Limitations of serial cloning in mammals

Introduction

Asexual reproduction, including cloning, enables many species such as potatoes to thrive through efficient mass reproduction systems^1,2. Even in animals, cloning from a body part, that is, producing offspring from somatic cells, is observed in a few lower species, such as pot worms and planaria^3,4. However, higher animals, e.g., mammals, cannot reproduce clonally from somatic cells in nature, and the reason for this remains unknown^5,6.

Until recently, mammalian cloning was not possible except in cases such as monozygotic twins⁷. However, following the development of nuclear transfer technology and the successful somatic cell cloning of frogs⁸, Dolly the cloned sheep⁹, and Cumulina the cloned mouse¹⁰, it has become possible to produce cloned animals even from terminally differentiated somatic cells¹¹. This technology can be used to rescue endangered species^12,13 or infertile individuals¹⁴ by increasing population numbers from somatic cells. Cloning is also thought to be useful for the large-scale production of superior-quality domesticated animals^15,16, potentially making expensive meat, such as Wagyu beef, more affordable^17,18.

However, mammalian cloning has not yet been put into practical use, not only due to its low cloning success rate, but also because of anomalies¹⁹ not observed in the cloning of plants or lower animals. These issues are believed to stem from failures in “reprogramming” the donor nucleus to reset the epigenetic state of differentiated somatic cells to that of a fertilized embryo; this occurs not directly due to DNA abnormalities, but the detailed mechanisms responsible remain unclear²⁰.

If DNA abnormalities do not occur and epigenetic errors do not accumulate, mammals should be able to maintain their species through cloning. However, early studies showed that serial cloning in any mammalian species reached its limit after only a few generations^{21,22,23,24,25,26}. This suggests that cloning or asexual reproduction may not be viable for maintaining mammalian species. Nevertheless, by improving cloning success using an epigenetic modification reagent that promotes nuclear reprogramming²⁷, we succeeded in serially cloning up to 25 generations of mice²⁸. Since the success rate improved slightly with each successive generation, we initially concluded that serial cloning could be continued indefinitely²⁸. However, after continuing the experiment for 20 years, the success rate began to decline and eventually could not be sustained beyond 58 generations.

In this study, we aimed to investigate in detail what occurred in re-cloned mice after extended serial cloning, and to clarify why mammals maintain their species through sexual rather than asexual reproduction.

Results

Success rate of serial cloning up to 58 generations

We began serial cloning experiments (Fig. 1a) in January 2005. The success rate of re-cloned mice gradually increased as generations progressed (up to 15.5% in the 26th-generation), and NGS analysis showed no difference between first-generation clones and 25th-generation re-clones. Therefore, we initially believed that serial cloning could probably continue indefinitely²⁸, and thus continued this experiment. To maintain consistent experimental conditions as much as possible, TW and SW always participated in the production of next generation re-cloned mice. Approximately 3-4 generations of re-cloned mice were produced each year. However, the cloning success rate gradually decreased with each successive generation (Fig. 1b). Although success rates varied between generations, the average success rate declined to 0.6% after 57 rounds of serial cloning, and all 58th-generation re-cloned mice died the next day. In total, more than 1200 cloned mice were produced from a single original donor mouse (Supplementary Table 1). The original donor mouse was female with an agouti coat color; therefore, all cloned mice were female and exhibited the same coat color as the donor. Whole-genome sequencing (WGS) analysis also confirmed that these clones originated from the donor mouse (see later).

**Fig. 1: Success rate of mouse serial cloning and their life span.**

Placenta and lifespan of re-cloned mice

All cloned mice have been found to possess abnormal placentas arising from epigenetic errors^29,30. We therefore measured the body and placental weights of most cloned mice at the time of cesarean section. As controls, body and placental weights were also measured for mice derived from in vitro fertilized control embryos (FC), newly generated first-generation cloned mice (NG1), and 25th-generation re-cloned mice derived from the cryopreserved cumulus cells of G24 cloned mice (CryoG25). The mean body weight across re-cloned generations was ~1.88 g, slightly heavier than that of FC mice (1.3 g) but similar to other control clones (1.68–1.96 g) (Supplementary Table 2). The mean placental weight was about 0.30 g, which is two to three times heavier than that of FC mice (0.11 g), but comparable to other control clones (0.25–0.29 g). However, neither body nor placental weight increased over successive generations, suggesting that abnormalities did not accumulate. Histological examination and measurement of the area of labyrinth of these placentas showed slightly expanded the area of labyrinth zone when increased the number of serial cloning but no significant differences between NG1, CryoG25, and G50 cloned/re-cloned mice (Fig. 1c, Supplementary Fig. 1), despite expansion of the spongiotrophoblast layer, an abnormality specific to cloned mice³¹. However, these placentas were significantly different compared to those of FC mice. Finally, we also examined the average lifespan of randomly selected re-cloned mice across generations. The average was approximately two years (Fig. 1d, Supplementary Table 3), similar to that of the control cloned mice, suggesting no abnormality.

Effect of trichostatin A (TSA) on re-cloned mice

One possible explanation for the decreased success rate in later generations is that TSA, a known enhancer of nuclear reprogramming²⁷, may lose its effectiveness in supporting the development of re-cloned embryos. When nuclear transfer was performed without TSA, the success rate for generating G51 clones was 1.6%, similar to that of NG1 (0.95%) and significantly lower than with TSA treatment (G51: 5.4%; NG1: 3.9%) (Fig. 2a, Supplementary Table 4). Thus, even after more than 50 nuclear transfer cycles, the somatic cell nuclei of G50 re-cloned mice retained sensitivity to TSA.

**Fig. 2: Quality comparison of FC embryos, NG1 clones, and G51 re-cloned embryos.**

Epigenetic abnormalities in re-cloned zygotes produced by G50 cloned mice

We next examined whether the decreased cloning success was due to accumulated epigenetic abnormalities, as cloned embryos are known to differ epigenetically from fertilized embryos^20,32,33. To investigate this, FC zygotes, NG1 clones, and G51 re-cloned zygotes were fixed and immunostained with antibodies targeting trimethylated histone H3 at lysines 4, 9, and 27 (H3K4me3, H3K9me3, H3K27me3), acetylated histone H3 at lysine 14 (acH3K14), and stained with DAPI. As shown in Fig. 2b, c, and Supplementary Fig. 2, the acetylation and methylation levels were similar between NG1 and G51 cloned/re-cloned zygotes, except for H3K9me3. However, both groups differed markedly from FC zygotes. These results suggest that epigenetic abnormalities did not accumulate in cloned/re-cloned embryos, at least at the one-cell stage.

In vitro development and quality of G51 cloned embryos

NG1 cloned embryos and G51 re-cloned embryos were cultured in vitro until they reached the blastocyst stage and were then compared with FC embryos. As shown in Fig. 2d and Supplementary Table 5, the developmental rate into expanded blastocysts was lower for NG1 and G51 embryos than for FC embryos. However, there were no differences between NG1 and G51 groups, suggesting that serial cloning up to 50 cycles did not impair in vitro embryo development.

To evaluate embryo quality, some blastocysts were analyzed for cell number and inner cell mass (ICM) allocation. ICM distribution did not differ clearly between groups (Fig. 2e, f, Supplementary Table 6). Although the average total cell number was significantly reduced in cloned/re-cloned embryos relative to FC embryos, no difference was observed between NG1 and G51 embryos. Interestingly, the average ICM cell number was slightly higher in cloned/re-cloned embryos compared to FC embryos.

RNA-seq analysis of G51 cloned blastocyst

To detect differences between FC embryos, NG1 clones, and G51 re-cloned embryos, we analyzed the whole-genome gene expression profiles of blastocysts using RNA-seq (Fig. 3a). Relatively well-developed blastocysts were selected, and RNA was extracted individually from each blastocyst. RNA-seq was performed, and the three groups were compared using principal component analysis (PCA) to identify gene expression differences. PCA showed that not only NG1 (green dots) and G51 (blue dots) clone/re-cloned embryos exhibited similar gene expression profiles, but these were also similar to those of FC (red dots) embryos.

**Fig. 3: Reproductive competence of cloned mice and the normality of their progeny.**

Production of offspring from re-cloned mice by sexual reproduction

Although serial cloning could not continue beyond the 58th generation, the re-cloned mice remained healthy except G58, raising the possibility that subsequent generations could be produced via sexual reproduction. Therefore, FC mice, as well as G20, G50, and G55 re-cloned mice, were mated with normal male mice to evaluate pregnancy and offspring production. There was no difference in age at first delivery among the groups, and litter sizes were similar between G20 re-cloned mice (9.9) and FC mice (10.3). However, litter sizes declined sharply in G50 and G55 re-cloned mice (2.8 and 2.2, respectively) (Fig. 3b, c, Supplementary Table 7). The number of oocytes recovered from each generation was recorded when collecting cumulus cells as donor cells (Supplementary Table 8). Although superovulation was triggered by hormonal treatment, the number of oocytes did not decrease even with further generation of re-cloned mice. Therefore, the decrease in litter size was not due to a reduced number of oocytes ovulated from the re-cloned mice.

Interestingly, when the next generation (grandchildren) was produced from offspring of G55 re-cloned mice, the litter size increased to 7.0 (Fig. 3c). Moreover, histological analysis showed that placentas from the grandchildren were clearly smaller than the large placentas characteristic of cloned mice and were nearly the same size as those from fertilized embryos (Figs. 1c and 3d, Supplementary Fig. 1).

Normality of re-cloned mouse oocytes

To examine whether oocytes derived from near-final-generation re-cloned mice were of reduced quality and/or prone to fertilization or developmental failure, we collected oocytes from CryoG25 and G53 re-cloned mice and subjected them to artificial activation and diploidization for parthenogenesis (Fig. 4a). If mutations accumulate through serial cloning, diploidization can reveal homozygous or defective mutations in oocytes, allowing assessment of their effects on early development. As a result, 91.3% of parthenogenetic embryos from BDF1 control mice developed into blastocysts, while only 12.7% of embryos from G25 re-cloned mice and none from G53 re-cloned mice developed into blastocysts (Fig. 4c, e, f, Supplementary Table 9).

**Fig. 4: Normality of oocytes collected from re-cloned mice.**

Next, oocytes from re-cloned mice were fertilized in vitro with sperm from normal males (Fig. 4a). Even if oocytes carried many lethal mutations, normal sperm genes could improve blastocyst development. As a result, 97.6% of BDF1 control embryos reached the blastocyst stage, compared with 39.7% from G25 re-cloned mice and 20.3% from G53 re-cloned mice (Fig. 4c, g, h, Supplementary Table 9). These results clearly indicate that the number of lethal mutations in oocytes increases with each generation of serial cloning.

Finally, to determine whether the abnormalities originated from the oocyte nucleus (Metaphase II; MII) or cytoplasm, we performed MII replacement between oocytes from G56 re-cloned mice and fresh BDF1 mice (Fig. 4b). When the MII from one BDF1 oocyte was removed and immediately transferred into same oocyte, 68.2% of parthenogenetic embryos developed into blastocysts, confirming normality of both nucleus and cytoplasm (Supplementary Table 10). When MII from G56 re-cloned oocytes was transferred into enucleated BDF1 oocyte cytoplasm, only 8.0% of embryos developed to blastocysts. Similarly, when BDF1 MII was transferred into enucleated G56 oocyte cytoplasm, blastocyst formation was only 6.8%. These results suggest that G56 oocytes are damaged in both nucleus and cytoplasm.

Whole-genome sequencing (WGS) analysis of re-cloned mice

To investigate genomic mutations accumulated across generations, we conducted WGS on re-cloned mice from various generations. Short-read NGS-based WGS (read coverage: >30×) was performed on ten individuals from generations G6 to G57 (Fig. 5a, Supplementary Table 11). In addition, three of these individuals (G6-2, G50-1, and G57-2) underwent long-read WGS (read coverage: >20×) using Oxford Nanopore Technology long-read data.

**Fig. 5: Mutations identified in re-cloned mice by WGS.**

Based on single nucleotide variant (SNV) profiles, we reconstructed the genetic lineage of each individual. As shown in Fig. 5a, yellow boxes indicate individuals analyzed by WGS, while black boxes denote estimated common ancestors. This phylogeny matched the experimental lineage, confirming that no genetic contamination occurred during the 20-year serial cloning process. SNVs accumulated over generations (Fig. 5b), with ~3700 SNVs and 80 insertion/deletions (indels; excluding repeat elements) identified from G1 to G57 (Table 1). On average, each generation acquired 69.4 SNVs and 1.4 indels. The SNV mutation signatures from G1 to G57 resembled those of naturally occurring germline mutations³⁴ (Fig. 5c, Supplementary Fig. 3).

Table 1 Comparison of mutations accumulated in serial nuclear transfer and long-term inbred mouse strains derived from natural mating

Full size table

From G1 to G57, we identified 80–84 de novo structural variants (SVs) (Supplementary Data 1). Table 2 lists 16 large SVs (LSVs) of ≥40 kb, which are more likely to affect phenotypes. Many SVs appeared between G25 and G45. Among them, loss of one X chromosome, loss of heterozygosity (LOH) on chromosome 4 (147.5 Mb), and a translocation between chromosomes 7 and 9 are likely highly deleterious for fitness. LSVs also occurred after G45. Chromosomal abnormalities, including the X chromosome loss and two translocations, were likewise confirmed in the karyotype analysis of G54 re-cloned mice (Fig. 5e, Supplementary Fig. 4). We measured telomere length using short-read WGS and Telomerecat software 4.0.0. Although telomere length estimates varied widely and were method-dependent across samples, significant telomere shortening between G6 and G55 is unlikely (Supplementary Table 12). Nanopore long-read WGS data also revealed no evidence of extreme telomere shortening at individual chromosome ends (≥10 kb in G57-2).

Table 2 Large-scale Structural Variants found in re-cloned mice were increased with increased serial cloning

Full size table

To evaluate the effects of accumulated mutations, we analyzed their effects on genes. Of the 4404 SNVs and small indels identified, 0.95% were classified by SnpEff as “High” (loss-of-function) or “Moderate” (mainly missense). Dividing mutations into two groups—those from G1 to G23 during increased cloning success and those from G23 to G57 during decreased success (Fig. 1b)—revealed a significantly higher proportion in the latter group (G1–G23: 0.52%; G23–G57: 1.19%; P = 0.034) (Supplementary Table 13). Consistent with this, we assessed potentially deleterious variants using Genome Evolutionary Rate Profiling (GERP) scores, which evaluate functional constraint across coding and noncoding regions. The proportion with GERP scores ≥2 was significantly higher in the latter group (G1–G23: 2.4%; G23–G57: 4.0%; P = 0.005, Fisher’s exact test) (Fig. 5d, Supplementary Data 1, Supplementary Table 14). Mutation spectra also shifted, with transversions rising from 44.0% in G1–G23 to 47.9% in G23–G57 (P = 0.015, Fisher’s exact test) (Supplementary Fig. 3). In terms of SVs, G57 re-cloned mice showed at least 16 variants affecting exons (Supplementary Data 1). Across generations, two large deletions were identified in an adjacent region, resulting in homozygous loss of Csmd1 gene function (Supplementary Fig. 5). Considering the complete mutation spectrum, ~30 loss-of-function mutations and ~50 missense mutations had accumulated in G57 mice, even when excluding X chromosome loss (Table 1). In this context of an extensive background mutation load, the emergence of additional deleterious mutations near the final generation, including Csmd1 disruption, may have exceeded a tolerable threshold, leading to lethality or severe developmental impairment, ultimately contributing to the collapse of cloning success (Supplementary Fig. 6). Considered together, these findings indicate that the progressive decline of serial cloning success is strongly associated with the accumulation of de novo mutations—including large SVs and harmful SNVs/indels after G23–G25—rather than the effect of any single mutation alone.

To compare mutation patterns in asexual versus sexual reproduction, we analyzed WGS data from two mouse lines subjected to 60 generations of sibling mating over ~20 years. These lines accumulated 739 and 752 SNVs, 22 and 23 INDELs and 29 and 29 SVs, respectively, with ~80% of autosomal variants being homozygous (Fig. 5b, Table 1). This corresponds to 22.3 SNVs, 0.68 INDELs and 0.91 SVs per generation, representing one-third of the SNV mutation rate in serial cloning and a slightly lower INDELs and SV occurrence frequency.

Finally, we performed WGS again on offspring produced via natural mating between G55 re-cloned mice and wild-type males (C57BL/6), as well as on their grandchildren. As expected from Mendelian inheritance, the mutations present in G55 were passed to both generations. LSVs transmitted across generations are listed in Supplementary Table 15. Notably, the translocation between chromosomes 7 and 9 in G55 was also found in G55 descendants, likely because both affected chromosomes were inherited together, preserving balance.

Discussion

It has long been unclear whether mammals, unlike plants and some lower animals, could sustain their species through clonal reproduction alone. In this study, we performed continuous somatic cell nuclear transfer cloning in mice for nearly two decades, covering 57 generations. Our findings show that clone-specific epigenetic abnormalities did not accumulate, but genomic mutations did, increasing the burden of deleterious variants. Ultimately, this accumulation drove a progressive decline in birth rates, making continued clonal reproduction impossible.

On average, each generation acquired 70 single-nucleotide variants (SNVs) and 1.5 structural variants (SVs), including large structural variants (LSVs). These mutation rates are comparable to those in certain mouse somatic cell lineages^35,36 and in the human germline³⁷, suggesting that the rate itself is not unusually high. Nonetheless, two factors appear central to the collapse of long-term clonal propagation: (1) the emergence of LSVs, which are rarely seen in the germline^37,38, and (2) the absence of chromosomal recombination in clonal reproduction, which prevents effective purifying selection and allows all classes of mutations to accumulate progressively.

A critical turning point occurred around generation G25. Before this stage, chromosomal abnormalities and coding mutations were efficiently eliminated, and birth rates even improved. During this period, although we were not able to detect adaptive mutations by WGS analysis, some mutations may have transiently enhanced cloning performance, or somatic cells that were more easily reprogrammable and capable of generating viable re-clones may have been preferentially selected, resulting in a positive selection effect. However, these advantages were not sustained. After G25, large and deleterious SVs, including loss of heterozygosity (LOH) and X chromosome loss, became prominent. The build-up of harmful variants appears to have outweighed adaptive effects, such as the positive selection effect around G25, resulting an irreversible decline in birth rates. The frequency of deleterious mutations nearly doubled, from 0.86 to 1.78 per generation (Table 1), in parallel with the onset of this irreversible decline. This transition suggests that the burden of accumulating mutations began to affect not only nuclear genomic integrity but also oocyte-level functions essential for development. Consistent with this interpretation, our MII replacement experiments indicate that late-generation oocytes exhibited compromised cytoplasmic competence in addition to nuclear defects.

This study was conducted over a period of ~20 years. Therefore, considering variations to the quality of culture medium or equipment, it was impossible to maintain completely identical experimental conditions throughout the entire period. However, two researchers (TW and SW) consistently participated in the nuclear transfer experiments across all generations. Furthermore, nuclear transfer experiments conducted during this period using BDF1 cumulus cells demonstrated no change to the birth rate of cloned mice, including the duration of the COVID-19 pandemic (Supplementary Table 16). Therefore, it is unlikely that these factors contributed to the decreased cloning success rate. Conversely, in this study, TSA treatment was conducted during nuclear transfer for production of all generations of re-cloned mice. TSA is a standard class I/II histone deacetylase (HDAC) inhibitor known to not only activate genes involved in transcription and promote nuclear reprogramming^27,39, but also reactivate endogenous retroviruses^40,41. However, since a very low concentration (50 nM) was used during nuclear transfer and lasted only about 8 h, it is unlikely that TSA treatment caused the high mutation rate of the re-cloned mice.

Late-generation clones except last generation that survived to birth were notably healthy, with normal lifespans, despite carrying numerous deleterious mutations. Although placental size tended to increase with advancing generations, embryonic development itself was similar between early and late groups. Likely, re-cloned embryos that acquired lethal mutations were lost after implantation, and only those without such mutations developed to term. This stringent selection probably ensured that surviving re-cloned mice remained healthy. WGS analysis was performed using samples collected from live-born re-cloned mice, so it may not have identify critical mutations that cause lethality in embryos or fetuses.

Even more striking was the preservation of reproductive capacity. Despite over 20 years of repeated clonal propagation without engagement of the germline or reproductive organs, all examined re-cloned mice retained fertility. Since sexual system function is not required for the survival of individual clones, it is improbable that only cloned embryos with normal reproductive system were selected and born. Instead, these results imply that the sexual reproductive system is unusually robust against mutational damage, even under conditions that foster accumulation of deleterious variants. Throughout this study, we consistently utilized oocytes collected from young individuals, and it is known that the oocytes of young individuals can effectively suppress age-related DNA damage⁴², and this ability may have been particularly effective against mutations in the sexual reproductive system.

The necessity of sexual reproduction has traditionally been explained by the Red Queen hypothesis, which posits that sexual reproduction is required to maintain genetic diversity and thereby enable species to persist under changing environmental conditions^6,43. Our study, however, reveals limitations of this framework. Re-cloned mice displayed nearly threefold higher mutation rates than their sexually reproducing counterparts, thus generating diversity even without recombination. Moreover, because our experiments were performed under specific-pathogen-free conditions with minimal environmental pressure, the failure of clonal reproduction in this setting cannot be attributed to the Red Queen hypothesis.

In contrast, our results align closely with Muller’s ratchet theory⁴⁴. This model predicts that in asexual lineages, deleterious mutations inevitably accumulate, ultimately producing mutational meltdown and extinction⁴⁵. Across more than 30,000 nuclear transfer attempts spanning 58 generations, our study provides the first empirical demonstration that mutational meltdown, as predicted by Muller’s ratchet, can indeed occur in mammals. In comparison, wild-type mice maintained through natural breeding showed no fertility decline across more than 60 generations, emphasizing that the mutational load of clonal reproduction exceeds the threshold compatible with long-term mammalian survival. At the same time, the observation that most re-cloned mice could remain healthy despite harboring numerous genetic alterations offers insight into the extent of mutation tolerance at the level of individual organisms.

Since the birth of Dolly the sheep⁹, cloning technology has been envisioned for diverse applications, including agricultural improvement, regenerative medicine, and the conservation of Earth’s rare genetic resources^46,47,48. Yet, these findings demonstrate that practical application of cloning of mammals requires a deeper appreciation of these biological constraints, warranting further research, and reaffirm the evolutionary inevitability that sexual reproduction is indispensable for the long-term survival of mammalian species.

Methods

Animals

A three-month old BD129F1 (BDF1 × 129/Sv) mouse was used as the original donor animal, and cumulus cells were used as nuclear donors. This mouse was produced in our laboratory using three-way crosses between the C57BL/6, DBA/2, and 129/Sv strains. During experimentation, C57BL/6J mice were maintained at a constant temperature (25 °C ± 1 °C) under a 12-h light/dark cycle, with ad libitum access to food and water. The donor mouse that showed the highest success rate in producing first-generation cloned mice (hereafter, “G1”) was selected as the original donor and was used to initiate the serial mouse cloning experiment.

To generate subsequent generations, recipient oocytes were collected from adult BDF1 females. The surrogate pseudopregnant females used as embryo transfer recipients were ICR strain mice mated with vasectomized males of the same strain. BDF1, C57BL/6J, and ICR mice were purchased from Shizuoka Laboratory Animal Center (Hamamatsu, Japan).

For all sexual reproduction experiments, animal protocols were approved by the Animal Committee of Osaka University (Osaka, Japan; Approval No. FBS-17-006) and the Experimental Animal Care Committee of the Radiation Effects Research Foundation (RERF; Hiroshima, Japan; Approval No. 2018-02). All experiments were conducted in accordance with the 1996 National Institutes of Health guidelines and the Guidelines for Animal Experiments of the RERF. During experimentation, C57BL/6J mice were maintained at a constant temperature (22 °C ± 1 °C) under a 12-h light/dark cycle, with ad libitum access to food and water.

Housing conditions and diet were kept constant throughout the experiment. On the day of experimentation or after completion of all procedures, mice were euthanized by CO₂ inhalation or cervical dislocation. All animal experiments followed the ARRIVE guidelines and were performed in accordance with the Guidelines for the Care and Use of Laboratory Animals, with approval from the Institutional Committee of Laboratory Animal Experimentation of the University of Yamanashi (Reference No. A29-24).

Oocyte collection

Female mice were superovulated by injection of 5 IU of equine chorionic gonadotropin, followed 48 h later by 5 IU of human chorionic gonadotropin (hCG). Cumulus–oocyte complexes (COCs) were collected from the oviducts 14–16 h later and transferred to a Falcon dish containing HEPES–CZB medium⁴⁹. To disperse the cumulus cells, COCs were placed in a 50 μL droplet of HEPES–CZB medium containing 0.1% bovine testicular hyaluronidase for 3 min. Cumulus-free oocytes were washed twice and transferred to 20 μL droplets of CZB medium⁵⁰ for culturing.

Collection of cumulus cells and tail tip fibroblasts

Cumulus cells were collected from donor or re-cloned mice using the same method as for oocyte collection. Fibroblasts were collected from the tail tips of G54 re-cloned mice for karyotype analysis as described previously⁵¹. Briefly, tail tips were skinned, cut into small pieces, then incubated in 5 mL DMEM (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal calf serum (Sigma-Aldrich). After 10–14 days at 37.5 °C in 5% CO₂ in air, proliferating fibroblasts were dissociated using trypsin and replated into larger dishes to expand the cell population.

Nuclear transfer

For nuclear transfer, oocytes were first transferred to a droplet (~10 μL) of HEPES–CZB containing 5 μg/mL cytochalasin B, placed under oil on the microscope stage. Each oocyte was held with an oocyte-holding pipette, and its zona pellucida was perforated by several piezoelectric pulses applied to the tip of an enucleation pipette¹⁰. The metaphase II chromosome spindle complex, visible as a translucent spot in the ooplasm, was drawn into the pipette with a small amount of ooplasm and gently pulled away until the cytoplasmic bridge was pinched off. Enucleated oocytes were then injected with a cumulus cell nucleus recovered from a 3-month-old re-cloned mouse, as described previously¹⁰. Briefly, the zona pellucida and the oocyte membrane were perforated by piezoelectric pulses, and the donor nucleus was inserted into the ooplasm. Reconstructed oocytes were transferred into CZB medium and incubated at 37 °C in 5% CO₂.

Effects of nuclear transfer techniques on birth rate

During this long-term study, we improved the birth rate of cloned mice by optimizing nuclear transfer methods, such as using different histone deacetylase inhibitor (HDACi) treatments instead of TSA^52,53. However, to isolate the effect of re-cloning, the standard method described above was used whenever possible. Some variations, such as in media quality or experimenter skill, were unavoidable over the study period and may have affected success rates or phenotypes. To control for these factors, we generated cloned mice from other BD129F1 donors as technical or time-matched comparisons (designated NG1). To assess the pure effect of serial nuclear transfer, cumulus cells from G24 re-cloned mice were frozen in June 2013 and thawed 8 years later to generate additional G25 mice (designated CryoG25), concurrently with G50 generation. Fertilized control (FC) mice with the same BD129F1 genetic background were also produced via intracytoplasmic sperm injection (ICSI), simulating equivalent in vitro manipulation and culture stress.

Activation of cloned embryos

Reconstructed oocytes were activated using 5 mM SrCl₂ in Ca²⁺-free CZB medium containing 50 nM trichostatin A (TSA) and 5 μM latrunculin A for 9 h^27,33,54. Pseudo-pronuclear formation was verified, and embryos were cultured for either 1 day before transfer or 96 h to evaluate blastocyst development.

Embryo transfer

On the day following mating with vasectomized males, 2-cell stage cloned embryos were transferred into the oviducts of pseudo-pregnant ICR females at 0.5 days post coitum (dpc)⁵⁵. On the transfer day, recipients were anesthetized by intraperitoneal injection of medetomidine, midazolam, and butorphanol. Five to eight embryos were transferred per uterine horn, and atipamezole was administered. At 19.5dpc, offspring were delivered by cesarean section and transferred to foster mothers that had delivered naturally. Fertility was assessed by mating offspring at three weeks of age.

Intracytoplasmic sperm injection (ICSI)

To generate genetically identical but fertilized control (FC) mice, oocytes were collected from BDF1 females and sperm from 129/Sv males, producing BD129F1 offspring as with the original donor. To match cloned embryo conditions, FC mice were generated via ICSI, with equivalent in vitro manipulation and 1 day of CO₂ incubation⁴⁹.

Histological analysis of placentae

Placentae from FC, NG1 (newly generated clone), G25, G50 re-cloned mice, and grandchild mice were embedded in OCT compound (Sakura Finetek, Japan). Blocks were cryosectioned at 7 μm (Thermo Fisher Scientific, USA) and mounted on Superfrost Micro Slides (Matsunami Glass, Japan). Sections were fixed in 4% paraformaldehyde for 10 min at room temperature, washed three times in PBS, incubated in 100 mM Tris-HCl (pH 9.5) for 10 min, and stained using NBT-BCIP (Vector Laboratories) for endogenous alkaline phosphatase (AP) activity. Sections were rinsed in Tris-HCl and counterstained with Nuclear Fast Red (Vector Laboratories). Trophoblasts in the labyrinth layers were detected by AP staining, and the area of AP staining-positive labyrinth layers were compared between natural, re-cloned mice and grandchild mice.

Immunostaining of re-cloned embryos at the one-cell stage

Ten hours post-activation, oocytes were fixed for 20 min at 25 °C in 4% (w/v) paraformaldehyde with 0.2% (v/v) Triton X-100 (Nacalai Tesque, Kyoto, Japan). Fixed oocytes were washed thrice in PBS with 0.25% BSA (Millipore, Darmstadt, Germany) and stored at 4 °C. Embryos were incubated with primary antibodies: rabbit anti-H3K4me3 (1:1000, Millipore, Darmstadt, Germany), rabbit anti-H3K9me3 (1:1000,Millipore, Darmstadt, Germany), or rabbit anti-H3K27me3 (1:1000, Millipore, Darmstadt, Germany) in PBS-BSA overnight at 4 °C. Antibodies were supplemented with 0.2% (v/v) Tween (Millipore, Darmstadt, Germany). After two PBS-BSA washes (15 min each), embryos were incubated for 45 min with Alexa Fluor Cy5-labeled goat anti-rabbit IgG (1:500, Abcam, Cambridge, UK). DNA was stained with 2 μg/mL DAPI (Molecular Probes Inc., OR, USA). The largest pseudo-pronucleus per embryo was imaged using confocal microscopy (FV-1000, Olympus Corp., Tokyo, Japan). Fluorescence intensity was quantified using Olympus Fluor View and NIH ImageJ.

Blastocyst immunostaining

To assess cloned blastocyst quality, cell counts were performed via immunofluorescence staining as described previously⁵⁶. Briefly, blastocysts were fixed for 20 min at 25 °C in 4% paraformaldehyde with 0.2% Triton X-100, washed thrice in PBS with 0.25% BSA, and stored at 4 °C. Primary antibodies: anti-CDX2 mouse monoclonal (1:500, BioGenex, CA, USA) for TE cells and anti-Nanog rabbit polyclonal (1:500, Abcam, Cambridge, UK) for ICM cells. Secondary antibodies: Alexa Fluor 488-labeled goat anti-mouse IgG and Alexa Fluor 568-labeled goat anti-rabbit IgG (1:500, Abcam, Cambridge, UK). DNA was stained with 2 μg/mL DAPI (Molecular Probes, OR, USA). Blastocysts were mounted on slides, and serial z-stacks were captured by confocal microscopy (FV-1200). Cell numbers were quantified using Olympus Fluor View.

Re-cloned mice oocyte quality examination

Oocytes were collected from BDF1, G25, and G53 re-cloned mice, and diploid parthenogenetic or fertilized embryos were generated to assess blastocyst development. Diploid parthenogenetic embryos were produced using the same method as that used to activate reconstructed oocytes after nuclear transfer. ICSI was performed using the same procedure as that for generating FC mice. The resulting parthenogenetic and fertilized embryos were cultured for five days, and their development to expanded blastocysts was evaluated. Next, to investigate the nucleus (metaphase II: MII) and cytoplasm of cloned mice oocytes separately, we performed MII replacement between oocytes from BDF1 and G56 re-cloned mice. This was followed by diploid parthenogenesis treatment to determine the rate of development to the blastocyst stage. MII extraction was performed using the same method as described for oocyte enucleation. However, when transferring MII to enucleated oocytes, the enucleation pipette was used to inject the MII into enucleated oocytes in cytochalasin B-containing medium, after which the hole was immediately sucked out to seal the cell membrane⁵⁷. After replacing the MII, all oocytes were allowed to rest in an incubator for at least two hours before activation.

RNA-seq of re-cloned blastocysts

Five blastocysts were sampled from each of the following groups: FC mice, NG1 cloned mice, and G50 re-cloned mice. Samples were frozen immediately in liquid nitrogen and stored at –80 °C until further analysis. RNA-seq libraries were prepared using the SMART-seq Stranded kit (Takara Bio, Shiga, Japan) and sequenced as 150-base single-end reads on the DNBSEQ-G50RS system (MGI Tech., Shenzhen, China). Sequence data were then mapped against the mouse reference genome sequence (GRCm39/mm39) using STAR⁵⁸. We then calculated the “transcripts per million” value for each gene using RSEM⁵⁹. To extract transcripts that often exhibited upregulation in blastocysts greater than twofold, we then calculated the fold change of each signal in all individuals within each group (i.e., NG1 and G50). These were then compared with the mean signal in the FC; the mean difference was then assessed using the DESeq2 R package⁶⁰.

WGS and mapping

Genomic DNA was extracted from spleen samples using smart DNA prep (m) (Analytik Jena) and Qiagen Genomic tip DNA extraction kits (QIAGEN). For short-read WGS, paired-end libraries were prepared with DNA samples without PCR amplification using a TruSeq DNA PCR-Free Library Prep Kit (Illumina). All libraries were then sequenced on NovaSeq 6000 or X plus platforms (Illumina) to obtain paired-end 151-bp reads. Short-read sequence data were then mapped to the mouse reference genome (UCSC mm10) using BWA-MEM version 0.7.17⁶¹ with the “-M” option. This mapping was also processed with Picard version 2.18.26 (broadinstitute.github.io/picard) to remove PCR duplicates. In the analysis of sib-mating mouse lines, we utilized the ancestral genome data (F0 mice) from our previous paper (Ancestor ConA2 male-female mice). For long-read sequencing, libraries were prepared from G6-2, G50-1, and G57-2 using the SQK-LSK114 kit (Oxford Nanopore) and sequenced on a PromethION with FLO-PRO114M flow cells in high-accuracy mode. Base calling was performed with Guppy version 6.4.2 and Dorado version 7.2 (https://community.nanoporetech.com/). Base-called reads were mapped to the mouse reference genome (UCSC mm10, excluding chrY) using minimap2 version 2.21 and version 2.28⁶² with the parameters “-ax map-ont --MD.”

Identification of de novo mutations

Genomic variants were called from short-read bam files using GATK HaplotypeCaller version 4.1.0.0⁶³ with a minimum base quality of 20. For accurate detection of somatic mutations, effective whole-genome coverage (EWC) regions were defined per sample based on the following criteria: (i) MQ60 read depths were between 50% and 300% of the chromosome-specific peak coverage; (ii) MQ60 reads comprised at least 80% of mapped reads at each site; and (iii) a minimum base quality of 20 was confirmed using SAMtools version 1.9⁶⁴. Only shared EWC regions were used for mutation analyses (excluding SVs), covering 76.4% of the diploid genome for serial nuclear transfer lines, and 78.6% for sib-mating mouse lines (Supplementary Table 11). All de novo mutations were identified from filtered variants using the following criteria: (1) located within EWC regions; (2) multiallelic variants excluded; (3) variants with 0.35 > AF ≥ 0.03 in G45, G57-1, and G57-2 but <0.03 in others were excluded to remove NovaSeq X Plus–specific noise; (4) minimum VAF = 0 in any clone; (5) Fisher strand bias (FS) < 23; (6) variants with AF ≥ 0.06 and <0.2 detected in ≥3 samples excluded; (7) max VAF ≥ 0.3 for SNVs and ≥ 0.25 for INDELs in G6, G25, G45, G50, G55 and G57. We also used the following INDEL-specific filters: (8) exclude variants in homopolymers (>3 bp) or short repeats (e.g., di-, tri-, tetra-, penta-, or hexa-). (9) Only manually confirmed variants in IGV were retained. (10) Finally, we retained variants that met either of the following: (i) VAF < 0.10 in G6-1; or (ii) VAF < 0.10 in G6-2 and, in G6-1, VAF ≥ 0.30 for SNVs or VAF ≥ 0.25 for INDELs. In the analysis of sib-mating mouse lines, we retained variants that satisfied the following: VAF < 0.06 in both the F0 male and F0 female, and—on autosomes—VAF ≥ 0.30 for SNVs or ≥0.25 for INDELs in F60 or F62; on sex chromosomes, VAF ≥ 0.80 in F60 or F62.

SV calling using short- and long-read data

SV calling of short-read data was conducted using Manta version 1.6.0⁶⁵, RUFUS version 1.0⁶⁶, and Delly version 0.8.6⁶⁷. SV calling by Manta was conducted using the following criteria: 1) The number of split reads for the “ALT” allele (SRA) was >3 in a mouse. 2) The minimum SRA number of other mice was zero. SV calling by RUFUS was conducted using the following criteria: 1) “SVTYPE” was described in the INFO column of the VCF file, or 2) the length of the “REF” allele or the “ALT” allele was >40 bp. SV calling by DELLY was conducted using the following criteria: 1) In all cases the number of variant reads (i.e., DV or RV) was >3 for a mouse. 2) The minimum number of variant reads in other mice was 0.

Next, we performed SV detection using long-read data to recover mutations missed in short read analyses. Mutation candidates were then extracted from G6-2 and G50-1 using sniffles version 1.0.12⁶⁸ with the parameter: “-s 3.” VCF files of SV calls from each sample were merged using SURVIVOR version 1.0.7⁶⁹ with the following parameters: max distance between breakpoints was 1000 bp and “types must match.” The merged VCF file was then used to perform a “force call” of SVs using sniffles. Force-called VCF files of each sample were again merged using SURVIVOR. Structural variants were called using the following inclusion criteria: (1) SVTYPE is TRA or SVLEN is ≤−40 or ≥40, (2) The number of reads supporting alt is 0 for G6-2, >5 for G50-1, and/or G57-2, and (3) depth is >10 and <100 for both G6-2 and G50-1. All candidates were then manually inspected by IGV, including short read results. Using this procedure five new de novo genes were detected. Three of these could be linked to the generation in which they arose using short read sample information. Finally, we note that this analysis was not limited to EWC regions.

Mutation analysis

The genetic lineage of each re-cloned mouse was inferred from its SNV profile. Assuming a constant number of SNVs per generation, lineage divergence points were identified. By excluding regions affected by LOH, large deletions, or X chromosome loss, we constructed a consistent phylogenetic relationship.

Mutational signature analysis and visualization of single-base substitutions were carried out using SigProfilerAssignment⁷⁰ (https://cancer.sanger.ac.uk/signatures/assignment/app/), with the parameters set as follows: Reference genome = “mm10/GRCm38,” Variant class = “SBS.” Reference mutational signatures = “COSMIC v3.4.” Type of study = “Whole-Genome Sequencing”⁷⁰. Functional annotations were performed with SnpEff version 4.3t⁷¹. De novo mutations were further annotated using GERP scores⁷². GERP score data derived from the mouse mm9 reference was obtained from the UCSC website (http://hgdownload.soe.ucsc.edu/gbdb/mm9/bbi/All_mm9_RS.bw).

Telomere length evaluation

Telomere length was assessed using WGS data from two G6 individuals, two CryoG25 individuals, and two G50 individuals by Telomerecat (version 4.0.0)⁷³ using the default parameters. Although we performed whole-genome sequencing on either the Illumina NovaSeq 6000 or NovaSeq X Plus platforms, only datasets produced on the NovaSeq 6000 were used for telomere length analysis in order to ensure methodological consistency and avoid potential platform-specific biases.

Karyotyping of G54 re-cloned mice

Chromosomes from tail-tip cells of two G54 re-cloned mice (G54-1 and G54-2) were analyzed by spectral karyotyping using fluorescence in situ hybridization (SKY–FISH) chromosome painting techniques (SCAS Ltd., Tokyo, Japan).

Statistical analyses

Most serial cloning experiments for each generation were repeated at least twice. Other experiments were independently performed at least three times by two to three experimentalists, and similar results were obtained regardless of the experimenter. Rates of embryo development, offspring birth, and body and placental weights were evaluated using chi-squared tests. Fluorescence intensity was analyzed by one-way ANOVA followed by Tukey’s multiple comparison tests as implemented in GraphPad Prism (GraphPad Software, USA). Principal component analysis (PCA) was performed using the DESeq2 R package. For all statistical tests, a value of P < 0.05 was considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All data are available in the main text, Supplementary Figs./tables, or source data file. NGS and WGS data have been deposited in DNA Data Bank of Japan Sequence Read Archive under accession numbers PRJDB37585 and PRJDB35828. Additional data related to this paper may be requested from the authors. Source data are provided with this paper.

References

Gerritsen, J. Sex and parthenogenesis in sparse populations. Am. Naturalist 114, 718–742 (1980).
Article Google Scholar
Lively, C. M. & Morran, L. T. The ecology of sexual reproduction. J. Evol. Biol. 27, 1292–1303 (2014).
Article CAS PubMed PubMed Central Google Scholar
Yoshida-Noro, C. & Tochinai, S. Stem cell system in asexual and sexual reproduction of Enchytraeus japonensis (Oligochaeta, Annelida). Dev. Growth Differ. 52, 43–55 (2010).
Article CAS PubMed Google Scholar
Nishimura, O. et al. Unusually large number of mutations in asexually reproducing clonal planarian Dugesia japonica. PLoS One 10, e0143525 (2015).
Article PubMed PubMed Central Google Scholar
Avise, J. C. Evolutionary perspectives on clonal reproduction in vertebrate animals. Proc. Natl. Acad. Sci. USA 112, 8867–8873 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
West, L. R. A pluralist approach to sex and recombination. J. Evolut. Biol. 12, 1003–1012 (1999).
Article Google Scholar
Blickstein, I. & Keith, L. G. On the possible cause of monozygotic twinning: lessons from the 9-banded armadillo and from assisted reproduction. Twin Res. Hum. Genet. 10, 394–399 (2007).
Article PubMed Google Scholar
Gurdon, J. B. Adult frogs derived from the nuclei of single somatic cells. Dev. Biol. 4, 256–273 (1962).
Article CAS PubMed Google Scholar
Wilmut, I., Schnieke, A. E., McWhir, J., Kind, A. J. & Campbell, K. H. Viable offspring derived from fetal and adult mammalian cells. Nature 385, 810–813 (1997).
Article ADS CAS PubMed Google Scholar
Wakayama, T., Perry, A. C., Zuccotti, M., Johnson, K. R. & Yanagimachi, R. Full-term development of mice from enucleated oocytes injected with cumulus cell nuclei. Nature 394, 369–374 (1998).
Article ADS CAS PubMed Google Scholar
Hochedlinger, K. & Jaenisch, R. Monoclonal mice generated by nuclear transfer from mature B and T donor cells. Nature 415, 1035–1038 (2002).
Article ADS CAS PubMed Google Scholar
Lanza, R. P. et al. Cloning of an endangered species (Bos gaurus) using interspecies nuclear transfer. Cloning 2, 79–90 (2000).
Article CAS PubMed Google Scholar
Loi, P. et al. Genetic rescue of an endangered mammal by cross-species nuclear transfer using post-mortem somatic cells. Nat. Biotechnol. 19, 962–964 (2001).
Article ADS CAS PubMed Google Scholar
Wakayama, S. et al. Propagation of an infertile hermaphrodite mouse lacking germ cells by using nuclear transfer and embryonic stem cell technology. Proc. Natl. Acad. Sci. USA 102, 29–33 (2005).
Article ADS CAS PubMed Google Scholar
Vajta, G. & Gjerris, M. Science and technology of farm animal cloning: state of the art. Anim. Reprod. Sci. 92, 211–230 (2006).
Article PubMed Google Scholar
Yang, X. et al. Risk assessment of meat and milk from cloned animals. Nat. Biotechnol. 25, 77–83 (2007).
Article CAS PubMed Google Scholar
Kato, Y. et al. Eight calves cloned from somatic cells of a single adult. Science 282, 2095–2098 (1998).
Article ADS CAS PubMed Google Scholar
Hoshino, Y. et al. Resurrection of a bull by cloning from organs frozen without cryoprotectant in a −80 degrees c freezer for a decade. PLoS One 4, e4142 (2009).
Article ADS PubMed PubMed Central Google Scholar
Malin, K., Witkowska-Pilaszewicz, O. & Papis, K. The many problems of somatic cell nuclear transfer in reproductive cloning of mammals. Theriogenology 189, 246–254 (2022).
Article CAS PubMed Google Scholar
Matoba, S. & Zhang, Y. Somatic cell nuclear transfer reprogramming: mechanisms and applications. Cell Stem Cell 23, 471–485 (2018).
Article CAS PubMed PubMed Central Google Scholar
Yin, X. J. et al. Production of second-generation cloned cats by somatic cell nuclear transfer. Theriogenology 69, 1001–1006 (2008).
Article CAS PubMed PubMed Central Google Scholar
Wakayama, T. et al. Cloning of mice to six generations. Nature 407, 318–319 (2000).
Article ADS CAS PubMed Google Scholar
Kubota, C., Tian, X. C. & Yang, X. Serial bull cloning by somatic cell nuclear transfer. Nat. Biotechnol. 22, 693–694 (2004).
Article CAS PubMed Google Scholar
Cho, S. K. et al. Serial cloning of pigs by somatic cell nuclear transfer: restoration of phenotypic normality during serial cloning. Dev. Dyn. 236, 3369–3382 (2007).
Article CAS PubMed Google Scholar
Kurome, M. et al. Production efficiency and telomere length of the cloned pigs following serial somatic cell nuclear transfer. J. Reprod. Dev. 54, 254–258 (2008).
Article PubMed Google Scholar
Yang, M. et al. Increased pregnancy losses following serial somatic cell nuclear transfer in goats. Reprod. Fertil. Dev. 30, 1443–1453 (2018).
Article PubMed Google Scholar
Kishigami, S. et al. Significant improvement of mouse cloning technique by treatment with trichostatin A after somatic nuclear transfer. Biochem. Biophys. Res Commun. 340, 183–189 (2006).
Article ADS CAS PubMed Google Scholar
Wakayama, S. et al. Successful serial recloning in the mouse over multiple generations. Cell Stem Cell 12, 293–297 (2013).
Article CAS PubMed Google Scholar
Wakayama, T. & Yanagimachi, R. Cloning of male mice from adult tail-tip cells. Nat. Genet 22, 127–128 (1999).
Article CAS PubMed Google Scholar
Inoue, K. et al. Loss of H3K27me3 imprinting in the Sfmbt2 miRNA cluster causes enlargement of cloned mouse placentas. Nat. Commun. 11, 2150 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Tanaka, S. et al. Placentomegaly in cloned mouse concepti caused by expansion of the spongiotrophoblast layer. Biol. Reprod. 65, 1813–1821 (2001).
Article CAS PubMed Google Scholar
Wang, F., Kou, Z., Zhang, Y. & Gao, S. Dynamic reprogramming of histone acetylation and methylation in the first cell cycle of cloned mouse embryos. Biol. Reprod. 77, 1007–1016 (2007).
Article CAS PubMed Google Scholar
Terashita, Y. et al. Latrunculin a treatment prevents abnormal chromosome segregation for successful development of cloned embryos. PloS One 8, e78380 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Uchimura, A. et al. Early embryonic mutations reveal dynamics of somatic and germ cell lineages in mice. Genome Res. 32, 945–955 (2022).
CAS PubMed PubMed Central Google Scholar
Matsuda, Y. et al. Spectra and characteristics of somatic mutations induced by ionizing radiation in hematopoietic stem cells. Proc. Natl. Acad. Sci. USA 120, e2216550120 (2023).
Article CAS PubMed PubMed Central Google Scholar
Sugo, N. et al. DNA polymerase beta suppresses somatic indels at CpG dinucleotides in developing cortical neurons. Proc. Natl. Acad. Sci. USA 122, e2506846122 (2025).
Article CAS PubMed PubMed Central Google Scholar
Porubsky, D. et al. Human de novo mutation rates from a four-generation pedigree reference. Nature 643, 427–436 (2025).
Article ADS CAS PubMed PubMed Central Google Scholar
Belyeu, J. R. et al. De novo structural mutation rates and gamete-of-origin biases revealed through genome sequencing of 2,396 families. Am. J. Hum. Genet. 108, 597–607 (2021).
Article CAS PubMed PubMed Central Google Scholar
Inoue, K. et al. Trichostatin A specifically improves the aberrant expression of transcription factor genes in embryos produced by somatic cell nuclear transfer. Sci. Rep. 5, 10127 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Bouyahya, A. et al. Pharmacological properties of trichostatin a, focusing on the anticancer potential: a comprehensive review. Pharmaceuticals 15, 1235 (2022).
Mueller, T., Hantsch, C., Volkmer, I. & Staege, M. S. Differentiation-dependent regulation of human endogenous retrovirus K sequences and neighboring genes in germ cell tumor cells. Front. Microbiol. 9, 1253 (2018).
Article PubMed PubMed Central Google Scholar
Dudko, N. et al. Oocyte age-dependent DNA damage can be reverted by the DNA repair competent karyoplasm of young oocytes. Aging Cell 25, e70300 (2025).
Valen, L. V. A new evolutionary law. Evol. Theory 1, 1–30 (1973).
Muller, H. J. The relation of recombination to mutational advance. Mutat. Res. 106, 2–9 (1964).
Article CAS PubMed Google Scholar
Michael, L. & Wilfried, G. Mutation load and the survival of small populations. Evolution 44, 1725–1737 (1990).
Article Google Scholar
Wakayama, S. et al. Production of healthy cloned mice from bodies frozen at −20 degrees C for 16 years. Proc. Natl. Acad. Sci. USA 105, 17318–17322 (2008).
Article ADS CAS PubMed PubMed Central Google Scholar
Mizutani, E. et al. Generation of cloned mice and nuclear transfer embryonic stem cell lines from urine-derived cells. Sci. Rep. 6, 23808 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Wakayama, S. & Wakayama, T. Can humanity thrive beyond the galaxy? J. Reprod. Dev. 71, 10–16 (2025).
Article CAS PubMed Google Scholar
Kimura, Y. & Yanagimachi, R. Intracytoplasmic sperm injection in the mouse. Biol. Reprod. 52, 709–720 (1995).
Article CAS PubMed Google Scholar
Chatot, C. L., Ziomek, C. A., Bavister, B. D., Lewis, J. L. & Torres, I. An improved culture medium supports development of random-bred 1-cell mouse embryos in vitro. J. Reprod. Fertil. 86, 679–688 (1989).
Article CAS PubMed Google Scholar
Wakayama, S., Ito, D., Hayashi, E., Ishiuchi, T. & Wakayama, T. Healthy cloned offspring derived from freeze-dried somatic cells. Nat. Commun. 13, 3666 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Ono, T. et al. Inhibition of class IIb histone deacetylase significantly improves cloning efficiency in mice. Biol. Reprod. 83, 929–937 (2010).
Article CAS PubMed Google Scholar
Van Thuan, N. et al. The histone deacetylase inhibitor scriptaid enhances nascent mRNA production and rescues full-term development in cloned inbred mice. Reprod. Camb. Engl. 138, 309–317 (2009).
Article Google Scholar
Terashita, Y. et al. Latrunculin a can improve the birth rate of cloned mice and simplify the nuclear transfer protocol by gently inhibiting actin polymerization. Biol. Reprod. 86, 180 (2012).
Article PubMed Google Scholar
Yang, L. L. et al. A novel, simplified method to prepare and preserve freeze-dried mouse sperm in plastic microtubes. J. Reprod. Dev. 69, 198–205 (2023).
Article CAS PubMed PubMed Central Google Scholar
Wakayama, S. et al. Healthy offspring from freeze-dried mouse spermatozoa held on the International Space Station for 9 months. Proc. Natl. Acad. Sci. USA 114, 5988–5993 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Hikichi, T. et al. Differentiation potential of parthenogenetic embryonic stem cells is improved by nuclear transfer. Stem Cells 25, 46–53 (2007).
Article CAS PubMed Google Scholar
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Article CAS PubMed Google Scholar
Li, B. & Dewey, C. N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinforma. 12, 323 (2011).
Article CAS Google Scholar
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Article PubMed PubMed Central Google Scholar
Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760 (2009).
Article CAS PubMed PubMed Central Google Scholar
Li, H. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics 34, 3094–3100 (2018).
Article CAS PubMed PubMed Central Google Scholar
McKenna, A. et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 20, 1297–1303 (2010).
Article CAS PubMed PubMed Central Google Scholar
Li, H. et al. The sequence alignment/map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
Article PubMed PubMed Central Google Scholar
Chen, X. et al. Manta: rapid detection of structural variants and indels for germline and cancer sequencing applications. Bioinformatics 32, 1220–1222 (2016).
Article CAS PubMed Google Scholar
Ostrander, B. E. P. et al. Whole-genome analysis for effective clinical diagnosis and gene discovery in early infantile epileptic encephalopathy. NPJ Genom. Med. 3, 22 (2018).
Article PubMed PubMed Central Google Scholar
Rausch, T. et al. DELLY: structural variant discovery by integrated paired-end and split-read analysis. Bioinformatics 28, i333–i339 (2012).
Article CAS PubMed PubMed Central Google Scholar
Sedlazeck, F. J. et al. Accurate detection of complex structural variations using single-molecule sequencing. Nat. Methods 15, 461–468 (2018).
Article CAS PubMed PubMed Central Google Scholar
Jeffares, D. C. et al. Transient structural variations have strong effects on quantitative traits and reproductive isolation in fission yeast. Nat. Commun. 8, 14061 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Diaz-Gay, M. et al. Assigning mutational signatures to individual samples and individual somatic mutations with SigProfilerAssignment. Bioinformatics 39, btad756 (2023).
Cingolani, P. et al. A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly. Austin 6, 80–92 (2012).
CAS PubMed PubMed Central Google Scholar
Davydov, E. V. et al. Identifying a high fraction of the human genome to be under selective constraint using GERP++. PLoS Comput. Biol. 6, e1001025 (2010).
Article PubMed PubMed Central Google Scholar
Farmery, J. H. R., Smith, M. L., Diseases, N. B.-R. & Lynch, A. G. Telomerecat: a ploidy-agnostic method for estimating telomere length from whole genome sequencing data. Sci. Rep. 8, 1300 (2018).
Article ADS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank Dr. S. Kishigami, Mrs. Y. Kanda for assistance in preparing this manuscript. This work was partially funded by the Research Fellowships of Japan Society for the Promotion of Science to D.I. (JP20J23364), A.U. (24K03096, 24H02061, 24H00752), S. W. (23K08843) to T.W. (23K18124 and 24K01779); the Naito Foundation and Takahashi Industrial and Economic Research Foundation (189) to S.W.; Asada Science Foundation and the Canon Foundation (M20-0008) to T.W.

Author information

Authors and Affiliations

Advanced Biotechnology Center, University of Yamanashi, Yamanashi, Japan
Sayaka Wakayama & Teruhiko Wakayama
Faculty of Life and Environmental Sciences, University of Yamanashi, Yamanashi, Japan
Daiyu Ito, Rei Inoue, Hirosuke Shiura, Takashi Kohda & Teruhiko Wakayama
Department of Animal Science and Biotechnology, School of Veterinary Medicine, Azabu University, Kanagawa, Japan
Masatoshi Ooga
Department of Molecular Biosciences, Radiation Effects Research Foundation, Hiroshima, Japan
Masaaki Toshishige, Yasunari Satoh & Arikuni Uchimura
KOKORO-Biology Group, Laboratories for Integrated Biology, Graduate School of Frontier Biosciences, Osaka University, Suita, Osaka, Japan
Arikuni Uchimura

Authors

Sayaka Wakayama
Daiyu Ito
Rei Inoue
Masatoshi Ooga
Masaaki Toshishige
Yasunari Satoh
Hirosuke Shiura
Takashi Kohda
Arikuni Uchimura
Teruhiko Wakayama

Contributions

S.W., A.U. and T.W. conceived and designed the study. S.W., D.I., R.I., M.O., M.T., Y.S., H.S., T.K., A.U. and T.W. performed experiments, analyzed the data, and interpreted the results. A.U. and T.W. wrote the manuscript. All authors read and edited the manuscript.

Corresponding authors

Correspondence to Sayaka Wakayama, Arikuni Uchimura or Teruhiko Wakayama.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Atsuo Ogura, Jie Xu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Wakayama, S., Ito, D., Inoue, R. et al. Limitations of serial cloning in mammals. Nat Commun 17, 2495 (2026). https://doi.org/10.1038/s41467-026-69765-7

Download citation

Received: 05 November 2025
Accepted: 10 February 2026
Published: 24 March 2026
Version of record: 24 March 2026
DOI: https://doi.org/10.1038/s41467-026-69765-7