Introduction
Palmer amaranth (Amaranthus palmeri S. Watson) is regulated as a noxious weed seed in many states, including Minnesota, Iowa, Wisconsin, North Dakota, South Dakota, and Ohio (USDA 2023), making it illegal to sell seeds that are contaminated with this species. Federal and Minnesota risk assessments detail the threat that this species poses to crop production throughout the United States and the need for regulation to prevent widespread establishment in the upper Midwest (MDA NWAC 2014; USDA 2020). Yu et al. (Reference Yu, Blair, Hardel, Chandler, Thiede, Cortilet, Gunsolus and Becker2021) reviewed the regulatory and management actions taken by the Minnesota Department of Agriculture (MDA) to prevent the establishment of this species in Minnesota. That review highlights the pathways for introduction, including seed, screenings, animal feed, and manure in Minnesota. A current map of the infestations in Minnesota, their status, and their sources is available on the MDA’s website (https://www.mda.state.mn.us/plants-insects/palmer-amaranth-location-sources).
Based on A. palmeri’s listing as a prohibited noxious weed seed in Minnesota, it is not legal to sell seeds, screenings, or animal feed that contain A. palmeri in Minnesota. The Association of Official Seed Analysts, Society of Commercial Seed Technologists, and MDA require that if Amaranthus seeds are identified in a seed lot during required federal testing, a warning is added to the seed label. Due to the regulations prohibiting A. palmeri in Minnesota, MDA requires further genetic testing of sampled seed lots if Amaranthus seeds are identified to determine the Amaranthus species. The MDA collaborated with the California Department of Food and Agriculture and the company Eurofins BioDiagnostics in 2017 to develop and test a method to identify individual Amaranthus seeds to the species level using polymerase chain reaction (PCR) and Sanger sequencing of the internal transcribed spacer region (ITS) (Price Reference Price2016; Yu et al. Reference Yu, Blair, Hardel, Chandler, Thiede, Cortilet, Gunsolus and Becker2021). More rapid and less expensive methods employing real-time polymerase chain reaction (qPCR) were published shortly afterward (Murphy et al. Reference Murphy, Plewa, Phillippi, Bissonnette and Tranel2017).
DNA-based tests to identify A. palmeri from among other Amaranthus species have been developed for individual plants or seeds. These include a restriction fragment length polymorphism assay (Wetzel et al. Reference Wetzel, Horak and Skinner1999) and a PCR-based assay (Wright et al. Reference Wright, Molin and Nandula2016). Some commercial and government labs initially sequenced the ITS gene region of individual seeds and compared the results to GenBank databases to differentiate A. palmeri from other Amaranthus spp. (Yu et al. Reference Yu, Blair, Hardel, Chandler, Thiede, Cortilet, Gunsolus and Becker2021). The development of a fluorescence-based qPCR assay for A. palmeri identification facilitated a faster sample testing turnaround time and opened up the possibility of testing groups (also called pools) of seeds up to 100 seeds (Murphy et al. Reference Murphy, Plewa, Phillippi, Bissonnette and Tranel2017). Competitive allele-specific PCR (KASP) has also been developed but requires different reagents and data analysis than the qPCR assay (Brusa et al. Reference Brusa, Patterson, Gaines, Dorn, Westra, Sparks and Wyse2021; Oliveira et al. Reference Oliveira, Gaines, Patterson, Jhala, Irmak, Amundsen and Knezevic2018).
This paper describes the MDA’s optimization and validation of a published qPCR genetic testing method to detect A. palmeri in individual or groups of Amaranthus spp. seeds. We evaluated the specificity of three markers for 17 non–Amaranthus palmeri species and determined their sensitivity and limit of detection in single seeds and pools of seed. We propose a sequential testing method wherein an initial marker is used to screen samples, and follow-up testing using an additional marker is done to confirm a positive result. This sequential approach lowers the risk of false positives using a qPCR approach.
Materials and Methods
Seed Source and Species Validation
The seeds of Amaranthus spp. were obtained from the U.S. Department of Agriculture–Agricultural Research Service (USDA-ARS) Germplasm Resources Information Network (GRIN) (USDA-ARS 2023). These seeds included: tumble pigweed (Amaranthus albus L.); sandhill amaranth (Amaranthus. arenicola I.M. Johnst.); southern amaranth [Amaranthus australis (A. Gray) Sauer]; prostrate pigweed (Amaranthus blitoides S. Watson); California amaranth [Amaranthus californicus (Moq.) S. Watson]; low amaranth (Amaranthus deflexus L.); spleen amaranth (Amaranthus dubius Mart. ex Thell.); hybrid amaranth (Amaranthus L. hybrid; 4 accessions); smooth pigweed (Amaranthus hybridus L.); Prince-of-Wales feather (Amaranthus hypochondriacus L.); A. palmeri; Powell’s amaranth [Amaranthus powellii S. Watson ssp. bouchonii (Thell.) Costea & Carretero]; redroot pigweed (Amaranthus retroflexus L.); spiny amaranth (Amaranthus spinosus L.); waterhemp [Amaranthus tuberculatus (Moq.) Sauer]; Watson’s amaranth (Amaranthus watsonii Standl.); and Wright’s amaranth (Amaranthus wrightii S. Watson) (Supplementary Table S1).
To evaluate marker sensitivity in single-seed and pooled-seed matrices, we used seeds from a single A. palmeri plant phenotypically identified by MDA scientists in the field. We confirmed that the seeds from this plant were positive for A. palmeri by testing 10 seeds individually at an external lab (Eurofins BioDiagnostics, River Falls, WI, USA). Amaranthus palmeri–negative seeds used for seed-pool testing were confirmed to not contain A. palmeri through testing of six sets of 100 random seeds by an external lab (University of Illinois Plant Clinic, Urbana, IL, USA).
Nucleic Acid Extraction
Seeds were added to a Lysing Matrix A tube with a second ceramic sphere added (MP Biomedicals, Santa Ana, CA, USA). Tubes were processed for 2 min using a Mini-Beadbeater™-16 (BioSpec Products, Bartlesville, OK, USA), and DNA was subsequently extracted using a MagMAX™ Plant DNA kit (Fisher Scientific, Hanover Park, IL, USA). Lysis buffer A contained 40 mM molecular biology grade dithiothreitol (Fisher Scientific). DNA was purified manually following the manufacturer’s directions and eluted with 50 µl of elution buffer (1 to 50 seeds) or 100 µl of elution buffer (50 to 100 seeds).
DNA was quantified using a Nanodrop™ 2000c Spectrophotometer (Fisher Scientific). Samples were diluted to 10 ng µl−1 with elution buffer immediately before qPCR reactions were run. Positive control dilution series were made by serial dilution of A. palmeri DNA to 10, 2, 0.08, 0.016, and 0.0032 ng µl−1 using elution buffer immediately before qPCR reactions were run. DNA was frozen, undiluted, at −20 C if not used the same day.
Oligos
Custom DNA oligos, including primers designed by Murphy et al. (Reference Murphy, Plewa, Phillippi, Bissonnette and Tranel2017) (Murphy forward and reverse) and KASP markers with the FAM or HEX tail removed from Oliveira et al. (Reference Oliveira, Gaines, Patterson, Jhala, Irmak, Amundsen and Knezevic2018) (A. palmeri Marker 1 [M1] forward and reverse) and Brusa et al. (Reference Brusa, Patterson, Gaines, Dorn, Westra, Sparks and Wyse2021) (A. palmeri Markers 2 (M2) and 3 (M3) forward and reverse) were synthesized by IDT (Integrated DNA Technologies, Coralville, Iowa, USA).
qPCR conditions
Twenty-microliter reactions containing 10 µl of 1X Applied Biosystems™ PowerUP™ SYBR™ Green 2X Master Mix (Fisher Scientific) were run for each set of markers. Unless defined otherwise, samples and controls were run in triplicate on an ABI QuantStudio™ 3 Real-Time PCR System (Fisher Scientific). The resulting data were analyzed using the QuantStudio Design and Analysis Software v. 1.2x (Life Technologies, Carlsbad, CA, USA).
Other Marker Cycle Optimization
Markers 1, 2, and 3 were tested for use in an SYBR green qPCR assay. Amaranthus palmeri, A. dubius, and A. spinosus were run in single reactions on an ABI QuantStudio™ 3 Real-Time PCR System (Fisher Scientific) using a standard cycle consisting of 50 C for 2 min, 95 C for 2 min, followed by 35 cycles of 95 C for 1 s, 52/54/56/58/60/62 C for 30 s.
Following the manufacturer’s recommendations (Applied Biosystems™ PowerUP™ SYBR™ Green 2X Master Mix), M2 and M3 primers were tested at varied concentrations. Duplicates of no template (water), 10 ng of A. palmeri DNA, and 2 ng of A. palmeri DNA were tested in a 20-µl reaction with 1 µl of DNA (or water), 10 µl of 1X Applied Biosystems™ PowerUP™ SYBR™ Green 2X Master Mix (Fisher Scientific), primers at 300, 500, or 800 nM, and molecular-grade water to 20 µl.
Optimized qPCR reaction conditions for M2 are 800 nM forward and reverse primers, while M3 uses 800 nM forward primer and 500 nM reverse primer.
qPCR cycles
The Murphy standard cycle consisted of 50 C for 2 min, 95 C for 2 min, followed by 35 cycles of 95 C for 15 s, 59 C for 1 min. The Murphy fast cycle consisted of 50 C for 2 min, 95 C for 2 min, followed by 35 cycles of 95 C for 1 s, 59 C for 30 s.
The M2 standard cycle consisted of 50 C for 2 min, 95 C for 2 min, followed by 40 cycles of 95 C for 15 s, 52 C for 15 s, 72 C for 1 min. The M2 fast cycle consisted of 50 C for 2 min, 95 C for 2 min, followed by 40 cycles of 95 C for 1 s, 52 C for 30 s.
The M3 fast cycle consisted of 50 C for 2 min, 95 C for 2 min, followed by 40 cycles of 95 C for 1 s, 54 C for 30 s. The M3 regular cycle consisted of 50 C for 2 min, 95 C for 2 min, followed by 45 cycles of 95 C for 15 s, 54 C for 15 s, 72 C for 1 min.
Seed-Pool Testing
Eighty-three blind samples were prepared for seed-pool testing, with the analyst running the assay not knowing which samples contained A. palmeri. Seed-pool sizes consisted of single seeds (20 samples), 25 seeds (20 samples), 50 seeds (22 samples), and 100 seeds (21 samples). Pools were spiked with 0 to 100 A. palmeri seeds (Table 1). The Murphy fast cycle was used to test all 83 seed-pool samples, while the M3 fast qPCR cycle was used to test 4 single-seed pools and 6 pools of 25, 50, and 100 seeds. For each sample, 10 ng of DNA was tested in triplicate.
Table 1. Composition of seed pools tested by real-time polymerase chain reaction (qPCR) includes individual (single) seeds and pools of 25, 50, or 100 seeds a
a The replicates per pool size (number of samples) and number of A. palmeri seeds used in each pool are included.
Limit of Detection
The detection limits of both the fast and regular cycles were tested for Murphy, M2, and M3 primers with the standard and fast cycles using 45 cycles and qPCR reaction conditions, both listed earlier. Amaranthus palmeri DNA was serially diluted using MagMAX™ Plant DNA kit elution buffer. Final concentrations tested included 10, 1, 0.05, 0.0125, 0.00625, 0.003125, and 0.0015625 ng µl−1. Concentrations were tested in duplicate wells for the six cycles described.
Primer Species Specificity
DNA was extracted from 5 seeds from each species of Amaranthus as described earlier and was run in triplicate wells for each set of primers. The qPCR cycles were completed using Murphy regular, Murphy fast, Brusa M2 regular, and Brusa M3 fast cycles.
ITS Sequence Analysis
The four accessions of hybrid Amaranthus and A. watsonii were further analyzed by amplifying the ITS1-5.8S rDNA-ITS2 region with ITS1 and ITS4 primers as previously described (White et al. Reference White, Bruns, Lee, Taylor, Innis, Gelfand, Sninsky and White1990) using Phusion green high-fidelity DNA polymerase (Fisher Scientific). PCR products were purified (PureLink™ PCR Purification Kit, Fisher Scientific) and sequenced bidirectionally (ACGT, Germantown, MD, USA). Geneious Prime® 2023.0.4 bioinformatics software was used to trim and assemble sequences. The four hybrid Amaranthus sequences and the A. watsonii sequences were submitted to GenBank.
The ITS region of all Amaranthus species used in this study were obtained from GenBank or from this study (four hybrid Amaranthus and A. watsonii). These regions were manually aligned to identify the primer binding sites for the Murphy primers.
Results and Discussion
Nucleic Acid Extraction
DNA extraction from Amaranthus spp. seeds was optimized for use with seed samples that contained between 1 and 100 seeds. Due to the hardness and small size of Amaranthus spp. seeds (0.9- to 1.7-mm diameter), Murphy et al. (Reference Murphy, Plewa, Phillippi, Bissonnette and Tranel2017) utilized an overnight incubation step to soften the seed coats and repeated the homogenization step to enhance tissue disruption to improve DNA extraction (Stallknecht and Schulz-Schaeffer Reference Stallknecht, Schulz-Schaeffer, Janick and Simon1993). We aimed to expedite testing turnaround time to complete extraction in 1 d. The addition of an extra ceramic bead to the homogenization tube to evenly grind all seeds without incubation in lysis buffer overnight was examined and shown to effectively disrupt the seed tissue for extraction of DNA. Nucleic acid concentrations for the different seed-pool sizes (1 to 100) showed consistent DNA concentrations for 1 and 25 seed-pool sizes, while larger seed-pool sizes had a greater concentration range (Figure 1). The differences in nucleic acid concentrations may be due to the range of seed sizes contained in different pools. Larger seeds would have more DNA available for purification.
Figure 1. DNA concentration (in ng µl−1) extracted from pools of Amaranthus seeds ranging in size from 1 to 100 seeds. The average DNA concentration extracted, maximum DNA concentration extracted, and minimum DNA concentration extracted are shown. Error bars represent the standard deviation of the mean (SE).
The variation in seed-pool size (1 to 100) and total DNA extracted from samples (8.7 to 328.2 ng µl−1) necessitated the application of a standard template concentration for use in qPCR. A concentration of 10 ng of template DNA was used for all samples and controls. Samples with higher DNA concentrations were diluted for use in qPCR testing.
The number of Amaranthus spp. seeds that are identified in seed noxious weed tests or screening exams can range from one to thousands of seeds (Yu et al Reference Yu, Blair, Hardel, Chandler, Thiede, Cortilet, Gunsolus and Becker2021). Elimination of an overnight incubation step before DNA extraction improves the testing lab’s flexibility and reduces turnaround time. Due to the sensitivity of qPCR, it is possible that seed-pool sizes larger than 100 seeds could also be tested using this method. Additional testing would be required to demonstrate whether this is feasible and what the maximum seed-pool size would be for accurate testing. The maximum seed-pool size is recommended to be 100 seeds until additional work is completed to show larger seed-pool sizes still efficiently detect A. palmeri.
Other Marker Cycle Optimization
Although there is a strong interest in detecting different Amaranthus species with a molecular method, the assay developed by Murphy et al. (Reference Murphy, Plewa, Phillippi, Bissonnette and Tranel2017) is the only qPCR method that has been published and widely adopted. Other molecular methods like KASP have been published (Brusa et al. Reference Brusa, Patterson, Gaines, Dorn, Westra, Sparks and Wyse2021). Based on the specificity results from the Murphy marker (see Primer Species Specificity section), we wanted to determine whether additional markers may be needed in regulatory testing. Three additional pairs of primers (M1, M2, M3) were modified for use in qPCR and tested to determine whether any of them would be suitable for use. Initial testing to determine the optimal annealing temperature showed that M1 did not work in an SYBR green qPCR, likely due to the very short target size, and it was not tested further (data not shown). M2 had the most discrete difference in quantification cycle (Cq) values between A. palmeri and other Amaranthus spp. using an annealing temp of 52 C, while M3 was best at 54 C (Supplementary Figures S1 and S2). The optimal primer concentrations for both M2 and M3 were evaluated, and M2 had the lowest Cq values with 800 nM forward and reverse primer, while M3 had the lowest Cq values with 800 nM forward primer and 500 nM reverse primer (Supplementary Table S2).
Seed-Pool Testing
The Murphy standard qPCR cycle was used to test all 83 seed pools and accurately detected the presence of A. palmeri in all seed pools from 1 to 100 seeds (Figure 2A). Seed pools of all sizes that contained 1 or more A. palmeri seeds consistently had Cq values less than 25. DNA extracted from pure A. palmeri samples resulted in the lowest Cq values. Amaranthus palmeri–positive samples ranged in Cq values from 16.2 (50 A. palmeri seeds) to 24.2 (1 A. palmeri seed in a 50-seed pool). Amaranthus palmer–negative samples had Cq values above 34.8, with many samples having a value of undetermined (35 cycles were run).
Figure 2. Average quantification cycle (Cq) values for Murphy and M3 real-time polymerase chain reaction (qPCR) with various seed-pool sizes. The average Cq values for Murphy qPCR (A) and M3 qPCR (B) are shown. Seed pools were composed of 1, 25, 50, or 100 Amaranthus seeds. The seeds pools contained 0, 1, 2, 25, 50, or 100 Amaranthus palmeri seeds (y axis). The Murphy cycle included 35 cycles, while M3 included 40 cycles. Error bars represent the standard deviation of the mean (SE).
The pooled-seed testing validation was also completed using M3. The M3 marker was tested using a subset of the 83 samples, including 4 single-seed pools and 6 pools of 25, 50, and 100 seeds containing 0 to 2 A. palmeri seeds (Figure 2B). Amaranthus palmeri–positive samples had Cq values less than 26.35, while negative samples had Cq values higher than 33.6.
Limit of Detection
The limit of detection was evaluated for Murphy, M2, and M3 primers. Both regular and fast qPCR cycles were tested. The Murphy fast cycle was the most sensitive cycle, yielding the lowest Cq values for all A. palmeri DNA concentrations tested (Figure 3). The Murphy regular cycle had lower Cq values than both M2 and M3 cycles. The M3 fast cycle had very similar Cq values to the M2 regular cycle (Figure 3). Although M2 and M3 yielded higher Cq values for A. palmeri–positive samples, it would still be feasible to use them for diagnostic testing using a higher Cq level for positive samples than is used for the Murphy qPCR test.
Figure 3. Limit of detection for Murphy, M2, and M3 real-time polymerase chain reaction cycles. Amaranthus palmeri DNA ranging in concentration from 0.0015625 to 10 ng µl−1 was tested using the regular and fast qPCR cycles for Murphy, M2, and M3. Error bars represent the standard deviation of the mean (SE).
Primer Species Specificity
We tested primer specificity using DNA extracted from seeds of 16 Amaranthus spp. (including two A. palmeri samples) and four Amaranthus hybrids (Figure 4). Both the regular and fast Murphy cycles detected A. palmeri with the lowest Cq values across the species tested, with Cq values of 15.61 and 14.84 (regular cycle) or 19.46 and 18.91 (fast cycle). However, both cycles also had amplification of A. dubius, A. spinosus, and A. watsonii. Amaranthus watsonii resulted in the lowest Cq values of 17.49 (regular cycle) and 21.24 (fast cycle), while A. spinosus had higher Cq values of 23.58 (regular cycle) and 27.85 (fast cycle). Similar to A. spinosus, A. dubius also had higher Cq values of 23.10 (regular cycle) and 26.86 (fast cycle). The sole use of the Murphy et al. primer may result in a false positive if A. dubius, A. spinosus, or A. watsonii is present in the sample, depending on the testing lab’s criteria for identifying positives. None of the four hybrids were amplified in either Murphy cycle. The fast cycle has a higher Cq value for A. dubius and could eliminate the potential false positive caused by that species. The strong amplification of A. watsonii, with Cq values similar to A. palmeri, suggests homologous ribosomal RNA single-nucleotide polymorphisms (SNPs) between A. watsonii and A. palmeri. Although the Murphy assay cannot differentiate between A. watsonii and A. palmeri, neither can the M2 or M3 primers.
Figure 4. Real-time polymerase chain reaction (qPCR) specificity to detect Amaranthus palmeri. The average quantification cycle (Cq) results for the Amaranthus species tested using Murphy, M2, and M3 are shown. Two cycle settings are possible using the QuantStudio 3 real-time PCR system and include the regular and fast cycles. The Murphy regular and fast cycles are presented, while only the optimized M2 regular and M3 fast cycles are presented. Error bars represent the standard deviation of the mean (SE). All data points had SE calculated, and points that appear to be missing them had very low error that does not extend beyond the data point’s visual.
The M2 regular and M3 fast cycles were tested against the same species. In addition to the Murphy assay, M2 and M3 also showed the highest non-target species amplification of A. watsonii. M3 had Cq values of 21.79 and 20.27 for A. palmeri and a Cq of 23.56 for A. watsonii. M2 had higher Cq values for A. palmeri of 25.20 and 23.88 with a similar spread to the Cq value of A. watsonii of 27.16. M2 also amplified A. arenicola, with a Cq of 27.42, which is a widely distributed native Amaranthus spp. in the United States. The amplification of A. arenicola limits the effectiveness of M2 for use in an SYBR green qPCR assay as a complementary test to the Murphy assay. The complete genome sequences are not available for either species, but it is likely that these Amaranthus species share sequence homology with A. palmeri in the DNA region the assays were designed to detect. The M3 marker may be the best additional marker for use in a secondary qPCR assay based on the amplification of only A. watsonii, low A. palmeri detection limit, and accurate detection of A. palmeri in pools of seeds. To implement the most risk averse strategy and limit the risk of false positives in our testing, we decided to take a sequential approach to testing and confirm any positive result identified by the Murphy marker with a follow-up test using the M3 marker. This approach should minimize the risk of a false positive in testing. Confirming a positive result using two independent genetic markers will limit an erroneous result in a regulatory setting.
Future studies may wish to explore the feasibility of utilizing both the Murphy marker and the M3 marker in a probe-based Taqman qPCR assay. Testing both markers simultaneously in one assay would allow for checking for the presence of two independent genetic markers in this group of closely related plants. Additional A. palmeri or other Amaranthus species–specific tests developed should include as many Amaranthus species as possible in the assay specificity testing, as GenBank does not contain complete genome sequences for this family. Analysis by computational simulations may not detect regions of high genetic homology between Amaranthus species, which could lead to false-positive results when the assay is deployed for real-world use.
ITS Sequence Analysis
The sequences of the ITS1-5.8S rDNA-ITS2 region corresponding to the region amplified by the Murphy qPCR assay were available for 16 out of 17 Amaranthus species tested. The sequences of the four hybrids were not available, and the deposited A. watsonii sequence had ambiguities in the forward primer binding site. Because the A. watsonii sequence for the ITS1-5.8S rDNA-ITS2 region was not available, we sequenced our accession and the four hybrid Amaranthus. The sequences were deposited in GenBank as: OQ996211.1 (hybrid Amaranthus), OQ996212.1 (hybrid Amaranthus), OQ996213.1 (hybrid Amaranthus), OQ996214.1 (hybrid Amaranthus), and OQ996215.1 (A. watsonii).
Analysis of the forward primer binding site for all Amaranthus species tested revealed that A. watsonii contained the A. palmeri specific SNP the Murphy primer was designed to target (Table 2). Amaranthus watsonii and A. palmeri have an identical sequence in the forward primer binding site, and alignment of the sequences showed these two species are identical for the entire gene region. The other species that exhibited amplification with the Murphy assay, A. spinosus, has one mismatch with the primer of two nucleotides in the middle of the primer binding site and one nucleotide mismatch in the SNP binding region of the primer, substituting a guanosine for a cytosine. This is likely the reason that A. spinosus amplifies with the Murphy assay, as both ends of the primer can bind and provide a template for qPCR replication. In addition to A. spinosus, A. dubius is also able to amplify with the Murphy assay. It is unknown why A. dubius is amplified with the Murphy assay, as the forward primer binding site has the same sequence as other species that do not amplify; it may be possible A. dubius contains some genetic diversity in that region, and the seed isolate we tested contains a different genetic sequence in this region than the one deposited in GenBank. It is also possible that the A. dubius accession in GRIN is not a pure species isolate and may be seed mix or hybridized with other species. Although A. dubius and A. spinosus amplify with the Murphy regular qPCR cycle, using the fast qPCR cycle decreases the amplification of these two species and would eliminate them as potential false positives. Amaranthus dubius and A. watsonii were not tested in the initial Murphy paper, so it is not unexpected to identify additional species with some level of cross-reactivity.
Table 2. Alignment of the Murphy forward primer with Amaranthus species tested
a Base-pair mismatches are presented in bold, underlined font.
In addition to amplifying with the Murphy assay targeting an ITS SNP, A. watsonii also amplifies strongly with M2 and M3. These additional markers targeted SNPs from different genomic regions, indicating A. watsonii and A. palmeri are very closely related. A review of these species shows that A. palmeri and A. watsonii have been described as sister species based on morphology since 1914 and have recently been a part of several sequencing-based phylogenetic studies of the Amaranthus genus (Brenan Reference Brenan1961; Raiyemo et al. Reference Raiyemo, Bobadilla and Tranel2023; Standley Reference Standley1914; Waselkov et al. Reference Waselkov, Boleda and Olsen2018). Analysis of six molecular markers across 58 Amaranthus species indicated the close genetic relatedness between A. palmeri and A. watsonii may be due to a chloroplast capture event and incomplete lineage sorting that are not observed in the greater Hybridus clade (Waselkov et al. Reference Waselkov, Boleda and Olsen2018). Based on their analysis of the male-specific region of the Y chromosome (MSY), Raiyemo et al. (Reference Raiyemo, Bobadilla and Tranel2023) propose two origins of dioecy for the dioecious Amaranthus spp. and suggest A. palmeri and A. watsonii have a different origin than the rest of the clade. The strong genetic homology observed between A. palmeri and A. watsonii is supported by their overlapping native habitat in arid regions of the southwestern United States and northern Mexico (Sauer Reference Sauer1955). Our sequencing of the A. watsonii ITS1-5.8S rDNA-ITS2 gene region revealed an identical sequence to A. palmeri and demonstrated the high genetic relatedness of the ribosomal nucleic acid between these species that could not be discerned by the Murphy, M2, or M3 qPCR assays tested.
Although A. watsonii would be a potential false positive for the Murphy, M2, and M3 qPCR assay, it currently is not known to be distributed outside its small native range and is not expected to be encountered in samples. Every method has limitations, and this false positive is one such limitation uncovered in this study. Future studies to identify other markers and develop A. palmeri–specific molecular diagnostics should include A. watsonii in the analysis to eliminate the possibility of a test reacting with it.
Practical Implications
Regulatory testing of commercially sold seeds requires the identification of contaminating noxious weed seeds to the species level. Molecular testing can assist in seed species identification for seeds obtained during visual purity and noxious weed exams. Method specificity and sensitivity are critical aspects of implementing and validating new methods. Thorough testing of three A. palmeri molecular methods using widely distributed Amaranthus species identified A. watsonii, a geographically isolated sister species to A. palmeri, as a potential false positive for all assays. Two assays were shown to be effective at identifying A. palmeri from DNA extracted in groups of up to 100 seeds. Regulatory programs need to be knowledgeable about the limitations of tests used in regulatory decisions. This work will support the MDA’s efforts to detect and limit the spread of A. palmeri in seed, screenings, feed, manure, and other pathways to protect our agricultural and natural lands.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/wsc.2024.58
Acknowledgments
The authors would like to thank the Minnesota Legislature and the Minnesota Seed Industry for supporting this study. The authors would like to recognize Serena McCoy with support from Tina Seeland who performed preliminary evaluation of DNA extraction methods for Amaranthus seeds in the MDA’s laboratory. The authors are especially grateful for the shared expertise and research provided by Donald Wyse, Jeff Gunsolus, Roger Becker, Tom Peters, and Anthony Brusa (University of Minnesota) who over the past decade have contributed to the work discussed in this article.
Funding
This research was financially supported by the Minnesota State Legislature through the Minnesota Department of Agriculture (MDA) Noxious and Invasive Plant Program, the MDA Seed and Plant Pathology Laboratory Unit, and the MDA Seed Regulatory Program. The MDA Seed Permit Program also provided support. This research received no specific grant from any funding agency or the commercial or not-for-profit sectors.
Competing interests
The authors declare no conflicts of interest.
Open access
