After decades of development, metabolomics are now widely used in clinical researches. In particular, solid-phase NMR has become a favorite, with which clinical researchers can obtain accurate information from tissue samples. More and more researchers have begun to engage in solid-phase NMR metabolomics research.
The key to good metabolomics research is to collect metabolic information from tissue samples in vivo as much as possible. However, due to the limited conditions, it is usually impossible to perform NMR testing immediately after the tissue sample is separated. Generally, the metabolic processes continue after tissue samples are removed from the body, and metabolic levels are affected by the sample storage mode. Therefore, tissue samples must be temporarily stored under conditions that minimize the loss of tissue metabolic information and minimize the changes in metabolic state. Inappropriate pretreatment methods of tissue samples may change the metabolic information in vivo, resulting in the acquisition of wrong metabolic information by NMR testing, which would greatly affect the results and even cause trial failure. Therefore, we try to find the optimal pretreatment method for colorectal tissue samples in this research.
The results showed that liquid nitrogen storage had the least impact on the metabolic information of colorectal cancer tissues, mainly by slow down the glucose metabolism and choline phospholipid metabolism process of the tissue samples after in vitro. Compounds with statistically significant difference included nonanoic acid, octadecanoic acid, ribose, heptadecanoic acid, pyruvate, succinic acid, sarcosine, glycine, creatine and lactic acid. Mapping these compounds into metabolic pathways, we found differences in glucose metabolism, choline phospholipid metabolism and possible corruption metabolic processes.
Glucose is the main source of energy for cell metabolism. It is mainly supplied via three metabolic pathways: glycogen synthesis, glycolysis and pentose phosphate (PPP)[50]. The carbohydrate metabolism process of cancer cells is different from that of normal tissue cells, and their rapid division and proliferation make cancer cells prefer glucose as energy supply materials. Cancer cells have extremely fewer organelles and are unable to carry out complex biochemical reactions. In the process of carbohydrate metabolism in cancer cells, the main metabolic mode is called Warburg effect, also known as aerobic glycolysis[51–53]. The level of Pyruvic and L-lactate, as the downstream metabolites of glucose metabolism, could reflect the speed of glucose metabolism. When the tissue samples is in vivo, glucose metabolism is under way all the time. But after removal from the body, this metabolic process continues or changes to produce some metabolites which is irrelevant to metabolic information in vivo. In this research, we found that the level of pyruvic and L-lactate are lower in liquid nitrogen group than that of the other groups. And as we can see in Fig. 3, pyruvic and L-lactate are down-stream metabolites of glucose metabolism, which means pyruvic and L-lactate are produced continuously along with glucose metabolism after in vivo, and the speed of pyruvic and L-lactate production in liquid nitrogen is slower than that of other groups. We can therefore conclude that the speed of glucose metabolism in liquid nitrogen is slower than that of other groups.
Choline, as the core metabolite of choline phospholipid metabolism, also had a lover level in liquid nitrogen than that of other groups, as shown in Fig. 3. We know that choline continues to decrease with choline phospholipid metabolism progresses. In this research, the level of glycerophosphocholine in liquid nitrogen is lowest among all groups, which means that the speed of choline consumption in liquid nitrogen is the slowest. We can conclude that liquid nitrogen storage could slow down or even stop the speed of choline phospholipid metabolism[54–56].
In addition, we also find that the levels of nonanoic acid, octadecanoic acid, heptadecanoic acid and ribose in the liquid nitrogen group were significantly different from those of the other four groups. Consulting relevant literature, we find that these compounds are related to the metabolic pathway of corruption. However, the specific pathway has not been reported in the literature. Based on the existing research data, we can make the following speculations: nonanoic acid, octadecanoic acid, heptadecanoic acid and ribose may be related to the corruption process of the tissue samples after in vivo. Liquid nitrogen not only can stop the internal metabolic process of the cell, but also the bacterial-related corruption. Specific metabolic processes, specific mechanisms and pathways need to be discussed after the relevant research is available[57–59].
This experiment manually screened 11 compounds with significant differences. However, there are 60 compounds with a VIP value greater than 1, and other compounds may play important roles in other pathways. Therefore, we used MATLAB to draw heat maps of these compounds to analyze the changing trend of these metabolites visually.
Although we obtained much useful information from this research, the following shortages are noted. The sample size is not large, and it can only be used as a catalyst for methodological innovation. There is no statistical difference in the changes of most metabolites, only if the sample size is expanded in the future can we continue to explore the metabolic discipline of related metabolites and consequently explain the corruption metabolic pathways.