Reproductive science and wildlife conservation: the knowns and unknowns!

Abstract

Graphical Abstract

Abstract

Conservation biologists specializing in reproductive science for the support of threatened species are hampered by many knowledge gaps. One of the objectives of this review is to point out a few of those gaps. We discuss the surprising observation that modern genomic technologies are revealing inaccuracies in the identification of cryptic species, which is critically important for wildlife biobank management or where reproductive interventions are required. If confronted with a familiar species, but of unfamiliar origin, it would be unwise to assume that every aspect of their reproductive biology is exactly as expected. Serious gaps in detailed knowledge can mislead the biologist, and new projects and species are therefore usually hampered by unforeseen problems. Developmental plasticity is one of the key processes behind species diversity, whereby instead of embryos always developing along inflexible pathways, it is now clear that embryonic development can be modified by modulating gene expression through epigenetic mechanisms. In some cases, animals can improve their survival by migrating to optimal feeding or breeding grounds. Others can exploit reproductive delays that relax the synchrony that normally forces mating, or insemination, to occur at the same time as ovulation. While genetic properties such as inbreeding are known to modulate reproductive success, more recent research shows that even the microbiome can influence mammalian fertility. It is therefore becoming clearer that not only has evolution resulted in a bewildering diversity of reproductive adaptations, but attempts to support wildlife have to be backed up by a great deal of fundamental knowledge.

Lay summary

All animal species, past and present alike, are the products of evolution, and although they have diversified in multiple directions, they currently occupy niches that optimize their future chances of survival. While animal evolution as an abstract concept usually brings to mind a series of changes taking place over long timescales, recent advances in developmental biology have shown that many animal species can also adapt surprisingly rapidly if their environment changes. For example, some tadpoles and fish embryos can sense that their watery habitat is drying out, whereupon they can either accelerate their development or bury themselves in mud until the waters return. How do they do it? While biologists are gradually discovering the mechanisms behind these, and other, unexpected responses confront biologists every day with knowledge gaps that have to be filled. Here, we explore some of these gaps and their influence on efforts to protect threatened species.

Introduction

The objective of this review is to offer an overview of major gaps in reproductive science to experts involved in the ongoing struggle to support the future survival of rare animal species. This is an aspect of conservation biology whose complexity is easily and consistently underestimated. Here, we draw attention to the diversity of reproductive strategies across different taxonomic groups, especially as that diversity is being continuously adapted in the face of relentless environmental challenges.

Although animal reproduction is a natural process, it can be fraught with hazards. Leaving aside predators, and increasing urbanization, which threaten individuals and populations, reproductive success currently has to be achieved in the face of accelerated climate change. This is disrupting the seasons and, hence, inducing significant changes to terrestrial and aquatic wildlife habitats. As a consequence, established food availability networks are being seriously disrupted, forcing many species to find effective solutions that mitigate problems associated with poor food availability. Some marine or avian species can migrate towards more suitable habitats, while many terrestrial species do not have that option (Torstenson & Shaw 2025). A few reptiles, especially turtles, have developed temperature-dependent sex determination mechanisms, whereby females can control nest locations, giving them the ability to control the nest temperature to some extent (Santidrián Tomillo & Spotila 2020). This also allows them to avoid extreme, and potentially lethal, incubation temperatures.

Incidentally, many female reptiles, fishes, birds, and a small number of mammals have developed a more sophisticated reproductive strategy. In species with the ability to store spermatozoa within the female reproductive tract, the synchrony between ovulation and mating is less critical for fertilization success than in those without that capacity.

Animal conservation requires multidisciplinary skills and approaches to cater for the diversity of locally and geographically different situations to be addressed. Translocations or physically moving animals from one place to another has been used widely, for different reasons and with different outcomes. Sometimes, that can be a popular and effective policy, readily exemplified by the reintroduction of beavers into some rivers in the United Kingdom. These animals have effectively re-engineered water flows and helped to protect against flooding. ‘Rewilding’ is a somewhat similar conservation effort, but it is often highly controversial. It typically involves introducing, or reintroducing, various species to regions that they had inhabited in the past. The controversies are most likely to arise when proposals involve human–wildlife interactions, such as reintroducing top predators (e.g. lynx or Scottish wildcats), into highly populated areas. Saving and reintroducing the black-footed ferret in the USA, which included significant assisted reproductive technology achievements (Santymire et al. 2014), was not universally welcomed by local ranchers.

Reproductive and developmental success both involve some degree of forecasting in the face of changing future environments and are currently a topic of massive interest. Many species have developed the ability to respond to, and avoid, various threats to their survival. Developmental plasticity is now known to be an extremely important component of the sometimes rapid, natural responses that species can deploy. Among the responses that animals have evolved over longer periods is the ability to adjust their breeding seasons in complex ways. These adjustments help to ensure that new generations are produced under optimal conditions.

Advances in gamete and embryo cryopreservation technology over the past few decades have been sufficiently effective to support various policies aimed at the long-term preservation of genetic materials in the so-called biobanks. The knowledge amassed by cryobiologists has been invaluable in this venture, but unfortunately, applying this technology across multiple species is plagued by species differences. This represents another, and major, source of ‘unknowns’ and means that establishing effective biobanking programmes can only be successful if accompanied by suitable research support and access to animals.

The establishment of conservation-focused biobanks for multiple wild animal species, some of which might focus entirely on locally adapted species, has highlighted the imperative need to accompany the collected biological samples with as much accurate ancillary information as possible. It is, nevertheless, the case that sometimes, the relevant information may not be reliable, and in this review, we discuss why this may be the case and why it is important. Another purpose of this review is to point out some of the sources of uncertainty (‘unknowns’) and encourage biobank scientists and conservationists to understand, or avoid, them as far as possible. Some of the ‘unknowns’ are far from obvious but may have potentially important impacts on the future survival and well-being of the species they intend to support.

Finally, we also consider both the risks and mitigations that must be addressed within the contexts of wildlife conservation and how some of the technologies might raise ethical questions that are sometimes controversial.

Impacts of translocations and rewilding

Some species respond to the warming climate by migrating to more suitable habitats. This approach to nature conservation has been emulated to some extent by programmes that involve deliberate and planned animal translocations and rewilding (Seddon 2010). Although these activities are well intentioned, and could be regarded as variants of conservation technology, the outcomes are fraught with uncertainties and challenges. This approach can also have problematic outcomes, one of which is the unintentional introduction of alien species into new and biologically naïve habitats (Seddon & Redford 2025). Animal translocations can also affect reproductive success by changing mating behaviour, competitiveness, and territoriality (Berger-Tal et al. 2020). Incidentally, the effects of animal translocations and introductions would be mirrored in the context of de-extinctions (Sandler 2017), although this has yet to be confirmed.

When species experience environmental change, they often respond with subtle, or even major, physiological changes. These can occur surprisingly quickly, even from one generation to the next. It is now recognized that these rapid adaptations can occur via mechanisms involving ‘epigenetic’ change and developmental plasticity, without necessarily altering the organism’s genomic DNA sequences. Recent advances in molecular genomics have identified, and partially filled, one of the important gaps in our knowledge of adaptation, as it is now known that epigenetic change can occur in many, if not all, species (Jeremias et al. 2018). This realization represents a major advance in knowledge about the evolution of species and how they have responded to the challenges of their specific environments. Multiple regulatory mechanisms involving, for example, histone modifications of the genome, DNA methylation, non-coding RNAs, and three-dimensional chromatin organization, can influence the way that genes are expressed. Some of these mechanisms have been understood for more than two or three decades, while others have only recently been recognised.

Surprisingly, it has recently been shown that well-focused and successful conservation programmes can themselves induce short-term epigenetic responses. Recent detailed epigenetic studies of the black-footed ferret, a species that almost became extinct in the 1980s but has since been bred successfully in captivity (Santymire et al. 2014), have shown that captive breeding induced complex epigenetic changes associated with reductions in male and female fertility (Ali et al. 2024, Bornbusch et al. 2024, Tennenbaum et al. 2024). These findings imply that species involved in captive breeding programmes are also adapting to environmental changes, including changes to their microbiome (Bornbusch et al. 2024). In fact, some researchers have coined the term ‘exposome’ (Ottinger & Geiselman 2025) to capture the principle that changed environments, caused by both anthropogenic and stochastic influences, elicit multiple changes to the survival prospects of all species. This principle opens a hitherto unsuspected window on the effectiveness and outcomes of wildlife conservation programmes.

Forecasting and responding to environmental change

Transgenerational reproductive forecasting, sometimes known as ‘anticipatory transgenerational phenotypic plasticity’, has been observed in several mammalian species, including humans, and helps the organism to match the path of embryonic development against expectations. This sophisticated aspect of developmental plasticity serves to optimize the survival of subsequent generations. In fact, this relationship is at the heart of reproductive seasonality, where species across many taxonomic groups have developed physiological strategies that help to synchronize breeding and parturition with optimal environmental conditions. If the forecast turns out to be inaccurate because the environment changes and defies expectations, the mismatch can have unfortunate and lifelong health consequences (Gluckman et al. 2019).

Organisms are now known to use such developmental forecasting to predict how to prepare for possible and stressful changes in nutrition, temperature, salinity, and even the presence of environmental toxicants. Such relationships are now well recognized in humans and other mammals, owing to a series of studies linking early life nutrition to the later onset of diseases in adulthood. The realization that there are systematic relationships between pre- and/or peri-conception nutritional conditions and chronic disease conditions in adults has stimulated a great deal of research in humans and other species. It has even been shown, in humans, that transgenerational effects on body condition in males are related to the father’s and grandfather’s smoking habits (Pembrey et al. 2006). In a similar vein, reproductive success in a wild desert rodent (Meriones crassus) was highest when the rates of flea infestation in mothers and grandmothers were similar. When both were either infested or non-parasitized, the pups would reach sexual maturity more quickly than those pups whose mothers’ infestation status did not match that of their grandmothers (Warburton et al. 2020).

Environmental influences on fish development have long been a topic of intense interest to researchers (reviewed by Jonsson & Jonsson (2014)). Multiple aspects of the life history of fishes, including growth and adult body size, sex ratio, egg size, lifespan, and tendency to migrate, are affected by early influences, and although the mechanisms responsible are poorly understood, it is becoming ever more apparent that epigenetics is a central theme. As an example, using sea bass (Dicentrarchus labrax) as a model species, Navarro-Martín et al. (2011) demonstrated sex ratio skewing in this species, where higher water temperatures during a critical period in early development induce higher proportions of male offspring. This was associated with increased DNA methylation of the aromatase promoter. Even the pre-fertilization temperature experienced by Atlantic salmon gametes has recently been shown to affect not only gamete performance but also offspring fitness (Graziano et al. 2023). For example, thermal stress in Atlantic cod embryos resulted in stage-specific changes in the expression profiles of 5-methyltransferase enzymes (Skjærven et al. 2014), while temperature changes altered the expression of microRNAs during embryonic development in Senegalese sole (Solea senegalensis) (Campos et al. 2014). These changes were associated with variations in growth rate, muscle development, and lipid metabolism. Importantly, these adaptations are due to changing seasons and annual weather variations, but not the extreme changes in climate.

Examples of natural adaptations: reproductive delays, in vivo sperm storage, and delayed implantation

The capacity to store spermatozoa within the female reproductive tract allows for a temporal separation between mating and ovulation, giving females the flexibility to mate at times that are not directly correlated with ovulation (Holt & Fazeli 2016, Liu et al. 2016). This can be advantageous in various scenarios, such as when mating opportunities are limited or when females elect to choose the fertilizing ejaculate, e.g. through cryptic female choice (Orr & Zuk 2014).

In the context of conservation breeding using assisted reproductive methods, the capacity for prolonged sperm storage is not necessarily an advantage. In principle, it might be imagined that inseminations could be performed without the necessity to synchronize them with ovulation. However, even though the mechanisms underpinning prolonged sperm survival are still poorly understood, it is known that they vary enormously between species. Research on domestic mammals, such as sheep, pigs, and dogs, has shown that the female reproductive tract has the capacity to identify, protect, and store a privileged subset of spermatozoa within the oviduct (Suarez 2016). Some birds are able to store spermatozoa for several weeks in the utero–vaginal junction of the oviduct, where they become sequestered within specialized sperm storage tubules (SSTs) (Birkhead & Møller 1993). Sperm survival within the SSTs is achieved by a reduction in the oxygen tension coupled with low metabolic activity and various biochemical and immunological changes (Kim & Kim 2025). Some reptiles, including snakes and turtles, also exploit SSTs in the context of prolonged sperm storage (Sever & Hamlett 2002), and some have the ability to keep the spermatozoa viable and fertile for several months and even years.

The evolution of sperm storage mechanisms in fishes has produced numerous, and sometimes surprising, outcomes. A recent study of 29 cartilaginous fish species (Jordan et al. 2021) identified the presence of stored spermatozoa within the oviductal gland of the reproductive tract. In contrast, many small viviparous fishes have evolved the ability to store spermatozoa within the ovary itself. This mode of sperm storage has recently been investigated in a detailed metabolomic and molecular study in black rockfish (Sebastes schlegelii) (Li et al. 2024), where one of the main findings was the significant upregulation of ovarian complement-related proteins and pattern recognition receptors. This was coupled with the remodelling of glycans on the surface of spermatozoa early in the storage process.

Despite the extensive biological diversity and complexity that face researchers who accept the challenges of fish conservation and reproduction, this is a vibrant and successful field of science. In fact, because small fishes are now used as model organisms in biomedical research, knowledge gained in that field is of value with respect to the conservation and biobanking of fish species (Bodenstein et al. 2023).

Many mammalian species exploit the phenomenon of embryonic diapause or delayed implantation, which is, similar to sperm storage in the female reproductive tract, a strategy to extend pregnancy and postpone birth until conditions are ideal for offspring survival (Fenelon & Renfree 2018). Embryonic diapause was first reported in roe deer in the 19th century. Since then, it has been described in more than 130 mammalian species. Species-specific physiological and cellular mechanisms are involved during this suspension of embryonic development. The morphology and the biology of the blastocysts can be classified in different categories: i) no growth in blastocysts with a zona pellucida (kangaroos), ii) no growth in blastocysts without a zona pellucida (mice and rats), iii) slow growth in blastocysts with a zona pellucida (mustelids, bears, and seals), iv) slow growth in two-layered blastocysts without a zona pellucida (roe deer), and v) implanted but undifferentiated (bats). In addition, embryonic diapause can be controlled either by lactation or by the season. Lactational embryonic diapause is a temporary, facultative pause in embryonic development triggered by the physiological stress of lactation to separate mating and birth. However, seasonal or obligate diapause is a regular, obligate pause in every pregnancy in a species, triggered by seasonal cues (such as photoperiod) to ensure birth during favourable conditions. The former is a flexible response to the maternal state, while the latter is a programmed, recurring reproductive strategy, tied to the annual cycle (Beachler 2024). There are still many questions to be answered and more species to be thoroughly studied. Research efforts have to better characterize embryology during diapause, while embryogenesis is either suspended or slowed down. The structure and function of the uterus during diapause remain to be described using new molecular tools. Similarly, embryo–uterus interactions during onset of diapause, diapause itself, and resumption of embryo development/implantation have to be deciphered. Overall, comparative explorations of new molecular players in distinct species associated with the use of cutting-edge approaches are necessary to make progress (Fenelon 2024).

Careful species identification is crucial for reproductive science and biobanking

The development of sophisticated techniques for species identification (e.g. genome sequencing, DNA barcoding, and machine learning) has resulted in the widespread realization that many related species are, in fact, genetically distinct but morphologically indistinguishable (these are known as ‘cryptic species’ (Hending 2025)). A classic example of this problem concerned pipistrelle bat species in the United Kingdom (Pipistrellus pipistrellus) (Barratt et al. 1997). It was shown that the wild population actually comprises two morphologically identical but genetically different subspecies; these use distinct echolocation frequencies, roost in separate locations, and their mitochondrial DNA sequences differ. More recently, many more cryptic species have been detected across multiple taxonomic groups and within widely different habitats. These include various mammals, as well as many fishes and amphibians, all of which are of key interest in terms of biobanking. It is therefore likely that such cryptic evolution will be problematic for the success of future biobanking activities if not recognized.

It is worth mentioning that species identification for threatened amphibians can be especially problematic. Besides being under threat from the fungal disease chytridiomycosis (Fisher et al. 2021), many amphibian species are cryptic, and therefore not easily identified, especially if they live in countries with high rates of new species discovery (e.g. Brazil, India, Peru, and Vietnam (Tapley et al. 2018)).

For many years, species identification practice has involved the collection and permanent preservation of ‘voucher’ specimens in museums. These specimens are regarded as providing the definitive example of a given species (Buckner et al. 2021). The voucher specimen permits taxonomists and other investigators to make detailed morphological observations and measurements so that other examples can be compared with the original(s) when necessary. It should be noted that collectors of voucher specimens of species such as amphibians, where larval development may result in sequential morphological changes, aim to preserve multiple examples of different larval stages to ensure highest accuracy of species descriptions (Grosjean et al. 2015). These observations are now frequently supplemented with original photographs, DNA barcodes, and important ancillary information about the origin of the specimen. Unfortunately, this ever-greater reliance on genetic databases has led fewer scientists to link their genomic data to the physical voucher specimens, something which eventually might reduce the reliability of species identification (Buckner et al. 2021).

Assisted reproduction and the development, diversity, and use of biobanks

Over the past two or three decades, the importance and conservation-relevance of establishing collections of ‘live’ germplasm from multiple species has become widely recognised. These collections, widely known as ‘biobanks’ or ‘genetic resource banks’, usually contain cryopreserved, or possibly desiccated, samples of reproductive cells and tissues from the species of interest. There is also an underlying expectation that the samples will eventually prove to be useful in some aspect of conservation biology.

Biobanks established with the explicit aim of supporting breeding programmes for rare and threatened species, whether or not via reproductive technologies, have to rely heavily on fundamental genetic, physiological and, increasingly, microbiological information if they are to be successful. Bearing in mind the wide taxonomic diversity of threatened species, and the scarcity of relevant species-specific information, it is worth emphasizing the need for locally accurate information (see Figs 1 and 2). For example, where exactly is the target species located? Are they in captivity or in the wild? Are the wild populations sufficiently well described and identified to support long-term natural breeding activities or are they expected to decline in size? Are the wild populations isolated and fragmented? Importantly, how much is known about the genetic characteristics of the population, especially heterozygosity? This is a measure of the genetic variation within the population of interest and, in many instances, can only be ascertained by sampling a representative subset of target individuals and using genetic techniques to check for differences, for example, in somatic or mitochondrial DNA sequence.

Figure 1.

Questions to ask for any wildlife samples offered to a biobank.

Figure 2.

Questions to ask when planning new sample collection in wildlife.

Importantly, the essential roles of complementary knowledge about reproductive physiology in diverse species have to be recognized, as encapsulated in the well-known phrase a ‘cheetah is not a cow’ coined by David Wildt from the Smithsonian’s National Zoo (Howard et al. 1986). One attitude among conservation scientists also was the understandable tendency to report the successful, but often isolated, births of captive wild animals without subsequent efforts to repeat the success. While these anecdotal births represented important scientific and technical breakthroughs, they did not represent the repeatability that is really needed in a technology that could make a material contribution to captive breeding, genetic management, and the long-term support of threatened species (Swanson 2023). Fortunately, it is now possible to report that this situation has recently been improved. Successful collaborations between cryobiologists and scientists from other specialities have improved the survival prospects for wild mammalian populations such as the giant panda and black-footed ferret. Similarly, considerable effort has been invested in attempts to develop sperm cryopreservation techniques for marsupials. However, although that goal has yet to be reached, intensive studies aimed at developing the essential physiological and anatomical knowledge required for undertaking successful conservation in koalas and other marsupials have resulted in an impressive body of knowledge that will be essential if the sperm preservation methods are eventually developed.

For balanced and up-to-date summaries of the role and prospects of biobanking, the reader is recommended to consult two recent, and highly realistic, critiques of the role of biobanks and advanced technologies in conservation biology (Bolton et al. 2022, Mastromonaco 2024). The role and significance of biobanks in wildlife conservation has also been discussed in a number of reviews and books published by the present authors These observations have direct relevance to the governance and routine practices of wildlife biobanking (Chen & Mastromonaco 2025), especially in the face of increased crypticity, where species identification must remain reliable and accurate.

Importance of gamete and embryo cryopreservation

The discovery 60 years ago that spermatozoa from some mammalian and avian species could not only survive being frozen and thawed but could also retain their viability and fertility was transformational (Polge et al. 1949). Animal breeding industries were suddenly able to establish banks of cryopreserved semen samples from genetically valuable livestock (mainly cattle and sheep). Furthermore, as part of this revolutionary development, the semen could be stored at low temperatures and sold and traded internationally, with the expectation that it would produce live offspring when used in association with artificial insemination (AI).

The value of using of semen cryopreservation, storage, and AI to assist captive breeding programmes for rarer species, including Speke’s gazelle, reindeer, and even budgerigars (Dott & Utsi 1973, Boever et al. 1980, Samour et al. 1988), later became more apparent and led Michael Soulé, the authoritative conservation biologist, to include ‘suspended ex situ programmes’ as one of his ‘eight paths’ to biotic survival (Soulé 1991). This category included seed banks and collections of cryopreserved germplasm, in biobanks. The concept behind the value and establishment of cryopreserved germplasm banks was also promoted in the 1970s by two pioneering Russian scientists, whose ambitious proposals included plans to establish frozen collections of genetic materials from multiple taxonomic groups, even including human gut bacteria and earthworms (see Veprintsev & Rott (1979)). At the San Diego Zoo, Kurt Benirschke created in the early 1970s the world’s first frozen zoo for the cryopreservation of genetic material from endangered species.

The principal objectives of establishing biobanks for wild animal species are well known and have been reviewed multiple times (see, for example, Saragusty et al. (2020), Holt & Comizzoli (2021), Bolton et al. (2022)). Existing biobanks include important historical collections of frozen, fixed, or dried materials that are, or have previously been, collected for many different reasons, not all of which are aimed at the conservation of species. These collections are valuable resources for biogeographic and evolutionary studies, as well as providing information about the appearance and spread of various infectious diseases (Radin 2015). Many research laboratories possess large collections of frozen animal tissues as well as associated samples, such as blood, urine, and faeces (Wandeler et al. 2007). However, in many instances, information about these collections is not widely available despite their value.

Addressing biosecurity and contamination concerns in wildlife biobanks

Wildlife biobanks face unique biosecurity and contamination challenges that require careful attention. Wildlife samples can carry pathogens that can pose risks to other animals and humans (zoonoses) (Corrales et al. 2023). Cross-contamination during collection, processing, or storage can compromise the integrity and usefulness of samples. Contaminated samples may lead to inaccurate research results or hinder future conservation efforts involving assisted reproductive technologies. In addition, improper waste disposal or accidental releases of pathogens can contaminate the surrounding environment. This can have far-reaching impacts on ecosystems, disrupting food chains and potentially affecting public health. Unauthorized access, theft, or misuse of samples poses a significant risk to the biobank integrity and potentially creates security concerns related to dual-use potential. Accidental releases of dangerous pathogens could also have catastrophic consequences. By implementing robust biosecurity measures, adhering to best practices, and fostering a culture of preparedness, wildlife biobanks can effectively manage risks and ensure the long-term viability of their collections, benefiting both wildlife and human health (Fujihara & Comizzoli 2025).

As it has been implemented in human, livestock, and laboratory model biobanks, wildlife biobanks have implemented a multi-faceted approach to minimize biosecurity and contamination risks, including i) strict access control of facilities and personnel training on biosecurity protocols, ii) enhanced sanitation and disinfection protocols of sample processing areas, iii) strict waste disposal and management, and iv) rodent and vector control.

However, there is still a need to foster more collaborations and data sharing with other biobank, researchers, and government agencies to share best practices and information on emerging threats. Likewise, it is necessary to develop contingency plans for potential biosecurity breaches, outbreaks, or environmental contamination events (Angeles & Catap 2023). Even if screening pathogenicity of samples is implemented, cross-contaminations within the same storage container remain possible. It therefore is recommended to plan for separate storage to avoid mixing samples potentially carrying pathogens (such as chytrid fungus in amphibians).

Some ethical issues that arise in the context of research in reproduction and conservation biology

The vast taxonomic diversity of threatened species means that conservation biologists have to face the reality that they will never be able to rescue and support them all. Priorities have to be set, and choices made. If researchers are especially focused on the use of exciting advanced reproductive technologies, such as nuclear transfer, intracytoplasmic sperm injection (ICSI), and CRISPR-based genome editing, whether in wild mammals, fishes, or amphibians, they will first need to conduct exploratory experiments, usually, but not always, in a laboratory.

These first steps will inevitably involve making a large number of decisions, often with independent input from external experts. For example, if anything invasive is contemplated, will anaesthesia be needed? If so, which type of anaesthetic will be most appropriate and safe? If wild species, such as fishes, are involved, it is often necessary to use some form of tagging to enable capture/recapture assessments. As these decisions require multidisciplinary expertise, responsible organisations increasingly find it useful to submit research proposals to advisory committees for advice.

As discussed above, the success of conservation interventions is likely to be unpredictable, thus rendering evidence-based judgements rather difficult. In an effort to improve this situation, the group led by Professor Bill Sutherland at the University of Cambridge has spent many years developing evidence-based approaches to conservation (see, for example, Sutherland (2022), White et al. (2022)). However, these approaches have yet to address the success or otherwise of methods involving advanced reproductive technologies.

Relevance of statistical power and ethics

Paradoxically, the difficulties associated with undertaking preliminary experiments with threatened species may be hampered simply because experimental science demands results that are based on robust, and adequately replicated, experiments. Experiments that fail to conform to this standard are therefore likely to be rejected by reviewers and journals. One important and relevant corollary of this situation is that it may sometimes be more ethically acceptable to increase, rather than reduce, the number of animals in an experiment. When animal numbers are too small to meet statistical requirements, such as sufficient ‘power’, the entire experiment may have to be disregarded as the chances of detecting ‘true’ experimental effects results in unreliable outcomes. This problem has recently been highlighted in several influential reviews (Button et al. 2013, Collins & Tabak 2014, Hankenson et al. 2024, Knudson 2025), in which some authors have even observed that many conclusions in biomedical research are probably false.

This is not always a problem, and indeed, some authors involved in amphibian sperm cryopreservation have taken positive steps to avoid the problems mentioned above. Given that the production of amphibian offspring in a laboratory setting usually results in large numbers of tadpoles, impressive improvements in the statistical analysis of well-designed studies of amphibian sperm cryopreservation methods for amphibian have recently been reported (see, for example, Poo et al. (2022), Upton et al. (2023)).

It is also relevant that researchers are more than ever being obliged to respect the principles of the 3Rs (replacement, reduction, and refinement) (Parker & Browne 2014). As discussed above, and by Lindsjo et al. (2016), applying the 3Rs to wildlife research can be often challenging. Indeed, the legalities involved in animal health and research can sometimes lead to seemingly paradoxical and unhelpful ethical regulations. For example, in the United Kingdom, opportunistic biobanking activities, such as retrieving reproductive and skin tissues for preservation from dead animals, are largely permitted without any restrictions. These animals might have died naturally within zoos and wildlife parks and might even have been valued for their outstanding genetic qualities. However, although biobank managers could be forgiven for thinking that it would have been helpful to take skin biopsies or collect semen before the animal had died, this is not currently permitted without a licence from the UK government.

It would also be helpful to collect semen and perform some simple experiments aimed at testing sperm cryopreservation protocols, but because this would be regarded as experimental research, it can only be conducted under strict conditions imposed by the UK government. In accordance with the best scientific practice, the experimental design would have to satisfy many technical conditions, but importantly, the researcher has to show that the design would involve enough individual animals to yield statistically valid results.

This regulatory framework also requires the researcher to write a very detailed protocol explaining the rationale for the study, as well as spelling out how the benefits are balanced against any likely harms. In order to obtain a licence that allows the study to proceed, the researcher must also attend several educational coursers on animal handling, ethics, and statistical analysis and must also pass relevant examinations.

Value of case studies

On a regular basis, opportunities arise to cryopreserve semen from individuals, or even small numbers of animals. Such situations happen for a variety of reasons, often during routine health checks under veterinary supervision. Individually, these opportunities can provide valuable data, especially if the results have not been obtained using consistent protocols. After some time, there may be sufficient individual data points that it would be possible to undertake simple data analyses. While these datasets were not developed as pre-planned experiments and may not always have included control treatments, their importance should not be dismissed.

The valuable data on Giant panda reproduction and breeding via AI, amassed in China, have produced valuable insights into the efficacy of different insemination techniques (Li et al. 2017, Martin-Wintle et al. 2019). The multiple insemination attempts, studied for more than 20 years, and which involved detailed urinary hormone analyses to establish optimal insemination timing (Martin-Wintle et al. 2019), could be regarded as a coherent series of individual case studies, even though they were not aiming to support or refute a specific hypothesis.

Less comprehensive datasets have also accrued over time, thus providing valuable insights into background information (such as the ageing, sexual development, population variability, and growth). Without such information, it would be difficult to plan robust and focused investigations.

The reader will realize from the description above that, given the need to perform ethically acceptable and statistically robust, experiments with rare and threatened species are complex. Consequently, it can be very difficult to undertake research aimed at using ambitious advanced reproductive technologies, such as restoring fertility through spermatogonial stem cell transplantation (Morimoto et al. 2025). While these constraints might seem to be unhelpful, it is nevertheless clear that the ethics framework illustrates the strong animal welfare culture that protects research animals (de Mori et al. 2024).

Conclusions and future perspectives

In this article, we have attempted to identify, emphasise, and understand the complexity that becomes apparent whenever conservation biologists work with wildlife species. Indeed, the amount of information about wild species that has been amassed over centuries is still usually insufficient when researchers begin to focus on solving conservation problems. Figure 3 represents an attempt to summarize and illustrate the variety of gaps (‘unknowns’) that still exist in this field, especially in the context of assisted reproduction in diverse species.

Figure 3.

List of unknowns in reproductive science and wildlife conservation.

As mentioned above in the context of translocations, relocations, and rewilding, animal conservation projects in general, despite being well meaning, can sometimes also be controversial and unpopular if they negatively affect local livelihoods. For example, while elephants are cherished and protected in Thailand, the local villagers do not necessarily appreciate the damage to crops that might also occur. Overcoming conflicts of this nature is not always easy, but positive resolutions are possible, especially if a conservation activity can result in clear benefits to the local population, and better still if the local populace is able to play a part themselves. One such project, which is particularly relevant to this article, concerns the development of coral cryopreservation methods. The success of this project, which was aimed at protecting, preserving, and restoring coral reefs (Hagedorn et al. 2019), is an excellent example of this principle that should be consider as a good model for future conservation efforts.

Declaration of interest

P Comizzoli is an Associate Editor of Reproduction & Fertility and was not involved in the review or editorial process for this paper, on which he is listed as an author.

Funding

This research did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.