Blueberries are rich in flavonoids, anthocyanins, phenolic acids, and other bioactive substances. Anthocyanins are important functional components in blueberries. We collected 65 varieties of blueberries to investigate their nutritional and functional values. Among them, Gardenblue had the highest anthocyanin content, with 2.59 mg/g in fresh fruit. After ultrasound-assisted solvent extraction and macroporous resin absorption, the content was increased to 459.81 mg/g in the dried powder. Biological experiments showed that Gardenblue anthocyanins (L¹) had antiproliferative effect on cervical cancer cells (Hela, 51.98 μg/mL), liver cancer cells (HepG2, 23.57 μg/mL), breast cancer cells (MCF-7, 113.39 μg/mL), and lung cancer cells (A549, 76.10 μg/mL), and no apparent toxic effects were indicated by methyl thiazolyl tetrazolium (MTT) assay, especially against HepG2 cells both in vitro and in vivo. After combining it with DDP (cisplatin) and DOX (doxorubicin), the antiproliferative effects were enhanced, especially when combined with DOX against HepG2 cells; the IC₅₀ value was 0.02 μg/mL. This was further evidence that L¹ could inhibit cell proliferation by inducing apoptosis. The detailed mechanism might be L¹ interacting with DNA in an intercalation mode that changes or destroys DNA, causing apoptosis and inhibiting cell proliferation. The findings of this study suggest that L¹ extract can be used as a functional agent against hepatoma carcinoma cells.

Blackberry fruit contains high levels of nutrients and phenolic compounds. Blackberry pomace accounts for 20~30% of its whole fruit during processing and is generally treated as fertilizer. Blackberry pomace has many seeds that contain carbohydrates, polyphenols, flavonoids, pectin, protein, and other bioactive nutrients. However, its functional properties and seed protein compositions have not been reported. We used a single-factor experiment, response surface, and Osborne isolate method to extract protein isolate, albumin, globulin, glutelin, and prolamin from blackberry seeds for the first time and evaluated their characteristics and functional properties. Glutelin and protein isolate showed good water-holding capacity, emulsification, and foaming capacity, while albumin and globulin showed good oil-holding capacity and thermal stability. They were found to have good antioxidant activities that might be good DPPH free radical scavengers, especially prolamin, which has the lowest IC₅₀ value (15.76 μg/mL). Moreover, globulin had the lowest IC₅₀ value of 5.03 μg/mL against Hela cells, 31.82 μg/mL against HepG2 cells, and 77.81 μg/mL against MCF-7 cells and a high selectivity index (SI), which suggested globulin had better anti-cervical, antihepatoma, and anti-breast activity but relatively low cytotoxicity. These seed proteins may have great prospects for the development and application of food and drugs in the future.

Blackberries have high nutritional value and strong biological activities, such as antiproliferative activity. Anthocyanins are important functional components in blackberries. We collected 25 kinds (lines) of blackberries from our nursery to investigate antiproliferative agents in natural foods. Among them, the Shuofeng variety had the highest anthocyanin content, with 2.54 mg/g of fresh fruit, which increased to 357.75 mg/g of dried powder through ultrasound-assisted solvent extraction and macroporous resin adsorption. Additional experiments showed that Shuofeng’s anthocyanin content had high anti-HepG2 activity in vitro and in vivo, as well as activity against Hela (68.62 μg/mL), HepG2 (55.85 μg/mL), MCF-7 (181.21 μg/mL), and A549 cells (82.01 μg/mL), as determined by MTT assay. It also had no apparent toxic effects. The combination of DDP and DOX significantly enhanced the antiproliferative activity of the four cell lines. The IC₅₀ value of Shuofeng’s anthocyanin content combined with DOX in HepG2 cells was the lowest at only 0.08 μg/mL, indicating that the combination of drugs had additive and synergistic effects. Shuofeng’s anthocyanin content might intercalate into DNA and alter or destroy DNA, causing apoptosis and inhibiting cell proliferation. Our results show that blackberry anthocyanins can inhibit the proliferation of cancer cells and their possible mechanisms. However, we must study the deeper mechanism and explore its targeting effects in the future.

Ellagic acid (EA) is present at relatively high concentrations in many berries and has many beneficial health effects, including anticancer properties. To improve the development and utilization of blackberry fruit nutrients, we divided Hull blackberry fruits into five growth periods according to color and determined the EA content in the fruits in each period. The EA content in the green fruit stage was the highest at 5.67 mg/g FW. Single-factor tests and response surface methodology were used to optimize the extraction process, while macroporous resin adsorption and alkali dissolution, acid precipitation, and solvent recrystallization were used for purification. The highest purity of the final EA powder was 90%. The anticancer assessment results determined by MTT assay showed that EA inhibited HeLa cells with an IC₅₀ of 35 μg/mL, and the apoptosis rate of the cells increased in a dose-dependent manner, with the highest rate of about 67%. We evaluated the changes in the mRNA levels of genes related to the EA-mediated inhibition of cancer cell growth and initially verified the PI3K/PTEN/AKT/mTOR pathway as the pathway by which EA inhibits HeLa cell growth. We hope to provide a theoretical basis for the deep exploration and utilization of this functional food.

Deep reinforcement learning is one of the core technologies, which leads to artificial intelligence. It has been used widely not only in robot, autonomous driving, power station control and human-computer interaction in mobile e-commerce platforms, but also in gambling games. The game software AlphaGo of artificial intelligence defeated Ke Jie in 2017. Since then, the study of artificial intelligence and deep reinforcement is widely concerned. We build a Neural Network model using TensorFlow and Monte Carlo tree search algorithm, so that agents can reach the level of playing chess with human beings in board games. The gobang interface is realized by Python, and the PhoenixGo recurrence is completed. The computation is reduced and the training effect of the agent is improved by optimizing the establishment of gobang strategic value network. Black hand first can achieve a 73% winning rate, and white hand first is 64%, when playing against human players.