Making Plants, Part 1: Sow What?

By David Harry, Ph.D.

Many of us take for granted the origins of common items we use or consume on a daily basis. Consider food, clothing, and paper products. Many of these items are derived directly from plants, or from animals which in turn depend on plant-based feeds. Clearly, we owe a great deal to plants and agriculture. But what is the origin of these plants we depend on? LettuceWhile many plants are reproduced from seeds, a thorough consideration of this topic is somewhat more complicated. (Or why else would I be writing about it?) As a biologist working to commercialize pongamia trees as an oilseed crop, I expend considerable thought evaluating alternative approaches for propagating and distributing plant materials. Seeds are but one of several options. Furthermore, there is a renewed interest in broader philosophical and socio-political questions around the marketing and patenting of seeds and other types of germplasm (cuttings, tubers, bulbs, plantlets, etc.) in commercial agriculture. Questioning current practices is often an effective way to increase overall awareness, but alternative approaches that superficially appear simple may be impractical or otherwise more complicated than they might seem.

This is the first of two blog postings in which I take a closer look at how crop plants are propagated, disseminated, and ultimately grown. In particular, I’ll compare and contrast the use of sexually produced seeds vs. asexual propagules such as bulbs, cuttings or grafts. How do such approaches differ, and what are their relative advantages and disadvantages? I’ll focus on seeds in this installment, and take up asexual methods next time.

Plants, Seeds, Or?
Many factors can influence choosing one type of propagule over another (seed, starter plant, etc.). Certainly one important factor is operational scale. You likely know someone (perhaps yourself) who enjoys puttering with landscaping in their yard, or perhaps someone whose goal is to produce the perfect juicy tomato. Smaller-scale farmers may take pride in producing a substantive proportion of their own food. Other smaller operators may sell produce in local farmers markets, or perhaps open their farm gates to the public by offering U-pick sweet-corn, berries, or miscellaneous fruits. Larger farming operations might specialize by growing a few crops for sale to co-ops, food processors, or elsewhere. A common thread among all these operations, regardless of their scale, is the need to cultivate plants. Where do all these plants come from?

Larger farms growing annual crops might work exclusively using seeds. Many crops such as corn, beans, soybeans, sorghum, and others, are readily grown from sown seeds BeanSeeds(e.g. drilled, broadcast, or burrowed into rows). Other crops such as lettuce or cucurbits (melons, cucumbers, pumpkins etc.) might be started in flats and later transplanted as individual starts. On the other hand, smaller operations such as hobbyists and backyard gardeners may chose convenience over price and purchase starter plants from a local nursery. Specifics are likely to vary from farm to farm depending on goals, local climate, facilities, and labor.

Most crop plants can be loosely classified as annual, biennial, or perennial, corresponding to plants completing their life cycle in one, two, or multiple years. (Flowering bulbs, like tulips, may seem to blur these boundaries. Because the bulbs are long-lived, such plants are considered perennials.) Many annual and biennial plants are grown from seeds, which for many species are nature’s way of initiating the next generation of plants. Propagation from seeds can be an ideal and economically efficient mechanism in the right circumstances, with radishes, beans, corn and grains serving as good examples.

Seedlings of some plants, like tomatoes, are initially grown in a protected environment before being transplanted into the field. Such extra efforts are often taken to overcome environmental limitations or length of growing season. Tomato seeds and young plants prefer warmer conditions, with transplanted starters providing a head-start for warmer weather later in the season. For hobbyists, it may simply make sense to pay a bit more for a starter plant rather than take on the extra effort of germinating seeds. On the other hand, larger-scale operators may germinate seeds in trays and produce their own starter plants. Plant productivity is another consideration. Healthy tomato plants can be highly productive, so backyard hobbyists may require only few plants to satiate their craving for fresh tomatoes. On the other hand, root plants such as radishes produce only one (typically small) edible unit per plant. I don’t believe I’ve ever encountered starter plants for radishes!

Historical Seed Production
Seeds are the most common vehicle for reproducing annual and biennial crop plants. (Potatoes, vegetatively propagated from tubers, are an exception in that they are grown as annuals, but are in fact perennial). Most seeds are purchased from seed suppliers, and yet all seeds are not equal. Over recent decades the pathway from seed provider to farmer has grown more complex.

Until the first half of the twentieth century, seeds for subsequent crops were typically collected by setting aside selected fruits, allowing them to fully ripen, and finally harvesting and processing mature seeds. Some farmers produced their own seeds, while others purchased or exchanged seeds with neighbors, or perhaps purchased seeds from smaller companies. Larger seed companies, at least as we know them today, did not yet exist. Farmers recognized and utilized different locally adapted strains (i.e. landraces), although they did not understand how the landraces had developed.

The science of plant breeding was in its infancy (see Kingsbury, 2009, Hybrid: The History & Science of Plant Breeding). By the1800’s, a small number of pioneering breeders and scientists were experimenting with interbreeding (crossing) plants by transferring pollen from one plant onto flowers of another. Some workers began stretching the boundaries of interbreeding by experimenting with hybridization, initially using “wide-crosses” between different races or in some cases between individuals of different species. (Luther Burbank, 1849-1926, is among the better known pioneers of hybridization).

The science of genetics, describing the inheritance of genes from parents to offspring, was born in the early 1900’s and quickly lead to the development and early commercialization of hybrid corn, beginning in the 1920s. The rate of adoption of hybrid corn EGE-USDA-Corn-Hybrid-2was astounding, growing from less than 10% of Iowa corn in 1935 to over 90% in 1939 (Crow, 1998). While enormously successful, the development of hybrid varieties of an outbreeding (or cross pollinating) crop like corn first requires creating inbred lines—a major undertaking of time and resources. Without going into specifics, hybrid varieties offer several advantages over non-hybrids: increased uniformity and year-to-year repeatability. But these advantages come at a cost to farmers—hybrids do not breed true, meaning that seeds from hybrid plants grow into offspring that are highly variable and therefore commercially undesirable. In short, each crop of hybrid plants must be replenished by sowing a freshly created generation of hybrid seeds. As farmers fully recognize, seed-saving from hybrid crops is not practical and farmers must purchase new hybrid seeds each year.

Modern Seed Production
Plant breeding, the development of new plant varieties, and the commercial distribution of seeds have changed substantially since hybrid corn was first commercialized in the 1920s. The success of hybrid corn prompted the development of hybrid varieties in many other crops. As technologies evolved, structural changes within the plant breeding industry also took place. Today’s breeding industry has undergone substantial consolidation as smaller seed companies were acquired by larger players such as Monsanto, DuPont Pioneer, and Syngenta. Only some of these changes were driven by the implementation of genetic engineering (GMO) technology, since to-date relatively few GMO crops have been commercialized (e.g. corn, soybeans, cotton, alfalfa, sugar beets and papaya). Large seed companies command a substantial market share among large farming operations, but there nevertheless remains a smaller (and growing) market for seeds targeted to smaller farmers and home gardeners. For example, along with growth of farmers markets and a renewed interest in locally-produced food, there has been a resurgence of interest in open-pollinated and heirloom vegetables, particularly among people favoring more traditional approaches to agriculture.

Not all seeds are equal. The following list illustrates the types of descriptions someone might encounter while flipping through their favorite seed catalog.

  • Hybrid (or F1): a variety produced by intercrossing two different parental varieties. These are often inbred lines of the same species, but in some instances could be from different species. Hybrids are often more vigorous and uniform than their non-hybrid counterparts. Hybrid varieties must be produced afresh each generation because hybrid parents do not “breed true,” their offspring are highly variable.
  • Open pollinated: This designation has no strict definition except that it encompasses plants that were pollinated naturally, without any assistance by people. Wind, insects, bees, hummingbirds, etc., are all natural pollinators. Plants that are open pollinated often include among their offspring a mix of individuals from self-pollinations (both male and female gametes originated from the seed parent) and others from outcrosses (female gamete contributed by the seed parent whereas the pollen originated from a second individual). Because of the genetic mixing that can occur within open pollinated varieties, they tend to exhibit less uniformity than other types of varieties. Furthermore, in order for open pollinated varieties to reproduce faithfully (remaining true-to-type), they should be isolated from related varieties to minimize the movement of stray pollen. Further complicating the make-up of this group, naturally selfing plants such as beans and peas are often included (see below).
  • Selfing variety: Selfing varieties include plants whose offspring are usually produced by self-pollinations, in which male gametes from the seed parent effect fertilization. Often, such pollinations occur within the same flower, sometimes before it even opens. Selfing varieties tend to “breed true,” meaning their offspring share many if not all attributes of the parental generation, even when planted near other selfing varieties. Pollen movement is generally of little concern. For a hobbyist gardener wishing to produce their own seeds for subsequent crops, selfing plants are ideal. Selfing varieties are typically not labeled as such (and are often unlabeled), but may be grouped in the broad category of open pollinated varieties.
  • Heirloom variety: These tend to be older varieties with historical significance or reputations. They are not hybrids, and instead are reproduced as open pollinated or selfed varieties. Over time, heirloom varieties from various producers may gradually drift apart, and in some instances, may receive a new designation (e.g. brandywine red vs. brandywine pink).

Concluding Remarks
Not all seeds are the same, and some types of seeds are better suited for certain purposes than others. A hobbyist gardener may give these differences little if any thought if their goal is simply to grow some fresh vegetables in the summer months. Nevertheless seed producers, represented as larger commercial producers, heirloom producers, or backyard seed savers, should all be aware of the differences and manage their seed crops accordingly. Differences in reproductive biology among plants and plant varieties can have major impacts on how seedling-derived offspring grow and yield. Perennial plants are subject to similar constraints and concerns, but their longer life-span (and often larger size) make it feasible to consider alternative propagation strategies. This will be the subject of my next installment. Stay tuned!

Genetic Predisposition and Weather Forecasts: Common Ground?

Author:  David Harry

Image

      It was nice to take a few days off during the holidays—a chance to catch up on some reading and head outdoors.  

     Among my readings, an intriguing article about genetic testing was published by Kira Peikoff in the New York Times (http://www.nytimes.com/2013/12/31/science/i-had-my-dna-picture-taken-with-varying-results.html).  Ms. Peikoff’s article described her personal experience after submitting her own DNA for testing by three companies.   The article seemed particularly timely since one of the companies had recently received a stern warning from the US FDA 

regarding its direct-to-consumer testing program (http://www.nytimes.com/2013/11/26/business/fda-demands-a-halt-to-a-dna-test-kits-marketing.html).  At first glance, Ms. Peikoff’s results appeared surprising Imagebecause they suggested conflicting interpretations regarding her genetic predisposition for certain afflictions.    

      While I found all this interesting, I was on holiday, the day was still young, and I had a chance to get outside.  I called up my favorite smartphone weather app, which provides access to forecasts wherever I happen to travel.  Lo and behold, two sources provided strikingly different forecasts for the day!  One called for partly sunny skies and little chance of rain; the other called for clouds and an 80% chance of rain.  What to do?  I grabbed my weather gear and headed outside. 

      Luckily for me, the weather that day turned out to be very pleasant.  My mind was free to wander, yet my curiosity had been piqued.  We readily accept discrepancies in weather forecasts as they happen all the time.  Why, then, should we be so surprised that genetic test results from different providers would yield different interpretations of genetic risks?  To elucidate this, let’s take a step back and think about what forecasts entail. 

Probability and Risk

      In the case of weather, meteorologists have amassed an impressive arsenal of instrumentation, computational power, atmospheric mechanics, and historical observations which are all combined using sophisticated probabilistic forecasting models.  Local forecasts reflect a greatly simplified interpretation of these complicated models, providing the likelihood of alternative weather scenarios (e.g. chance of rain or whatever) in the hours or days ahead.  In my experience, weather forecasts have improved substantially in recent decades, but even so, I tend to give more credence to short-term (next few days) than I do to long-range forecasts.  Why?  Different forecasting models predict different outcomes, and differences among the models are more pronounced for longer-range forecasts.  Forgive the cliché, but it doesn’t take a rocket scientist to figure this out.  Simplistically, weather forecasts help us evaluate our individual circumstances so that we may balance risks and plan activities in the future. 

     At one level, weather forecasts and predictive genetic testing share common themes.  Both provide insights into the likelihood of future events based on appropriate and available predictive models.  Yet whereas meteorologists have benefitted from decades of modern climatological observations and experience (not to mention hundreds of years of anecdotal history), the science of predictive personal genomics is still in its infancy.  The first draft sequence of the human genome became available only in 2000, and most gene-based risk assessments have been developed since then.  Moreover, gene-based predictions typically evaluate different genetic markers (e.g. DNA signposts), a fundamental problem which has been likened to determining a novel’s plot after reading  random letters drawn from a small percentage of its words.  Different readers, with access to different letters and words, are likely to derive different interpretations.  Another fundamental difference is timeframe.  As discussed, weather forecasts for the immediate future are often given more credence than are long-range forecasts.  Predictive genetic tests attempt to assess the likelihood of outcomes that, should they occur, may not become obvious for many years.  Even for those relatively few genes for which strong disease relationships have been established (e.g. BRCA genes with cancer risk), many other non-genetic factors modulate the likelihood of disease.  The vast majority of tested genes are at best only loosely tied to a genetic disease, and for these, the intricate complexities of “nature vs. nurture” are even greater.  Does this imply that genetic diagnostics are meaningless?  Perhaps not if the results are interpreted in the proper context.  But how can any one individual interpret their personal results? 

 Statistical Forecasts and Individual Consequences

     Both weather models and genetic models predict alternative outcomes with varying degrees of likelihood.  But if weather models can (and often do) produce incongruous forecasts, why not genetic models?  Forecasts attempt to provide information to evaluate possibilities—they do not claim to tell us what will happen.  Over the years, we have realized that weather forecasts are often helpful, and yet they are far from perfect.  Genetic testing, especially of the sort provided by personal genomic service companies (as in Ms. Peikoff’s example), has a long way to go to improve reliability and consistency. 

     To evaluate the overall merits of genetic testing, two additional issues come to mind.  The first concerns methods to evaluate the accuracy or precision of the estimated likelihoods.  Political pollsters, for example, routinely evaluate the precision of their estimates.  A polling result of, e.g. 46% +/-2% conveys a sense that the actual likelihood is somewhere between 44-48%, whereas a result of, say, 46% +/- 25% conveys a sense that the results are unreliable.  Might such estimates of precision become available for predictive genetic tests?  Time may tell.             

     Another factor, which in my opinion is both more subtle and even more important, concerns the meaning of probabilistic interpretations as applied to populations vs. individuals.  Statistical models (i.e. of genetic predispositions) may be meaningful when applied to larger samples, but individuals want to know what such results mean for themselves.  Even though a given genetic predisposition may indicate a low overall risk of disease (at a population level), such likelihoods are meaningless to those individuals who do end up suffering the affliction.  For people, evaluating genetic predispositions must also keep individual consequences clearly in mind. 

Application to Breeding

Image     Time and again, practices in agricultural genetics and breeding are informed by advances in human genetics. As with previous blog posts, I often conclude by evaluating my musings in light of agricultural applications.  Relative to human medical genetics, an obvious advantage in breeding is that, at least for the most part, we need not overly concern ourselves with individual plants or animals.  We are certainly interested in changing (i.e. “improving”) the overall average performance of a population (as a collection of individuals), but the individuals themselves are typically dispensable.  Hence, provided genetic models are “reasonably” accurate, they can be used to our advantage.  Because predictive models deal with probabilities, certain individuals with above-average genetic merit will sometimes be discarded, and likewise, others with below-average genetic merit will be retained.  The basic principles of building and refining genetic models are similar for both medical geneticists and agricultural geneticists, yet the take-home message is that utilizing genetic testing is more straightforward for agricultural geneticists than it is for medical geneticists.  Yes agricultural geneticists have much to learn (and gain) by applying genetic predictions, and are lucky in that ethical considerations are not nearly as consequential as in  disciplines such as medical genetics.   

 

David Harry, Ph.D., is TerViva’s R&D Director

Genetic Testing: A Common Thread in Breast Cancer and Agriculture

TIME Magazine, May 27, 2013

TIME Magazine, May 27, 2013

By David Harry, TerViva

Angelina Jolie’s personal decision to undergo a preventative double mastectomy became a very public discussion topic after she published an op-ed piece in the May 14, 2013, New York Times (http://www.nytimes.com/2013/05/14/opinion/my-medical-choice.html?_r=0).  TIME magazine (May 27, 2013) followed suit by featuring the story on its cover.  Ms. Jolie’s story is valuable because it provides an instructive example to help others begin to understand how genetic testing can help assess risk.  But it also provides a learning opportunity for better understanding how genetic testing can be applied in agriculture.

Ms. Jolie sought out genetic testing because her family history presented strong evidence of an inherited predisposition for hereditary breast and ovarian cancer (HBOC, http://www.cancer.net/cancer-types/hereditary-breast-and-ovarian-cancer):  her mother died of breast cancer at age 56; her maternal grandmother died of ovarian cancer at age 45; and within weeks after Ms. Jolie’s announcement, her maternal aunt died of breast cancer at age 61 (http://en.wikipedia.org/wiki/Angelina_Jolie#Cancer_prevention_treatment).   Ms. Jolie’s genetic test revealed that she carries the defective (cancer causing) version of the BRCA1 gene associated with HBOC.  Ms. Jolie subsequently opted to undergo a double mastectomy dramatically reducing her overall risk of developing breast cancer.  The overwhelming public reaction has been an outpouring of support, coupled with praise for Ms. Jolie’s decision to go public in order to help others.

Extrapolating from medical genetics to other applications such as agriculture requires some explanation.  First, it’s important to realize that the same rules of inheritance apply equally to plants, animals, and humans.  Likewise, the process of interpreting these rules, and coupling family history with genetics, follows incredibly similar rationales and offer similar predictive opportunities in both medicine and in agriculture.

Ms. Jolie’s journey illustrates how, in light of her family history and genetic testing results, her medical advisors estimated she had an 87% chance of developing breast cancer.  Such predictions are not always as straight forward. Over the past two decades, the roles of certain genes in breast cancer have been increasingly understood.  BRCA1 and BRCA2 play particularly significant roles in cancer because they affect DNA repair, but the involvement of other genes (http://www.cancer.gov/cancertopics/factsheet/Risk/BRCA) means that completing a battery of  genetic tests does not ensure accurately predicting the likelihood of developing cancer.  Genetic counseling is typically recommended since a large number of factors must be considered, including the interaction of an individual’s personal history in conjunction with complex environmental influences.  Needless to say, amassing the resources to evaluate overall risk is no trivial matter, and deciding how to act on this information involves balancing many factors.

All medical decisions, with or without genetic testing, are made from the perspective of an individual.  In light of all pertinent medical information, and after balancing all other factors, what course of action is best for an individual patient?   Each decision, whatever its outcome, is highly individual, being made by a patient in consultation with his/her medical advisor and family.

In contrast to human medicine, agriculture decisions are typically neither individual nor personal.  In agriculture, decisions typically involve evaluating alternatives affecting groups of plants or animals (e.g. populations, herds, fields, orchards, etc) to select the most economical means to achieve a given commercial or environmental outcome.   For example, the typical mission of breeders and geneticists is to somehow shape the genetic trajectory of a population so that it best conforms to a targeted goal—commercial or otherwise.  Genetic predictions can be influenced by family history (e.g. field performance of siblings), and increasingly, augmented using results from DNA-based genetic testing.

Two young pongamia trees with contrasting phenotypes.  Which might perform better in the long-run?

Two young pongamia trees with contrasting phenotypes. Which might perform better in the long-run?

Genetic screening, using methods akin to human medical genetics, is being widely used in diverse agricultural applications.  Some of these involve assessing the role of one or a few genes, combining perhaps dozens of markers.  Other applications are being developed that simultaneously evaluate hundreds to thousands of DNA-based differences, and then combine this information with massive datasets on plant or animal performance.

For pongamia, we are constantly on the look-out for cost effective ways to evaluate which trees to select, propagate, and distribute, using genetic markers as one of many inter-related approaches.  Genetic testing will rarely supply unambiguous predictions, so our goal is to stack the odds (i.e. like a winning hand in a card game) to provide the most likely collection of pongamia varieties for a given set of circumstances.

David Harry, Ph.D., is Director of Research and Development for TerViva.  His background encompasses research and management positions in the public, private, and academic sectors, working primarily to integrate novel genetic applications with applied breeding in plants and animals.  David has a B.S. and M.S. in forestry, and a Ph.D. in Genetics from UC Berkeley.