Appendix D: Transgenic Clones
The current risk assessment is limited to address the risk of clones from non-transgenic cells. Although not within the purview of this analysis, the results of a number of studies that address either transgenic animal clones or transgenic and nontransgenic animal clones (Hill et al. 1999; the "ACT series" including Cibelli et al.1998, Lanza et al. 2000, and Lanza et al. 2001 for cattle; McCreath et al. 2000 and Denning et al. 2001 in sheep; Baguisi et al. 1999 and Keefer et al. 2001a in goats; Carter et al. 2002, Lai et al. 2002, and Lee GS et al. 2003 in swine) are presented here to clarify the relative utility of such studies for assessing potential risk(s) associated with non-transgenic somatic cell nuclear transfer (SCNT).
Many reviews of SCNT outcomes cite these papers as indicative of the severity of adverse outcomes associated with cloning, or a demonstration of the positive outcomes that can come from cloning. Hill et al. (1999), in particular, is often cited as the seminal "adverse outcome" paper for cloning. On the other hand, if only the "final report" paper of the ACT series, Lanza et al. 2001, were read, it would not be possible to know that the cells from which the cattle were cloned were indeed transgenic. This distinction only becomes apparent when the earlier papers are also reviewed. Most recently, headlines were generated when Pearson (2003) reported on "sudden death syndrome" in pig clones; CVM reviewed the paper that describes their generation and noted that these pig clones carried two distinct transgenes (Lee GS et al. 2003).
Because these animals are transgenic clones, it is not possible to determine whether adverse outcomes result from the direct effect of the expression of the transgenic construct, pleiotropic effects resulting from insertion of the construct, the SCNT process, or some interaction of any or all of these processes. For example in a comparison of 34 cells lines, Forsberg et al. (2002) reported that when otherwise similar cells are used as donors for SCNT, those that are transgenic (for a selectable marker gene) result in lower pregnancy initiation (22 percent vs. 32 percent) and calving (3.4 percent vs. 8.9 percent) rates. The authors hypothesize that the additional culturing required togenerate transgenic cells, selection of transgenic lines, or the DNA construct itself, could be responsible for the lower rates.
CVM thus assumes that transgenic clones occupy a different "risk space" from "just clones." Conversely, an argument can be made that if no adverse outcomes are detected, then for these animals, neither process sufficiently perturbed development to induce anomalies. We have included those studies in the overall risk assessments if such results were obtained. Nonetheless, because these studies occupy such a large segment of the cited literature, a few are presented here to illustrate the range of responses noted, with the appropriate caveats for interpretation.
Hill et al. 1999
Hill et al. 1999 reports on a group of 13 transgenic clones of a Holstein bull. Twelve Brangus cows carrying 13 fetuses cloned from Holstein cells were originally included in the study, although three of these transgenic clone fetuses died prior to the perinatal period (defined in this paper as two weeks prior to anticipated delivery and a few days thereafter), and one cow aborted at eight months of gestation. Two cows developed hydroallantois and were delivered by C-section; four others were also delivered by this method due to subjective judgment regarding fetal size. The two remaining pregnancies delivered vaginally. Birth weights of the transgenic clones ranged from 44-58.6 kg (average Holstein male calf weight is in the range of 40-50 kg), and cited as within the weight range of in vitro produced embryos. Five of the eight live born clones were judged to be normal within four hours of birth based on clinical signs and blood gas measurements. Three of the eight were immediately diagnosed with neonatal respiratory distress. One of these calves died from pulmonary hypertension, pulmonary surfactant deficiency, and elevated systemic venous pressure at day 4. The other three animals recovered. Two of the five fetuses that did not survive to birth also exhibited signs of pulmonary hypertension and placental edema at necropsy. Another clone died at 6 weeks of age with signs of respiratory distress; subsequent field necropsy suggested dilated cardiomyopathy, although no definitive diagnosis could be made.
The Advanced Cell Technology Series
The series of papers from the Advanced Cell Technology group (Cibelli et al. 1998, Lanza et al. 2000, and Lanza et al. 2001) on the health of clones are similar to that of Hill et al. (1999) in that the animals presented are clones that are derived from transgenic cells. Interpretation of any adverse outcomes is thus also confounded by the potential role of the transgene and its insertion.
The results of these studies are summarized in Lanza et al. (2001), in a short overview with accompanying supporting documentation provided by the journal in electronic form. Of 30 fetal transgenic clones that developed to term, 24 were reported healthy at 1-4 years of age, but five died within 7 days of cardiopulmonary difficulties that the authors speculated were secondary to placental insufficiencies. The sixth animal died at day 149 due to enteric disease, lymphadenopathy, and exhibited mild placental edema and high fever at birth. Problems observed at birth included placental edema, including edematous cotyledons (attachment sites of the placenta to the uterus), labored breathing, froth and fluid in the lungs, pulmonary edema, pneumonia, high fever, septicemia, lethargy, abdominal distention, masses in the abdomen, liver damage due to hypoxia, and heart abnormalities.
Birth weights of the survivors were reported as 45 ± 2 kg (this paper cites normal as 43 kg). An unspecified number exhibited pulmonary hypertension and respiratory distress at birth. Presumably, they received supportive care at that time. Another unspecified number were also reported as experiencing fever following vaccination. This is not an atypical response among calves receiving vaccinations, as stimulating a potent immune response is likely to produce at least a mild local and systemic (fever) reaction in the animals (Roth 1999).
Physical and veterinary examination of surviving animals aged 1-3 years were reported as normal and included temperature, pulse, and respiratory rate. No abnormalities were detected in general appearance, on auscultation (listening to breathing, heart beat, and digestive sounds), and behavior appeared normal. Puberty onset was reported to occur at the expected time, and fertility appeared to be normal. At the time of publication (2001), two of the animals had delivered apparently normal progeny.
Clinical chemistry parameters evaluated for these animals included electrolytes, urea, creatinine, glucose, bilirubin, aspartate aminotransferase (AAT), sorbitol dehydrogenase (SGT), albumin, globulin, and total protein. Globulin and total protein measurements were reported in the publication as "slightly below normal." All other measurements were reported to be within normal range. Hemograms (analysis of cellular components of blood) were all reported as normal: hematocrit, hemoglobin, red blood cells, mean red cell volume, mean red cell hemoglobin concentration, and white blood cell numbers and differentials were within normal ranges. Blood gases were also within normal ranges. To examine immunocompetence in the clones, peripheral blood lymphocytes from the transgenic clones and conventional Holsteins were compared to determine whether the same ratio of cell surface markers were present, and if the transgenic clone cells responded to mitogen challenge in the same way as cells from conventional Holsteins. No significant differences were observed between the cell surface markers or cellular responses of cells from conventional animals or clones.
In the early spring of 2003, an interview of an ACT executive reported in the lay press indicated that two animals from this cohort had developed significant health problems. One animal was reported to have developed a tumor, and the other was diagnosed as having neurological problems. The first animal apparently died during surgery to remove the tumor, and no further information is available on the potential causes of the tumor. The second animal was later diagnosed as being positive for Johne’s disease (Mycobacterium paratuberculosis), an infectious, chronic, progressive disease that often presents with chronic diarrhea and eventual cachexia (general physical wasting and malnutrition). It is therefore unlikely that this animal’s symptoms were due to either cloning or transgenesis. We are unaware of any other adverse outcomes associated with these animals.
Carter et al. 2002
Carter et al. 2002 reported on the overall health status of transgenic swine clones produced from cells transfected with green fluorescent protein (GFP). The 10 transgenic piglet clones from three litters were followed for the first six months of life.
Five of the ten transgenic swine clones died or were euthanized during the study. Two piglets died of congestive heart failure at 7 and 35 days of age, two others died from bacterial infections at 3 and 116 days of age. The fifth animal died at 130 days of age, following a history of chronic diarrhea, decreased growth and vitamin E deficiency. The remaining five piglets were reported as healthy and growing similarly to conventional animals housed in the same facility at the conclusion of the study. Behavior was reported as "consistent with pigs of their age group."
Average birth weight of the transgenic clones (1,312 g) was similar to average birth weights of conventional piglets from similar genetic background (1,450 g). Average daily weight gain for transgenic clones through the first 16 weeks was (461 g) relative to the herd average (594 g), which the authors considered as within the normal range.
Some of the piglets displayed physical defects. These included two piglets with contracture of the flexor tendons, another piglet with five digits on a forelimb (four digits are normal) and an enlarged dewclaw. Another piglet with low birth weight was described as having short legs and a large, round chest.
Hematology and blood clinical chemistry data were collected beginning at 2 days of age and every two to four weeks until 24 weeks of age. Most hematological variables were similar to the comparator group, except for hemoglobin, hematocrit, and plasma total protein. Mild anemia and low blood protein concentration were observed for the first four weeks, but both these conditions resolved by eight weeks of age. The authors stated that decreased hematocrit and hemoglobin values are common in piglets reared in confinement, and that these symptoms are generally treated with iron dextran. Similarly, clinical chemistry results indicated decreased levels of albumin and globulin during the first four weeks in the transgenic clones relative to comparators, but these values were back within the normal range by eight weeks of age. The authors attributed the decreased protein and globulin values to the decreased colostrum intake of the newborns as the surrogate sow bearing them did not initiate normal lactation, and piglets were dosed with colostrum at some unspecified point after birth.
Seven of the transgenic clones were evaluated for cardiac function. Although no physical defects were found, one piglet had evidence of mitral insufficiency (a condition in which the mitral valve of the heart does not close all the way during contraction, resulting in regurgitation of some of the blood in the left ventricle), and dilation of the left atrium and ventricle. This piglet and two other clones had reduced cardiac output values compared to control piglets, but did not display clinical signs of cardiac disease. Although similar cardiac abnormalities have been noted in conventional swine, the incidence is reported to be very low (Carter citing Hsu et al. 1982). These developmental defects appear to be similar to those noted in cattle clones (see Critical Biological Systems discussions).
Lai et al. 2002
This study was reported in a brief communication, and a limited amount of data was presented. Piglets were generated from cell lines (derived from inbred miniature pigs) in which the a-1-3-galactosyltransferase gene was interrupted by the insertion of a gene sequence in order to create a-1-3-galactosyltransferase "knock-outs." The a-1-3-galactosyltransferase gene codes for a protein that causes hyperacute rejection of swine organs when transplanted into primates. "Knocking out" the expression of this gene increases the suitability of these animals to be used as donors of organs for human transplant patients.
Six piglets were born from two litters. All but one of the piglets had low birth weights compared to the breed average (115 to 650 g vs. 860 g). One piglet from each litter died shortly after birth from what the authors termed "respiratory distress syndrome." A third piglet died at 17 days of age during a routine blood draw, and was diagnosed at necropsy with a dilated right ventricle and thickening of the heart wall. Other abnormalities noted in these surviving transgenic piglets included flexor tendon deformities in three animals; abdominal ascites, enlarged right ventricle, pulmonary hypertension in one animal; and ocular defects and lack of patent ear canals in another animal. The authors attributed these abnormalities to failures in reprogramming during the SCNT process rather than the genetic engineering process, as they did not see a consistent phenotype across the piglets.
Lee GS et al. (2003)
Recently, Pearson (2003) reported that the University of Connecticut laboratory that had generated four transgenic swine clones had announced that the three (of four) surviving piglets died suddenly of heart failure at less than six months of age. The fourth piglet died at three days due to infection and abnormal spine development (Lee GS et al. 2003). Because of the transgenic nature of the animals (they carried genes for human clotting factor IX and porcine lactoferrin, an iron transport protein found in blood), it is not possible to attribute the deaths solely to cloning. It is unknown whether any cardiac abnormalities were detected in these animals prior to their deaths, or if any measurements of cardiac function were made.
Denning et al. 2001
Denning et al. 2001 were unsuccessful in producing viable knock-out sheep lacking either the a-(1,3)-galactosyl transferase (GGTA1) or the prion protein (PrP) gene using gene targeted fetal fibroblasts and SCNT. Reconstructed embryos were either incubated for six days (n=48) or overnight (n=93) in synthetic oviductal fluid with bovine serum albumin (concentration not specified). Embryos incubated overnight in vitro were then embedded in 1 percent agar chips in phosphate buffered saline and transferred to the ligated oviduct of an estrus-synchronized ewe for six days. A total of 120 morula or blastocyst stage embryos were transferred to 78 estrus-synchronized Finn Dorset ewes as final recipients. It is not clear from this paper how many of the transferred embryos had been incubated in vitro. Although 39 pregnancies were diagnosed at gd 35, only eight were maintained to term, resulting in four live births. Three of the four live-born lambs died shortly after birth. The fourth lamb survived 12 days before it was euthanized after developing dyspnea (difficulty breathing) due to pulmonary hypertension and right-sided heart failure. The authors attributed the abnormalities observed to the nuclear transfer procedure, as they were similar to results obtained with non-transgenic NT lambs.
McCreath et al. 2000
McCreath et al. 2000 inserted a promoter-less neomycin selectable marker between the ovine a1(I)-procollagen translational stop and polyadenylation signal1 in male and female ovine fetal fibroblast cultures. Four transgenic female fibroblast cultures were selected as nuclear sources for SCNT, due to their vigor and normal chromosome number. A total of 80 morula and blastocyst stage embryos were transferred to recipient ewes. No description of post-fusion incubation or estrous cycle status of recipient ewes was provided in this report. Fourteen lambs were born alive; seven of these lambs died within 30 hours of birth. Four more lambs died between 3 days and 12 weeks of age. Three lambs survived and were described as thriving at one year of age. Necropsy of lambs that died in utero or after birth revealed a number of abnormalities including a high incidence of kidney defects (frequently renal pelvis dilation) and liver and brain abnormalities (not specified). The authors attribute these abnormalities to either cell treatment or the NT procedure, because the necropsy findings were similar to a previous nuclear transfer study using the same cell lines.
Baguisi et al. 1999
In this study from Genzyme Transgenics, six cell lines were established from 35- and 40-day old fetuses that resulted from the mating of a transgenic buck (carrying a human antithrombin III (hAT) gene with a goat ß-casein promoter) to a non-transgenic doe. This study differs slightly from several other transgenic cloning studies reported here, in which the gene was inserted into the cell lines before the cultures were established. Clone embryos were cultured on goat oviduct epithelial cells for 48 hours (2-16 cell stage) before being transferred to estrus synchronized recipient does. Although overall cloning efficiency was low (3/112 embryos transferred resulted in live births), all pregnancy losses occurred prior to 60 days of pregnancy. There were no stillbirths and no abnormalities observed in the live-born kids. Kids weighed between 2.35 and 3.5 kg, within the normal birth weight for dairy goats, and are reported as healthy.
Keefer et al. 2001a
In this study from the Nexia Biotechnologies laboratory, goat fetal fibroblasts were transfected2 with green fluorescent protein (eGFP) and neomycin resistance genes. These are commonly used as markers to demonstrate that transgenes have been inserted and are being expressed. Twenty seven NT embryos were produced with the transfected cells, and an additional 70 non-transgenic NT embryos were constructed and transferred into 13 estrus synchronized recipient does. The authors did not specify how many embryos (transgenic or non-transgenic) were transferred to each doe. Five non-transgenic male clones and one transgenic female clone were born alive. Three of the non-transgenic clones died of bacterial infections, but the single female transgenic clone lived and showed no signs of abnormalities. The kids were all within the normal birth weight range (1.5 to 3.1 kg) for goats at that facility, and no abnormalities were observed in the placentae.
Behboodi et al. 2004
The authors compared development of embryos cultured with oviductal cells in vitro vs. embryos cultured in vivo. Embryos were constructed using skin fibroblasts of transgenic goats. Only embryos cultured in vivo resulted in pregnancies. Two of these pregnancies were lost early in gestation (after 30 days gestation), and four other pregnancies were carried to term. Two surrogate does delivered stillborn kids 2-3 days after their due dates; the other two does delivered healthy kids (one per each doe) at term. The two live clones weighed 3.8 and 4.1 kg at birth, and were within the normal birth weight range for their breed (Saanen). Clones were weaned at 8 weeks of age, and had similar growth rates compared to age-matched AI derived Saanen kids born at the same facility (14.5 and 18.1 kg for clones vs. 14.88 ± 1.98 kg for AI comparators). Pathology on the dead fetuses indicated diffuse atelectasis (lung collapse) and the presence of amniotic fluid in the lungs. No bacterial or viral cause for the deaths of these clones could be identified. The presence of amniotic fluid in the lungs suggests that the clones attempted to breathe prematurely, a sign of fetal stress which sometimes occurs around the time of birth.
Behboodi et al. 2005
The authors evaluated health, growth, reproduction and lactation in four female goat clones generated from two transgenic fetal cell lines (one cell line coding for glycosylated and the other for non-glycosylated protein). A total of seven clones were carried to term. One clone from the glycosylated group was still born with evidence that the umbilical vessel had ruptured. Two clones died at birth (one from each of the transgenic lines) after failing to breathe on their own, despite attempts at manual ventilation. Thus, two clones from each transgenic line survived to adulthood. There were no differences in birth or weaning weights among the four surviving clones or their age-matched comparators. Transgenic clones exhibited enlarged umbilical stumps (two live and one stillborn kid), "tendon laxity" (three of the four live-born clones), and minor generalized edema (number of clones affected not indicated). These conditions resolved without intervention. The four does were bred and produced nine kids, compared to five kids produced by comparators. Clones expressing the glycosylated version of the protein lactated only briefly, but the does expressing the non-glycosylated protein had normal lactation length and milk yields.
This study is the only one we encountered that presented hematology and blood clinical chemistry data for four goat clones. These data are presented in comparison to four age-matched comparators and values from the literature (Pugh 2002). It is unclear whether or not the comparators in this study were also transgenic, whether they were the same breed as the clones, or how they were generated. Hematology values were similar between clones and comparators, and all hematology values fell within the published range. For clinical chemistry, 18/24 values were not significantly different between clones and their age-matched comparators. Of the 19 clinical chemistry values for which published ranges were available, 18 of the values for clones and comparators fell within the published range. The one value out of the published range was creatine kinase (244.6 vs. 204.4 IU/L for clones and comparators). However, values between clones and comparators were not statistically different. The study does not specify the age of the goats at time of blood sampling, so it is difficult to interpret the high values for CK in these animals compared to the published range.
This study is unique among reports of goat clones because it is the first to indicate possible signs of LOS in goat clones (enlarged umbilici, failure to initiate breathing, tendon problems). It is interesting to note that similar signs have not been noted in non-transgenic goats. We should also note that clinical signs in the four surviving clones resolved, and their health, growth, reproduction, and hematology, clinical chemistry values indicate that even these transgenic clones are apparently normal.
Landry et al. 2005
The authors reported on growth (weight gain, wither and hip height change) and endocrine profiles of two lines of transgenic goat clones. Group 1 consisted on five does carrying the AT-III gene with a ß-casein promoter inserted into cells of a female Toggenburg (dairy breed). The gene inserted into the second line of goats (Group II; n=2) was not identified, but the cells used for cloning were from a female Saanen (also a dairy breed). Non-transgenic, non-clone comparators (n=7) were Boer X Spanish crossbred meat-type does (Group III). The authors did not report on overall health of the clones. One female in each group of clones died prior to the end of the study; one died due to an accident, the other due to a ruptured abomasum. Neither death appears to be related to cloning. Both groups of clones were within range for their breed for birth weight, and appeared to grow normally. Interpretation of hormone profiles (GH, IGF-I, T3, and T4) is difficult due to the fact that the clones and comparators were of different breed (purebred vs. crossbred) and type (dairy vs. meat) backgrounds. However, for most of the hormones assayed, the values for clones fell within the range of values for comparators. The one exception is insulin, which resulted in an extremely low value in blood samples of comparators, and may have been the result of difficulties with the assay.
Melican et al. 2006
With the long-term goal of producing caprine milk containing recombinant therapeutic proteins, the authors demonstrated successful reproduction and lactation in transgenic dairy goats produced by SCNT. Does were produced using primary fetal cells harvested from day 35-40 fetuses. These cells were co-transfected with DNA fragments encoding the heavy and light immunoglobulin chains of three different monoclonal antibodies, plus neomycin resistance as a selectable marker. Two transgenic does were hormonally induced to lactate at two to three months of age and produced small amounts of milk (10-20 ml per day) for 30-40 days. These does were subsequently bred and, at 18 months of age, gave birth to 6 kids (4 were transgenic). Following parturition, the 2 transgenic does produced an average of 2.2 liters of milk per day for 3-6 months. A third transgenic doe, at age 23 months, also lactated following the birth of one transgenic kid, producing 2.6 liters of milk per day for one month prior to being dried off. The milk of all 3 transgenic does contained the antibody for which they were transgenic. These results demonstrate reproductive capacity in cloned female goats, and the ability of these goats to produce normal quantities of milk following parturition.
Conclusions Regarding Transgenic Clones
The experience of these cohorts of transgenic clones can be summarized as follows:
- A relatively large fraction of transgenic fetal bovine clones in cohorts surviving to late gestation presents with severe and often fatal difficulties. Some of these are qualitatively similar to those observed in cattle and sheep clones that are not derived from transgenic cells. Due to the many other variables that have been altered in the generation of these animals, at this time it is not possible to attribute these abnormalities to either of the processes (cloning or transgenesis) or their combination.
- Some animals in both cattle cohorts are born with varying degrees of initial respiratory or other physiological distress. Supportive care appears to allow most of these animals to survive to adulthood, although some animals that initially survive can succumb to possible sequellae up to six weeks later.
- Animals surviving to adulthood in the ACT cohort that appear to be healthy on visual inspection also exhibit physiological values that generally fall within normal ranges. CVM is unaware of an update of the health status of the Hill et al. cohort.
- Animals in the ACT cohort surviving to reproductive maturity appear to be capable of bearing normal offspring, although it is not clear whether the offsprings’ health has been examined in a rigorous manner.
- Two severe adverse outcomes have been noted for the ACT cohort. Both cloning and transgenesis can likely be ruled out as causes for one (Johne’s disease) and no causal agent or process has been associated with the neoplasm found in the other.
- The appearance, behavior, and physiological function of the animals that survive suggest that even the "riskiest" set of clones (i.e., transgenic clones) can develop into normally functioning animals. These results are consistent with the analysis of non-transgenic clones, and provide additional confidence that rigorous monitoring and responsible husbandry of such animals can allow for the selection of animals that are healthy.
- Abnormalities for transgenic sheep clones appear similar to reports for non-transgenic sheep and cattle clones.
- Goats appear to suffer fewer adverse effects compared to sheep and cattle. Of the reports reviewed, only one cohort exhibited clinical signs of LOS.
- Abnormalities reported for transgenic swine clones are similar to those reported for transgenic and non-transgenic cattle clones. CVM is aware of only one report in non-transgenic swine clones (Park MR et al. 2005) in which clones exhibited similar health problems; however, in vitro methods used in this study likely influenced the outcome of swine clones in this study.