Loop-mediated isothermal amplification (LAMP) assay coupled with gold nanoparticles for colorimetric detection of Trichoderma spp. in Agaricus bisporuscultivation substrates

Cultivated edible mushrooms are a substantial nutritional source that holds economic significance in various regions across the world1,2. Asian countries are the largest mushroom producers contributing up to 76% of the global production (China alone produces about 35%), followed by Europe (17.2%) and United States (5.9%)3. The global market value of fresh mushrooms was estimated to US$ 38 billion in 20184, and is expected to increase in future due to the changes in population lifestyle and growing consumers’ food awareness. Mushrooms embody a rich source of healthy nutrients and are a staple in the human diet. They are characterized by a low-calorie content, are devoid of saturated fat and cholesterol, and encompass all essential amino acids. Apart from their nutritional value, the appeal of mushrooms lies in antioxidant activity and therapeutic properties. Additionally, their distinctive taste and unique texture make them an attractive choice for inclusion as a food ingredient or as a substitute for food additives5,6,7, both in traditional food products and in the rising field of alternative proteins.

Modern mushroom farming elicits a number of environmental impacts which can be moderated through the circular economy principles based on upcycling of organic waste in the cultivation process. Button mushrooms (Agaricus bisporus) as one of the most consumed species are cultivated using compost, an organic substrate created through a thermophilic microbial process involving crop residues and underutilized wood (or other organic waste), nitrogen-containing additives (typically poultry or horse manure), seed meal, or synthetic nitrogen sources like urea or ammonium nitrate, along with gypsum. During the fruit-bearing phase of button mushroom cultivation, it is essential to add a layer of peat, known as "casing soil," on the surface of the composted substrate. This practice enhances the conditions for optimal mushroom growth as illustrated in Fig. 1A.

Figure 1

Full size image

(A) Illustration of principle stages in the mushroom cultivation process. (B) Images of button mushroom-growing substrate contaminated with Trichoderma harzianum.

During the production of button mushrooms, the emergence of green mold in the nursery stands out as a prevalent issue. This disease, attributed to the rapid invasion of fungi of the genus Trichoderma in cultivation bags significantly hinders the fructification of mushrooms (Fig. 1B). When compost or casing are infected with an aggressive species, such as Trichoderma harzianum, no mushroom is produced on the infected area since Trichodermafeeds on an assortment of other fungi8 and utilizes the majority of available nutrients due to its high capability for nutrient uptake9. Trichoderma spores get introduced into mushroom-growing facilities through contaminated spawn, compost, casing soil and wood. The spread of green mold is facilitated by contaminated tools, substrate, and clothing of employees. Additionally, it can be transmitted through contaminated air and insect vectors, such as sciarid mushroom flies10. The early detection of Trichoderma spp. on the substrate for growing edible mushrooms including A. bisporus as well as other species susceptible to Trichoderma infection such as oyster mushroom (Pleurotus ostreatus) and shiitake (Lentinula edodes), is crucial11,12. Early detection is a key step because Trichoderma spp. becomes visible only after the formation of dark green spores (Fig. 1B), when it is already too late to prevent the edible mushroom yield loss. Trichoderma has been estimated to affect 10–20% of the total worldwide mushroom production every year, resulting not only in economic losses amounting to several billion dollars, but also in diminished food supplies13,14,15.

The main measures to minimize Trichoderma contamination in the mushroom farms rely on strict hygiene practices, implementing treatments with disinfectants, and applying fungicides. However, usage of fungicides is not possible in the case of organic production16,17,18. Therefore, Trichoderma infestation is particularly detrimental for organic mushroom farms, as they have limited options to counter it. One promising option may be new methods for early Trichoderma detection which would allow organic producers to exclude infected substrate from further use, before significant losses in mushroom yield occur. Recently we have reviewed methods for monitoring populations of Trichoderma in mushroom-farming conditions10. Briefly, traditional methods for detection and identification of Trichodermaspecies in soil are based on isolation techniques using selective media. Unfortunately, this approach is laborious and not adapted for the fastidious nature of some strains. Immunological tests are not adapted for soil analysis because isozyme analysis and serology are nonspecific at the isolate level for Trichoderma species. Finally, only a few studies have been published about PCR-based methods to target Trichoderma species19,20,21. Although PCR-based assays are sensitive and nucleic acid genetic markers can be highly specific, this approach may provide false-negative responses because of the sensitivity of DNA polymerase to the inhibitors present in soil and to low quantity of Trichoderma genomic DNA when isolated in the early phase of disease.

Given the extent of yield loss in mushroom production caused by the onset of green mold disease, there is an urgent need for a new method that will enable early, rapid, and specific detection of Trichoderma spp., especially during organic cultivation of button mushrooms. Here, we report a novel loop-mediated isothermal amplification (LAMP) assay coupled with gold nanoparticles (AuNPs) for highly sensitive detection of green mold in A. bisporuscultivation substrates (casing soil and compost). LAMP is an isothermal amplification technique that shows a higher degree of specificity and sensitivity compared to PCR due to the larger number of target-specific primers and exceptional resistance to contaminants that inhibit PCR reaction22,23. AuNPs were used to allow a colorimetric read-out of the LAMP reaction indicating the presence or absence of Trichoderma species. Due to its specificity and simplicity to perform, we expect the developed method of detection of green mold disease in its early phases to be of high value for all mushroom producers.

LAMP assay in naturally infected samples

Before conducting LAMP reactions, primer set specificity for the chosen target tef1 gene was tested in silico for potential cross-reaction with other species (Supplementary Data S1). The tef1 gene was chosen as a secondary fungal DNA barcode for Trichoderma species. Additionally, to confirm that the developed LAMP primers are genus-specific, primer specificity was tested in silico by using tef1 genes originated from different Trichodermaspecies (Supplementary Data S2). The position and direction of all LAMP primers within the tef1 gene are shown in Supplementary Fig. S1. The multiple sequence alignment in Supplementary Data S1 shows that there was no similarity with fungal and bacterial species widely represented in the soil materials such as compost and casing soil.

The specificity of the LAMP primer set was then experimentally validated using gDNAs extracted from T. harzianum (as a model for highly aggressive Trichoderma spp.) isolated from the substrate samples, non-target fungal strains, and soil-borne bacterium Bacillus subtilis. The conditions for LAMP detection of tef1 were optimized to complete in 30 min at 65 °C (Fig. 2A). The amplification was confirmed through positive bands in the 2% gel electrophoresis (Fig. 2B, Supplementary Fig. S8). No band was observed on negative controls confirming that positive LAMP reaction occurred exclusively when Trichoderma gDNA was used as the template. These LAMP conditions were applied for development of the LAMP-AuNPs colorimetric assay for Trichoderma detection.

Figure 2

Full size image

Validating the specificity of LAMP reaction for Trichoderma harzianum detection: (A) by using Genie® III instrument and (B) on 2% agarose gel. Concentrations of gDNAs extracted from Trichoderma harzianum, Aspergillus fumigatus, Alternaria alternata, Cladosporium allicinum and Bacillus subtilis were: 75 ng/µL, 72.30 ng/µL, 28.67 ng/µL, 18.66 ng/µL, 26.01 ng/µL, respectively.

One of the most critical parameters in detection assays is the sensitivity. To check the sensitivity of the LAMP method, LAMP primers were used to amplify the target T. harzianumgDNA from tenfold serial dilutions (Supplementary Fig. S7).

AuNPs production and characterization

Assays based on colloidal AuNPs allow a naked-eye visualization of the results because AuNPs color depends on their size, shape, and inter-particle distance24,25. AuNPs produced by the citrate reduction method were stabilized in solution of wine-red color (Fig. 3A). TEM observations showed that the AuNP morphology was uniform and spherical of about 20 nm diameter (Fig. 3B). Applying the multipole scattering theory on UV–Vis spectrum of synthetized AuNPs26 (Supplementary Fig. S2) the calculated particle size was 22.1 nm and their concentration 3.7 × 1012 NPs/mL (Supplementary Table S1). Moreover, particles had a quite narrow size distribution with a hydrodynamic diameter of about 20 ± 3 nm, as estimated from DLS measurements (Fig. 3C). Obtained AuNPs were stable at 4 °C for at least six months. The addition of 5 mM MgCl2 induced the color change of the solution to violet (Fig. 3A), due to the ionic strength-induced cutoff of electrostatic repulsive forces between nanoparticles causing their aggregation. AuNPs aggregation was confirmed by DLS measurements because the hydrodynamic diameter shifted to 170 ± 5 nm (Fig. 3C). Naked-eye observation of AuNPs was possible due to the phenomenon of local surface plasmon resonance27. The initial solution of a red color had plasmonic band at ∼520 nm, while the solution with aggregated AuNPs of a blue-to-violet color had plasmonic band at ∼630 nm (Fig. 3D). 

Figure 3

Full size image

Salt control aggregation of AuNPs. (A) Color differences between stable (no MgCl2 added) and aggregated AuNPs (5 mM MgCl2 added); (B) Transmission electron microscopy (TEM) of synthetized AuNPs; (C) Dynamic light scattering (DLS) of stabilized and aggregated AuNPs and (D) UV–Vis spectra of stabilized and aggregated AuNPs; (E) Schematic presentation of the colorimetric sensor coupled with loop-mediated isothermal nucleic acid amplification system.

Colorimetric detection of LAMP products

The principle of the LAMP-AuNPs assay is described in Fig. 3E. Initially, the LAMP products were incubated with colloidal AuNPs to enable their interaction. Oligonucleotides have a strong tendency to stably adsorb on the surface of AuNPs through electrostatic interactions, as previously described in reference25. Since the adsorption of amplicons on AuNPs protects nanoparticles from salt-induced aggregation, the addition of MgCl2 induced no color change in solution containing LAMP products. However, the color of AuNPs incubated with negative LAMP reaction, containing no amplicons, was sensitive to MgCl2 addition and became violet.

For the sensitive detection of LAMP products, AuNPs should be aggregated properly in response to a very small amount of salt. In a high salt condition, citrate-capped AuNPs readily aggregate and the number of amplicon molecules needed to suppress the aggregation is high. Therefore, to enhance the sensitivity of the colorimetric detection, we first optimized the concentration of MgCl2. The impact of salt concentration on the aggregation of AuNP was tested in 150 µL volume containing 2 × 109 NPs. This volume was chosen in order to adapt the assay to a multiplex 96-well microplate format. When 1, 5, or 10 μL of MgCl2 of varying molarities (2 mM, 20 mM, and 2 M) were added to the AuNP solutions, the agglomeration of gold nanoparticles was observed across all tested concentrations obtained with 20 mM MgCl2 stock solution i.e., from 0.13 mM to 1.33 mM final concentration (Fig. 4A). We, thus, proceeded to utilize 20 mM MgCl2 as a salt stock solution in subsequent tests with LAMP products. When AuNPs were preincubated with 2 µL of LAMP products (32 ng/µL in final concentration), 3 min after the addition of up to 0.4 mM MgCl2, the solution retained a red color, indicating the prevention of salt-induced aggregation due to the adsorption of amplicons onto AuNPs (Fig. 4B). Therefore 0.4 mM MgCl2 was chosen for the colorimetric sensor construction.

Figure 4

Full size image

Optimization of colorimetric detection of LAMP products. (A) Impact of salt concentration on the aggregation of AuNPs. (B) Impact of salt concentration on the aggregation of AuNPs hybridized with LAMP amplicons. (C) Optimization of incubation time and temperature to allow AuNP hybridization with LAMP amplicons. The numbers stand for concentrations of LAMP amplicons in ng/μL. Circled in red are chosen conditions.

Next, various temperatures and time conditions were tested to improve binding of the LAMP products to gold nanoparticles. The corresponding results, depicted in Fig. 4C, indicate that incubation at 65 °C for 10 min was the most optimal for the adsorption of LAMP products with AuNPs and was used in further experiments.

To assess the specificity of the LAMP-AuNP assay, AuNPs were incubated with a positive LAMP reaction obtained using gDNA of T. harzianum isolated from substrate used for A. bisporus cultivation. Negative LAMP reactions were performed without DNA template, or with non-specific gDNA from B. subtilis together with LAMP master mix (WarmStart LAMP Reagent containing deoxynucleotides dNTPs, Bst 2.0 WarmStart DNA Polymerase, isothermal amplification buffer, LAMP primers and PCR-grade water). As shown in Fig. 5, the color of solutions and spectrophotometric measurements indicated that only AuNPs incubated with positive LAMP were not aggregated upon MgCl2 addition. These results indicate that the color remained red in a specific way as a result of the binding between LAMP amplicons and AuNPs.

Figure 5

Full size image

Detection of LAMP products. (A) Absorption at 630 nm (A630) for the detection of the positive LAMP reaction and negative controls. (B) Absorbance spectra of the positive reaction and tested control coupled with AuNPs. Dashed lines present citrate-capped AuNPs (red) and salt-aggregated AuNPs (blue).

To determine the minimum detectable amount of the LAMP product by the AuNPs-based assay, tests were carried out with tenfold serial dilutions of positive LAMP reaction products obtained with T. harzianum gDNA. In parallel, serial dilutions of the no-template reaction and negative LAMP reaction with non-specific gDNA of B. subtilis were also tested. In case of the positive LAMP reaction a limit of detection of 24 ng/µL was observed by naked eye (Fig. 6A). At this amplicon dilution there was a clear difference in absorption intensity at 630 nm between the positive LAMP reaction and all tested negative controls (Fig. 6B). The comparison of signal intensities at A630 showed that the distinction between positive and negative LAMP reactions is not possible directly in crude samples but requires a ten-time dilution. This can be explained by the high concentration of reactants and primers in the LAMP reaction and master mix that prevent salt-induced AuNPs aggregation. In diluted samples the concentration of reagents and primers decreased and AuNPs aggregation is controlled only by the presence of amplicons. Figure 6C indicates that the LAMP-AuNPs assay was sensitive in solution containing amplicons at concentrations ranging from 240 to 24 ng/µL. Further dilution of samples overly decreased the number of amplicons and no suppression of AuNP aggregation was possible at the given salt concentration.

Figure 6

Full size image

Sensitivity of the LAMP-AuNP assays. (A) Color differences in positive (above) and negative (below) LAMP reactions detection using the colorimetric test at different dilutions. Circled in red are dilutions providing discriminations between positive and negative LAMP reactions. (B) The absorbance spectra of the AuNPs assay performed with two-time dilution of positive (red) and negative (blue) LAMP reaction; (C) The absorbance intensity at 630 nm (A630) of various dilutions of positive (red) and negative (blue) LAMP-AuNP assays.

Overall, to observe the color change from red (indicating a positive result—the presence of Trichoderma) to blue (indicating a negative reaction—the absence of Trichoderma) using the LAMP-AuNPs assay, it is necessary to test several dilutions at the same time, for which the 96-well plate was found to be suitable. It is also necessary to strictly follow good laboratory practice in order to avoid potential cross-contamination and/or false-positives. On this note, it is important to add that LAMP has been known for producing false-positives, however it has been shown that the longer the duration of reaction, the higher proportion of false-positives, and the earliest false-positives appeared after 45 min28. Therefore, the time-gating strategy we used (the assay duration is 30 min) should prevent the formation of false-positives. Further optimization may be needed to enable a proper translation of the developed LAMP-based assay to the fully POC-applicable kit.

In summary, to the best of our knowledge, there is currently no other publication describing LAMP-based assay for detection of Trichoderma on a genus-specific level, performed using the newly designed and validated LAMP primers, label-free AuNPs and direct visualization of the positive LAMP reaction, yielding detection sensitivity by LAMP of at least 917 fg/µL (Supplementary Fig. S7).

In this work, we combined the merit of LAMP and colorimetric monitoring of amplicons using AuNPs in a high-throughput format to detect Trichoderma. We demonstrated the efficiency of the assay on the soil-borne T. harzianum species isolated from substrates from organic button mushroom farming. The test shows potential for being routinely performed for screening mushroom substrates (compost and casing soil). However, since the test is based on salt-induced AuNPs aggregation and localized surface plasmon phenomenon, its working conditions depend strongly on the size and shape of used AuNPs25,29,30,31,32,33. Changing the type of AuNPs will demand a new optimization of salt-concentration and time of reaction to enable naked-eye detection. The colorimetric signal provides a simple and quick result acquisition independent of the number of samples present on the plate.

The portability and naked-eye visualization of results may help future development of a high-throughput sensitive Trichoderma spp. detection directly at the farm. The test can be easily utilized as a smartphone-based colorimetric test since many phone applications are available that can image and analyze 96 well plates (such as e.g. free app MyLight). This assay represents a good alternative to other analytical methods coupled with LAMP and reports an innovative concept for Trichoderma spp. detection.

Reagents and kits

Sodium citrate, gold (III) chloride solution (HAuCl4), GelRed Nucleic Acid Stain, and magnesium chloride (MgCl2) were purchased from Sigma-Aldrich (Saint-Quentin-Fallavier, France). Hydrogen chloride (HCl) and nitric acid (HNO3) was purchased from VWR (Strasbourg, France) and Merck (Saint-Quentin-Fallavier, France), respectively. LAMP primers were synthetized by Integrated DNA Technologies (Coralville, Iowa, USA) and resuspended in MilliQ water. Malt Agar was obtained from Torlak Institute of Immunology and Virology (Belgrade, Serbia), Rose Bengal Agar (RBA) was purchased from Merck (Darmstadt, Germany), and Tryptone Soya Agar (TSA) from Millipore (Burlington, USA).

The DNeasy Blood & Tissue Kit was purchased from Qiagen (Düsseldorf, Germany). Plant/Fungi DNA Isolation Kit was obtained from Norgen Biotek (Thorold, Canada), and GenElute™ Soil DNA Isolation Kit was purchased from Sigma-Aldrich (LLC, Germany). WarmStart LAMP Kit (DNA & RNA) was purchased from New England Biolabs (Massachusetts, USA).

Microorganisms and DNA extraction

The soil-borne T. harzianum was isolated through cultivation and co-cultivation process from the button mushroom substrate samples (compost/casing soil) kindly provided by a local organic mushroom producer in Serbia. Samples of compost and casing soil, from which a pure culture of T. harzianum was isolated, were collected from 10 cultivation bags at the conclusion of the button mushroom cultivation process. At this stage the highest concentration of Trichoderma spores in the substrate was expected. Samples for Trichoderma isolation were cultivated on the same day. After cultivation and co-cultivation, pure isolates were obtained. Initially, cultures were grown on RBA at 26 °C for 5 to 7 days. Subsequently, fungal cultures which have grown from the samples were transferred to Malt Agar and allowed to grow at 26 °C for another 5–7 days to obtain pure cultures. T. harzianumwas identified by sequencing (Novogene sequencing services) after DNA extraction from the pure cultures using GenElute™ Soil DNA Isolation Kit. Aspergillus fumigatus, Cladosporium allicinum, and Alternaria alternata were used as negative controls (Table 1). B. subtilis, used as a soil bacterial control strain, was cultivated in TSA at 37 °C overnight. All species and the strain used in this study are given in Table 1<