Summary

Introduction

The production of high amounts of phytogenic biomass with high quality or high bioavailability of valuable nutrients is one of the most important challenges to meet future demand (SCAR 2008; Flachowsky 2008; The Royal Society 2009). The world population is predicted to grow from presently 7 billion to about 9 billion in 2050, and the demand for food of animal origin may double (Steinfeld et al. 2006; Godfray et al. 2010), driven by increasing income from productive employment (Keyzer et al. 2005) and preference for “Western style of life” in many developing countries. Food of animal origin like poultry meat and eggs contributes to meet the human requirements in amino acids and many trace nutrients. The production of food of animal origin requires vast resources (e.g. Flachowsky 2002, 2011) especially in terms of arable land for feed production. Figure 1 shows the effects of population growth on the availability of arable land per person and the number of people to be fed per ha arable land during the time from 1950 to 2050. Furthermore, feed/food production causes emissions with greenhouse gas potential such as carbon dioxide (CO2) from fossil fuel, methane (CH4) from the enteric fermentation (esp. ruminants) and from the excrement management as well as nitrogen-compounds (NH3, N2O) from the protein metabolism in the animals (see DEFRA 2006; Flachowsky and Hachenberg 2009, FAO 2010; Grünberg et al. 2010; Leip et al. 2010). Additional arable land will be needed to produce biofuel and material for the industry, competing with land use for feed production. Therefore plant breeding and cultivation are the focal points for global



feed and food security in the years ahead. High yielding plants with low external inputs of limited natural resources should be the main goals of plant breeding in the future. So-called “Low Input Varieties” should use unlimited resources such as sunlight or sun energy, nitrogen (N2) and CO2 as plant nutrients from the atmosphere to the highest possible level and should use limited resources such as agricultural area, water, fossil energy, phosphorus etc. as effectively as possible (see Table 1). The biodiversity of microorganisms, plants and animals offers an extremely large gene pool which has been already used by traditional plant breeding and which could be used more intensively in the future. Apart from traditional breeding, plant biotechnology apparently has a potential to contribute to the objective of “Low Input Varieties”. The cultivation of GMP increased worldwide from about 1.7 (in 1996) to nearly 148 million ha (in 2010), representing about 10% of arable land (James 2011). In % of the global GM area, the most important GM-crops are currently soybeans (60), corn (24), cotton (11) and canola (5) (Figure 2).

Definitions

Presently, most of the Genetically Modified Plants (GMP) are modified for agronomic traits (see Figure 2) such as increased tolerance against insects or higher resistance against insecticides or pesticides. Such plants are characterized by so-called input traits (GMP of the first generation) without substantial changes in composition and/or nutritive value. Such plants can be considered as substantially equivalent to their isogenic counterpart (OECD 1993).





GMP of the second generation (with output traits) should contain more nutrients or less anti-nutritive substances. Such plants (feeds) are not substantially different in composition from their counterpart. GMP offer a wide range of application in animal nutrition. Seeds and by-products from food and biofuel industry are the most important feedstuffs for poultry.

Based on the present (public) situation animal nutritionists are to address the following aspects:



In Europe the safety of GMP for humans, animals and the environment is assessed by the Panel for Genetically Modified Organisms (GMO-Panel) of the European Food Safety Authority (EFSA, located in Parma; Italy), based on various Guidance documents (e.g. EFSA 2006, 2008). The EU-Commission is responsible for the risk management.

Compositional analysis

Composition analysis of feeds from GMP is the starting point for nutritional assessment. There are different recommendations for compositional analysis of GMP for feed groups (e.g. concentrates, forages etc.) and for animal groups (e.g. ruminants and non-ruminants), as shown for non-ruminants in Table 2. Between 60-100 ingredients of transgenic, isogenic and commercial varieties will be determined to compare the composition of plants and feeds from plants. In addition the newly expressed protein(s) and their degradability (mostly in vitro) will be determined.

No additional animal studies are recommended (EFSA 2006, 2008) if the GMP are substantially equivalent to their isogenic counterpart in the case of GMP of the 1st generation. Nevertheless many feeding studies with feeds from GMP of the 1st generation have been carried out during the last few years. Incidentally, all these studies can contribute substantial information to feed science, which has been dramatically neglected during the last 30 years.

Feeding studies

Types of studies

Details of sampling for animal feeding studies, handling of samples and preparation of samples for animal feeding studies are described by ILSI (2007) and are shown in Figure 3. Feeding studies with laboratory animals and with food producing (target) animals can be done with various objectives to answer different questions (Table 3; see also Flachowsky and Wenk 2010). Many studies were done with laboratory animal models for toxicity testing of single substances (single dose toxicity testing, repeated dose toxicity testing, reproductive and development toxicity testing, immunotoxicity testing etc.; EFSA 2008). Laboratory animals were also used for the safety (and nutritional) assessment of the whole GM-food and feed (in general 90-day feeding studies to detect unintended effects, sub-chronic animal tests, allergenicity tests; for margins of safety etc.; EFSA 2008; 2011; OECD 1995).

Studies with target animals are more of nutritional concern. The type of study depends on the type of genetic modification in plants, the availability of GM-feed and further factors (see Tables 3 and 4).

Results of feeding studies with GMP of the 1st generation

In previous studies the authors compared only the composition and the nutritive value of one feed (e.g. transgenic origin) with another one (e.g. isogenic counterpart) and neglected the considerable biological range described e.g. in the OECD-consensus documents (OECD 2001a, 2001b, 2002a, 2002b, 2003, 2004a, 2004b, 2004c, 2005) or other feed value tables. In general GMP´s of the first generation were essentially equivalent to their isogenic counterparts. Under some cultivation conditions the mycotoxin contamination of GMP feed was lower than in feed from non-GM plants. For example, Bt maize is less severely attacked and weakened by the European corn borer and might have a greater resistance to field infections, particularly to Fusarium fungi, which produce mycotoxins. Evidence of reduced mycotoxin contamination in GMP has been demonstrated in some, but not all studies, as summarized by Flachowsky et al. (2005a). In long-term studies, numerous researchers investigated the influence of levels of corn borer infestation of isogenic and Bt hybrids on mycotoxin contamination. Most researchers reported a lower level of mycotoxin contamination in the transgenic hybrids, over a considerable geographical and time range of observations (Figure 4).





In early feeding studies with food producing animals, feeds from GMP of the first generation were only compared with their isogenic counterparts to demonstrate equivalence (OECD 1993). Later studies included three or more commercial varieties to measure also the biological range of various measurements. In recent years about 150 feeding trials with food producing animals were reported in peer reviewed papers and summarized in several reviews (see above).

The inclusion of commercial varieties in such studies as recommended by ILSI (2007) and EFSA (2008) may contribute to a more biologically relevant assessment of the results of animal feeding studies (e.g. Lucas et al. 2007; McNaughton et al. 2007; see Table 5).





Apart from a statistically significant small increase in relative liver weight (p<0.05) of female broilers, no relevant differences between transgenic maize (DAS-59122-7) and its isogenic counterpart were found in this feeding trial. The inclusion of several commercial non-GM-varieties in the field and in animal feeding studies should help to avoid wrong conclusions from experimental data. Long term feeding studies cover a very long period of the life or the whole lifespan of the animals. Results from such studies and multi-generation studies may include not only the animals´ growth performance, but also their health and reproductive performance (BEETLE 2009) in response to being fed high amounts of GM-feed. In laboratory studies, no negative effects on reproductive traits were found in rodents fed with Bt-corn, glyphosate tolerant soybeans or GM-potatoes compared with their conventional counterparts (Brake and Everson 2004; Kilic and Akay 2008; Rhee et al. 2005). The results of two multi-generation studies at our Institute with laying hens (Halle et al. 2006) and quails (Figure 5) showed no differences in production and reproduction performance between laying quails fed diets containing 50% Bt maize vs. diets containing 50% isogenic maize Table 6 summarizes results from feeding trials with different poultry species and categories, comparing feeds of GMP of the first generation (plants with input traits) with their isogenic counterparts. The absence of biologically relevant adverse effects in poultry studies is not surprising in view of the compositional equivalence between feeds from isogenic and transgenic plants and the general observation that GMP of the 1st generation are comparable with plants from traditional breeding.

Results of feeding studies with GMP of the 2nd generation

During the last few years much attention has been spent to develop GMP, in which significant intended alterations in composition have been achieved in order to enhance the nutritional properties or health benefits. Examples of nutritionally improved GMP are given in Table 7.





New experimental designs are necessary for nutritional assessment of GMP of the 2nd generation (Flachowsky and Böhme 2005; ILSI 2007; EFSA 2008, 2011; Flachowsky and Wenk 2010) to test the significance of higher concentrations of valuable substances such as nutrients or nutrient precursors or lower concentrations of undesirable ingredients. An experimental design to demonstrate the bioavailability of a nutrient precursor is shown in Table 8.





Table 9 shows an example to determine the ß-carotene conversion from maize into vitamin A in Mongolian gerbils.





Adequate studies are necessary to demonstrate the effects of other newly expressed nutrients or higher levels of nutrients such as amino acids (Lucas et al. 2007), fatty acids (Meja et al. 2010), nonessential substances like enzymes or essential oils (Zhang et al. 2000).

The introduction of new gene fragments may trigger the expression of new substances, which were never before in such plants. A recent example is the introduction of genes which express two desaturases in soybeans with the consequence to synthesize C18:4 n-3 octadecatetraenoic acid, also known as stearidonic acid (SDA; see Figure 6). This long-chain omega-3 fatty acid is one of the precursors for the formation of the long chain omega-3 polyunsaturated fatty acids 20:5 n-3 eicosapentaenoic acid (EPA) and 22:6 n-3 docosahexaenoic acid (DHA) which are essential for human and animal nutrition and have potential health benefits (Gebauer et al. 2006, Ursin 2003; Harris et al. 2008, Whelan et al. 2009).

The SDA-content of such soybean oil may vary between 20 and 30%. Rymer et al. (2011) added 45 (grower) and 50g (finisher) soybean oil containing 24% SDA to broiler feed and confirmed results from lactating cows (Bernal-Santos et al. 2010): increased concentration of SDA, EPA and DHA in various meat samples, compared to conventional soybean oil. Even higher EPA and DHA concentrations were achieved with fish oil supplementation, but the fishy taste was not acceptable. Gibbs et al. (2010) suggested the int roduction of SDA in broiler feed as a possibility to increase the longchain n-3 PUFA intake of humans.

Fate of transgenic DNA and newly expressed proteins

The intake of feeds from GMP results in the ingestion of transgenic DNA and newly expressed protein(s). Several studies were conducted to trace their fate during food/feed processing and when passing through the gastrointestinal tract of animals, and the extent to which transgenes or their products may be incorporated into animal tissues. Table 10 shows the influence of various processing conditions on some DNA fragments of rapeseed. Higher temperatures and extraction contributed to the degradation of DNA fragments. There is agreement among authors (Mazza et al. 2005, Sharma et al. 2006, Alexander et al. 2007) that recombinant DNA would be processed during feed treatment (ensiling, extraction etc.; see Table 10) and in the gut in the same manner as genetic material from endogenous feed, as shown in feeding studies with non-ruminants at our Institute (see Table 11) and



several other institutions. Small DNA fragments from isogenic and transgenic plants could be detected in blood, spleen, liver and kidney (Mazza et al. 2005).









Newly expressed proteins show similar chemical and physiological properties, including microbial and enzymatic degradation (Hammond 2008), as native plant proteins (Alexander et al. 2007).

Future tendencies

Presently many GMP containing stack events are being developed and already in cultivation (Figure 7). That means for example, the plants are resistant against insects and tolerant against insecticides. There are already plants in the pipeline containing up to eight stacks. In the future we may expect GM-plants with changed composition (2nd generation of GMP), more resistant against biotic and abiotic stressors such as drought and saline soils and more efficient in using limited natural resources (Low Input Varieties; see Table 12).

Conclusions

“Green” biotechnology should be considered as a method of plant breeding. Presently, the breeders improve resistance and tolerance of plants against insects, herbicides and/or insecticides (plants of the 1st generation) or influence the composition of GMP by increasing valuable nutrients and/or decreasing anti-nutritive substances (plants of the 2nd generation). Many new developments, including changes in composition, are in the pipeline by different companies. Furthermore, GMP´s are being developed to improve their agronomic properties such as drought resistance and salt tolerance (abiotic stressors; see Table 12).

Assessing the nutritive value and the safety of feeds from plant breeding and dealing with GM-animals are real challenges for animal nutritionists in the future (Figure 8). Various types of studies are necessary to answer all the questions and to contribute to a better public acceptance of such plants and animals (see Tables 3, 4, 8 and 9).





Presently, 10% of the global arable land is cultivated with GM-plants of the first generation, which have been tested in about 150 feeding studies with food producing animals. No biologically relevant effects have been described in peer reviewed papers where the authors compared feed from GMP with their isogenic counterpart and commercial varieties if fed to broilers or other food producing animals.

GMP for more efficient use of limited resources such as water, arable land, fertilizers etc. are under development (see Table 12), but not yet in cultivation. Development of such plants is a real challenge for plant breeders all over the world for substantial contributions to global food security (Table 13). Safety and nutritional assessment of GMP and feeds from GMP are a substantial prerequisite for feeding such products to food producing animals and for a better acceptance in the society