tRNAVal mitochondrial marker for the identification and phylogenetic analysis of shrimp species belonging to the superfamily Penaeoidea

Analytical Biochemistry 391 (2009) 127–134

Contents lists available at ScienceDirect

Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio

Evaluation of a novel 16S rRNA/tRNAVal mitochondrial marker for the identification and phylogenetic analysis of shrimp species belonging to the superfamily Penaeoidea Pilar Calo-Mata a, Ananias Pascoal a, Inmaculada Fernández-No a, Karola Böhme a, José M. Gallardo b, Jorge Barros-Velázquez a,* a

Department of Analytical Chemistry, Nutrition and Food Science, School of Veterinary Sciences/College of Biotechnology, University of Santiago de Compostela, Campus Universitario Norte, E-27002 Lugo, Spain Department of Food Technology, Institute for Marine Research (IIM-CSIC), Higher Council for Scientific Research, E-36208 Vigo, Spain

b

a r t i c l e

i n f o

Article history: Received 28 March 2009 Available online 18 May 2009 Keywords: Mitochondrial DNA Penaeid shrimp Decapoda 16S rRNA Cytochrome c oxidase I (COI) Species identification Food authenticity PCR–RFLP Phylogenetic analysis

a b s t r a c t In this study, we infer the phylogenetic relationships within commercial shrimp using sequence data from a novel mitochondrial marker consisting of an approximately 530-bp region of the 16S ribosomal RNA (rRNA)/transfer RNA (tRNA)Val genes compared with two other mitochondrial genes: 16S rRNA and cytochrome c oxidase I (COI). All three mitochondrial markers were considerably AT rich, exhibiting values up to 78.2% for the species Penaeus monodon in the 16S rRNA/tRNAVal genes, notably higher than the average among other Malacostracan mitochondrial genomes. Unlike the 16S rRNA and COI genes, the 16S rRNA/tRNAVal marker evidenced that Parapenaeus is more closely related to Metapenaeus than to Solenocera, a result that seems to be more in agreement with the taxonomic status of these genera. To our knowledge, our study using the 16S rRNA/tRNAVal gene as a marker for phylogenetic analysis offers the first genetic evidence to confirm that Pleoticus muelleri and Solenocera agassizi constitute a separate group and that they are more related to each other than to genera belonging to the family Penaeidae. The 16S rRNA/tRNAVal region was also found to contain more variable sites (56%) than the other two regions studied (33.4% for the 16S rRNA region and 42.7% for the COI region). The presence of more variable sites in the 16S rRNA/tRNAVal marker allowed the interspecific differentiation of all 19 species examined. This is especially useful at the commercial level for the identification of a large number of shrimp species, particularly when the lack of morphological characteristics prevents their differentiation. Ó 2009 Elsevier Inc. All rights reserved.

Decapoda marine shrimp, especially those belonging to the superfamily Penaeoidea, represent the majority of the world’s commercially important shrimp species, accounting for more than 80% of the wild catch [1]. Among them, shrimp of the families Penaeidae and Solenoceridae are valuable resources for fisheries and aquaculture in both tropical and subtropical regions. The identification and characterization of shrimp traditionally relied on morphometric analysis; however, it is well known that such characteristics are environmentally influenced [2]. Accordingly, genetic studies of shrimp populations have been conducted using a variety of approaches such as isozyme protein electrophoresis [3], microsatellite analysis [4,5], and allozyme analysis [6,7]. Nuclear genes such as the 28S ribosomal RNA (rRNA),1 18S rRNA, 5.8S rRNA, and * Corresponding author. Fax: +34 982 252195. E-mail address: [email protected] (J. Barros-Velázquez). 1 Abbreviations used: rRNA, ribosomal RNA; ITS1, internal transcribed spacer 1; mtDNA, mitochondrial DNA; COI, cytochrome c oxidase subunit I; cytb, cytochrome b; tRNA, transfer RNA; PCR, polymerase chain reaction; SNP, single nucleotide polymorphism; RFLP, restriction fragment length polymorphism; SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; NJ, neighbor-joining. 0003-2697/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2009.05.020

internal transcribed spacer 1 (ITS1) have also been considered for the study of phylogenetic relationships among shrimp [8–10]. Nevertheless, the unique properties of mitochondrial DNA (mtDNA), especially the fact that its evolutionary rate is 10 times faster than that of nuclear DNA, make it a particularly useful tool in phylogenetic analyses. The most commonly used molecular markers include the 16S rRNA, 12S rRNA, and cytochrome c oxidase subunit I (COI) genes, the mtDNA control region, and (more recently) the cytochrome b gene (cytb) [1,6,11–15]. Such mtDNA markers have been used not only for the phylogenetic classification of natural shrimp populations [1,5,6,16] but also for the elucidation of genetic divergence between captive populations [17,18]. Among the mtDNA markers, the 16S rRNA and COI have been previously used for inferring phylogenetic relationships among shrimp species. On the one hand, studies indicate that the 16S rRNA region exhibits a low rate of evolution [19], meaning that it is especially useful for the interspecific differentiation of species rather than for the intraspecific differentiation of specimens. Accordingly, certain studies based on the 16S rRNA gene have established evolutionary relationships among shrimp at different

128

Identification and phylogenetic analysis of shrimp / P. Calo-Mata et al. / Anal. Biochem. 391 (2009) 127–134

taxonomic levels [20–23]. On the other hand, COI has proven to be a robust evolutionary marker for determining both intraspecific and interspecific relationships in a variety of marine mollusks [1,24]. In the same vein, a previous report showed high genetic variation in the mitochondrial COI gene within some shrimp species [24]. Thus, the genetic divergences observed in the COI gene among different shrimp species are consistently higher than those reported for the 16S rRNA gene, with the former exhibiting sufficient nucleotide diversity to allow the detection of different haplotypes [25]. In this study, we investigated a novel mitochondrial marker in the 16S rRNA/transfer RNA (tRNA)Val region and compared it with the COI and 16S rRNA genes in an effort to elucidate which of these markers constitutes a better molecular tool both for the basic investigation of the phylogenetic relationships among shrimp species and to assess the authenticity and traceability of most commercially relevant species. Materials and methods Shrimp species examined In this study, we compared a 530-bp fragment of the 16S rRNA/ tRNAVal region with nucleotide sequences of two mitochondrial genes, namely a 449-bp fragment of the COI and 476 bp of the 16S rRNA, both of which were reported previously by others and, therefore, were available for download from the GenBank database. The shrimp species that we evaluated account for the most marketed species. The 24 species of shrimp (superfamily Penaeoidea) included in this study belong to two families: Penaeidae and Solenoceridae. Their scientific names, commercial names, and accession numbers are compiled in Table 1. These included genera belonging to the family Penaeidae, such as Farfantepenaeus, Fenneropenaeus, Litopenaeus, Penaeus, Marsupenaeus, Melicertus, Metapenaeus, and

Parapenaeus, as well as the family Solenoceridae, represented by the genera Solenocera and Pleoticus. DNA extraction, amplification, sequencing, and restriction fragment length polymorphism analysis Extraction and amplification protocols were as described elsewhere [14]. Briefly, samples of 250 mg of skeletal muscle from each shrimp specimen were obtained. In all cases, three different specimens were considered for each species. DNA was extracted by means of a commercial kit (DNeasy Tissue Kit, Qiagen, Darmstadt, Germany), as described previously [14]. The primers used for the polymerase chain reaction (PCR) amplification and sequencing of the 16S rRNA/tRNAVal genes were 16ScruC4 (50 -AATATGGCTG TTTTTAAGCCTAATTCA-30 ) and 16ScruC3 (50 -CGTTGAGAAGTTCG TTGTGCA-30 ), which were constructed on two well-conserved regions of the 16S rRNA/tRNAVal genes. Such primers allowed the amplification of a 515- to 535-bp fragment of the 16S rRNA/ tRNAVal mtDNA genes in the penaeid shrimp species considered, as described elsewhere [14]. Sequencing was performed in both directions. Prior to sequencing, the PCR products were purified by means of the ExoSAP-IT kit (GE Healthcare, Uppsala, Sweden). Direct sequencing was performed with the BigDye Terminator Cycle Sequencing Kit (version 3.1, Applied Biosystems, Foster City, CA, USA). The same primers used for PCR were also employed for the sequencing of both strands of the PCR products. Sequencing reactions were analyzed in an automatic sequencing system (ABI 3730XL DNA Analyzer, Applied Biosystems) provided with the POP-7 system. Single nucleotide polymorphism (SNP) events in DNA sequences were carefully reviewed by eye using Chromas software (Griffith University, Queensland, Australia). Sequence alignment was accomplished using ClustalX 1.8 software [26]. Sequence homologies were searched using the BLAST tool (National Center for Biotechnology Information).

Table 1 Shrimp species considered in this study, common names, 3-alpha codes, 16S rRNA/tRNAVal, 16S rRNA, and COI sequence GenBank accession numbers. FAO species code

16S rRNA/ tRNAVal Accession Nos.

16S rRNA Accession Nos.

COI Accession Nos.

Family Penaeidae (penaeid shrimp) Farfantepenaeus aztecus Northern brown shrimp Farfantepenaeus brasiliensis Redspotted shrimp Farfantepenaeus brevirostris Crystal shrimp Farfantepenaeus californiensis Yellowleg shrimp Farfantepenaeus notialis Southern pink shrimp Fenneropenaeus chinensis Fleshy prawn Fenneropenaeus indicus Indian white prawn Fenneropenaeus merguiensis Banana prawn Litopenaeus setiferus Northern white shrimp Litopenaeus stylirostris Blue shrimp Litopenaeus vannamei Whiteleg shrimp Penaeus monodon Giant tiger prawn Penaeus semisulcatus Green tiger prawn Marsupenaeus japonicus Kuruma prawn Melicertus latisulcatus Western king prawn Metapenaeus ensis Greasyback shrimp Metapenaeus affinis Jinga shrimp Metapenaeus sp Metapenaeus shrimp nei Parapenaeus longirostris Deep-water rose shrimp Parapenaeus fissuroides –

ABS PNB CSP YPS SOP FLP PNI PBA PST PNS PNV GIT TIP KUP WKP MPE MTJ MET DPS PAF

EF589699a EF589701a EF589700a AY046912 EF589694a NC_009679 EF589688a EF589692a AJ297971 AY046913 EF589702a AF217843 EF589704a NC_007010 EF589708a NA NA EF589713a EF589715a NA

AF192051 AF192054 NA AY046912 NA AF245113 AF279815 AF279814 AJ297971 AF255054 AY344189 NC_002184 AY744268 AY742276 AF279821 AF279810 AY264904 NA NA AY264909

AF279834 AF029393 NA AY135197 X84350 DQ778650 AF279837 AY143990 NA AY135191 NC_009626 NC_002184 AF279831 NC_007010 AF279845 AF279830 AY264889 NA NA AY264894

Family Solenoceridae (solenocerid shrimp) Pleoticus muelleri Argentine red shrimp Solenocera agassizi Kolibri shrimp Solenocera crassicornis Coastal mud shrimp Solenocera koelbeli Chinese mud shrimp

LAA SOK SOJ SKO

EF589716a EF589719a NA NA

NA NA AY264915 AF105038

NA NA AY264902 AF105049

Family Palinuridae (spiny lobsters) Panulirus japonicus Japanese spiny lobster

NUJ

NC_004251

NC_004251

NC_004251

Taxonomic designation

Note: NA, not available. a This study.

Common name

129

Identification and phylogenetic analysis of shrimp / P. Calo-Mata et al. / Anal. Biochem. 391 (2009) 127–134 Table 2 Summary of sequence variation features of 16s rRNA/tRNAVal, 16s rRNA, and COI genes of the shrimp species analyzed.

16S rRNA/tRNA 16S rRNA COI

Val

Total sites

Variable sites

Percentage of variable sites

Informative sites

Percentage of informative sites

Average A+T content (%)

550 (554) 476 (482) 449 (449)

308/550 (356/554) 159/476 (199/482) 192/449 (179/449)

56.0 (64.2) 33.4 (41.2) 42.7 (39.9)

241/550 (260/554) 127/476 (127/482) 152/449 (151/449)

43.8 (46.9) 26.6 (25.7) 33.8 (33.6)

75.6 66.2 62.9

Note: The values within parentheses are obtained when the out-group Panulirus japonicus is included.

Predicted restriction fragment length polymorphism (RFLP) patterns were produced in silico using the web-based software In Silico [27]. PCR products and restriction fragments were processed in 2.5% horizontal agarose (MS-8, Pronadisa, Madrid, Spain) gel electrophoresis. PCR–RFLP analyses were carried out by agarose electrophoresis in 2.5% gels and by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) in 15% ExcelGel homogeneous gels (GE Healthcare) at 15 °C in a Multiphor II electrophoresis unit (Amersham Biosciences, Uppsala, Sweden). The latter gels were stained using a standard silver staining protocol (Amersham Biosciences). When required, PCR products were purified from the agarose gels by means of the MinElute Gel Extraction Kit (Qiagen). Image analysis of the electrophoretic gels was performed with 1-D Manager software (TDI, Madrid, Spain).

parsimony and distance analyses. The extent of sequence disparity between specimens was calculated by averaging pairwise comparisons of sequence differences across all specimens. Results Sequence data and phylogenetic analysis Alignment of all sequences was simple and unambiguous for all molecular markers and resulted in 550 sites for 16S rRNA/tRNAVal, 476 sites for the 16S rRNA gene, and 449 sites for the COI gene (Table 2). Aligned sequences were significantly AT rich, exhibiting values up to 78.2% for the species Penaeus monodon in the 16S rRNA/tRNAVal genes, up to 68.2% for P. monodon in the 16S rRNA gene and 66.1% for Farfantepenaeus notialis in the COI gene (Table 3).

Bioinformatic study of genetic distances and phylogenetic analysis Analysis of 16S rRNA/tRNAVal variation among shrimp species All nucleotide sequences were aligned using ClustalX 1.8 software [26] and then manually adjusted. Forward and reverse strands were combined, and sequences were visually checked for errors. For phylogenetic analysis, sequences were first aligned among themselves and then to the out-group. Phylogenetic analysis was conducted in MEGA 4.0 [28]. Phylogenetic relationships were estimated using neighbor-joining (NJ) analysis [29]. NJ analysis was performed using Jukes–Cantor and pairwise deletion options [30]. Bootstrapping (1000 replicates) was performed to assess the confidence level at each branch. The nucleotide distance matrices were created under a maximum likelihood correction [31]. Trees based on genetic distances were constructed by the NJ method [29]. Panulirus japonicus, a Decapoda species belonging to the family Palinuridae, was used as an outgroup due to the fact that it is closely related to members of the Penaeidae and Solenoceridae families yet still represents a separate taxon. Two different types of phylogenetic analysis were conducted using the transitions plus transversions data set: maximum

The 550 sites of the 16S rRNA/tRNAVal gene studied in 19 species contained 308 variable sites (56.0%) and 241 parsimony informative sites (43.8%) (Table 2). Remarkably, it contained 22 of 550 four-degenerate sites. The majority of variable sites occurred in the third codon position. The average distance in all Penaeoidea species was 0.230 and ranged from 0.057 within the species Farfantepenaeus californiensis and Farfantepenaeus aztecus to 0.377 between Litopenaeus vannamei and Metapenaeus sp. (Table 4). Interestingly, we found a high rate of guanine-to-adenine substitutions with a value of 23.43% and, similarly, a rate of 25.67% for cytosine-to-thymine substitutions (Table 5). Analysis of 16S rRNA variation among shrimp species The 476 sites of the 16S rRNA gene analyzed contained 159 variable sites (33.4%), and 127 parsimony informative sites (26.6%) (Table 2). This fragment of the 16S rRNA contained 51 of 476

Table 3 Nucleotide composition of the 16S rRNA/tRNAVal, 16S rRNA, and COI genes of the shrimp species analyzed by means of MEGA 4.0 software [28]. T(U)

C

A

G

Total

T-1

16S rRNA/tRNAVal gene [GIT] 37.8 6.8 [SOK] 34.4 5.7 [KUP] 31.3 8.3 Average 34.2 6.9

40.4 41.2 42.7 41.4

15.0 18.7 17.6 17.5

527 524 527 528.1

38.9 38.4 37.6 39.1

16S rRNA [GIT] [SKO] [PNV] Average

gene 34.0 33.9 32.0 33.3

12.6 11.9 13.9 12.7

34.2 32.4 33.0 32.9

19.2 21.8 21.0 21.1

468 463 466 464.9

COI gene [SOP] [KUP] [MPE] Average

37.6 35.6 31.4 36.0

16.3 17.4 21.8 19.0

28.5 27.6 28.1 26.9

17.6 19.4 18.7 18.1

449 449 449 435.0

C-1

A-1

G-1

Position #1

T-2

8.6 6.8 8.4 7.9

37.7 37.3 35.4 36.7

14.9 17.5 18.5 16.2

175 177 178 177.2

35.8 32.0 27.3 31.6

34.4 32.5 33.8 33.4

11.5 13.0 11.5 11.7

35.0 31.8 34.4 34.3

19.1 22.7 20.4 20.6

157 154 157 155.6

46.0 40.7 30.7 41.4

11.3 14.0 24.0 17.4

39.3 36.7 38.7 35.7

3.3 8.7 6.7 5.5

150 150 150 145.1

C-2

A-2

G-2

Position #2

T-3

6.3 4.7 8.1 6.1

43.2 44.2 46.5 42.7

14.8 19.2 18.0 19.5

176 172 172 174.3

38.6 32.6 28.8 31.7

33.1 34.2 30.1 32.5

14.6 13.5 16.0 15.1

32.5 30.3 32.1 30.9

19.7 21.9 21.8 21.5

157 155 156 156.3

25.3 24.7 22.0 24.2

14.0 14.7 18.0 15.9

28.0 28.0 27.3 27.6

32.7 32.7 32.7 32.3

150 150 150 145.4

C-3

A-3

G-3

Position #3

5.7 5.7 8.5 6.8

40.3 42.3 46.3 44.8

15.3 19.4 16.4 16.8

176 175 177 176.6

34.4 35.1 32.0 34.0

11.7 9.1 14.4 11.3

35.1 35.1 32.7 33.6

18.8 20.8 20.9 21.1

154 154 153 153

41.6 41.6 41.6 42.5

23.5 23.5 23.5 23.6

18.1 18.1 18.1 17.5

16.8 16.8 16.8 16.4

149 149 149 144.6

Note: All frequencies are given as percentages. The three species exhibiting the higher, moderate, and lower AT values, followed by the average for all species, are presented for each mitochondrial marker.

130

Identification and phylogenetic analysis of shrimp / P. Calo-Mata et al. / Anal. Biochem. 391 (2009) 127–134

Table 4 16S rRNA/tRNAVal gene distances among shrimp species analyzed by pairwise distance calculation using the model of nucleotide maximum composite likelihood including transitions and transversions. KUP

WKP

PNI

PBA

FLP

GIT

YPS

ABS

SOP

CSP

PNB

PNS

PNV

PST

TIP

MET

DPS

LAA

SOK

[KUP] [WKP] [PNI] [PBA] [FLP] [GIT] [YPS] [ABS] [SOP] [CSP] [PNB] [PNS] [PNV] [PST] [TIP] [MET] [DPS] [LAA] [SOK]

0,128 0,253 0,253 0,243 0,254 0,258 0,282 0,263 0,224 0,279 0,235 0,258 0,263 0,216 0,364 0,374 0,332 0,353

0,224 0,222 0,234 0,246 0,231 0,254 0,255 0,212 0,235 0,221 0,229 0,260 0,227 0,371 0,330 0,307 0,343

0,085 0,093 0,155 0,148 0,163 0,181 0,147 0,167 0,141 0,162 0,149 0,164 0,333 0,300 0,279 0,276

0,122 0,154 0,156 0,174 0,198 0,146 0,173 0,158 0,170 0,148 0,147 0,335 0,286 0,272 0,255

0,156 0,147 0,162 0,203 0,157 0,179 0,151 0,163 0,153 0,157 0,357 0,316 0,283 0,286

0,184 0,205 0,235 0,167 0,205 0,173 0,181 0,186 0,183 0,363 0,348 0,309 0,304

0,057 0,132 0,119 0,123 0,146 0,143 0,150 0,173 0,365 0,322 0,324 0,315

0,156 0,112 0,118 0,159 0,161 0,148 0,185 0,374 0,324 0,330 0,315

0,153 0,143 0,185 0,177 0,187 0,190 0,366 0,370 0,326 0,340

0,092 0,120 0,111 0,116 0,156 0,346 0,318 0,280 0,279

0,140 0,137 0,147 0,170 0,366 0,321 0,296 0,297

0,088 0,121 0,163 0,366 0,332 0,275 0,308

0,117 0,177 0,377 0,334 0,336 0,316

0,193 0,354 0,302 0,301 0,291

0,352 0,314 0,279 0,316

0,247 0,299 0,297

0,251 0,272

0,143

four-degenerate sites. We found a 19.26% rate of guanine-to-adenine substitutions and a comparable rate of cytosine-to-thymine substitutions of 19.06% (Table 5). Both of these values were lower than the corresponding rates detected in the 16S rRNA/tRNAVal region.

The 449 sites of the COI gene investigated in shrimp species showed no insertions or deletions across taxa and contained 92

variable sites (42.7%) and 152 parsimony informative sites (33.8%) (Table 2). A high rate of substitution of cytosine by thymine (31.27%) was determined (Table 5). The average distance in all Penaeoidea species was 0.196 and ranged from 0.000 within species of Solenocera koelbeli and Solenocera crassicornis to 0.254 between both F. aztecus and Farfantepenaeus brasiliensis and Metapenaeus affinis (see Supplementary Table 5 in supplementary material). The complete nucleotide compositions for all shrimp species and mitochondrial markers are presented as Supplementary material.

Table 5 Pattern of nucleotide substitution detected in the 16S rRNA/tRNAVal, 16S rRNA, and COI genes.

Comparative phylogenetic analysis of shrimp species as determined by the three mitochondrial markers

Maximum composite likelihood estimate of the pattern of nucleotide substitution

Evidence for three major clades within the shrimp species analyzed was obtained by nucleotide sequence data analysis for all three mitochondrial markers. Thus, the topologies resulting from NJ analysis (Figs. 1–3) show that the members of the ‘‘old genus” Penaeus studied in this work, including the genera Farfantepenaeus, Litopenaeus, Penaeus, and Fenneropenaeus, constitute a major clade regardless of the mitochondrial marker considered. Likewise, the 16S rRNA/tRNAVal mitochondrial region analyzed in this work allowed the designation of a second clade that encompassed the genera Marsupenaeus and Melicertus (Fig. 1). These results were in agreement with the work previously reported by other authors in which the 16S rRNA (Fig. 2) and COI (Fig. 3) genes were evaluated [21,22]. Remarkably, we found that the 16S rRNA/tRNAVal marker examined in this work provided new information that may help to clarify some weak points detected when using the 16S rRNA and COI genes. Thus, the 16S rRNA/tRNAVal marker allows the definition of a third clade that is branched into two smaller groups: one that includes the genera Metapenaeus and Parapenaeus and another that includes genera belonging to the family Solenoceridae: Solenocera and Pleoticus (Fig. 1). On the one hand, both 16S rRNA (Fig. 2) and COI (Fig. 3) markers grouped the genera Metapenaeus and Parapenaeus in two different subclades, whereas the 16S rRNA/ tRNAVal marker groups both of them in the same subclade. Hence, according to the 16S rRNA and COI markers, the genus Parapenaeus would be less related to the genus Metapenaeus, with both of them belonging to the family Penaeidae, than to the genus Solenocera, which belongs to the family Solenoceridae (Figs. 2 and 3) [22]. In contrast, the 16S rRNA/tRNAVal marker provided a closer relationship between Metapenaeus and Parapenaeus than between the

Analysis of COI variation among shrimp species

16S rRNA/tRNAVal

A

T

A T C G

— 7.03 7.03 23.43

6.11 — 25.67 6.11

1.29 5.42 — 1.29

10.38 3.12 3.12 —

16S rRNA A T C G

— 7.08 7.08 19.26

6.97 — 19.06 6.97

2.65 7.25 — 2.65

12.12 4.46 4.46 —

— 4.94 4.94 8.31

6.98 — 31.27 6.98

3.69 16.51 — 3.69

5.79 3.44 3.44 —

COI A T C G

C

G

Note: Each entry shows the probability of instantaneous substitution from one base (row) to another base (column). Only entries within a row should be compared. Rates of different transitional substitutions are shown in bold font, and those of transversional substitutions are shown in italic font. The nucleotide frequencies for 16S rRNA/tRNAVal are 0.401 (A), 0.348 (T/U), 0.074 (C), and 0.178 (G). The nucleotide frequencies for 16S rRNA are 0.335 (A), 0.329 (T/U), 0.125 (C), and 0.211 (G). The nucleotide frequencies for COI gene are 0.259 (A), 0.366 (T/U), 0.193 (C), and 0.181 (G). The transition/transversion rate ratios for 16S rRNA/tRNAVal are k1 = 3.332 (purines) and k2 = 4.199 (pyrimidines), and the overall transition/transversion bias is R = 0.856. The transition/transversion rate ratios for 16S rRNA are k1 = 2.719 (purines) and k2 = 2.736 (pyrimidines), and the overall transition/transversion bias is R = 0.91. The transition/transversion rate ratios for COI are k1 = 1.682 (purines) and k2 = 4.478 (pyrimidines), and the overall transition/transversion bias is R = 1.311, where R = [AGk1 + TCk2]/[(A+G)(T+C)]. There were totals of 485, 449, and 408 positions for the 16S rRNA/tRNAVal, 16S rRNA, and COI genes, respectively, in the final dataset. Codon positions included were 1st, 2nd, 3rd, and noncoding. All positions containing gaps and missing data were eliminated from the dataset (complete deletion option). All calculations were conducted in MEGA 4.0 [28].

Identification and phylogenetic analysis of shrimp / P. Calo-Mata et al. / Anal. Biochem. 391 (2009) 127–134

131

YPS F. californiensis

99 60

ABS F. aztecus

55

SOP F. notialis

71

PNB F. brasiliensis CSP F. brevirostris

82

PST L. setiferus PNS L. stylirostris

75

94

PNV L. vannamei

85

77

TIP P. semisulcatus GIT P. monodon

45

FLP F. chinensis

60

PNI F. indicus

82

PBA F. merguiensis

56

KUP M. japonicus WKP M. latisulcatus MET Metapenaeus sp.

99 90

DPS P. longirostris LAA P. muelleri

75

SOK S. agasizzi

99

NUJ P. japonicus. 0.05

Fig. 1. Topologies resulting from the phylogenetic analysis of the nucleotide sequences of the 515- to 535-bp 16S rRNA/tRNAVal mitochondrial genes in all shrimp species considered and the out-group by means of the NJ method. Numbers above and below branches indicate bootstrap values from NJ analysis.

92 99

PNB F. brasiliensis YPS F. californiensis ABS F. aztecus

97

PST L. setiferus PNS L. stylirostris

99

28

PNV L. vannamei

80

FLP F. chinensis

30

PBA F. merguiensis

99 93

41 100

PNI F. indicus GIT P. monodon TIP P. semisulcatus KUP M. japonicus

51 99

56

WKP M. latisulcatus

PAF P. fissuroides SOJ S. crassicornis 100

SKO S. koelbeli MPE M. ensis

100

MTJ M. affinis NUJ P. japonicus.

0.05

Fig. 2. Topologies resulting from the phylogenetic analysis of the nucleotide sequences of the 476-bp 16S rRNA mitochondrial gene in all shrimp species considered and the out-group by means of the NJ method. Numbers above and below branches indicate bootstrap values from NJ analysis.

latter and Solenocera (Fig. 1), a result that seems to be more in agreement with the taxonomic status of these genera. On the other hand, the 16S rRNA/tRNAVal marker categorizes the Pleoticus and Solenocera species, belonging to the family Solenoceridae, in the same subclade (Fig. 1). Remarkably, species belonging to the genus Pleoticus had not been considered in the most recent and complete phylogenetic studies on shrimp species based on 16S rRNA and COI genes [21,22].

A tree considering the combined data from all three regions was also constructed for the 14 shrimp species sequenced in these three genes (Fig. 4). In general terms, the conclusions extracted from this tree are quite in agreement with those provided by each mitochondrial marker individually. However, the phylogenetic status of Penaeus semisulcatus in this tree (Fig. 4) was more in agreement with the results obtained with either the 16S rRNA/tRNAVal (Fig. 1) or COI (Fig. 3), whereas in the case of Fenneropenaeus chinensis the rRNA/

132

Identification and phylogenetic analysis of shrimp / P. Calo-Mata et al. / Anal. Biochem. 391 (2009) 127–134

PNB F. brasiliensis

97 53

YPS F. californiensis ABS F. aztecus

47

SOP F. notialis PNV L. vannamei

93

25

PNS L. stylirostris GIT P. monodon

67

TIP P. semisulcatus PBA F. merguiensis PNI F. indicus

77 17 39 100

50

FLP F. chinensis WKP M. latisulcatus

81

KUP M. japonicus PAF P. fissuroides

87

28 100

SKO S. koelbeli SOJ S. crassicornis

MPE M. ensis MTJ M. affinis

100

NUJ P. japonicus. 0.02

Fig. 3. Topologies resulting from the phylogenetic analysis of the nucleotide sequences of the 449-bp COI mitochondrial gene in all shrimp species considered and the outgroup by means of the NJ method. Numbers above and below branches indicate bootstrap values from NJ analysis.

YPS F. californiensis

50 100

ABS F. aztecus PNB F. brasiliensis

100

66

PNS L. stylirostris PNV L. vannamei

100

GIT P. monodon FLP F. chinensis

99 80

PNI F. indicus

99

PBA F. merguiensis

100

100

TIP P. semisulcatus KUP M. japonicus WKP M. latisulcatus

100

MPE M. ensis 100

MTJ M. affinis NUJ P. japonicus.

0.05

Fig. 4. Topologies resulting from the phylogenetic analysis of the combined nucleotide sequence data of the 16S rRNA/tRNAVal, 16S rRNA, and COI mitochondrial genes in the common shrimp species and the out-group by means of the NJ method. Numbers above and below branches indicate bootstrap values from NJ analysis.

tRNAVal (Fig. 1) and 16S rRNA (Fig. 2) genes gave more reliable results than the COI gene (Fig. 3). Interspecific differentiation of shrimp species by PCR–RFLP analysis All sequences belonging to each of the three mitochondrial markers considered in this study were analyzed in silico to select restriction enzymes that might be useful for the interspecific differentiation by PCR–RFLP analysis. Enzymes that produced the greatest pattern variability among different species were selected for further analysis. All species assessed in this study for each mitochondrial marker were aligned and analyzed to confirm the existence of sufficient sequence variation among species to develop a PCR–RFLP method for their interspecific differentiation.

The 16S rRNA/tRNAVal region was found to contain more informative restriction sites, allowing the clear differentiation of all species studied and even allowing intraspecific differentiation [14] when sequences were analyzed by the In Silico bioinformatic tool [27]. Thus, from all mtDNA markers analyzed in this study, the approximately 530-bp fragment of the 16S rRNA/tRNAVal genes provided the best results, allowing the identification of all 19 shrimp species considered (Table 1). This fragment, when subjected to cleavage with AluI, TaqI, and HinfI, provided species-specific restriction patterns that had been reported to allow the identification of 18 shrimp species [14]. In this study, F. chinensis represents a new species that may also be differentiated from all other shrimp species previously analyzed by PCR–RFLP. Thus, F. chinensis may be detected by generic primers 16ScruC4/16ScruC3,

Identification and phylogenetic analysis of shrimp / P. Calo-Mata et al. / Anal. Biochem. 391 (2009) 127–134

giving a 522-bp amplicon that is cleaved in three fragments of 285, 160, and 77 bp with AluI, in two fragments of 300 and 123 bp with TaqI, and in two fragments of 402 and 120 bp with HinfI, thereby constituting a species-specific restriction pattern. Discussion In an effort to improve the current standard of the phylogenetic classification of shrimp, we amplified and sequenced a novel mitochondrial marker corresponding to an approximately 530-bp region of the 16S rRNA/tRNAVal gene in shrimp. In addition, other sequences within this region and two other mitochondrial markers corresponding to the 16S rRNA gene (476 bp) and COI gene (449 bp) of shrimp were downloaded from the GenBank database, and each region was aligned independently. The similarity of nucleotide sequences among species varied from 91% to 98% (interspecific variability), whereas within species (intraspecific variability) the similarity was approximately 99% (data not shown). These changes were mostly transitional substitutions in the third position of the codon. Each genus revealed little internal variation, suggesting that the species within each genus are closely related. However, we found consistent sequence differences of one to five nucleotides among even the most closely related species, indicating the usefulness of the 16S rRNA/tRNAVal, 16S rRNA, and COI markers for species identification. Of the three regions, the 16S rRNA/tRNAVal region appeared to provide the greatest phylogenetic resolving power, probably due to the higher interspecific variability (Table 2). The overall AT content in all of the genes analyzed in our study proved to be higher than the average among other Malacostracan mitochondrial genomes. Other authors found that protein genes from F. notialis and Artemia franciscana shared nearly identical AT content (64.0%), which was much lower than in insects (73.5%) and much higher than in Panulirus argus genes (58.1%) [32]. Other studies reported GC contents near 28.52% (71.48% AT) for F. californiensis, 28.79% (71.21% AT) for L. vannamei, 28.7% (71.3% AT) for Litopenaeus stylirostris, and 23.5% (76.5% AT) for P. monodon [33]. The high AT content (63%) of the COI fragment (Table 3) was consistent with descriptions by other authors [32] and with arthropod mtDNA sequences [34]. To our knowledge, the 16S rRNA/tRNAVal marker used in this work provides the first genetic evidence confirming that Pleoticus muelleri and Solenocera agassizi constitute a separate group and that they are more closely related to each other than to genera belonging to the family Penaeidae. However, it would be desirable to sequence P. muelleri, S. agassizi, and other Solenocera spp. with a view to strengthen the placement of the genus Pleoticus, for which our work constitutes the first phylogenetic study to date. With respect to the usefulness of the mitochondrial markers for the interspecific differentiation of shrimp species, a previous report described the usefulness of a 1.38-kb mitochondrial region that comprised fragments of the 16S rRNA and 12S rRNA genes and the entire tRNAVal region for phylogenetic analysis of shrimp [33]. However, this study considered only three Eastern Pacific species: F. californiensis, L. vannamei, and L. stylirostris. Furthermore, the study was focused only on phylogenetic aspects, not on their species identification [33]. To our knowledge, the only previous molecular method for the PCR–RFLP identification of five shrimp species—P. monodon, P. semisulcatus, L. vannamei, Fenneropenaeus merguiensis, and Marsupenaeus japonicus—is based on COI, cytochrome oxidase II (COII), and 16S rRNA mitochondrial genes [35]. The authors of this study based this identification method on the restriction analysis of (i) a 1550-bp mitochondrial fragment with endonucleases DdaI, SspI, and VspI for the species P. monodon and P. semisulcatus and (ii) 560 bp of the 16S rRNA mitochondrial gene

133

with endonucleases AluI, MboI, SspI, and VspI for the species P. monodon, P. semisulcatus, L. vannamei, F. merguiensis, and M. japonicus. In addition, a method for the generic detection of crustacean DNA based on a PCR–RFLP approach has also been proposed [36]. In this method, the authors amplified a 205-bp region of the 16S rRNA mitochondrial gene and described a PCR–RFLP method for the detection of potentially allergenic crustacean proteins found in food products. The species studied comprised only 4 species of shrimp, 3 species of crab, and 2 species of lobster and crawfish. In our study, the 16S rRNA/tRNAVal marker evaluated allowed the differentiation of up to 19 commercially relevant species, constituting a valuable tool for both traceability and authenticity purposes. Conclusions Among the three molecular mitochondrial markers compared in this study for the interspecific differentiation of shrimp species, the 16S rRNA gene is relatively more conserved than the COI gene, whereas both of them are more conserved than the novel 16S rRNA/tRNAVal region considered. Remarkably, the 16S rRNA/tRNAVal region analyzed in this study provided new evidence of the phylogenetic status of the genera Metapenaeus, Parapenaeus, Solenocera, and (especially) Pleoticus that both complements and differs from previous reports based on the 16S rRNA and COI mitochondrial genes. A PCR–RFLP method based on the 16S rRNA/tRNAVal region of penaeid shrimp is best suited to study the prevalence of known species in commercial shrimp or prawn products in a non-expertise-based work environment. In conclusion, the 530-bp 16S rRNA/tRNAVal mitochondrial region investigated in this study enriches current knowledge of the phylogenetic relationships among this group of species and also provides a robust tool for the identification of the most relevant shrimp for food authentication purposes. Acknowledgments The authors thank Julio Maroto (CETMAR, Vigo, Spain) for management of the collection of specimens for this study and the staff of the Museum of Natural Sciences (Berlin, Germany) for kindly providing Fenneropenaeus indicus specimens. Thanks are also due to Francisco Barros (Unidad e Medicina Molecular, Fundación Pública Galega de Medicina Xenómica, Santiago de Compostela, Spain) for his excellent technical assistance with mtDNA sequencing and to Carmen Pineiro (IIM–CSIC, Vigo, Spain) for her help in penaeid shrimp classification. The authors acknowledge the financial support from the National Food Program of the INIA (Spanish Ministry for Education, project CAL-03-030-C2-1) and from the PGIDIT Research Program in Marine Resources (project PGIDIT04RMA261004PR) of the Xunta de Galicia (Galician Council for Industry Commerce and Innovation). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ab.2009.05.020. References [1] J.D. Baldwin, A.L. Bass, B.W. Bowen, W.H. Clark, Molecular phylogeny and biogeography of the marine shrimp Penaeus, Mol. Phylogenet. Evol. 10 (1998) 399–407. [2] T.E. Bowman, L.G. Abele, Classification of recent crustacea, in: L.G. Abele (Ed.), The Biology of the Crustacea, vol. 1: Systematics, the Fossil Record and Biogeography, Academic Press, New York, 1982, pp. 1–27. [3] S. Lavery, D. Staples, Use of allozyme electrophoresis for identifying two species of penaeid prawn postlarvae, Aus. J. Mar. Freshwater Res. 41 (1990) 259–266.

134

Identification and phylogenetic analysis of shrimp / P. Calo-Mata et al. / Anal. Biochem. 391 (2009) 127–134

[4] A.O. Ball, S. Leonard, R.W. Chapman, Characterization of (GT)n microsatellites from native white shrimp (Penaeus setiferus), Mol. Ecol. 7 (1998) 1251–1253. [5] R. Maggioni, A.D. Rogers, N. Maclean, Population structure of Litopenaeus schmitti (Decapoda, Penaeidae) from the Brazilian coast identified using six polymorphic microsatellite loci, Mol. Ecol. 12 (2003) 3213–3217. [6] J. Gusmao, C. Lazoski, A.M. Sole-Cava, A new species of Penaeus (Crustacea: Penaeidae) revealed by allozyme and cytochrome oxidase I analyses, Mar. Biol. 137 (2000) 435–446. [7] C.N. Zitari-Chatti, A. Elouaer, K. Said, Genetic variation and population structure of the caramote prawn Penaeus kerathurus (Forskäl) from the eastern and western Mediterranean coasts in Tunisia, Aquacult. Res. 39 (2008) 70–76. [8] M.L. Porter, M. Perez-Losada, K.A. Crandall, Model-based multi-locus estimation of decapod phylogeny and divergence times, Mol. Phylogenet. Evol. 37 (2005) 355–369. [9] W. Kim, L.G. Abele, Molecular phylogeny of selected decapod crustaceans based on 18S rRNA nucleotide sequences, J. Crust. Biol. 10 (1990) 1–13. [10] W. Wanna, W. Chotigeat, A. Phongdara, Sequence variations of the first ribosomal internal transcribed spacer of Penaeus species in Thailand, J. Exp. Mar. Biol. Ecol. 331 (2006) 64–73. [11] T.D. Tzeng, S.Y. Yeh, C.F. Hui, Population genetic structure of the kuruma prawn (Penaeus japonicus) in East Asia inferred from mitochondrial DNA sequences, ICES J. Mar. Sci. 61 (2004) 913–920. [12] S. Klinbunga, D.J. Penman, B.J. McAndrew, A. Tassanakajon, Mitochondrial DNA diversity in three populations of the giant tiger shrimp Penaeus monodon, Mar. Biotechnol. 1 (1999) 113–121. [13] J. Quan, X.M. Liu, Z. Zhuang, J. Deng, J. Dai, Y.P. Zhang, Phylogenetic relationships of 12 Penaeoidea shrimp species deduced from mitochondrial DNA sequences, Biochem. Genet. 42 (2004) 331–345. [14] A. Pascoal, J. Barros-Velázquez, A. Cepeda, J.M. Gallardo, P. Calo-Mata, A polymerase chain reaction–restriction fragment length polymorphism method based on the analysis of a 16S rRNA/tRNAVal mitochondrial region for species identification of commercial penaeid shrimps (Crustacea: Decapoda: Penaeoidea) of food interest, Electrophoresis 29 (2008) 499–509. [15] A. Pascoal, J. Barros-Velázquez, A. Cepeda, J.M. Gallardo, P. Calo-Mata, Identification of shrimp species in raw and processed food products by means of a PCR–RFLP method targeted to cytochrome b mitochondrial sequences, Electrophoresis 29 (2008) 3220–3228. [16] D. Bouchon, C. Souty-Grosset, R. Raimond, Mitochondrial DNA variation and markers of species identity in two penaeid shrimp species: Penaeus monodon Fabricius and Penaeus japonicus Bate, Aquaculture 127 (1994) 131–144. [17] K. Iguchi, K. Watanabe, M. Nishida, Reduced mitochondrial DNA variation in hatchery populations of ayu (Plecoglossus altivelis) cultured for multiple generations, Aquaculture 178 (1999) 235–243. [18] M. Sekino, M. Hara, N. Taniguchi, Loss of microsatellite and mitochondrial DNA variation in hatchery strains of Japanese flounder Paralichthys olivaceus, Aquaculture 213 (2002) 101–122. [19] A. Meyer, DNA technology and phylogeny of fish, in: A.R. Beaumont (Ed.), Genetics and Evolution of Aquatic Organisms, Chapman & Hall, London, 1994, pp. 220–290. [20] E. Garcia-Machado, N. Dennebouy, M.O. Suarez, J.C. Mounolou, M. Monnerot, Mitochondrial 16S rRNA gene of two species of shrimps: sequence variability and secondary structure, Crustaceana 65 (1993) 279–286.

[21] S. Lavery, T.Y. Chan, Y.K. Tam, K.H. Chu, Phylogenetic relationships and evolutionary history of the shrimp genus Penaeus s.l. derived from mitochondrial DNA, Mol. Phylogenet. Evol. 31 (2004) 39–49. [22] C.M. Voloch, P.R. Freire, C.A.M. Russo, Molecular phylogeny of penaeid shrimps inferred from two mitochondrial markers, Genet. Mol. Res. 4 (2005) 668–674. [23] S.R. Palumbi, J. Benzie, Large mitochondrial differences between morphologically similar penaeid shrimp, Mol. Mar. Biol. Biotechnol. 1 (1991) 27–34. [24] B.S. Baldwin, M. Black, O. Sanjur, R. Gustafson, R.A. Lutz, R.C. Vrijenhoek, A diagnostic molecular marker for zebra mussels (Dreissena polymorpha) and potentially co-occurring bivalves: mitochondrial COI, Mol. Mar. Biol. Biotechnol. 5 (1996) 9–14. [25] A.K. de Francisco, P.M. Galetti, Genetic distance between broodstocks of the marine shrimp Litopenaeus vannamei (Decapoda, Penaeidae) by mtDNA analysis, Genet. Mol. Biol. 28 (2005) 258–261. [26] J.D. Thompson, D.G. Higgins, T.J. Gibson, CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties, and weight matrix choice, Nucleic Acids Res. 22 (1994) 4673–4680. [27] J. Bikandi, R. San Millan, A. Rementeria, J. Garaizar, In silico analysis of complete bacterial genomes: PCR, AFLP–PCR, and endonuclease restriction, Bioinformatics 20 (2004) 798–799. [28] K. Tamura, J. Dudley, M. Nei, S. Kumar, MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0, Mol. Biol. Evol. 24 (2007) 1596–1599. [29] N. Saitou, M. Nei, The neighbor-joining method: a new method for reconstructing phylogenetic trees, Mol. Biol. Evol. 4 (1987) 406–425. [30] S. Kumar, K. Tamura, I.B. Jakobsen, N. Masatoshi, MEGA2: Molecular Evolutionary Genetics Analysis Software Version 2.1, Arizona State University, Tempe, 2001. [31] J. Felsenstein, Evolutionary trees from DNA sequences: a maximum likelihood approach, J. Mol. Evol. 17 (1981) 368–376. [32] E. Garcia-Machado, N. Dennebouy, M.O. Suarez, J.C. Mounolou, M. Monnerot, Partial sequence of the shrimp Penaeus notialis mitochondrial genome, C. R. Acad. Sci. III 319 (1996) 473–486. [33] L.E. Gutiérrez-Millan, A.B. Peregrino-Uriarte, R. Sotelo-Mundo, F. VargasAlbores, G. Yépiz-Plasencia, Sequence and conservation of a rRNA and tRNAVal mitochondrial gene fragment from Penaeus californiensis and comparison with Penaeus vannamei and Penaeus stylirostris, Mar. Biotechnol. 4 (2002) 392–398. [34] G.S. Spicer, Phylogenetic utility of the cytochrome oxidase gene: Molecular evolution of the Drosophila buzzatii species complex, J. Mol. Evol. 41 (1995) 749–759. [35] B. Khamnamtong, S. Klinbunga, P. Menasveta, Species identification of five penaeid shrimps using PCR–RFLP and SSCP analyses of 16S ribosomal DNA, J. Biochem. Mol. Biol. 38 (2005) 491–499. [36] J.L. Brzezinski, Detection of crustacean DNA and species identification using a PCR-restriction fragment length polymorphism method, J. Food Prot. 68 (2005) 1866–1873.