Bulletin Number 5. July 31th, 2011
Dr. Peter C. Romaine J.B. Swayne Chair and Professor Department of Plant Pathology The Pennsylvania State University University Park, PA 16802 USABirth of the recombinant DNA era.
In the early 1970's, scientists first succeeded at splicing viral and bacterial DNAs in the test tube, heralding the birth of the recombinant DNA era, commonly referred to as genetic engineering, gene splicing, and transgenics. This new biotechnology found immediate application in the production of pharmaceuticals, where synthesis of protein-based drugs by genetically modified (GM) microbes provided a quantum leap in efficiency over the laborious extraction of often miniscule amounts from animal tissues..
It was during the 1980's when the potential of the burgeoning discipline of genetic engineering was first brought to bear on the improvement of agricultural productivity. The discovery of techniques to transfer genes to the major agronomic crops from unrelated species provided breeders with new vistas for increasing the efficiency of food crop production. Today, GM crop plants have been developed, and in some examples commercialized, for increased resistance to insects, pathogens, and herbicides, improved tolerance to drought and cold, and enhanced nutritional qualities.
Enter the common cultivated mushroom.
For almost as long as scientists have been introducing genes into crop plants using transgenic breeding, others have attempted with limited success at developing a gene transfer method for the common cultivated mushroom, Agaricus bisporus. In 2000, a post-doctoral fellow working in my laboratory, Xi Chen, devised an easy and effective Agrobacterium-mediated "fruiting body" gene transfer method holding the promise of a powerful tool for the genetic improvement of A. bisporus (Chen et al. 2000). Xi discovered that the lamellar tissue in the fruiting body was highly receptive to DNA transfer mediated by the bacterium, Agrobacterium tumefaciens.
Agrobacterium tumefaciens is a common soil inhabitant with a worldwide distribution. It causes crown gall disease on hundreds of woody and herbaceous plant species, but most commonly pome and stone fruits, brambles, and grapes. In its normal life cycle, the bacterium transfers a tiny piece of its plasmid DNA into the plant DNA, resulting in the formation of galls. These galls serve as specialized factories for the mass production of the bacterium. Over the years, scientists developed disarmed strains that were incapable of inducing galls, but retained the ability to transfer DNA. Thus, a natural biological process was harnessed to create a bacterial delivery system for moving genes into plants, as well as fungi (De Groot et al. 1998).
In the experiments carried out by Xi Chen, a small ring of DNA carrying a gene for resistance to the antibiotic, hygromycin B, was transferred to a disarmed strain of Agrobacterium. The antibiotic resistance gene is referred to as a selectable marker, because mushroom cells receiving this gene from the bacterium become marked by the resistance trait and can be selected based on the ability to grow on an otherwise lethal hygromycin B-amended medium. The end result was a mushroom strain having the newly acquired trait of hygromycin B resistance. Such a strain had absolutely no commercial value, but rather the resistance trait was simply a research tool that allowed us to easily determine if the bacterium had transferred the gene to A. bisporus, and exactly how efficiently it did so under different experimental conditions.
Our "fruiting body" gene transfer method entailed excising lamellar tissue from a fruiting body approaching maturity, but with the veil intact, so as to ensure some degree of sterility. The tissue was then sectioned into small pieces and vacuum-infiltrated with a suspension of Agrobacterium carrying the hygromycin B resistance gene. The lamellar explants and bacterium were co-cultivated on a defined growth medium for 3 to 4 days, during which time the bacterium transferred the antibiotic resistance gene to the mushroom genome. Because not all lamellar cells receive a copy of the resistance gene, those that have can be distinguished from those that have not by the ability to proliferate on the antibiotic-amended medium. After 7 days of incubation, mycelium of A. bisporus appeared growing at the edges of some of the lamellar explants, and after 21 days upwards of 95% of the tissue pieces had regenerated into discernible cultures. At this point, the GM cultures could be transferred to a standard nutrient medium and used to prepare grain spawn in the conventional manner for the production of fruiting bodies.
The results of cropping trials carried out during the early 2000's at the Penn State Mushroom Research Center demonstrated that hygromycin B-resistant GM lines of A. bisporus mirrored their parental commercial hybrid off-white strain in colonizing the growth substrate. Moreover, GM lines produced fruiting bodies having a normal morphology and some yielded on a par with the parental commercial strain. Stable inheritance of the antibiotic resistance gene in the fruiting bodies was easily confirmed by the re-growth of pileus and stipe tissue explants on a hygromycin B-amended medium. The results of these trials were critical, albeit somewhat predicted, because they established for the first time that a foreign gene could be introduced into A. bisporus without having a detrimental effect on its vegetative and reproductive characteristics.
3. Following the path of lesser resistance
Agaricus bisporus, is one of the most intensively cultivated and extensively researched edible mushroom species in the world (Chang 2005). It is popularly known as the white button and brown portabella and crimini. Despite more than one-half century of commercial cultivation and scientific investigation, advances in the genetic enhancement of this crop species has been impeded by its recalcitrant genetics (Summerbell et al. 1989; Van Griensven 1991). Virtually all white strains of A. bisporus commercially cultivated throughout the U.S., Canada, Mexico, and Eastern and Western Europe today represent largely clonal derivatives of the Horst 'U1' and 'U3' hybrid off-white/white strains that were introduced to the industry in the 1980's. These two strains were the products of an ambitious breeding effort led by Dr. Gerda Fritche, Horst Mushroom Experimental Station, The Netherlands (Fritche 1983).
Dr. Fritche intercrossed smooth-white and off-white strains to develop hybrids that combined the favorable high-yield and mechanical harvesting characteristics of the off-white strain with the smooth round dense cap and bruising resistance of the white strain. The 'U1' and 'U3' strains had completely dominated the market for the white button variety within 10 years of their release, because of their compellingly superior agronomic traits. Ironically, the overwhelming success of these strains created a near global mushroom monoculture that remains precarious from the standpoint of pathogen and pest susceptibility, while limiting the range of cultural characteristics available to the grower. A strong movement afoot, however, now seeks to broaden the genetic diversity of commercial strains by utilizing a largely untapped wild germ plasm collection procured by Dr. Richard Kerrigan, Sylvan Inc., under the Agaricus Recovery Program (ARP) (Kerrigan 2004). A recent success story is the strong acceptance by the U.S. mushroom industry of the high-yielding Amycel 'Heirloom' portabella strain that was bred by a hybridization approach using the ARP wild germ plasm (Robles and Lodder 2010).
The advent of a facile genetic transformation method for A. bisporus offered an alternative approach to traditional breeding for broadening the genetic base of commercial mushroom strains. The prospect of achieving for the mushroom what had been accomplished for crop plants, particularly resistance to fly pests and viral pathogens, was now technically feasible. However, my fervor for transgenically breeding the mushroom was quenched by the cool disinterest of the commercial mushroom breeders. It quickly became apparent that if the transgenic methodology was to be used for a practical application, then our research would have to take on a new direction.
In 2005, we refocused our research effort on the use of A. bisporus as a factory for the manufacture of commercially valuable proteins, such as biopharmaceuticals (protein-based drugs) and industrial enzymes. A post-doctoral fellow in my laboratory at the time, Carl Schlagnhaufer, succeeded in expressing the first biopharmaceutical in the mushroom, a commercial protein with a market value of US$600 million. Growing GM mushrooms as a factory for protein-based drugs circumvented the consumer acceptance issues surrounding GM food. Further, inherent characteristics of A. bisporus and its cultivation scheme were amenable to the commercial-scale manufacture of biopharmaceuticals. For example, the highly efficient Agrobacterium-mediated gene transfer technique facilitated the large-scale generation of GM cultures. Using a mature technology from the food industry, a starter culture could be rapidly expanded by vegetative propagation for the production of a large quantity of GM fruiting body biomass. A typical 32-day cropping cycle yielded 30 kg of biomass/m2 of growing area at a cost of less than US$2.00/kg (USDA 2010). Finally, mushroom cultivation was a readily scalable process that could be carried out under containment in an HVAC-controlled, HEPA-filtered, enclosed facility.
Accelerated manufacture of biopharmaceuticals
In 2006, I co-founded Agarigen Inc., a Penn State spin-out company dedicated to harnessing A. bisporus as a workhorse for the mass-manufacture of commercialized proteins. In the following year, the company was awarded a multi-year research contract from the Department of Defense -- Advanced Research Projects Agency (DARPA) under the Accelerated Manufacture of Pharmaceuticals (AMP) program. The mushroom was one of five organisms-- tobacco, Neurospora crassa, Pseudomonas bacterium, and shrimp-- selected for the development of a high-capacity high-speed, low-cost manufacturing system for protein-based drugs. A team of 12 Agarigen scientists quickly found itself immersed in a fast-paced, intensive, and expansive research effort, exploring and defining molecular tactics for expressing recombinant genes in the high-biomass fruiting body of A. bisporus. Using a high-throughput scheme, thousands of GM lines were generated and screened for the production of recombinant proteins. More than 40 genes encoding a broad diversity of prokaryotic and eukaryotic proteins with multimeric structure, disulfide bonding, and glycosylation were successfully expressed in the mushroom. Several mushroom-made recombinant proteins were extracted, purified, and shown to be fully functional when compared to their native counterparts. Intrexon Corporation, which acquired Agarigen in early 2011 (Intrexon 2011), is now gearing up for the expanded production of a protein for one its first clients, with several other recombinant proteins in development.
The need for speed - elucidating the contribution of the spawn and CI in the organogenesis of the fruiting body
Commercial cultivation of A. bisporus is carried out using a bi-layered substrate consisting of a lower layer of compost and upper layer of neutralized peat ("casing"). The compost is seeded with either sterilized rye or millet grain colonized by A. bisporus ("spawn"). After two weeks, when the substrate has become completely colonized by A. bisporus, the casing is overlaid on the compost bed. A common practice in the North American mushroom industry is to seed the casing with a second inoculant consisting of a nutrient-fortified particulate matrix colonized by A. bisporus ("CI"). This practice results in the rapid and uniform development of the fruiting bodies. Mycelia colonizing the casing and compost anastomose at the interface of the two layers, creating a singular fungal network throughout the cultivation medium. Fruiting bodies first appear at about 17 days after application of the casing layer, and continue to develop at weekly intervals during a three-week harvest period.
Under the DARPA AMP program, we were tasked with devising a 12-week manufacturing cycle, starting with the input of the gene encoding a protein and ending with the output of the protein in purified form. One potential timesaving solution was to use a dual-inoculant strategy in which the casing layer was inoculated with a GM CI and overlaid on compost that had been pre-colonized by a commercial non-GM spawn. In effect, this would reduce the manufacturing timeline by the two-week compost colonization period. We conducted these studies using GM lines expressing a bacterial beta-glucuronidase (GUS) reporter protein. Fruiting body tissues expressing GUS developed a blue-green color when incubated for several hours in a colorless substrate of the enzyme (Fig. 1). Comparing the level of GUS enzyme activity in fruiting bodies grown from both a GM GUS spawn and CI with those grown from a non-GM spawn and GM GUS CI would provide a measure of the efficacy of the dual-inoculant strategy for reducing the manufacturing timeline. Moreover, the results of these studies would shed light on the relative contribution of the spawn-derived mycelium and CI-derived mycelium in the formation of the fruiting body. Mushroom growers were of the opinion that the CI was the "brain center" controlling mushroom development, although exactly to what extent was strictly conjecture.
Fig. 1 Non-genetically modified (Non-GM) fruiting body (left) and genetically
The findings of a three-year investigation proved the dual-inoculant strategy to be a viable option for shortening the timeline for mushroom-based manufacture of recombinant proteins. More importantly, emanating from these studies were two surprising discoveries. First, the genotype of the fruiting body was determined completely by the genotype of the A. bisporus strain used as the CI in the upper casing layer. And second, GUS protein that was synthesized in GM mycelium colonizing the lower compost layer was transported up and into a non-GM fruiting body developing from a non-GM mycelium in the casing layer. To our knowledge, long-distance translocation of protein in filamentous fungi had not been described.
Figure 2 summarizes the genotype and phenotype of fruiting bodies grown from different combinations of non-GM and GM GUS inoculants for the cultivation substrates. Fruiting bodies developing from both non-GM spawn and CI tested negative by PCR analysis for the GUS gene and negative for GUS enzyme activity. Fruiting bodies produced with both GM GUS spawn and CI tested positive for both the GUS gene and enzyme activity. Unexpectedly, fruiting bodies grown from a GM GUS spawn and non-GM CI expressed high-level GUS enzyme activity, but did not contain the GUS gene. These fruiting bodies also lacked the GUS transcript, as determined by RT-PCR analysis, indicating it was the GUS protein that was translocated from the mycelium colonizing the compost to the developing fruiting body. We interpreted our findings to suggest that the recruitment of functional protein from the vegetative mycelial network in the compost offers A. bisporus a more favorable conservation of metabolic resources and increased rate of development relative to a total dependence on the de novo synthesis of protein within the fruiting body.
Cultivation of Agaricus bisporus: Fruit tree culture by another name?
The unexpected discovery that the genotype of A. bisporus colonizing the upper casing layer governed the genotype of the fruiting body, independent of the genotype colonizing the lower compost layer, creates an opportunity to explore the propagation scheme used in fruit tree culture for the agronomic improvement of the mushroom. A common practice is to graft the rootstock of one cultivar to the scion of another to bring together the desirable features of the two cultivars in one tree (Fig. 3). Similarly, a "companion-inoculant" strategy could be investigated using the mushroom in which the two substrate layers are inoculated with different, but complementary, strains of A. bisporus. For example, the CI strain could be selected for imparting desirable reproductive traits, such as color, size, and quality of the fruiting body, whereas the spawn strain could be chosen for its superior vegetative traits, such as rate of growth, thermal tolerance, green mold resistance, and efficient utilization of the substrate. The companion-inoculant method would permit the combined use of two strains to achieve a set of vegetative and reproductive traits that otherwise might be difficult to breed in a single strain. The two genetically disparate strains with complementary traits will need to anastomose efficiently to form a fully functional single mycelial network in the cultivation substrate. Nonetheless, this strategy could well be a fruitful line of investigation that is ripe for development!
