Cannabis Seeds Authentication by Chloroplast and Nuclear DNA Analysis Coupled with High-Resolution Melting Method for Quality Control Purposes

Abstract

Background:

Cannabis plants and their seed have been used in many cultures as a source of medicine and feeding during history. Today, there is an increasing demand for cannabis seeds for medical use. Moreover, a seed sales market with no legal regulations has also grown. This may pose some issues if a quality control is not set in place. Identification of cannabis strains is important for quality control purposes in a nonregulated growing market and in cases of illegal traffic and medical use. Owing to the high price as a pharmacological drug, commercial products of cannabis plants and seeds for medical users are often subjected to adulterations, either when packing or distributing certified seeds in the market.

Materials and Methods:

Cannabis commercial seeds and cannabis seeds for medical use were analyzed with high-resolution melting (HRM) analysis using barcoding markers. Humulus lupulus L. plants from a local market were used as outgroup control. DNA barcoding uses specific regions of the genome to identify differences in the genetic sequence of conserved regions such as internal transcribed spacer (ITS) and rbcL. DNA barcoding data can be generated with real-time polymerase chain reaction combined with HRM analysis to distinguish specific conserved DNA regions of closely related species. HRM analysis is the method of choice for rapid analysis of sequence variation.

Results:

The melting temperature (T_m) of homogeneous packages was consistent with single genotypes. However, packages containing contaminating seeds showed T_m differences of 0.2°C on average.

Conclusions:

An effective, rapid, and low-cost method based on ITS nuclear DNA and on chloroplast rbcL regions for screening and detection of contamination in commercial cannabis seeds was developed and applied for the analysis of different samples. This approach can be used as a quality control tool for cannabis seeds or other plant material.

Introduction

The medical properties of Cannabis species have been known for centuries, but the interest on active secondary metabolites as alternative therapies for diverse pathologies has grown in recent years.¹ This use of cannabis for therapeutic purposes has recently become the focus of different medical, political, and economic groups. This potential use has generated a strong impact worldwide concerning public health and the commerce of cannabis products. Indeed, this has led to legislation changes in some countries. Six countries have legalized cannabis for recreational use and 45 countries for medical use, and recently, The World Health Organization removed cannabis from the “Drug” category.

There is an estimate of 260 million cannabis consumers worldwide, who spend ∼350 billion USD annually on products associated with cannabis.² Technological advances and the open availability of new scientific studies have allowed a better understanding of the pharmacological use of cannabis. Genetic identification of different cannabis strains is critical not only for quality control purposes in a nonregulated growing market, but for forensic purposes and for medical use as well. Its medical use has been focused on the treatment of skin pathologies, seizures, lack of appetite, neurodegenerative diseases, multiple sclerosis, as well as for the management of chemotherapy side effects^3–6 and chronic pain treatment.⁷ Although medical cannabis is available in many countries, there is still some controversy about its psychoactive properties and the addiction to Δ-9-tetrahydrocannabinol (Δ-9-THC).⁸ Cannabis use can affect the endocannabinoid system, which is directly involved in brain development. This could produce cognitive disorders related to day-to-day consumption and especially in cases of long-term cannabis use.⁹ In this way, various seed banks have developed strains with low content of nonpsychoactive cannabinoids and high content of Cannabidiol (CBD) with potential therapeutic use. Moreover, several strains with different THC/CBD ratios have been developed.¹⁰ However, commercial cannabis plants and seeds are often subjected to adulterations during packaging. This poses a problem since these plant products require a certificate of authentication. Various methods of identification have been developed to authenticate medical plant species such as morphological analysis, chemical profiling, and DNA-based molecular analysis.¹¹ Although up to date there is no standardized method for rapid identification of cannabis, the use of internal transcribed spacer (ITS) regions has been recently reported for the identification of cannabis from seizures.¹² Forensic identification of cannabis usually consists on the detection of the presence of Δ-9-THC and cystolithic hairs on the leaves of the plant,¹³ but genetic identification has not been taken into account yet. The study of conserved regions of the genome for identification has allowed broadening the spectrum of analysis in a number of species. This has been applied on both the botanic and forensic fields. These molecular methods can detect differences at the DNA level and they offer numerous advantages over the conventional phenotype-based approaches.^14,15 Barcoding is the analysis of specific conserved regions of the genome to identify genetic sequences of different species. The term “DNA barcode” for global species identification was first coined by Hebert in 2003.¹⁶ DNA barcoding can be used for identification of species and also for differentiation between populations of the same species. The ITS2 region is one of the most used universal barcode for plant species identification.^14,17,18 Barcoding universal primer sets have been developed for studies of plants of medical use¹⁹ and the ITS2 region has been shown to distinguish a wide range of plant species.^14,20,21 DNA barcoding data can be generated with real-time polymerase chain reaction (PCR) combined with high-resolution melting (HRM) analysis to distinguish closely related species. HRM analysis is the method of choice for rapid analysis of sequence variation within PCR amplicons by determining their melting temperature (T_m).²² T_m depends on the base composition of amplicons. T_m differences can be detected by monitoring the fluorescence changes during PCR cycles using real-time PCR differential analysis.

Genotypes are then differentiated by their characteristic melting curves, visualized by the loss of fluorescence as the DNA duplex melts²³ HRM analysis allows genotyping of plant, fungus, and animal species. DNA sequence differences, such as single nucleotide polymorphisms and small insertions and deletions (Indels), can be detected based on the location of a differential peak and the shape of the melting transition curves.^15,24 The ITS2 region barcode marker was chosen because it is widely used for phylogenetic reconstructions at both genus and species levels.^25,26 In addition, the chloroplast ribulose bisphosphate carboxylase large chain gene (rbcL) is another barcode region used in most land plants due to the ease of amplification, the availability of databases^21,27–29 and its potential application for authenticity testing.³⁰ Several other cannabis chloroplast markers have been used for forensic purposes (e.g., to establish the geographical origin of seizures).¹⁸ Therefore, we proposed a method based on the ITS2 nuclear DNA region and rbcL, a chloroplast marker, combined with HRM-PCR, for detection of adulterants in cannabis seeds commercial strains. This article describes a case of adulteration of cannabis seeds due to lack of control during the packaging process. Cannabis sativa L. seeds from different sources were analyzed. The objective of this study was to authenticate cannabis strains using a real-time PCR and HRM-based molecular method. Using ITS2 and rbcL barcoding HRM, it was possible to detect the presence of different strains in commercial cannabis seed used for medicinal purposes. This method can be very valuable in terms of traceability and authentication of strains in cannabis commercial seeds.

Materials and Methods

Samples

Ten cannabis sample sets for medical use, containing five seeds each, were donated from collections of medicinal users (N=50). Commercial seeds of four different C. sativa L. strains, Lemon Haze, AK, Amnesia, and London Cheese, were purchased from a local coffee shop in southern Chile (N=12; control group). Commercial Hops (Humulus lupulus L.) leaves were used as an outgroup control.³¹

DNA extraction and quantification

DNA extraction from plant material (100 mg) was performed using the Qiagen DNeasy Plant mini kit, according to manufacturer's instructions (Qiagen, Hilden, Germany). Plant material was previously powderized before extraction. DNA was quantified using a fluorometer Qubit 2.0 following manufacturer's protocol (Thermo Fisher Scientific). Genomic DNA integrity was confirmed by agarose gel electrophoresis. DNA extracts were diluted to 20 ng/μL and stored at −20°C until further analysis.

PCR amplification

After DNA extraction, a PCR reaction was performed on a SimpliAmp thermocycler (Applied Biosystems) under the following conditions: 40 cycles of 94°C for 30 sec, 56°C for 30 sec, and 72°C for 30 sec, then a final extension step of 72°C for 2 min for ITS2 marker and an initial denaturation 95°C followed of 30 cycles of 55°C for 1 min, and 72°C for 1 min for the rbcL marker. Amplicon size and primer specificity were confirmed by 1.8% agarose gel electrophoresis.

Real-time PCR and HRM analysis

Real-time PCR reactions were conducted using the KAPA SYBR^® FAST qPCR Kit (Merck, Darmstadt, Germany) in a total volume of 15 μL on a Rotor-Gene Q Real-time Thermocycler (Qiagen). For ITS2 region, a PCR was performed according to the following conditions: an initial denaturing step of 94°C for 3 min followed by 40 cycles of 94°C for 30 sec, 56°C for 30 sec, and 72°C for 30 sec, then a final extension step of 72°C for 2 min. ITS2/ITS3 and rbcL forward and reverse primers were used at a concentration of 300 nM. ITS2F 5′-ATGCGATACTTGGTGTGAAT-3′ ITS3R 5′-GACGCTTCTCCAGACTACAAT-3′ rbcLa-F: 5′-ATGTCACCACAAACAGAGACTAAAGC-3′, rbcLa-R: 5′-GTAAAATCAAGTCCACCRCG-3′.^14,30 For rbcL, the amplification conditions were as follows: an initial denaturation at 95°C followed of 30 cycles of 55°C for 1 min, and 72°C for 1 min.³² The fluorescent data were acquired at the end of each extension step during PCR cycles. Before HRM, the products were denatured at 95°C for 5 sec, and then annealed at 50°C for 30 sec to randomly form DNA duplexes. HRM was performed as follows: pre-melt at specific melting temperature for each marker for 90 sec, and melt at a ramp of 10°C in a range of 0.1°C increments every 2 sec. The fluorescent data were acquired at the end of each increment step. End-point fluorescence level was acquired after the melting process by holding at 60°C for 5 min. All experiments were run in duplicate. A threshold cycle (Ct) parameter of 19±4 cycles was set at a threshold of 0.01 of the normalized fluorescence. Melting curve analysis was performed with the Rotor-Gene Q software v. 2.3.5 (Qiagen).

Statistical analysis

Melting Derivate Data for the ITS2 and rbcL were used for the comparative analysis. The first step was to calculate triangular similarity matrix based on Euclidean distance and then a multidimensional metric scaling (mMDS) was performed. This analysis determines the relationships between the profiles at different temperatures for each marker. To compare the response profiles, a Permanova test³³ was performed, which allows a comparison test (defined a priori), using permutation/randomization methods on a similarity matrix. A k-means clustering was used to detect classes through a set of quantitative variables.³⁴

Results

A nuclear region (ITS2) and a chloroplast (rbcL) were selected for the HRM analysis. DNA extracted from all seeds yielded a specific amplification product for both regions. PCR product sizes were determined to be 150 and 600 bp for ITS2 and rbcL regions, respectively. These sizes are in accordance to previously published sequence data. PCR amplicons were analyzed to determine their specific T_m. Results are represented by means of conventional derivative plots. The melting curve is generated by slowly melting the DNA through a range of temperatures in the presence of a dsDNA binding dye. Distinguishable melting curves were obtained for H. lupulus and the cannabis samples. T_ms resulted to be 88.8°C and 84.4°C for H lupulus and 87.9°C and 83.8°C for cannabis samples for ITS2 and rbcL, respectively (Fig. 1). All cannabis seeds provided by customers yielded different melting curves.

FIG. 1.

Bar-HRM on cannabis samples and control (Humulus lupulus L) using HRM analysis. (A) Melting curve profile using ITS region. (B) Normalized curve profile of PCR products of ITS region. (C) Melting curve profile using rbcL marker. (D) Normalized curve profile of PCR products of rbcL marker. HRM, high-resolution melting; ITS, internal transcribed spacer; PCR, polymerase chain reaction.

The analysis of ITS2 region yielded different outcomes for the four analyzed strains. Two of the strains resulted to have the same genetic profile (Fig. 2A, B) and the other two appeared to be genetically distinguishable (Fig. 2C, D).

FIG. 2.

HRM analysis using the ITS2 nuclear marker. (A) Melting curve profile of three cannabis seeds with same sequence. (B) Normalized curve profile of three cannabis seeds with same sequence. (C) Melting curve profile of two cannabis seeds with same sequence and one seed with different sequence. (D) Normalized curve profile of two cannabis seeds with same sequence and one seed with different sequence.

These differences depend on GC content, length of amplified product, and sequence. The undistinguishable seeds resulted to have a consistent Tm of 88.0°C. However, seeds with different genetic profiles resulted to have a ΔT_m > 0.2°C (87.9°C for sample 1°C vs. 88.1°C for samples 2 and 3) (Fig. 2).

Analysis of HRM using rbcL locus was equally powerful and discriminatory when compared with those of ITS2 region. ITS2 results were confirmed through rbcL analysis. Analysis of derivative melt curve resulted in consistent T_ms, within the range 83.7.0–84.1°C. Three cannabis seeds resulted to have the same rbcL sequence (Fig. 3A, B) for two strains and seeds from other two strains displayed different sequences (Fig. 3C, D).

FIG. 3.

HRM analysis using the rbcL chloroplast marker. (A) Melting curve profile of three cannabis seeds with same sequence. (B) Normalized curve profile of three cannabis seeds with same sequence. (C) Melting curve profile of two cannabis seeds with same sequence and one seed with different sequence. (D) Normalized curve profile of two cannabis seeds with same sequence and one seed with different sequence.

The mMDS analysis showed the relationships between the seeds' profiles at different temperatures for each gene (Fig. 4). Four sets of three seeds from different strains were included in the analysis. All seeds from sets #2 and #4 resulted to form a cluster, whereas seeds from sets #1 and #3 appeared to be genetically heterogeneous.

FIG. 4.

(A) mMDS represent seed samples of the four different packages used in the study. Each point represents a seed for ITS region. (B) mMDS represent seed samples of the four different packages used in the study. Each point represents a seed for rbcL marker. mMDS, multidimensional metric scaling.

Permanova analysis allowed to establish significant differences (p<0.05) and the clustering k-means allowed to assign samples from sets #2 and #4 to specific clusters. However, samples from sets #1 and #3 were not able to be assigned to the same cluster due to their genetic sequence difference.

These results demonstrated that seeds from sets #2 and #4 appeared to be genetically homogeneous, whereas samples from sets #1 and #3 seemed to be genetically heterogeneous. Results from sets #1 and #3 clearly indicated the presence of a mixture of seeds.

Discussion

We described a case in which 10 samples of seeds from medicinal users and four strains of cannabis seeds were analyzed by HRM coupled with barcoding. HRM analysis is a method that measures the dissociation rates of double-stranded DNA into single-stranded DNA at increasing temperatures.¹⁵ HRM involves accurate precise monitoring of fluorescence changes caused by the release of an intercalating DNA dye from double-stranded DNA during its denaturation caused by increased temperatures.²² These analyses allowed to determine the genetic homogeneity in the seeds contained in each package of a certain strain. This authentication is critical to ensuring the quality of medical cannabis seed. Various methods of identification have been developed to authenticate medical plant species such as morphological analysis, chemical profiling, and DNA-based molecular analysis. However, up to date there is no standardized method for rapid identification of cannabis.¹³ We used a nuclear and a chloroplast marker ITS and rbcL, respectively, as barcode regions to genotype these species.^14,18,19,21

DNA sequencing technology is relatively expensive and time consuming. Thus, Bar-HRM is a better option for a rapid screening of cannabis strains. However, HRM shows some limitations when the genetic variation is extremely small.³⁵ The Bar-HRM analysis using two loci was performed in one “closed” tube and the results were available for analysis at the end of the run. Four strains were purchased from a specialized store that distributes cannabis seeds for recreational and medicinal use. Molecular authentication showed that cannabis seed are good sources of DNA, providing full and amplifiable genetic material that can be used efficiently for authentication analysis by HRM. Distinction down to genus and—in many cases—species level is possible based on melting temperatures (T_m) of specific PCR products.³⁶ In this study, the application of HRM using ITS and rbcL regions allowed the authentication and discrimination between cannabis genotypes. The ITS2 region was selected as a barcode marker because is widely used for phylogenetic reconstructions at both genus and species levels.^25,26,37 In addition, rbcL is another barcode region used in most land plants due to the ease of amplification and availability of databases.^21,27 We have used two genetic markers to authenticate the contents of four packages of cannabis seeds for medical purposes. The use of a combination of barcode regions is common and recommended for plant identification.³⁸ In the context of authenticating commercial cannabis seeds, our results are the basis for initiating new explorations of chloroplastic and nuclear regions.³⁹ The aforementioned obtain highly decisive information taking into account variable sites and additional loci. To confirm that each package contained seeds of the same strain, we decided to analyze randomly purchased commercial products. In some cases, some cannabis users reported that different phenotypes were observed from seeds purchased as belonging to a single strain. The robustness of such analyzes would also be strengthened by the exploration of markers in organelles and by increasing the universe of samples to be incorporated in the studies.⁴⁰ New seed banks and an increased number of cannabis seed users force the scientific community to increase its efforts to support the authenticity of the material. Our results proved that seeds had different genetic sequences. This possible adulteration may be attributed to poor packaging or a mistake during package processing. Each package of each strain contains three seeds each. Surprisingly, it was found that two out of four packages contained mixtures of seeds. We proposed a fast easy low-cost method to distinguish the presence of mixtures in seed packages. This can certainly help confirm the authenticity or at least the homogeneity of the seed contents of a given strain. Although this analysis can be done at any stage of plant development, it was demonstrated that seeds were a good source of amplifiable DNA. The homogeneous packages had the same T_m for each of the three seeds of that strain. In contrast, packages containing mixed seeds displayed different T_ms. This allowed distinguishing the different genotypes in the same package. Analysis of HRM using rbcL locus was consistent with that of the ITS region.

Conclusion

In this study, Bar-HRM analysis was proposed to be a fast and accurate technique for authentication testing of cannabis seed of commercial strains. Here, we describe the development of a Bar-HRM method for adulteration testing of cannabis seeds contained in individual packages. These packages correspond to different cannabis strains. This method allowed detecting heterogeneity in two out of four cases in the contents of these packages. This is undoubtedly important at the time of consumption of these plants by medicinal users. In future studies, we will continue to explore the different seed banks commonly used by users of cannabis and incorporating new DNA markers.

Abbreviations Used

Δ-9-THC

Δ-9-tetrahydrocannabinol

CBD

cannabidiol

HRM

high-resolution melting

ITS

internal transcribed spacer

mMDS

multidimensional metric scaling

PCR

polymerase chain reaction

rbcL

ribulose bisphosphate carboxylase large chain gene

Author Disclosure Statement

We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this study that could have influenced the results.

Funding Information

This study was partially funded by #2020EM-LA-03. Dirección de Investigación, Vicerrectoria de Investigación y Posgrado. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the article.

Cite this article as: Anabalón L, Solano J, Encina-Montoya F, Bustos M, Figueroa A, Gangitano D (2022) Cannabis seeds authentication by chloroplast and nuclear DNA analysis coupled with high-resolution melting method for quality control purposes, Cannabis and Cannabinoid Research 7:4, 548–556, DOI: 10.1089/can.2020.0168.