Researchers have already used sperm-like cells grown from stem cells to produce live, fertile mice, but the technology remains years from safe use in humans. The gap between that mouse milestone and a human clinic is enormous, filled with unsolved problems in epigenetics, safety testing, and regulation. Still, the science is advancing fast enough that fertility specialists, conservation biologists, and ethicists are all actively preparing for a future in which sperm can be made in a dish from almost any cell in the body.
How Scientists Make Sperm in a Lab
The basic idea is to coax stem cells through the same developmental steps that normally happen inside the testes, but to do it in a culture dish instead. In 2006, a team working with mouse embryonic stem cells managed to guide those cells through meiosis, the specialized cell division that halves the chromosome count, and produce haploid male gametes. When those lab-derived cells were injected into mouse eggs, the resulting embryos developed normally, and live pups were born.
A decade later, a Chinese group refined the approach considerably. Starting again with mouse embryonic stem cells, they walked the cells through each checkpoint of natural sperm development: erasing and resetting chemical tags on the DNA, pairing up chromosomes, and shuffling genetic material between them. The spermatid-like cells that emerged at the end of this process were injected into mouse eggs and produced viable, fertile offspring, meaning those mice could themselves go on to reproduce.
These mouse experiments remain the clearest proof that lab-made male gametes can do the full job of fertilization and support normal development. No equivalent result exists yet in humans or other primates, but the mouse work provides a detailed roadmap for what each stage of the process should look like.
Where Human Research Stands
Translating mouse success to human cells has proved far harder. In 2012, researchers showed that human embryonic stem cells and human induced pluripotent stem cells (iPSCs, which are ordinary body cells reprogrammed back to a flexible state) could be pushed to form spermatid-like cells without any genetic engineering. These cells expressed proteins found exclusively in spermatids and sperm, and they showed DNA imprinting patterns similar to those in natural human sperm at certain key genetic locations.1PubMed Central. Direct Differentiation of Human Pluripotent Stem Cells into Haploid Spermatogenic Cells That was encouraging, but these cells were never tested for actual fertilization ability, and there is no ethical framework that would currently allow such a test in humans.
More recently, a 2024 preprint described a system in which human iPSCs were differentiated into primordial germ cell-like cells and then placed inside reconstituted testicular tissue made from mouse fetal cells. After transplantation into mice, these hybrid structures generated spermatogonia, the precursor cells that normally give rise to sperm, along with early-stage spermatocytes that closely resembled their natural counterparts. The same group achieved similar results with rhesus monkey iPSCs, an important step because any future clinical path will require primate testing before human trials.2bioRxiv. Generation of spermatogonia from pluripotent stem cells in humans and non-human primates
So the current state of play in humans is this: researchers can produce cells that look and behave like several intermediate stages of sperm development, but nobody has yet completed the entire journey from stem cell to a fully functional human sperm cell. The final steps of meiosis and the physical maturation that turns a round spermatid into a swimming sperm cell remain stubbornly difficult to replicate outside the body.
Testicular Organoids and the Problem of Environment
Sperm do not develop in isolation. Inside the body, they depend on an intricate support system: Sertoli cells that nurse them, Leydig cells that produce testosterone, and peritubular cells that provide structural scaffolding. Recreating this environment in the lab has become a research focus in its own right. Teams have built three-dimensional testicular organoids containing all four major cell types, and these miniature structures continuously produced testosterone and showed molecular signs that a small fraction of the germ cells within them had begun the transition from diploid to haploid.3Biology of Reproduction. Three-dimensional testicular organoid: a novel tool for the study of human spermatogenesis and gonadotoxicity in vitro
These organoids serve a dual purpose. Beyond being a potential incubator for lab-grown sperm, they function as toxicity screening tools. When exposed to chemotherapy drugs, the organoids responded in a dose-dependent way, and they tolerated higher concentrations than flat cell cultures before dying. That makes them useful for testing which cancer treatments are most damaging to fertility, a question that matters a great deal to young patients facing chemotherapy.
Scaling up stem cell cultures is another engineering challenge. Conventional lab dishes support modest cell densities, but automated feeding systems and specialized culture surfaces have boosted mouse embryonic stem cell densities several-fold while maintaining the cells’ ability to become any cell type.4PubMed. High density cultures of embryonic stem cells That kind of bioprocess optimization will be essential if lab-grown gametes are ever to move from a research curiosity to a clinical product.
What This Could Mean for Male Infertility
The most immediate clinical audience for synthetic sperm is men with nonobstructive azoospermia, a condition in which the testes produce little or no sperm. This accounts for roughly 60% of all azoospermia cases and represents one of the most severe forms of male infertility. The current standard of care involves surgically extracting tiny amounts of tissue from the testes and searching for usable sperm, but even with the most refined surgical techniques, sperm is found only about half the time.5PubMed Central. Future prospects for the advancement of treatment of men with NOA: focus on gene editing, artificial sperm, stem cells, and use of imaging For the other half, there is currently no way to have a genetically related child.
If functional sperm could be derived from a skin biopsy or a blood sample reprogrammed into iPSCs, it would bypass the testicular defect entirely. The man’s own genome would still be represented in the resulting gamete, preserving the genetic connection to any offspring. This is the application that drives much of the funding and attention around in vitro gametogenesis, and it is the one that fertility patients and their doctors ask about most frequently.
There is also interest in combining lab-grown gametes with gene-editing tools. A recent narrative review explored how CRISPR-based editing could theoretically be used alongside regenerative approaches to correct genetic causes of infertility before deriving gametes, though the authors stressed that safety, efficacy, and ethical integrity would need to be firmly established before any such combination entered practice.6PubMed. Genome Editing for Fertility: Unlocking the Promise of CRISPR/Cas9 in Addressing Male Infertility – A Narrative Review
Conservation of Endangered Species
The applications extend well beyond human medicine. For critically endangered animals with tiny populations and dwindling genetic diversity, the ability to generate gametes from preserved cells could be transformative. Researchers have proposed combining two technologies: creating gametes from iPSCs derived from banked tissue samples, and then using surrogate mothers of a related, more abundant species to carry pregnancies. The immunological trick involves placing the endangered species’ embryo inside a trophoblast shell from the surrogate’s own species, essentially fooling the surrogate’s immune system into accepting the pregnancy.7PubMed. Reproductive biotechnology and critically endangered species: Merging in vitro gametogenesis with inner cell mass transfer
An early demonstration of cross-species germ cell production has already been achieved in rodents. When rat embryonic stem cells were injected into genetically modified mouse embryos that could not produce their own germ cells, the resulting chimeras grew testes that contained exclusively rat-origin sperm. Those rat spermatozoa were capable of fertilizing oocytes.8Stem Cell Reports. Germline blastocyst complementation of the male niche in intra- and interspecies rodent chimeras The implications are significant: if a closely related common species can serve as a living bioreactor for producing an endangered species’ gametes, frozen cell lines from animals that died years ago could theoretically contribute offspring to the population. A 2025 review highlighted breakthroughs in differentiating iPSCs into functional gametes as a way to circumvent the chronic shortage of donor eggs and the genetic bottlenecks that threaten small populations.9PubMed Central. Application of induced pluripotent stem cells in the conservation of endangered animals
The Epigenetic Problem
The single biggest scientific obstacle to clinical use is not making cells that look like sperm. It is making sure those cells carry the right pattern of chemical marks on their DNA and on the histone proteins around which DNA wraps. These marks, collectively called the epigenome, control which genes are active in the embryo and throughout life. During natural sperm development, old marks are erased and new ones are laid down in a precise sequence. Getting that sequence wrong, even slightly, can have consequences that persist for generations.
Mouse studies have demonstrated this in dramatic fashion. When researchers disrupted histone modifications in developing sperm by overexpressing a single enzyme, the offspring suffered severe developmental defects and reduced survival. Those problems persisted into subsequent generations even when the enzyme was no longer present, carried forward by altered molecular profiles in the sperm.10PubMed. Disruption of histone methylation in developing sperm impairs offspring health transgenerationally The finding is sobering: errors in the epigenome are not just one-generation problems. They can echo forward in ways that are hard to predict and harder to reverse.
Studies of men with impaired sperm production have found that imprinting errors, where specific genes carry the wrong methylation pattern, do occur in their spermatogenic cells, including in elongated spermatids and spermatozoa.11PubMed. DNA methylation imprinting errors in spermatogenic cells from maturation arrest azoospermic patients If natural spermatogenesis can go wrong in this way, the concern is that an artificial process, which lacks the full regulatory environment of the testis, would be even more prone to such errors. Verifying epigenetic fidelity in lab-derived gametes before they are ever used clinically is considered a non-negotiable safety requirement by the research community.
Same-Sex and Solo Reproduction
One of the most discussed possibilities is the creation of sperm from female cells, or eggs from male cells. A systematic review noted that artificial sperm has been generated from female iPSCs in early-stage research.12Human Reproduction Update. Artificial gametes: a systematic review of biological progress towards clinical application If this were ever perfected, two women could theoretically have a child genetically related to both of them, with one partner’s cells providing the sperm and the other contributing the egg. Similarly, male cells could in principle be coaxed to produce eggs, enabling two men to be genetic co-parents of the same child.
The biological barriers here are substantial. Female cells carry two X chromosomes and no Y chromosome, meaning any sperm derived from them could only pass on an X. Children of two genetic mothers would always be female. Male-derived eggs face an even steeper challenge because the Y chromosome lacks much of the genetic material present on the second X that appears to play a role in oocyte development. These are not insurmountable problems in theory, but they are barely past the proof-of-concept stage, and no one has demonstrated functional cross-sex gametes in any mammalian species.
The social dimension is equally complex. Surveys from Japan found that while more than half of respondents accepted the use of lab-derived gametes by infertile heterosexual married couples, the same respondents largely rejected their use by unmarried or same-sex individuals.13PubMed Central. Public attitudes in Japan toward the reproductive use of gametes derived from human-induced pluripotent stem cells In Belgium, attitudes were more permissive: women, younger respondents, and non-heterosexual individuals were significantly more likely to support same-sex parenthood through these technologies than men, older respondents, and those with religious affiliations.14Reproductive BioMedicine Online. Enthusiasm, concern and ambivalence in the Belgian public’s attitude towards in-vitro gametogenesis The variation suggests that even if the science were solved tomorrow, social acceptance would vary enormously by country and community.
Ethical and Regulatory Guardrails
The International Society for Stem Cell Research, which sets widely referenced guidelines for the field, has been explicit about where lines should currently be drawn. Research that creates embryos from lab-derived gametes and maintains them up to 14 days in a dish is considered permissible, provided it undergoes specialized ethical review. But transferring such an embryo into a human uterus for pregnancy is currently classified as a prohibited activity. The society’s reasoning is straightforward: there is no compelling evidence that lab-derived human gametes are safe for reproduction, given unresolved concerns about epigenetic and genetic abnormalities. Even in mice, where the technology is most advanced, lab-derived oocytes and sperm-like cells are reported to be lower quality than their naturally produced equivalents.15Stem Cell Reports. Human embryo research, stem cell-based embryo models and in vitro gametogenesis: Considerations leading to the revised ISSCR guidelines
The guidelines also address some of the more exotic possibilities that make ethicists uneasy. Experiments designed to transfer a human lab-derived embryo into an animal uterus are banned outright, as is breeding animal chimeras that contain human cells capable of forming human gametes. These restrictions reflect a broad international consensus rather than one country’s regulatory framework.
Beyond the laboratory rules, legal questions pile up quickly. Could a person create gametes from someone else’s discarded cells without consent? How would parentage be established when a child’s genetic material came from reprogrammed skin cells rather than conventional gametes? What about posthumous reproduction? Survey data suggest that a majority of adults of reproductive age would want a surviving partner to be able to use their gametes after an unexpected death, with about 70% of men and 58% of women between 18 and 44 expressing that wish.16Oxford Academic. Creating life after death: should posthumous reproduction be legally permissible without the deceased’s prior consent? Lab-derived gametes would make such scenarios easier to carry out and harder to regulate, especially if sperm could be generated from preserved tissue long after death.
The Designer Baby Concern
Nearly 70% of Japanese survey respondents opposed using lab-derived gametes to design a baby, even among those who otherwise supported the technology for infertility treatment.13PubMed Central. Public attitudes in Japan toward the reproductive use of gametes derived from human-induced pluripotent stem cells The worry is not hypothetical. If gametes can be produced in bulk from iPSCs, it becomes theoretically possible to generate many embryos from the same pair of genetic parents, screen them extensively, and select for desired traits. Current IVF already allows preimplantation genetic testing, but the number of embryos is limited by how many eggs can be retrieved in a cycle. Lab-grown gametes could remove that bottleneck.
A 2025 perspective in Trends in Genetics went further, arguing that fully synthetic DNA technologies could eventually allow gametes to be assembled with precise control over genetic inheritance, raising questions about human identity, genetic selection, and what the authors called “evolutionary boundaries.”17Trends in Genetics. Synthetic gametes and the non-identity problem: the babies of tomorrow That scenario is far more speculative than iPSC-derived gametes, but it illustrates why policymakers are watching the field closely. The distance between treating infertility and engineering offspring is measured partly in technology and partly in regulation, and the technology tends to move faster.
What Stakeholders Actually Want
When diverse groups including patients, clinicians, ethicists, and advocacy organizations were surveyed about their hopes and concerns for in vitro gametogenesis, the responses clustered around three themes. People hoped for higher reproductive success rates than current assisted reproduction offers, less physical and emotional suffering during treatment, and greater social inclusion for people who cannot currently have genetically related children. Their concerns focused on equity (who would be able to afford it), safety (whether the offspring would be healthy), and the broader societal implications of decoupling reproduction from the body so completely.18Stem Cell Reports. Anticipating in vitro gametogenesis: Hopes and concerns for IVG among diverse stakeholders
The equity concern deserves attention because existing assisted reproduction is already expensive and unevenly accessible. A single IVF cycle can cost thousands of dollars, and lab-derived gametes would add layers of cell culture, quality testing, and epigenetic verification on top of that. Whether the technology would eventually reduce costs by eliminating surgical sperm retrieval and hormone stimulation, or increase them by adding complexity, is genuinely unclear. The Belgian survey found that respondents with self-reported fertility problems were more supportive of the technology than the general public, which makes intuitive sense: the people who stand to benefit most are the ones least likely to object.14Reproductive BioMedicine Online. Enthusiasm, concern and ambivalence in the Belgian public’s attitude towards in-vitro gametogenesis
Timelines and Open Questions
Predicting when synthetic sperm might reach the clinic is a fool’s errand, but the trajectory of the science offers some rough signposts. The mouse work progressed from first proof of concept in 2006 to fully characterized, meiosis-complete gametes producing fertile offspring by 2016.19PubMed. In vitro-differentiated embryonic stem cells give rise to male gametes that can generate offspring mice20Cell Stem Cell. Complete Meiosis from Embryonic Stem Cell-Derived Germ Cells In Vitro Human research, by contrast, is still working on reliably producing the precursor cell stages. If the human timeline follows a similar pace, functional human sperm-like cells in a lab dish might be achievable within the next decade, but demonstrating that those cells are safe enough for reproduction would take substantially longer.
Several problems have no clear solution yet. The epigenetic fidelity question is the most pressing: no one has developed a comprehensive, cost-effective way to verify that every relevant chemical mark on a lab-derived gamete’s DNA matches the natural pattern. The long-term health of offspring is another unknown that can only be assessed over years or decades, which does not lend itself to quick clinical trials. And the regulatory framework is still catching up. Most countries do not have laws specifically addressing gametes derived from stem cells, which means the rules are being written in real time as the science advances.
For patients waiting for this technology, the honest message from the field is one of cautious optimism. The fundamental biology works in mice. The human biology is yielding to persistent effort. But the gap between a spermatid-like cell in a dish and a healthy baby is one that researchers are unwilling to rush across, and for good reason.