Led by a Nobel Prize winner, a modular click compound library is constructed. "Deep Synthesis" has obtained tens of millions of yuan in angel round financing. | 36Kr exclusive.
Text｜Hu Xiangyun
Editor｜Hai Ruojing
36Kr has learned that the molecular building block company "Deep Synthesis" has recently completed a tens of millions of yuan angel round of financing. This round of financing is invested by Houxue Capital, Lihe Finance, and Xiaomiao Langcheng Investment. The raised funds will be mainly used for the construction of an automated click chemistry laboratory, as well as the establishment of an azide compound library and an AI data base.
Deep Synthesis officially started operations in 2024. Its establishment is mainly based on the Nobel Prize-winning achievement of "click chemistry" and relies on the advantageous disciplines, software and hardware facilities of Shanghai Jiao Tong University to achieve the transformation of scientific and technological achievements. The company's founder, Professor Dong Jiajia, once worked in the team of K. Barry Sharpless, the father of click chemistry in the United States and a Nobel Prize winner. Later, he joined the Institute of Organic Chemistry of the Chinese Academy of Sciences through the "Hundred Talents Program" and in 2022, he joined Shanghai Jiao Tong University as a tenured professor.
Breaking Through the "Last Difficulty" of Click Chemistry Technology
Currently, with the increasing cross-penetration between disciplines, more efficiently achieving molecular functions through synthesis has become an inherent demand of the synthetic discipline. In 1999, Professor Karl Barry Sharpless from the Scripps Research Institute in the United States proposed a modular synthesis method. He believed that synthetic chemists should use a few, highly efficient, and highly specific chemical reactions to modularly connect building block molecules, thereby efficiently achieving molecular functions, and named this method Click Chemistry.
The most representative reaction in click chemistry is the copper(I)-catalyzed cycloaddition reaction of terminal alkynes and azides (CuAAC), which was independently reported by Sharpless and the Danish scientist Meldal in 2002. Azides and terminal alkynes remain stable under the vast majority of chemical conditions, but they can be efficiently and specifically converted into 1,3-substituted triazoles under the catalytic conditions of copper(I), with the characteristics of mild conditions, high yield, high chemical selectivity, and no interference from water and oxygen.
Currently, CuAAc has been widely used in many fields such as organic synthesis, medicinal chemistry, chemical biology, and materials chemistry. With this research, Sharpless and three other scientists jointly won the 2022 Nobel Prize in Chemistry.
However, the CuAAC reaction also has some objective problems. For example, the synthetic space of this reaction is greatly limited by the availability of azide compounds and terminal alkyne compounds. Due to potential toxicity and the explosion risk in the preparation process, azide reagents are not easily obtainable (previously, only a few hundred azides were commercially available, unable to cover a diverse molecular structure space), which also leads to the fact that the biggest use of this method before is to link molecules under complex conditions, rather than the traditional diverse synthesis.
What Deep Synthesis is trying to solve is precisely this problem. Currently, relying on the research of Professor Dong Jiajia, the company has invented a "modular click compound library construction technology", which uses fluorosulfonyl azide to convert primary amine compounds (the most commercially available and numerous chemical building blocks (100,000+)) into azide compounds in a high-throughput, modular, and low-cost manner, thereby solving the scientific problem of the difficulty in obtaining azide compounds.
"Because the risk of explosion of azide compounds in the pure state is much greater than that in the diluted state after being dissolved in a solvent, in specific practice, we directly synthesize azides in a 96-well plate to avoid the separation and purification of azide compounds, keeping them in a dissolved and diluted state, thereby minimizing the possibility of explosion to the greatest extent." Professor Dong Jiajia explained.
It is reported that in the past using traditional synthesis methods, the number of molecules synthesized per person per day is generally only a few. However, under the "modular click compound library construction technology", it is possible to achieve a higher chemical synthesis throughput of thousands or even tens of thousands of molecules per person per day, greatly increasing the synthesis rate.
At the same time, thanks to the two-step continuous click reaction, the product does not need to be separated, which greatly reduces the separation cost and time. In addition, using this technology, chemical synthesis can be carried out in 96/384-well plates, and the compounds synthesized in real time can be directly used for biological phenotype screening, achieving a seamless connection between high-throughput synthesis and high-throughput screening.
Several Lead Compounds Have Been Screened Out
Professor Dong Jiajia introduced that currently, Deep Synthesis has carried out multiple project collaborations with pharmaceutical companies and research institutes to find better active molecules and probe molecules for important physiological processes, and several lead compounds with high biological activity for different disease models have been screened out.
With the application of the "modular click compound library construction technology", compared with traditional CRO enterprises, the product forms of Deep Synthesis mainly present three new characteristics:
First, there are more types of drugs that can be produced. "We can try conventional macromolecular and small molecular drugs, as well as cosmetic products, pesticides, veterinary drugs, etc. Currently, we have approximately 380,000 compounds in the library, with a base iteration of about 5,000 - 6,000, and the expected annual compound production in the future can reach the million level."
Secondly, it lies in the optimization of efficiency and cost. Deep Synthesis can deeply participate in the entire process of preclinical drug discovery, especially in the lead drug discovery stage. Deep Synthesis can quickly and abundantly provide a low-cost active small molecule library to achieve the effect of cost reduction and efficiency increase. This technology has made relevant progress in the fields of anti-tumor, anti-viral, and anti-tuberculosis bacillus. "At present, our customer customization needs can basically be delivered within 2 weeks, and the single-well cost can be reduced to within 100 yuan."
Third, innovation in the business model. Currently, the commercialization attempts of Deep Synthesis mainly focus on two aspects: On the one hand, it is to provide drug development services for pharmaceutical companies, research institutes, etc. as usual; on the other hand, the company is also laying out more possibilities for cooperation in the "AI for science" aspect, such as correcting model parameters by providing effective wet experiment underlying data to empower the vertical field.
But different from the traditional CRO model, the technical intervention gives Deep Synthesis a relatively stronger say in the commercial share.
Take the anti-tumor molecular glue product cooperation between the company and Shanghai Ruijin Hospital as an example. Deep Synthesis provided more than 50,000 active small molecules to Shanghai Ruijin Hospital in a short period of time and screened out a molecular glue with high anti-tumor activity (DC50 is 5 pmol), achieving a joint patent application by both parties (each accounting for 50%). "Under the profit-sharing model, no matter at what stage we conduct sales or disposals, the income will be evenly divided between the two parties."
In the business exploration of AI for science, for a long time, the difficulty that plagues many AI pharmaceutical companies is that most of the data available for their reference and associated learning comes from public literature, and most of the algorithms trained are also completed based on this, resulting in it being difficult to obtain good results when the model is actually applied to a specific drug development scenario.
"To a certain extent, we can assume the role of the data base for AI. By customizing wet experiments for it and reaching the required data volume for its training, the final trained model effect will be much better than using traditional methods." Professor Dong Jiajia said.
It is understood that in this regard, Deep Synthesis has already achieved cooperation with relevant AI pharmaceutical enterprises.
It is also reported that the "Click Innovation Incubator" centered around Deep Synthesis has recently been established in Minhang, Shanghai. It is reported that this incubator will provide a full life cycle incubation service chain around click chemistry technology. By building a technical support system, an investment service system, and a low-cost operation system, a closed-loop incubation will be formed. Currently, Shanghai Zhitai, Beijing Anji Bai'ao Bio, and other enterprises have completed their settlement.
Investor Views:
The Houxue Capital said that as a strategic leading industry in China, the synthetic biomedical industry is becoming the commanding height of the great power game and an important engine to promote economic development in the new era. The Deep Synthesis team is a rare team with strong R & D capabilities and rich industrialization practical experience. Houxue firmly believes that with the world-leading "click chemistry" technology, Deep Synthesis can achieve the dual optimization of cost-effectiveness and performance in drug research and development, and is expected to change the industrial pattern monopolized by Europe and the United States, becoming a leader in the front end of the domestic new drug research and development industry chain. Houxue is very honored to participate in this round of financing of Deep Synthesis. We will work hand in hand with the enterprise to strive to practice the development of new-quality productive forces.
The Lihe Finance said that we highly recognize the deep technical accumulation of Deep Synthesis in the field of "click chemistry" and the huge commercial potential contained in its business. The compound library is the core underlying asset of pharmaceutical companies, and the construction cost is huge. Based on "click chemistry", Deep Synthesis can efficiently and modularly convert primary amines into azide compounds. This technology shows significant advantages in synthesis efficiency and cost control, greatly reducing the threshold of new drug research and development. Currently, Deep Synthesis has quickly entered the top customers in the industry, and is also laying out the "AI for Science" business to provide real training data for AI to empower new drug research and development. We believe that Deep Synthesis is expected to bring changes to the industry with its technological innovation.
The Zizhu Xiaomiao Fund said that AI for Science is developing rapidly, but AI cannot replace the physical process of compound screening, and also lacks a large amount of real data for training. Click chemistry is expected to become the infrastructure for AI research and a catalyst to accelerate its development. According to the research, only the third-generation click chemistry truly has the potential for large-scale commercialization, and Deep Synthesis is the only team in the world with patents, materials, and process technologies, with extremely high barriers. We are very optimistic about the company's commercial prospects.