General Secretary Xi Jinping stressed that "the security of seed sources is related to national security, and we must be determined to develop the seed industry in the mainland to achieve self-reliance and self-improvement in seed science and technology, and independent and controllable seed sources". Feeling the importance of the current "germplasm innovation" or "provenance innovation" or "germplasm improvement" and the great attention of the state and the industry, I have made some thoughts on germplasm innovation, and shared a few of them with you for discussion.
1. Carry out germplasm improvement according to the breeding objectives of maize varieties
The goal of germplasm improvement needs to be closely combined with the goal of maize breeding, and the two cannot be separated. There are many specific goals for breeding, but in general, they mainly focus on three aspects: high yield, environmental adaptability, and production demand. Yield is the primary trait, the importance of high yield is self-evident; adaptability includes the adaptation or resistance to light, temperature, water, diseases, etc., is an important basis for stable yield, is the guarantee of achieving large area of high yield, some studies have shown that the history of maize breeding for decades is mainly to improve the ability to resist stress; production demand mainly includes field operation needs and nutritional quality consumption needs, etc., such as the current suitable mechanized operation is the most important production demand, which derives early maturity, dense tolerance, Lodging resistance and other important traits, and the seed production ability of the female parent has become a very important trait for the needs of seed production yield and efficient reduction of seed production cost. These traits were overlooked or undervalued in previous breeding. Good varieties are the organic combination of these "three sexes", and germplasm improvement needs to be targeted to adapt to the characteristics of corn planted in relevant ecological environment areas.
Corn has been bred for tens of thousands of years. The early stage was mainly domestication, from the grass to the primitive maize, and then to be taken around the world to be "domesticated" (for the time being, the word "domestication") into different local germplasm. This process is primarily a process of adaptability. For example, the Tang Sipingtou germplasm and the Luda red bone germplasm in the mainland have good adaptability to the local environment, and the European duromeric germplasm has very good adaptability in northern Europe and the cold zone in the northern part of the mainland. These germplasms were greatly improved by inbred line breeding and reincarnation improvement in the later stage, and the corresponding germplasm groups or heterosis groups were formed.
Modern breeding is a process of synergistic improvement of high and stable yields. Previously, I put forward the idea that fertility x adaptability can be used as a theoretical basis for the design of the heterogeneity model, and its basic assumption is that one of the lines focuses on contributing to the high yield (usually the mother parent), especially in the case of higher and higher seed production costs, and the other department focuses on contributing adaptability to resist the pressure of complex environments and improve environmental resilience. The key contribution here is not the only meaning, but the two lines have their own emphasis on the basis of a certain yield, so as to ensure that the two lines accumulate different types of alleles, so as to ensure a certain genetic distance between the two lines, and ensure a reasonable heterosis. The more it is in the harsh environment, the more valuable it is, such as the extremely early maturing area and the Huang-Huai-Hai summer sowing area. According to this line of thinking, the concept of heterogeneous mode will not be constrained. In recent years, none of the varieties that have appeared in the Huang-Huai-Hai region are not BSSS/NSSS models, which also verifies the feasibility of high yield x adaptability as a heterogeneous model to some extent.
2. Germplasm innovation should pay attention to the utilization of germplasm in the United States
Local germplasm has very good diversity and adaptability, but it is also difficult to improve. For quick success, attention needs to be paid to the rational use of American germplasm. In view of the importance of American germplasm to germplasm improvement, it is necessary to think about this question before carrying out germplasm innovation or before designing ideas, why is the world's breeding more or less inseparable from North American germplasm? I think there are three important reasons:
(1) The public sector in the United States was the first to start the theoretical and practical exploration of maize breeding, which is theoretically leading and germplasm-leading in the world. For example, world-famous inbred lines such as B73 and Mo17 have been selected and bred and spread to temperate regions around the world. The yields of these materials are very good, significantly higher than those of local germplasm. The mainland introduced these two inbred lines in the 70s of the last century, B73 was not able to be well applied because it was not very good in terms of adaptability, while Mo17 showed good yield and adaptability, which had a huge impact on the mainland corn breeding, since Mr. Li Jingxiong introduced Mo17 from the United States, there are more than 90 corn monocrosses bred by inbred lines Mo17 in the country, the most famous is Zhongdan No. 2. The United States has made several rounds of improvements on the basis of Mo17 and formed a good germplasm, but China has not had a few particularly successful cases to this day, which is worthy of our reflection.
(2) The germplasm level of the United States is leading in the world, which is mainly due to the development of commercial breeding. The United States is the first country to start commercial breeding, leading by technology, and comprehensively using advanced breeding technology to carry out germplasm improvement, and it is continuous reincarnation improvement. Therefore, the germplasm level is the best in the world. The level of corn germplasm represented by Monsanto and Pioneer can be said to represent the highest level at present.
Continental maize breeding attaches great importance to the utilization of American germplasm, and there are very successful representative varieties, of which three germplasms are of far-reaching significance to mainland maize breeding: 478 germplasms, 78599 germplasms, and X germplasms. 478 types of germplasm produced representative varieties Zhengdan 958, 78599 germplasms produced representative varieties Nongda 108, and X germplasm produced representative varieties Jingke 968. It is of great significance to carefully review and analyze the history of these three types of maize germplasm and analyze their contingency and inevitability, which is of great significance to our maize germplasm innovation ideas. Of course, we can't always rely on this kind of use of American hybrids to select lines, lack of original innovation, and is not conducive to the formation of a core heterogeneous model suitable for the national conditions of the mainland. However, in recent years, many varieties have been selected and bred by foreign companies in China, which provides a very rare germplasm basis, and the selection of lines from hybrids is one of the important ways of germplasm innovation at this stage.
and (3) the impact of the U.S. maize heterogeneity model on the world
Due to the successful breeding of B73 and Mo17 lines, as well as the large-scale promotion of B73xMo17 hybrids, the heterogeneous model of B73xMo17 has gradually formed in the United States. The restriction of intellectual property rights in commercial breeding has forced various companies to carry out germplasm improvement and innovation in different directions from these two lines, and the overall BSSS X Lancaster heterogeneity model has been formed. Later, a new heterosis group of Iodent was formed, and the current BSSS x Iodent model gradually prevailed. With the promotion of varieties by multinational companies around the world, the superiority of these varieties is very obvious. The example of Xianyu 335 is a good example, and if it were not for the fact that this variety is not suitable for the climate of the Huang-Huai-Hai region, its influence on the mainland corn industry would be even more unpredictable.
There are two ways to utilize American germplasm in germplasm innovation, one is to use American hybrid lines, which have the potential to form new germplasm groups, and the other is to use American inbred lines, especially expired protection inbred lines (so that there is no intellectual property risk) to improve the fertility and adaptability of the chassis. Again, it is emphasized that it is important to use the germplasm of the United States, not that the germplasm of the United States must be used for germplasm improvement.
3. Germplasm innovation should be carried out around a certain heterogeneous model
Although germplasm innovation is more about some specific agronomic traits, such as improving drought resistance, high temperature tolerance, low temperature germination tolerance, protein content level, etc., these innovative materials will eventually enter the breeding process, so from the beginning to comprehensively consider the subsequent breeding applications, so the heterogeneous mode is the first important factor to consider. In other words, germplasm innovation should be organically integrated with product creation, rather than waiting for the creation of a new class of germplasm material to be created and then consider how to use it. This is the key reason why some germplasm innovative materials have been difficult to utilize. Although China's germplasm innovation is very successful, such as 478 classes, 78599 classes, and X classes, these have formed a new heterogeneous model, and the further improvement effect on this basis is not obvious, and it is basically a new model to replace the old model of the variety replacement process. Although the model is important, the germplasm improvement around a certain model is even more important, which is very worthy of our thought.
The hybrid model of multinational seed industry giants still adheres to the BSSS x NSSS model, and this model is in the trend of "disintegration" of local models around the world, and the mainland is no exception. For example, the Lanka x Tang Si Pingtou model in the northeast of the mainland was basically disintegrated after Xianyu 335 entered, and the proportion of Huang Huaihai's Reid x Tang Si Pingtou model was also greatly reduced under the use of pioneer germplasm, and it is difficult to say whether it will be unified to the BSSS/NSS model in the future. The IodentxFlint model in Europe is now also gradually transitioning to the BSSS/NSSS model, and the very early maturing regions of the continent are estimated to be no exception. In the past, it should be said that the region with the most complex heterogeneous model in the tropics has emerged a variety of models. Among them, the Suwan x Non-Suwan model is very famous, and the Suwan germplasm is also regarded as a successful model of tropical germplasm improvement. However, the transition to BSSS/NSSS is now also evident in the tropical regions. It can be seen that BSSS/NSS is becoming the mainstream global heterogeneous model. In this context, how we should carry out germplasm improvement and innovation is also worthy of our in-depth thinking.
The heterogeneous model is relative, there is no absolute heterogeneous mode, and there is no right or wrong. Continuous improvement along a heterogeneous model can gradually increase the yield level of the variety. Take the Reid/Tang model in the Huang-Huai-Hai region of the mainland as an example, at least today it can still compete with the BSSS/NSSS model. Continuous improvement along the Reid/Tang model will certainly be very competitive, but the key reason is that the speed of improvement is slower than that of multinational seed giants, so it will gradually fall behind, and the original hybrid model will be eliminated. The global heterogeneous model is mainly due to the formation of multinational seed giants in the practice of promoting varieties around the world, so that the cost performance is the highest for them, and the varieties that can be cultivated to adapt to different regions around the world at the minimum cost.
However, there are also many drawbacks to the global heterogeneous model, the diversity of germplasm has been greatly reduced, the homogenization trend of varieties is obvious, and the ecological vulnerability has been significantly improved. There are already many scientists who are paying close attention to this issue and calling for it. The United States has launched a GEM program to introduce tropical germplasm into the male or female parent to improve the genetic diversity of temperate germplasm. The diversity of Pioneer's parent stock has also increased a lot due to the integration of Amargo germplasm, which is also an important process of germplasm innovation.
Will we really be able to find an independent path in the future? The answer is yes, and we should be interested in taking a different path of breeding innovation. But there is also a premise, that is, we need to carry out continuous genetic improvement in a certain direction (fixed heterogeneous model), and the speed of improvement should be similar or even faster than that of multinational seed giants. Pioneer's Iodent germplasm group is a good example. The University of Minnesota, a public research institution, was the first to innovate the Iodent germplasm group, and Pioneer followed with a number of improvements that led to success. We should aspire to increase the genetic base of maize germplasm by creating a few more germplasm groups similar to Iodent, as the University of Minnesota and Pioneer did in the past (even if we concentrate the efforts of the whole country to do one, which will affect the next few decades).
4. The main task of germplasm innovation
Different germplasms represent different genetic compositions or different groups. Germplasm often represents different traits or genetic backgrounds. For example, Tang Sipingtou germplasm refers to some inbred lines with common characteristics derived from Tang Sipingtou materials. The concept of germplasm is more associated with key trait traits or germplasm sources. For example, shade-tolerant germplasm (key feature), suitable for machine harvesting germplasm (key feature), rust resistant germplasm (key feature), Tang Sipingtou germplasm (germplasm source), 78599 germplasm (germplasm source), North American germplasm and so on. The focus of the concept of germplasm resources lies in the word resource. Not all materials can be called resources, but materials with special value can be called resources. Only after the resources have been fully identified can the germplasm be truly formed.
There is also a distinction between germplasm groups and heterosis groups. A type of germplasm material with obvious common or similar characteristics can be called germplasm group, but heterosis group needs to have a relatively high level of such material and have special compatibility with materials from other germplasm groups.
Germplasm improvement and germplasm innovation. Literally, improvement is the small adjustment and improvement on the basis of the original, while innovation has the meaning of starting from scratch. Therefore, germplasm improvement is the genetic improvement of a certain type of germplasm, so as to form better germplasm material, which has a broader meaning. Germplasm innovation is more innovative, creating new traits or new groups. For example, high-oil corn is a classic germplasm innovation work, there is no high-oil corn in the world, but after continuous gene polymerization to form high-oil corn. The concepts of germplasm improvement and germplasm innovation are sometimes difficult to distinguish, especially some small innovations are difficult to distinguish whether they belong to innovation or improvement, and of course, it is not necessary to distinguish them so clearly.
To do a good job in germplasm innovation in the mainland, there are five important tasks that need to be done:
☞ (1) Collection, preservation and accurate identification of germplasm resources
By collecting germplasm resources at home and abroad, aiming at the production demand and industrial demand to carry out accurate identification, the breeding materials urgently needed for breeding are screened, such as high salt and alkali tolerant materials, high protein, and high oil materials.
☞（2）地方种质的挖掘利用
There are many local germplasms in the mainland, which contain a large number of excellent genes, but they are rarely used in current breeding. Due to the low overall yield of these materials and the difficulty of direct utilization, certain strategies need to be adopted, such as crossbreeding with modern inbred lines, to breed intermediate materials that can be directly used in breeding. This is a systematic project to tap the breeding and utilization potential of local germplasm in mainland China. It is suggested that the top-level design should be carried out at the national level, and the utilization value and potential of local germplasm in mainland China should be systematically explored, and different intermediate materials should be formed, and these materials should be stored in the national germplasm bank for future use.
☞ (3) Gene-oriented breeding material creation
Using modern breeding methods such as key genes or gene editing, some new variants or new materials that are difficult to obtain in nature are created. This was previously difficult to operate, but the development of modern breeding techniques has provided such an opportunity.
☞ (4) Germplasm expansion
Broaden the germplasm base by hybridizing with materials that are genetically distant. Among them, tropical germplasm amplification is a very important content. Tropical germplasm contains a large number of stress resistance genes, which is of great significance for the amplification of temperate germplasm. However, tropical germplasm has light and temperature sensitivity and other utilization obstacles, which is difficult to use directly, so tropical germplasm and temperate inbred lines can be used to cross to screen offspring families. The GEM program in the United States is mainly about doing just that.
☞ (5) The use of commercial hybrids to select innovative core germplasm
As mentioned earlier, the importance of American germplasm to mainland breeding has been discussed, and it can be said that every corn renewal is related to the use of foreign germplasm, and it is a direct selection of American hybrids. This is still true even for the next breed change. In recent years, more and more varieties have been approved by foreign companies in China, which also provides a rare opportunity for germplasm innovation.
5. Path selection and strategy for the improvement of inbred line reincarnation
Although intelligent design breeding will be the direction of the development of breeding technology in the future, at least at present, or even for a long time in the future, it will not be possible to achieve artificial synthesis or directional superposition of genes. The selection process of inbred lines is still dominated by genetic recombination, supplemented by genetic modification (gene editing or transgenic). Therefore, it is necessary to consider the path and strategy of germplasm improvement.
The essence of inbred line breeding is a process of genetic recombination, through which the aggregation of advantageous alleles or haplotypes is carried out, and due to the limited exchange of each round of recombination, it often has to go through multiple rounds of recombination, and different derived lineages are formed. For example, looking back at the pedigree of inbred lines, U8112 derived 478, followed by Z58, and a number of inbred lines were also bred on the basis of Z58. The inbred line breeding lineage of the multinational seed industry giant may be longer. For example, the female parent of the famous Xianyu 335, PH6WC from PH09B/PH01N, PH09B from PHP38/PHHB9, PHHB9 from PHW52/PHG86, PHW52 from B73/PHG39, and PHG39 from Amargo germplasm. This gives us a look at the entire selection lineage: B73→ PHW52→ PHHB9→ PH09B→ PH6WC, which have undergone at least four rounds of improvement. In addition, PH6WC is also an inbred line selected by Pioneer Company 30 years ago, which shows the gap between us and foreign companies in germplasm improvement.
The above-mentioned PH6WC breeding can be regarded as an innovative chain of germplasm improvement, and each inbred line in the chain can be called a "nodal inbred line", and each nodal inbred line has its own chance and inevitability. For example, Pioneer Company has selected PHG39 and PHG86 on the basis of Amargo germplasm, of which PHG39 has greater influence and contribution, using PHG39 to breed PHP38, PHR61, PHT11, PHW52 and other inbred lines, and on this basis, the second round of inbred lines such as PHBW8, PHHB9, PHRE1, PH07D and so on. It can be seen that starting from a certain inbred line (here PHG39 is taken as an example), multiple inbred lines are often selected in each round, and then the second, third or even fourth rounds of breeding are carried out on this basis. When we look back at this process, we will find that not every inbred line can be successfully continued, and it is the successful breeding of some (and maybe even very few) inbred lines that leads to continuous genetic improvement, so I will call it "node inbred lines" for the time being, and only those that can derive better inbred lines can be called "node inbred lines". Therefore, I believe that the key to germplasm improvement is to be able to breed these nodal inbred lines, which is the path of germplasm improvement. This is a theoretical problem worth studying in breeding, and it is also one of the important design contents of intelligent design breeding in the next step.
Although it is not accurate, for a moment a brief distinction is made between the concepts of path and strategy problems. The path question is to answer whether to use a particular line to select a line to derive a better line, or to choose which line from the offspring to continue a new round of improvement to ensure continuous genetic progress. The question of strategy is more about how to breed the desired line faster and better. If we compare germplasm improvement to climbing a mountain, the goal is to climb to the top of the mountain, and the path problem is to choose which road to climb the mountain from, and the strategy problem is what kind of equipment to carry to climb the mountain faster. PH6WC has been used in mainland production for many years, and there are a large number of breeders in the country who are improving this inbred line, and the manpower, material and financial resources invested are also huge, so it is very necessary to find out what kind of genetic improvement has been carried out on this basis, and what is the effect of genetic improvement? This is of great benefit to thinking about the theory and technology of maize breeding in the mainland. Another example is C7-2, due to the success of Zhengdan 958, the national efforts to improve C7-2 are also estimated to be large, but so far there seems to be no particularly famous inbred line that uses C7-2 for breeding, so a question worth reflecting on is whether there is a problem with the path (C7-2 should not be selected for breeding) or a problem with the strategy (there is a problem with the breeding method). I believe that thinking about these questions is very important for us to further develop germplasm improvement. In the next step, it is of great significance to improve the efficiency of breeding if we strengthen the study of genome structure and recombination law, combined with the prediction of gene function, and put forward guidance and suggestions for the path of genetic improvement.
6. Technical methods and organization management of germplasm innovation
Traditional germplasm innovation often only relies on cross-breeding, radiation mutagenesis, distant hybridization and other technical means, and the efficiency of identifying and polymerizing genes of target traits is low. In recent years, the continuous progress of germplasm innovation technology methods, especially the development of haploid technology, genome-wide selection technology, gene editing technology, artificial intelligence big data technology, etc., provides a powerful means for us to accurately and rapidly identify and aggregate target genes. However, there are still many issues worth studying about what kind of technical methods or technical systems should be established. In other words, how to quickly improve germplasm still has a problem of methodological optimization.
Haploid technology and genome-wide selection technology will become the core technology of germplasm innovation in the future. The advantage of haploid technology is that the pure line is fast and large-scale. How to quickly determine the utilization value of each DH line depends on traditional multi-environment testing, so the whole genome selection technology can help to carry out high-throughput evaluation of DH lines. The organic integration of these two technologies will establish a new breeding process system. In June 2023, we successfully developed a medium-density (20K) breeding chip that will serve this goal well, and several units have already used this chip and responded well. At the first National Symposium on Engineered Breeding of Maize Haploid organized last year, 30,000 and 300,000 plans were issued, and in the next five years, 30,000 inbred lines will be identified by 20K chips and 300,000 DH lines will be identified by 3K chips. Then, these data are aggregated to form a data platform, and intelligent design breeding should be carried out on this basis, which should be very helpful to accelerate the efficiency of germplasm innovation. We invite all breeders to join us in achieving this goal!
The design of the target genotype and the design of genetic improvement pathways are the most challenging. I think the biggest obstacle to germplasm innovation is not germplasm resources or breeding methods, but often we don't know what kind of materials to create to meet the market demand in 5 or 10 years. The target trait can be described, but it is difficult to define the target genotype, and it is impossible to estimate that there is no single answer. Unlike inbred crops, the heterogeneity pattern of hybrid crops complicates this problem. Taking the Huanghuaihai summer sowing area as an example, good varieties can be bred along the model of Pioneer Company, and good varieties can also be bred along the model of Zhengdan 958, which shows that the standard answer is not unique. Nevertheless, it is worth considering trying to anticipate or define the genotype of the target breed or inbred line. This shows the importance of the heterogeneous model, if the heterogeneous model is not fixed, even if the genotype of the target variety can be quantitatively described, the genotype changes of the parents are massive, and it is impossible to continuously improve. Defining the genotypes of the two parents along a certain pattern can help to improve breeding efficiency. I think that at this stage, intelligent design breeding is intelligent design breeding under a certain heterogeneous mode, if China does not solve this problem, the application of intelligent breeding will be empty talk. Therefore, in retrospect, we still need to think about whether to follow the foreign BSSS/NSS model or create our Chinese model, and on this basis, we need to consider the issue of intelligent design breeding.
The organization and management of germplasm innovation requires the concerted efforts of the whole industry. Public welfare institutions should do some long-term things, such as the excavation and utilization of local germplasm, the expansion of tropical germplasm, the excavation and utilization of commercial hybrid germplasm, and so on; enterprises and institutions need to complement each other's advantages, work together, and jointly tackle key problems to carry out the cultivation of backbone inbred lines, and on this basis, seed enterprises can carry out reincarnation improvement around the backbone inbred lines, gradually form their own material system, and enhance domestic and even international competitiveness.
Intellectual property protection is the fundamental prerequisite for germplasm innovation. The important reason why China's seed industry is not strong is that the intellectual property protection system is not perfect. Steal someone else's inbred line or take it and make it your own with a little embellishment, without paying any price. In such an environment, it is difficult for seed enterprises to continue to invest in breeding scientific research, and it is difficult to ensure effective innovation. It is gratifying that since the start of the seed industry revitalization action, great progress has been made in this area in recent years, but there is still a lot of room for improvement, especially the protection system of substantial derivative varieties (EDV) has not yet been truly implemented, which requires the joint efforts of the whole industry.