A
Bridging the gap in embryogenesis research: North Bornean orangutan-derived blastoids as models for human developmentNicoleLilingTAY1EmailNicole.tay@mandai.org.sg
QiuyeBAO1EmailQiuye.bao@mandai.org.sg
UmaSangumathiKAMARAJ1,3EmailUma_kamaraj@bii.a-star.edu.sg
YingyingZENG1EmailZeng0114@e.ntu.edu.sg
KaWaiWONG1EmailGary.wong@ucl.ac.uk
ChristinaYing1EmailChristina.lim@mandai.org.sg
YanLIM1EmailObgmac@nus.edu.sg
JonathanYuin12EmailYhloh@imcb.a-star.edu.sg
HanLOH12EmailOz.pomp@mandai.org.sg
1
A
A
A
A
Delia Hwee Hoon CHUA28,9 Soon Chye NG* Soonchyeng@gmail.com 1,2,8,10 Oz POMP
3Mandai Nature80 Mandai Lake Road729826Singapore
4Institute of Molecular and Cell Biology61 Biopolis Drive138673ProteosSingapore
5Institute of OphthalmologyUniversity College LondonEC1V 9ELLondonUK
6STAR Office of Sustainability4 Fusionopolis Way, Level 11138635KinesisSingapore
7Mandai Wildlife Group80 Mandai Lake Road729826Singapore
8Nanyang Technological University50 Nanyang Ave639798SingaporeSingapore
9University of Adelaide5005AdelaideSAAustralia
10National University of Singapore, Yong Loo Lin School of Medicine119074Singapore
11Department of Biological SciencesNational University of Singapore117543 10Singapore
12Sincere Healthcare Group8 Sinaran Drive307470Singapore
Nicole Liling TAY Nicole.tay@mandai.org.sg 1,2
Qiuye BAO Qiuye.bao@mandai.org.sg 1,2
Uma Sangumathi KAMARAJ Uma_kamaraj@bii.a-star.edu.sg 2
Yingying ZENG Zeng0114@e.ntu.edu.sg 2
Ka Wai WONG Gary.wong@ucl.ac.uk 2,3
Christina Ying Yan LIM Christina.lim@mandai.org.sg 1,2
Winnie Koon Lay TEO Winnie_teo@a-star.edu.sg 2,4
Huili YEO Huili.yeo@mandai.org.sg 1,2
Kanchana PUNYAWAI Kanchana.punyawai@mandai.org.sg 1,2
Delia Hwee Hoon CHUA Delia.chua@mandai.com 5
Chia-da HSU Chiada.hsu@mandai.com 5
Shangzhe XIE Shangzhe.xie@mandai.com 1,5,6,7
Mahesh CHOOLANI Obgmac@nus.edu.sg 8
Jonathan Yuin-Han LOH Yhloh@imcb.a-star.edu.sg 2,8,9
Soon Chye NG* Soonchyeng@gmail.com 1,2,8,10
Oz POMP* Oz.pomp@mandai.org.sg 1
1. Mandai Nature, 80 Mandai Lake Road, Singapore 729826
2. Institute of Molecular and Cell Biology, 61 Biopolis Drive, Proteos, Singapore 138673
3. University College London, Institute of Ophthalmology, London EC1V 9EL, UK
4. A*STAR Office of Sustainability, 4 Fusionopolis Way, Level 11, Kinesis, Singapore 138635
5. Mandai Wildlife Group, 80 Mandai Lake Road, Singapore 729826
6. Nanyang Technological University, Singapore, 50 Nanyang Ave, Singapore 639798
7. University of Adelaide, Adelaide SA 5005, Australia
8. National University of Singapore, Yong Loo Lin School of Medicine, Singapore 119074
9. National University of Singapore, Department of Biological Sciences, Singapore 117543
10. Sincere Healthcare Group, 8 Sinaran Drive, Singapore 307470
*Corresponding author
ABSTRACT
A
A
Understanding early primate embryogenesis remains limited by ethical and technical constraints. To address this, we established a stem cell-based embryo model from the North Bornean orangutan (Pongo pygmaeus pygmaeus). Using integration-free induced pluripotent stem cells (iPSCs), we generated blastoids that self-organized into structures resembling human blastocysts in morphology, lineage composition, and transcriptomic identity. Initial aggregation of primed iPSCs produced blastoids with organized lineage segregation. We enhanced this approach by combining isogenic iPSCs maintained under different culture conditions at optimized ratios, generating "bi-blastoids" with high formation efficiency, balanced lineage specification, and dimensions matching human blastocysts. Single-cell RNA sequencing revealed epiblast-, trophectoderm-, and hypoblast-like populations with transcriptomic profiles closely resembling human E6-7 blastocysts. Comparative analysis demonstrated that orangutan blastoids exhibited stronger transcriptional fidelity to human embryos than existing macaque models, establishing their unique value for developmental studies. Functional validation confirmed that orangutan blastoids could generate embryonic stem cell-like and trophectoderm stem cell-like lines, recapitulate trophoblast differentiation and secrete chorionic gonadotropin. However, primitive endoderm/hypoblast-like cells failed to form a distinct compartment, mirroring limitations observed across other primate models. This orangutan blastoid system provides a less ethically restrictive platform for studying human-relevant embryogenesis while maintaining high developmental fidelity. Beyond developmental insights into primate development, this model may inform future reproductive interventions for critically endangered great apes as conservation technologies advance.A
A
INTRODUCTION
Human embryo development is a complex and dynamic process, and understanding how cells assemble into functional embryos is fundamental for uncovering the origins of developmental disorders 1. Yet, despite advances in imaging and molecular profiling, a comprehensive view of the mechanisms underlying peri-implantation human embryogenesis remains elusive due to limited access to samples and stringent ethical constraints 2.
Embryological studies have relied heavily on model organisms such as the mouse, which are inexpensive and reproduce efficiently. These studies have revealed essential principles of early development 3–5. Yet, their applicability is limited by key interspecies differences 6–8. Although human and mouse blastocysts appear morphologically similar at pre-implantation, they differ profoundly at the cellular and molecular levels, with distinct gene expression and signalling pathways 5,9–11. These differences become even more pronounced during post-implantation development and placentation 12. These differences restrict murine models from fully capturing human-specific biology.
Progress in human in vitro culture systems has yielded valuable insights into blastocyst and peri-implantation stages 13–18. However, embryo availability remains limited, manipulation is technically challenging, and ethical considerations limit experimental scope 19–23. The peri-implantation stage, which is largely inaccessible to study, is particularly critical, as abnormalities during this stage underlie infertility, pregnancy loss, and developmental condition 24–29.
To overcome these barriers, stem cell-based embryo models have emerged. Initial efforts used embryonal carcinoma-derived embryoid bodies 30,31, and embryonic stem cells (ESCs) which could complement natural embryos through blastocyst chimerism 32 or tetraploid complementation 33. However, these approaches still relied on natural embryos.
More recently, the ability of pluripotent stem cells (PSCs) to self-organize in response to mechanochemical cues has enabled the creation of “blastoids” – 3D structures that mimic key features of blastocyst development. These models mimic many aspects of blastocyst development and can be observed and manipulated in high throughput, offering complex spatiotemporal insights into lineage segregation, cellular mechanics, implantation, and epigenetic regulation 34–36. While containing both embryonic and extra-embryonic cell types, these structures can provide complex spatiotemporal molecular information on the development of the blastocyst and peri-implantation embryo, including lineage segregation, cellular mechanics, implantation, the effect of epigenetic abnormalities, and many other processes 37.
In humans, blastoids have been generated from partially reprogrammed fibroblasts 38, naïve induced pluripotent stem cells (iPSCs) 39–41 and expanded potential / extended pluripotent stem cells (EPSCs) 42,43. Despite variations between protocols, these structures resemble human blastocysts in morphology, size, cell number, lineage allocation, markers, transcriptomic profiles, and functional potential, including stem cell derivation and implantation modelling. Importantly, they bypass some limitations of using human embryos for research, while enabling applications such as drug and therapeutic screening 41,44.
Despite their potential, human blastoid studies remain ethically and legally constrained. Although the International Society for Stem Cell Research (ISSCR) recently lifted the blanket “14-day rule” restricting embryo research 45, in vitro culture beyond 14 days is still evaluated on a case-by-case basis, and in vivo transplantation into a uterus remains forbidden. Thus, both human and murine models remain limited in their ability to capture peri-implantation development in full biological context.
NHPs offer a promising alternative. Compared to human models, they have less stringent ethical concerns and provide greater sample availability, while allowing studies to extend into gastrulation and later stages 46. Progress in NHP embryology 21,47 and in vitro techniques 48 support their value for modelling human embryonic development, given the conservation of key developmental mechanisms amongst primates 11,49. The recent creation of cynomolgus macaque blastoids 50 further demonstrates the utility of such systems.
Among > 300 primate species, Great Apes (family Hominidae) are the closest relatives of humans, diverging only 12–16 million years ago compared to ~ 25–33 million years for macaques (Macaca sp.) 51–53. Developmental and placental differences between the macaques and humans 11,12 underscore the need for a Great Ape model 54. Yet, invasive studies in living Great Apes are prohibited, leaving a critical gap for cellular approaches.
Research on early embryonic development in orangutans (Pongo sp.) is limited by their critically threatened status, the scarcity of embryos, and ethical restrictions. All three orangutan species are listed as Critically Endangered (CR) under the International Union for Conservation of Nature (IUCN) Red List 55 and share a close phylogenetic relationship with humans 56,57. Their embryogenesis and placentation are comparable to humans and other apes 58–61, yet little is known about their reproductive physiology, and assisted/advanced reproductive technologies (ART) approaches have seen limited success 62, although new information and techniques will emerge as zoological institutions progress in this field.
Here, we report the first successful generation of blastoids from the North Bornean orangutan (Pongo pygmaeus pygmaeus). These blastoids closely recapitulated human counterparts in morphology, organization, and transcriptomic identity. Functional validation through stem cell derivation and in vitro culture confirmed their developmental potential. The creation of orangutan blastoids establishes a new comparative system for studying primate embryogenesis while offering a less ethically restrictive alternative to human models. Beyond advancing developmental biology, this system may hold future potential for conservation, should supporting technologies mature. Orangutan iPSC and blastoid platforms could one day inform reproductive interventions or assist in developing improved ART for critically endangered great apes, and in the long term, might provide last-resort options for species preservation.
RESULTS
Formation of orangutan blastoids
To model preimplantation development, we employed the AggreWell platform to aggregate feeder-dependent footprint-free induced pluripotent stem cells (F+iPSCs) under limiting dilution (Suppl. Figure 1). Optimal blastoid formation was achieved at a seeding density of 25–30 cells per microwell (Suppl. Figure 2a). By one day post-aggregation (DPA), uniform spheroids had formed, which expanded progressively until 5 DPA (Fig. 1b; Suppl. Figure 2b). At 6 DPA, aggregates acquired a cystic architecture resembling human blastocysts, including a fluid-filled cavity, an outer epithelial layer, and an inner cell mass-like cluster (Fig. 1a, c).
Immunostaining confirmed lineage segregation, with expression of epiblast (EPI; OCT4, SOX2), trophectoderm (TE; GATA2, CDX2), and primitive endoderm / hypoblast (PE/HYPO; SOX17, GATA4) markers in distinct regions (Fig. 1d; Suppl. Figure 2c). The spatial organization mirrored that of human blastocysts 49,63–66. By contrast, aggregates derived from feeder-free orangutan iPSCs (F–iPSC) formed irregular cysts with poorly defined patterning (Suppl. Figure 2b, c) and were excluded from downstream analyses. Consistent with observations in human and NHPs 67, aggregation of F+iPSCs led to upregulation of pluripotency and early lineage-associated transcripts (Suppl. Figure 2d), supporting the notion that orangutan blastoids undergo self-organization through conserved developmental programs. Collectively, these findings establish orangutan blastoids as a tractable in vitro system to investigate early embryogenesis in this endangered primate.
Transcriptome analysis
To resolve cellular identities at single-cell resolution, we performed scRNA-seq on 210 orangutan blastoids, generating transcriptomes from 7,028 cells. Dimensionality reduction and clustering identified seven transcriptionally distinct populations (Fig. 1e; Suppl. Figure 3a, b). Marker analysis revealed populations corresponding to EPI (POU5F1, SOX2, KLF4, DPPA2/3/5, DNMT3L, PRDM14, KLF17, FGF4, GDF3), TE (CDX2, GATA2/3, PLAC8, KRT7/8/18/19, CD24), and PE/HYPO (SOX17, GATA4/6, PDGFRA, COL4A1) lineages (Suppl. Figure 3c).
Correlation analyses indicated that orangutan blastoids aligned more closely with human blastoids and E6 human blastocysts (Fig. 1g). Comparison with cynomolgus macaque (Macaca fascicularis) blastocyst datasets 68 further showed strong transcriptional similarity between orangutan and human blastoids and blastocysts (Fig. 1h), reinforcing the orangutan as a valuable model for human embryonic development.
Notably, POU5F1 and SOX2 displayed heterogeneous expression across clusters, consistent with their roles in multiple lineages during preimplantation development 65. Cluster identities were assigned by cross-referencing with transcriptional profiles from human blastocysts and blastoids 5,17,38,65,69 (Fig. 1f). These results confirm that F+iPSC-derived blastoids encompass the three principal lineages of the preimplantation embryo and recapitulate early cell fate decisions observed in human and other primates.
Formation of bi-line orangutan blastoids
Given the divergent characteristics of F+blastoids and F–blastoids, where F+iPSCs supported organized lineage segregation, while F–iPSCs produced poorly structured cysts with a bias toward the TE-like identity, we hypothesized that combining both cell types could improve structural integrity and lineage balance.
Serial dilutions of F+iPSCs and F–iPSCs revealed that aggregation at a 1:3 ratio markedly improved blastoid formation efficiency compared with either population alone (Suppl. Figure 4a). The resulting chimeric “bi-line” blastoids (bi-blastoids) were composed of fewer cells than F+blastoids and were intermediate in size between F+ and F–blastoids, closely resembling the dimensions of human blastoids (Suppl. Figure 4b-d). Bi-blastoids expressed core pluripotency markers (NANOG, POU5F1, SOX2, KLF4) as well as naïve pluripotency markers (DPPA3, DPPA5, DNMT3L, KLF5, KLF17, PRDM14, SUSD2) at levels comparable to mono-blastoids (F+ or F–blastoids) (Fig. 2b). Importantly, TE (CDX2, GATA2/3, KRT8/18, PLAC8, TFAP2C, TF63, HAVCR1), PE/HYPO (SOX17, GATA6, PDGFRA), and post-implantation lineage markers including amnion (GABRP, BMP4) and primordial germ cells (PRDM1) were all upregulated relative to mono-blastoids (Fig. 2b), indicating enhanced lineage diversification.
As with mono-blastoids, bi-blastoids displayed spatial organization resembling human blastocysts, with peripheral GATA2 + and CDX2 + TE-like cells, centrally located SOX2 + EPI-like cells, as well as SOX17 PE/HYPO-like cells and a blastocoel-like cavity (Fig. 2c; Suppl. Figure 4e). To track lineage contribution, F-iPSCs stably expressing GFP were aggregated with F+iPSCs at the 1:3 ratio to form bi-blastoidsGFP(Fig. 2a). GFP + cells localized predominantly to the outer TE layer, co-expressing CDX2 and the tight-junction marker ZO-1, and were rarely observed within NANOG + or SOX2 + EPI-like cells or GATA6+/SOX17 + PE/HYPO-like cells. This pattern indicates a bias of the GFP + F-iPSC fraction toward the TE lineage (Fig. 2d; Suppl. Figure 4e).
Single-cell transcriptomics of orangutan bi-blastoids
To resolve lineage specification within bi-blastoids, we performed scRNA-seq on approximately 200 aggregates, generating > 8,000 single-cell transcriptomes. Dimensionality reduction and clustering revealed nine transcriptionally distinct clusters (Fig. 3a, b). Marker analysis revealed populations corresponding to EPI (POU5F1, SOX2, KLF4, DPPA2/3/5, DNMT3L, PRDM14, KLF17, FGF4, GDF3), TE (CDX2, GATA2/3, PLAC8, KRT7/8/18/19, CD24), and PE/HYPO (SOX17, GATA6/4, PDGFRA, COL4A1) lineages (Suppl. Figure 5a). Benchmarking against transcriptional signatures of human blastocysts and blastoids 5,17,38,65,69 confirmed lineage identities for each cluster (Fig. 3b, c). Differentially expressed genes across each cluster were examined (Fig. 3e).
Comparative analysis with orangutan blastoids, human blastoids, and human blastocysts 38,65, revealed broad transcriptional overlap (Suppl. Figure 5b). Similar to blastocysts, and in contrast to mono-blastocysts, bi-blastoids lacked intermediate populations (Fig. 3d; Suppl. Figure 5b). Furthermore, they exhibited lineage distributions and gene expression pattern closely resembling human blastocysts at E6-7 (Fig. 3d, f).
Derivation of ESC-like and trophectoderm stem cell-like cells
Natural blastocysts provide pluripotent and trophectoderm stem cells (TSCs). To evaluate whether orangutan blastoids harbour expandable stem cell populations, we plated bi-blastoidsGFP under lineage-specific culture conditions. In iPSC medium, alkaline phosphatase-positive ESC-like cells (ESCLCs) emerged, displaying the characteristic morphology of primed pluripotent colonies – compact and flat with sharp borders and a high nuclear-to-cytoplasmic ratio (Fig. 4a). Transcriptomic analysis confirmed robust expression of primed pluripotency markers at levels comparable to parental iPSCs (Fig. 4b; Suppl. Figure 6a). These ESCLC lines exhibited minimal expression of TE-associated transcripts (Suppl. Figure 6b). Immunostaining further validated expression of pluripotency markers (NANOG, OCT4, SOX2, SSEA-4) in the absence of GFP signal, consistent with their derivation from the F+iPSC fraction (Fig. 4d).
When cultured under TSC-supportive conditions, bi-blastoidsGFP gave rise to trophectoderm stem cell–like cells (TSCLCs) exhibiting the cobblestone morphology typical of human TSCs (Fig. 4a). Transcriptomic profiling revealed upregulation of trophoblast-associated markers relative to iPSCs and ESCLCs (Fig. 4c; Suppl. Figure 6b), while immunostaining confirmed expression of TE markers (CDX2, GATA2, GATA3) (Fig. 4d). Under these conditions, pluripotency markers (NANOG, OCT4, SOX2, SSEA-4) were not detected. Most TSCLC clones were GFP⁺, although a minority of GFP⁻ clones were also obtained, suggesting that F⁺iPSCs, when aggregated, can give rise to both TE- and EPI-like lineages, whereas F⁻iPSCs show limited potential to generate EPI-like cells.
Collectively, these results confirms that orangutan blastoids can generate both ESCLC and TSCLCs, further validating their quality.
In vitro implantation assay
During the implantation process, trophoblasts of a blastocyst attach to the uterine wall, and after adhesion, the inner cell mass forms the pro-amniotic cavity before initiating gastrulation 5. To evaluate the peri-implantation potential of the orangutan blastoids, we adapted a human blastoid attachment assay 13,14,38. Bi-blastoidsGFP adhered within 24–48 h, flattened, and produced outgrowths (Suppl. Figure 7a). Immunostaining identified OCT4 + and SOX2 + EPI-like cells at the centre and CDX2 + and GATA3 + TE-like cells at the periphery, mirroring spatial organisation in the 3D blastoids (Suppl. Figure 7b). A minor population co-expressed SOX2 and CDX2, suggestive of an amnion-like identity 41.
By day 5 post-attachment, subsets of TE-like cells showed positive staining for a HLA-G analogue and CGβ, markers associated with maternal-foetal tolerance and early pregnancy maintenance (Fig. 5a-c). ELISA and commercial pregnancy tests confirmed CGβ secretion into conditioned media, indicating functional syncytiotrophoblast-like differentiation (Fig. 5d). These findings were reproducible in mono-blastoids (Fig. 5e). Together, these data demonstrate that orangutan blastoids can recapitulate key cellular and molecular events of blastocyst implantation. The identification of post-implantation markers indicates that the orangutan blastoids are capable of progressing beyond pre-implantation stages, exhibiting hallmarks of peri-implantation development in vitro.
DISCUSSION
This study represents the first successful generation of integration-free orangutan iPSCs and their subsequent use to generate blastoids. Orangutan blastoids consisted of lineage-representative populations with transcriptomic profiles resembling those of human blastocysts and blastoids, capturing several functional hallmarks of peri-implantation development, with transcriptomic profiles resembling those of human blastocysts and blastoids. Notably, orangutan blastoids exhibited greater transcriptional similarity to human blastocysts than those of macaque models, underscoring their distinctive value as a comparative system for studying human blastocyst development and implantation.
A
Protocols were adapted from the pioneering human blastoid study 38. The bi-blastoid protocol achieved a maximum efficiency of 13.1%, close to the reported median efficiency of published human blastoids (median = 14.4%, range 1.9–89%) 38–43,67. While most prior studies predominantly used naïve PSCs or EPSCs to initiate blastoid formation, this study demonstrates that primed iPSCs can also support blastoid generation. This suggests that the blastoid formation conditions may have driven partial conversion of primed iPSCs toward a naïve-like state. Such primed-to-naïve transitions may represent a rate-limiting step, accounting for reduced efficiency, but also highlight a tradeoff: primed iPSCs are easier to maintain than naïve cultures, potentially broadening accessibility of blastoid approaches.The bi-blastoid protocol markedly outperformed mono-blastoid conditions. Interestingly, F+iPSCs predominantly contributed to EPI-like cells, while F−iPSCs favoured TE-like differentiation, despite being derived isogenically from the same parental clone. This suggests that culture conditions, rather than clonal genetics, shaped lineage bias. We propose that F−iPSCs transiently adopted a naïve-like state, expanding developmental potential but subsequently “downgraded” to a primed state predisposed toward TE. Rather than enhancing blastoid formation, this shift skewed lineage orientation, preventing proper blastoid architecture. These findings highlight plasticity of primate pluripotency states and suggest that lineage orientation can be deliberately programmed through culture conditions.
Although average cell numbers were comparable across orangutan blastoids, human blastocysts, and human blastoids 38, their physical dimensions differed. Orangutan mono-blastoids displayed significantly larger diameters than mature human blastocysts (p = 0.0084), while orangutan bi-blastoids closely matched human blastocyst dimensions 70. Together with their higher formation efficiency, these findings indicate that the bi-blastoid represents a more accurate and reliable orangutan model of blastocyst development.
A persistent limitation of blastoid models is incomplete lineage specification. Across species, blastoids frequently lack defined PE/HYPO compartments 38–40,42,43,71 and often include non-relevant cell types, including uncommitted PSCs 39,40, lineage intermediates expressing multiple lineages markers 38,40,42), or cell types associated with later developmental stages 38,72. In orangutan blastoids, PE/HYPO-like cells were detected by transcriptomics and immunostaining but failed to organize into a distinct compartment, and diverged from human profiles. This limitation mirrors other primate models and underscores the challenge of fully recapitulating extraembryonic lineages in vitro.
Cell-type annotation in embryo models is strongly influenced by available reference datasets 73,74. While extensive single-cell datasets exist for mouse and primate development, equivalent data for wildlife species remain scarce. Defining lineages in stem cell-based embryo models without species-specific reference remains contentious 15. Interspecies comparisons, applied here, provide partial solutions 64,69,75, and human embryo reference maps remain fragmented, limiting lineage resolution. Nevertheless, recent NHP blastoid models 50,76 and the orangutan model described here showed strong transcriptional resemblance to human blastoids and blastocysts, particularly at E7, underscoring their relevance for comparative studies and the urgent need to expand reference datasets across primates.
Functionality evaluation of blastoid-derived cells extends beyond morphological comparisons. Orangutan blastoids generated ESCLCs and TSCLCs, paralleling ESCs and TSCs from blastocysts. Unlike mouse PSCs, which are largely incapable of TE differentiation, human and other great ape PSCs possess this ability 77–80, facilitating the formation of TE-containing blastoids using only PSCs to start. However, establishing robust, long-term TSCLC cultures remains a barrier, limiting the model's broader applicability.
Functional validation was further explored using an implantation assay. In vitro implantation assays revealed that orangutan blastoids adhered, differentiated, and expressed markers of post-implantation lineages, recapitulating key features of uterine attachment. Yet, the absence of maternal-foetal crosstalk limits the modelling of later events such as mural TE development, parietal endoderm formation, and ecoplacental cone establishment. Incorporating hormonally stimulated endometrial models, as has been applied to human blastoids, could improve fidelity 41.
This orangutan blastoid model bridges a crucial gap between human and non-primate systems. It offers an ethically viable and scientifically robust platform for studying early embryogenesis without embryo destruction. Although current human blastoid models cannot progress beyond early stages, further advances may eventually extend developmental capacity and raise ethical considerations. In this context, great ape blastoids represent a valuable compromise – scientifically informative while ethically viable.
Perspective – Expanding the Conservation Toolkit
The blastoid system – also referred to as integrated stem cell-based embryo models, synthetic embryos, or blastocyst-like structures – offers unique opportunities for conservation. For critically endangered species, where natural embryos are scarce or inaccessible, blastoids may provide an unprecedented window into developmental stages otherwise unobtainable.
Here, we established the first integration-free iPSC lines from P. p. pygmaeus, creating a renewable biobanking resource. Importantly, our findings demonstrate that lineage bias can be programmed by culture conditions, raising the possibility of engineering interspecies blastoids in which ICM-like cells from endangered species are combined with TE-biased cells from surrogate species. This would form analogues of embryos derived from blastocyst-based ICM transplantation or tetraploid complementation, and such strategies may help overcome barriers to interspecies implantation.
A
Hence, integrated stem cell-based embryo models may expand the conservation toolkit beyond artificial insemination, IVF, and somatic cell nuclear transfer, which are constrained by gamete and embryo availability. Stem cell-derived blastoids may therefore offer a scalable alternative, particularly when generated from cryopreserved somatic cells, enabling recovery of genetic diversity from deceased individuals.Beyond their potential to generate embryos, blastoids allow systematic exploration of interspecies compatibility, trophoblast biology, and host-surrogate interactions, which are crucial for advancing cross-species ART. Coupled with biobanking initiatives, iPSC-based embryo modelscan safeguard threatened taxa today while providing tools for future reproductive technologies.
Amid the accelerating biodiversity crisis, stem cell-based embryo models should be viewed not only as experimental platforms for developmental biology but also as integral tools for conservation. When integrated with reproductive technologies and species recovery strategies, these models provide a forward-looking approach with tangible potential to mitigate biodiversity loss.
MATERIALS AND METHODS
Ethics statement
A
This project was approved by the Mandai Wildlife Group (MWG) Research Panel under the project code MWG230104. No animals were harmed in the preparation of this manuscript.Access to skin samples
The Institute of Molecular and Cell Biology – Endangered Species Conservation via Assisted Reproduction (IMCB-ESCAR) laboratory is part of the Mandai Wildlife Group, which is the steward of Mandai Wildlife Reserve, home to Singapore Zoo, Night Safari, River Wonders, Bird Paradise, and Rainforest Wild Asia. Skin samples were obtained from only one female North Bornean Orangutan (Pongo pygmaeus pygmaeus), who died of natural causes at 6 years of age.
Derivation and culture of North Bornean orangutan primary fibroblasts
A
At post-mortem, the animals’ skin from the ear was cleaned with 70% ethanol and shaved before aseptic surgical preparation of the area of interest. A sterile scalpel blade was used to harvest a 3cm x 3cm sample of full-thickness skin. In sterile conditions, any remaining fur, fat, and epidermis were removed from the skin sample, and the sample was cut into smaller pieces. After washing in PBS, the pieces were incubated in DMEM with 1X Antibiotic–Antimycotic (Gibco, 15240062), at room temperature for 30 min. After further washing in PBS, the pieces were placed in a 0.1% gelatin-coated sterile tissue culture dishes and cultured in P0 media; P0 media contained Dulbecco’s Modified Eagle Medium (DMEM) (Gibco, 11966025), 20% Fetal Bovine Serum (FBS) (Gibco, 26140079), 1X Minimum Essential Medium (MEM) non-essential amino acids (NEAA) (Gibco, 11140076), 1X Penicillin–Streptomycin (Gibco, 15140122) (equivalent to 100U/ml Penicillin and 100µg/ml Streptomycin), and 2mM L-glutamine (Gibco, A2916801). All cell cultures were incubated at 37° Celsius, 5% CO2, and 20% O2, unless otherwise described. When confluent, fibroblasts were passaged with 0.25% trypsin and maintained on 0.1% gelatin-coated sterile tissue culture dishes with fibroblast media (FM); FM contained DMEM, 10% FBS, 1X MEM NEAA, 100µM β-mercaptoethanol (Gibco, 21985022), 1X Penicillin–Streptomycin, and 2mM L-glutamine. For efficient reprogramming, fibroblasts were transduced or transfected at passage 8 or earlier.Episomal reprogramming for iPSCs
Cell counting was performed with the Countess™3 Automated Cell Counter (Invitrogen, AMQAX2000) and verified manually. 600,000 fibroblasts were transfected with 2ug of each plasmid pCXLE-OCT3/4-p53shRNA, pCXLE-hSK, pCXLE-hUL, and pCXWB-EBNA-1 (Addgene #27077, #27078, #27080, and #37624 respectively 81). Electroporation was performed via the Neon™ Transfection System (Invitrogen, MPK5000, MPK1025) at 1650V, 10ms, and 3 pulses. Transfected cells were seeded onto 0.1% gelatin-coated 6cm dishes and cultured in FM without Penicillin–Streptomycin (Day 0). 24h later (Day 1), the media was changed to FM containing 10ng/ml bFGF. On Day 2, the media was changed to FM containing 10ng/ml bFGF and 0.5mM sodium butyrate, and was refreshed daily. On Day 7, cells were passaged and split 1:4 into 6-well plates coated with irradiated CF1 mouse embryonic fibroblast (iMEF) (Gibco, A34180) feeder layers, and cultured in a 1:1 ratio of FM and PluriSTEM™ Human ES/iPS Cell Medium (MERCK, SCM130). On Day 9, the media was changed to PluriSTEM™ only, and was refreshed every other day. iPSC colonies started forming by Day 12. iPSC colonies were picked from Day 21, and expanded and subsequently maintained in 6-well plates coated with feeders in PluriSTEM™. iPSCs were passaged with ReLeSR™ (STEMCELL Technologies, #100–0484) every 4–6 days and split 1:10–20 into PluriSTEM™, with 10µM Y-27632 for the 1st 24h post-passaging; PluriSTEM™ was changed daily or every other day. iPSCs were stable and could be passaged for more than 20 passages. Six iPSC lines were generated and one was used for further analysis.
Sendai viral reprogramming for iPSCs
250,000 fibroblasts were seeded onto 0.1% gelatin-coated 3 cm dishes and cultured in FM. 24h later (Day 0), fibroblasts were transduced with Sendai virus from the CytoTune™-iPS 2.0 kit (Invitrogen, A16517) containing expression cassettes for human KLF4, OCT3/4, SOX2, and c-MYC, in 1 ml of FM. On Day 1, the media was changed to FM containing 10ng/ml bFGF and 0.5 mM sodium butyrate, and was refreshed daily. On Day 7, cells were passaged and split 1:4 into 6-well plates coated with iMEF feeder layers or Matrigel®, and cultured in FM with bFGF. On Day 8, the media was changed to homemade pluripotent stem cell media (bFGF-media) and refreshed daily. bFGF-media contained DMEM, 15% ESC FBS (Gibco, 16141002), 1X MEM NEAA, 100µM β-mercaptoethanol, 1X Penicillin–Streptomycin, 2mM L-glutamine, 10ng/ml bFGF (Gibco, PHG0264), and 10ng/ml hLIF (PeproTech, 3000525UG). iPSC colonies were picked from Day 14, and expanded and subsequently maintained in 6-well plates seeded with feeders in bFGF-media (F+iPSCs), or on Matrigel®-coated plates in StemFlex™ Medium (Gibco, A3349401) (F−iPSCs). iPSCs were passaged with ReLeSR™ (STEMCELL Technologies, #100–0484) every 4–6 days and split 1:10–20 into their respective media, with 10µM Y-27632 for the 1st 24h post-passaging; media was changed daily or every other day. iPSCs were stable and could be passaged for more than 20 passages. Ten iPSC lines were generated and one was used for further analysis. Detection of Sendai viral genome transgenes was performed following manufacturer’s instructions, with transgene-specific primers for the specific detection of transgenes carried by the Sendai reprogramming vectors.
Constructing GFP+ F−iPSCs
F−iPSCs were passaged at 60–80% confluence. 24h (Day 0) later, F−iPSCs were transduced with 6ul Lentivirus-packed pSin-GFP-Fg (Addgene #174307). On Day 2, 0.8µg/ml puromycin was applied for selection. GFP+ F−iPSC conlonies were picked from Day 14, and expanded and subsequently maintained in 6-well plates coated with Matrigel® in StemFlex™ Medium with 0.6ug/ml puromycin.
Alkaline phosphatase staining
Alkaline phosphatase staining was performed using Vector® Blue Substrate Kit (Alkaline Phosphatase) according to the manufacturer’s instructions (Vector Laboratories, SK-5300).
Karyotyping
Cells were treated overnight with 2.5mg/ml 5-bromo-2-deoxyuridine and 10ug/ml colcemid. Cells were then dissociated with 0.5% trypsin–EDTA for 10 min, collected into a centrifuge tube, and treated with 5 ml hypotonic solution containing 75mM KCl and 0.8% sodium citrate for 20min. Cells were then fixed with a series of 3:1 methanol and acetic acid fixative.
A
G-banding and chromosome analyses were performed in accordance with standard procedures: cell suspensions were dropped onto clean wet slides and allowed to air dry, before being placed in a 90 degree Celsius oven prior to staining using GTG-banding techniques; 20 cells were analyzed and 5 full karyotypes were performed per sample. Reference karyograms were taken from the Atlas of Mammalian Chromosomes 82. Chromosomes were counted and checked for numerical and/or structural abnormalities.Embryoid body formation and differentiation into three germ layers
iPSCs were passaged and split 1:3 into ultralow attachment 24-well plates. Cells were cultured in bFGF-media without bFGF nor hLIF for 7 days. EBs formed were either maintained in suspension for another 7 days before being harvested for RNA extraction, or transferred to a 0.1% gelatin-coated plate for another 7 days before being fixed for immunostaining.
Blastoid formation
Blastoid formation using F+iPSCs, F−iPSCs, or a combination of both was performed in a similar manner to published iBlastoid protocol 38. AggreWell™400 24-well or 6-well culture plates (STEMCELL Technologies, #34421) were prepared according to the manufacturer’s instructions as above. ReLeSR™ was added to wells containing iPSC colonies for 1 min before aspiration. The wells were then incubated for 2 mins at 37° Celsius. TrypLE™ was then used to flush out iPSCs sans iMEF feeder cells when present, and cells were incubated in TrypLE™ for 2 mins at 37° Celsius to promote singularization of iPSCs. iPSCs were resuspended in 1ml of iBlastoid media (iBM) with Y-27632; iBM contained 36.6% DMEM/F12 GlutaMAX (Gibco, 10565018), 38.8% Advanced DMEM/F12 (Gibco, 12634010), 12.1% Neurobasal medium, 10.1% conditioned FBS, 0.075% BSA, 1.25mM L-glutamine, 0.625X Penicillin-Streptomycin, 50µM β-mercaptoethanol, 0.75X Insulin-Transferrin-Selenium-Ethanolamine (ITS-X) (Gibco, 51500056), 0.125X N-2 Supplement, 0.125X B-27™ Supplement, 12.5µM N-acetyl-L-cysteine (Sigma-Aldrich, A7250), 4nM β-oestradiol (Sigma-Aldrich, E4389100MG), 100ng/ml Progesterone (Sigma-Aldrich, P7556100MG), 375.2ng/ml L-ascorbic Acid, 2µM CHIR99021, 0.5µM A83-01 (Sigma-Aldrich, SML0788), 1µM SB431542 (STEMCELL Technologies, 72234), 0.8mM Valproic Acid (VPA) (Sigma-Aldrich, P4543), 50ng/ml EGF (Sigma-Aldrich, E5036500UG), and 10ng/ml BMP4 (Miltenyi Biotec, 130111168). Cell numbers were quantified, and a minimum cell viability of 80% was ensured before proceeding for blastoid culture. iPSCs at defined ratios for 25 cells per microwell (1200 microwells in an AggreWell™400 24-well plate, 5900 microwells in an AggreWell™400 6-well plate), were collected and transferred into each AggreWell™ well. The plate was centrifuged at x300g for 3 mins and transferred back into the incubator 37° Celsius, 5% CO2, and 5% O2. 24h later (Day 0), iBM without Y-27632 was carefully replaced. 156h after initial cell seeding (Day 6.5), aggregates were collected from each AggreWell™ well using wide-bore P1000 tips. Visual analysis and picking of cavitated blastoids were performed under a stereomicroscope. Efficiency of blastoid formation was calculated by dividing the number of cavitated structures collected from each AggreWell™ well by the number of microwells in each AggreWell™ well.
Morphometric analyses of blastoids
Diameters of blastoids were measured using ImageJ (NIH, version 1.54f) 83 with the x-axis determined as a line drawn between the widest section of the blastoid parallel to the edge of the inner cell mass-like structure adjacent to the blastocoel, and the y-axis determined as a line drawn between the widest section of the blastoid perpendicular to the x-axis. Number of cells per blastoid were calculated by first washing a number of collected blastoids in 0.04% BSA in PBS, followed by digestion in a 1:1 mix of Accumax™ (STEMCELL Technologies, #07921) and Versene Solution (Gibco, 15040066) at 37° Celsius in a thermoshaker for 30 mins or until blastoids have been dissociated into single cells. The cell suspension was then subjected to centrifugation and resuspension to achieve an appropriate dilution for cell counting. The total number of cells calculated to be collected was then divided by the number of blastoids dissociated to determine the number of cells per blastoid.
Derivation of stem cells from blastoids
To derive ESCLCs and TSCLCs, blastoids were collected and deposited onto iMEF feeders in bFGF-medium supplemented with 10µM Y-27632, or in trophectoderm stem cell (TSC) medium, and refreshed daily without Y-27632. TSC medium contained DMEM/F12 GlutaMAX, 0.2% characterized FBS (HyClone, SH30071.03), 0.3% BSA, 0.1mM β-mercaptoethanol, 0.5% Penicillin-Streptomycin, 1% ITS-X, 0.5mM VPA, 50ng/ml EGF, 2µM CHIR99021, 1.5µg/ml L-ascorbic Acid, 0.5µM A83-01, 1µM SB431542, 5µM Y-27632. ESCLCs and TSCLCs were passaged with ReLeSR™ or TrypLE™ respectively every 4–6 days and split 1:10–20 into their respective media, which was refreshed daily or every other day. 4–8 ESCLC and TSCLC lines were generated.
In vitro attachment assay
The extended in vitro attachment assay was adapted from previously published protocols 13,14,38. Blastoids were collected and deposited onto 0.1% gelatin- or Matrigel®-coated optical-grade tissue culture plates, and cultured in in vitro culture medium 1 (IVC1); IVC1 contained Advanced DMEM/F12, 1% ITS-X, 2mM L-glutamine, 0.5% Penicillin-Streptomycin, 20% characterized FBS, 25µM N-acetyl-L-cysteine, 8nM β-oestradiol, and 200ng/ml Progesterone. 48h later, media was refreshed to in vitro culture medium 2 (IVC2); IVC2 was identical to IVC1 with the exception of FBS, which was replaced by 30% knockout serum replacement (Gibco, 10828028). Blastoids were cultured in this manner for 5 days.
hCG-analogue assay
To assess the presence and concentration of chorionic gonadotrophin (CG) produced by syncytiotrophoblast-like cells, IVC1 and IVC2 were collected from the in vitro attachment assay at Days 2 and 5 respectively, and subjected to commercially available human pregnancy test trips (Guardian, 622125) or a human CG (hCG) beta DuoSet solid phase sandwich enzyme-linked immunosorbent assay (ELISA) (R&D Systems, DY9034-05), following manufacturer’s instructions.
A
Quantitative RT-PCRTotal RNA was extracted using Monarch® Total RNA Miniprep Kit (New England BioLabs Inc., T2010) and reverse transcribed into cDNA with iScript™ Reverse Transcription Supermix (Bio-Rad, 1,708,840). qPCR was performed using KAPA SYBR® FAST qPCR Master Mix (2X) Universal (Kapa Biosystems, KK4618) with 3.3ng cDNA with a primer concentration of 115nM. Primers are listed
A
in Supplementary Table 1; these were designed to be specific only to predicted orangutan sequences. Samples were run in either duplicates or triplicates and GAPDH and ACTB were used as internal controls. CFX Maestro Software was used for qPCR analyses.Immunofluorescence staining
Attached cells were washed in PBS, floating organoids were washed in 0.04% BSA in PBS. Fixing was performed with 4% PFA for 10-12min. Blocking was performed at room temperature for 2h; blocking buffer contained 2% BSA, 0.3% triton X-100, in PBS. Incubation with primary antibodies was performed at 4° Celsius overnight, before washing with 0.1% tween-20 in wash buffer. Incubation with secondary antibodies and DAPI was performed in the dark at room temperature for 1h. Images were acquired on a fluorescence microscope (Nikon ECLIPSE Ts2, 4X, 10X, 20X) or on an inverted confocal microscope (ZEISS LSM 700, 10X, 20X). ImageJ or ZEISS ZEN Microscopy Software (blue edition) (RRID:SCR_013672) were used for fluorescence or confocal microscopy images respectively, to overlay images and to improve visualisation of the scale bars when necessary. Antibodies are listed in Supplementary Table 2.For blastoids, each immunostaining condition was performed in at least 10 blastoids. Images are only presented if > 50% of the blastoids examined demonstrated spatial localization of embryonic and extraembryonic lineages that resembled those of blastocysts.
Single cell RNA-sequencing library generation
For single cell RNA-sequencing (scRNA-seq), 210 F+blastoids and 200 bi-blastoids were selected on the basis of morphological assessment. The selection criteria for blastoids were the presence of a blastocoel, the presence of an ICM-like structure, and similarity in size to human blastocysts. The selected blastoids were dissociated to obtain single-cell suspensions as described before, with the addition of 0.05X DNAse I (STEMCELL Technologies, #07900). Cell numbers were quantified, and a minimum cell viability of 70% was ensured before proceeding for library generation. For each type of blastoid, approximately 10,000 cells were loaded into a Chromium Next GEM Chip G (10X Genomics, 2000177) for isolation, encapsulation, and construction of gel beads in-emulsions with the Chromium controller (10X Genomics, 1000204) according to the manufacturer’s instructions; GEM was checked for homogenous emulsion and no clog formation, indicating successful GEM generation. Libraries were generated according to the manufacturer's instructions using the Chromium Next GEM Single Cell 3’ GEM and Library & Gel Bead Kit v3.1 (10X Genomics, PN-1000121). Libraries were pooled and sequencing was performed by Mayo Clinic, on the NovaSeq 6000 System (Illumina) at a read depth of approximately 700M pair-end reads of 150bp per sample/library.
Single cell RNA-sequencing analyses
With 10X Genomics data obtained, alignment and UMI counting were executed using the Cell Ranger (RRID:SCR_017344) program under default parameters and the ponAbe3 genome. The feature-barcode count matrix generated by Cellranger was utilized for downstream analyses with Seurat in R. During quality control, cells expressing fewer than 500 genes or having over 10% of mitochondrial gene content were considered abnormal and excluded. Genes expressed in fewer than three cells were excluded. Uniform manifold approximation and projection (UMAP) visualization was performed with 20 dimensions following dimensionality reduction using principal component analysis (PCA). Differentially expressed genes (DEGs) in each cluster or cell type were identified using the FindAllMarkers function. DEGs between samples were identified using the FindMarkers function with Wilcoxon’s rank sum tests, requiring a minimum upregulation of 0.1 log-fold and an adjusted p-value threshold of 0.01. Average expression values for each cell type in each sample were computed. These steps were performed in collaboration with a bioinformatician.
Integrated single cell RNA-sequencing analyses
For the integration of publicly available datasets from human blastocysts 65 and blastoids 38, Seurat’s FindIntegrationAnchors and IntegrateData functions were employed. UMAP was used for dimensionality reduction. Transcriptome correlation was calculated using integrated gene expression values, and average expression values for each cell type in each sample were computed. Based on these average expression values, PCA and hierarchical clustering were conducted across samples. These steps were performed in collaboration with a bioinformatician.
Bulk RNA-sequencing analyses
Total RNA was extracted using Monarch® Total RNA Miniprep Kit. Qualitative control was performed to ensure that total RNA samples were ≥ 400ng in ≥ 20µl for a concentration of ≥ 20ng/µl, with an RNA Integrity Number of ≥ 6.8 (Agilent 2100™ Bioanalyzer) and a purity (OD260/280 and OD260/230) of ≥ 2.0 (NanoDrop™ ND-1000 Spectrophotometer, Thermo Scientific, RRID:SCR_016517).
For bulk RNA-seq, sequencing was performed by Novogene, at 40M PE150 reads per sample. Reads were mapped to the ponAbe3 genome using STAR with default parameters. The GTF annotation file for genes was downloaded from the UCSC genome browser. Cufflinks (v2.2.1) was used for gene assembly and quantification to generate fragments per kilobase per million mapped reads (FPKM) tables based on mapped BAM files and the GTF annotation file. Transcript counts were computed using Htseq (v.0.12.4). DESeq2 was employed for differential gene analysis in samples with replicates, with a p-value cutoff of 0.05 and a fold change cutoff of 1.5. For samples without replicates, differential gene analysis was performed using EdgeR, with a p-value cutoff of 0.05. These steps were performed in collaboration with a bioinformatician.
Principal component analysis
Principal component analysis (PCA) was conducted on RNA-seq counts with at least five mapped reads across at least 10% of samples. Batch effects were corrected using the ComBat function from the sva Bioconductor package (v.3.34.0). PCA analysis was performed using the FactoMineR R package. A cluster dendrogram was plotted following hierarchical clustering with the hclust function.
Statistics and reproducibility
A
The experiments were not randomized, and investigators were not blinded to allocation during experiments and outcome assessments. Data and statistical analyses presented, if not otherwise stated, were plotted and performed using GraphPad Prism. For qPCR, relative normalized gene expression data between two genes were compared using a two sample t-test with a two-tailed p threshold of 0.05, with a Levene’s test for equality of variances with a p threshold of 0.05; relative normalized gene expression data between more than two genes were compared using a one-way ANOVA with a p threshold of 0.05, followed by a post-hoc Tukey’s multiple comparisons test with a p threshold of 0.05. For efficiency, morphology, and ELISA data, a Shapiro-Wilk test for normality was employed with a p threshold of 0.05; normally distributed data was compared using a one-way ANOVA with a p threshold of 0.05, followed by a post-hoc Tukey’s multiple comparisons test with a p threshold of 0.05; and non-normally distributed data compared using either a Mann-Whitney test or a Kruskal-Wallis test with a p threshold of 0.05, followed by a post-hoc Dunn’s multiple comparisons test with a p threshold of 0.05; means were compared with reference means using a one-sample two-tailed t-test with a p threshold of 0.05.Fig. 1
Orangutan iPSCs can form blastoids that encompass the three principal lineages of the preimplantation embryo and recapitulate early cell fate decisions observed in humans. (A) Orangutan primary dermal fibroblasts were reprogrammed to form integration-free iPSCs, which could self-aggregate under specific conditions to form blastoids. (B) Day 0 and Day 6 of orangutan blastoid formation. (C) Orangutan blastoids resemble a hatched human blastocyst. (D) Immunostaining confirming localization of OCT4 (EPI-marker) to cells in the ICM-like compartment, and CDX2 (TE-marker) to cells in the TE-like compartment, of orangutan blastoids. (E) UMAP of the transcriptome of single cells originating from orangutan blastoids, demonstrating 7 cell clusters. (F) UMAP of the transcriptome of single cells originating from orangutan blastoids, annotated; intmd = intermediate. (G) Correlation analysis and dendrogram demonstrating relative transcriptomic similarity of orangutan blastoids to other human blastocysts and blastoids. (H) PCA and correlation analysis demonstrating relative dissimilarity of macaque blastocysts to Great Ape blastocysts and blastoids.
Fig. 2
Aggregation of orangutan F+iPSCs and F−iPSCs at a 1:3 ratio formed blastoids with improved efficiency. (A) Orangutan primary dermal fibroblasts were reprogrammed to form integration-free iPSCs, which were maintained in different conditions, and when aggregated at a specific ratio could self-aggregate under specific conditions to form bi-blastoids. (B) Quantitative RT-PCR (qRT-PCR) analysis for the expression of core pluripotency (NANOG, POU5F1, SOX2, KLF4), naïve pluripotency (DPPA3, DPPA5, DNMT3L, KLF5, KLF17, PRDM14, SUSD2), TE (CDX2, GATA2, GATA3, KRT8, KRT18, PLAC8, TFAP2C, TF63, HAVCR1), PE/HYPO (SOX17, GATA6, PDGFRA), amnion (GABRP, BMP4), and PGC (PRDM1) markers in aggregates harvested from bi-blastoid formation and their starting iPSCs, normalized to GAPDH and ACTB, and controlled to fibroblasts (for pluripotency markers) or bi-blastoids (for TE, PE/HYPO, amnion, and PGC markers). (C) Immunostaining confirming localization of OCT4 (EPI-marker) to cells in the ICM-like compartment, and CDX2 (TE-marker) to cells in the TE-like compartment, of orangutan bi-blastoids. (D) Immunostaining demonstrating localization of SOX2 to GFP− cells in the ICM-like compartment, surrounded by localization of CDX2 to GFP+ cells in the TE-like compartment, in orangutan bi-blastoids.
Fig. 3
Orangutan bi-blastoids exhibited lineage distributions closely resembling human E6-7 blastocysts. (A) Violin plot demonstrating exclusion of cells containing less than 500 genes per cell in scRNA-seq analysis of orangutan bi-blastoids. (B) UMAP of the transcriptome of single cells originating from orangutan bi-blastoids, demonstrating 9 cell clusters. (C) UMAP of the transcriptome of single cells originating from orangutan bi-blastoids, annotated. (D) Different proportions of cell types in human blastocysts, macaque blastocysts, human blastoids, and orangutan bi-blastoids; inter = intermediate. (E) Top 9 DEGs in each cluster in bi-blastoids. (F) Correlation analysis and dendrogram demonstrating relative transcriptomic similarity of orangutan mono- and bi-blastoids to other human blastocysts and blastoids.
Fig. 4
Orangutan blastoids can generate both ESCLCs and TSCLCs. (A) Under lineage-specific culture conditions, orangutan bi-blastoids were able to generate AP-positive ESCLCs, as well as TSCLCs. (B) qRT-PCR analysis for the expression of core pluripotency markers in ESCLCs, normalized to GAPDH and ACTB, and controlled to fibroblasts. (C) qRT-PCR analysis for the expression of TE-associated markers in TSCLCs, compared to F+iPSCs, F−iPSCs, and ESCLCs. (D) Immunostaining confirming positive expression of pluripotency and TE-associated markers in ESCLCs and TSCLCs respectively.
Fig. 5
Orangutan blastoids are capable of progressing beyond pre-implantation stages, exhibiting hallmarks of peri-implantation development in vitro. (A) Immunostaining demonstrating cytotrophoblast-like cells expressing CK7, and extravillous trophoblast-like cells expressing an HLA-G analogue, in orangutan bi-blastoids after in vitro attachment assay. (B) Immunostaining demonstrating polar trophectoderm-like cells expressing NR2F2, and CGβ secreted by syncytiotrophoblast-like cells, in orangutan bi-blastoids after in vitro attachment assay. (C) These are markers associated with maternal-foetal tolerance and early pregnancy maintenance. (D) ELISA measurements of the concentration of the protein CGβ secreted into IVC medium of attached and unattached bi-blastoids at Day 2 and 5 of extended in vitro attachment assay. (E) ELISA measurements of the concentration of the protein CGβ secreted into IVC medium of attached and unattached mono-blastoids at Day 2 and 5 of extended in vitro attachment assay.
AUTHORSHIP
NLT: conceptualization, methodology, software, validation, formal analysis, investigation, data curation, writing (original draft), visualization. QB: methodology, validation, formal analysis, investigation. UMK, YZ: methodology, software, formal analysis, investigation, visualization. KWW: methodology, investigation. CYYL, WKLT, HY: investigation, project administration. KP: investigation. DHHC, CH: resources. SX: writing (review & editing). MC, JYL, SCN: conceptualization, supervision. OP: data curation, writing (original draft), visualization, supervision.
A
A
A
COMPETING INTERESTSThe authors declare no competing interests.
REFERENCES
1.
Khokha MK, Liu KJ, Wallingford JB (2020) Challenges and opportunities at the interface of birth defects, human genetics and developmental biology. Development 147:dev197871
2.
Nicolas P, Etoc F, Brivanlou AH (2021) The ethics of human-embryoids model: a call for consistency. J Mol Med (Berl) 99:569–579. https://doi.org/10.1007/s00109-021-02053-7
3.
Quinn P, Horstman FC (1998) Is the mouse a good model for the human with respect to the development of the preimplantation embryo in vitro? Hum Reprod 13:173–183. https://doi.org/10.1093/humrep/13.suppl_4.173
4.
Cha J, Sun X, Dey SK (2012) Mechanisms of implantation: strategies for successful pregnancy. Nat Med 18:1754–1767
5.
Molè MA, Weberling A, Zernicka-Goetz M (2020) Comparative analysis of human and mouse development: From zygote to pre-gastrulation. Curr Top Dev Biol 136:113–138
6.
Rossant J, Tam PPL (2017) New Insights into Early Human Development: Lessons for Stem Cell Derivation and Differentiation. Cell Stem Cell 20:18–28. https://doi.org/10.1016/j.stem.2016.12.004
7.
Rossant J, Tam PPL (2018) Exploring early human embryo development. Science 360:1075. https://doi.org/10.1126/science.aas9302
8.
Shahbazi MN (2020) Mechanisms of human embryo development: from cell fate to tissue shape and back. Development 147. https://doi.org/10.1242/dev.190629
9.
Rayon T et al (2020) Species-specific pace of development is associated with differences in protein stability. Science 369:eaba7667
10.
Matsuda M et al (2020) Species-specific segmentation clock periods are due to differential biochemical reaction speeds. Science 369:1450–1455
11.
Nakamura T, Fujiwara K, Saitou M, Tsukiyama T (2021) Non-human primates as a model for human development. Stem Cell Rep 16:1093–1103. https://doi.org/https://doi.org/10.1016/j.stemcr.2021.03.021
12.
Chavatte-Palmer P, Tarrade A (2016) Placentation in different mammalian species. Ann Endocrinol (Paris) 77:67–74. https://doi.org/10.1016/j.ando.2016.04.006
13.
Deglincerti A et al (2016) Self-organization of the in vitro attached human embryo. Nature 533:251–254. https://doi.org/10.1038/nature17948
14.
Shahbazi MN et al (2016) Self-organization of the human embryo in the absence of maternal tissues. Nat Cell Biol 18:700–708. https://doi.org/10.1038/ncb3347
15.
Zhou F et al (2019) Reconstituting the transcriptome and DNA methylome landscapes of human implantation. Nature 572:660–664. https://doi.org/10.1038/s41586-019-1500-0
16.
Tyser RCV et al (2020) A spatially resolved single cell atlas of human gastrulation. bioRxiv, 2020.2007.2021.213512 https://doi.org/10.1101/2020.07.21.213512
17.
Xiang L et al (2020) A developmental landscape of 3D-cultured human pre-gastrulation embryos. Nature 577:537–542. https://doi.org/10.1038/s41586-019-1875-y
18.
Li L-C et al (2021) Single-cell patterning and axis characterization in the murine and human definitive endoderm. Cell Res 31:326–344. https://doi.org/10.1038/s41422-020-00426-0
19.
O'leary T et al (2012) Tracking the progression of the human inner cell mass during embryonic stem cell derivation. Nat Biotechnol 30:278–282
20.
Sunde A et al (2016) Time to take human embryo culture seriously. Hum Reprod 31:2174–2182
21.
Ma H et al (2019) In vitro culture of cynomolgus monkey embryos beyond early gastrulation. Science 366. https://doi.org/10.1126/science.aax7890
22.
Gerri C et al (2020) Initiation of a conserved trophectoderm program in human, cow and mouse embryos. Nature 587:443–447
23.
Stamatiadis P et al (2022) TEAD4 regulates trophectoderm differentiation upstream of CDX2 in a GATA3-independent manner in the human preimplantation embryo. Hum Reprod 37:1760–1773
24.
Barker DJ et al (1993) Fetal nutrition and cardiovascular disease in adult life. Lancet 341:938–941
25.
Zinaman MJ, Clegg ED, Brown CC, O’Connor J, Selevan SG (1996) Estimates of human fertility and pregnancy loss. Fertil Steril 65:503–509
26.
Norwitz ER, Schust DJ, Fisher SJ (2001) Implantation and the survival of early pregnancy. N Engl J Med 345:1400–1408
27.
Wilcox AJ, Baird DD, Weinberg CR (1999) Time of implantation of the conceptus and loss of pregnancy. N Engl J Med 340:1796–1799
28.
Gluckman PD, Hanson MA (2008) Developmental and epigenetic pathways to obesity: an evolutionary-developmental perspective. Int J Obes 32:62–S71
29.
Jarvis GE (2016) Early embryo mortality in natural human reproduction: What the data say. F1000Research 5
30.
Martin GR, Evans MJ (1975) Differentiation of clonal lines of teratocarcinoma cells: formation of embryoid bodies in vitro. Proceedings of the National Academy of Sciences 72, 1441–1445
31.
Martin GR, Wiley LM, Damjanov I (1977) The development of cystic embryoid bodies in vitro from clonal teratocarcinoma stem cells. Dev Biol 61:230–244
32.
Bradley A, Evans M, Kaufman MH, Robertson E (1984) Formation of germ-line chimaeras from embryo-derived teratocarcinoma cell lines. Nature 309:255–256. https://doi.org/10.1038/309255a0
33.
Nagy A, Rossant J, Nagy R, Abramow-Newerly W, Roder JC (1993) Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc Natl Acad Sci U S A 90:8424–8428. https://doi.org/10.1073/pnas.90.18.8424
34.
Rivron N, Fu J, SnapShot (2021) Embryo models. Stem Cell Reports 16, 1142–1142 e1141 https://doi.org/10.1016/j.stemcr.2021.04.012
35.
Fu J, Warmflash A, Lutolf MP (2021) Stem-cell-based embryo models for fundamental research and translation. Nat Mater 20:132–144. https://doi.org/10.1038/s41563-020-00829-9
36.
Rossant J, Tam PPL (2021) Opportunities and challenges with stem cell-based embryo models. Stem Cell Rep 16:1031–1038. https://doi.org/10.1016/j.stemcr.2021.02.002. https://doi.org/
37.
El Azhar Y, Sonnen KF (2021) Development in a Dish-In Vitro Models of Mammalian Embryonic Development. Front Cell Dev Biol 9:655993. https://doi.org/10.3389/fcell.2021.655993
38.
Liu X et al (2021) Modelling human blastocysts by reprogramming fibroblasts into iBlastoids. Nature 591:627–632. https://doi.org/10.1038/s41586-021-03372-y
39.
Yu L et al (2021) Blastocyst-like structures generated from human pluripotent stem cells. Nature 591:620–626
40.
Yanagida A et al (2021) Naive stem cell blastocyst model captures human embryo lineage segregation. Cell Stem Cell 28:1016–1022
41.
Kagawa H et al (2022) Human blastoids model blastocyst development and implantation. Nature 601:600–605. https://doi.org/10.1038/s41586-021-04267-8
42.
Fan Y et al (2021) Generation of human blastocyst-like structures from pluripotent stem cells. bioRxiv, 2021.2003.2009.434313 https://doi.org/10.1101/2021.03.09.434313
43.
Sozen B et al (2021) Reconstructing aspects of human embryogenesis with pluripotent stem cells. Nat Commun 12:5550
44.
Zhao C et al (2022) Single-cell multi-omics of human preimplantation embryos shows susceptibility to glucocorticoids. Genome Res 32:1627–1641
45.
Lovell-Badge R et al (2021) ISSCR Guidelines for Stem Cell Research and Clinical Translation: The 2021 update. Stem Cell Rep 16:1398–1408. https://doi.org/10.1016/j.stemcr.2021.05.012
46.
Tomoda K, Kime C (2021) Synthetic embryology: Early mammalian embryo modeling systems from cell cultures. Dev Growth Differ 63:116–126. https://doi.org/10.1111/dgd.12713
47.
Niu Y et al (2019) Dissecting primate early post-implantation development using long-term in vitro embryo culture. Science 366. https://doi.org/10.1126/science.aaw5754
48.
Sankai T (2000) In vitro manipulation of nonhuman primate gametes for embryo production and embryo transfer. Exp Anim 49:69–81
49.
Nakamura T et al (2016) A developmental coordinate of pluripotency among mice, monkeys and humans. Nature 537:57–62. https://doi.org/10.1038/nature19096
50.
Li J et al (2023) Cynomolgus monkey embryo model captures gastrulation and early pregnancy. Cell Stem Cell 30:362–377
A
51.Pääbo S (2003) The mosaic that is our genome. Nature 421:409–412. https://doi.org/10.1038/nature01400
A
52.Gibbs RA et al (2007) Evolutionary and biomedical insights from the rhesus macaque genome. Science 316:222–234. https://doi.org/10.1126/science.1139247
A
53.Perelman P et al (2011) A molecular phylogeny of living primates. PLoS Genet 7:e1001342
54.
Ross C, Boroviak TE (2020) Origin and function of the yolk sac in primate embryogenesis. Nat Commun 11:3760
55.
Ancrenaz M et al (2016) Pongo pygmaeus
56.
Grehan JR, Schwartz JH (2009) Evolution of the second orangutan: phylogeny and biogeography of hominid origins. J Biogeogr 36:1823–1844. https://doi.org/10.1111/j.1365-2699.2009.02141.x
57.
Locke DP et al (2011) Comparative and demographic analysis of orang-utan genomes. Nature 469:529–533. https://doi.org/10.1038/nature09687
58.
Gould KG (1983) Ovum recovery and in vitro fertilization in the chimpanzee. Fertil Steril 40:378–383
59.
Puschmann W (1989) Zootierhaltung. Dt. Landwirtschaftsverl
60.
Pope CE et al (1997) (Wiley Online Library
61.
Carter AM, Pijnenborg R (2011) Evolution of invasive placentation with special reference to non-human primates. Best Pract Res Clin Obstet Gynecol 25:249–257
62.
Silber SJ, Barbey N, Lenahan K, Silber DZ (2013) Applying clinically proven human techniques for contraception and fertility to endangered species and zoo animals: a review. J Zoo Wildl Med 44
63.
Yan L et al (2013) Single-cell RNA-Seq profiling of human preimplantation embryos and embryonic stem cells. Nat Struct Mol Biol 20:1131–1139
64.
Boroviak T et al (2015) Lineage-specific profiling delineates the emergence and progression of naive pluripotency in mammalian embryogenesis. Dev Cell 35:366–382
65.
Petropoulos S et al (2016) Single-cell RNA-seq reveals lineage and X chromosome dynamics in human preimplantation embryos. Cell 165:1012–1026
66.
Stirparo GG et al (2018) Integrated analysis of single-cell embryo data yields a unified transcriptome signature for the human pre-implantation epiblast. Development 145:dev158501
67.
Imamura S et al (2022) Non-cell-adhesive hydrogel promotes formation of human blastoids from primed human pluripotent stem cells. bioRxiv, 2006. 2023.497328 (2022)
68.
Hu Y et al (2021) Single-cell analysis of nonhuman primate preimplantation development in comparison to humans and mice. Dev Dyn 250:974–985
69.
Blakeley P et al (2015) Defining the three cell lineages of the human blastocyst by single-cell RNA-seq. Development 142:3151–3165. https://doi.org/10.1242/dev.123547
70.
Shapiro BS, Daneshmand ST, Garner FC, Aguirre M, Thomas S (2008) Large blastocyst diameter, early blastulation, and low preovulatory serum progesterone are dominant predictors of clinical pregnancy in fresh autologous cycles. Fertil Steril 90:302–309
71.
Rivron NC et al (2018) Blastocyst-like structures generated solely from stem cells. Nature 557:106–111. https://doi.org/10.1038/s41586-018-0051-0
72.
Guo G et al (2021) Human naive epiblast cells possess unrestricted lineage potential. Cell Stem Cell 28, 1040–1056 e1046 https://doi.org/10.1016/j.stem.2021.02.025
73.
Posfai E, Lanner F, Mulas C, Leitch HG (2021) All models are wrong, but some are useful: Establishing standards for stem cell-based embryo models. Stem Cell Rep 16:1117–1141
74.
Zhao C et al (2007) Reprogrammed blastoids contain amnion-like cells but not trophectoderm. biorxiv, 2021.2005. 442980 (2021)
75.
Meistermann D et al (2021) Integrated pseudotime analysis of human pre-implantation embryo single-cell transcriptomes reveals the dynamics of lineage specification. Cell Stem Cell 28, 1625–1640. e1626
76.
Huang T et al (2025) Inhibition of PRC2 enables self-renewal of blastoid-competent naive pluripotent stem cells from chimpanzee. Cell stem cell 32, 627–639. e628
77.
Xu RH et al (2002) BMP4 initiates human embryonic stem cell differentiation to trophoblast. Nat Biotechnol 20:1261–1264. https://doi.org/10.1038/nbt761
78.
Amita M et al (2013) Complete and unidirectional conversion of human embryonic stem cells to trophoblast by BMP4. Proc Natl Acad Sci U S A 110:E1212–1221. https://doi.org/10.1073/pnas.1303094110
79.
Cheung VC et al (2021) Pluripotent stem cell-derived endometrial stromal fibroblasts in a cyclic, hormone-responsive, coculture model of human decidua. Cell Rep 35
80.
Jang YJ, Kim M, Lee BK, Kim J (2022) Induction of human trophoblast stem-like cells from primed pluripotent stem cells. Proc Natl Acad Sci U S A 119:e2115709119. https://doi.org/10.1073/pnas.2115709119
81.
Okita K et al (2011) A more efficient method to generate integration-free human iPS cells. Nat Methods 8:409–412. https://doi.org/10.1038/nmeth.1591
82.
O'Brien SJ, Menninger JC, Nash WG (2006) Atlas of mammalian chromosomes. Wiley
83.
Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25 years of image analysis. Nat Methods 9:671–675. https://doi.org/10.1038/nmeth.2089