiate into several cell types. In this study, the in vitro development of NT embryos using porcine EG cells and their derivative neural precursor (NP) cells was investigated, thus eliminating any variation in genetic differences. The rates of fusion did not differ between NT embryos from EG and NP cells; however, the rate of cleavage in NT embryos derived from EG cells was significantly higher (p < 0.05) than that from NP cells (141/247 [57.1%] vs. 105/228 [46.1%]). Similarly, the rate of blastocyst development was significantly higher (P < 0.05) in NT using EG cells than the rate using NP cells (43/247 [17.4%] vs. 18/228 [7.9%]). The results obtained from the present study in pigs demonstrate a reduced capability for nuclear donor cells to be reprogrammed following the differentiation of porcine EG cells. Undifferentiated EG cells may be more amenable to reprogramming after reconstruction compared with differentiated somatic cells.     

Development of the techniques for nuclear transfer in pigs

Nuclear transfer in pigs was developed in the late 1980's. The techniques were based on previous studies in frogs, mice and cattle. Within stage nuclear transfer, pronuclear exchange, was followed by the transfer of nuclei from cleavage stage embryos. While these have resulted in term development, many problems remain. Recently progress on the problem of inadequate oocyte activation has been made and now there can be a refocus on the other aspects of the nuclear transfer procedure. The emphasis in developing the cloning/transgenic technology is easily justified, not so much by the ability to produce genetically identical animals for production agriculture, but for the potential to use a cell line that can be genetically engineered prior to the nuclear transfer. Pigs with specific genetic modifications will have a great impact on production agriculture as well as human medicine.     

Production of transgenic goats expressing human coagulation factor IX in the mammary glands after nuclear transfer using transfected fetal fibroblast cells

There are growing numbers of recombinant proteins that have been expressed in milk. Thus one can consider the placement of any gene of interest under the control of the regulatory elements of a milk protein gene in a dairy farm animal. Among the transgene introducing techniques, only nuclear transfer (NT) allows 100?% efficiency and bypasses the mosaicism associated with counterpart techniques. In this study, in an attempt to produce a transgenic goat carrying the human coagulation factor IX (hFIX) transgene, goat fetal fibroblasts were electroporated with a linearized marker-free construct in which the transgene was juxtaposed to ??-casein promoter designed to secret the recombinant protein in goat milk. Two different lines of transfected cells were used as donors for NT to enucleated oocytes. Two transgenic goats were liveborn. DNA sequencing of the corresponding transgene locus confirmed authenticity of the cloning procedure and the complementary experiments on the whey demonstrated expression of human factor IX in the milk of transgenic goats. In conclusion, our study has provided the groundwork for a prosperous and promising approach for large-scale production and therapeutic application of hFIX expressed in transgenic goats.     

Comparison of the effect of cortisol on aromatase activity and androgen metabolism in two human fibroblast cell lines derived from the same individual

The effect of preincubation with cortisol on estrogen and androgen metabolism was investigated in human fibroblast monolayers grown from biopsies of genital and non-genital skin of the same person. The activity in the cells of aromatase, 5 alpha-reductase, 17 beta-hydroxysteroid oxidoreductase and 3 alpha-hydroxysteroid oxidoreductase was investigated by isolating estrone, estradiol, estriol, dihydrotestosterone, androstanedione, androsterone, 3 alpha-androstanediol, testosterone and androstenedione after incubation of the cells with [14C]testosterone or [14C]androstenedione. For experiments with 14C-labeled substrate the cells were incubated in medium, charcoal stripped of steroids without Phenol Red. Preincubation from 6 to 36 h with cortisol in concentrations of 10(-8) - 10(-6) M showed maximal stimulation of aromatase activity after 12 h preincubation with cortisol in concentrations of 0.5-1.0 x 10(-6) M in both cell lines. When preincubation with cortisol was omitted no estrogen synthesis was detected. The formation of androgen was not altered after preincubation with cortisol. Pronounced differences were found in estrogen and in androgen metabolism in the two cell lines suggesting a local regulation of the hormonal environment. The aromatase activity, which is low in many tissues could be stimulated by cortisol without altering the androgen metabolism was found to be a suitable system for investigations of the cellular interconversion of androgens and estrogens and for investigations of the in vitro regulation of the enzymes involved.