As a larger fungus, mushrooms are widely distributed worldwide with high nutritional value and pharmacological effects (Lei et al., 2020). According to statistics, there are currently more than 14,000 species of mushrooms in the world, and more than 3,800 species are known in China, of which approximately 480 are poisonous (Wu et al., 2019). Many edible mushrooms have similar morphological characteristics to poisonous mushrooms, and thus, it is difficult to distinguish poisonous mushrooms by the naked eye (Yang, 2013). In China, many food poisoning incidents are caused by poisonous mushrooms every year (Chen et al., 2014; Sun et al., 2018). According to statistics, there were 276 independent mushroom poisoning incidents in 2019, involving 769 patients from 17 provinces and 22 deaths, with an overall mortality rate of 2.86% (Haijiao et al., 2020). Among them, Chlorophyllum molybdites is a typical type of mushroom that is prone to be eaten by mistake. C. molybdites is widely distributed throughout the year in different regions of China, which has become the most poisonous mushrooms closest to humans (Soltaninejad, 2018). In 2018, 55 mushroom poisoning incidents caused by accidental eating of C. molybdites involving at last 133 patients were reported in China (Soltaninejad, 2018).

The mushroom poisoning incidents have become one of the serious food safety issues in China. For different types of poisonous mushrooms, they contain various toxins that result in different kinds of toxic symptoms. Mis-eating of C. molybdites mainly causes serious gastrointestinal discomfort that can last from 1 to 6 h after consumption. The toxin is a protein called molybdophyllysin (Yamada et al., 2012). If not treated in time, it will cause multiple organ failure and is life-threatening. Therefore, the rapid and accurate identification of mushroom species is necessary for proper clinical treatment. In food poisoning incidents caused by poisonous mushrooms, it has been found that mushroom samples lose their original morphological features after being cooked and digested. Therefore, identifying mushrooms by morphological identification methods is impossible. Various methods for mushroom species identification have been developed, in which the detection of mushroom toxins and DNA-based molecular biology methods are the most widely used (Li et al., 2016). With the discovery of various toxins in mushrooms, methods for identifying mushroom species by detecting toxins have been developed, such as chromatography, liquid chromatography diode array detection, inductively coupled plasma atomic emission spectrometry, and other technologies (Ahmed et al., 2010; Yao et al., 2018; Wurita et al., 2019; Numanoglu Cevik, 2020). However, most of these methods require expensive instruments, a long detection cycle, and high detection costs, which are not suitable for application in primary regions. The rapid detection of toxins represented by lateral flow immunoassays has the advantages of low cost, rapid detection, and intuitive results, but the toxin antibodies are not stable and are difficult to prepare, and the sensitivity and specificity need to be improved for practical applications, especially with respect to processed products (Tripathi et al., 2018). As the basis of the genetic material of organisms, nucleic acid would not be affected by age, season, developmental stage, and environment factors, and has a high thermal stability that could not be interfered by alterations during processing (Hebert et al., 2003; Garnica et al., 2016). With the development of polymerase chain reaction (PCR), a variety of diagnostic methods have been derived, such as real-time PCR, polymerase chain reaction–restriction fragment length polymorphism (PCR-RFLP), and DNA barcoding (Sugano et al., 2017; Kondo et al., 2019; Wu et al., 2019). However, these methods usually rely on expensive instruments and are time-consuming, which were technically challenging to achieve on-site rapid detection. Therefore, the development of simple, efficient, accurate, and sensitive mushroom species identification methods would be helpful for timely symptomatic treatment in the early onset of poisoning.

Isothermal amplification methods have been widely used for species identification of pathogenic fungi in recent years (He et al., 2019; Qiu et al., 2019; Bodulev and Sakharov, 2020). Among them, loop-mediated isothermal amplification (LAMP) is the most widely used one, which has received increasing attention (Lee, 2017). LAMP was first proposed by Notomi et al. (2000). The LAMP reaction contains a set of four to six specific primers, which can quickly identify six to eight regions on the target sequences and bind to strand displacement active DNA polymerase. LAMP reaction can be carried out at a constant temperature (60–65°C), generating a vast number of amplicons (Lee, 2017). There are various detection methods for LAMP products; among them, the colorimetric method is easily used in on-site detection. The reaction result of the colorimetric method is visible to the naked eye, and it can be realized by only relying on a water bath without expensive equipment (Wang et al., 2014). After more than 20 years of development, LAMP has been widely used in the field of pathogenic microorganism rapid text (Lv et al., 2020), clinical disease diagnosis (Wang W.T. et al., 2019), food hygiene inspection (Zhu and Wang, 2018), and environmental monitoring (Becherer et al., 2020), in view of its efficient, specific, and sensitive detection capabilities, as well as a wide range of applications and the capacity of this technology to be performed outside of the routine laboratory environment. In terms of mushroom species identification, Vaagt et al. (2013) first applied the LAMP-based technique to the detection of Amanita phalloides. He et al. (2019) established the LAMP method for 10 poisonous Amanita detections. To date, there is no report on the detection and identification of C. molybdites.

In this study, the LAMP assay based on the colorimetric method targeting the internal transcribed spacer (ITS) was established for the rapid detection of C. molybdites from edible mushrooms with similar shapes and other common toxic mushrooms. In addition, considering the practical application of this method, mushroom processing and digestion in the human body were simulated. The LAMP detection method of C. molybdites established in this study is of great significance for the rapid identification of the toxic mushrooms in poisoning events, the targeted treatment after poisoning, and the prevention and diagnosis of poisoning.

Mushroom gatherers may mistakenly identify mushrooms from the species of C. molybdites as edible, thus causing food poisoning. In this study, we established a LAMP method that is relatively simple and field-adaptable.

One of the challenges in establishing an efficient LAMP detection assay is the selection of species-specific amplification targets (Tomita et al., 2008). Mitochondrial DNA has been widely studied in species for the selection of target genes based on its multiple copies, matrilineal inheritance, and rapid evolution rate (Dulay et al., 2020; Mesic et al., 2020). Among these, the ITS gene has been proven to have high intraspecific differences and low interspecies variation. Schoch et al. (2012) tested six target regions (SSU, LSU, ITS, RPB1, RPB2, and MCM7) commonly used in the molecular identification of fungi. They proposed the use of ITS genes as the official fungal amplification target. This proposal was supported by the scientific community and later used as the Barcode of Life⁴ by the Consortium. ITS gene fragments are commonly used in the detection and identification of mushroom species (Garnica et al., 2016; Badotti et al., 2017; Wang S. et al., 2019). A rich and complete database is the basis for the design of specific primers. Fungi are a group of organisms that show great diversity in morphology, life cycle, and ecology; the current research on fungi is very limited. In molecular biology studies on fungal species, a large number of species are only represented by a single strain or specimen, and thus, nucleotide variation data for within-species and close-source species comparisons are lacking (Xu, 2016). Information on such variations is essential to distinguish closely related species (Xu, 2016). As one of the fungi with large species abundance, mushrooms are also faced with the problem of lack of genetic information. Therefore, we first examined the integrity and availability of nucleic acid sequences of Chlorophyllum before designing the specific C. molybdites LAMP primer. According to the literature, there are about 20 species of Chlorophyllum recorded in the world (Vellinga, 2002; Ge et al., 2018). We searched the ITS sequences of these species through NCBI, and all the species were successfully searched. Therefore, in the present study, we selected the ITS gene as the molecular marker to design species-specific LAMP primer sets.

In the selection of binding sites for LAMP primers, sufficient numbers of mismatches with non-target species should be confirmed. For example, rainbow trout (Oncorhynchus mykiss) LAMP primers designed by Xiong et al. (2020) had a total of 43 mismatch sites against their near-source species (Salmo salar, Oncorhynchus gorbuscha, Oncorhynchus tshawytscha, Oncorhynchus keta, Oncorhynchus masou, and Oncorhynchus nerka), resulting in high specificity for the rainbow trout. The primer set designed in this study had several mismatches against other species of Chlorophyllum at the 3′ end of C. mol-F3/B3 and the 5′ end of C. mol-F1c/B1c regions (Figure 1), demonstrating the high specificity of the C. molybdites primer set.

The sensitivity of LAMP detection is usually 10–100 times higher than that of conventional PCR assays (Wang et al., 2015). The method proposed in the present study proved to have high sensitivity and could detect 1 pg of C. molybdites DNA. This conclusion was confirmed by multiple independent repeated experiments (Figure 4). In related studies, the LAMP detection method established by Chen Zuohong’s research team from Hunan Normal University for lethal Amanita species had a detection limit of 10 pg (He et al., 2019). The sensitivity of the LAMP assay might be affected by many factors. It has been reported that the addition of loop primers could increase the sensitivity and specificity of the LAMP reaction (Nagamine et al., 2002). Nagamine et al. (2002) found that significant amplification of 10³ lambda DNA could be observed with loop primers. While removing loop primers, the signal was dispersed among 10³ copies within 52–92 min, which proved that significant DNA amplification required at least 10⁴ copies, indicating a higher sensitivity of the LAMP reaction with loop primers. The detection limit of C. molybdites in our study also confirmed this. In addition, some researchers found that the amount of DNA templates larger than 200 ng had an inhibitory effect on the LAMP reaction (Njiru et al., 2008). In the current study, when the content of C. molybdites DNA added to the reaction system was 100 ng, the LAMP reaction was inhibited, and the color of the final product was an intermediate color of pink and yellow instead of yellow, which was consistent with the findings of Njiru et al. (2008) (results were not shown in Figure 4). This is presumably due to the complexity of the mushroom matrix, which contains many polysaccharides and polyphenols (Abubakar et al., 2018). Polyphenols can be oxidized after a long time, with irreversibly bound proteins and nucleic acids forming high molecular weight complexes (Newbury and Possingham, 1977). Polysaccharides can hinder the activity of polymerase and inhibit the amplification reaction of LAMP (Fang et al., 1992). Therefore, during detection, attention should be paid to controlling the content of the DNA template in the reaction system.

Visualization of LAMP amplification results is more conducive to rapid field detection. A variety of LAMP rapid visual inspection methods have been developed in recent years. Among them, the most commonly used methods are turbidity detection and color change (Tomita et al., 2008; Denschlag et al., 2013). The turbidity detection method is simple and cost-saving. However, this method has low detection sensitivity. When the concentration of the amplified product is low, the result is not always visible (Njiru et al., 2008). The detection method based on a color change has proven to be an effective alternative method. Many metal ion indicators, such as hydroxyl naphthol blue (HNB) and calcein, are used as indicative dyes for LAMP amplification (Pang et al., 2019; Chen et al., 2020; Wigmann et al., 2020). Whether the reaction has occurred can be determined according to the color change in the tubes. The reaction results are visible to the naked eye, and furthermore, there is no need to open the tubes during the reaction process, which can effectively avoid aerosol contamination and the probability of false positives (Goto et al., 2009). However, the LAMP amplification products are not obvious in terms of color changes with the HNB and calcein dyes. The WarmStart^® Colorimetric LAMP 2X Master Mix (New England Biolabs, United States) used in this study contains a visible pH indicator. During the LAMP reaction, the pH of the reaction solution tends to decrease because of the production of protons and the subsequent enhancement of DNA polymerase activity, such that the color of the solution changes from pink to yellow. The color change is obvious, and the reaction result is visible to the naked eye under daylight conditions, which is consistent with the results of Daskou et al. (2019). In addition, considering that researchers are affected by subjective preference, we put the reaction tube on white paper after amplification and used the white color of the paper as a control to ensure that the difference in color between positive and negative tubes became more pronounced. Furthermore, double-blind experiments were carried out in the laboratory, and all observers had no disagreement with the test results. The LAMP visual detection method established in this study will provide ideas for the further development of microfluidic chips and integrated on-site rapid detection methods.

For mushroom poisoning incidents, in the examination of clinical patients and forensic investigation, the detection of vomit, feces, and stomach contents is required. In order to ensure the feasibility of this method, we simulated the process of mushroom processing and digestion in human gastric juice. In this study, we successfully detected C. molybdites in the mixed mushrooms digested with artificial gastric juice. High temperature, pressure, and the digestion of gastric juice may cause some degree of genomic DNA degradation and fragmentation. Previous studies have confirmed that DNA degradation would not affect the detection of fragments shorter than 300 bp (Goldstein and Desalle, 2003). Gausterer et al. (2012) used direct PCR to investigate the stomach contents in a forensic case and successfully detected the digested yew plant. Others had proved that DNA was still present in traditional food processing styles, and it could be amplified in dried mushrooms and even be incubated by gastric juices (Gausterer et al., 2014). The LAMP method for C. molybdites detection established in this study had the same conclusion. In conclusion, our study provided an evidence that the LAMP assay could be used for the rapid and sensitive detection of C. molybdites in various matrices, including raw, boiled, and digested mushrooms with single or mixed species.

According to the results of this present study, molecular biology methods focused on nucleic acid are of great necessity to detect mushrooms of C. molybdites. More specifically, the LAMP method targeting the mitochondrial ITS gene was established for the on-site rapid detection of C. molybdites. High specificity and sensitivity (the LOD of detectable DNA was 1 pg) for the LAMP method were confirmed in the present study. Finally, the established method was successfully applied to the boiled and digested mushroom mixtures. These characteristics indicated that the LAMP assay for species-specific detection was useful for identifying C. molybdites in clinical treatment and forensic analysis.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AC conceived of and designed the experiments. ZZ and XZ provided some Chlorophyllum molybdites materials and identified the species. ET completed the morphological identification of the mushroom samples. HL and JZ completed the analysis of ITS DNA sequences of all mushrooms. NW, JG, WY, and RX carried out the LAMP assay. NW and AC wrote the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.