Somatic nuclear transfer in livestock species

In this review, we discuss the recent advances in somatic cell nuclear transfer (NT) in sheep, cattle, goats, swine and rabbits. Numerous advances have been reported as this technique has developed over the last five years. In the first part of this review, we describe the reported data pertaining to each of the species mentioned above. Theories are offered to explain the different results seen between different species and cell types. One of the main aspects of somatic cell NT, the production of transgenic animals will also be reviewed. Future applications of this powerful technique are discussed. This review concludes with a discussion of some of the problems observed in animals produced using NT as well as possible Solutions for these challenges.


Introduction
Nuclear transfer procedures originally required donor-cells from undifferentiated early embryonal stages.The only exception was that cells from the inner cell mass (ICM)those cells present at the blastocyst stage of embryo development that are destined to give rise to the fetus itself -could support development and lead to the birth of live offspring (COLLAS and BARNES, 1994).In 1996 -a milestone in the history of NT -the theory that early embryonal stages were required for successful nuclear transfer became obsolete.CAMPBELL et al. (1996b) were the first to use differentiated embryonal cells, that were kept in culture for several passages as donor cells for NT.These experiments opened the new aera of somatic cell nuclear transfer.New and exciting applications of the NT technique becameapparent.ltappeared NT could beusedto facilitate the production of trans-genic animals, enhance the reproduetive Performance in endangered species (LANZA et al., 2000b) and eventually be used to retrieve extinet species.From 1996 on, reports of somatic cell nuclear transfer using a variety of differentiated embryonal, fetal and adult donor cells have been published.In this review, we summarize the advances that were achieved in sheep, cattle, goats, swine and rabbits in the last five years (Table 1).We focus on the observed differences in developmental potential observed among different types of donor cells and describe reports of transgenic animals produced with the aid of NT.Many reviews relating to the field of NT exist (COLMAN, 1999;KÜHHOLZER and PRATHER, 2000;WOLF et al., 2000).The aim of this review is to present a discussion of recent advances in NT from different points of view: cell cycle, cell types, synchronization between cytoplast and donor cell, maturation and in vitrosystems, artificial activation as well as new findings about mitochondrial DNA and telomeres.

Sheep
As with many other bio-techniques (embryo Splitting: WILLADSEN, 1979; NT using blastomeres: WILLADSEN, 1986), the experiments of somatic cell NT sueeeeded first in the sheep.Compared to cattle, this species has the advantage that its generation interval is significantly shorter; sheep can be maintained with minimal efforts, making experiments more time-and cost-efficient.CAMPBELL and his colleagues (1996) tried to establish embryonic stem cells from the inner cell mass, but they were not able to maintain them in an undifferentiated State.These embryonic cells were used as donor cells in their NT experiments rcgardless of their State of differentiation.Most scientists were surprised in 1996 when this group published the birth of five lambs, two of which survived, that were born following the transfer of these NT derived embryos to Surrogates (CAMPBELL et al., 1996b).The following year the same group (WILMUT et al., 1997) surprised the public and the scientific Community with another unbelievable report: The announcement of the birth of Dolly, a healthy lamb born following transfer of NT embryos, where the donor cells were derived from the differentiated adult mammary gland.Although this experiment was questioned in the beginning by some other groups, a number of reports followed describing comparable findings using somatic cell NT and therefore indirectly proved that the experiments in Wilmut's study were sound and could be replicated.CAMPBELL et al. (1996b) andWILMUT et al. (1997) stressed that the cell cycle arrest in the GO-stage is a major factor that enables reprogramming of a differentiated cell nucleus following NT.It was suggested to culture the cells for a certain period of time in medium with a reduced serum content (starvation), thereby arresting them before DNA-replication (pre-S-phase) in an inactive State (quiescence, G0).
Other factors are also important to improve the efficiency and reduce the cost of producing offspring by NT.One of them is the source of oocytes.It dramatically reduces the expenses of experiments when slaughterhouse ovaries are used as source of oocytes.These oocytes can be matured in vitro thereby eliminating the expenses of maintaining and treating live animals for the sole purpose of ooeyte donation.In a comparative study WELLS et al. (1997) investigated the results of somatic nuclear transfer when in vitro versus in vivo matured oocytes were used as reeipient cytoplasts.When using in vitro matured oocytes, the proportion of embryos that developed to term was reduced compared to in vivo oocytes.In addition the lamb that was derived from in vitro matured oocytes died within 10 min of delivery.However, this report does demonstrate that in vitro matured oocytes are a viable alternative to using surgically collected, in vivo matured oocytes.In 1999, SHIELS et al. (1999a, b) analysed the length of telomeres of Dolly, the sheep derived from a mammary gland cell.Telomeres are DNA repeats at the end of eukaryotic chromosomes and they shorten with every cell division, thereby being an indicator of cellular age and senescence.While conventional DNA polymerases are unable to completely replicate telomeres, their lengths can be re-established by an enzyme called telomerase, which is only active in early embryos and germ cells Telomeres not only protect chromosomal ends from degradation, they also facilitate completton of chromosomal DNA replication and contribute to chromosome positioning within the nucleus (KIERZENBAUM, 2000).Surprisingly, the length of the telomeres in Dolly was comparable to that found in the progenitor mammary tissue and significantly shorter than comparable telomeres of age-matching sheep (SHIELS et al., 1999a, b).They concluded that telomeres in NT embryos could not be restored to füll length because these embryos did not pass through the germ line.This would mean, that the genetic age of nuclear transfer animals is significantly older than their phenotypic age.However, this Observation of the telomeres in Dolly may not be the general case for all cloned mammals.In the next section the apparent contradictory findings from telomere studies in cloned cattle are discussed.These conflicting results highhght the need for further studies to fully understand the events that occur during NT and to eliminate or minimize problems associated with NT-animals.Since eukaryotic cells contain two distinct genomes, nuclear DNA and mitochondrial DNA (mtDNA), it was of high interest to determine the origin of the mtDNA in animals produced by somatic NT.Since mtDNA is maternally inherited after normal fertihzation, nuclear transfer where the donor cell is fused to the reeipient cytoplast should result in mtDNA heteroplasmy.However, when EVANS et al. (1999) analysed the mtDNA of Dolly and nine other NT sheep derived from fetal cells, only ooeytedenved mtDNA could be detected.Interestingly and similar to analysis of telomeres, controversial results were obtained in cattle (see below).

Cattle
The first report ofthe birth of calves generated by somatic NT was published in 1998 by CIBELLI and coworkers. In this study CIBELLI et al. (1998) used cycling cells without any synchronization treatment.Up to that time the prevailing opinion was that donor cells had to be arrested in the presumptive GO-stage of the cell cycle to allow complete reprogramming of differentiated donor cell nuclei.A study by ZAKHARTCHENKO et al. (1999c) that compared cycling and serum starved fetal fibroblasts as donor nuclei revealed NT embryos produced with serum starved cells developed to the blastocyst stage at a higher rate (39% vs 20%) and resulted in more pregnancies after transfer (7/9 vs 1/3) when compared to cycling cells.In contrast, a higher proportion of pregnancies has lost when embryos were derived from serum-starved cells (7/7) compared to cycling cells (0/1).HILL et al. (1999aHILL et al. ( , 2000b) also found better in vitro development when serum starved fetal cells were used.However, the same group (HILL et al., 2000b) could not find any positive effect of serumstarvation on the development to the blastocyst stage when adult cells were used.It should be mentioned that every existing synchronization protocol for cycling somatic cells results in a small proportion of synchronized cells positioncd in a stage of the cell cycle other than the desired one.Additionally, individual cells used as donor-karyoplasts cannot be analysed prior to NT.Therefore, no one has been able to definitely prove that an embryo or offspring was derived from a donor cell that was in a certain stage ofthe cell-cycle.The advances in somatic cell NT are demonstrated by a variety of reports that followed.Within the last three years scientists have been able to produce offspring derived from primordial germ cells (ZAKHARTCHENKO et al., 1999b), embryonic germ cells (BRINK et al., 2000), various newbom and adult tissues (KATO et al., 1998;WELLS et al., 1999;KATO et al., 2000) and most surprisingly, even from white blood cells (GALLI et al., 1999).In contrast to findings in sheep (SHIELS et al" 1999a, b), LANZA et al. (2000a) found that NT procedures in cattle were capable of resetting and even extending the life span of somatic cells.These conflicting results could be attributed to species-speeifie differences in telomerase activity in early embryos, or to the use of donor-cells at different stages of their lifespan.While the study in sheep used serum-starved cells, LANZA et al. (2000a) used fibroblast cells that were close to cellular senescence.Accordingly, BETTS et al. (2001) found that cell passage and culture in medium with low serum concentrations reduced telomerase activity by 30 to 50%, compared to nontreated cells.It was demonstrated that cloned embryos inherit genomic modifications (shortened telomeres aecording to age and culture period) from their donor-cells.However, these modifications are subsequently reversed during early development of cloned embryos, thereby being reprogrammed and rejuvenated.Another report strengthens these findings.In cattle XU and YANG (2001) analysed telomerase in bovine NT-embryos and detected its activity in all stages, with highest levels found at the blastocyst stage.KUBOTA et al. (2000) were able to produce live offspring derived from donor cells after long-term culture, indicating cellular aging must be reversed to a certain extent in these experiments.It was also possible to produce live offspring from aged individuals (21 year-old bull), which also indicates that NT might successfully reverse aging (HILL et al., 2000b).STEINBORN et al. (2000) analyzed mtDNA of calves derived from fetal and adult fibroblasts as well as mammary gland cells.In seven out of ten analysed clones the donor-reeipient ratios of parental mtDNAs before fusion remained the same throughout embryogenesis and development to term.However, tissues of the other 3 clones showed significant reduction or absence of donor mtDNA due to unknown mechanisms.This is in contrast to findings in sheep where no mtDNA heteroplasmy could be detected in somatic NT-offspring.The report by STEINBORN et al. (2000) suggests that mtDNA heteroplasmy does not impede normal development and may be related to the high pregnancy losses observed in somatic NT-studies.It also indicates that there might be a Variation in the mechanisms involved in the inheritance of mtDNA.Therefore, further studies are required to enlarge our understanding of this subjeet.

Goats
Only two reports on somatic NT in goats have been published so far.In the first report BAGUISI et al. (1999) compared the use of metaphase II (MIT) versus pre-activated oocytes as reeipient cytoplasts.Both stages were recovered out of the oviduets from superovulated does.Some of the recovered oocytes already extruded a second polar body, which was interpreted as being a morphological sign of pre-activation.It was concluded that the flushing of the oviduets and short-term culture in Ca 2+ -containing medium probably induced activation.Another group of oocytes was pre-activated by ethanol treatment.While one offspring was derived from a Mll-oocyte and a set of twins was born after using self-activated oocytes, no term-development was observed when ethanol-activated oocytes were used as reeipient cytoplasts.Since only a low number of NTs were reported, it cannot be exeluded, that also the use of chemically pre-activated oocytes might result in live offspring.For NT using embryonic blastomeres that are mainly in S-phase ofthe cell-cycle, it was found that pre-activated cytoplasts are the universal reeipient (CAMPBELL et al., 1996a).In pre-activated cytoplasts, the DNA is replicated aecording to the cell-cycle stage and, therefore correct ploidy can be maintained.However, the use of differentiated donor cells seems to require high concentrations of maturation/mitosis/meiosis promotion factor (MPF) lo be properly reprogrammed (CAMPBELL, 1999).This and the fact that the majority of differentiated eultured cells are in pre-S-phase explain, why most scientists prefer the use ofthe combination of MH-cytoplasts and donor cells in Gl or GO.In contrast to the report by BAGUISI et al. (1999) who used in vivo matured oocytes, KEEFER et al. (2001) recovered immature oocytes by laparoscopic follicular aspiration and subsequently used them as reeipient cytoplasts after in vitro maturation.As another difference, KEEFER et al. (2001) synchronized their donor-fibroblasts by culture to confluency in addition to subsequent culture in low serum, thereby increasing the probability that cells in the GO-stage were used.Six kids were born from 2 different cell lines; three of them died within one month after delivery but the remaining kids appear to be healthy and are presently older than one year.

Swine
Since the first NT-piglet derived from a blastomere was born (PRATHER et al., 1989), many years passed without a major success in porcine nuclear transfer.Although there were serious efforts in many groups to produce pigs by somatic nuclear transfer, results were not comparable to those in sheep or cattle.It became clear, that porcine oocytes are harder to activate when compared to other species; therefore, different protocols were developed that improved this part.Besides electrical activation, a variety of chemical treatments were tested.In contrast to fertilization, which is followed by Ca 2+ -oszillations, electrical activation induces only a single Ca 2+ -transient  (reviewedby MACHATY and PRATHER, 1998).MACHATY et al. (1997) finally established a protocol for chemical activation which mimicked fertilization-events: The sulfhydryl reagent thimerosal induced repeated Ca 2+ transients, and in combination with subsequent incubation in dithiotreitol (DTT) porcine oocytes were successfuUy activated.Also nuclear transfer embryos that were activated aecording to this protocol developed to the blastocyst stage (KÜHHOLZER et al., 2000).However, no development to term could be established.Finally in 2000, two independent groups were able to produce piglets by somatic NT.In the first report (POLEJEAVA et al., 2000) the technique of serial NT was used to circumvent the problem of artificial activation.After the initial NT using granulosa cells reconstrueted embryos were eultured over night and subsequently the pronuclei of the developing embryos were removed and fused to enucleated zygotes.Thereby the cytoplast was activated by fertilization and provided optimal conditions for the transferred pronucleus.After transfer of 400 embryos into 7 reeipients five healthy piglets were born.In contrast to this report, ONISHI et al. (2000) used fetal fibroblasts as nuclear donors and transferred them by direct intracytoplasmic injection.Three to four hours after injection, reconstrueted embryos were activated by exposure to electrical pulses, followed by incubation in cytochalasin-B to prevent cytokinesis.This delayed activation enabled a longer exposure ofthe donor nucleus to the MH-cytoplast (high levels of MPF), which should facilitate correct/complete reprogramming.Similar to Polejacva's report, where embryos were transferred immediately (within 2 hours) after fusion, Onishi transferred 110 embryos at an early stage (2-to 8-cell stage) into 4 reeipients, which resulted in the birth of a single piglet.In both studies, ovulated oocytes were used as reeipient cytoplasts, which provided the optimal maturation.However, to reduce the cost of NT, it is necessary to establish in vitro Systems that are suitable to provide sufficient numbers of matured oocytes for NT.Shortly after the first two reports of somatic NT in the pig, BETTHAUSER et al. (2000) were able to produce nuclear transfer piglets from in vitro matured sow-ooeytes.Another big success in their study was, that the protocol for NT was reproducible and NT-piglets derived from fetal fibroblasts and cells from the genital ridge were born from two separate pregnancies with another five ongoing pregnancies reported.In addition to the successful production of piglets, BETTHAUSER et al. (2000) investigated the feasibility of sow versus gilt oocytes for their in vitro system: for both, IVF and NT about twice as much embryos developed to blastocysts when sow oocytes were used as opposed to gilt oocytes, an Observation that was also described by KÜHHOLZER et al. (2001).While neither of these reports describe significant changes of existing protocols, it is obvious that the management of reeipient herds, synchronization and embryo transfer might have been an underestimated factor for the success of producing NT-offspring.An increase in the number of embryos transferred per reeipient (more than 100) resulted in a significant improvement ofthe efficiency of somatic NT.Surprisingly, after more than 10 years of failure, a variety of groups sueeeeded to produce live NT-piglets using different donor cell types and different activation protocols.

Rabbits
Intcrestingly, so far there is no report ofthe successful production of live offspring after somatic nuclear transfer in rabbits.This is especially surprising, since the rabbit is a species with a very high fecundity when compared to other livestock species and gives outstanding efficiencies in other bio-techniques like pronuclear injections or superovulation and embryo transfer in general (reviewed by BESENFELDER et al 1998).When using the established technique of NT of embryonic blastomeres no major problems were observed like it was in the pig (HEYMAN et al 1990) However, there are still significant improvements required in this species.The lack of NI-offspring is also in contrast to the good in vitro development of rabbit parthenogenetic embryos or embryos produced by somatic NT (MITALIPOV et al 1999;YIN et al., 2000;DINNYES et al., 2001;NOWSHARI et al., submitted) This contrad.ct.on po.nts out clearly, that it is a must, that new or mores sensitive analysis and detection methods have to be found to be able to better evaluate embryos So far we have the opportunity to evaluate embryos morphologically, count the number of nuclei of various embryonal stages, differentiate between inner cell mass and trophectoderm by differential staining or analyse the expression ofa variety of genes But still, all of these methods are not sufficient to predict if embryos will be capable of developing into viable fetuses and subsequently result in live, healthy offspring.Production of transgenic animals Transgenic farm-animals have been produced since 1985 (BREM et al., 1985;HAMMER et al., 1985) by microinjeetion into the pronuclei of zygotes, however, the efficiency remained low.Somatic NT using transgenic donor cells also prevents the incidence of mosaicism and -when using homologous recombination -of insertional mutations or problems of varying gene-expression due to position effects.The main interests in generating transgenic animals lies in the production of pharmaceuticals or other proteins of interest, supply of organs for xenotransplantation or modifications that improve features that are of economic interest (reviewed by FRENCH et al., 1998;BRINK et al., 2000;PIEDRAHITA, 2000;WOLF et al., 2000).There are many reports about in vitro development of NT-embryos derived from transfected donor-cells.In contrast, reports of transgenic NT-offspring are limited.As reviewed in this manuscript, lots of different cell-types resulted in the birth of offspring after somatic NT, however, fetal fibroblasts seem to be the cell-type of choice for the production of transgenic NT-animals with the exception of one report using transfected embryonic germ cells (Table 2).The culture period of donor-cells gives the opportunity of transfection and subsequent analysis prior to use in nuclear transfer.There is also the possibility to get multiple identical offspring from (male) transgenic founder animals by breeding females and recover their fetuses.From such fetuses one can establish fetal cell lines and use them (after analysis to confirm the presence of the transgene) as donors for NT.The first report using this system was published by BAGUISI et al. (1999).They produced transgenic goats from fetal fibroblasts that were derived from a transgenic fetus, carrying the recombinant human antithrombin III (rhAT).This method has the disadvantage that one does not want to sacrifice a female founder to obtain fetal tissue, in addition to the requirement of existing transgenic animals to begin with.The in vitro transfection of an existing cell line is a more advanced technology.Although successful modification of cells with subsequent selection results in cells of a higher passage number and more generation intervals, there are still reports of live transgenic animals that were derived from donor cells, that were transfected in vitro.Again the first animal produced by somatic nuclear transfer using transfected fetal fibroblasts was a sheep.SCHNIEKE et al. (1997) transfected fetal fibroblasts with the gene for the human blood-clotting factor IX (FIX) and subsequently used these cells for NT.The lambs that were generated in this study not only integrated, but also expressed the transgene.CIBELLI et al. (1998) reported the birth of the first transgenic calves that were produced by somatic nuclear transfer.They used a marker gene (ß-galactosidaseneomyein resistance, ß-GEO) to prove the feasibility of this technique in cattle.After these initial experiments further studies followed that produced a transgenic goat (enhanced green fluorescent protein -eGFP: KEEFER et al., 2001) and a transgenic calf (p77 -bovine proehymosin: ZAKHARTCHENKO et al, submitted) derived from transfected cells.In contrast to reports in other species, this transgenic goat did not express the transgene.Since the transfection method used in the reports mentioned above resulted in random Integration and a viable number of copies that were integrated, the use of homologous recombination for the production of transgenic animals is of great interest Homologous recombination or gene targeting require long-term culture and result in a high number of population doublings and were therefore only successfuUy performed in mural embryonic stem (ES) cells.ES cells are not yet available in livestock.Thus scientists tned to modify fetal fibroblasts by homologous recombination.Again the first success was reported in sheep: McCREATH et al. (2000a, b) were able to establish efficient and reproducible gene targeting in fetal fibroblasts.A therapeutic transgene (human cd-antitrypsin, ATT) was successfuUy inserted at the ovine al(I) procollagen locus.These gene-targeted cells were subsequently used in NTexpenments that resulted in the birth of 12 lambs, of which three survived.Table 2 summarizes the different techniques that resulted in the birth of transgenic NToffspring.Future perspectives Unfortunately, and regardless ofthe big aecomplishments in the last couple of years, it cannot be ignored, that there are still many problems in connection with nuclear transfer (HILL et al., 1999a, b;RENARD et al., 1999;HILL et al, 2000b;NUMABE et al, 2000).Not only a high percentage of pregnancies are lost but also a variety of abnormalities are found in NT-offspring.Before the technology reaches a State, where its use will be practical and cost-efficient, significant improvements in maturation, donor-cell preparation, activation and in vitro Systems are required.It is essential, that protocols for defined in vitro Systems will be developed and/or improved, so' that variable factors can be minimized.New studies on gene expression (MEMILI and FIRST, 1999;DeSOUSA et al 1999-LAURINCIK et al, 2000;YOUNG et al, 2001) are a start in expanding the knowledge and understanding of reprogramming and subsequent events that happen during and after somatic cell nuclear transfer.