The unique role of seed banking and cryobiotechnologies in plant conservation

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors, Plants, People, Planet © New Phytologist Trust 1USDA-ARS National Center for Genetic Resources Preservation, Fort Collins, CO, USA 2Center for Conservation and Research of Endangered Wildlife (CREW), Cincinnati, OH, USA


Societal Impact Statement
Eroding plant diversity has serious implications for the well-being of humanity and our planet. Conserving plants ex situ requires technologies that are rapidly advancing and readily accessible. The alarming loss of plant habitats has spawned global investment in technologies that focus on either conventional freezer storage, which exploit seed adaptations to survive drying, or on cryogenic platforms, that ensure long-term survival of germplasm that is not amenable to conventional methods. Increasing evidence that germplasm survives for decades provide proof of concept, but also warns of the limited utility of stored germplasm that is not returned to the Earth.

Summary
A future sustainable world requires concerted efforts to conserve plant biodiversity.
Using an integrated approach, botanic gardens, arboreta, universities, governmental agencies, and non-governmental organizations are addressing that challenge. Here, we summarize some of the technological advances, in an ever-growing toolbox, that increase the scope of taxa that are conserved ex situ as well as the lifespans of diverse plant tissues that can be used as germplasm. Seed banking continues to be a powerful and efficient tool. Seeds that tolerate extreme drying and low temperature will likely survive at least 100 years using conventional conditions of a common freezer.
The extreme tolerance of seeds among diverse taxa has led to the global growth of seed banks to over 1,750 currently, and the conservation of over 50,000 species. Not all plants produce seeds or seeds that survive freezer conditions. Predictive models provide insight into the extent of taxa needing alternative strategies and an initial list of such species is available. These "exceptional" species require cryobiotechnologies (cryogenic storage in liquid nitrogen and in vitro technologies), which provide effective, long-term ex situ conservation for a wide variety of tissues beyond seeds. The application of cryobiotechnologies increases the potential for conserving all plant biodiversity. Restoration of plant biodiversity into the future will require institutional collaborations among living collections, seed banks, and cryobanks to ensure technology transfer, information gathering and sharing, and capacity building in centers of biodiversity.

| INTRODUC TI ON
Ex situ conservation creates reserves of plant diversity through human involvement, cooperation, and emerging technologies.
Botanical gardens or field genebanks grow plants in their living collections to provide materials for observation, research, and education (Westwood, Cavender, Meyer, Smith, 2020). A complementary strategy stores germplasm (i.e., propagules that can regenerate a plant) in a state of "suspended animation," possibly for decades or centuries. Seed banks are a classic example of quiescent collections, initiated mostly to serve agriculture by providing genetic resources for crop breeding (Vavilov, 1987). The technologies and successes of early seed banks encouraged broader application for conservation (Chapman, Miles, & Trivedi, 2019;Falk & Holsinger, 1991;O'Donnell & Sharrock, 2017). Now, nearly 1,750 seed banks exist globally and house about 6 million accessions (i.e., samples having unique identifier information especially taxon, date, and location; Byrne et al., 2018;Hay & Probert, 2013;O'Donnell & Sharrock, 2018). An estimated 50,000-60,000 taxa are now represented in today's germplasm banks; about 15,000-20,000 taxa in agricultural genebanks (e.g., the USDA National Plant Germplasm System houses nearly 14,000 species (https:// npgsw eb.ars-grin.gov/gring lobal/ query/ summa ry.aspx) in its base collection and many more species in research collections) and about 45,000-55,000 taxa in conservation-based genebanks (O'Donnell & Sharrock, 2018). Species collections that support agriculture often have hundreds to thousands of accessions representing cultivars, genetic stocks, and wild relatives of crops. In contrast, there are usually fewer accessions per species in conservation collections, partly because collecting from the wild is more expensive, and also because accessions from wild populations are harder to manage in genebanks, and use afterwards, due to their heterogeneity, small sample size, unknown growth requirements, and additional needs for non-seed conservation methods (Pence, 2011;Walters, 2015a;Walters, Richards, & Volk, 2018).

| WHY ARE S EEDS THE PREFERRED G ERMPL A S M TO BANK?
Seeds are highly adapted to survive harsh conditions and establish when conditions improve. Survival of complete drying is a near-ubiquitous adaptation, possibly derived from the earliest land-inhabiting plants (Gaff & Oliver, 2013). Most seeds acquire desiccation tolerance as they accumulate food reserves within embryo cells (Vertucci & Farrant, 1995;Wang, Liu, Song, & Møller, 2015). Packing the cytoplasm with dry matter maintains cell structures when water is displaced (Walters, 2015b).
Drying seeds stop metabolizing and become essentially quiescent, or cryptobiotic, at ambient relative humidity (Walters, Hill, & Wheeler, 2005). Cryptobiosis in animals, such as tardigrades and brine shrimp, has captured considerable attention because the organism is alive, but doesn't show it by respiring, responding to environmental stimuli, or growing (Clegg, 2001;Wright, 2001). In plants, the tendency is widespread during sexual reproductive phases and rarer in vegetative cells (Costa et al., 2016;Dinakar & Bartels, 2013). Quiescent collections can store tremendous diversity relatively inexpensively, often using small spaces and standard procedures and achieving long shelflife (Hay & Probert, 2013;Li & Pritchard, 2009;Walters et al., 2018).
But, shelf-life is not indefinite. Quiescent organisms eventually lose the ability to recover and, in the case of seeds, cannot germinate even when required conditions are met (Fleming, Hill, & Walters, 2019;Rajjou & Debeaujon, 2008;Walters, 2015aWalters, , 2015bWalters, Hill, et al., 2005). In other words, the utility of the sample becomes limited by its achieved lifespan during storage. That survival duration (i.e., longevity) depends on storage conditions, namely, the relative humidity (RH) and temperature (Walters, 2015a;Royal Botanic Gardens Kew, 2019). Longevity of the orthodox seed storage category can be predicted by models for temperatures between 60°C and −20°C and RH between 80% and 30%, and approximately doubles for each 6°C-10°C drop in temperature (Walters, Wheeler, & Stanwood, 2004;Fleming et al., 2019;Royal Botanic Gardens Kew, 2019). Longevity of orthodox seeds during freezer storage remains mostly conjectural; the earliest experiments were started just 50-60 years ago, providing insufficient time to detect change.
Lacking empirical proof that freezer storage was effective, seed banks established before 1970, used refrigerated storage at 5°C-10°C.
In the 1980s, a few seed banks pioneered the use of freezer storage at −18°C to −20°C based on the possibility that seed lifespans would increase 4-to 16-fold, a possibility that has been confirmed for a few samples (i.e., Fleming et al., 2019;Walters et al., 2004). Freezer storage is now the international standard for "conventional" seed banking (CPC, 2019;FAO, 2014;MSBP, 2015;Figure 1a,b). We can project that properly dried orthodox seeds that survive for 25-50 years under refrigerated conditions will survive for 100-200 years in the freezer (Walters, Wheeler, & Grotenhuis, 2005). The long survival period of diverse germplasm that is protected from vulnerabilities inherent to living collections and is readily available when needed, at an affordable cost, makes this ex situ conservation strategy implementable at large and small institutions (Li & Pritchard, 2009).
Seeds are preferred germplasm because they are a complete organism, rather than a gamete, and there are many species that naturally survive drying and so naturally enter into a quiescent state.
However, pollen of many plant species also tolerates desiccation (Franchi et al., 2011) and so can be a candidate for "conventional" freezer storage. As the male gametophyte, pollen is analogous to K E Y W O R D S ex situ conservation, exceptional species, genebanks, germplasm, in vitro, longevity, preservation, seed storage behavior sperm, which is the most commonly banked germplasm for animals (Mazur, Leibo, & Seidel Jr., 2008). Indeed, pollen is increasingly valued as a germplasm form for ex situ conservation, especially when trying to restore diversity to a population of trees or when trying to cross individuals that flower asynchronously (Towill & Walters, 2000).
However, orthodox pollen is not stored using conventional methods, even though longevity exhibits similar temperature and moisture dependencies as seeds (Buitink, Leprince, Hemminga, & Hoekstra, 2000;Walters, Hill, et al., 2005). This is because orthodox pollen is inherently short lived, with many species surviving just weeks to months in the refrigerator at optimum RH (Dafni & Firmage, 2000). A 4-to 16-fold increase in longevity, achieved by placing these grains in the freezer, would only provide viable germplasm for a few years at best. This illustration of the limited utility of freezer storage of pollen underscores the time frame for survival expected from seed banking using conventional methods: seeds are expected to survive several decades. In fact, germplasm in quiescent collections should outlive counterparts stored as living and growing specimens.

| WHEN DRY G ERMPL A S M DOE S N ' T S TORE WELL
There is increasing recognition that some seeds and other germplasm forms can be intrinsically short-lived, much as is observed for pollen (previous paragraph). Examples include seeds of Salix (willow), Ulmus (elm), some orchids, and some species native to Hawaii Chau et al., 2019;Popova et al., 2013;Whigham, O'Neill, Rasmussen, Caldwell, & McCormick, 2006), as well as spores from ferns and mosses that have relatively short lifespans (Ballesteros, Hill, & Walters, 2017;Walters, Hill, et al., 2005).
Thermal behavior of TAG may explain low shelf-life in the freezer of many temperate nut species (e.g., Juglans, Carya, Castaena, and Corylus) and tropical species (e.g., Cuphea, Elais, and Hevea), as well as seeds from numerous Hawaiian species (Chau et al., 2019). We recently showed that seeds from safflower (Carthamus tinctorius) showed low temperature sensitivity after 20-30 years of storage (Fleming et al., 2019). TAG with high levels of saturated or monounsaturated fatty acids tends to be viscous and crystallizes near 0°C to −10°C. Analogous questions about stability are studied in foods, pharmaceuticals, and materials which seek to reliably predict expiration dates or onset of poor functionality (Yoshioka & Aso, 2007).
In these highly controlled synthetic systems, the formulation and the method of processing are critical in achieving a desired product that lasts for the expected time. Molecules in solids (also called glassy matrices) are immobilized by crowding, and irregular spacing allows movement over short distances. This minute motion is the root cause of degradation (Bhattacharya & Suryanarayanan, 2009), with glasses characterized as "fragile" being extremely vulnerable to slight warming or added water (Yoshioka & Aso, 2007).  (Ballesteros et al., 2018). Storage at −80°C also appears to be effective, although there is considerably less empirical or theoretical support for this storage platform. Use of LN provides an alternative F I G U R E 2 Schematic phase diagram illustrating safe (orange) and lethal (yellow) environmental conditions for germplasm storage based on interactions between moisture and temperature. Metabolism occurs in hydrated cytoplasm (green region) and quiescence is induced by solidification, when germplasm is dried and/or cooled. But tolerance of germplasm to these environmental stresses varies considerably, and response to environmental treatments must be considered to effect successful preservation. Orthodox physiology (represented by the bean icon) describes germplasm that survives longer when dried or cooled, treatments that encourage glass formation (solid blue curve). Once solidified, ice formation is not a risk and aging kinetics follow classic temperature dependencies. The papaya icon reflects germplasm with cytoplasm that doesn't form a stable solid at freezer temperatures, probably for many reasons, one being crystallization of lipids. This germplasm survives exposure to LN temperatures without added treatments; care to cool and warm moderately rapidly (within 1-4 min) has aided recovery (unpublished). The acorn icon represents germplasm, like recalcitrant seeds, that survive considerable drying but not enough to solidify cytoplasm. To safely cool to a solid (orange region), this germplasm must traverse the lethal ice zone (yellow region) within about ½ s, which is facilitated by reducing the size of the specimen to just the excised embryonic axis. Alternative treatments are needed for highly hydrated cells that are also extremely sensitive to water stress. Application of cryoprotectant solutions and rapid cooling to limit exposure time within the lethal ice zone is effective strategies ( to mechanical freezers that are subject to warming during electrical outages.  Figure 2). However, cooling of whole recalcitrant seeds, which tend to be large, within half a second is physically impossible and the ability to surgically remove the embryonic axis and recover it in vitro presented breakthrough technology (Walters et al., 2013).

| S EED REC ALCITR AN CE
This dependency of in vitro and cryotechnologies spawned the evolution of a new scientific discipline called cryobiotechnologies (Pence et al., 2020;Pritchard, 2018 (Benson, 2000).

| E XCEP TI ONAL S PECIE S AND PROS PEC TS FOR G ERMPL A S M BANK ING
Storing seeds in the freezer is not a suitable ex situ conservation strategy for many species, and we call these species "exceptional" (Pence, 2013 for the conservation of these species, and cryobiotechnologies may be essential to obtain source materials, recover thawed samples, or assess longevity potential. Species experiencing reproductive failure in the wild, in which plants do not produce pollen or seeds, require advanced cryobiotechnologies, as do agricultural varieties that must be propagated clonally (Engelmann, 2011;Pence, 2013;Pence et al., 2020). Mating When seeds are not available, the best option is to obtain cuttings of existing plants and propagate them, usually using microculture (i.e., in vitro) techniques. Most commonly, in vitro shoot cultures are initiated and propagated, and these provide tiny (1-2 mm) shoot tips, which are dissected from the shoot and cryopreserved ( Figure 1d). In addition, somatic embryos and tips from root cultures can also be utilized for storage in liquid nitrogen (Carneros, Hernández, Toribio, Díaz-Sala, & Celestino, 2017;Reed, 2008;Simão et al., 2018). Healthy explants and the right medium for growth and proliferation are required for this approach, and species vary in the ease with which such cultures are initiated and grown, either because of endophytes that can contaminate the cultures, the production of inhibiting phenolic compounds in some species that are released upon wounding, or the need of some species for specialized growth medium and conditions (Benson, 2000;Niedz & Bausher, 2002). These biotechnologies can also be used to establish in vitro, living collections that are not at risk from biotic and abiotic stresses outdoors; however, maintenance of these collections are labor intensive and many groups seek to cryopreserve their in vitro collections to gain the advantage of quiescence (Pence, 2011(Pence, , 2013. Despite the challenges, a wide range of species have been cryopreserved as in vitro tissues (Reed, 2008), and reports of longterm (1-2 decades) survival through LN storage are now being reported (Caswell & Kartha, 2009;Pence et al., 2017Pence et al., , 2020. These in vitro methods require a major investment of time and resources before cryopreservation procedures can be implemented, but they offer a viable option when other methods are not workable (Dulloo et al., 2009;Li & Pritchard, 2009;Pence, 2011).
Successful cryopreservation of explants follows the same principle as seeds and pollen: stabilize cellular structure and limit chemical change by solidifying the cytoplasm. Cytoplasmic solidification through air drying causes lethal cell shrinkage and so partial dehydration is accomplished chemically by adding osmotica and cryoprotectant molecules that remove and replace water without massive cellular deformation. In some procedures, partial air drying also helps to remove water that would otherwise form ice. Protecting molecules also delay freezing of the remaining water molecules so that cells can be cooled at achievable rates of a few hundred °C/ sec (Volk & Walters, 2006). Vitrifying agents, such as glycerol, ethylene glycol, and dimethyl sulfoxide, have the capacity to induce glassy matrices within cytoplasm without inducing major structural changes to the cell (Volk & Walters, 2006). These compounds, along with sucrose, are major constituents in PVS2 (Plant Vitrification Solution 2), one of the most effective and broadly applicable cryoprotectant solutions used for plant cryopreservation (Sakai, Hirai, & Niino, 2008).
Approaches in the cryopreservation toolbox are varied and flexible. Deciding on a method will depend primarily on the type of tissue available. Can seeds be cryopreserved whole, or is embryo excision or dormant bud storage necessary? If pollen is available, can it be stored? What in vitro methods are needed? Choosing the most effective method requiring the least input of time will allow efficient use of resources and help ensure the conservation of more species (Pence et al., 2020;Smith & Pence, 2017). While the cryogenic genebank is a more recent addition to the conservation toolbox, recognition of the need for increased global capacity is growing, and the Exceptional Plant Conservation Network was established to focus attention on species that cannot be conserved in conventional seed banks (https://cinci nnati zoo.org/epcn).

| THE " E XIT PL AN "-C A S E S TUD IE S OF THE B ENEFITS OF G ENEBANK ED PL ANT MATERIAL
Seed banks (and more generally genebanks) for plants are a recognized strategy to conserve genetic diversity, providing insights about Earth's biodiversity, humans' need for biodiversity, and the consequences for our planet if we do not protect biodiversity. The concepts presented in this review demonstrate that it is possible to achieve a quiescent state (i.e., preserve) in diverse plant germplasm; however, germplasm will not survive indefinitely. To ensure a return on investment of genebanking, the material must eventually be removed from the genebank and used. Those responsible for genebanking may not be the same as those using the materials; successful genebanking requires partnerships among different groups that have different missions. There are a number of works that have been published over the years to describe these partnerships as well as best practices to bank genetic resources that are "fit for purpose" (Center for Plant Conservation, 2019;Commander et al., 2018;Maschinski & Haskins, 2012;Offord & Meagher, 2009;Pritchard & Probert, 2003;Walters et al., 2018).  "time capsules" to evaluate temporal changes in an approach referred to as "resurrection ecology" (Etterson, Franks, et al., 2016;Etterson, Schneider, Soper Gorden, & Weber, 2016;Franks et al., 2008;McGraw, Vavrek, & Bennington, 1991). The Project Baseline collection (basel inese edbank.org) includes numerous populations per species, making it possible to examine interactions among selection stressors, population history, and genetic composition to influence the rate and extent of evolutionary response across species' ranges.
Over the next 50 years, this unique seed collection will provide material to conduct resurrection experiments that will illuminate how anthropogenic and natural disturbance, including climate change, is driving evolution in wild plant species across both time and space.
The Notably, these conservation methods will be at the heart of an historic, large-scale genebanking project funded by the State of California to collect seed from over 700 currently unbanked rare plant taxa for storage at botanical gardens in the California Plant Rescue collaborative.

| CON CLUS ION
The rapidly developing technologies that allow plant germplasm banking under conventional or cryogenic platforms broaden the possibilities and promise that humans can safeguard the Earth's trove of plant diversity. But saving these invaluable resources in freezers and cryovats isn't enough. Stored germplasm that is forgotten or neglected will eventually die and will not serve a conservation goal. Therefore, an integrated approach that combines preservation technologies with botanical expertise and restoration ecology, as well as organizations from public and private sectors, is essential to protect plants and ensure they keep a home in the world that relies on plants for its own existence.

D I SCL A I M ER
Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the US Department of Agriculture. USDA is an equal opportunity provider and employer.