RESEARCH

Genetic Diversity and Structure of Pepper

(Capsicum Annuum L.) from Northwestern Mexico
Analyzed by Microsatellite Markers
Antonio Pacheco-Olvera,* Sergio Hernndez-Verdugo, Vctor Rocha-Ramrez,
Antonio Gonzlez-Rodrguez, and Ken Oyama

ABSTRACT
The analysis of the variability and genetic structure of wild and landrace populations of pepper
(Capsicum annuum L.) is important for the management and conservation of valuable genetic
resources and to understand the consequences
of domestication on the patterns of neutral
genetic variation. For this purpose, 12 populations of wild peppers, 3 landrace populations
and 7 hybrid populations from northwestern
Mexico were studied using microsatellites. On
average, 3.62 alleles per locus were detected
in the wild relatives, 3.37 in the landraces, and
3.08 in the hybrids. According to the average
values of expected heterozygosity (He), slightly
greater genetic diversity was found among the
wild relatives (He = 0.466) than in the hybrids
(He = 0.440) or the landraces (He = 0.422). In
terms of the average number of alleles per
locus and the average expected heterozygosity, reductions of 8.18 and 10.25% were found
in the genetic diversity of the landraces and
hybrids, respectively, with respect to the wild
populations. The genetic differentiation among
the populations was the highest among hybrids
(GST = 0.324), followed by landraces (0.309) and
wild relatives (the lowest, at 0.208). Cluster analysis clearly demarcated the wild relatives and
domesticated populations into different groups.
The high levels of genetic diversity found among
C. annuum in northwestern Mexico suggest that
the wild and landrace populations are a valuable
resource that should be conserved.

A. Pacheco-Olvera and S. Hernndez-Verdugo, Facultad de Agronoma,

Univ. Autnoma de Sinaloa, Culiacn, Sinaloa, Mxico, Carretera Culiacn-Eldorado km 17.5; V. Rocha-Ramrez, A. GonzlezRodrguez, and K. Oyama, Centro de Investigaciones en Ecosistemas,
Univ. Nacional Autnoma de Mxico (UNAM), Antigua Carretera a
Ptzcuaro No 8701, Colonia San Jos de la Huerta, 58190, Morelia
Michoacn, Mxico. Received 14 June 2011. *Corresponding author
(apo@uas.uasnet.mx).
Abbreviations: A, average number of alleles per locus; F IS, inbreeding coefficient; F IT, total inbreeding; F ST, genetic differentiation; GD,
genetic distance; G ST, coefficient of genetic differentiation; H E HO
P, percentage of polymorphic loci; PCR, polymerase chain reaction;
RAPD, random amplified polymorphic DNA; SSR, simple sequence
repeat, or microsatellite; UPGMA, unweighted pair group method
with arithmetic averaging.

enetic resources have been utilized through programs of

domestication and improvement (Berthaud, 1997), promoting the use and distribution of plants under different ecological,
cultural, and technological conditions in relation to the centers of
origin (Harlan, 1992; Casas and Barbera, 2002). Artificial selection has produced morphological, physiological, and genetic
changes that modify mating systems and the genetic structure
of the populations of genetic resources (Hawkes, 1983; Doebley,
1989). Domestication is a relatively recent occurrence (the past
10,000 yr), and for this reason, it is expected that crops will exhibit
less genetic variation and little neutral genetic differentiation with
respect to their progenitors, despite the large morphological differences among crops and their wild relatives (Doebley, 1989, 1992).
The genus Capsicum originated in South America and includes 29
Published in Crop Sci. 52:231241 (2012).
doi: 10.2135/cropsci2011.06.0319
Published online 11 Nov. 2011.
Crop Science Society of America | 5585 Guilford Rd., Madison, WI 53711 USA
All rights reserved. No part of this periodical may be reproduced or transmitted in any
form or by any means, electronic or mechanical, including photocopying, recording,
or any information storage and retrieval system, without permission in writing from
the publisher. Permission for printing and for reprinting the material contained herein
has been obtained by the publisher.

CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

231

species (Hunziker, 1979; Eshbaugh 1980; Pickersgill, 1984;

Hernndez-Verdugo et al., 2001), of which 5 (C. annuum,
C. chinense, C. frutense, C. baccatum, and C. pubescens) have
been domesticated (Pickersgill, 1997). Peppers were one of
the first plants to be domesticated in the Americas, and they
are used worldwide as a spice, condiment, and a vegetable
(MacNeish, 1964). Capsicum annuum is the most economically important species and includes the majority of peppers
in the Americas, Africa, and Asia as well as several nonpungent sweet materials that grow in temperate regions of
Europe and North America (Pickersgill, 1997). The wild
populations of this species are known as Capsicum annuum
var. glabriusculum (Heiser and Pickersgill, 1975; HernndezVerdugo et al., 1999), which is considered the progenitor of
domesticated peppers. These populations, which have the
common name chiltepin or piquines, are widely distributed throughout Mexico. Individual plants are generally
found under trees in tropical deciduous forests, orchards,
and grazing pastures (Hernndez-Verdugo et al., 1999).
All the wild forms are perennials, herbs, or climbers.
Their fruits are small, red, and spicy, and they are eaten by
birds, which spread their seeds (Vzquez-Dvila, 1996). Capsicum annuum var. annuum has been collected, cultivated, and
consumed in Mexico for hundreds of years. Plantings began
approximately 8000 yr ago, followed by the cultivation and
definitive domestication of the spicy fruits approximately
6000 yr ago (Perry and Kent, 2007). Since pre-Columbian
times indigenous communities have selected the characteristics with the greatest agricultural importance, such as the
shape and size of the fruit (Pickersgill et al., 1979), which has
resulted in high morphological variation and a large number of landrace varieties that are adapted to local conditions.
The landraces are native varieties that have been cultivated
for long periods of time with intermediate but stable performance under a system of agricultural management. These
varieties constitute an important genetic resource that is in
danger of being lost because these crops are being displaced
by improved hybrids with higher yields (Fri, 1986; Zeven,
1998), whose demand in the market has increased over the
last 30 yr. However, these hybrid varieties have the disadvantage of being susceptible to plagues and diseases inherent
to the crop. Therefore, a study of the variability and genetic
structures of wild and landrace populations is important for
the management and conservation of this valuable genetic
resource. Such variability constitutes an important genetic
reservoir that can help solve agricultural problems, for example, by providing resistance to attack from plagues and diseases (Hernndez-Verdugo et al., 1998; Votava et al., 2005).
Species of the genus Capsicum have been studied using
morphological, cytogenetic, and molecular markers with
different results depending on the origin of the materials
(germplasm bank), the level of domestication of the species
(wild and cultivated), and the genetic markers used (McLeod
et al., 1983; Loaiza-Figueroa et al., 1989; Conicella et al.,
232

1990; Lefebvr et al., 1993; 2001; Prince et al., 1992, 1995;

Hernndez-Verdugo et al., 2001; Oyama et al., 2006).
Recently, Aguilar-Melndez et al. (2009) examined
the nucleotide diversity of the sequences of the Dhn (dehydrin, an osmotic stress-related gene), G3pdh (gene encoding glyceraldehyde 3-phosphate dehydrogenase), and Waxy
(gene sequence encoding granule-bound starch synthase)
genes in 58 semi-wild populations and 22 domesticated
populations. The genetic diversity was found to be similar
among semi-wild populations (populations with morphological characteristics that are similar to those of wild populations and that are grown with minimal management) and
domesticated populations of Capsicum.
The objective of this study was to compare the population genetic variation and structures of wild relatives,
landraces, and hybrid populations of C. annuum to understand the consequences of the domestication process on
the patterns of neutral genetic variation. For this purpose,
microsatellites (also called simple sequence repeats, or
SSRs) were used, as these tools have been applied to a
great variety of plants owing to the high degree of polymorphism, codominance, and reproducibility (Haghnazari et al., 2005). In addition, the identification and
characterization of numerous microsatellite loci for the
Capsicum genus have recently been reported (Lee et al.,
2004; Minamiyama et al., 2006; Portis et al., 2007; Huang
et al., 2000; Ince et al., 2010).

MATERIALS AND METHODS

Plant Materials
The plant material was sampled in northwestern Mexico, in
the states of Nayarit, Sinaloa, and Sonora. In total, 12 wild
populations of C. annuum were collected, covering a latitudinal interval of approximately 767 km from 2124 to 2711 N
(Table 1, Fig. 1). Eleven of the wild populations were sampled
in deciduous tropical forests in the states of Sonora and Sinaloa,
and the other population was sampled within a commercial
orchard at the location of El Llano in the state of Nayarit. The
domesticated populations of C. annuum were sampled in the
crop fields of Sinaloa and correspond to the landrace and hybrid
groups. The landrace peppers belong to the Tequila, Cascabel,
and Cola de Rata types. The hybrids correspond to the commercial types, Hngaro, Guajillo, Anaheim, Cambray, Serrano,
Poblano, and Chilaca. Leaves were sampled from 20 plants that
were randomly selected from each population. The leaves were
immediately frozen in liquid nitrogen and fi nally transported
to the laboratory and stored at 80C until DNA extraction.

DNA Extraction and Amplication

of Microsatellites
The total DNA was extracted from young leaves using the
method of Lefort and Douglas (1999). Initially, 14 microsatellite primer pairs developed for the Capsicum genus were tested,
but only 7 were selected to be scored in all sampled individuals
on the basis of the reproducibility of the results and the levels of

WWW.CROPS.ORG

CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

variation revealed. These seven were the following: EPMS342,

EPMS417, EPMS440, EPMS480, EPMS501 (Nagy et al., 2007),
EPMS643, and EPMS919 (Portis et al., 2007). Polymerase chain
reactions (PCR) were performed using a PCR multiplex kit
(QIAGEN, Hombrechtikon, Switzerland) with 5-L reactions as
follows: 1X PCR multiplex Master Mix, 2 M of each primer,
distilled water, and 20 ng of DNA. The temperature profi le for
PCR amplification consisted of a cycle of 95C for 5 min, followed by 35 cycles of 94C for 1 min, 60C for 1.5 min, and
72C for 1 min. After cycling, there was a final elongation step at
72C for 5 min. The PCR-amplified products were analyzed by
capillary electrophoresis using an ABI 3100-Avant (Applied Biosystems, Hitachi, Japan) automated sequencer. Each of the samples analyzed in the sequencer contained 10 L of formamide, 0.4
L of Gene Scan 500 LIZ, and 1.0 L of the PCR product. For
visualization of the alleles, the Peak Scanner program version 1.0
(Applied Biosystems) was used.

Statistical Analyses
The genetic diversity of the populations was quantified according to the average number of alleles per locus (A), the percentage
of polymorphic loci (P), the observed heterozygosity (HO), and
the expected heterozygosity (H E) using the GDA program version 1.0 (Lewis and Zaykin, 2001). To assess whether the seven
loci analyzed provide independent information, a test of pairwise
linkage disequilibrium was conducted using Arlequin software
version 3.0 (Excoffier et al., 2005). The genetic structure of the
populations was calculated with Wrights F statistics (FIS, FST, and
FIT) using GenAlex software versions 6.4 (Peakall and Smouse,
2006). To determine whether the FIS and FIT values were significantly different from zero with respect to the Hardy-Weinberg
equilibrium, a X 2 test was performed using the following equation: X2 = F2N(k1) with [k(k1)]/2 degrees of freedom, where
N is the size of the sample and k is the number of loci analyzed (Li

Table 1. Collection sites of Capsicum annuum populationsstudied.

Population Abbreviation Status

Lat

Long

Mocuzary
Estacin
Cuaternaria
Potrero de
Alcantar
Rancho el
Coyote

MCU
EST

Wild
Wild

27 11 48.60 109 07 25.62

27 05 00.00 109 33 00.00

POA

Wild

26 54 00.00 108 53 00.00

RCY

Wild

26 53 55.50 108 46 28.80

Tehueco
Lode Vega
Pilares
Chapeteado
Alcoyonqui
Tabal
Otates
Llano
Tequila
Cascabel
Cola de Rata

TEH
LVE
PIL
CHA
ALC
TAB
OTA
LLA
TEQ
CAS
CRA

Wild
Wild
Wild
Wild
Wild
Wild
Wild
Wild
Landrace
Landrace
Landrace

26 20 00.00
26 22 00.00
25 04 00.00
24 29 00.00
24 54 00.00
24 24 00.00
23 02 00.00
21 24 00.00
22 43 07.50
22 43 07.50
22 41 01.80

108 45 00.00
108 40 00.00
107 21 00.00
107 26 00.00
107 12 00.00
107 05 00.00
105 55 00.00
105 10 00.00
105 10 00.00
105 51 29.58
105 47 48.84

Hungaro
Guajillo
Anahim
Cambray
Serrano
Poblano
Chilaca

HUN
GUA
ANA
CAM
SER
POB
CHI

Hybrid
Hybrid
Hybrid
Hybrid
Hybrid
Hybrid
Hybrid

22 45 19.68
22 41 01.80
23 01 25.20
22 53 47.22
23 12 00.18
23 01 25.20
22 43 53.16

105 52 43.02
105 47 48.84
106 10 16.62
105 58 16.56
106 13 49.50
106 10 16.62
105 50 50.04

and Horvitz, 1953). To determine the significance of FST, the test

X 2 = 2NFST(k1) was performed with [(k1)(s-1)] degrees of freedom, where s is the number of subpopulations (Workman and
Niswander, 1970). The genetic diversity within each subpopulation (HS) was calculated, along with the genetic diversity of

Figure 1. Map showing the distribution of populations of Capsicum annuum studied in Northwestern Mexico.
CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

WWW.CROPS.ORG

233

the total population (HT), the total genetic diversity distributed

among populations (D ST), and the measurement of population
differentiation (G ST) for the seven loci (Nei, 1978), where G ST
= D ST/HT and D ST = HT HS. These statistics were estimated
using FSTAT software version 2.9.3.2 (Goudet, 1995). In addition, a nonparametric analysis of molecular variance (AMOVA)
was performed as described by Stewart and Excoffier (1996)
using four hierarchical levels: (1) wild vs. landrace vs. hybrid
populations, (2) wild populations, (3) landrace populations, and
(4) hybrid populations. The analysis was performed using Arlequin software version 3.0 (Excoffier et al., 2005). Finally, Bayesian clustering was performed using the STRUCTURE program
(Pritchard et al., 2000), where all individuals of the three groups
of peppers were analyzed together under a mixed model with
correlated allelic frequencies. To obtain the most probable K
value (number of genetic groups), values of K from 1 to 5 were
tested, with 10 independent runs for each K. The longitude of
the run was 500,000 steps followed by 106 iterations. The K value
with the greatest probability was calculated estimating the maximum value of the K statistic, according to Evanno et al. (2005).
The genetic similarity relationships among the populations
under study were estimated by the unweighted pair group method
with arithmetic averaging (UPGMA) using Neis genetic distances (1972), calculated with the Tools for Population Genetics
Analyses (TFPGA) program version 1.3 (Miller, 1997). Similarly,
an Analysis of Principal Coordinates was performed with Neis
genetic distances (1972) using the GenAlex software version 6.4
(Peakall and Smouse, 2006). With the goal of testing the isolation by distance model in the wild populations, the correlation
between pairwise genetic distances and pairwise geographic distances among populations was analyzed with a Mantel test using
TFPGA software version 1.3 (Miller,1997).
To determine the possible genetic barriers in the distribution of the wild populations of C. annuum from the northwest
of Mexico, the Monmoniers maximum difference algorithm
was applied (Monmonier, 1973) with the BARRIERS program version 2.2 (Manni et al., 2004). For the analysis of the
barriers, Neis genetic distances (1983) were employed and estimated for the 12 wild populations.

RESULTS
Population Genetic Variation
In wild populations, an average number of alleles per locus
(A) of 3.6 was detected with the seven SSR markers, with
a range from 2.6 (Mocuzari) to 5.0 (Estacin Cuaternaria).
In the landrace populations, A was 3.4, with a range from
3.0 (Tequila) to 3.8 (Cola de rata), and in the hybrid populations A was 3.1, with a range from 2.0 (Poblano) to 4.0
(Cambray) (Table 2). The percentage of polymorphic loci
(P) was 89.3% among the wild populations, with a range
from 71.4% (Alcoyonqui and Tabal) to 100% (various
populations). Among landrace populations, the average
was 80.2% and the minimum and maximum values were
71.4 (Cascabel) and 85.7% (Tequila), respectively. In the
hybrid populations, the average was 93.2%, with a minimum of 66.7 (Poblano) and a maximum of 100% (various
populations). The average HO and H E were, respectively,
234

0.225 and 0.466 for the wild populations, 0.201 and 0.422
for the landrace populations, and 0.237 and 0.440 for the
hybrid populations (Table 2). The differences in the average values of A, P, HO, and H E among the three groups of
pepper populations were not significant. Among the three
groups of pepper populations that were analyzed, the
average H E value exceeded the HO values, which indicates
that there is an excess of homozygous individuals. The
test for pairwise linkage disequilibrium after a Bonferroni
correction was nonsignificant in all cases, indicating that
the seven analyzed loci provide independent information.

Genetic Differentiation among Populations

The average HT obtained with the seven microsatellite loci
for the wild populations was 0.658, with a range from 0.216
(EPMS480 locus) to 0.885 (EPMS417). For the landrace
populations, the average HT was 0.648, with a range from
0.461 (EPMS342) to 0.821 (EPMS417). For the hybrid peppers, the average HT was 0.725, with a range from 0.625
(EPMS440) to 0.868 (EPMS919) (Table 3). The average
intrapopulation genetic diversity for each group was highest
for the wild peppers (HS = 0.501), followed by the hybrid
populations (HS = 0.493) and then by the landraces (HS=
0.460). The average coefficient of genetic differentiation
(G ST) was 0.208 for the wild populations, with a range from
0.037 (EPMS440) to 0.295 (EPMS919). For the landraces,
the average was 0.309, with a range from 0.033 (EPMS501)
to 0.710 (EPMS480). For the hybrid populations, the average
G ST value was 0.324, with a range from 0.192 (EPMS440)
to 0.474 (EPMS342) (Table 3). The average inbreeding
coefficient (FIS) within the wild populations was positive
and statistically different from zero (0.412) (Table 4). The
only value that was not different from zero was found for
the EPMS440 locus, which indicates Hardy-Weinberg
equilibrium. The landrace populations presented positive
values that were different from zero at all of their loci, with
an average FIS value of 0.542. The hybrid populations presented positive values that were different from zero at five
loci. The total inbreeding (FIT) among the three groups of
peppers varied from 0.709 among the landrace peppers to
0.560 for the wild peppers. These results indicate that there
is a high level of genetic differentiation among the populations of C. annuum for the three levels of domestication.
The genetic differentiation was greatest among the hybrid
populations (FST = 0.424), followed by the landraces (FST=
0.396), and finally the wild populations (FST=0.297). The
analysis of molecular variation at the three hierarchical levels (Table 5) showed that most of the variation occurred
within the populations (74.2%). However, a considerable
proportion of the variation was found between populations
(15.7%). The remaining variation (10.1%) was distributed
among the three population groups (wild, landraces, and
hybrids), indicating a clear distinction among these groups.

WWW.CROPS.ORG

CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

Relationships between Populations

The cluster analysis performed with the UPGMA algorithm generated a dendrogram that divided the populations into two genetic groups: group I, which was formed
by two landrace populations mixed with the hybrids; and
group II, which was formed by the wild populations, the
Tequila landrace population, and the Chilaca hybrid (Fig.
2). The parental lines of commercial hybrids are unknown.
It is possible that the Chilaca variety was obtained from
parental lines that were introgressed with genes from wild
populations. For the Neis (1972) genetic distance (GD), all
the pepper populations presented an average value of 0.703,
with a minimum value of 0.112 and a maximum value of
2.975. Within the wild populations, the average GD value
was 0.455, with a minimum distance of 0.112 (EST and
PIL) and a maximum distance of 1.121 (LLA and LVE). The
landrace populations presented an average distance of 0.745,
with a minimum distance of 0.566 (CAS and CRA) and a
maximum distance of 0.955 (CAS and TEQ). The hybrid
populations presented an average distance of 0.908, with a
minimum value of 0.402 (HUN and CHI) and a maximum
value of 2.201 (SER and ANA). The landrace populations
were separated from the wild ones by a distance of 0.290,
and the hybrid populations were separated from the wild
ones by a distance of 0.453. Finally, the hybrid populations
were interlaced with the landrace populations.
In the principal coordinate analysis, coordinate 1
explained 45.0% and coordinate 2 explained 16.3% (Fig. 3) of
the total variation. The greatest discrimination was observed
only for the axes of the two first main coordinates. The axis
of coordinate 1 clearly separated the wild populations from
the landrace and hybrid populations, yet the landrace populations were grouped with the hybrid populations, reflecting
similarity among these two groups of domesticated peppers.
The analysis with the STRUCTURE program and the
subsequent evaluation of the K statistic indicated that K = 2
is the most probable number of clearly differentiated genetic
groups of wild and domesticated C. annuum in northwestern
Mexico. One group is formed by the wild populations and
the other by the landrace and hybrid populations (Fig. 4).
However, in this test, the Tequila landrace and the Chilaca

Table 2. Genetic variation parameters for wild, landrace, and

hybrid populations of Capsicum annuum studied: A, number
of alleles per locus; P, percentage of polymorphic loci; HO,
observed heterozygosity; and HE, expected heterozygosity.
Population

Wild
Mocuzary
Estacin Cuaternaria
Potrero de Alcantar
Rancho el Coyote
Tehueco
Lode Vega
Pilares
Chapeteado
Alcoyonqui
Tabal
Otates
Llano
Average
Landraces
Tequila
Cascabel
Cola de Rata
Average
Hybrids
Hungaro
Guajillo
Anaheim
Cambray
Serrano
Poblano
Chilaca
Average

HO

HE

2.57
5.00
3.86
3.14
3.29
2.71
4.00
3.86
4.71
3.86
3.29
3.14
3.62

85.71
100.00
85.71
100.00
85.71
100.00
100.00
85.71
71.43
71.43
100.00
85.71
89.29

0.253
0.315
0.197
0.255
0.238
0.215
0.196
0.209
0.246
0.167
0.214
0.196
0.225

0.384
0.592
0.533
0.446
0.461
0.447
0.509
0.473
0.452
0.516
0.448
0.327
0.466

3.00
3.29
3.83
3.37

85.71
71.42
83.33
80.16

0.263
0.168
0.172
0.201

0.449
0.358
0.459
0.422

2.86
3.42
2.86
4.00
3.43
2.00
3.00
3.08

85.71
100.00
100.00
100.00
100.00
66.67
100.00
93.20

0.232
0.271
0.302
0.179
0.277
0.111
0.289
0.237

0.466
0.519
0.468
0.473
0.461
0.191
0.504
0.440

hybrid population presented evidence of introgression by the

wild relatives. The test of isolation by distance performed
only with the wild populations indicated a significant and
positive correlation between the geographic distance matrix
(Km) and the genetic distance (r = 0.476; P = 0.03) (Fig. 5).
Using the BARRIERS program, five genetic barriers in the
distribution of the wild populations of C. annuum in northwestern Mexico were found with bootstrap-support greater
than 59% (Fig. 6). The first barrier (Barrier 1) separated the

Table 3. Total (HT ), within (HS), and between (DST ) population genetic diversity and the genetic differentiation coefcient (GST )
for wild, landrace, and hybrid populations of Capsicum annuum.
Wild
Locus
EPMS480
EPMS501
EPMS342
EPMS417
EPMS643
EPMS440
EPMS919
Mean
SD

Landraces

Hybrids

HT

HS

DST

GST

HT

HS

DST

GST

HT

HS

DST

GST

0.216
0.878
0.771
0.885
0.843
0.237
0.776
0.658
0.298

0.189
0.681
0.549
0.714
0.595
0.229
0.548
0.501
0.209

0.027
0.197
0.223
0.171
0.248
0.009
0.229
0.158
0.099

0.123
0.224
0.289
0.194
0.294
0.037
0.295
0.208
0.099

0.574
0.649
0.461
0.821
0.798
0.650
0.582
0.648
0.127

0.166
0.628
0.284
0.685
0.627
0.412
0.415
0.460
0.195

0.408
0.022
0.177
0.136
0.171
0.238
0.168
0.188
0.117

0.710
0.033
0.383
0.166
0.214
0.366
0.288
0.309
0.214

0.606
0.769
0.644
0.861
0.700
0.625
0.868
0.725
0.110

0.391
0.583
0.339
0.650
0.386
0.504
0.595
0.493
0.122

0.215
0.185
0.305
0.210
0.314
0.120
0.272
0.232
0.070

0.355
0.241
0.474
0.244
0.448
0.192
0.314
0.324
0.108

CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

WWW.CROPS.ORG

235

Table 4. Wrights F statistics of wild, landrace, and hybrid Capsicum annuum populations.
Locus
EPMS480
EPMS501
EPMS342
EPMS417
EPMS643
EPMS440
EPMS919
Mean
SD

FIS
0.172***
0.604***
0.651***
0.547***
0.479***
0.064
0.475***
0.412***
0.260

Wild
FIT

FST

Landraces
FIT

FIS

FST

Hybrids
FIT

FIS

FST

0.312***
0.713***
0.809***
0.693***
0.696***
0.027

0.169***
0.274***
0.454***
0.323***
0.416***
0.070

0.856***
0.522***
0.500***
0.515***
0.466***
0.408***

0.961***
0.556***
0.712***
0.642***
0.767***
0.641***

0.728***
0.071*
0.425***
0.263***
0.564***
0.394***

0.476***
0.687***
0.064
0.134
0.402***
0.288***

0.678***
0.776***
0.549***
0.558***
0.689***
0.599***

0.385***
0.282***
0.518***
0.489***
0.480***
0.437***

0.673***
0.560***
0.283***

0.377***
0.297***
0.138

0.525***
0.542***
0.144

0.682***
0.709***
0.129

0.331***
0.396***
0.210

0.698***
0.393***
0.249

0.812***
0.666***
0.103

0.378***
0.424***
0.082

*Signicant at the 0.05 probability level.

** Signicant at the 0.01 probability level.
***Signicant at the 0.001 probability level.

Table 5. Analysis of molecular variance (AMOVA) conducted on seven simple sequence repeat loci of wild, landrace, and hybrid
Capsicum annuum populations.
Sources of variation
Wilds vs. Landraces vs. Hybrids
Among populations
Within populations
Wilds
Among populations
Within populations
Landraces
Among populations
Within populations
Hybrids
Among populations
Within populations

Df

Variance components

% Total

2
19
624

Sum of squares
17123.863
35880.797
162911.60

35.45589
55.41416
261.07628

10.07
15.75
74.18

P < 0.0001
P < 0.0001
P < 0.0001

11
338

19356.601
72752.001

52.96179
215.24261

19.75
80.25

P < 0.0001
P < 0.0001

2
83

2131.91
38103.067

21.2162
459.07309

4.42
95.58

P < 0.0001
P < 0.0001

6
203

14392.286
52056.533

71.40927
256.43612

21.78
78.22

P < 0.0001
P < 0.0001

Figure 2. Dendrogram based on Neis (1972) genetic distance, applying the unweighted pair-group method with arithmetic averaging
(UPGMA) clustering algorithm, between wild (W), landrace (L), and hybrid (H) populations of Capsicum annuum.

LLA population from the Sinaloa and Sonora populations.

The second barrier (Barrier 2) is located in the central part of
the state of Sinaloa, separating the TAB populations from the
CHA, PIL, and ALC populations. The third barrier (Barrier
3) is located between the OTA populations and the rest of
236

the populations present in Sinaloa and Sonora. The fourth

barrier (Barrier 4) separated the populations of the north
of Sinaloa and Sonora. Finally, the fifth barrier (Barrier 5)
is located between the CHA population and the PIL and
ALCpopulations.

WWW.CROPS.ORG

CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

Figure 3. Associations between 22 Capsicum annuum populations revealed by principal coordinate analysis of Neis (1972)
geneticdistances.

Figure 5. A Mantels test for correlation between genetic distance

(Nei, 1972) and geographic distance (km) showed a moderate
correlation, r = 0.476, P < 0.030 from 10,000 randomizations.

Figure 4. (A) Mean and standard deviation in InP (D) for ve

independent runs of STRUCTURE plotted against the number of
genetic groups (K) used in the analysis. (B) Values of K plotted
against K. In both cases the peak indicates the most probable
number of genetic groups given the data.

DISCUSSION
In this study, we analyzed the variation and genetic structure of wild, landrace, and hybrid populations of C. annuum of northwestern Mexico. The results indicated high
levels of genetic diversity both within and among the populations of C. annuum. The genetic variation in our study
was greater than that found in previous studies of wild and
domesticated populations of the Capsicum genus in Mexico
that were performed with samples from germplasm banks
with isozyme markers (Loaiza-Figueroa et al., 1989).
CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

The results of the current study more closely coincide

with previous studies performed on natural populations
with isozymes and random amplified polymorphic DNA
(RAPD) (Hernndez-Verdugo et al., 2001; Oyama et al.,
2006). The genetic diversity of the domesticated peppers
as compared with the wild ones has slightly decreased at
the seven loci studied, both in terms of allelic diversity
and in terms of expected heterozygosity. However, hybrid
populations showed higher expected heterozygosity than
landraces. This could be a result of heterozygote formation during the breeding process of hybrid varieties. In
contrast, landraces are maintained through generations of
open pollination, including self-pollination. Among the
landrace populations, we found an average reduction of
8.2% in genetic diversity, and in the hybrid populations
we found a reduction of 10.3%. A similar pattern was
found with RAPDs (Oyama et al., 2006), with a genetic
erosion of only ~8%. Similarly, for three functional genes
(Dhn, G3ph, and Waxy), the domesticated peppers have

WWW.CROPS.ORG

237

Figure 6. Map showing the distribution of Capsicum annuum used in the present study (abbreviations indicate the collection sites of
populations). Each pie chart represents the proportions in each population of the two genetic groups as assigned by the program
STRUCTURE. Red and green represent the genetic groups corresponding to wild and domesticated populations, respectively, and blue
lines represent calculated barriers based on genetic distances with the BARRIERS software.

retained 91% of the diversity found in wild C. annuum

peppers in Mexico (Aguilar-Melndez et al., 2009). This
reduction in genetic diversity among domesticated populations may be associated with the domestication process.
Our results show a positive coefficient of inbreeding for
all the populations, with the landrace populations presenting
the highest average value (0.542) and the hybrid populations
presenting the lowest average value (0.393). These levels of
inbreeding may be due to the reproduction system of this
self-pollinating species, which shows only 7.8 to 38.6% cross
pollination (Ballester and de Vicente, 1998). However, these
high to moderate levels of inbreeding may also be the product
of the limited movement of genes among populations (Elam,
1998), which by itself may cause significant substructuring
and an increase in inbreeding (Wright, 1943; Rohlf and Schnell, 1971; Turner et al., 1982; Soakal and Wartenberg, 1983;
Hamrick and Loveless, 1986). In contrast to previous studies
of isoenzymes of wild and domesticated C. annuum in northwestern Mexico (Hernndez-Verdugo et al., 2001), where
the average values of genetic differentiation were 0.056 for
the wild populations and 0.167 for the domesticated populations, in this study we found that the domestication process
has changed the distribution of the genetic variation of the
domesticated populations in comparison with the wild populations. In the wild populations, we found average GST values
of 0.208 and FST values of 0.287, while in the landraces peppers we found an average GST value of 0.309 and an average
FST value of 0.396. Finally, we found greater differentiation
238

among the modern pepper varieties (hybrids), with an average GST value of 0.324 and an average FST value of 0.424. It is
possible that this genetic differentiation among domesticated
populations is increasingly more intense than that observed
among the wild populations. The average GST values in this
study are similar to the average values obtained with RAPDs
(Oyama et al., 2006) among wild and domesticated populations of C. annuum in northwestern Mexico. Normally a
genetic differentiation among populations (FST) greater than
0.15 is considered to indicate considerable genetic differentiation (Frankham et al., 2002).
The differences in the GST averages among the isoenzymes and the DNA markers (RAPDs and microsatellites)
may be due to the variation of some enzymatic loci that are
not selectively neutral and may determine deviations in the
estimates of the parameters that describe the genetic structure
of the populations (Ferguson et al., 1998). On the basis of a
matrix of genetic distances (Nei, 1972), a dendrogram was
constructed with the UPGMA method. This dendrogram
separated the pepper populations into two groups, one grouping the wild populations and the other the cultivated pepper
populations, with the exception of the Tequila (landrace) and
Chilaca (hybrid) populations, which were grouped with the
wild populations. The principal coordinate analysis showed
tight clustering among the wild populations, while among
the landrace and hybrid populations there is greater separation, reflecting greater genetic differentiation among these
populations. An analysis was performed with the Bayesian

WWW.CROPS.ORG

CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

algorithm implemented by the STRUCTURE program,

with the goal of determining the most probable number
of genetic groups. This analysis produced a most probable
value of K = 2. However, this analysis also suggested various
degrees of mixing among the two main genetic groups (wild
and domesticated). The significance of Mantels test reflected
isolation by distance in the structure of the wild populations
being studied, where the populations that are the most distant
from each other are genetically more different than populations that are geographically closer. The natural isolation by
distance results in limited gene flow, where the probability of
gene flow between two populations is a function of the geographic distance between them. In the selfing populations
of wild peppers, where little or no pollen flow is present,
the gene flow must be produced by the movement of seeds,
particularly through fruit consumption, with the seeds later
disseminated by birds (Laborde and Pozo-Campodnico
1982; Pozo-Campodnico et al., 1991; Vzquez-Dvila,
1996). The results obtained from the BARRIERS program
identified important barriers that separate the majority of the
wild populations through the gradient of their distribution
with bootstrap support of >59%. These barriers may possibly
be due to the course of some riverbeds as well as to certain
changes in vegetation altitude and climate that may result in
some process of differentiation among the populations.

CONCLUSIONS
The populations of C. annuum of northwest Mexico may
be classified into two genetic groups. The first group is
formed by wild populations and the second group is formed
by landraces and hybrids. The genetic variation within
the wild, landrace, and hybrid groups is high, although
when comparing the landrace and hybrid population with
the wild pepper populations, there is a slight reduction in
A (alleles per locus) and in H E (expected heterozygosity).
The genetic differentiation among the domesticated pepper populations (C. annuum) is greater than that among
the wild populations.
Acknowledgments
We thank the National Council of Science and Technology,
CONACYT, Mxico, for financial support (research project
106129) as well as Universidad Autnoma de Sinaloa, project
PROFAPI 2008/084. We are grateful to Juan Pealoza Ramirez,
Ana Luisa Albarrn L, and Maria L. Herrera Arroyo for computer support and suggested improvements to the manuscript.

References
Aguilar-Melndez, A., L. Peter, L. Mikeal, and K. Seung-Chul.
2009. Genetic diversity and structure in semiwild and domesticated chiles (Capsicum annuum; Solanaceae) from Mexico.
Am. J. Bot. 96:11901202. doi:10.3732/ajb.0800155
Ballester, J., and M.C. de Vicente. 1998. Determination of F1
hybrid seed purity in pepper using PCR-based markers.
Euphytica 103:223226. doi:10.1023/A:1018372523343
CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

Berthaud, J. 1997. Strategies for conservation of genetic resources

in relation with their utilization. Euphytica 96:112.
doi:10.1023/A:1002922220521
Casas, A., and G. Barbera. 2002. Mesoamerican domestication and
diff usion. p. 143162. In P.S. Nobel (ed.) Cacti: Biology and
uses. The Univ. of California Press, Los Angeles, CA.
Conicella, C., A. Errico, and F. Saccardo. 1990. Cytogenetic and
isozyme studies of wild and cultivated Capsicum annuum.
Genome 33:279282.
Doebley, J. 1989. Isozymic evidence and evolution of crop plants.
p. 165191. In E.D. Soltis and P.M. Soltis (ed.) Isozymes in
Plant Biology. Discorides, Portland, OR.
Doebley, J. 1992. Molecular systematic and crop evolution. p. 202
222. In P.S. Soltis, D.E. Soltis, and J.J. Doyle (ed.) Molecular
systematic of plants. Chapman & Hall, New York, London.
Elam, D. 1998. Population genetics of vernal pool plants: Theory, data and conservation implications. p. 180189. In C.W.
Witham, E. Baunder, D. Belk, W. Ferren, and R. Ornduff (ed.)
Ecology, conservation and management of vernal pool ecosystems. California Native Plants Society, Sacramento, CA.
Eshbaugh, W.H. 1980. The taxonomy of the genus Capsicum
(Solanaceae). Phytologia 47:153166.
Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number of clusters of individuals using the software Structure:
A simulation study. Mol. Ecol. 14:26112620. doi:10.1111/
j.1365-294X.2005.02553.x
Excoffier, L., G. Laval, and S. Schneider. 2005. Arlequin ver. 3.0:
An integrated software package for population genetics data
analysis. Evol. Bioinform. Online 1:4750.
Fri, M. 1986. Pepper (Capsicum annuum L.). p. 344362. In Y.P.S.
Bajaj (ed.) Biotechnology in Agriculture and Forestry 2.
Springer-Verlag, Berlin Heidelberg.
Ferguson, M.E., H.J. Newbury, N. Maxted, B.V. Ford-Lloyd, and
L.D. Robertson. 1998. Population genetic structure in Lens
taxa revealed by isozyme and RAPD analysis. Genet. Resour.
Crop Evol. 44:257269.
Frankham, R., J.D. Ballou, and D.A. Briscou. 2002. Introduction
to conservation genetics. Cambridge, UK.
Goudet, J. 1995. Fstat version 1.2: A computer program to calculate F-statistics. J. Hered. 86:485486.
Haghnazari, A., R. Samimifard, J. Najafi, and M. Mardi. 2005.
Genetic diversity in pea (Pisum sativum L.) accessions detected
by sequence tagged microsatellite markers. J. Genet. Breed.
59:145152.
Hamrick, J.L., and M.D. Loveless. 1986. The influence of seed dispersal mechanisms on the genetic structure of plant populations.
p. 211223. In A. Estrada and T.H. Fleming (ed.) Frugivores
and seed dispersal. Junk Publ., The Hague, the Netherlands.
Harlan, J.C. 1992. Crops and man, second ed. ASA and CSSA,
Madison, WI.
Hawkes, J.G. 1983. The diversity of crop plants. Harvard Univ.
Press, Cambridge, MA.
Heiser, C.B., and B. Pickersgill. 1975. Names for the bird peppers
(Capsicum-Solanaceae). Baileya 19:151153.
Hernndez-Verdugo, S., R.G. Guevara-Gonzlez, R.F. RiveraBustamante, C. Vzquez-Yaes, and K. Oyama. 1998. Los
parientes silvestres del chile (Capsicum spp.) como recursos
genticos. Bol. Soc. Bot. Mex. 62:171181 (In Spanish).
Hernndez-Verdugo, S., P. Dvila, and K. Oyama. 1999. Sntesis del conocimiento taxonmico, origen y domesticacin del
gnero Capsicum. Bol. Soc. Bot. Mex. 64:6584 (In Spanish).
Hernndez-Verdugo, S., R. Luna-Reyes, and K. Oyama. 2001.

WWW.CROPS.ORG

239

Genetic structure and differentiation of wild and domesticated populations of Capsicum annuum from Mxico. Plant
Syst. Evol. 226:129142. doi:10.1007/s006060170061
Huang, S., B. Zhang, D. Milbourne, L. Cardle, G. Yang, and
J. Guo. 2000. Development of pepper SSR markers from
sequence databases. Euphytica 117:163168.
Hunziker, A.T. 1979. South American Solanaceae: A synoptic survey. p. 4985. In J.K. Hawkes, R.L. Lester, and A.D. Skelding
(ed.) Biology and taxonomy of Solanaceae. Linnean Society
Symposium, Series, no. 7. Academic Press, New York.
Ince, A.G., M. Karaca, and A. Naci. 2010. Polymorphic microsatellite markers transferable across Capsicum species. Plant Mol.
Biol. Rep. 28:285291. doi:10.1007/s11105-009-0151-y
Laborde, C.J., and O. Pozo-Campodnico. 1982. Presente y
pasado del chile en Mxico. Publicacin Especial Num. 85.
SARH-INIA. Mxico.
Lee, J.M., S.H. Nahm, Y.M. Kim, and B.D. Kim. 2004. Characterization and molecular genetic mapping of microsatellite
loci in pepper. Theor. Appl. Genet. 108:619627. doi:10.1007/
s00122-003-1467-x
Lefebvr, V., A. Palloix, and M. Rives. 1993. Nuclear RFLP
between pepper cultivars (Capsicum annuum L.). Euphytica
71:189199. doi:10.1007/BF00040408
Lefebvr, V., B. Goffinet, J.C. Chauvet, B. Caromel, P. Signoret,
R. Brand, and A. Palloix. 2001. Evaluation of genetic distances
between pepper inbred lines for cultivar protection purposes:
Comparison of AFLP, RAPD and phenotypic data. Theor.
Appl. Genet. 102:741750. doi:10.1007/s001220051705
Lefort, F., and G.C. Douglas. 1999. An efficient micro-method of
DNA isolation from mature leaves of four hardwood tree species Acer, Fraxinus, Prunus and Quercus. Ann. For. Sci. 56:259
263. doi:10.1051/forest:19990308
Lewis, P.O., and D. Zaykin. 2001. Genetic data analysis: Computer
program for the analysis of allelic data. Version 1.0 (d16c).
Free program distributed by the authors over the Internet
from http://lewis.eeb.uconn.edu/lewishome/software.html
(verified 27 Oct. 2011).
Li, C.C., and D.G. Horvitz. 1953. Some methods of estimating
inbreeding coefficient. Am. J. Hum. Genet. 5:107117.
Loaiza-Figueroa, F., K. Ritland, C.J. Laborde, and S.D. Tanksley. 1989. Patterns of genetic variation of the genus Capsicum (Solanaceae) in Mexico. Plant Syst. Evol. 165:159188.
doi:10.1007/BF00936000
MacNeish, R.S. 1964. Ancient mesoamerican civilization. Science
143:531537. doi:10.1126/science.143.3606.531
Manni, F., E. Gurard, and E. Heyer. 2004. Geographic patterns
of (genetic, morphologic, linguistic) variation: How barriers
can be detected by Monmoniers algorithm. Hum. Biol.
76:173190. doi:10.1353/hub.2004.0034
McLeod, M.J., S.I. Guttman, W.H. Eshbaugh, and R.E. Rayle.
1983. An electrophoretic study of evolution in Capsicum
(Solanaceae). Evolution 37:562574. doi:10.2307/2408269
Minamiyama, Y., M. Tsuro, and M. Hirai. 2006. An SSR-based
linkage map of Capsicum annuum. Mol. Breed. 18:157169.
doi:10.1007/s11032-006-9024-3
Miller, M.P. 1997. Tool for Population Genetic Analyses (TFPGA)
version 1.3. A Windows program for the analysis of allozyme
and molecular population genetic data.
Monmonier, M. 1973. Maximum-difference barriers: An alternative
numerical regionalization method. Geogr. Anal. 3:245261.
Nagy, I., A. Stgel, Z. Sasvri, M. Rder, and M. Ganal. 2007.
Development, characterization and transferability to other

240

Solanaceae of microsatellite markers in pepper (Capsicum annuum L.). Genome 50:668688. doi:10.1139/G07-047
Nei, M. 1972. Genetic distance between populations. Am. Nat.
106:283292. doi:10.1086/282771
Nei, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89:583590.
Oyama, K., S. Hernndez-Verdugo, C. Snchez, A. GonzlezRodrguez, P. Snchez-Pea, J.A. Garzn-Tiznado, and A.
Casas. 2006. Genetic structure of wild and domesticated populations of Capsicum annuum (Solanaceae) from northwestern
Mexico analyzed by RAPDs. Genet. Resour. Crop Evol.
53:553562. doi:10.1007/s10722-004-2363-1
Peakall, R., and P.E. Smouse. 2006. GENALEX 6: Genetic analysis in Excel. Population genetic software for teaching and
research. Mol. Ecol. Notes 6:288295. doi:10.1111/j.14718286.2005.01155.x
Perry, L., and V. Kent. 2007. Precolumbian use of chili peppers
in the valley of Oaxaca, Mxico. Proc. Natl. Acad. Sci. USA
104:1190511909. doi:10.1073/pnas.0704936104
Pickersgill, B., C.B. Heiser, and J. McNeill. 1979. Numerical taxonomic studies on variation and domestication in some species of Capsicum. p. 679700. In J.G. Hawkes, R.N. Lester,
and A.D. Skelding (ed.) The biology and taxonomy of the
Solanaceae. Academic Press, London.
Pickersgill, B. 1984. Migration of chili peppers, Capsicum spp., in
the Americas. p. 105123. In D. Stone (ed.) Papers of the Peabody Museum of Archeology and Ethnology, Vol. 76. Harvard Univ. Press.
Pickersgill, B. 1997. Genetic resources and breeding of Capsicum
spp. Euphytica 96:129133. doi:10.1023/A:1002913228101
Portis, E., I. Nagy, Z. Sasvri, A. Stgel, L. Barchi, and S. Lantery.
2007. The design of Capsicum spp. SSR assays via analysis of in silico DNA sequence, and their potential utility for
genetic mapping. Plant Sci. 172:640648. doi:10.1016/j.
plantsci.2006.11.016
Pozo-Campodnico, O., H. Montes, and J. Redondo. 1991. Chile
(Capsicum spp.). p. 217238. In Avances en el estudio de los
recursos fitogenticos de Mxico, Sociedad Mexicana de
Fitogentica A.C. Mexico. (In Spanish).
Prince, J.P., E. Pochard, and S.D. Tanksley. 1992. Restriction fragment length polymorphism and genetic distance among Mexican accessions of pepper. Genome 36:404417. doi:10.1139/
g93-056
Prince, J.P., V.K. Lackney, C. Angeles, J.R. Blauth, and M.M.
Kyle. 1995. A survey of DNA polymorphism within the genus
Capsicum and the fi ngerprinting of pepper cultivars. Genome
38:224231. doi:10.1139/g95-027
Pritchard, J.K., M. Stephens, and P. Donnelly. 2000. Inference of
population structure using multilocus genotype data. Genetics 155:945959.
Rohlf, F.J., and G.D. Schnell. 1971. An investigation of the
isolation-by-distance model. Am. Nat. 105:295324.
doi:10.1086/282727
Soakal, R.R., and D.E. Wartenberg. 1983. A test of spatial autocorrelation analysis using an isolation-by-distance model.
Genetics 105:219237.
Stewart, C.N., and L. Excoffier. 1996. Assessing population
genetic structure and variability with RAPD data: Application to Vaccinium macrocarpon (American Cranberry). J. Evol.
Biol. 9:153171. doi:10.1046/j.1420-9101.1996.9020153.x
Turner, M.E., J.C. Stephens, and W.W. Anderson. 1982. Homozygosity and patch structure in plant populations as a result

WWW.CROPS.ORG

CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

of nearest-neighbor pollination. Proc. Natl. Acad. Sci. USA

79:203207. doi:10.1073/pnas.79.1.203
Vzquez- Dvila. M.A. 1996. El amash y el pistoqu: Un ejemplo
de la etnoecologa de los chontales de Tabasco, Mxico. Etnoecolgica 3:5963.
Votava, E., B. Jit, and Bosland. 2005. Genetic diversity of chile
(Capsicum annuum var. annuum L.) landraces from northern
New Mexico, Colorado, and Mexico. Econ. Bot. 59:817.

CROP SCIENCE, VOL. 52, JANUARY FEBRUARY 2012

doi:10.1663/0013-0001(2005)059[0008:GDOCCA]2.0.CO;2
Workman, P., and J. Niswander. 1970. Populations studies on
southwestern Indians tribes. Local genetic differentiation in
the Papago. Am. J. Hum. Genet. 22:2449.
Wright, S. 1943. Isolation by distance. Genetics 28:114138.
Zeven, A.C. 1998. Landraces: A review of defi nitions and classifications. Euphytica 104:127139. doi:10.1023/A:1018683119237

WWW.CROPS.ORG

241