• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Animal & Veterinary

  • Print
  • Share
  • E-mail

Section Contents Menu

Safety & Health

Chapter VI: Food Consumption Risks

A. Potential Hazards and Food Consumption Risks Associated with Food Products from Animal Clones and their Progeny

1. Assumptions

This Chapter of the Risk Assessment is focused on food safety concerns, and assumes that any clones or their products would be subject to the same local, state, and federal laws and regulations as conventional food animals or their products. These assumptions exclude animals with gross anomalies that would not enter the human food supply (although they might be rendered). It also assumes that any hazards arising from the consumption of products derived from animal clones would result from epigenetic dysregulation of the genome of the developing animal, as described in Chapter IV.

The focus of this analysis is the identification of potential subtle hazards in otherwise healthy-appearing animals. Using the Critical Biological Systems Approach (CBSA), we evaluated animal health data on as fine a level of resolution as possible. This includes individual animals or even individual analytes per animal in order to have a sensitive screen for adverse outcomes (and thus food consumption risks). Thus, although some of the data in this chapter reprises information previously addressed in Chapters IV and V, the methods by which the data were evaluated differed. Although Chapter V addressed risks to the health of animals involved in the cloning process, the focus of Chapter VI is to identify whether any adverse outcomes observed in clones provide insight into the identification of food consumption hazards. The emphasis here is not on macroscopic adverse outcomes observed in clones and animals produced using other ARTs, but on the search for potential subtle food consumption hazards, which can be thought of as alterations in the physiology of normal and healthy-appearing clones that may indicate food consumption risks. A second step involves determining whether any of the adverse outcomes noted differ from those identified in conventionally bred animals to determine whether cloning poses any unique food consumption risks.

The search for these subtle hazards requires analysis of physiological parameters in blood and tissues (e.g., clinical chemistry, hematological measurements, hormone levels). The working assumption is that these subtle hazards, which are likely the result of inappropriate or incomplete genetic reprogramming, could lead to alterations in the expression of key proteins that could affect the nutritional content of food.

Chapter VI also describes the Compositional Analysis (See Chapter III), which includes all of the information that we could identify on the composition of meat or milk from clones or their progeny. Much of this information has been published or made available since 2005, and tends to evaluate very similar compositional components; much of it is on animals for which physiological data are also available.

2. Critical Biological Systems Approach to Clones of Cattle, Swine, Sheep, and Goats

Chapter V and VI review the health outcomes reported for clones of cattle, swine, sheep, and goats. Over 2,000 references were identified in our literature searches; closer examination revealed that approximately 400-450 papers were useful to the understanding of the subject, and a smaller fraction of those actually cited papers were cited for information on the health of clones or the composition of their food products. Many of these reports are on the same cohorts of animals, but concentrate on different measurements or life stages. Several are reviews of adverse outcomes that have been observed in individual animals or cohorts of animals, but do not provide new data. As indicated previously within the Risk Assessment and detailed in Appendix D, some of the animals on which reports are provided are somatic cell nuclear transfers of transgenic cells, thereby actually being reports on transgenic animal clones. These have been included in the food consumption risk assessment when they provide corroborative information, and the transgenic status of the animals has been indicated when that information is available.

The following section reviews the available information on animal cloning by species, sorting the information into developmental node-specific groupings. This approach was most applicable to bovine clones, where there is significantly more information compared to other species. For those species where information is very limited, such as sheep, the available information is presented as a single unit.

a. Bovine Clones

The largest number of publicly available publications and data sources address clones of dairy and beef cattle. Many reports on effects noted in the Cell Fusion/Reprogramming, Embryo/Fetal, and Perinatal periods tend to come from the early cloning experiments. Others test hypotheses regarding some component of the SCNT process (e.g., cell cycle, cell source, culture conditions, epigenetic reprogramming (see Chapter IV)), and either do not result in live births, or result in very few live births. Very few systematically evaluate the health of the animals, many simply state that “animals appear normal and healthy” or that “no differences were observed between clones and controls.” CVM has extracted as much information as possible from these studies, and has incorporated its findings into the appropriate Developmental Nodes. Some of these papers, particularly those addressing methodological developments, do not provide information directly applicable to the risk assessment; these have not been cited.

During the course of preparing this risk assessment, clone producers shared information on various cloning outcomes with CVM. The most comprehensive dataset was generated in response to preliminary presentations of the risk assessment methodology by FDA staff at various scientific meetings. In particular, one clone producer, Cyagra, Inc., attempted to gather information on all of the cattle clones that it had produced, including animals that did not survive or that were culled for various reasons. In some cases, this has proved impracticable due to the dispersal of clones to their ultimate owners. The Cyagra dataset is the most comprehensive survey of the health status of cattle clones that has been assembled, and this information has been incorporated into this Risk Assessment. Many of these data have been compiled with data on other clones produced by Cyagra in different parts of the world, and have been published in summary form in a peer-reviewed publication (Panarace et al. 2007), discussed previously in Chapter V, and at the relevant points in this chapter. Details on the animals, the methods used to collect and interpret the data, and the actual data themselves can be found in Appendix E. Cyagra also collected data on the composition of meat from several clones; these data are also in the Appendix.80

The information provided by Cyagra differs from that presented in the peer-reviewed literature for several reasons:

  • The data were collected specifically to address issues raised in this risk assessment, and thus are not part of a hypothesis-testing study, or written to provide examples of novel or unusual events;
  • They have not been peer-reviewed outside CVM (to the Center’s knowledge);
  • They include individual animal data; and
  • They are far more extensive with respect to the number of clearly non-transgenic animals evaluated (n=78 surviving and tracked animals), and the number of observations on individual animals than any other study or series of studies from a particular laboratory.

i. Cell Fusion, Nuclear Reprogramming, and Embryonic and Fetal Development in Bovine Clones81 (Developmental Node 1)

SCNT is a relatively inefficient process. “Successful” event estimates can be based on the number of fused cells, implanted blastocysts, or pregnancies confirmed at a specific day of gestation, and range from one in one thousand (usually based on fused cells) to one in four (confirmed pregnancy at gestation day 60). The former estimates include the earliest reports of SCNT, as well as studies testing various methodological variables, and reflect the “technology development” nature of the reports. Cloning efficiency in cattle, defined as the number transferred embryos resulting in live calves, has been estimated as 9-13 percent (Wells 2005, Panarace et al. 2007). In a large, 5-year study of bovine clones, Panarace et al. (2007) noted that cloning efficiency can vary from 0 to 45 percent depending on the cell line used for SCNT. When measured from the detection of an established pregnancy in the surrogate dam, the success rate can range from 1-2 percent (as reviewed in NAS 2002b) to approximately 20-25 percent as related to CVM by commercial cloning ventures.

Lack of success at the cell fusion stage is likely due to several factors, the most significant of which are technological (e.g., damage to the oöcyte or donor cells) or biological (e.g., incorrect reprogramming of the genome of the donor cells (Chapter IV) or possible lack of synchrony between donor cell and oöcyte). An alternative justification proposed by Hochedlinger and Jaenisch (2002) among others, is that the extremely low success frequency is a reflection of the inability of all but “stem cells” of various degrees of pluripotency to be reprogrammed, and the serendipitous outgrowth of such cells selected at random for use as donor cells. Regardless of the explanation, few fused donor/oöcyte pairs survive to divide or to become established as pregnancies in surrogate dams.

The following overview of methods that may affect success rates of SCNT are included to allow the reader to understand that there are many different components that may influence cloning efficiency. It is important to remember, however, that the goal of this chapter of the risk assessment is to identify and characterize potential subtle hazards in clones and to determine whether they pose food consumption risks.

(a) Peer-reviewed Publications

The following section provides summaries of studies that contribute to identifying some of the factors that may contribute to successful nuclear transfer at the earliest developmental node. It is intended to be illustrative, and not comprehensive.

Effect of the Zona Pellucida. The importance of the zona pellucida in embryo development is not clear, and there are conflicting outcomes in different studies evaluating its role. Dinnyes et al. (2000) compared developmental rates of cattle oöcytes subjected to SCNT, parthenogenetic activation, or in vitro fertilization. For the oöcytes undergoing SCNT (n=106), 74 percent fused, 90 percent of fused embryos cleaved by Day 2, and 29 percent of cleaved embryos developed to blastocysts. Eighty-one percent of parthenotes82 (early embryos arising from parthenogenetic activation) (n=47) incubated in 5 percent CO2 in air cleaved by Day 2 of the experiment, but only 17 percent developed into blastocysts. Parthenotes (n=98) incubated in 5 percent O2, 5 percent CO2 and 90 percent N2 had a 79 percent cleavage rate on Day 2, and a 32 percent survival to blastocyst stage. By comparison, in vitro fertilized oöcytes (n=98) had a 69 percent cleavage rate by Day 2, and 35 percent developed to blastocysts. Because parthenotes are “clones” that have not undergone nuclear transfer, the zona pellucida of the embryo is not disrupted. This disruption has been hypothesized to be a possible cause of early embryo failure in nuclear transfer (NT) embryos. The lack of difference in development to blastocyst between SCNT, parthenotes and IVF embryos cultured under the same conditions suggests that disruption of the zona pellucida may not be an important factor in early loss of SCNT embryos. Conversely, Ribas et al. (2006) noted no difference in development to blastocyst in zona-free vs. zona-intact IVF mouse embryos, although the authors stated that zona-free blastocysts were smaller and more irregular than zona-intact embryos. None of the embryos in this study were transferred to recipients for gestation, however, so further development could not be assessed. In another study involving IVF-derived sheep embryos, Ritchie et al. (2005) transferred eight zona-free embryos to four surrogate ewes. One of these pregnancies progressed to term and resulted in a live lamb.

Cell Culture Conditions. Several laboratories have attempted to optimize culture conditions to improve cloning efficiency (Kubota et al. 2000; Li GP et al. 2004; Park ES et al. 2004b; Du et al. 2005). These manipulations have included addition of various compounds to culture media, co-culture with “feeder cells,” and serum starvation. Results of these studies have been mixed, as described below.

In order to study the influence of culture conditions of donor cells used for SCNT, Kubota et al. (2000) used fibroblasts derived from a skin biopsy obtained from a 17 year old Japanese Black beef bull. Donor cells for nuclear transfer were obtained from cultures that had undergone 5 (n=570), 10 (n = 269), or 15 (n = 264) passages.83 All cultures were serum starved prior to nuclear transfer, except that cells from passage number five were divided into two groups, one of which was serum-starved (n=288), and the other was not (n=282). There were no differences among groups for fusion or cleavage rates, but development to blastocyst stage was lower in cells from Passage 5, relative to cells from the higher passage rates, regardless of whether or not the cells were serum starved. A total of 54 blastocysts were transferred to 36 recipient cows. Fifteen cows were diagnosed pregnant, of which nine spontaneously aborted between 39 and 123 days of pregnancy. All three of the pregnancies derived from Passage 5 cell cultures spontaneously aborted. Six calves derived from the two more extensively passaged cultures were delivered at term, two from cultures that had undergone 15 passages, the other four from cells that had undergone 10 passages. Two calves derived from Passage 10 donor cells died shortly after birth. In this study it appears that cells that have been more extensively passaged make better donors than less extensively passaged cells. The biological basis for this is not clear, unless cells that have been passaged more extensively in culture somehow become more amenable to epigenetic reprogramming.

In another study of culture conditions, Li GP et al. (2004) compared development of SCNT embryos co-cultured with bovine cumulus cells or with one of two different types of serum (fetal calf serum (FCS) or bovine serum albumin (BSA)) for seven days. The rates of cleavage, morula and blastocyst formation were similar across treatment groups. Fewer blastocysts in the FCS group exhibited normal chromosomal ploidy compared to the BSA group (24/41 or 58.5 percent vs. 24/35 or 68.6 percent), but both of the serum supplemented groups performed poorly compared to the cumulus cell co-culture group, in which 34/42 (80.9 percent) of blastocysts had normal ploidy.

Park ES et al. (2004b) noted that although not effective in improving embryo development alone, the combination of ß-mercaptoethanol (ME) and hemoglobin (Hb) enhanced the rate of development of NT embryos to the morula stage compared to unsupplemented media (19/57 vs. 55/85). Development to blastocyst, however, was similar between untreated controls and either the combined treatments or ME or Hb supplementation alone (16/57 vs. 18/99, 15/95, and 40/104 for control, Hb, Me and Hb + ME, respectively). Similarly, Du et al. (2005) found no beneficial effect of adding phytohemagglutinin-L (PHA) to culture media for survival, cleavage or blastocyst formation of NT embryos. From a total of 324 fused embryos, three live calves were born: two from the PHA group and one from the untreated group.

Heterogeneity of Fusion Components. Hiendleder et al. (2004b) studied how differences between nuclear and oöplasm sources can influence SCNT outcomes by using three breeds of cattle (Brown Swiss, Dwarf Zebu, and two varieties of Simmental) as oöcyte sources and granulosa cells from a Brown Swiss cow as the source for somatic cells. Four groups of SCNT embryos were produced. All pregnancies were terminated at 80 days gestation and uterine contents collected to determine the number of viable fetuses. Details on individual fetuses were not discussed, but the authors noted that SCNT fetuses in general were heavier, had a larger thorax circumference, and a reduced crown rump length: thorax ratio (a standard measure of body size) compared to AI fetuses. The proportion of viable fetuses was significantly affected by source of oöplasm, and was higher for fetuses produced using Dwarf Zebu oöplasts than the other three sources. The lowest viability was noted for one, but not both, of the Simmental sources. Interestingly, the difference between the two Simmental sources for viability was significantly different. No details regarding the oöcyte donors, other than breed, were provided, so there is no way to determine if other factors (e.g., age of the oöcyte donor cows, nutritional status, health history, or size of follicles collected) might have influenced fetal viability. The authors also compared mitochondrial DNA sequences between the two Simmental oöcyte sources, and noted extensive polymorphism in coding and non-coding regions of the two mitochondrial genomes. Although there has been speculation that mitochondrial dimorphism may affect development of SCNT embryos, only one study was identified that looked specifically at mitochondrial effects on embryo development (Takeda et al. 2005). Also of interest, when fetal morphology was compared in the Hiendleder et al. study, hybrid fetuses (reconstructed using either Zebu or Simmental oöplasm) were not significantly different in size compared to AI fetuses of the same gestational age; however, fetuses produced using the same breed as source of both oöplasm and nucleus (Brown Swiss) exhibited fetal overgrowth. The Brown Swiss cows that were used as sources of oöcytes were different individuals from the Brown Swiss donor of the nuclear DNA. The authors do not report whether they compared mitochondrial DNA of the nuclear donor with that of any of the Brown Swiss oöcyte donors.

Source of Donor or Recipient Cells. Tissue source of nuclear donor cells can also affect development and survival of NT embryos. Galli et al. (1999) used bovine blood lymphocytes as nuclear donors. Lymphocytes, involved in the immune system, must undergo rearrangement of their DNA in order to produce immunoglobulins. Panelli et al. (2004) examined tissues of four aborted NT fetuses and the chondrocytes of the single surviving clone from the Galli et al. experiments. The results were compared to chondrocytes from three non-clone bulls (how the comparator bulls were generated is not described). The aborted fetuses exhibited DNA rearrangement in brain cells that was typical of terminally differentiated lymphocytes, but the surviving clone showed no rearrangement in chondrocytes isolated from his sperm, similar to chondrocytes collected from non-clone bulls. Based on this small dataset, the authors suggested that although terminally differentiated cells can sustain development through the late fetal stage, cells more amenable to reprogramming (dedifferentiation), such as stem cells, were more likely to result in live clones.

Xue et al. (2002) reported on the relative success rates associated with generating clone embryos from three different tissues collected from a 13 year old Holstein cow. In their hands, ovarian cumulus cells had the highest rate of development to blastocyst (57 percent, n=92), compared to skin fibroblast cells (34 percent, n=110) and mammary epithelial cells (23 percent, n=96). Six term pregnancies resulted following transfer of ovarian cumulus nuclear transfer (NT) embryos to recipient cows (5.5 percent, n=109), and four (7 percent, n=57) term pregnancies resulted from skin fibroblast NT embryos. None of the embryos generated from mammary epithelial cells resulted in a term pregnancy when transferred to recipient cows (n=34). The expression of X-chromosome linked genes in various tissues from deceased animals and conventional controls, and from the placentae of surviving clones was also investigated. Results indicated that X-chromosome inactivation occurred normally in the surviving female clones, but was incomplete in the clones that died. Embryo samples were taken to determine if there were differences in cell counts in embryos from parthenotes and SCNT-derived embryos at the same stage of development. Cell numbers for NT embryos were lower compared to parthenotes at all stages examined (Day 5 morula: 35.1 ± 1.1, n=48 for NT vs. 43.5 ± 1.5, n=58 for parthenotes; Day 7 blastocyst: 81.0 ± 3.7, n=46 for NT vs. 93.8 ± 5.6, n=48 for parthenotes). The importance of differences in cell numbers is not clear from this study, as mammalian parthenotes generally do not develop to term. Cell counts of IVF embryos, which would have been a more informative comparison, were not provided.

Gong et al. (2004b) compared granulosa cells from adult cattle of two different breeds (Holstein and Chinese red-breed yellow cattle), skin fibroblasts from two individual Holsteins and a Holstein fetus, and oviductal cells from a Holstein fetus for development and survival through the birth of clones. The rate of blastocyst formation was lowest for one of the two adult skin fibroblast sources (253/906 blastocysts/fused couplets or 27.9 percent), although the other adult fibroblast cell line was comparable to the fetal fibroblast cell line (52/132 or 39.4 percent vs. 1294/3412 or 37.9 percent). Fetal oviductal cells had the highest rate of blastocyst formation in this experiment (456/1098 or 41.5 percent). A total of 346 Day 7 blastocysts were transferred to 171 recipients. Pregnancy rate at day 60 was 34.5 percent (59/171), with 25 surrogates carrying 27 calves to term. Because of the small numbers of calves delivered at term, no differences could be detected among donor cell sources for live birth. Of the 27 calves born, eight died during the perinatal period, and another seven died at later stages. Seven of the calves died of causes associated with LOS (hepatic, cardiac, or gastro-intestinal defects, respiratory distress), and eight animals apparently died due to management errors. It is not clear what portion of the perinatal deaths were due to birth defects/respiratory failure or management errors. Birth weights of calves were not reported.

Some authors have suggested that the stage of the cell cycle may also influence cloning outcomes. However, results in different laboratories (Wells et al. 2003b; Urakawa et al. 2004; Ideta et al. 2005) using cells in different stages have been mixed. Wells et al. (2003b) compared putative G0 cells (cells that apparently were not dividing) to G1 phase (cells that had begun dividing) cells for SCNT. They noted high early pregnancy losses, but no losses after 120 days of gestation, and no reported hydrops in the G0 group. In contrast, G1 phase cells had higher losses to term (21/43 pregnancies lost after 120 days gestation) and higher incidence of hydrops (18/43 (42 percent) of pregnancies), but higher post natal survival than clones from G0 cells. In contrast to the Wells et al. study, Urakawa et al. (2004) reported success using fetal fibroblast donor cells in the G1 phase. However, it should be noted that Urakawa et al. used only G1 phase cells, and did not compare to other stages of development. Two cell lines were used, derived from fetuses with the same dam but two different bulls. Ten blastocysts were transferred into ten recipients, resulting in nine live calves. According to the authors, calving was “uneventful.” Differences were noted between cell lines, in that three calves resulting from one of the lines tended to be heavier at birth than the six calves of the other cell line used (actual birth weights not provided). One of these three heavy-weight calves died after two days without standing. The authors do not report on the health or survival of the remaining eight calves beyond the first six days of life. Ideta et al. (2005) compared development of embryos constructed with G1 or M phase fetal fibroblasts, and noted that G1 SCNT embryos had higher rates of development to blastocyst than M phase cells (31 vs. 16 percent). Only five surrogate cows received embryos in the Ideta et al. study, of which three were diagnosed pregnant on day 30 of gestation, and one live calf was delivered. All of the transferred embryos were developed from G1-phase somatic cells. The single calf died two days after birth. Health of the surrogate dams, method of delivery, and birth weight of the single calf were not reported in this study. Based on these studies, two of which used only embryos developed from G1 phase cells, it is not possible at this time to determine the influence of the stage of the donor cell cycle on subsequent development of the embryo/fetus.

It has been proposed that cellular quiescence facilitates reprogramming of the somatic cell nucleus (Wilmut et al. 1997; Wilmut and Campbell 1998) and thus may increase the efficiency of SCNT. To test this postulate, Lawrence et al. (2005) used serum starvation to induce quiescence in bovine granulosa cells prior to their use as donor cells for SCNT. No developmental differences were found in vitro between embryos derived from serum-starved and serum-fed cells. More heifers receiving clones from serum-starved cells were confirmed pregnant (9/13, 67 percent vs. 11/25, 44 percent), but embryonic loss between days 29 and 50 of pregnancy was greater in the serum-starved group (88 percent vs. 36 percent). In this study, only one fetus from the serum-fed group survived to term. 84Most of the remaining fetuses were lost due to complications from hydroallantois, and two died = 6 d after Caesarian delivery (C-section). The authors concluded that the use of serum-fed granulosa cells was associated with a high incidence of losses during the third trimester due to hydroallantois and fetal overgrowth.

Embryo and Fetal Development. Early pregnancy failures in bovine clones are thought to be a function of incorrect reprogramming of the donor cell that manifest as lethal developmental defects (see Chapter IV). Some of those developmental defects may manifest as difficulties in placentation. For example, Hill et al. (2000b) noted that placentae from gestation day 40-50 clone embryos were hypoplastic (low cell density), and had poorly developed cotyledons. (In ruminants, the cotyledon is the fetal part of the junction between the maternal and fetal sides of the placenta where nutrients and wastes are exchanged.) Additional placental anomalies in first trimester aborted fetal clones may include decreased numbers of placentomes (the junction of maternal and fetal components of the ruminant placenta that serve to transport nutrients into and waste out of the fetal environment), and poor formation of blood vessels in the placenta. In contrast, Lee RS et al. (2004) noted that although fewer cotyledons were present in SCNT placentae compared to AI and IVF placentae at day 50 of gestation, vascularization was very good, and appeared more developed in SCNT compared to AI or IVF placentae. Edwards et al. (2003) also studied this phenomenon in transgenic and non-transgenic bovine clones and observed that approximately 50 percent of transferred embryo clones established a pregnancy when measured by the presence of a heart beat between gestational days 29-32. This rate was compared favorably to that observed for non-clone IVF embryos. Edwards et al. (2003) noted that 50-100 percent of embryo clones spontaneously aborted between 30–60 days of pregnancy. Dindot et al. (2004) have noted more than 80 percent of hybrid bovine clone pregnancies (Bos gaurus X Bos taurus) were lost between gestational days 30 and 60. Evaluation of the early placental structures at gestational day 40 indicated an absence of cotyledons in each clone pregnancy, while the control (AI) fetuses had between 4 and 25 cotyledons per pregnancy). Pace et al. (2002), in a study that included transgenic clones, noted that the fetal abortion rate prior to gestational day 60 was 67 percent. A comparison of the crown-rump length of calved and aborted clone fetuses with AI-generated fetuses from gestational day 25 to gestational day 70 indicated that prior to abortion fetuses grew at the same rate.

Later pregnancy failures are thought to be a function of developmental defects, including placentation abnormalities. Heyman et al. (2002a), for example, compared pregnancy loss between gestation day 90 and calving among clones derived from adult somatic cells, fetal somatic cells, blastomere nuclear transfer (BNT), and in vitro fertilization (IVF) animals. They noted that the somatic cell clones showed a pregnancy loss incidence of approximately 44 percent and 33 percent, while BNT clones were lost in only 4 percent of the pregnancies, and the IVF control group lost no pregnancies.

Abnormal placentation can, however, result in the birth of a viable clone (Hill et al. 2000b). In this case, one of six transgenic fetal clones detected at 40 days of gestation continued to develop to term, and when delivered vaginally weighed 37.7 kg, within the normal weight range for Holstein calves (35 to 45 kg). The calf was considered normal based on physical examination at birth. It suckled normally, and at the time of publication, was two years old and considered to be normal. The placenta was similar in weight for term Holsteins (4.3 kg vs. mean expected weight of 5.6 kg). Its structure, however, was highly abnormal, with only 26 cotyledons present, of which only 12 were judged to have been functional. These were enlarged, and the authors hypothesized that this increased size allowed the normal development of the calf. The authors also note that pregnancies resulting from IVF have also been reported to contain fewer placentomes85 than those of conventional cattle. As discussed in Chapter III, the role of transgenesis in the development of this pregnancy cannot be determined. Batchelder (2005), however, working with non-transgenic clones, also noted fewer and larger placentomes in placentae of eight live-born clones compared to AI and ET comparators.

To gain a better understanding about the relationships between fetal and placental growth in bovine clone pregnancies complicated by hydroallantois, Constant et al. (2006) examined placentae from pregnant cows between days 180 and 280 of gestation. Placentae were obtained from 10 AI pregnancies, 6 IVF pregnancies, and 18 SCNT pregnancies. All surrogate dams bearing SCNT-derived pregnancies in this study were diagnosed with hydroallantois and were humanely slaughtered within 1-2 weeks of diagnosis. No differences in morphology or stereology were found between AI and IVF pregnancies, so these groups were combined into a single control group for purposes of statistical evaluation. One of the key qualitative findings of this study is that placental overgrowth in clone pregnancies preceded fetal overgrowth. Before day 220, fetal weights were similar in SCNT and control pregnancies despite evidence of abnormal placental development in SCNT pregnancies (fewer placentomes and higher mean placentome weight). After day 220, fetal weights were significantly higher in SCNT pregnancies and the ratio of fetal weight to total placentome weight was decreased compared to controls. No major histological abnormalities were observed in SCNT placentomes, but after day 220, growth of the fetal component of the placenta, particularly fetal connective tissue, was favored over the maternal component. Based on these findings, the authors suggest that placental overgrowth in SCNT pregnancies is due to factors inherent to the placenta and is not simply driven by fetal overgrowth. The authors also suggest that the term “large placenta syndrome” may be better than “large offspring syndrome” to describe the complications commonly associated with SCNT pregnancies.

The molecular mechanisms underlying placental overgrowth in SCNT pregnancies are unclear at this time. Based on studies in the mouse, it has been suggested that placental overgrowth associated with nuclear transfer may be caused by inappropriate epigenetic reprogramming (DNA methylation), leading to dysregulated expression of genes that regulate placental development. This hypothesis is consistent with the suggestion of Constant et al. (2006) that placental overgrowth associated with SCNT in cattle is mediated by placenta-specific factors.

(b) Summary for the Embryonic/Fetal Developmental Node in Bovine Clones (Developmental Node 1)

This period manifests the highest degree of risk for the developing clone. The probability of an SCNT-embryo implanting, and the subsequent likelihood of an implanted clone embryo surviving and continuing to develop appropriately are low. Various investigators have attempted to understand the role of various components of the donor/recipient/cell culture system that comprises the “cloning unit” to improve efficiency with different sources of nuclear or oöcyte donors or by manipulating the culture conditions. These studies have been met with mixed results. Lack of success can be attributed to failure of the genome to be reprogrammed (Chapter IV), including failure of the embryo to begin dividing and implant in the uterus, and failure of development in the first trimester (likely due to defects in reprogramming that manifest as poor placentation or other defects that do not allow the fetus to develop), or physical damage to the early embryo. Some may be due to the types of cells used as donors for nuclear transfer. Difficulties that may persist in later pregnancy are largely associated with placentation anomalies that may co-develop with Large Offspring Syndrome (LOS) (see Chapter V). Nonetheless, some of these early embryos do divide, implant, develop, and give rise to live animals, as discussed in the subsequent Developmental Nodes.

ii. Perinatal Development in Bovine Clones (Developmental Node 2)

In the early studies of the technology, relatively high perinatal losses were reported. Deaths generally resulted from phenomena associated with LOS, including poor development of the respiratory and cardiovascular systems. (For a more complete description, refer to Chapter V.) In general, animals with LOS tend to have high birth weights (ranging from 20-50 percent greater than breed averages), poorly developed and sometimes edematous (fluid-filled) lungs and other tissues, and heart malformations and malfunctions. These animals may also have kidney and liver anomalies, and may initially exhibit difficulties in maintaining homeostatic functions such as body temperature and glucose metabolism. The latter are discussed in more detail later in this section. As the expertise develops, however, more animals are either born with no apparent defects, or have perinatal supportive care and survive to grow into healthy cattle.

(a) Peer-reviewed Publications

Most of the adverse outcomes that have been reported result in loss of the fetus before birth, although there is another period of loss after delivery, usually within the first few days of life. Reproducible sets of adverse outcomes have been observed, including LOS and gross morphological abnormalities that may result in pregnancy loss either early in gestation or late in gestation. For example, contracture of tendons has been noted in some clones. None of the abnormalities noted in animal clones are unique to animals derived by SCNT; all have been observed in natural reproduction, as well as in ARTs such as AI and IVF (reviewed by Cibelli et al. 2002, Pace et al. 2002, Farin et al. 2006, and in Chapter V).

Despite the initial frequency of publications describing adverse outcomes of SCNT, two classes of successful outcomes actually predominate at birth. The first includes animals that may require assistance with delivery and immediate post-natal support in maintaining oxygenation and body temperature. Among others, Cibelli et al. (2002) noted that adverse effects associated with abnormal placental functions in the birth of a group of transgenic clones can be mitigated by intensive veterinary care immediately following birth. One bull clone described by Hill et al. (2000a) required considerable veterinary support immediately after birth due to respiratory problems (immature lungs and pulmonary hypertension), lack of suckling reflex, apparent Type I diabetes, and other health problems. According to this report, the calf improved rapidly, and the diabetes resolved (the calf was able to maintain normal blood glucose and insulin levels) by two months of age. This animal has fully recovered, and is reported to be a vigorous and healthy bull (PIFB 2003).

The second set of successful outcomes consists of those animals born with relatively little assistance (due to the high cost of developing animal clones, most are delivered via planned C section, and may require more supportive care than animals derived from more conventional breeding techniques), and appear to be normal and healthy (see especially the Cyagra database (Appendix E)). Although many reviews attribute the difference in birth weight to various degrees of LOS, higher birth weights may also be due to the greater care afforded surrogate dams carrying animal clones relative to standard husbandry of conventional animals. Alternatively, birth weight may be related to genetics of the nuclear donor. No data were found on birth weights of nuclear donors, but studies indicate that birth weight is heritable (Knight et al. 2001; Chapter V).

Forsberg et al. (2002) reported the production of 103 cattle clones, of which 47 were produced from non-transgenic cells and 56 from transgenic cells, including a Holstein bull calf generated by recloning an embryo derived from genital ridge cells. Of five pregnancies initiated from that recloning, two aborted prior to gestational day 30, one pregnancy was terminated at gestational day 203 due to hydrops, one set of twins died at birth due to the surrogate dam’s ketosis, and the fifth gave rise to “Gene,” the first cattle clone not produced from an embryonic cell line.86 Little further information on Gene’s birth status, growth, or development is found in the peer-reviewed literature, except that as of the end of 2001, when the Forsberg et al. manuscript was accepted for publication, Gene had matured into “a healthy, fertile bull.” In a separate recloning trial described in this report, fibroblast cell lines derived from another fetal clone were used as donors to generate 28 blastocysts that were then transferred into 14 surrogate dams. Nine pregnancies were initiated. Four of those pregnancies went to term, and five calves (three singletons and one set of twins) were produced.

Forsberg et al. (2002) also used cells from adult animals as donors for SCNT. Ear cells from a bull (age and breed not specified) were used to generate 32 embryo clones that were transferred into 17 surrogate dams, of which 10 became pregnant. Five pregnancies were lost prior to gestational day 60, and two more were terminated due to hydramnios or hydroallantois (these conditions are also referred to as hydrops). Three live animals were born, but one was euthanized at 11 days of age due to a heart defect. In a separate trial described in the same paper, cumulus cells from an in vivo matured oöcyte from a 17 year old cow were used to initiate 11 pregnancies, from which three calves were born. Although information on the health status of many of these animals is not available, 15 of these animals were bred, gave birth, and their milk studied by Walsh et al. (2003) (See Compositional Analysis Method - Section 3).

In addition, Pace et al. (2002) of the same group reported on the development of 117 cattle clones from the reconstructed embryo stage through to lactation. These animals were born between January 1998 and February of 2000. Some of the cell lines from which these animals were developed were transgenic (Forsberg et al. 2002), and 75 percent of the resulting clones were transgenic. Because this report does not distinguish individual animals by cell source, it is not possible to determine which of the animals are transgenic. Interpretation of adverse outcomes should therefore be considered within the context of the discussion of transgenic animals in Appendix D. Of the 117 clone births, 106 were born alive, and 82 remained alive at the time of publication. Birth weights of the surviving clones ranged from 11-72 kg, with an average birth weight of 51 ?± 14 kg. The distribution of birth weights was skewed in excess of birth weight ranges for conventional Holsteins.

Pace and his colleagues (2002) divided the calf clone deaths into preventable and non-preventable causes (summarized in Table VI-1). Of the 24 animals that did not survive, 12 died between post partum days 1-5, nine died between days 6-122, and three died at more than 123 days of age. Many of the animals appear to have experienced complications resulting from enlarged umbilici, and three of the deaths were directly related to this condition. For subsequent births, this condition was managed by prophylactically tying or clamping off the umbilical arteries. Difficulties with the umbilicus were also observed at levels apparently higher than in conventional animals by Kishi et al. (2000); Gibbons et al. (2002); Cyagra (Appendix E); Edwards et al. (2003); and Batchelder (2005).


Table VI-1: Summary of Causes of Death of Calf Clones
(adapted from Pace et al. 2002)
Non- Preventable Deaths
Physiological System Involved Calves (n) Age at death (days) Birth Weight (kg) Observations
Multiple dysfunctions 3 1-2 11-63 Failure of most systemic functions
Placental 2 1 50-59 Apparent premature separation of placenta
Respiratory 1 3 62 Lung immaturity, meconium aspiration at birth
Digestive 2 78-122 52-60 Chronic diarrhea (n=1); Intussusception of small intestine with obstruction (n=1)1
Circulatory 1 42 52 Congenital heart defect
Nervous 1 154 51 Hydrocephalus
Musculoskeletal 1 298 44 Developmental orthopedic disease
Preventable Deaths
Physiological System Involved Calves (n) Age at death (days) Birth Weight (kg) Observations
Placental 3 1 53-69 Extensive internal bleeding from enlarged umbilicus
Respiratory 3 1-5 48-66 Developed pneumonia (n=2); Premature induction of labor 16 days early, immature lungs (n=1)
Digestive 5 5-90 59-72 Clostridial infection (n=1); Developed abomasal ulcers2 from eating wood chips (n=2); Bloat (n=2)
Musculoskeleta 1 328 42 Injury, dislocation of patella
Urinary 1 112 59 Pyelonephritis3 probably secondary to umbilical infection
1 Intestinal intussusception is the collapse of one portion of the intestine into another, like a telescope, often resulting in the obstruction of the intestine.
2 The abomasum is the fourth compartment of the stomach of cattle, similar to the human stomach in function.
3 Pyelonephritis is an inflammation of the kidney brought on by bacterial infection.

Chavatte-Palmer et al. (2002) described the gross pathology of 16 abnormal (LOS) SCNT-produced fetuses and stillborn calves. (This report also compares the clinical, hematological, and endocrine characteristics through two months of age in 21 apparently normal cattle clones with the same parameters in calves produced by IVF or AI; detailed discussion of these data can be found the Juvenile Development Node.)  The study included 11 abnormal fetuses and 1 normal SCNT fetus between 154 and 245 days of gestation, recovered either at slaughter or after spontaneous abortion, and five abnormal stillborn SCNT calves obtained at term. Three IVF-produced calves, one abnormal calf recovered at 242 days of gestation and two normal calves recovered at term, served as control material. All 16 of the abnormal SCNT fetuses (and placentae) showed some degree of edema due to hydrops.  Pregnancy outcomes and gross pathological findings in the clone calves are summarized in Table VI-2.


Table VI-2: Summary of Pathologies observed in Abnormal Fetuses and Stillborn Calves produced by SCNT
(adapted from Chavatte-Palmer et al. 2002)
Pregnancy Outcome IVF Controls SCNT
Abnormal Fetuses or Stillborns NR 11/12 exhibit "pathological gestation;" 1 animal sacrificed for control.
5 term stillborn (gd* 274.4 ?± 2.6).
Abdominal ascites and edema.
7 fetal membranes show large edematous cotyledons, and lower mean number of placentomes.
Mean and median weight of placentomes higher than for normal pregnancies and controls.
Kidney defects:
Fetus: 1 enlarged
Stillborn: both autolyzed.
1 apparently normal fetus had "seemingly small kidneys."
1 large fatty liver in fetus; seemingly large amount of fat surrounding abdominal organs in "several" fetuses (number not specified). No other gross morphologic abnormalities in other organs.

In a follow-up study by this same group, including animals from the 2002 study (Chavatte-Palmer et al. 2004), the authors noted a 76 percent survival rate (44/58) among clones following the first week after birth. Causes of death during the neonatal period included hyperthermia, umbilical hernia, respiratory problems, ascites (abnormal fluid accumulation) in the chest and abdomen, fatty liver, limb deformities, various digestive tract problems, and abnormal or degenerating kidneys.

Reports from research groups noting no differences between clones and naturally bred animals provide very few details about the health status of the clones. For example, Kubota et al. 2000 reported that although 30 blood measurements were taken on four clone calves, and that they observed no differences between the clones and their age-matched peers, neither the nature nor the numerical values of the measurements were provided.

Matsuzaki and Shiga (2002) evaluated the potential link between endocrine status and perinatal difficulties in Japanese Black clone calves delivered via C-section (selected by the investigators on the basis of a comparison of fetal size and maternal pelvic diameter, or rapidly expanding hydroallantois) relative to clones delivered vaginally, or Japanese Black calves produced via AI, and IVF calves born via spontaneous vaginal delivery. Birth weight, plasma cortisol levels, adrenocorticotropic hormone (ACTH), and components of the insulin-like growth factor signal transduction pathway (IGF) were evaluated. Average birth weights of clones delivered by C-section were heavier than AI controls; average birth weights of vaginally delivered clones and IVF animals were intermediate compared with C-section clones and AI control animals. Clones delivered by C-section had lower cortisol and IGF-1 levels than AI and in vitro produced controls, similar ACTH levels, and had more IGF binding protein-2 (IGFBP-2) relative to controls. The authors concluded that in C-section delivered clones the expected prepartum rise in plasma cortisol did not occur, and that these animals failed to initiate the switch to extra-uterine IGF-1 system during late gestation. Four of five C-section delivered clones died within the first week following birth; one of the eight vaginally delivered clones died in that same time period of unspecified causes.

In their first study, Kato et al. (1998) reported that eight of 10 blastocysts derived by SCNT from a Japanese Black beef cow completed gestation and were born. Seven were delivered vaginally, while one was delivered by emergency C-section due to dystocia. Two of the calves were born prematurely. Four of the eight calves died. No abnormalities were noted, and the authors attributed the deaths to “environmental factors” as described in Table VI-3.

In a second publication, Kato et al. (2000) reported the production of 13 surviving clones of 24 deliveries of Japanese Black and Holstein donor cells. Pregnancy duration was approximately equivalent to that of the donor cell breed, except that “a few” recipient cows had shorter gestations. Calves were either born vaginally or delivered via C-section; no criteria were given for the decision to perform C-section. Seven animals were either stillborn or died at delivery.


Table VI-3: Summary of Clone Outcomes
(source Kato et al. 1998)
Calf Gestation Length (days)1 Vaginal (V)/ Cesarean (C) Delivery Birth Weight (kg)2 Status at Publication Cause of Death
1 242 V 18.2 Alive NA3
2 242 V 17.3 Alive NA
3 266 V 32.0 Dead (day 3) Pneumonia apostematosa from heatstroke
4 267 V 17.3 Dead (day 0) Inhalation of amniotic fluid
5 267 V 34.8 Dead (day 0) Inhalation of amniotic fluid
6 276 V 23.0 Alive NA
7 276 V 27.5 Alive NA
8 287 C 30.1 Dead (day 0) Dystocia and delayed delivery
1 Average gestation length for Japanese Black cattle: 286.6 ± 0.9 days
2 Average weight of Japanese Black calf at birth: 27.0 ± 0.8 kg
3 NA = not applicable

Two clones died during C-section due to dystocia, but presented no gross abnormalities. One clone born appeared normal at birth but died 19 days later from septicemia. Six dead clones had significant morphological abnormalities of the kidney or outer extremities, including severe tendon contracture. One clone was born disemboweled, and another had a “warped” face. All of these abnormal births were attributed to infection with Akabane virus, a known teratogen (birth defect inducer), as antibodies to the virus were detected in the serum of afflicted animals. Mean body weights of clones were higher than those of controls,87 with nine clones exceeding the mean body weight of controls by >40 percent. Interestingly, Kato et al. report on the unusual appearance of some male clones derived from a bull that was 10 years of age when cells were taken for donors in the SCNT process. At birth, the bull calves were reported to exhibit “an adult appearance, displayed as many wrinkles in the skin, thick bone structure and rough hairs resembling those of adult males.” They speculate that these might result from mutations in the donor cells that increase with age or to telomere length.

In the Kubota et al. (2000) study of clones from the 17 year old Japanese Black bull described in the Cell Fusion/Fetal Developmental Node (Developmental Node 1), two calves died shortly after birth, one of which was diagnosed as having Akabane Virus. The other died due to complications following a difficult delivery (dystocia). Four others survived, and were reported to be healthy and normal. The average gestation periods for the clone pregnancies was 294 days (range of 291-299 days), which was nine days longer than the breed average of 285 days. Average birth weight of the clones was 36 kg (range of 30.7 to 42.5 kg), approximately 20 percent heavier than the breed average of 30 kg. 

Kishi et al. (2000) used fibroblast cells from ear punches of Holstein or Japanese Black cattle, and somatic cells isolated from the colostrum of mammary gland epithelial (MGE) cells from Holstein cows as SCNT donors. Of the 45 embryos implanted into 31 recipients, three pregnancies were confirmed on gestation day 60, and two calves were born from colostrum derived MGE cells. One clone was delivered at 279 days of pregnancy by C-section and weighed 44 kg; the other was vaginally delivered after induction of parturition at 280 days of gestation and weighed 45 kg. For the fibroblast-derived clones, 43 embryos were implanted into 37 recipients. Five pregnancies were confirmed on gestational day 60, and 2 calves were born (one Holstein and one Japanese Black). The clone derived from the Japanese Black fibroblast died six hours after birth due to internal hemorrhage of the umbilical artery. Two of the Holstein clones (the origin of the cells is unclear) received blood transfusions due to anemia at some unspecified time after birth. The three remaining Holsteins (presumably including the post-transfusion clones) were reported as “normal and healthy.”

A series of papers (Taneja et al. (2000); Tian et al. (2000); Xu and Yang (2001); Enright et al. (2002); Govoni et al. (2002); Xue et al. (2002); Savage et al. (2003)) has been published on a group of female Holsteins cloned from a 13 year old cow by the laboratory of X. Yang at the University of Connecticut. Most of these studies report on the birth and later development of these calves, and are discussed in the sections appropriate to those developmental nodes.
In a meeting abstract, Taneja et al. (2000) described the premature delivery of 10 Holstein clones and the supportive care that they required. Normal gestation length for a Holstein averages 282 days (range 280 to 285 days). All the calves born in this study were premature (average gestation length 266.6 ± 2.0 days), regardless of whether labor was induced or occurred naturally. Three cows initiated labor spontaneously at 263.0 ± 3.8 days gestation. Twin calves born to one surrogate dam were stillborn, with one requiring manual delivery. One of the calves in the spontaneous labor group was delivered by C-section, showing signs of stress and hypothermia (body temperature <100ºF). This calf was hospitalized after 36 hours, when it began running a fever. A chest x-ray revealed immature lung development, and blood gas measurements indicated low blood oxygen concentration. The calf also underwent surgery for an umbilical abscess and for patent urachus (the canal connecting the bladder with the umbilicus) on day 6, after which it recovered and survived. The last calf born in the spontaneous labor group was delivered vaginally with some assistance, was diagnosed with immature lung development and low blood oxygen concentration; it died within 12 hours of birth. Necropsy of this calf indicated bacterial infection and septicemia, as well as immature lung development. The remaining five surrogate dams were treated with dexamethasone 17 hours prior to planned C-sections. Four single calves and a pair of twins were born in the induced labor group. Two calves were delivered vaginally without assistance at 8 and 15 hours post induction treatment. The first calf (born after eight hours) was healthy and did not require supportive care. The second calf (born after 15 hours) died three hours after birth; necropsy revealed that it had died of hypoxia and immature lungs. A set of twin calves and another single calf were delivered by C-section. One of the twin calves and the singleton survived, while the other twin and another single calf died soon after birth. Necropsy revealed that they had inhaled mecomium (the first intestinal discharge that normally occurs after birth that can appear in the amniotic fluid if the fetus is distressed) and the lungs failed to inflate completely. All but one of the surviving calves required supportive care ranging from supplemental oxygen to surgery. The four surviving clones were the subject of additional studies by this lab, including Enright et al. (2002) and Govoni et al. (2002). In the study by Xue et al. (2002) comparing the relative effectiveness of different cell types as donors for SCNT, four of the six calves from the ovarian cumulus group survived the perinatal period; all four of the calves born from donor skin fibroblast cells died. All deaths occurred within 24 hours of birth due to respiratory distress.

Batchelder (2005, 2007a, 2007b) reported on the birth of eight clones (three Hereford and five Holstein) and nine comparators produced by AI (n=3) or ET (n=6). She noted an interaction between outcome and cattle breed, such that Hereford clones were heavier (range 50.0 to 71.0 kg; n=3) than their breed-matched ET comparators (range 31.5 to 48.0 kg; n=3), while Holstein clones had similar birth weights to their breed-matched ET comparators (37.1 vs. 39.4 kg). Neonatal clones had lower values for RBC, WBC and hematocrit at birth and for the first hour, but values were similar to comparators thereafter. Mean rectal temperatures were similar between clones and comparators at birth and declined rapidly within 1 h. This decline was more pronounced in clones (mean 1-h rectal temperatures were 101.4 Fº in clones vs. 102.9 Fº in comparators). After 6 h, thermoregulation was similar between clones and comparators.  Extensive clinical chemistry analyses indicated similar values between clones and comparators for most of the parameters measured, with the exception of measures of carbohydrate metabolism. Clones exhibited lower blood glucose and lactate levels than comparators during the first 24 hours, but were similar to comparators by 48 hours. Plasma concentrations of fructose in clones were higher than comparators, resulting in a nearly 2-fold higher ratio of fructose to glucose in clones both at birth and 6 h later. The authors postulate that hyperfructosemia in newborn cloned calves reflects higher fructose concentrations in utero and may be related to the abnormal placental morphology (fewer total placentomes with a flattened shape and increased mass and surface area) observed in this and other studies of SCNT calves.  No differences were noted between clones and comparators for plasma protein parameters, electrolytes and minerals, acid-base parameters, blood gases, or parameters indicative of kidney, liver and muscle function.  In both clones and comparator calves, plasma immunoglobulins were undetectable at birth and increased rapidly after calves consumed colostrum, indicating that passive transfer of immunity occurred normally in the clones.

Although Batchelder noted several clinical signs often associated with LOS in both Holstein and Hereford clones (delayed time to suckle and stand, hypoglycemia, forelimb flexor tendon contracture, enlarged umbilicus, patent urachus, and respiratory distress), many of the same signs were noted in the AI-derived comparator group in this study (see Chapter V for more details). During the first 48 h, spontaneous urination was rarely observed in clones; manual stimulation was needed to evacuate urine. In this study all clones survived the first 48 hours after birth, but two clones were lost between 72 hours and six days of age. All comparator calves survived.

Wells et al. (2004) reported that a total of 133 clone calves were delivered as a result of 988 embryo transfers of somatic cell nuclear transfers (SCNT) using adult and fetal donor cells. Embryonic cloning resulted in 27 delivered clone cattle from 210 embryos derived from embryonic blastomeres (ENT). Both techniques were reported to result in a live birth success rate of 13 percent. Approximately two thirds of these calves survived to weaning (3 months of age).

Yonai et al. (2005) reported on the growth, reproduction, and lactation of clones whose nuclear donors were a high milk performance 13 year old Holstein and a six year old Jersey that had previously been used for embryo transfer. These animals had previously been characterized as having shortened telomeres, but are otherwise indistinguishable from cattle of presumably normal telomere length (Miyashita et al. 2002). (Discussions of growth and reproductive and lactational performance of these clones are found in Developmental Nodes 3, 4, and Compositional Analysis, respectively). Table VI-4 summarizes the success rates for the two breeds of dairy clones. All embryos, regardless of the breed of the donor cows, were implanted into multiparous Holstein surrogate dams. One of the recipients of Holstein embryos had twin calves. The overall success rates, as measured by surviving calves as a function of embryos implanted were approximately 5 and 10 percent for the Holsteins and Jerseys, respectively.

The authors state that although there is an approximately two-fold difference in the production rates between breeds, this difference is not statistically significant due to the low numbers in the study. The abortion rate in the surrogate dams carrying Holstein clones was approximately two times higher than the Jersey group (68.4 percent v 31.8 percent). No dystocia was noted in surrogates carrying Jersey clones; incidence of dystocia in the surrogates carrying Holsteins was not reported. The authors attribute the differences in outcomes to the smaller size of the Jersey fetuses relative to the Holstein fetuses. Gestational periods and birth weights were reported as being within normal ranges for dairy cows of these breeds. Although there was more variability in birth weights of the Holstein clones than the Jerseys, no symptoms of LOS were noted in these two clone cohorts. The authors note that although cell culture conditions have been implicated as a potential source of large calves, the two cell lines used for nuclear transfer were cultured under identical conditions, implying that differences between the cell lines (i.e., heredity) was likely responsible.


Table VI-4: Success Rates for Implantation Through Delivery for Holstein and Jersey Clones
(source Yonai et al. 2005)
Jersey Embryos
Holstein Embryos
Embryos Transferred
Pregnancy Detected at 40 - 60 days
7 (31.8%)
18 (28.6%)
Failure to Reach Term
1 (14.3%)
11 (61.1%)
Calves Delivered
6/22 (27.3%)
8/63 (11.1%)
Surviving Calves from Transferred Embryos
4/37 (10.8%)
6/124 (4.8%)
Production Rate from Recipients
4/22 (18.2%)
6/63 (9.5%)
Average Birth Weights ± SD kg (ranges)
29.4 ± 1.5 (27.5-31.0)
36.2 ± 7.7 (27.0-47.0)

In summary, the survival rate of clones appears to be in the range of 5-18 percent, depending on how it is calculated. Many of the perinatal clones die of complications or sequellae of LOS. Newborn cattle clones may be more physiologically fragile than their comparators, and differences between clones and comparators include body weight, body temperature, alterations in the amounts of circulating IGF-II, leptin, growth hormone, T4, and differences in mean erythrocyte volume either on the day of birth or shortly thereafter. None of the differences between clones and AI- or IVF-derived controls persisted through the longest observation period (up to three months) (Chavatte-Palmer et al. 2002; 2004), and most resolved within a week or two of birth (Hill et al. 1999 (for transgenic clones); Enright et al. (2002); Govoni et al. (2002); and Tian et al. (2001)) (See subsequent discussions in the sections on the appropriate developmental nodes).

(b) Cyagra Dataset: Perinatal Cohort88

Of the 134 clones in the Cyagra dataset that were born or delivered, 103 animals (or 77 percent) were alive three days after birth. The remaining 31 were stillborn, died, or were euthanized within three days of birth. Details on health and survival of conventional, age-matched comparators (comparators) are not available. At the time that data were collected on these animals (late March 2003), 67 were alive (64 percent of those surviving to 48 hours, or 50 percent of those born or delivered). Eight animals died between 4 and 149 days of birth. The problems noted at the time of birth and the causes of death for those clones not surviving are summarized in Table E-2 of Appendix E: The Cyagra Dataset. Some animals required supportive care immediately after birth (e.g., glucose, warming, or supplemental oxygen), and many (n=26) received umbilical surgery after birth.

Blood was drawn for clinical chemistry and hematology for 10 clones within a few hours (or in some cases, minutes) of birth. The actual measurements provided by the Cornell Animal Health Diagnostic Laboratory are found in Appendix E, Tables E-100a (clinical chemistry), and E100b (hematology). Charts E-100, E-101, E-102, E-110, E-111, and E-112 compare these values with the comparator population reared on the same farms and the Cornell Reference Values and are also found in Appendix E, along with all of the data from which they were generated.

Ninety percent of the total clinical chemistry values of the clones were within the range of values exhibited by the comparators, and 90 percent of the hematology values were within the comparator range. Twenty-seven of the 33 analytes (substances that were measured, such as sodium, cholesterol, or liver enzyme activity) had either no differences or one difference relative to the comparators (Chart E-101). The remaining six analytes tended to be more variable between clones and comparators. Liver values (AST, GGT, cholesterol, bile acids (hBA)) were lower in several clones, for reasons likely related to the placental/umbilical abnormalities, or transitions from fetal to adult circulation. GGT levels were also low relative to the comparators, probably related to blood sampling prior to colostrum intake, whereas comparators were administered colostrum prior to blood draw. None of the out-of-range values of these analytes poses any particular concern for food safety, as they are relatively close to the comparator range.

Blood cell parameters in the neonatal clones were also very similar to those of the comparators. Fifteen of the 17 analytes had either no differences or just one difference between the two groups (Chart E-111). With the exception of one clone that was infected with rotavirus and subsequently died, all red blood cell parameters were within the range of the comparator group. Three clones had white blood cell counts that were lower than the comparator range. One clone was infected with rotavirus but survived, indicating that at least in that animal, the immune system was functioning appropriately. There did not appear to be an increased incidence of infection in these animals, except where infection was associated with umbilical difficulties, also indicating that the immune systems were functioning appropriately.
“Sentinel” markers were sought that might predict a successful outcome for perinatal clone calves. Based on the literature and the Cyagra data, it does not appear that any one analyte or analyte profile is predictive of whether a particular animal, or indeed, the entire cohort of animals will develop into normal, fully functioning, healthy animals. The laboratory data are consistent with the hypothesis that animals that look and behave normally are normal with respect to laboratory values, implying that consideration of the complete dataset on an individual animal is the best predictor of the health of that animal. Further, the seven surviving Cyagra clones that were sampled twice (# 71, 72, 73, 78, 79, 119, and 132) provide the baseline data for a small subcohort of animals for which there are laboratory measurements at two different time points, as described more fully in the Juvenile section.

(c) Unpublished data

In response to requests by CVM, various groups involved in cloning submitted unpublished data. One such group, a commercial cloning company, submitted body temperature, pulse and respiration rates on 19 cattle clones (breed(s) and gender not identified) during the first 72 hours of life. These data has been discussed in greater detail in Chapter V. Body temperatures were elevated during the observation period (mean 103°F at birth; 102.7°F at 72 hours); heart rates appeared to increase (95.2 beats/min at birth; 138.6 beats/min at 72 hours); while respiration rates remained fairly constant (53.9 breaths/min at birth; 53.1 breaths/min at 72 hours). It is often difficult to evaluate data on heart rate and respiration in livestock, since the stress of handling tends to increase these rates. Body temperature in neonatal clones appears to be quite variable, with some studies reporting hyperthermia (Chavatte-Palmer et al 2002; Batchelder 2005) which may persist through the first 50 to 60 days of life and then appears to normalize.

Another cloning firm presented birth records on two Holstein heifer clones delivered by C-section. The calves weighed 45 and 47.7 kg at time of delivery, within the normal range for Holstein cattle; body temperatures were 100 and 102.6°F at birth, slightly below and above normal (101.5°F) for cattle. These two calves were otherwise normal, according to the veterinarian’s notes and limited blood chemistry (See Chapter V for details).
(d) Summary for Perinatal Developmental Node in Bovine Clones (Developmental Node 2)

The combined information from the peer-reviewed literature and the Cyagra dataset indicates that, overall, the health of newborn clones tends to be more unstable than their comparators, with a higher incidence of perinatal death. Abnormalities noted among both dead and surviving clones include respiratory distress, organ malformations, flexor tendon contracture, and umbilical difficulties. Some animals succumbed to infection, but there does not appear to be a decrease in immune function in the population of clones at the perinatal stage. Despite the increased incidence of perinatal death and abnormalities in newborn clone calves, none of the adverse outcomes observed are qualitatively different from adverse outcomes that have been observed in natural breeding or other ARTs. It is therefore unlikely that any unique food consumption risks have been introduced into these animals. 

iii. Juvenile Development in Bovine Clones (Developmental Node 3)

Most of the information on this developmental node has been extracted from publications that primarily address the perinatal period.

(a) Peer-reviewed Publications

For purposes of following the cohorts of animals, these reviews have been grouped by institution.

The Institut National de la Recherche Agronomique (INRA) Studies: Renard et al. 1999 and Chavatte-Palmer et al. 2002, 2004

Renard et al. (1999) reported one case of lymphoid hypoplasia in a clone generated from cells in an ear biopsy of an animal that had herself been the product of blastomere (or embryo) nuclear transfer (BNT). An echocardiogram performed on the animal immediately after birth revealed an enlarged right ventricle of the heart. The animal was treated with an angiotensin converting enzyme (ACE) inhibitor and given diuretics for one month, at which time the condition was reported to be resolved. Blood samples taken every two days after birth revealed relatively high reticulocyte counts and immature blood cells in the blood during the first three weeks of life. Lymphocyte (white blood cell) counts were also reported as normal for about a month after birth, but counts fell rapidly after that time. Hemoglobin levels in the animal also decreased at about day 40. On day 51, the animal died from severe anemia. Histological examination of the calf revealed hypoplasia (lack of development) of the thymus, spleen, and lymph nodes or global lymphoid aplasia (absence of lymphoid cells in all organs in which they would likely be found) that likely began at birth. No evidence for the endogenous synthesis of immunoglobulin G was detected. Bovine Viral Diarrhea virus, which has been known to induce thymic atrophy, was ruled out. SCNT was implicated as the cause of the lymphoid aplasia, possibly due to the selection of a cell with a mutation responsible for the expression of the portion of the genome governing lymphoid development, or lack of appropriate reprogramming of the somatic cell nucleus. In a follow-up study by this group (Chavatte-Palmer et al. 2004) an additional four clones were diagnosed with thymic aplasia. Histological examination of the thymus glands of these calves indicated abnormal tissue organization, suggesting the aplasia was the result of epigenetic errors. It is not clear from the late report whether these four clones were also the result of serial cloning. To our knowledge, this is the only laboratory reporting thymic aplasia in clones. Three other calves in this cohort died suddenly with few or no clinical signs: two died of diarrhea, and one died without any apparent cause.

To investigate if apparently normal calf clones share similar clinical and endocrine characteristics with conventionally-bred calves, Chavatte-Palmer et al. (2002) monitored 21 clones from birth to two months of age. Data from this part of the study are summarized in Table VI-5.  (Some of the results of this study are discussed in the Perinatal Developmental Node).  Blood samples were collected daily for the first week of life, weekly for the first month, and every two weeks thereafter.

For the first week after birth, the mean rectal body temperature was higher in clones than AI controls, and some temperature spikes (up to 41º C; normal temperature is considered to be approximately 38.5 ºC in dairy cows) were observed. Elevated temperatures in the clones persisted for 24-36 hours, and were not sensitive to pharmacological intervention. Animals were cooled by wrapping in wet towels and providing ventilation, although they did not appear to be distressed during the temperature spikes. No bacterial infection was detected, and no changes in hematology or clinical chemistry were observed. The authors state that the mean temperature remained elevated for 50 days, although data are only provided for the first week.

Hematologic parameters evaluated in this study included red blood cell count (RBC), hematocrit (HC), hemoglobin (Hb), and counts of white blood cells (WBC), including differentials (counts of the distributions of populations within the overall category of white cells). Mean cell volume (MCV) was higher in clones than AI controls, and the neutrophil:lymphocyte ratio at birth was higher in clones than in AI controls. The authors state that this ratio reflects the normal increase in fetal cortisol production at birth that occurs as a result of adrenocortical maturation. As previously mentioned, one clone presented with lymphoid aplasia (Renard et al. 1999) and had decreased lymphocyte and RBC counts. Measurements of all other hematologic parameters in clones were similar to those in AI controls. Clinical chemistry values were within normal limits. Thus, with the exception of the aplastic clone (Renard et al. 1999), no clinically relevant findings accompanied these measurements over the time period of the study. (For a discussion of the nature and relevance of these tests, refer to Appendix F).

Thyroxine (T4) is an important, stimulatory regulator of metabolic rate. Because metabolic rate is (usually) correlated with body temperature, elevated T4 secretion may be implicated as a cause of hyperthermia in clone calves. However, plasma T4 levels were lower in clones than controls during the first two weeks of life, after which they were similar to levels in controls. Chavatte-Palmer et al. (2002) noted that lower plasma T4 levels coupled with elevated body temperatures in young calves are consistent with studies on brown adipose tissue (BAT) by Carstens et al. (1997b). Brown adipose, found in neonates of many mammalian species, serves to generate heat.

Table VI-5: Summary of Clinical, Hematologic, and Hormonal Characteristics from Birth to Two Months of Age in Apparently Normal SCNT Clone Calves Versus AI and IVF Controls
(adapted from Chavatte-Palmer et al. 2002)
Outcome AI or IVF Controls SCNT Comment
Live Births
  Total  Caesarian   delivery Vaginal delivery

n=176 not specified

not specified

n=21; 7 fetal origin; 13 adult origin 20 (18 at term, 2 were 1 week before term).


Clones delivered via C-section when natural calving had not occurred by gd 282. All calves survived to at least 2 mo of age.
Body Weight at Birth (kg)

43.7?± 2.7


55.1?± 2.7; n= 26

Higher compared with and AI and IVF calves (P<0.01)

No significant difference between AI and IVF.
Body Temperature (BT) at birth Lower than SCNT (approximately 38 to 39.5ºC)

Mean rectal BT higher in SCNT than controls in 1st week, and until 50 days.

Data provided for only 1st week.

Peak temperature spike approximately 41o C.

No accompanying clinical signs.

Comparison between n=10 NT and n=10 combined AI (8) and IVF (2).


Not sensitive to NSAID; regulated by using wet towels and ventilation.

Hematologic Parameters RBC,




Differentials Mean cell Parameters


Mean cell volume (43.59± 0.60 fl).

Neutrophil: lymphocyte ratio at birth 3.14 ± 1.1; higher than SCNT.

n=21 live clones.

Mean cell volume (50.07 ± 1.29 fl) higher than AI.

Neutrophil: lymphocyte ratio at birth 6.28 ± 0.9; higher than AI.

1 animal with lymphoid aplasia (Renard et al. 1999), sudden decrease in lymphocyte and RBC counts.

Measured only at birth.







NR All values within normal limits; individual data not provided. Measured only at birth.
Thyroxine (T4) n=4 n=7; lower than AI controls for days 1-15 (P<0.05). Approximate kinetics the same as AI (rapid decrease from birth to d 4, then constant low level (~15-25 pmol/l) to day 15. Measured for 2 months to determine whether associated with hyperthermia.




n=5; No diff. from SCNT.

Lower than SCNT at birth and d 15.

No difference from SCNT.

n=7; no difference from AI.

Higher than AI at birth and day 15 (P<0.05).

No difference from AI.

Measured from day of birth until 80 days of age.
Leptin n=5; Lower than SCNT animals, and less inter-animal variability. n=6; higher than controls during first week after birth (P<0.01). More inter-animal variability and changes in absolute response in SCNT animals. Levels revert to normal in amount and amplitude after one week. Measured from day of birth until 28 days of age.
Growth Hormone


no difference

from SCNT


no difference

Same as leptin assay.
Pre- and Post-Prandial Insulin and Glucose n=6; No significant differences between AI and clones at 1 or 8 days of age.

Some clones presented with hypoglycemia and hypothermia during first 24 h post partum.

No significant differences between clones and AI after the first 24 hours.

Measured at 1 and 8 days of age.
Cortisol (ACTH Induction)

n=2; C-sect,

n=6; natural birth. Basal levels in C-section births lower than natural birth.

n=11; C-section.

n=1; natural birth.

No significant differences between clones and controls.

Basal and stimulated levels in C-section births lower than natural birth.

Increase in plasma cortisol in response to ACTH stimulation reflects appropriate adrenal maturation and function.

Lower basal cortisol values probably due to C-section and not NT or IVF.

AI = artificial insemination
IVF = in vitro fertilization
NR = not reported
C-section = Caesarian delivery
gd = gestational day
NSAID = non-steroidal anti-inflammatory drug

to keep the newborn warm during cold stress. Metabolism in BAT is stimulated by norepinephrine, and norepinephrine secretion is increased by stress (e.g. cold stress). In cattle, BAT usually disappears following the neonatal period (Blaxter 1989). The findings of Carstens et al. (1997b) indicated that, in contrast to most tissues, thermogenesis in BAT is suppressed by elevated T4 (see discussion in Chapter V). It is therefore plausible that the hyperthermia observed in the calf clones by Chavatte-Palmer et al. (2002) and other investigators is independent of T4, and may instead be explained by increased BAT metabolism. However, in the absence of empirical data on BAT metabolism (e.g. norepinephrine levels as an indirect measure), this idea remains speculative.

Chavatte-Palmer et al. (2002) investigated levels of several hormones that regulate growth; insulin like growth factor (IGF)-1, IGF-2, IGF binding protein (IGFBP), leptin, and growth hormone.There were no differences between clones and AI controls in concentrations of IGF-I, IGFBP, or growth hormone.  Levels of IGF-II in clones were higher than controls at birth, but were lower in clones on day 15. Leptin levels were higher in clones than controls during the first week of life, but were similar in clones and controls for the remainder of the study. As leptin is produced by adipocytes (fat cells), the authors speculate that increased leptin levels in clones may reflect greater amounts of adipose tissue, consistent with the higher body weight of clone calves, and supported by the authors’ subjective postmortem observations of more intra-abdominal fat in clones compared to non-clone calves.

As a measure of possible metabolic disturbances possibly related to the increased size of clones, Chavatte-Palmer et al. (2002) assayed plasma concentrations of insulin and glucose, both before and after feeding, and one and eight days of age. There were no differences between clones and AI calves in either pre- or post-prandial concentrations of insulin and glucose.

Finally, Chavatte-Palmer et al. (2002) measured cortisol secretion, both basal and in response to challenge with adrenocorticotropic hormone (ACTH, the pituitary hormone that induces the production of cortisol).  In cattle, as well as many other mammalian species, secretion of cortisol may be increased by stress. In prematurely born animals, the cortisol responses to ACTH challenge are decreased. Thus, challenge with ACTH provides a measure of adrenal maturation and function. Relative to calves born vaginally, cortisol levels (basal and ACTH-stimulated) in calves born by C-section were lower in both clone and non-clone calves. At one, seven, and 30 days of age, all of the calves exhibited similar cortisol levels following challenge with ACTH, indicating that maturation of the adrenal axis was normal in clone calves. These results also provide evidence that overgrowth of clone calves in utero is not due to fetal adrenal dysfunction.

To summarize, the study of Chavatte-Palmer (2002) indicates that there are some physiological indices that differ between apparently normal clones and controls during the transition from the perinatal to the early juvenile period.  However, most of these differences were resolved within the first 15 days of age. Of the parameters that were different in clones, even the most persistent, elevated body temperature, was resolved after 50 days. The authors conclude that, based on their data, apparently healthy clones should not be considered “physiologically normal animals until at least 50 days of age.”

In 2004, Chavatte-Palmer et al. published results from the first year of a three-year study designed to prospectively address the health of clones. They noted that for the first 65 days after birth, clones (n=25) had slightly lower hemoglobin levels than AI comparators (n=19), although the hemoglobin levels of the clones were still considered within the normal range. The lower levels persisted for the first 65 days after birth before reaching the same levels as the AI comparators.  This finding reinforced the opinion of the group at INRA that clones could not be considered physiologically normal for the first two months of life. 

Additional results from the three-year study were published by the group at INRA in 2007 (Heyman et al.). Twenty-one heifer clones of four different genotypes and 19 controls were studied between 4 and 36 months of age.  Findings during the juvenile period will be discussed here, while observations relevant to reproductive development and the post-pubertal phase will be discussed in the appropriate developmental nodes. All but one animal survived the juvenile period of the study; one control heifer was euthanized at seven months due to extreme laxity of the flexor tendon. Using daily gain and feed intake as indicators, no differences in growth rate were observed between clones and control heifers up to 15 months of age. The authors state that repeated clinical evaluations revealed no differences in cardiovascular, respiratory and locomotive functions, clinical biochemistry, or immune parameters (data not shown).  

It has been hypothesized that clones may be more sensitive to stress than conventional animals.  To address this hypothesis, the group at INRA measured concentrations of cortisol and catecholamine metabolites every two months in a group of juvenile clones (n=5) and a group of age-matched controls (n=5). They found that although cortisol concentrations were more variable in clones, blood cortisol concentrations were similar between clones and controls during the juvenile period (4 to 12 months of age). Similarly, no differences were found between clones and controls in urinary concentrations of epinephrine and norepinephrine. The authors cite these findings as evidence that the clones were not suffering from chronic stress.   

The University of Connecticut Studies: Govoni et al. 2002; Enright et al. 2002; and Savage et al. 2003.

Govoni et al. (2002) investigated the degree to which the somatotropic axis89 in Holstein clone heifers (n = 4) developed normally compared to AI-produced age-, gender- and breed-matched controls (n = 4). Serum samples were collected monthly from five to 14 months of age. Normal pulsatile patterns of GH secretion were observed in both clones and controls. Averaged across all time points, concentrations of GH were similar between clones and controls, but the patterns of secretion over time were different. GH levels declined in controls between five and 14 months of age, but concentrations of GH did not change in the clones during the same period, resulting in higher GH concentrations in clones at 9, 10, and 11 months of age. Parallel increases in IGF-I were observed in both groups over time, but clones had lower IGF-I concentrations compared to age matched controls from 5 to 11 months and at 14 months of age. As IGF-I is involved in development of ovarian follicles and uterine growth (Le Roith et al. 2001), lower circulating IGF-I levels may be partially responsible for the later onset at puberty observed in this group of clones (Enright et al. 2002).  It is important to note, however, that the concentration of IGF-I required for normal sexual development is not known.

Clones secreted 5-fold more GH in response to GHRH compared to controls. When GHRH and somatostatin were administered together, there was less inhibition of GHRH-stimulated GH release in clone heifers compared to controls. Therefore, the elevated GH concentrations in the clones in this study may be indicative of an increased responsiveness to GHRH. 

IGF Binding Proteins (IGFBPs) are responsible for transporting IGF in the blood and extend the half life of IGF-1. Concentrations of IGFBP-2 were static and not different between clone heifers and controls, but levels of IGFBP-3 were lower in clones compared to controls. This result parallels the lower concentrations of IGF-1 in the clone heifers.

Govoni et al. (2002) note that although they observed differences between clones and comparators in concentrations of GH, IGF-1, and IGFBP-3, concentrations of all three of these endocrine parameters were within the ranges previously reported in conventional cattle at similar ages.The authors therefore concluded that the clones exhibited age-appropriate development of the somatotropic axis and, overall, clones and comparator heifers were developmentally similar over time. The authors also point out that the clones in their study were derived from a single nuclear donor (cow) with superior genetic merit (high milk production). There is evidence that genetic differences in cattle may be identified by GHRH-stimulated GH secretion (Lovendahl et al. 1996), and bulls of genetically superior bulls secrete more GH in response to GHRH (Kazmer et al. 1992; Zinn et al. 1994). Moreover, cows with high milk production have greater concentrations of GH than do cows with low production (Hart et al. 1975). Therefore, Govoni et al. (2002) speculate that the differences they observed in GH, IGF-1, and IGFBP-3 levels may be explained by the difference in genetic merit between the identical clones and their comparators.

 Savage et al. (2003) evaluated the behavior of the clones and age-matched controls described by Govoni et al. (2002). Between 32 and 36 weeks of age, there were no differences in weight or height between the clones (205.5 ± 9.9 kg; 117.0 ± 1.8 cm) and controls (211.4 ± 7.4 kg; 119.5 ± 1.4 cm). All calves were raised together under the same management conditions. Based on a series of studies evaluating approach to other animals and novel objects, clones exhibited age-appropriate behaviors, but were reported to be more aggressive and inquisitive than controls, and spent more time grooming and socializing. Clones tended to spend less time in playful behavior than controls. Review of records on the cow that served as the donor for the clones indicated that she had displayed similarly aggressive and inquisitive behavior as a young animal, suggesting that at least some of these behavioral traits may be genetically controlled. Clones spent more time in proximity to adult animals in an adjacent pen (which also housed the nuclear donor), and in proximity to the feed bunk compared to control animals. In general, clones were reported to spend more time with each other rather than socializing with control animals, with the authors speculating as to whether clones exhibit some form of genetic kinship recognition. Nonetheless, the overall conclusion of this study was that the clones behaved normally.

Other Studies

Wells et al. (2004) and Wells (2005) followed the growth and maturity of cattle clones generated at their facility in New Zealand through approximately four years of age. Approximately 80 percent of the clones delivered alive at term survived the first 24 hours of live. They reported that two-thirds of the 20 percent that died was due to spinal fractures syndrome or to deaths from dystocia, associated with LOS (Wells 2005). Another 15 clones died in the time period before weaning, most commonly of musculoskeletal abnormalities, including tendon contracture and chronic lameness, and umbilical infections, attributable to complications of LOS. They also reported two clones dying as the result of bloat, and an unspecified number of clones dying due to endophyte toxicity after eating fungus-infected ryegrass. Bloat and other gastrointestinal disorders have been reported by others (Cyagra 2003, Appendix E; Batchelder 2005), but also may result from feeding or grazing management problems. Wells et al. use the phrase “clonal family” to refer to clones derived from a particular donor, and note that the bloat and susceptibility to endophyte toxicity was restricted to one clone family, and likely due to their genetics. Another clone family consisting of three clones (and five half-siblings produced by AI) survived with no health anomalies and at the time of reporting was 18 months old. Other health problems observed during the juvenile period included anemia, chronic heart failure, and degenerative nephrosis, problems that have also been noted by other researchers (Chavatte-Palmer et al. 2004). Additional deaths were categorized as being due to misadventure and accidental deaths due to clostridial disease, parasitism, and over feeding. Surviving animals from this group were characterized with respect to general health and physiological measurements; these are found in the discussion of Developmental Node 5 (Post-Pubertal Maturation).

Similar to Chavatte-Palmer et al. (2002), Batchelder (2005) also noted periodic moderate to severe hyperthermia in Holstein and Hereford clones up to 60 days of age. As with the Chavatte-Palmer clones, the Batchelder clones also showed no indication of infection, were unresponsive to anti-inflammatory drugs, and their behavior was unchanged; the hyperthermia also resolved spontaneously.

In their study of Japanese Black beef cattle clones described in the section on the Perinatal Developmental Node, Kato et al. (1998) reported that all of the clones that survived the perinatal period were alive and healthy at 85 and 120 days of age. In the subsequent study (Kato et al. 2000) of 13 clones that survived the perinatal period, 12 clones were alive and healthy at 117-350 days, and one clone died at three months “for no clear reason.”

Kubota et al. (2000), in their study of four surviving clones of a 17 year old Japanese Black bull, reported that the clones were 10-12 months of age at the time of publication. Based on veterinary examinations, growth curves, and 30 blood parameters no differences were found between the clones and their age-matched peers. No data were provided in the publication. Other groups have also reported normal growth rates for cattle clones (Wells et al. 2004; Heyman et al. 2004).

Yonai et al. (2005) (previously mentioned in the Perinatal Developmental Node) studied the growth of Holstein (n= 6) and Jersey (n=4) clones with shortened telomeres. Clones were given at least two liters of warmed colostrum immediately after birth, fed colostrum twice a day for the first five days of life, and monitored for physiological functions until they stabilized. Clones were fed according to the guidelines presented by the US National Research Council Nutrient Requirements of Dairy Cattle (1989). From Day 5 through Day 45, calves were given milk replacer twice daily, and offered calf starter pellets, hay and water during this time. After Day 45, all calves (clones and comparators) were weaned from milk replacer, and their feed gradually changed from calf starter pellet to formula feed over a two week period. Calves were fed 2-3 kg/day of formula feed, hay and water from Day 60 until one year of age. For the first 45 days after birth, the clones were reared in individual calf huts, after which they were reared together with other calves produced by AI or embryo transfer. Calves were held in a large pen in mixed groups of clones and age-matched comparators during the weaning period. After weaning, groups of 10-20 animals were moved into pens, and after one year of age, all animals were moved to a free-stall barn for heifers. Table VI-6 summarizes the average daily body weight gain of the clones from birth to two years of age. Body weights were collected monthly from birth to one year of age, and every three months between 15 and 24 months.


Table VI-6: Average Daily Gain (kg/day) for Holstein and Jersey Clones
(source Yonai et al. 2005)
months of age
  0-3 3-6 6-9 9-12 12-15 15-18 18-21 21-24  
Jersey Clones (n=4)
Mean 0.49 0.73 0.67 0.53 0.49 0.56 0.51 0.40  
SD 0.02 0.02 .11 0.06 0.05 0.17 0.16 0.18  
Holstein Clones (n = 6)
Mean 0.72 1.17 0.82 0.85 0.90 0.97 0.68 0.58  
SD 0.14 0.12 0.08 0.11 0.10 0.26 0.11 0.27  
SD = Standard Deviation

The authors report that the average daily gain for the clones was greater than that of the standard of each breed. For the Holstein clones, the average bodyweights conformed to the standard during the first three months of age, but exceeded the standard after five months, while the Jersey clones exceeded the body weight of the Japanese Feeding Standard for Dairy Cows throughout the measured time period. The Holstein clones’ body weights were approximately equivalent to that of the donor animal until 18 months of age, but exceeded it thereafter. The Jersey clones exceeded the body weights of the donor from birth to two years of age. The animals were reported as healthy with normal growth throughout this time period. No deaths were reported after the perinatal period.

There are other reports of clones that appear to be healthy at birth but unexpectedly die some time later. Gibbons et al. (2002), for example, reported a clone dying at 60 days of age due to respiratory and digestive problems. As mentioned above, Kato et al. (2000) also reported the death of a clone at 3 months. Chavatte-Palmer et al. 2004; Wells et al. 2004; Batchelder 2005 have also noted early deaths, but their cause(s) have not been clearly linked to cloning. The degree to which these unexpected deaths in cattle are related to cloning, or some disease process that is independent of cloning, is not clear. Ogunuki et al. (2002) have noted shorter life spans in some of their mouse clones; the cause of death appears to be due to liver damage, pneumonia, or neoplasia. The relevance of mouse models to domestic livestock has been discussed in Chapter IV.

(b) Cyagra Data: 1-6 Month Age Cohort

The calves from the Cyagra dataset most closely correlating to the Juvenile Developmental Node are the 46 clones and 47 comparators found in the 1-6 month of age group. Tables E-200a and E-200b, and Charts E-200, E-201, E-202, E-210, E-211, and E-212 describe CVM’s analyses of the information.

In general, these clones appeared normal, although some anomalies were noted on physical examination. These may be related either to cloning or to the genetics of the animal that was being propagated. None are unique to clones, although their frequency appears to be higher in clones than in calves produced using other forms of reproduction (see Chapter V). One of the clones was culled for poor conformation (the physical appearance of the animal), a matter of potential business importance to the producer, but likely having no impact on either food or animal health. Conventional animals with poor conformation are generally not used in selective breeding programs, and may be culled; it is likely that breeders will put similar limitations on clones as well. Several of the clones experienced serious problems resulting from umbilical abnormalities, including enlargement, excessive bleeding, and infection of the navel. These were resolved surgically. In addition, three cases of cryptorchidism (undescended testicle) were identified in calves from the same cell line. Although this condition is relatively uncommon in conventional animals, it is observed with some frequency, and is thought to be hereditary.

Interestingly, three clones derived from the same Jersey cow cell line presented with very different phenotypes. Clones # 87, 88, and 89 were within 10 days of age of each other when they were weighed and blood samples drawn (131-141 days old). All three required umbilical surgery. The oldest, clone #87, weighed 282 pounds. Clone #88, who was a day younger, weighed 197 pounds, and the youngest (at 131 days of age) weighed 215 pounds. Otherwise, the animals were healthy on physical examination. A fourth clone from this cell line died at birth from LOS-related complications.

Measurements of analyte levels in the entire 1-6 month old cohort were generally very close to those measured in the comparators (Chart E-201). In aggregate, 96 percent of the total analyte values for clones were within the range of the comparators. A few were out of range: glucose values were above the range of the comparators in six of the 42 likely valid measurements (four were considered artifactual). In order to determine whether the hyperglycemia was transient or sustained, urinalysis results were checked for the clones with elevated blood glucose levels. As none of those tests were positive for glucose (the renal threshold for glucose in cattle is approximately 100 mg/dl: i.e., if blood levels of glucose exceed 100 mg/dl for any appreciable time, glucose spills over into the urine), it is unlikely that the higher blood glucose levels (88-123) had been sustained long enough to allow for spillover into the urine. Most likely, these were transient elevations resulting from proximity to a meal or as a short-lived response to stress (as in being restrained for blood draws).

The hemograms for the cohort did not reveal any significant health concerns. None of the clones were anemic, and there was no depression of cellular immune function. Some of the clones had individual values that were outside the range of the comparators, but these were not judged to pose either an animal health or food consumption risk (see Appendix E for a more complete discussion).

It is important to note that although this time period appears to be relatively short, it spans an important developmental transition period for ruminants. Calves that are closer to one month of age are still primarily milk-fed, while those closer to six months of age have mostly transitioned to a more adult diet, and function as ruminants. The youngest animals are in a very rapid growth phase, while the older animals in the range, although still growing, are doing so at a slower rate. Because young animals are growing rapidly, measures of bone growth such as calcium, phosphate, and alkaline phosphatase might be expected to be higher in younger compared to older animals. Comparison of both the clone and comparator laboratory values to the Cornell Reference Range (which is derived from adult cattle) (Charts E-200 and E-202) indicates that many of the clones and comparators exhibit calcium, phosphate, and alkaline phosphatase levels that exceed the Cornell Reference Range. This finding is consistent with higher rates of growth in young calves relative to adults, and provides confidence that clones and comparators are exhibiting similar, normal physiological responses to growth stimuli. Review of Chart E-201 reveals that clone alkaline phosphatase values are almost entirely within the range of the comparators (38 of 46 values). Most of the clones whose alkaline phosphatase levels exceeded the comparator range were the youngest animals.

Another set of physiological parameters that varies with age can be seen in total protein, globulin, and albumin levels. These measurements reflect, among other things, the immune status of the animal. Immediately after birth, globulin levels, which are largely comprised of immunoglobulins, are derived almost entirely from colostrum (the antibody-rich first “milk” to be secreted by mammals). “Passive immunity” is conferred by the ingestion and intestinal absorption of immunoglobulin-rich maternal colostrum. In the two to four months after birth, the calf’s own immune system begins to develop its production of immunoglobulins, as the circulating supply of maternally-derived immunoglobulins in the calf’s blood wanes. This phenomenon can be observed in Charts E-200 and E-202 (Clones: Reference Range (1 to 6 months) and Comparator Population: Reference Range). Clone and comparator globulin values are low relative to the Cornell lab reference range because that reference range is derived from adult animals with fully functional endogenous immunoglobulin production. The clone and comparator calves in this cohort have not fully started to produce their own antibodies from their own B-lymphocytes. Review of Chart E-201 (Comparison of Clones to Comparator Population), however, indicates that there were few differences between the clones and the comparator population, reflecting the appropriate age-related lag accompanying the decrease in passive acquired immunity and endogenous immunoglobulin production. The globulin levels that are different between clones and comparators reflect this age-related physiological phenomenon. Clones #72 and 73 were among the youngest in the one to six month old group, and thus would be expected to have lower globulin levels. Comparison of the globulin value for clone #100 (174 days of age, globulin of 4.6g/dL) with clone #72 (48 days of age and globulin level of 1.6 g/dL) clearly demonstrates the age-related changes in the analyte, and appropriately reflects the normal developmental increase in endogenous globulin production.

Sub-Cohort Analysis

Examination of the subcohort of seven clones (# 71, 72, 73, 78, 79, 119, 132) at two time frames (birth and the 1-6 months of age) allows the determination that appropriate age-related physiological changes are occurring in the clones on an individual animal basis, rather than on a population basis. For example, gamma glutamyl transferase (GGT) values appear low relative to comparators in “within 24 hours of birth” time period for four of these seven clones. This likely reflects the difference in timing between when the blood samples were drawn for clones and comparators (Clones had their blood samples drawn prior to colostrum administration, while comparators had their blood drawn some time after being fed colostrum). As colostrum has high intrinsic GGT activity, the difference between the two groups may be due to its effective absorption of GGT by the comparators. GGT values normalized by the time of the second blood draw for three of these animals, and were only slightly lower (4U/L vs. the comparator range of 5-32 U/L) in the remaining clone at Day 48.

At birth, some of the clones in this sub-cohort had measures of liver function out of the comparator range (lower AST, and low bile acid or cholesterol levels). Low cholesterol is associated with retained fetal circulation in the livers of young animals. Were these low cholesterol levels to continue into the next developmental node, there might be cause for concern, but given that they normalized at the time of the second blood draw, there is little reason to expect that the lower values in these very young clones pose a health risk. The low levels at birth are more likely a reflection of the changeover from fetal to neonatal circulation, possibly exacerbated by the clones’ unusually large umbilical vessels, which often required surgical correction. The lower bile acid and AST values observed would also be related to the transition from fetal to neonatal circulation, and are not indicative of any disease state. All of these values normalized by the second measurement, as did additional analyte levels that were out of range for individual clones perinatally (low CK, TIBC, and iron). These measurements reflect normal adaptive physiological processes and not pathologic or disease states, and instead provide evidence of the “normalization” of the clones as they matured.

A few laboratory measurements appeared outside the range of the comparators in some of the clones at the time of the second measurement, but these do not appear to have clinical relevance. Complete blood count information is only available for four of the seven clones measured at both time points, and do not appear to be reflective of clinical problems. For a more complete discussion of these data, see Appendix E.

(c) Unpublished data

Full hematology and clinical chemistry screens on three pre-pubertal bull clones (aged 5 to 7 months old) were shared with CVM by a private veterinary clinic (Appendix G, Tables G-3a-3c). The clones were described as being clinically, physically and behaviorally normal, with normal growth rates and size. Blood samples were taken three times over a six week period. All of the clinical chemistry data, with the exception of one, were within normal published ranges or within the comparator range for the testing laboratory. Just as for the physiological data shared by Cyagra, the reference range for the testing laboratory was for an older cohort of animals (that were also female), and were not age-appropriate. The one analyte that fell outside a reference ranges occurred in a single sample in one bull clone, and was a low cholesterol value. All measurements in the subsequent sample from this bull clone were within normal ranges.

(d) Summary for Juvenile Developmental Node in Bovine Clones (Developmental Node 3)

Numerous studies have provided information about the physiology of bovine clones during the juvenile period. Some juvenile bovine clones exhibiting LOS at birth eventually succumb to its sequellae. Surviving clones, however, appear to grow and develop normally. Detailed review of laboratory results from several studies indicates that, overall, the physiology of juvenile clones reflects normal, appropriate responses to ongoing growth and developmental signals, and for most of the juvenile period, bovine clones are functionally indistinguishable from non-clones. A possible exception is the period of transition from Perinatal to the Juvenile node, during which some physiological differences between clones and conventional cattle have been identified.  Almost all of these differences are transient and are resolved in the first two weeks of life.  Thermoregulation in clone calves appears to normalize within two months of birth (Chavatte-Palmer et al. 2002; Cyagra 2003). Although there may be some physiological differences between clones and their comparators during the transition between the perinatal and juvenile developmental nodes, none of these differences indicate the presence of any subtle or frank food consumption hazards. 

iv. Reproductive Development and Function in Bovine Clones (Developmental Node 4)

(a) Peer-reviewed Publications

The number of studies that explicitly address the reproductive function of bovine clones is smaller than studies of other endpoints. Puberty onset has been reported as either “within normal limits” or somewhat (days) later in clones than controls. The Cyagra data received do not explicitly address the question of puberty onset or reproductive capability.

Reproductive Function of Female Clones

In a study of reproductive function in bovine clones, Enright et al. (2002) at the University of Connecticut evaluated the same clones and controls previously reported on by Xue et al. (2002) and Govoni et al. (2002). They reported that heifer clones reached puberty at a later age than controls (314.7 ± 9.6 days vs. 272 ± 4.4 days), and were reported as having higher body weights at first estrus (336.7 ± 13 vs. 302.8 ± 4.5 kg). No differences were noted between clones and controls in estrous cycle length, development of ovarian follicles, or profiles of hormonal changes. Three of the four clones and all four control heifers became pregnant following AI, although number of inseminations was not reported. Daily hormone profiles of lutenizing hormone (LH), follicle stimulating hormone (FSH), estradiol, and progesterone were similar between clones and controls. The cause of reproductive failure in one clone could not be determined; although this animal had reproductive hormone profiles similar to the other animals in the study, and no physical abnormalities could be found upon veterinary examination, poor signs of estrus were observed. This heifer did eventually conceive and produce a calf (Tian et al. 2005, further discussed below). The cause for the later age and higher weight of clones at the time of puberty is difficult to explain. The authors speculated that as the later onset of puberty can be genetically controlled in some cattle breeds, these clones may be expressing the genetics of the donor animal. Given that no records of age at puberty were kept for the donor cow, it is not possible to draw any conclusions regarding that hypothesis.

Heyman et al. (2002a) reported that from a group of clones derived from adult cells, five remaining animals were healthy and normal (one clone died of severe anemia (Renard et al. 1999), as previously discussed in the Perinatal section). They noted that some of the females were more than one year old at the time of publication and were cycling normally, but no data were provided. In a follow-up study (Heyman et al. 2004) the authors stated that female clones at the INRA facility generally began cycling at 10 months of age, and demonstrated estrous behavior by 12 months of age, within the normal range for their breed (Holstein). Ten female clones were bred by AI to the same non-clone bull. All 10 heifers conceived and produced live, apparently normal calves. Birth weight of progeny was 43.9 ± 4.1 kg, and gestation length was 281 ± 3.9 days, within the normal range for Holstein cattle.

The timing of puberty in clones at the INRA facility was compared to control heifers by Heyman et al. (2007). From eight to 15 months of age, 10 clones of three different genotypes and 11 AI-derived control heifers were observed twice daily for estrus, and plasma progesterone was measured at 10-day intervals. Onset of estrous cycles occurred 63 days later in clones compared to control heifers (419.3 ± 42.5 days vs. 356.5 ± 50.5 days, respectively) and cloned heifers were 56 kg heavier than controls at the time of puberty (359.0 ± 38.9 kg vs. 303.0 ± 22.3 kg, respectively).  These differences cannot be attributed to differences in growth rate since daily gain and feed intake up to 15 months of age were similar between groups.  Mean estrous cycle length was similar between clones and controls.  These results are consistent with those from the cohort of clones at the University of Connecticut (Enright et al. 2002) and provide further evidence, at least for Holstein heifers, that cloning may alter the timing of puberty.

Wells et al. (2004) reported conception rate to two AI was 83 percent (25/30) for Holstein heifer clones, compared to 90 percent (9/10) for as small group of heifers produced by AI. Gestation length was slightly longer for clones (n=16) than for nine comparators (287 ± 3 vs. 281 ± 3 days), but within the normal range for Holsteins. Wells (2005) notes that despite variations in gestation length, only conventional levels of animal management and husbandry are required for the calving of heifer clones, indicating that the signals for induction of parturition and actual birth are functioning appropriately. Although most of the clones were separated from their offspring soon after birth, as is conventional in dairy practice, those dams that were not separated from their progeny exhibited normal maternal behavior and successfully reared their young.

Forsberg et al. (2002) reported that Gene, the bull calf described previously, matured into a “healthy, fertile bull that has sired calves by artificial insemination and in vitro fertilization.” Specific data on measures of reproductive function were not provided.

Kato et al. (2000) report that one of the clones derived from a Holstein cumulus cell was artificially inseminated, conceived, and gave birth to a normal calf

The University of Connecticut (Tian et al. 2005) also reported first lactation milk yields and SCC for four clones and their non-clone comparators, indicating that lactation curves were similar for both groups. Total milk production for the first lactation was not different between clones and comparators (8,646 ± 743.8 kg vs. 9,507.8 ± 743.8 kg). One clone gave birth prematurely to a stillborn calf, did not have complete udder development, and produced approximately 30 percent less milk during her first lactation compared to her clone mates. Overall, SCC was low for both clones and comparators (based on Figure 2b of the paper: ~ 40 x 103 vs. 35 x 103 cells/mL), indicating a functional immune system, mammary gland, and low disease incidence. The role of good husbandry can also not be ruled out in this observation.

Heyman et al. (2007) reported that one of the 21 heifer clones heifers in their three-year prospective study at INRA died shortly after calving. The exact cause of death is not specified, but the authors point out that this heifer calved during the unusually hot summer months of 2003.