Content uploaded by Kelly Vining
Author content
All content in this area was uploaded by Kelly Vining on May 31, 2019
Content may be subject to copyright.
RESEARCH ARTICLE
Genetic diversity survey of Mentha aquatica L. and Mentha
suaveolens Ehrh., mint crop ancestors
Kelly J. Vining .Iovanna Pandelova .Kim Hummer .Nahla Bassil .
Ryan Contreras .Kristin Neill .Hsuan Chen .Amber N. Parrish .
Bernd Markus Lange
Received: 7 August 2018 / Accepted: 4 February 2019
ÓSpringer Nature B.V. 2019
Abstract The Mentha germplasm collection housed
at the USDA National Clonal Germplasm Repository
is a valuable source of diversity for genetic studies and
mint breeding. We surveyed phenotypes and geno-
types of accessions belonging to two species ancestral
to commercial peppermint: M. aquatica and M.
suaveolens. Morphology, ploidy, essential oil compo-
sition, and relative Verticillium wilt resistance were
assessed. Genotyping with simple sequence repeat
(SSR) markers was performed in order to establish a
set of informative markers for distinguishing acces-
sions from each other. M. suaveolens accessions were
triploid or tetraploid, while M. aquatica accessions
were octoploid or nonaploid. Holoploid genome sizes
differed significantly among accessions within both
species. Half of the M. aquatica accessions had (?)-
menthofuran as the primary oil constituent, while
other accessions showed atypical oil profiles. Most M.
suaveolens accessions had high levels of either
piperitenone oxide, (-)-carvone, or trans-piperite-
none oxide. M. aquatica accessions showed a range of
Verticillium wilt resistance to susceptibility, while
most M. suaveolens accessions were highly wilt-
resistant. Results from genotyping the accessions with
nine SSR markers distinguished three groups: one
mainly M. suaveolens, one mostly M. aquatica, and
one with a mixture of the species. This study enables
updates of accession descriptions in the Germplasm
Resources Information Network database, and
increases the utility of the Mentha collection to the
research community.
Keywords Mentha Mint Verticillium SSR
Polyploidy Essential oil
Introduction
Peppermint, Mentha 9piperita L., and spearmint, M.
spicata L., are the predominantly cultivated species
for mint oil production in the US (https://www.nass.
usda.gov/). In 2011, the United States produced about
7 million pounds (3175 metric tonnes) of peppermint
oil and 2.2 million pounds (998 metric tonnes) of
Electronic supplementary material The online version of
this article (https://doi.org/10.1007/s10722-019-00750-4) con-
tains supplementary material, which is available to authorized
users.
K. J. Vining (&)I. Pandelova R. Contreras
K. Neill H. Chen
Department of Horticulture, Oregon State University,
Corvallis, OR 97331, USA
e-mail: kelly.vining@oregonstate.edu
K. Hummer N. Bassil
USDA-ARS-National Clonal Germplasm Repository,
33447 Peoria Rd., Corvallis, OR 97333, USA
A. N. Parrish B. M. Lange
Institute of Biological Chemistry and M.J. Murdock
Metabolomics Laboratory, Washington State University,
Pullman, WA 99164-6340, USA
123
Genet Resour Crop Evol
https://doi.org/10.1007/s10722-019-00750-4(0123456789().,-volV)(0123456789().,-volV)
spearmint oil. Since then, US peppermint production
has dropped below 6 million pounds (2721 metric
tonnes), while the production of mint oil from corn-
mint, M. arvensis L., particularly in India, has been
increasing steadily over the past decade to over
50,000 tons. The current demand for mint oil and
derived products is expected to grow by 3–5% per year
(FAOSTAT 2018), and the development of high-
yielding, sustainable production practices is therefore
a high priority for the mint industry.
Besides global competition, which has influenced
mint prices and acreages, US production of pepper-
mint has been challenged significantly by Verticillium
wilt, caused by the fungal pathogen Verticillium
dahliae. While cultural means, such as the use of
certified clean planting stock, crop rotation, and
chemical applications, can reduce the disease inci-
dence, genetic resistance would be highly desirable for
perennial peppermint cropping systems. Due to the
complex genetics of mint, breeding approaches toward
cultivar development have thus far been lacking
positive outcomes.
Commercially grown peppermint cultivars are
sterile hybrids originating from the crosses of water-
mint, M. aquatica L., an octoploid (2n=8x= 96), and
M. spicata, a tetraploid (2n=4x= 48), and are
primarily hexaploid (2n=6x= 72) with some septa-
ploid (2n=7x= 84) or enneaploid (2n=9x= 108)
types (Tucker and Naczi 2007). Cultivated ‘Native’
spearmint is a sterile, triploid hybrid. Thus, conven-
tional breeding for improving disease resistance with
the present genotypes is not direct (Dung et al. 2010).
However, mint hybrids with peppermint-like oil
profiles can be recreated by crossing high menthofuran
M. aquatica with high menthone M. spicata (Tucker
and Naczi 2007). One strategy that has been explored
for breeding disease-resistant cultivars involves going
back to original species germplasm (Tucker 2012).
Identification of wilt-resistant genotypes with appro-
priate chemical profiles is a step toward interspecific
hybridizations aimed at creating wilt-resistant ‘Mitc-
ham’-type peppermints. To predict which genotypes
can be successfully crossed, accurate ploidy assess-
ments are essential but are currently not available for
all relevant genotypes. Past efforts also revealed
further complications, including transgressive segre-
gation, cytomixis, and complement fractionation
(Tucker 2012). To address these phenomena, it will
be important to develop molecular markers that can
identify parental contributions, and ultimately target
specific genes. DNA fingerprinting of germplasm is a
first step in that direction.
The objectives of this study are to assess ancestral
M. aquatica and M. suaveolens Ehrh. genotypes for
genome size and ploidy level, to evaluate phenotypic
diversity regarding essential oil composition and
resistance to Verticillium wilt, and to develop a set
of diagnostic microsatellite or simple sequence repeat
(SSR) markers that can distinguish each accession in
these species from the USDA Mentha collection.
Materials and methods
Plant material
This study is focused on accessions of M. aquatica and
M. suaveolens maintained as plants at the National
Clonal Germplasm Repository (NCGR) in Corvallis,
OR, USA (Table 1). The accessions were mainly
collected from countries in western and central
Europe, although one M. aquatica accession was from
Brazil, and one M. suaveolens accession was from
Argentina. For genome sizing, oil analyses and DNA
fingerprinting, 24 accessions of M. aquatica and 23
accessions of M. suaveolens were processed. Verticil-
lium wilt screening was conducted on 24 M. aquatica
accessions and 21 M. suaveolens accessions.
Genome size analysis
Flow cytometry was used to assess holoploid (2C)
genome size of individual accessions of mint by
comparing mean relative fluorescence of the sample
against an internal standard, Pisum sativum ‘Ctirad’,
with a known genome size of 8.76 pg (Greilhuber
et al. 2007). A total of 47 accessions representing two
species were evaluated using flow cytometric analysis
of nuclei stained with 40,60-diamidino-2-phenylindole
(DAPI) (CyStain ultraviolet Precise P; Partec). For
each sample, three young, fully expanded leaves were
collected to represent a random sample of nuclei. Each
sample was prepared by co-chopping 1–2 cm
2
of
tissue from both mint and the internal standard with a
double-sided razor blade in a polystyrene petri dish
containing 400 lL of nuclei extraction buffer solution
(Cystain Ultraviolet Precise P Nuclei Extraction
Buffer; Sysmex, Go
¨rlitz, Germany). Buffer containing
123
Genet Resour Crop Evol
Table 1 Mentha aquatica and M. suaveolens accessions included in this study
Plant
inventory
(PI)
Corvallis
mint
number
CMEN
Plant name Improvement status 2n
a
Origin
557570 107.001 M. aquatica 10172 Du. aq. Wild material 96 Netherlands
557571 108.001 M. aquatica 10174 2n aq. Unknown—possible
selection
96 Europe
557572 109.002 M. aquatica 10176 Unknown—possible
selection
96 Germany
557573 110.002 M. aquatica 10177 Unknown—possible
selection
– Germany
557574 111.001 M. aquatica 10179 Wild material – England
557575
b
112.001 M. aquatica 10180 Breeding material
(selection from material
from England)
96 Michigan
557576
b
113.001 M. aquatica 10181 Wild material 96 France
557577 114.001 M. aquatica 10183 Probable wild material;
cultivated at Kew
Gardens
– Europe
557578
b
115.001 M. aquatica 10184 Wild material – Brazil
557985 116.001 M. var. citrata 10185 2n cit. Breeding material – Michigan
557986 117.001 M. 9piperita var. crispa 10187
Cal. Nor
Breeding material – Oregon
557987 118.001 M. 9piperita var. crispa
d
10188 Probable wild material – Bulgaria
557988 119.001 M. 9piperita var. crispa 10189 Selection of citrata – Oregon
557989 120.003 M. 9piperita var. crispa 10190 Wild material – Michigan
557990 121.001 M. 9piperita var. crispa 10191 Wild material – Michigan
557991 122.001 M. 9piperita var. crispa 10192
Eng. #4
Wild material – England
557994 125.001 Orange Mint Selection of citrata – Europe*
557579 471.001 M. aquatica 10479 Breeding material – Michigan
557964 483.001 Mitcham sel. 80-338C Selection 2236 of
Irradiated Mitcham
– Michigan
557582 566.001 M. aquatica Bachmann #1 Wild material 96
c
Netherlands
557583 568.001 M. aquatica Bachmann #3 Wild material – Netherlands
617488 678.001 M. 9piperita var. crispa 14 Unknown – Sakhalin
684902 724.001 M. aquatica Czech Republic Wild material 48 Czech Republic
684904 728.001 M. aquatica # 7 Wild material 48 Czech Republic
557891
b
6.001 M. suaveolens 10014 2n rot. Wild material 24 Europe
557892
b
7.001 M. suaveolens 10015 4n rot. Breeding material 24 Michigan
557893 8.001 M. suaveolens 10016 French rot. Wild material 24
c
France
557894
b
9.001 M. suaveolens 10017 Du. 6 rot. Wild material 24 Netherlands
557896 11.001 M. suaveolens 10019 Fest. rot. Breeding material 24 Argentina
557897 12.001 M. suaveolens 10020 Tom. rot. Breeding material 24 Istanbul
557898 13.001 M. suaveolens 10021 Wild material 24
c
Southern Europe
557900 15.001 M. suaveolens 10025 Wild material 24 Europe
557998 28.001 M. suaveolens 10052 Breeding material Suspected 36 cultivated from Maine
nursery
123
Genet Resour Crop Evol
chopped leaf tissue was passed through a 50-lm gauze
filter (Partec Celltrics, Mu
¨nster, Germany) into a 3.5-
mL plastic tube (Sarstedt Ag & Co.; Nu
¨mbrecht,
Germany). Next, 1.6 mL of DAPI stain was added to
the nuclei suspension. All samples were analyzed
using a Partec PA II (Partec) flow cytometer except
four accessions (CMEN 48.001, CMEN 116.001,
CMEN 263.001, and CMEN 491.001), which were
analyzed using a Partec CyFlow Space (Partec). A
minimum of 3000 nuclei were analyzed per sample
with average coefficient of variation (CV) for each
fluorescence histogram less than 10. Relative 2C
genome size was calculated as DNA content of
standard 9(mean fluorescence value of sample/mean
fluorescence of standard). Monoploid genome sizes
were calculated by dividing each sample’s 2C genome
size by inferred ploidy. Monoploid genome sizes for
each species were separately subjected to analysis of
variance using PROC ANOVA using SAS software
(SAS 9.4, Cary, NC) and was significant (P\0.0001)
for both species, respectively. Means were then
separated using Tukey’s HSD at a= 0.05. Monoploid
genome size was then compared between species
using SAS software to perform a ttest.
Cytology
Root tips were placed in 2 mM 8-hydroxyquinoline
solution for 3 h at 4 °C. Next, root tips were rinsed in
distilled water before being fixed overnight in a
Farmer’s solution (3 ethanol: 1 acetic acid). Roots
were then rinsed in distilled water and transferred to a
70% ethanol solution for storage of up to several
months at 4 °C. Four root tips were placed on a slide
and treated with the enzyme solution described by
Lattier et al. (Lattier et al. 2017) for one and half hour
at 37 °C. One to two drops of modified Farmer’s
solution (3 methanol: 1 glacial acetic acid) were added
to the root tip, then root tip cells were separated by
tapping the root tip with a metal spatula (Chen et al.
2015). A drop of modified Farmer’s solution was
added to each corner of the slide, and the solution was
Table 1 continued
Plant
inventory
(PI)
Corvallis
mint
number
CMEN
Plant name Improvement status 2n
a
Origin
557999 48.001 M. suaveolens 10084 Wild material – Pennsylvania
558000 179.001 M. suaveolens 10297 Breeding material – California
557656 204.001 M. suaveolens 10365 Breeding material – Michigan
557902 201.001 M. suaveolens 10356 Breeding material – Michigan
557655 203.001 M. suaveolens 10363 Breeding material – Michigan
557657 205.001 M. suaveolens 10366 Breeding material – Michigan
557683 261.001 M. suaveolens 10502 Breeding material – Michigan
557684 263.001 M. suaveolens 10510 Breeding material – Michigan
557685 265.001 M. suaveolens 10514 Breeding material – Michigan
557904 436.001 M. suaveolens 10362 Breeding material – Michigan
557905 491.001 M. suaveolens 10357 Breeding material – Michigan
557907 538.001 M. suaveolens 32/86 Open-pollinated from
botanical collection
24 France
557908 543.001 M. suaveolens 19/79 Open-pollinated from
botanical collection
24 Portugal
557638 573.001 M. suaveolens No. 27 Open-pollinated from
botanical collection
24
b
Tashkent, Uzbekistan
557912 587.001 Variegata Pineapple Mint Selection of pineapple
Mint
– Oregon
a
Base chromosome number is x= 12;
b
used for initial screening of SSRs;
c
Chambers and Hummer (1994)
d
Formerly considered M. aquatica L. var. crispa Benth. (1833)
123
Genet Resour Crop Evol
immediately lit by quickly waving through an alcohol
lamp. Excess liquid was tapped off of the slides, and
the slides were allowed to air-dry overnight at 37 °C.
Air dried slides were submerged in a 5.7% solution of
Giemsa Stain, Modified Solution (Sigma-Aldrich) for
15 min, then quickly rinsed in water, and again air-
dried at 37 °C. Images were taken using a light
microscope at 400 9magnification (Axio Imager A1;
Zeiss).
Oil composition
Leaf material (approximately 7 g) representing the
entire shoot was collected at roughly 10% flowering
and transferred to a 500 mL round bottom flask with
100 mL of deionized water and camphor as an internal
standard (final concentration at 15 ng/lL) for subse-
quent hydrodistillation using a modified Likens-
Nickerson apparatus and n-hexane as carrier solvent
(Ringer et al. 2003). Three biological replicate sam-
ples were processed for each accession. An aliquot of
the n-hexane fraction was transferred to a 2-mL glass
vial for further analysis. Analytes were separated
using a DB-WAX UIcolumn (60 m 90.25 mm;
0.25 lM film thickness, J&W Scientific) installed in
a gas chromatograph with flame ionization detector
(GC-FID) (6890 N, Agilent Technologies). The initial
oven temperature of 85 °C was held for 4 min,
followed by stepwise gradients to 130 °C (at 3 °C/
min and a hold for 15 min), then to 235 °C (at 8 °C/
min and a hold for 12 min). Front inlet and detector
temperatures were 250 °C, the inlet was operated in
splitless mode, the injection volume was 1 lL, and He
carrier gas flow was set to 2.0 mL/min. The FID was
operated with a hydrogen flow of 30 mL/min, an air
flow of 400 mL/min and He makeup flow rate of
15 mL/min. Metabolite identification was achieved by
comparison with the chromatographic retention char-
acteristics of authentic standards and quantitation
preformed based on appropriate calibration curves.
Full oil profiling data are provided in Supplemental
Table S1.
Verticillium wilt susceptibility screening
Six replicate cuttings per accession were rooted in cell
packs in autoclaved Sungro
Ò
Professional Growing
Mix (Sungro Horticulture, Agawam, MA). During
rooting, they were incubated for approximately
2 weeks in a growth chamber under fluorescent
lighting (91 lmol/m
2
) with cool temperatures and a
short-day light regime (22 °C, 10 h light/20 °C, 14 h
dark) to maintain vegetative growth and prevent
flowering. In parallel, Verticillium dahliae was grown
in Difco
TM
Czapek-dox broth (Becton, Dickinson and
Company, Sparks, MD) on a shaker at 200 RPM at
room temperature. The liquid medium (200 mL) was
inoculated with three to four *0.5 cm
2
agar plugs
from a V. dahliae Czapek-dox agar plate. Inoculum
preparation and root-dip inoculation of rooted cuttings
were performed as previously described (Vining et al.
2005). Inoculated cuttings were randomly replanted in
new flats in autoclaved Sungro
Ò
Professional Growing
Mix, and incubated in the growth chamber for at least
4 weeks to observe symptom development. Verticil-
lium wilt disease symptoms used for rating relative
susceptibility were stunting, chlorosis, and crescent
leaf appearance relative to control plants. Each plant
was rated on a scale of zero to four, where a zero rating
indicated no visible symptoms, and a four rating
indicated a dead plant.
DNA extraction and genotyping
Fresh actively growing leaves of mint were collected
in from the OSU back-up collection of the USDA-
ARS National Clonal Germplasm Repository (NCGR)
(Table 1). Up to 50 mg of leaf tissue from each sample
was homogenized with an MM 301 Mixer Mill
(Retsch International, Haan, Germany) and DNA
was extracted using a modified Puregene (Gentra
Systems Inc. Minneapolis, MN, USA) protocol used
routinely in the NCGR lab (Gilmore et al. 2011).
Forty-five simple sequence repeats (SSR) targeting
long core repeats (C3 bp) were identified in the draft
genome sequence of M. longifolia L. Huds (Vining
et al. 2017). Primer pairs were designed with
BatchPrimer3 (You et al. 2008) and used to screen a
testing panel of eight mint accessions consisting of
three accessions each of M. suaveolens and M.
aquatica in addition to two accessions of M. longifolia
from South Africa which included CMEN 585 (the
accession used for the draft mint genome, Vining et al.
2017) and CMEN 584. An M13 tail (TGTAAAAC-
GACGGCCAGT) was added to the 50end of each
forward primer for economic genotyping (Schuelke
2000). Three SSR markers (prefix EMM) identified by
Kumar et al. (Kumar et al. 2015) to amplify in four
123
Genet Resour Crop Evol
species (M. arvensis,M. citrata,M. longifolia, and M.
spicata) were also used.
The 15 lL PCR reaction contained 1 9reaction
buffer, 2 mMMgCl
2
, 0.2 mMdNTPs, 0.12 lMM13-
tagged forward primer, 0.5 lMreverse primer,
0.5 lMfluorescent WellRED tag (Beckman Coulter,
Inc., Fullerton, CA, USA), 0.025 U of GoTaq Hot Start
polymerase (Promega Corp. Madison, WI, USA), and
4.5 ng genomic DNA. DNA was amplified in an
Eppendorf Gradient thermocycler (Eppendorf, West-
bury, NY, USA) or an MJ Research Tetrad thermo-
cycler (BioRad, Hercules, CA, USA). The PCR
protocol consisted of a ‘‘touchdown’’ program with
an initial denaturation cycle at 94 °C for 3 min
followed by ten cycles of 40 s at 94 °C; 45 s at
62 °C, decreasing 1 °C each cycle; and 45 s at 72 °C.
PCR continued for an additional 20 cycles of 40 s at
94 °C; 45 s at 52 °C; and 45 s at 72 °C; followed by
eight more cycles of 40 s at 94 °C; 45 s at 53 °C; and
45 s at 72 °C. The program was terminated at 72 °C
for 30 min. The PCR success and polymorphism of the
SSR was first evaluated by 2% agarose gel elec-
trophoresis and visualized with ethidium bromide.
PCR products were pooled and separated by capillary
electrophoresis with a Beckman CEQ 8000 (Beckman
Coulter, Inc.). Allele sizing and visualization were
performed using the fragment analysis module of the
CEQ 8000 software. Individuals were scored for
presence or absence of PCR products by grouping
alleles into bins of less than one base pair. The PCR
products of the nine SSRs used for fingerprinting the
47 accessions of M. aquatica and M. suaveolens used
in this study were pooled into two multiplexes using
the conditions described above.
Cluster analysis
Genetic relationships among mint accessions were
evaluated by constructing a neighbor joining (NJ)
dendrogram based on Bruvo’s distance (Bruvo et al.
2004) using the R packages ‘ape’ (v 5.0; Paradis et al.
2004) and ‘poppr’ (v 2.6.1, Kamvar et al. 2014,2015).
Null alleles were coded as ‘NA’ and unknown alleles
caused by null allele masking or unknown dosage
were coded as ‘000’ using the ‘recode polyploids’
command (Zurn et al. 2018). The ‘bruvo.boot’ com-
mand was used to produce a neighbor joining tree with
the ‘njs’ algorithm from ‘ape’ and Bruvo’s distance
assuming an infinite allele sharing model. Bootstrap
support of 2000 permutations was used (Online
Resource 2). Bruvo’s distance was chosen for its
ability to calculate distances for mixed-ploidy popu-
lations as observed on these mint samples.
Results
Morphological characters
Distinguishing morphological characters of the M.
aquatica and M. suaveolens accessions are listed
(Table 2). Representatives of each species are shown
(Fig. 1), and herbarium specimens of all accessions
are provided (Figures S1 and S2). The overall growth
habit for M. aquatica was either upright (11 acces-
sions) or spreading (five accessions). Most M. suave-
olens had an upright growth habit (15 accessions),
although two were intermediate between Upright and
spreading. For both species, leaf shapes ranged from
ovate to lanceolate, with ovate being the most
common shape. Leaf margins were serrate for all
accessions of both species, and leaf tips ranged from
acute to obtuse. M. aquatica corollas were purple, with
subtle variations in shade, while M. suaveolens
corollas were either purple or white. Both species
exhibited male sterility in the form of underdeveloped
stamens, with short filaments and shriveled, brown
anthers. This was more common in M. suaveolens (six
accessions) than in M. aquatica (three accessions).
Ploidy and genome size analysis
Holoploid genome sizes of M. suaveolens ranged from
0.85 to 1.13 pg among diploids and from 1.48 to
1.65 pg among triploids (Table 3). There were 12
diploids and 11 triploids among the accessions stud-
ied. Our findings differed from previous reports, which
found only diploids among these accessions (Cham-
bers and Hummer 1994). We confirmed ploidy levels
inferred from flow cytometry calculations by counting
chromosomes of two diploids (CMEN 9.001 and
CMEN 28.001) and one triploid (CMEN 6.001)
(Fig. 2). Monoploid (1Cx) genome sizes were signif-
icantly different (P\0.0001) among the 20 acces-
sions and ranged from 0.42 to 0.56 pg. Accession
CMEN 491.001 had a substantially lower holoploid
genome size (0.85 pg) than the mean of the other 11
diploid accessions (1.08 pg) and a significantly lower
123
Genet Resour Crop Evol
Table 2 Morphological characters differentiating Mentha aquatica and M. suaveolens accessions
Corvallis mint number
CMEN
Species Corolla
color
Developed
anthers
Growth habit Leaf shape Leaf
tip
107.001 M. aquatica Purple Y Upright Lanceolate Obtuse
108.001 M. aquatica Purple Y Upright Lanceolate Obtuse
109.002 M. aquatica Purple Y Upright Ovate-
lanceolate
Obtuse
110.002 M. aquatica Purple N Spreading Ovate Obtuse
111.001 M. aquatica Purple N Spreading Ovate Obtuse
112.001 M. aquatica Purple Y Spreading Ovate Obtuse
113.001 M. aquatica Purple N Upright Ovate Obtuse
114.001 M. aquatica Purple Y Upright Ovate Acute
115.001 M. aquatica Purple Y Upright Ovate Obtuse
116.001 M.aquatica Purple N Upright Ovate Obtuse
117.001 M. aquatica var
citrata
Purple N Upright Ovate Acute
118.001 M. aquatica var
citrata
Purple N Upright Ovate Acute
119.001 M. aquatica var
citrata
Purple N Upright Ovoid-ovate Acute
120.003 M. aquatica var
citrata
Purple N Upright Ovate Acute
121.001 M. aquatica var
citrata
Purple N Upright Ovate Obtuse
122.001 M. aquatica var
citrata
Purple N Upright Ovate Obtuse
125.001 Orange Mint Purple N Upright Ovate-
lanceolate
Acute
471.001 M. aquatica Purple N Upright Ovate Acute
483.001 Mitcham sel.
80-338C
Purple N Upright Ovate Obtuse
566.001 M. aquatica Purple Y Spreading Ovate Acute
568.001 M. aquatica Purple N Spreading Ovate Acute
678.001 M. aquatica Purple N Upright Ovate Acute
724.001 M. aquatica Purple N Upright Lanceolate Acute
728.001 M. aquatica Purple N Upright Lanceolate Acute
6.001 M. suaveolens Purple Y Upright Ovate Obtuse
7.001 M. suaveolens Purple Y Upright Ovate-
lanceolate
Obtuse
8.001 M. suaveolens Purple Y Upright Ovate Obtuse
9.001 M. suaveolens Purple Y Upright Ovate-
lanceolate
Obtuse
11.001 M. suaveolens White Y Upright Ovate-
lanceolate
Acute
12.001 M. suaveolens White N Upright Ovate-
lanceolate
Acute
13.001 M. suaveolens White Y Upright Ovate Obtuse
15.001 M. suaveolens White Y Upright Lanceolate Acute
28.001 M. suaveolens White N Upright Ovate-
lanceolate
Obtuse
123
Genet Resour Crop Evol
monoploid genome size (0.42 pg) than all other
accessions (0.52 pg).
There were 16 octoploids, two tetraploids, and six
eneaploid among the accessions of M. aquatica.
Holoploid genome size ranged from 3.20 to 3.51 pg
among octopoids, from 1.87 to 1.91 pg for tetraploids,
and from 3.68 to 3.80 pg for eneaploid (Table 4). Five
of the six eneaploid identified were M. aquatica var.
citrata and accession CMEN 483.001 was identified as
Mitcham sel. 80-338C. Monoploid genome sizes were
significantly different (P\0.0001) and ranged from
0.40 to 0.48 pg among the 24 accessions. Ploidy levels
of accessions for which previous reports existed were
consistent with prior results (Chambers and Hummer
1994). Chromosome counts for two octoploids
(CMEN 108.001 and CMEN 115.001), one tetraploid
(CMEN 724.001), and one eneaploid (CMEN
119.001) supported the inferred ploidy levels from
flow cytometry. Mean monoploid genome sizes were
significantly different (P\0.0001) for M. aquatica
(0.424 pg) and M. suaveolens (0.525 pg).
Oil composition
The essential oil composition of about one half of the
accessions labeled as M. aquatica (CMEN 107.001,
CMEN 108.001, CMEN 109.001, CMEN 110.001,
CMEN 111.001, CMEN 112.001, CMEN 114.001,
and CMEN 115.001) was fairly uniform, with high
levels of (?)-menthofuran (C65% of detected
volatiles) and smaller quantities of (-)-limonene
(3–14%) and 1,8-cineole (5–12%) (Table 5and
Fig. 3). Non-typical oils were recovered from acces-
sions CMEN 568.001 [relatively high 1,8-cineole
(31%) at the cost of (?)-menthofuran (30%)], CMEN
113.001, CMEN 125.001 and CMEN 566.001 (rich in
a-terpinyl acetate at C60%), CMEN 483.001 [high
linalool (62%) and linalyl acetate (11%)], CMEN
471.001 (high pulegone at [70%), CMEN 724.001
and CMEN 728.001 (both accumulate C65% car-
vone), and CMEN 678.001 [complex oil with consid-
erable quantities of 1,8-cineole (17%), menthone
(27%) and menthol (10%)] (Table 5and Fig. 3). Four
accessions listed as M. aquatica var. citrata (CMEN
Table 2 continued
Corvallis mint number
CMEN
Species Corolla
color
Developed
anthers
Growth habit Leaf shape Leaf
tip
48.001 M. suaveolens 10084 White Y Upright Ovate-
lanceolate
Acute
179.001 M. suaveolens Purple Y-Few Upright Lanceolate Acute
201.001 M. suaveolens White N Upright Ovate Acute
203.001 M. suaveolens White N Upright Ovate-
lanceolate
Obtuse
204.001 M. suaveolens Purple N Upright Lanceolate Acute
205.001 M. suaveolens White N Upright Ovate Obtuse
261.001 M. suaveolens Purple N Upright Ovate-
lanceolate
Acute
265.001 M. suaveolens White N Upright Lanceolate Acute
436.001 M. suaveolens White N Upright Ovate Obtuse
491.001 M. suaveolens White N Upright Ovate Obtuse
538.001 M. suaveolens Purple Y Spreading/
Upright
Ovate Obtuse
543.001 M. suaveolens Purple Y Spreading/
Upright
Ovate Obtuse
573.001 M. suaveolens Purple Y Upright Ovate Obtuse
587.001 M. suaveolens Purple N Upright Lanceolate Obtuse
123
Genet Resour Crop Evol
117.001, CMEN 118.001, CMEN 119.001, and
CMEN 120.001) contained linalool (48–55%) and
linalyl acetate (17–24%) as principal volatile con-
stituents. Two accessions labeled as M. aquatica var.
citrata (CMEN 121.001 and CMEN 122.001) accu-
mulated (?)-menthofuran (68 and 69%, respectively)
(Table 5). The oil of the CMEN 116.001 accession
was characterized by a prevalence of piperitenone
oxide (41%) and trans-piperitone oxide (30%).
The most common oil type among accessions
labeled as M. suaveolens (type I) was characterized
by high levels of piperitenone oxide (72–89% in
CMEN 006.001, CMEN 008.001, CMEN 011.001,
CMEN 013.001, CMEN 201.001, CMEN 203.001,
CMEN 204.001, CMEN 205.001, CMEN 436.001,
CMEN 538.001, and CMEN 587.001) (Table 5and
Fig. 3). Oil with high (-)-carvone content (C64%)
(type II) was obtained from six accessions (CMEN
012.001, CMEN 015.001, CMEN 028.001, CMEN
048.001, CMEN 179.001, and CMEN 573.001). The
oil of another class of M. suaveolens accessions (type
III) had trans-piperitone oxide as principal constituent
(56 and 60% in CMEN 007.001 and CMEN 543.001,
respectively) (Table 5and Fig. 3). A mixture of
piperitenone oxide, (-)-carvone and trans-piperitone
oxide (in order of abundance) was found in the oil of
CMEN 009.001. High pulegone levels ([82%) were
detected in the oil of CMEN 263.001 and CMEN
491.001. Piperitenone (82%) was the principal con-
stituent of the oil of accession CMEN 265.001.
Complete results of oil analyses for both species are
given in Supplemental Table S1.
Verticillium wilt susceptibility screening
Accessions with mean susceptibility scores between
1.0 and 2.0 were considered moderately resistant to
Verticillium wilt disease, while accessions with mean
ratings\1.0 were considered highly resistant. Acces-
sions with 2.0–2.5 mean ratings were categorized as
moderately susceptible, and accessions with mean
ratings [2.5 were categorized as highly susceptible.
Both M. aquatica and M. suaveolens showed consid-
erable variation in disease susceptibility. However, the
overall profile of disease susceptibility differed
between species. In M. aquatica, 14 of the 24
accessions (58%) were moderately or highly resistant,
four were moderately susceptible, and five were highly
susceptible. In contrast, in M. suaveolens, 19 of the 21
accessions were highly resistant, and only two were
moderately susceptible (Figs. 4,5).
Oil constituents and Verticillium resistance
correlation analysis
A correlation analysis was performed to assess if the
occurrence of specific essential oil constituents may
affect Verticillium wilt resistance. In this analysis we
focused on signature monoterpenes ((?)-menthofuran
for M. aquatica; linalool and linalyl acetate for M.
aquatica var. citrata; and piperitenone oxide, (-)-
carvone and trans-piperitone oxide for M. suaveolens
(type I, II and III, respectively). Each signature oil
constituent accumulated to high levels in only a few
Fig. 1 Representative flower and leaf phenotypes of Mentha
species. aPI 557892 CMEN 007.001 M. suaveolens.bPI
557894 CMEN 009.001 M. suaveolens.cPI 557570 CMEN
108.001 M. aquatica;dPI 557990 CMEN 121.001 M. aquatica
123
Genet Resour Crop Evol
accessions for which highly variable Verticillium
susceptibility scores were determined (Fig. 6;
Figure S3).
DNA fingerprinting
Screening the 8-member testing panel with the 48 SSR
primer pairs identified: 25 polymorphic primers; 2
monomorphic (C9039 and C9174); and 21 that did not
amplify (Supplemental Table S2). Products from all
25 SSRs that were polymorphic were pooled based on
size into non-overlapping size fragments and sepa-
rated by capillary electrophoresis on the Beckman
CEQ 8000 as described above to further assess
polymorphism and ease of scoring. C235 was
polymorphic only in M. longifolia (3 alleles, 272,
275, and 280 bp) but monomorphic in M. aquatica and
M. suaveolens (292 bp). C4299 and C4604 generated
two alleles each and were not used further due to low
polymorphism (Supplemental Table S2). Thirteen
primer pairs were difficult to score and were discarded
while nine primer pairs were polymorphic and
appeared easy to score and were thus used in two
multiplexes to genotype all 49 samples included in this
study.
Two (EMM_010 and EMM_032) of the three tri-
nucleotide-containing SSRs reported as cross trans-
ferable across M. arvensis,M. citrata,M. longifolia,
and M. spicata by Kumar et al. (2015) did not amplify
across the testing panel of M. aquatica,M. suaveolens
Table 3 Ploidy levels of Mentha suaveolens accessions surveyed for this study
Accession Ploidy—previously reported Ploidy—current study
a
2C (pg) 1Cx(pg)
b
CMEN 6.001 2x3x1.60 ±0.03 0.53 abcde
CMEN 7.001 2x2x1.09 ±0.02 0.49 e
CMEN 8.001 2x
c
2x1.12 ±0.02 0.56 ab
CMEN 9.001 2x2x1.13 ±0.01 0.56 a
CMEN 11.001 2x3x1.65 ±0.01 0.55 abc
CMEN 12.001 2x3x1.55 ±0.02 0.52 cde
CMEN 13.001 2x
c
2x1.04 ±0.01 0.52 bcde
CMEN 15.001 2x3x1.54 ±0.05 0.51 cde
CMEN 28.001 2x2x1.04 ±0.01 0.52 bcde
CMEN 48.001 – 3x1.51 ±0.02 0.50 de
CMEN 179.001 – 3x1.48 ±0.01 0.49 e
CMEN 201.001 – 2x1.07 ±0.00 0.53 abcde
CMEN 203.001 – 2x1.08 ±0.03 0.54 abcd
CMEN 204.001 – 2x1.09 ±0.01 0.54 abcd
CMEN 205.001 – 3x1.59 ±0.01 0.53 abcde
CMEN 263.001 – 3x1.66 ±0.02 0.55 abc
CMEN 265.001 – 3x1.62 ±0.01 0.54 abcd
CMEN 436.001 2x3x1.57 ±0.03 0.52 abcde
CMEN 491.001 – 2x0.85 ±0.02 0.42 f
CMEN 538.001 – 2x1.06 ±0.01 0.53 abcde
CMEN 543.001 2x2x1.09 ±0.02 0.55 abc
CMEN 573.001 2x3x1.54 ±0.01 0.51 cde
CMEN 587.001 – 2x1.05 ±0.01 0.52 abcde
Detailed information for each is listed in the USDA-GRIN database, https://npgsweb.ars-grin.gov/
a
Base chromosome number is x= 12. Ploidy inferred from 2C genome size divided by mean 1Cxgenome size (0.5)
b
Monoploid genome sizes followed by different letters are statistically different based on Tukey’s HSD (a= 0.05)
c
Chambers and Hummer (1994)
123
Genet Resour Crop Evol
and M. longifolia and will not be useful in these
species (Supplemental Table S2). The third SSR,
EMM_055 amplified across the species but was
difficult to score reliably and was thus excluded from
the analysis.
C667, C4701 and C2645 were the most polymor-
phic primer pairs among the nine used based on
number of alleles (15, 17, and 14, respectively) and
Nei’s allele diversity (0.92, 0.88, and 0.83, respec-
tively) while C4573 was the least polymorphic SSR
with 4 alleles and a Nei’s diversity of 0.57 (Table 6).
AB
CD
EF
G
Fig. 2 Photomicrographs of representative Mentha accessions
at various ploidy levels. aCMEN 9.001 (2n=2x= 24)
bCMEN 28.001 (2n=2x= 24) cCMEN 6.001
(2n=3x= 36) dCMEN 724.001 (2n=4x= 48) eCMEN
108.001 (2n=8x= 96) fCMEN 115.001 (2n=8x= 96). Scale
bar 5 lm
123
Genet Resour Crop Evol
Only two samples represented M. longifolia in this
study. Therefore, monomorphism in these two sam-
ples at C870 and C4573 may not be observed when
evaluating a larger number of accessions of this
species. As expected from having only two samples,
the average number of alleles and Nei’s diversity were
low in M. longifolia at 2.78 and 0.68, respectively and
a better assessment of their polymorphism in this
species needs further study. The average number of
alleles and Nei’s allele diversity in M. aquatica and in
M. suaveolens were relatively high, exceeding or
equal 7 and 0.7, respectively.
The number of private alleles or species-specific
alleles were 2 in M. longifolia,26inM. aquatica and
20 in M. suaveolens (Supplemental Table S3). In M.
longifolia, one private allele each were found at C4701
and C5388. The number of private alleles ranged from
0to6inM. suaveolens and were found at all but C852
and C4573 while they ranged from 1 to 4 in M.
aquatica and were found at each SSR.
Hierarchical clustering using Bruvo’s distance into
a neighbor joining tree identified three major groups
that consisted primarily of the M. aquatica and the M.
suaveolens species and a mixed group that contained
the two M. longifolia accessions (Fig. 7). The M.
aquatica group contained all accessions classified in
this species except for two accessions of M. aquatica
from the Czech Republic (CMEN 724 and CMEN
Table 4 Ploidy levels of Mentha aquatica accessions surveyed for this study
Accession Ploidy—previously reported Ploidy—current study
a
2C (pg) 1Cx(pg)
b
CMEN 107.001 8x8x3.29 ±0.04 0.41 cd
CMEN 108.001 8x8x3.39 ±0.01 0.42 cd
CMEN 109.002 8x8x3.31 ±0.05 0.41 cd
CMEN 110.002 – 8x3.26 ±0.01 0.41 cd
CMEN 111.001 – 8x3.51 ±0.04 0.44 bc
CMEN 112.001 8x8x3.41 ±0.06 0.43 cd
CMEN 113.001 8x8x3.44 ±0.05 0.43 cd
CMEN 114.001 – 8x3.43 ±0.07 0.43 cd
CMEN 115.001 – 8x3.44 ±0.07 0.43 cd
CMEN 116.001 – 9x3.71 ±0.02 0.41 cd
CMEN 117.001 – 9x3.69 ±0.03 0.41 cd
CMEN 118.001 – 9x3.79 ±0.05 0.42 cd
CMEN 119.001 – 9x3.75 ±0.04 0.42 cd
CMEN 120.003 – 8x3.49 ±0.04 0.44 bc
CMEN 121.001 – 8x3.23 ±0.07 0.40 d
CMEN 122.001 – 8x3.31 ±0.03 0.41 cd
CMEN 125.001 8x
c
8x3.41 ±0.06 0.43 cd
CMEN 471.001 – 8x3.41 ±0.04 0.43 cd
CMEN 483.001 – 9x3.68 ±0.05 0.41 cd
CMEN 566.001 – 8x3.20 ±0.02 0.40 d
CMEN 568.001 – 8x3.41 ±0.07 0.43 cd
CMEN 678.001 4x9x3.80 ±0.02 0.42 cd
CMEN 724.001 4x4x1.87 ±0.01 0.47 ab
CMEN 728.001 4x1.91 ±0.04 0.48 a
Detailed information for each is listed in the USDA-GRIN database, https://npgsweb.ars-grin.gov/
a
Base chromosome number is x= 12. Ploidy inferred from 2C genome size divided by mean 1Cxgenome size (0.4356)
b
Monoploid genome sizes followed by different letters are statistically different based on Tukey’s HSD (a= 0.05)
c
Chambers and Hummer (1994)
123
Genet Resour Crop Evol
Table 5 Main volatile constituents of essential oils distilled from M. aquatica and M. suaveolens accessions
Identifier
in Mentha
collection
Oil constituent (% of total)
(-)-
Limonene
1,8-
Cineole
(?)-
Mentho-
furan
Linalool Linalyl
acetate
trans-
Piperitone
oxide
(-)-
Carvone
Piperitenone
oxide
Other
Mentha aquatica
CMEN
107.001
6.6 ±0.6 10.8 ±2.5 69.4 ±5.0 0.7 ±0.2 – – – 0.2 ±0.2 12.3
CMEN
108.001
6.7 ±0.4 10.6 ±1.6 71.2 ±4.1 0.6 ±0.2 – – – – 10.9
CMEN
109.001
11.2 ±0.4 9.8 ±0.6 66.9 ±1.1 0.4 ±0.1 – – – – 11.7
CMEN
110.001
3.2 ±0.3 10.0 ±2.1 71.5 ±4.7 – – – – – 15.3
CMEN
111.001
3.1 ±0.2 10.8 ±0.7 68.9 ±1.7 0.2 ±0.2 – – – – 17.0
CMEN
112.001
3.0 ±0.3 11.5 ±1.6 68.5 ±4.1 – – – – – 17.0
CMEN
114.001
4.7 ±0.1 9.0 ±0.7 65.7 ±3.3 0.2 ±0.1 – – – – 20.4
CMEN
115.001
9.4 ±0.1 5.4 ±0.3 32.7 ±1.5 – – – – – 52.5
CMEN
568.001
5.8 ±0.3 30.7 ±5.9 29.6 ±14.1 – – – – – 33.9
CMEN
483.001
0.7 ±0.1 5.3 ±0.9 – 62.3 ±0.3 11.4 ±4.0 – – – 20.3
CMEN
724.001
11.1 ±9.9 0.1 ±0.1 – – – – 70.5 ±4.1 – 18.3
CMEN
728.001
9.8 ±0.5 3.8 ±0.6 – – – – 64.8 ±4.1 – 21.6
CMEN
113.001
10.4 ±3.6 6.1 ±1.3 0.2 ±0.1 0.2 ±0.1 – – – – 83.1
CMEN
125.001
8.6 ±2.0 6.8 ±0.7 0.2 ±0.1 0.2 ±0.1 – – – – 84.2
CMEN
566.001
13.7 ±0.1 8.5 ±0.1 – – – – – – 77.8
CMEN
471.001
0.8 ±0.1 0.5 ±0.1 0.1 ±0.1 0.1 ±0.1 – 0.1 ±0.1 – – 98.4
CMEN
678.001
3.8 ±0.1 16.6 ±0.1 1.9 ±0.1 0.5 ±0.1 – 2.6 ±0.1 0.1 ±0.1 0.1 ±0.1 74.9
Mentha aquatica var. citrata
CMEN
117.001
0.8 ±0.1 0.3 ±0.1 – 52.2 ±1.5 19.6 ±0.7 – – – 27.1
CMEN
118.001
0.9 ±0.1 0.3 ±0.1 – 50.2 ±2.3 17.5 ±0.8 – – – 31.1
CMEN
119.001
0.9 ±0.1 1.9 ±0.2 0.1 ±0.1 55.0 ±1.8 20.1 ±6.1 – – – 22.8
CMEN
120.001
1.2 ±0.1 2.6 ±0.7 – 47.8 ±4.6 23.6 ±2.4 – – – 24.8
123
Genet Resour Crop Evol
Table 5 continued
Identifier
in Mentha
collection
Oil constituent (% of total)
(-)-
Limonene
1,8-
Cineole
(?)-
Mentho-
furan
Linalool Linalyl
acetate
trans-
Piperitone
oxide
(-)-
Carvone
Piperitenone
oxide
Other
CMEN
121.001
6.9 ±0.1 11.8 ±0.7 68.3 ±2.8 0.7 ±0.1 – – – – 12.3
CMEN
122.001
5.5 ±0.4 12.2 ±0.8 69.2 ±0.7 0.7 ±0.1 – – – – 11.7
CMEN
116.001
6.3 ±0.7 0.1 ±0.1 – 5.2 ±5.2 2.3 ±2.3 29.8 ±4.2 0.1 ±0.1 41.0 ±7.4 15.2
Mentha suaveolens
CMEN
006.001
2.1 ±0.5 0.1 ±0.1 – 0.1 ±0.1 – 0.5 ±0.1 – 88.5 ±1.2 8.7
CMEN
008.001
1.6 ±0.5 0.1 ±0.1 – 0.1 ±0.1 – 16.7 ±3.6 – 71.8 ±4.4 9.7
CMEN
011.001
1.6 ±0.3 1.2 ±0.1 – 0.1 ±0.1 – 7.8 ±3.2 – 75.1 ±5.4 14.2
CMEN
013.001
1.3 ±0.1 – – 0.1 ±0.1 – – – 77.8 ±1.0 20.8
CMEN
201.001
1.0 ±0.3 0.1 ±0.1 – – – – – 85.0 ±1.4 13.9
CMEN
203.001
0.7 ±0.1 0.1 ±0.1 – 0.1 ±0.1 – – – 88.8 ±0.9 10.3
CMEN
204.001
1.2 ±0.5 2.2 ±1.8 – 0.1 ±0.1 – – – 77.8 ±2.7 18.7
CMEN
205.001
1.2 ±0.4 1.3 ±2.1 – 0.1 ±0.1 – – – 80.6 ±5.9 16.8
CMEN
436.001
1.2 ±0.1 1.5 ±0.1 – – – – – 86.7 ±0.7 10.6
CMEN
538.001
1.1 ±0.4 0.1 ±0.1 – – – 0.1 ±0.1 – 77.7 ±0.7 21.0
CMEN
587.001
1.0 ±0.3 0.1 ±0.1 – – – 0.2 ±0.2 – 84.8 ±0.1 13.9
CMEN
012.001
10.7 ±0.3 3.5 ±0.2 – 0.1 ±0.1 0.1 ±0.1 – 74.5 ±2.4 0.1 ±0.1 11.0
CMEN
015.001
10.9 ±1.5 4.2 ±0.7 – 0.1 ±0.1 0.1 ±0.1 – 73.6 ±1.1 0.1 ±0.1 11.0
CMEN
028.001
11.3 ±1.7 2.8 ±0.3 – 0.1 ±0.1 – – 72.3 ±0.5 – 13.5
CMEN
048.001
22.1 ±1.5 0.7 ±0.2 0.2 ±0.1 – – – 64.6 ±1.2 – 12.4
CMEN
179.001
18.1 ±3.7 3.3 ±0.8 – – – – 65.3 ±1.7 – 13.3
CMEN
573.001
15.0 ±2.2 4.3 ±0.5 – – – 0.3 ±0.1 66.1 ±0.9 – 14.3
CMEN
007.001
2.1 ±0.5 1.7 ±0.3 – 0.1 ±0.1 0.1 ±0.1 56.4 ±1.1 – 3.0 ±1.7 36.6
CMEN
543.001
2.1 ±0.4 0.1 ±0.1 – – – 59.7 ±6.1 0.2 ±0.2 19.3 ±7.6 18.6
CMEN
009.001
5.4 ±1.0 0.1 ±0.1 – 0.1 ±0.1 – 15.9 ±8.2 29.6 ±18.1 36.0 ±16.4 12.9
123
Genet Resour Crop Evol
728) that were tetraploid based on cytology. These
accessions were indicated by the donor as possible
hybrids and grouped with M. suaveolens accessions
and not with the M. aquatica clade. Another M.
aquatica accession, CMEN 678 also did not group
with the M. aquatica accessions but was grouped with
the mixed group that contained the two M. longifolia
accessions, CMEN 724 from the Czech Republic and
one M. suaveolens accession, CMEN 7.
Table 5 continued
Identifier
in Mentha
collection
Oil constituent (% of total)
(-)-
Limonene
1,8-
Cineole
(?)-
Mentho-
furan
Linalool Linalyl
acetate
trans-
Piperitone
oxide
(-)-
Carvone
Piperitenone
oxide
Other
CMEN
263.001
0.9 ±0.1 0.2 ±0.1 0.1 ±0.1 0.1 ±0.1 – – – – 98.7
CMEN
491.001
1.3 ±0.4 – – – – – – – 98.7
CMEN
265.001
0.6 ±0.2 0.1 ±0.1 – 0.1 ±0.1 – – – 0.7 ±0.1 98.5
Fig. 3 Overview of p-menthane monoterpene biosynthesis in
the genus Mentha, emphasizing the reactions and metabolites of
relevance for essential oils of M. aquatica and M. suaveolens.
Known enzymes are indicated in blue font. CDH, (-)-carveol
dehydrogenase; ISPD, (-)-trans-isopiperitenol dehydrogenase;
ISPR, (-)-trans-isopiperitenone reductase; LS, (-)-limonene
synthase; L3H, (-)-limonene 3-hydroxylase; L6H, (-)-
limonene 6-hydroxylase; MFS, (?)-menthofuran synthase
123
Genet Resour Crop Evol
Discussion
Understanding the diversity of available germplasm is
essential to utilization of plant accessions in breeding.
Toward this end, we assessed phenotypic, chemotypic,
and molecular genetic diversity of two Mentha species
that are ancestral to commercial peppermint: M.
aquatica and M. suaveolens. A similar assessment
for the third ancestral species, M. longifolia, was
previously reported (Vining et al. 2005). A summary
of findings for each species follows.
Mentha aquatica
The M. aquatica accessions had uniformly purple
flowers, but varied moderately in growth habit and leaf
morphology (Table 2). Oil composition differed
between the M. aquatica and the M. aquatica var.
citrata subgroup, with the former typified by (?)-
menthofuran, the latter by linalool and linalyl acetate.
While most of the M. aquatica accessions were
octoploid, agreeing with previous work, we also found
five eneaploid M. aquatica accessions, none of which
had been previously reported. Jedrzejczyk and Rewers
(Jedrzejczyk and Rewers 2018) recently reported
genome sizes for 34 accessions of various mint
species, including two genotypes each of M. aquatica
and M. suaveolens, both of which agreed with our
report. However, their report included no inferences
regarding ploidy, nor were any chromosome counts
performed. Harley and Brighton (Harley and Brighton
1977) provided an extensive report of chromosome
numbers and indicated that M. aquatica sampled over
a wide range of the species’ geographic distribution
and morphological variation included only octoploid
accessions (2n=8x= 96). However, Harley and
Fig. 4 Relative ratings of Verticillium wilt susceptibility for M.
aquatica accessions. Means and standard deviations are shown
for three biological replicates per accession. The rating scale
ranges from zero (no visible disease symptoms) to four (dead
plant). Supscripts above bars indicate significant differences
according to a Tukey’s test
123
Genet Resour Crop Evol
Brighton cite a report of 2n=9x= 108 from Graham
(Graham 1958), who concludes that the accession in
question likely is a backcross of M. 9maximilianea
F.W. Schultz (M. aquatica xM. rotundifolia) with M.
aquatica.
Genotyping revealed a single, major group that
included both the M. aquatica and the M. aquatica var.
citrata accessions. Three pairs of accessions of M.
aquatica were closely related and distinguished by a
single allele each: CMEN 116, CMEN 117; CMEN
110, CMEN 111; and CMEN 121, CMEN 122. The
consecutive local numbers of each pair of closely
related accessions indicate they are in neighboring
pots and may suggest plant contamination, which will
be investigated further.
The M. aquatica accessions displayed a range of
Verticillium wilt resistance to susceptibility, similar to
that found previously in M. longifolia germplasm
(Vining et al. 2005).
Two accessions that are listed in the Germplasm
Resources Information Network (GRIN) as M. aquat-
ica, CMEN 724 and CMEN 728, were tetraploid,
whereas all of the other M. aquatica accessions were
found to be octoploid or eneaploid. DNA fingerprint-
ing placed these two accessions closer to M. suave-
olens than to M. aquatica, each in a distinct cluster.
Both accessions are likely Mentha hybrids that are not
closely related to each other.
Mentha suaveolens
The M. suaveolens accessions had mostly upright
growth habits, but variation in leaf morphology and
flower color (Table 2). Oil types fell into three major
Fig. 5 Relative ratings of Verticillium wilt susceptibility for M.
suaveolens accessions. Means and standard deviations are
shown for three biological replicates per accession. The rating
scale ranges from zero (no visible disease symptoms) to four
(dead plant). Supscripts above bars indicate significant differ-
ences according to a Tukey’s test
123
Genet Resour Crop Evol
categories: piperitenone oxide, (-)-carvone, and
trans-piperitenone oxide.
There were 11 triploid M. suaveolens accessions,
six of which had previously been reported as diploid
(Chambers and Hummer 1994). Ploidy levels have
implications for fertility, and therefore for intra- and
interspecies hybridization. Interspecific hybridization
among Mentha species with different ploidy levels is
common and can yield unpredictable results including
ploidy series and aneuploidy (Tucker 2012). This
study expands on and complements previous work
assessing ploidy of Mentha accessions at the Corvallis,
Oregon NCGR (Chambers and Hummer 1994).
In contrast to M. aquatica, most of the M. suave-
olens accessions were highly Verticillium wilt-resis-
tant. Genes involved in Verticillium wilt resistance
were initially identified in tomato (Kawchuk et al.
2001), and subsequent studies in tomato and cotton
have provided insight into the signaling pathways
mediating the wilt resistance response. The Ve1 gene,
first cloned from tomato (Kawchuk et al. 2001),
encodes a cell surface leucine-rich receptor protein
that functions to initiate a signaling cascade dependent
on EDS1 (Enhanced Disease Susceptibility 1), NDR1
(Non-race-specific Disease Resistance 1), MKK2
(Mitogen-activated protein Kinase Kinase 2), and
positive regulators of Ve1 (Fradin et al. 2009). In
tomato, the signaling pathway overlaps with that of
Cladosporium fulvum resistance (Fradin et al. 2009).
In cotton, homologs of Ve1,NDR1 and MKK2, have
been shown to be required for Verticillium wilt
resistance (Gao et al. 2011). Downstream generic
disease resistance responses involving the salicylic
acid (SA) and jasmonic acid (JA) pathways have also
been implicated (Shaban et al. 2018). Ve1 homologs
0.0
1.0
2.0
3.0
4.0
0 20406080100
Verticillium susceptibility score [scale 0-4]
Oil constituents [% of total oil]
Fig. 6 Correlation analysis of oil constituents versus Verticil-
lium susceptibility scores. The rating scale for Verticillium
susceptibility ranges from zero (no visible disease symptoms) to
four (dead plant). Symbols: (?)-menthofuran, filled circles;
linalool and linalyl acetate, hollow squares; trans-piperitone
oxide, hollow triangles; (-)-carvone, hollow diamonds; and
piperitenone oxide, letter ‘x’
Table 6 Allelic diversity among 24 M. aquatica, 23 M. suave-
olens and 2 M. longifolia accessions surveyed in this study.
SSR motif, allele size range, number of alleles and Nei’s
genetic diversity across all samples are listed. For each species,
the number of alleles and Nei’s genetic diversity (in parenthe-
sis) are provided for each SSR
SSR Motif Allele range No of alleles Nei allele diversity No. of alleles (Nei allele diversity) in
across samples across samples across samples M. longifolia M. aquatica M. suaveolens
C667 (CAT)11 149–284 15 0.92 5 (0.93) 12 (0.88) 13 (0.90)
C852 (AGC)5 174–186 5 0.66 2 (0.67) 3 (0.60) 5 (0.64)
C870 (CCT)4 273–329 13 0.83 1 (0) 11 (0.74) 7 (0.63)
C1739 (CAT)11 191–223 10 0.78 3 (1) 9 (0.82) 5 (0.72)
C2645 (GAA)9 243–314 17 0.88 4 (1) 13 (0.88) 11 (0.74)
C3787 (ATTT)4 132–190 10 0.78 2 (0.67) 8 (0.75) 4 (0.60)
C4573 (CCT)5 175–188 4 0.57 1 (0) 3 (0.57) 4 (0.60)
C4701 (TTGTAA)9 107–172 14 0.83 4 (1) 11 (0.83) 10 (0.80)
C5388 (CAGCAC)9 121–174 7 0.68 3 (0.83) 3 (0.53) 5 (0.68)
Average 10.56 0.77 2.78 (0.68) 8.11 (0.73) 7.11 (0.70)
c
Fig. 7 Neighbor-joining tree showing based on SSR genotyp-
ing of Mentha accessions
123
Genet Resour Crop Evol
123
Genet Resour Crop Evol
have been cloned from M. longifolia and from
cultivated mints (Vining and Davis 2009). Given the
apparent cross-talk among different disease resistance
response pathways in plant Verticillium wilt resis-
tance, it is possible that different disease resistance
genes or alleles confer the wilt resistance phenotype in
different Mentha species.
Both M. aquatica and M. suaveolens had male-
sterile accessions, with underdeveloped stamens lack-
ing pollen. These accessions have the potential to be
convenient female parents in crosses, as painstaking
emasculation of mint flowers can be avoided. The
genetic basis of this male sterility trait has yet to be
determined in mint.
Current status of SSR markers for differentiating
mint accessions
The nine SSRs developed in this study separated the
accessions mostly according to species and not based
on ploidy, oil composition, or geographical origin.
They amplified across the three species and identified
each accession as unique. Out of 45 SSR primer pairs
designed based on the available assembly of the M.
longifolia, 24 (53.3%) amplified and were polymor-
phic in all three species. This polymorphism rate may
be improved when using the newer assembly of M.
longifolia and after in silico detection of polymor-
phism by aligning SSR sequences of the reference
genome to that of re-sequenced mint accessions as
recently reported by Bhattarai and Mehlenbacher
(2017) in hazelnut and Iorizzo et al. (Iorizzo et al.
2011) in carrot.
Currently, a single study has identified and evalu-
ated 54 SSRs for mint isolated from a public database
of M. 9piperita sequences (Kumar et al. 2015). Of
these 54 SSRs, 33 were polymorphic in 13 accessions
M. 9piperita sequences but revealed low informa-
tion content as determined by an average of 2.33
alleles/SSR and an average expected heterozygosity
and polymorphism information content (PIC) of 0.21,
and 0.25, respectively. The low information content
could have been the result of the small number of
accessions in that study. Of the six SSRs that cross-
amplified in the four species evaluated by Kumar et al.
the three that amplified trinucleotide-containing SSRs
were evaluated in this study and either did not amplify
across the M. suaveolens and M. aquatica accessions
in the evaluation panel, or were difficult to score and
could not be used in this study. While in this study the
information content of the SSRs was considerably
higher than that observed in the Kumar et al. study
based on the higher average number of alleles/SSR
(10.56 vs. 2.33) and expected heterozygosity (0.77 vs.
0.21), the number of SSRs in Mentha is very low and
more SSRs need to be identified and evaluated across
Mentha species.
Lack of correlation between oil constituents
and Verticillium resistance
No correlation was found between the presence of
specific oil constituents and Verticillium wilt resis-
tance. Therefore, there does not appear to be a genetic
linkage between these characters, which is advanta-
geous for the breeding and selection of mint cultivars
that incorporate a combination of desirable oil com-
position and wilt resistance.
Altogether, this study highlights the importance of
the Mentha collection to continuing genetic and
genomic studies, and to mint breeding efforts.
Compliance with ethical standards
Conflict of interest The authors declare no conflict of interest.
References
Bhattarai G, Mehlenbacher SA (2017) In silico development and
characterization of tri-nucleotide simple sequence repeat
markers in hazelnut (Corylus avellana L.). PLOS ONE
12(5):e0178061. https://doi.org/10.1371/journal.pone.
0178061
Bruvo R, Michiels NK, D’Souza TG, Schulenburg H (2004) A
simple method for the calculation of microsatellite geno-
type distances irrespective of ploidy level. Mol Ecol
13(7):2101–2106
Chambers HL, Hummer KE (1994) Chromosome counts in the
Mentha collection at the USDA: ARS National Clonal
Germplasm Repository. Taxon 43(3):423. https://doi.org/
10.2307/1222717
Chen H, Chung M-C, Tsai Y-C, Wei F-J, Hsieh J-S et al (2015)
Distribution of new satellites and simple sequence repeats
in annual and perennial Glycine species. Bot Stud 56(1):22
Dung JK, Schroeder BK, Johnson DA (2010) Evaluation of
Verticillium wilt resistance in Mentha arvensis and M.
longifolia genotypes. Plant Dis 94(10):1255–1260
FAOSTAT (2018). http://www.fao.org/faostat/en/#data.
Accessed 31 July 2018
Fradin EF, Zhang Z, Juarez Ayala JC, Castroverde CDM, Nazar
RN et al (2009) Genetic dissection of Verticillium wilt
123
Genet Resour Crop Evol
resistance mediated by tomato Ve1. Plant Physiol
150(1):320–332. https://doi.org/10.1104/pp.109.136762
Gao X, Wheeler T, Li Z, Kenerley CM, He P et al (2011)
Silencing GhNDR1 and GhMKK2 compromised cotton
resistance to Verticillium wilt. Plant J 66(2):293–305.
https://doi.org/10.1111/j.1365-313X.2011.04491.x
Gilmore BS, Bassil NV, Hummer KE (2011) DNA extraction
protocols from dormant buds of twelve woody plant gen-
era. J Am Pomol Soc 65(4):201–206
Graham RA (1958) Mint notes, VII: Mentha 9maximileanea
F. Schultz in Britain
Greilhuber J, Temsch EM, Loureiro JCM (2007) Nuclear DNA
content measurement. Flow cytometry with plant cells.
Wiley, London, pp 67–101
Harley RM, Brighton CA (1977) Chromosome numbers in the
genus Mentha L. Bot J Linn Soc 74(1):71–96
Iorizzo M, Senalik DA, Grzebelus D, Bowman M, Cavagnaro
PF et al (2011) De novo assembly and characterization of
the carrot transcriptome reveals novel genes, new markers,
and genetic diversity. BMC Genomics 12:389. https://doi.
org/10.1186/1471-2164-12-389
Jedrzejczyk I, Rewers M (2018) Genome size and ISSR markers
for Mentha L. (Lamiaceae) genetic diversity assessment
and species identification. Ind Crops Prod 120:171–179.
https://doi.org/10.1016/j.indcrop.2018.04.062
Kamvar ZN, Tabima JF, Gru
¨nwald NJ (2014) Poppr: an R
package for genetic analysis of populations with clonal,
partially clonal, and/or sexual reproduction. PeerJ 2:e281
Kamvar ZN, Brooks JC, Gru
¨nwald NJ (2015) Novel R tools for
analysis of genome-wide population genetic data with
emphasis on clonality. Front Genet 6:208
Kawchuk LM, Hachey J, Lynch DR, Kulcsar F, van Rooijen G
et al (2001) Tomato Ve disease resistance genes encode
cell surface-like receptors. Proc Natl Acad Sci USA
98(11):6511–6515. https://doi.org/10.1073/pnas.
091114198
Kumar B, Kumar U, Yadav HK (2015) Identification of EST–
SSRs and molecular diversity analysis in Mentha piperita.
Crop J 3(4):335–342. https://doi.org/10.1016/j.cj.2015.02.
002
Lattier JD, Chen H, Contreras RN (2017) Improved method of
enzyme digestion for root tip cytology. HortScience
52(7):1029–1032
Paradis E, Claude J, Strimmer K (2004) APE: analyses of
phylogenetics and evolution in R language. Bioinformatics
20(2):289–290
Ringer KL, McConkey ME, Davis EM, Rushing GW, Croteau R
(2003) Monoterpene double-bond reductases of the (-)-
menthol biosynthetic pathway: isolation and characteriza-
tion of cDNAs encoding (-)-isopiperitenone reductase and
(?)-pulegone reductase of peppermint. Arch Biochem
Biophys 418(1):80–92
Schuelke M (2000) An economic method for the fluorescent
labeling of PCR fragments. Nat Biotech 18:233–234.
https://doi.org/10.1038/72708
Shaban M, Miao Y, Ullah A, Khan AQ, Menghwar H et al
(2018) Physiological and molecular mechanism of defense
in cotton against Verticillium dahliae. Plant Physiol Bio-
chem 125:193–204. https://doi.org/10.1016/j.plaphy.2018.
02.011
Tucker IAO (2012) Genetics and breeding of the genus Mentha:
a model for other polyploid species with secondary con-
stituents. J Med Active Plants 1(1):7
Tucker AO, Naczi RF (2007) Mentha: an overview of its clas-
sification and relationships. In: Lawrence BM (ed) Mint:
the genus Mentha. CRC Press, Boca Raton, pp 1–39
Vining K, Davis T (2009) Isolation of a Ve homolog, mVe1, and
its relationship to Verticillium wilt resistance in Mentha
longifolia (L.) Huds. Mol Genet Genomics
282(2):173–184. https://doi.org/10.1007/s00438-009-
0454-6
Vining KJ, Zhang Q, Tucker AO, Smith C, Davis TM (2005)
Mentha longifolia (L.) L.: a model species for mint genetic
research. HortScience 40(5):1225–1229
Vining KJ, Johnson SR, Ahkami A, Lange I, Parrish AN et al
(2017) Draft genome sequence of Mentha longifolia and
development of resources for mint cultivar improvement.
Mol Plant 10(2):323–339. https://doi.org/10.1016/j.molp.
2016.10.018
You FM, Huo N, Gu YQ, Luo M, Ma Y et al (2008) BatchPri-
mer3: a high throughput web application for PCR and
sequencing primer design. BMC Bioinform 9(1):253
Zurn JD, Carter KA, Yin MH, Worthington M, Clark JR et al
(2018) Validating blackberry seedling pedigrees and
developing an improved multiplexed microsatellite fin-
gerprinting set. J Am Soc Hortic Sci 143(5):381–390
Publisher’s Note Springer Nature remains neutral with
regard to jurisdictional claims in published maps and
institutional affiliations.
123
Genet Resour Crop Evol
A preview of this full-text is provided by Springer Nature.
Content available from Genetic Resources and Crop Evolution
This content is subject to copyright. Terms and conditions apply.
