Professional Documents
Culture Documents
Garcia Rodriguez Et Al 2011 Mexican Sardine Morphometrics
Garcia Rodriguez Et Al 2011 Mexican Sardine Morphometrics
Fisheries Research
journal homepage: www.elsevier.com/locate/fishres
A study of the population structure of the Pacific sardine Sardinops sagax (Jenyns,
1842) in Mexico based on morphometric and genetic analyses
Francisco Javier García-Rodríguez a,∗ , Silvia Alejandra García-Gasca b , José De La Cruz-Agüero a ,
Víctor Manuel Cota-Gómez a
a
Centro Interdisciplinario de Ciencias Marinas, Instituto Politécnico Nacional, Departamento de Biología Marina y Pesquerías, Colección Ictiológica, Av. Instituto Politécnico Nacional
s/n, Col. Playa Palo de Santa Rita 23096, Apdo, Postal 592, La Paz, Baja California Sur, Mexico
b
Centro de Investigación en Alimentación y Desarrollo, A.C. Av. Sábalo – Cerritos s/n, Estero del Yugo, Apdo, Postal P 711, Mazatlán, Sinaloa, Mexico
a r t i c l e i n f o a b s t r a c t
Article history: Several studies on the Pacific sardine Sardinops sagax have focused on the identification of stock compo-
Received 25 April 2010 sition and boundaries, using morphometric and genetic analysis. In this study, geometric morphometric
Received in revised form 4 November 2010 body landmarks and control region mtDNA sequences were used to examine the population structure
Accepted 4 November 2010
of sardines along the Pacific coast of the Baja California Peninsula. Samples from commercial landings
in Ensenada (ENS), Baja California, and Bahia Magdalena (BM), Baja California Sur, were obtained during
Keywords:
2006–2007. The population hypotheses tested were based on the distribution of sea surface temperature
Pacific sardines
(SST) along the coast, which was previously used to define stocks. A total of 275 sardines from ENS and
Sardinops
Mitochondrial DNA
119 from BM were used in morphometric analysis. Fifty-three sequences from ENS and 106 from BM were
Control region used for genetic comparisons. Morphometric results showed differences among the three groups based
Genetic differentiation on SST, suggesting the existence of different morphotypes. Percentage of molecular variance explained
Morphometric analysis by the differences among three groups was significantly different from zero. However, the distribution of
Mexico haplotypes in the groups did not show a clear phylogeographic pattern. Additionally, mismatch distribu-
tions supported relatively similar historical demographic events in the three groups. Although evidence
of phenotypic groups along the Pacific coast of the Peninsula was found, current molecular data did not
clearly support the existence of a phylogeographically structured population.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction ture studies on the Pacific sardine in Mexican and California have
been based on various kinds of information, including tagging infor-
Biological and ecological knowledge about natural resources is mation (Clark, 1945) vertebral counts (Clark, 1947; Wisner, 1960),
relevant for devising management strategies, especially in species spawning areas (Marr, 1960), blood groups (Sprague and Vrooman,
with important conservation or commercial status. The Pacific 1962; Vrooman, 1964), size-at-age (Wolf and Daugherty, 1964),
sardine Sardinops sagax (Jenyns, 1842) is distributed from the morphometric data (De La Cruz-Agüero and García-Rodríguez,
southeastern coast of Alaska to the northwestern coast of Mexico, 2004), genetic analysis (Hedgecock et al., 1989; Lecomte et al.,
including the Gulf of California (Kramer and Smith, 1971). It is one of 2004; Gutiérrez-Flores, 2007) and cohort analysis (Félix-Uraga
the most important schooling pelagic species along the west coast et al., 1996).
of North America and is captured in Mexico, near Ensenada, Bahia Relevant information about the movements of fish and abun-
Magdalena, and Guaymas (Lluch-Belda et al., 1986). The Pacific dances of the Pacific sardine populations was obtained from the
sardine is a commercially valuable species. Capture records show intensive tagging program carried out between 1936 and 1944
fluctuations over time, with a near collapse during the mid 20th (Clark, 1945). This study found that fish tagged north of Bahia Sebas-
century (D.F.O., 2004). tian Viscaino, Mexico, were caught at northern sites, supporting
Several studies on the Pacific sardine have focused on the iden- the existence of only one stock with a distribution from British
tification of stocks. This is relevant because specific biological and Columbia to northern and central Baja California. No evidence was
ecological information of each stock is used to implement harvest found of movement toward the north of individuals tagged in Bahia
strategies aimed at achieving sustainable exploitation. Stock struc- Magdalena (Clark, 1945). Studies carried out on the seasonal and
geographic distributions of larvae show that the principal spawning
areas are centered in three areas (Marr, 1960): off Central California
∗ Corresponding author. Tel.: +52 612 12 25344; fax: +52 612 12 25322. in April (Lynn, 2003); near Bahia Magdalena, Baja California Sur, in
E-mail address: fjgarciar@ipn.mx (F.J. García-Rodríguez). summer; and in the Gulf of California in fall and winter (Aceves-
0165-7836/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.fishres.2010.11.002
170 F.J. García-Rodríguez et al. / Fisheries Research 107 (2011) 169–176
Table 1
USA Sampling sites. Data sets are grouped according to sampling site and sea surface
temperature (SST). Upper data were used for morphometric analysis. Lower data
were used for genetic analysis. Cold Ensenada (CEN), Temperate Bahia Magdalena
(TBM) and Warm Bahia Magdalena (WBM).
Northern 1
zone Sites Date SST Groups n Total
2.1. Sampling
Fig. 3. Sampling zones along of the year and groups defined by zone and sea surface temperature (SST) in Ensenada and Bahia Magdalena from June 2006 to May 2007.
Dotted lines represent the stock limits previously suggested (Félix-Uraga et al., 2005).
the shape, two templates of the digital image were constructed the principal components (PC) had significantly different vari-
to provide guidelines of equal angular spacing to identify points ances (Anderson, 1958). PC significant scores were used to compare
along the body curves using the program MakeFan (H.D. Sheets, groups using ANOVA. A matrix of the assignments was constructed
http://www2.canisius.edu/∼sheets/morphsoft.html). First, a tem- to complement previous analysis by assigning each specimen to
plate was constructed based on the landmarks at the end of the one of the known groups (based on the Mahalanobis distance from
branchiostegals rays, on the snout, and at the origin of the dor- the specimen to the mean value of the nearest group).
sal fin. A second template was based on landmarks located at Since the CVA suggested significant differences among groups,
the origin of the anal fin, the end of the dorsal fin, and the partial Procrustes distance means (PPDMs) were calculated to
origin of the upper lobe of the caudal fin. Additional points (semi- perform paired comparisons. The significance of the test was
landmarks) were digitized at the intersection of the curve and the based on bootstrapping to determine whether the observed
lines of the templates. Thus, we constructed the digitized con- F-value could have been produced by chance, taking into
figurations using 18 points, 16 along the contour and two on account the distribution of bootstrapped F-values. This analy-
the side of the fish (Fig. 4) using the program TpsDig Ver 1.4 sis was carried out using the TwoGroup6 software (IMP, H.D.
(Rohlf, 2004). A superimposition method based on generalized Pro- Sheets, http://www2.canisius.edu/∼sheets/morphsoft.html). Dis-
crustes analysis (GPA) was used to remove differences attributed tances obtained were used to construct an unrooted tree based
to the position, orientation, and scale between configurations. on Neighbor-Joining (NJ) using Phylip Ver 3.6 (Felsenstein, 2005).
Semi-landmarks were submitted to the alignment algorithm to Finally, the thin-plate spline interpolating functions were used to
reduce effects of the arbitrary selection of a limited number of visualize shape changes.
points to represent the curves, using the program SemiLand6 (H.D.
Sheets, http://www2.canisius.edu/∼sheets/morphsoft.html). Once
the semi-landmarks were aligned, they were treated as points in 2.3. Genetic analysis
landmark data sets.
Partial warp scores (the contribution that each partial Caudal fin samples were collected in 1.5 mL microtubes contain-
warp makes to the total deformation) were obtained from ing absolute ethanol and stored at −20 ◦ C until laboratory analysis.
the Thin Plate Spline interpolation function using IMP pro- Total DNA was isolated by taking ∼0.5 g of caudal fin and using
grams. They were subjected to a Principal Component Anal- the “salting out” method (Miller et al., 1988). Isolated DNA was re-
ysis (PCA) and Canonical Variates Analysis (CVA) using PCA- suspended in 100 L deTE (Tris–EDTA pH 8.0). A fragment of the
Gen6n and CVAGen6m software, respectively (IIMP, H.D. Sheets, control region (D-loop) of mtDNA was amplified using the primers
http://www2.canisius.edu/∼sheets/morphsoft.html). We used the reported by Bowen and Grant (1997). PCR amplification was car-
Chi square procedure in the PCAGen6n program to test whether ried out in 12.5 L reactions containing 1× PCR buffer with 1.5 mM
MgCl2 (Clontech), 131.25 M of each dNTP, 0.4 M of each primer,
0.5 U Advantage Taq DNA polymerase (Clontech), and 1 L of DNA.
The PCR setup consisted of an initial denaturation step at 94 ◦ C
for 2 min, followed by 35 cycles at 94 ◦ C for 1 min, 55 ◦ C for 1 min,
and 68 ◦ C for 2 min. Amplification products were purified using the
4 5 6
3 7 8
2
9
1 18 10 Qiagen MiniElute kit following the instructions suggested by the
11
manufacturer, and sequenced using an automated DNA sequencer
(LICOR IR2 ).
17
14
13 12 Sequences were aligned and edited using the BioEdit soft-
16 15
ware (Hall, 1999), which uses the ClustalW algorithm. Haplotype
and nucleotide diversity were calculated for each data set using
Fig. 4. Schematic representation of the Pacific sardine showing points used for mor-
Arlequin 3.0 (Excoffier et al., 2005). Population genetic struc-
phometric geometric analysis. Points 2, 3, 4, 5, 8, 9 and 10 were found using two
reference systems. The first was based on points 1, 6 and 16, and the second was ture was analyzed using the Analysis of Molecular Variance
based on points 7, 11 and 14. (AMOVA), which estimates the proportion of genetic variation
172 F.J. García-Rodríguez et al. / Fisheries Research 107 (2011) 169–176
Table 3
Nucleotide substitutions for Parsimoniosus sites in Northern and Southern sampling sites based in a fragment of the control region of mtDNA. Position numbers correspond
to the site in the 500 bp sequence. Dashes represent similarities to the consensus (cons).
Position Zones
111111222222223333344
366000126022366772236745
HAPLOTYPE 758689059529825386945752 NORTH SOUTH
#cons CCTTCTGTAAGGGGGATAAGCGAG
#H65 ----T-----A------------- 4
#H101 -TC-T---GG------C-GAT--- 2
#H116 -----CA------------A---- 3
#H133 ------A-GG--A-----GAT--- 2
#H134 --CC--A--GAA------G-TA-- 2
#H135 --C--CA---A-A---CGGAT--- 2
#H140 ----T---GGA--A----GAT--- 2
#H141 -T---CAC-GA--A-G------G- 2
#H142 -------C-G------C--A-A-- 2
#H143 --C-TCA-G---A-----G-TAG- 2
F.J. García-Rodríguez et al. / Fisheries Research 107 (2011) 169–176 173
Fig. 6. Tree of morphometric divergence and configuration means of each morphotype. Morphotype WBM was morphometrically the most different. Sardines from the
northern zone (CEN) showed a less depressed body shape than those from the southern zone (WBM). Vectors represent the direction and magnitude of deformations taking
as reference the overall mean configuration (black dots).
uals were excluded from the analysis because the sampling date Table 5
Genetic diversity of DNA sequences used in genetic studies of the Pacific sardine.
was unknown. Therefore, this genetic analysis was based on 153
sequences (47 from CEN, 33 from TBM, and 73 from WBM). All of Mitochondrial gene H Reference
the excluded specimens had unique haplotypes. Nucleotide diver- Cytochrome b 0.89 0.0050 Lecomte et al. (2004)
sity was relatively larger in the CEN group and smaller in the WBM NAD6 0.94 0.0097 Gutiérrez-Flores (2007)
group (Table 4). NAD5 0.96 0.0101 Gutiérrez-Flores (2007)
AMOVA revealed significant genetic differences among Control Region 0.99 0.0182 Present study
Table 4
Samples size (n), number of haplotypes (nh), haplotype diversity (h) and nucleotide diversity () and standard deviation (SD) for each SST-group. Cold Ensenada (CEN),
Temperate Bahia Magdalena (TBM) and Warm Bahia Magdalena (WBM).
The weak phylogeographic structure detected in this study may be Clark, F.N., 1945. Results of tagging experiments in California Waters on Sardine
due to the high migratory ability of adult sardines (Clark, 1945) (Sardinops caerulea). Fish. Bull. 61, 93pp.
Clark, F.N., 1947. Analysis of populations of the Pacific sardine on the basis of verte-
and to the highly dynamic meso-scale currents in the Califor- bral counts. Fish. Bull. 65, 26pp.
nia Current System (Maluf, 1983; Kessler, 2006), which disperses D.F.O., 2004. Pacific Sardine. DFO Can. Sci. Advis. Sec. Stock Status Report 2004/037.
eggs and larvae. The Pacific sardine is a multiple-batch spawner De La Cruz-Agüero, J., García-Rodríguez, F.J., 2004. Morphometric stock structure
of the Pacific sardine Sardinops sagax (Jenyns, 1842) off Baja California, Mexico.
(Torres-Villegas, 1997), thus the potential for random genetic dif- In: Elewa, A.M. (Ed.), Morphometrics: Applications in Biology and Paleontology.
ferentiation from genetic drift may be reduced by overlapping Springer-Verlag, New York, NY, pp. 115–124.
generations. Retention eddies, on the other hand, may localize Emmett, R.L., Brodeur, R.D., Miller, T.D., Pool, S.S., Krutzikowsky, G.K., Bentley, P.J.,
Mccrae, J., 2005. Pacific sardine (Sardinops sagax) abundance, distribution, and
larvae long enough to produce both detectable morphological dif- ecological relationships in the Pacific northwest. CalCOFI Rep. vol. 46, 122–
ferences among areas and chaotic genetic structure. 143.
The demographic history, inferred from mismatch distribution, Excoffier, L., Laval, G., Schneider, S., 2005. Arlequin ver 3.0: An integrated software
package for population genetics data analysis. Evol. Bioinform. Online 1, 47–
also suggests similar evolutionary events among the three mor-
50.
photypes, although weak demographic expansion gradients were Félix-Uraga, R., Alvarado, R.M., Carmona, R., 1996. The sardine fishery along the west
noticed from North to South (Fig. 7). A right-shifted unimodal coast of Baja California, 1981–1994. CalCOFI Rep. 37, 188–192.
mismatch distribution found toward the northern zone suggests Félix-Uraga, R., Gómez-Muñoz, V.M., García-Franco, W., 2004. On the existence of
Pacific sardine groups off the West coast of Baja California and Southern Cali-
that the northern group represents an older demographic expan- fornia. CalCOFI Rep. 45, 146–151p.
sion than southern groups (see Rogers and Harpending, 1992). Félix-Uraga, R., Gómez-Muñoz, V.M., Quiñonez-Velázquez, C., Melo-Barrera, F.N.,
Similar northern-to-southern gradients were found by Lecomte Hill, K.T., García-Franco, W., 2005. Pacific sardine (Sardinops sagax) stock dis-
crimination off the West coast of Baja California and Southern California using
et al. (2004). otolith morphometry. CalCOFI Rep. 46, 113–121.
Our results suggest that at least three morphotypes occur on Felsenstein, J., 2005. PHYLIP (Phylogeny Inference Package) version 3.6. Distributed
the western coast of the Baja California Peninsula, which may cor- by the author. Department of Genome Sciences, University of Washington, Seat-
tle.
respond to the three stocks proposed by Félix-Uraga et al. (2004, Gutiérrez-Flores, C., 2007. Estructura genética poblacional de la sardina del pacífico
2005). Moreover, the morphometric data suggest that the most nororiental Sardinops sagax caeruleus. Tesis de Maestría, CICESE, Ensenada, B.C,
similar morphotypes were CEN and TBM. The morphotype pre- p. 112.
Hall, T.A., 1999. BioEdit: a user-friendly biological sequences alignment edit and
sumably originating from the Gulf of California (WBM) showed the analysis program for Windows 95/98/NT. Nucleic Acids Sympos. Ser. 41, 95–
greatest morphological differentiation. In agreement with previous 98.
analyses, morphological differences may be more related to phe- Hedgecock, D., 1994. Temporal and spatial genetic structure of marine animal pop-
ulations in the California Current. CalCOFI Rep. 35, 73–81.
notypic plasticity than to the genetic variation (De La Cruz-Agüero
Hedgecock, D., Hutchinson, E.S., Li, G., Sly, F.L., Nelson, K., 1989. Genetic and morpho-
and García-Rodríguez, 2004). Higher discrimination of morpho- metric variation in the Pacific sardine, Sardinops sagax caerulea: comparisons
type WBM, should therefore, be associated with environmental and contrasts with historical data and with variability in the northern anchovy,
factors in the Gulf of California. Common events caused by the Cal- Engraulis mordax. U.S. Fish Bull. 87, 653–671.
Hedgecock, D., Hutchinson, E.S., Li, G., Sly, F.L., Nelson, K., 1994. The central stock
ifornia Current and other oceanographic processes may promote of northern anchovy (Engraulis mordax) is not a randomly mating population.
the greater morphological similarity between CEN and TBM. The CalCOFI Rep. 35, 121–136.
results obtained from the partial warps analysis suggest an appar- Hewitt, R., 1981. Eddies and speciation in the California Current. CalCOFI Rep. XXII,
96–98.
ent morphological north-to-south pattern; individuals were more Hubbs, C.L., 1960. The marine vertebrates of the outer coast, Symp: the biogeography
compressed in the Gulf of California than in the Pacific. of Baja California and Adjacent Seas. Syst. Zool. 9 (3 and 4), 134–147.
Future analyses should be focused in the identification of Karl, S., Avise, J.C., 1992. Balancing selection at allozyme loci in oyster: implications
from nuclear RFLPs. Science 256, 100–102.
morphologic criteria to distinguish individuals of different morpho- Kessler, W.S., 2006. The circulation of the eastern tropical Pacific: a review. Prog.
types. Since phylogeographic structure is not clear, it is presently Oceanogr. 69, 181–217.
not possible to support the idea of a limited gene flow of the Koehn, R.K., Newell, R.I.E., Immerman, F.W., 1980. Maintenance of an aminopepti-
dase allele frequency cline by natural selection. Proc. Natl. Acad. Sci. U.S.A. 77,
Pacific sardine along its distribution area. Future genetic compar- 5385–5389.
isons considering samples from three different zones (northern Kramer, D., Smith, P.E., 1971. Seasonal and geographic characteristics of fishery
zone, southern zone, and Gulf of California) grouped on the limits resources, California current. Region VII. Pacific Sardine Com. Fish. Rev. 33 (10),
7–11.
of SST, and larger sample sizes for comparing temporal variation
Larson, R.J., Julian, R.M., 1999. Spatial and temporal genetic patchiness in marine
intra SST stock should clarify our results. populations and their implications for fisheries management. CalCOFI Rep. 40,
94–99.
Acknowledgements Lecomte, F., Grant, W.S., Dodson, J.J., Rodríguez-Sánchez, R., Bowen, B., 2004. Liv-
ing with uncertainty: genetic imprints of climate shifts in East Pacific anchovy
(Engraulis mordax) and sardine (Sardinops sagax). Mol. Ecol. 13, 2169–2182.
This study was funded by grants from the Secretaría de Investi-
Lluch-Belda, D., Crawford, R.J.M., Kawasaki, T., MacCall, A.D., Parrish, R.H., Schwart-
gación y Posgrado – Instituto Politécnico Nacional (SIP-20071113, zlose, R.A., Smith, P.E., 1989. World-wide fluctuations of sardine and anchovy
SIP-20080573 and SIP-20090333) and with the support of J.R. stocks: the regime problem. S. Afr. J. Mar. Sci. 8, 195–205.
Lluch-Belda, D., Magallon, F.J., Schwartzlose, R.A., 1986. Large fluctuations in the sar-
Torres-Villegas. FJGR and JDA thank the grants by EDI-IPN, COFAA-
dine fishery in the Gulf of California: possible causes. CalCOFI Rep. 27, 136–140.
IPN, and SNI-CONACYT. We thank the “Colección Ictiológica”, Lynn, R.J., 2003. Variability in the spawning habitat of Pacific sardine (Sardinops
CICIMAR-IPN. We thank Stewart Grant for commenting on an ear- sagax) off southern and central California. Fish. Oceanogr. 12 (3), 1–13.
lier draft of the paper and three anonymous reviewers for their Maluf, L.Y., 1983. The physical oceanography. In: Case y, T.J., Cody, M.L. (Eds.), Island
Biogeography in the Sea of Cortez. University of California Press, Los Angeles,
valuable suggestions and criticism. CA, pp. 26–44.
Marr, J.C., 1960. The causes of major variations in the catch of the Pacific sardine,
References Sardinops caerulea (Girard). In: Rosa, H., Murphy, G.I. (Eds.), Proceedings of the
World Scientific meeting on the Biology of Sardines and Related Species, vol. III.
Aceves-Medina, G., Jimez-Rosenberg, S.P.A., Hinojosa-Medina, A., Funes-Rodriguez, Food and Agriculture Organization of the United Nations, pp. 667–791.
R., Saldierna-Martinez, R.J., Smith, P.E., 2004. Fish larvae assemblages in the Gulf Miller, S.A., Dykes, D.D., Polesky, H.F., 1988. A simple salting out procedure for
of California. J. Fish Biol. 65, 832–847. extracting DNA from human nucleated cells. Nucleic Acids Res. 16, 1215.
Anderson, T.W., 1958. An Introduction to Multivariate Analysis. Wiley. Murphy, G.I., 1967. Vital statistics of the Pacific sardine (Sardinops caeruea) and the
Bowen, B.W., Grant, W.S., 1997. Phylogeography of the sardines (Sardinops population consequences. Ecology 48, 731–736.
spp.): assessing biogeographic models and population histories in temperate Pogson, G.H., Mesa, K.A., Boutilier, R.G., 1995. Genetic population structure and gene
upwelling zones. Evolution 51, 1601–1610. flow in the Atlantic cod, Gadus morhua: a comparison of allozyme and nuclear
Butler, J.L., Smith, P.E., Lo, N.C.H., 1993. The effect of natural variability of life- RFLP loci. Genetics 139, 375–385.
history parameters on anchovy and sardine population growth. CalCOFI Rep. Rogers, A.R., Harpending, H., 1992. Population growth makes waves in the distribu-
34, 104–111. tion of pairwise genetic differences. Mol. Biol. Evol. 9, 552–569.
176 F.J. García-Rodríguez et al. / Fisheries Research 107 (2011) 169–176
Rohlf, F.J., 2004. TpsDIG Version 1.40. Department of Ecology and Evolution, State Valentine, J., 1966. Numerical analysis of marine molluscan ranges on the extrat-
University of New York at Stony Brook, New York. ropical northeastern Pacific shelf. Limnol. Oceanogr. 11, 198–211.
Smith, P.E., 2005. A History for subpopulation structure in the Pacific sardine Vrooman, A.M., 1964. Serologically differentiated subpopulations of the Pacific sar-
(Sardinops sagax) population off Western North America. CalCOFI Rep. 46, dine, Sardinops caerulea. J. Fish. Res. Board Can. 21, 691–701.
75–82p. Waples, R.S., 1998. Separating the wheat from the chaff: patterns of genetic differ-
Sprague, L.M., Vrooman, A.M., 1962. A racial analysis of the Pacific sardine (Sardinops entiation in high gene flow species. J. Hered. 89, 438–450.
caerulea), based on stidies of erythrocyte antigens. Ann. N. Y. Acad. Sci. 97, Wisner, R.L., 1960. Evidence of northward movement of stocks of Pacific sardine
131–138. based on the number of vertebrae. CalCOFI Rep. 8, 75–82.
Torres-Villegas, J.R., 1997. La reproducción de la sardina monterrey Sardinops Wolf, R., Daugherty, A.E., 1964. Age and length composition of the sardine catch off
caerulea (Girard, 1854) en el Noroeste de México y su relación con el ambiente. the Pacific Coast of the United States and Mexico in 1961 and 1962. Calif. Fish
Ph.D. thesis, Universidad Politécnica de Cataluña. Game 50, 241–242.