Molecular Phylogeny of Philippine Freshwater
Sardines Based on Mitochondrial DNA
I. E. Samonte, R. C. Pagulayan, and W. E. Mayer
The commercially important Sardinella species (family Clupeidae or herrings) usu-
ally thrive in marine environments. An exception is Sardinella tawilis of Taal Lake,
Batangas, Philippines, the only known freshwater sardine. This species is believed
to have immigrated from Balayan Bay to the lake when it was formed in the course
of volcanic eruptions some 240 years ago. To determine the relationship of S. taw-
ilis to the marine species S. albella, S. ﬁmbriata, and S. longiceps from the Balayan
Bay we sequenced 358 bp of the cytochrome b gene and the mitochondrial control
region. The cytochrome b gene was highly conserved and contained little phylo-
genetic information. The control region sequences, however, demonstrated two
highly diversiﬁed main haplotypes grouping S. tawilis with S. albella, as shown by
maximum parsimony and neighbor-joining analysis. The haplotypes are character-
ized by the presence of an 81 bp indel and up to eight 35 bp tandem repeat ele-
ments. The repeat copy number varied within individuals of S. tawilis and S. albella,
thus showing heteroplasmy in these two species only. The analysis of two sub-
populations of S. tawilis revealed restricted substitutions that may indicate the be-
ginning of genetic differentiation of the two subpopulations.
Members of the family Clupeidae, or her- by the collapse of a large volcanic crater
rings (order Clupeiformes, subclass Actin- and by subsidence, and whose morphom-
opterygii), are ﬁsh with a compressed etry has been modiﬁed by subsequent vol-
streamlined body, a single soft-rayed dor- canic activities, especially by a powerful
sal ﬁn, and protruding scales. They are eruption in 1754 ( Hargrove 1991). The 10th
found worldwide in temperate and tropi- century lake, connected to Balayan Bay
cal waters where they mainly move in through a wide channel, has now rear-
large schools feeding on plankton. Most of ranged its shape and narrowed its outlet to
the species live in the marine environ- form the Pansipit River, the sole outlet from
ment, but some, like the American gizzard its southwest corner ( Kira 1995). Like oth-
shad (Dorosama cepedianum), occur only er members of the genus, tawilis has an ob-
in freshwater systems. The family Clupei- long to nearly elongated body, equal jaws,
dae includes the subfamilies Alosinae or large mouth, lacks lateral lines, and has
shads, the Pellonulinae or freshwater her- thin deciduous scales (Munro 1955). Be-
rings, and the Clupeinae with the genera cause of its small size (10–20 cm) it does
Clupea, Sardina, and Sardinops. not command much interest economically,
Little is known about the phylogenetic but its year-round presence makes it im-
relationships among the commercially im- portant to ﬁshers who depend solely on
portant Sardinella species ( Family Clupei- the lake for their daily subsistence. Interest
dei, Suborder Clupeoidei), especially at the in this species has arisen mainly because
From the Max-Planck-Institut fur Biologie, Abteilung Im- molecular level. In the Philippines, eight of a recent report of human intervention in
mungenetik, Corrensstrasse 42, D-72076 Tubingen, Ger- members of this genus were identiﬁed and the lake which threatens its ﬁshery re-
many (Samonte and Mayer), and the Institute of Biol-
ogy, University of the Philippines–Diliman, Diliman,
collectively called sardines (Conlu 1986). sources ( Villanueva et al. 1996). Also the
Quezon City, Philippines (Pagulayan). We thank Dr. Jan All of these species are thriving in the ma- recent reclassiﬁcation of the species from
Klein for his support and stimulating discussions. I.E.S. rine environment except Sardinella tawilis Harengula to Sardinella on the basis of mor-
was supported by a short-term scholarship from the
Deutscher Akademischer Austauschdienst ( DAAD) and (Whitehead 1985), the only known fresh- phoanatomic features makes it an interest-
the Engineering and Science Education Program ( ESEP) water sardine, endemic to Taal Lake, Batan- ing subject of a molecular study ( FishBase
of the Department of Science and Technology ( DOST ), gas, Philippines. Known locally as tawilis, 1995).
Philippines. Address correspondence to Werner E.
Mayer at the address above or e-mail: werner.mayer@ this species is believed to have immigrated Analyses of the mitochondrial DNA
tuebingen.mpg.de. from Balayan Bay to Taal Lake. The said (mtDNA) sequences have proved to be a
2000 The American Genetic Association 91:247–253 lake was a caldera lake that formed partly valuable tool in addressing the problems
Table 1. Primers used for the ampliﬁcation of the cytochrome b gene and the control region of the
Primer designation Sequence
cytbLa 5 -CCCCTCAGAATGATATTTGTCCTCA-3
cytbHa 5 -CCATCCAACATCTCAGCATGATGAAA-3
L15926a 5 -TTACACCAGTCTTGTAAACC-3
H16498b 5 -CCTGAAGTAGGAACCAGATG-3
Q 5 -GGGCGGATCCCACCACTAGCTCCCAAA-3
B 5 -ACGCTGGAAAGAACGCCCGGCATGG-3
J 5 -TTTGGTTCCTATTTCAGGGCCA-3
N 5 -GGGGCGCGGATCCCATCCTAATATCTTCAG-3
A88c 5 -GCAAAACTCTCACCAACTCATC-3
A87c 5 -GGGGTTTCATGGTGAACT-3
Primer designed by Kocher et al. (1989) except that the restriction site was deleted.
Primer designed by Meyer and Wilson (1990).
Internal primers designed from the generated sequences. The remaining primers are those used by Lee at al.
Figure 1. Geography of Taal Lake showing the Pan-
sipit River and portions of Balayan Bay and Batangas
Bay. The location of the Talisay and Agoncillo parts is
that S. tawilis faces. Features of this mol- 1995). The variable portions of the control indicated.
ecule, such as its compact organization region contain varying numbers of tandem
(Russell 1994), its elevated rate of muta- repeat sequences ( Broughton and Dowl-
tions of S. tawilis were obtained from the
tion ( Brown 1983), its primarily maternal ing 1994; Lee et al. 1995) which are the
Talisay and Agoncillo portions of Taal
inheritance, and the absence of recombi- targets of frequent mutations, particularly
Lake ( Figure 1). S. albella and S. ﬁmbriata
nation (Moritz et al. 1987), make it a valu- deletions, insertions, and duplications
from Balayan Bay were procured from ﬁsh
able tool in evolutionary and population ( Brown 1983; Harrison 1989). Its fast evo-
vendors in Lemery, Batangas. S. longiceps,
studies. lutionary rate makes the control region a
caught in Cebu Bay, was obtained from the
Fish mtDNA ranges in length from 15.2 suitable marker for resolving relatively re-
Malabon Fish Port in Malabon, Metro Ma-
to 19.8 kb ( Billington and Hebert 1991) cent divergences of closely related species
nila. Muscle and gonads were dissected
and contains information specifying the and populations ( Zischler et al. 1995).
out and preserved in 85% ethanol. These
two rRNAs (16s and 12S), 22 tRNAs, 13 Analyses of mitochondrial genes have
were then brought to the Max-Planck In-
polypeptides, as well as the dissociation been facilitated by the discovery of the
stitute for Biology for analysis.
( D) loop, also referred to as the control polymerase chain reaction (PCR), which
region. The proteins it encodes include enables speciﬁc sections of the genes to
NADH dehydrogenase, cytochrome oxi- be routinely investigated (Carr and Mar-
Total DNA of the ethanol-preserved sam-
dase, cytochrome b (cytb), and ATPase 6 shall 1991; Kocher et al. 1989) from minute
ples was extracted according to the
and 8 (Gyllensten and Wilson 1987). Of quantities of tissue samples (McVeigh et
NucleoSpin C T protocol (Macherey-Na-
these, cytochrome b ( Barlett and David- al. 1991) to yield high resolution informa-
gel, Duren, Germany). MtDNA was extract-
son 1991) together with the control region tion about any variation that may exist in
ed using the modiﬁed phenol extraction
( Zhu et al. 1994) have become widely the DNA ( Hartley et al. 1992).
procedures of Gonzales Villasenor et al.
used molecular markers. In vertebrates, The aim of this study was to determine
(1990), Chapman and Powers (1984), and
the rate of the cytb gene evolution is esti- the phylogenetic relationship of S. tawilis
Bernatchez et al. (1988). The extracted
mated to be between 1% and 2.5% substi- to the marine sardines S. albella (White-
mtDNA was precipitated with ethanol and
tutions per one million year ( Irwin et al. head and Wongratana 1986), S. ﬁmbriata
resuspended in 50 l 10 mM TE buffer (pH
1991; Martin et al. 1992) which means that (Whitehead 1985), and S. longiceps (White-
8.5). Yields were typically 800 g DNA/ml
approximately one hundred thousand head 1985), and the genetic variability be-
years are needed before one might conﬁ- tween populations of S. tawilis from the
dently expect a substitution to differenti- mtDNA sequences. The latter information
ate closely related mtDNA types (Palumbi was analyzed for ﬁsh management purpos-
The primer sets used to PCR amplify (Saiki
and Kessing 1991). This relatively slow es.
et al. 1988) a segment of the mitochondrial
evolution of the cytb gene makes it a suit-
cytochrome b gene (cytb) and the mito-
able marker for resolving deeper phylo-
Materials and Methods
genetic relationships among different taxa
(Meyer and Wilson 1990). Unlike cytb, the Collection of Fish Samples
mitochondrial control region does not Four Philippine sardine species, S. tawilis,
code for proteins (Avise et al. 1987) and S. albella, S. ﬁmbriata, and S. longiceps,
has only short sequence blocks conserved were collected and transported to the Nat-
among distant taxa ( Brown et al. 1993; ural Sciences Research Institute’s labora-
Clayton 1982; Lee et al. 1995; Saccone et tory for identiﬁcation and dissection. Rau
Figure 2. Schematic representation of the mitochon-
al. 1987), presumably because of relaxed and Rau (1980) and Conlu (1986) served drial control region. Primers used for PCR and se-
functional constraints (Alves-Gomes et al. as bases for identiﬁcation. Two popula- quencing are indicated by arrows.
248 The Journal of Heredity 2000:91(3)
chondrial control region are shown in Ta- ware (version 0.6, written by Don Gilbert, populations of S. tawilis, as shown in Fig-
ble 1 and in Figure 2. Hot-Start PCR was Indiana University, Bloomington, IN) and ure 2.
performed in a 50 l reaction containing the alignments were then corrected man- The complete control region sequence
Mg2 -free PCR buffer [60 mM Tris, 15 mM ually. For the control region analyses, Alo- of the Philippine sardines is 804 nucleo-
( NH4)2SO4, pH 8.5 ( Invitrogen, Leek, The sa sapidissima, Alosa pseudoharengus, and tides long for S. ﬁmbriata and S. longiceps,
Netherlands)], 250 M dNTPs, 0.5 M of Clupea harengus as the most closely relat- while it ranges from 1231 to 1418 nucleo-
each primer, 1.0 l DNA template, and 2.5 ed sequences in the GenBank database tides in S. albella and the two populations
units Taq polymerase (Pharmacia Bio- served as outgroups. Maximum parsimony of S. tawilis due to the presence of an 81
tech). Magnesium was supplied to the re- (MP) analysis was performed using the bp insertion and several copies of a 35 bp
actions by adding HotWax-Mg2 beads ( In- branch and bound search algorithm of the repetitive element near the tRNAPro region
vitrogen). Ampliﬁcation was performed in PAUP 3.1.1 software (Swofford 1993). The ( Figure 3). At its 5 end the 81 bp insertion
the PTC-200 thermal cycler (MJ Research, topology of the consensus tree was eval- repeats the 32 bp preceding the insertion
Watertown, MA, and Biozym, Hess. Old- uated by 500 bootstrap replications. site. The copy number of the 35 bp repet-
endorf, Germany). The ampliﬁcation pro- Neighbor-joining ( NJ) trees (Saitou and itive element varied from ﬁve to eight in
ﬁle consisted of a 3-min preliminary de- Nei 1987) were constructed from a Kimura the sequenced fragments. This variation
naturation at 94 C, followed by 30 cycles two-parameter distance matrix ( Kimura was also observed within individuals,
of denaturation at 94 C for 30 s, primer an- 1980) using the MEGA program ( Kumar et thereby producing more than one form of
nealing at 48 C–55 C for 45 s, and primer al. 1993). Input orders of the taxa were var- mtDNA per ﬁsh. To verify this phenome-
extension at 72 C for 2 min. An extension ied to test the robustness of the trees. The non of heteroplasmy, the PCR products
step at 72 C for 5 min completed the re- reliability of the tree topology was as- from tissue samples of all species and
action. After thermal cycling, the PCR sessed by 500 bootstrap replications ( Fel- from cloned control region fragments were
products were separated on 1% agarose senstein 1985). The number of transitions electrophoresed through a 2.5% NuSieve 3:
gel ( NEEO ultra-quality, Roth, Karlsruhe, and transversions was calculated using 1 agarose gel with high resolving power.
Germany) and retrieved using the QIA- the MEGA program. The analysis revealed three to ﬁve bands
quick Gel Extraction Kit (QIAGEN, Hilden, ranging from 800 to 1100 bp in mtDNA
Germany). samples from S. albella and S. tawilis, but
Results only single bands in S. longiceps or S. ﬁm-
Cloning briata and in the samples of cloned control
The puriﬁed PCR products were ligated Nucleotide Variation in the Cytochrome region fragments from either species, thus
into pGEM-T Easy Vector (Promega, Mann- b Gene conﬁrming the presence of heteroplasmic
heim, Germany) and transformed into E. The ampliﬁcation of a 359-nucleotide long mtDNA in S. albella and S. tawilis and ex-
coli JM109 according to manufacturer’s in- cytb fragment from ﬁshes of two popula- cluding PCR artifacts.
structions. Plasmid DNA from recombi- tions of the freshwater sardine species S. Comparison of the sardine control re-
nant clones was isolated using the QIA- tawilis from Taal Lake and three marine gion sequences in Figure 3 shows low
GEN Plasmid Mini Kit. sardine species from the Balayan Bay similarity between the two main haplo-
produced identical sequences except for typic sequences. The highest variability
Sequencing single individual-speciﬁc substitutions resides in the ﬁrst 940 nucleotides fol-
Single-stranded plasmid DNA was se- (GenBank accession codes AF1014944– lowed by low variability in the rest of the
quenced using ALF DNA sequencer (Phar- AF104948). Only a single transition at po- sequence. Conserved sequence blocks
macia Biotech) and/or LI-COR Automated sition 166 was shared by two individuals (CSBs), CSB-D and CSB-2, homologous to
Sequencer 4200 (MWG-Biotech, Ebers- from the northern population of S. tawilis those of other vertebrates ( Lee et al.
berg, Germany). For ALF, sequencing re- ( Talisay). It occurred at a nonsynony- 1995), are present together with the
actions were prepared using either the mous site leading to the conservative GTGGG-box.
AutoRead 200 Sequencing Kit (Pharma- amino acid replacement Ile to Val.
cia Biotech) or the SequiTherm EXCEL Sequence-Based Phylogeny
Long-Read Premix DNA Sequencing Kit- Nucleotide Variation in the Control Control region. Only mitochondrial control
ALF ( Epicentre Technologies, Biozym). LI- Region sequences were informative for phyloge-
COR sequencing reactions were prepared For the ampliﬁcation of the mitochondrial netic analysis. The control region analysis
using the SequiTherm Long-Read DNA control region of the different sardine spe- was divided into two parts. First, the seg-
Sequencing Kit-LC ( Epicentre Technolo- cies, distinct primer sets had to be ap- ment between the tRNAPro gene and the
gies, Biozym) with primers labeled with plied, suggesting haplotypic variation be- conserved sequence blocks was used for
two different ﬂuorescent dyes ( IRD 700 tween species. The primer pair L15926 the assessment of the phylogenetic posi-
and IRD 800, MWG-Biotech). and H16498 ampliﬁed the ﬁrst two-thirds tion of the sardines because of the large
of the control region of S. ﬁmbriata and S. number of informative sites and because
Phylogenetic Analysis/Sequence longiceps to produce a fragment of 600 bp, corresponding sequences from related
Analysis whereas the primer combination Q and B ﬁsh species were available in the GenBank
Only sequences obtained by at least two generated the corresponding product in S. sequence database. Second, all homolo-
independent PCRs were considered. The albella and the two populations of S. taw- gous segments of the control region were
cytb and the control region sequences ilis. The remaining portions were ampliﬁed used for a more detailed analysis within
were aligned together with related se- with the primer pairs A87/A88 for the two the genus Sardinella. In the ﬁrst part, the
quences available from the GenBank da- populations of S. tawilis, J/A88 for all the analysis was restricted to the alignable
tabase with the help of the SeqPup soft- sardines, and J/N for S. albella and the two segments (positions 699 to 961 in Figure
Samonte et al • Philippine Freshwater Sardine Phylogeny 249
Figure 3. Sequence alignment of the sardine mitochondrial control region. Representative sardine sequences were aligned to control region sequences of A. sapidissima, A.
pseudoharengus, and C. harengus. The sardine sequences are distinguished by a number for the individual specimen, separated by a dot from the sequence number. The two
subpopulations of S. tawilis are indicated by a T ( Talisay) and an A (Agoncillo). A dash (–) indicates identity to the sequence on top, an asterisk (*) an indel, a dot (.) missing
information, a ( N) undetermined nucleotide at the primer annealing site. The 32 bp segment preceding the 81 bp insertion site is underlined, solid bars indicate the 35 bp
repeat units, the shaded bar the central conserved region. Conserved sequence blocks are boxed. The sequences reported in this article have been deposited in the GenBank
database (accession codes AF104949–AF104969).
3) of the two main sardine haplotypes and cluster together in one branch, and the while others seem to represent intraspe-
the sequences of the most closely related freshwater species S. tawilis and the ma- ciﬁc variants.
Clupeidae ( herrings), Alosa sapidissima rine species S. albella group in a separate
(GenBank accession U12061), A. pseudo- branch. The clades are supported by high
harengus (accession U12067), and Clupea bootstrap values.
harengus (accession U12062), excluding all To evaluate the relationships within the Structural Analysis
repeat elements. The alignment of the re- genus Sardinella, the nonrepeat portions The ampliﬁed segments of the Sardinella
maining 263 nucleotides was used to re- of the control region of S. tawilis and S. mtDNA showed features characteristic of
construct phylogenetic trees by the neigh- albella were aligned (positions 45–160, the respective regions. The selective pres-
bor-joining and maximum parsimony 241–487, 594–1058) . A total number of 30 sure exerted on the protein-coding cytb
methods ( Figure 4). Both trees show es- variable nucleotide sites was found in the gene kept its sequence conserved among
sentially the same topology and reveal the 626 nucleotides considered, some of the sardines. In stark contrast, in the con-
divergence of the sardine control region which distinguish S. albella from S. tawilis trol region, even primer sets known to am-
into two major haplotypes evident already (substitutions at positions 100, 263, 279) plify mtDNA from species as distantly re-
from the presence or absence of the re- or the two subpopulations of S. tawilis lated as teleost ﬁshes and primates failed
peat elements: S. ﬁmbriata and S. longiceps from each other (positions 778, 846), to amplify two of the sardine species. The
250 The Journal of Heredity 2000:91(3)
Figure 4. (A) Maximum parsimony tree. The tree was
constructed from alignable segments of clupeid control
regions using the branch-and-bound option of the PAUP
program. Numbers at the nodes indicate the percent
recovery in 500 bootstrap replications.
due to multiple copies of repeat units. The
ﬁrst of these repeat units, an 81 bp inser-
tion, was found in some S. tawilis sequenc-
es and in S. albella. The second repeat el-
ement, a tandem repeat of 35 bp, was
present in up to eight copies in S. tawilis
and S. albella. The single repeat units
showed low or no sequence variation from
each other indicating a recent divergence,
whereas the similarity to the monomeric
element in S. ﬁmbriata and S. longiceps
was rather low. To calculate the ratio of
transitions to transversions in the control
region, 626 repeat-free nucleotides from S.
tawilis and S. albella were aligned. Six to
17 transitions and zero to ﬁve transver-
sions were observed. The mean ratio of
transitions to transversions in the control
region thus was 4.3, suggesting no satu-
ration effects in the Sardinella sequences.
Copy number polymorphism ( heteroplas-
my) of the 35 bp repeat element was found
Figure 3. Continued.
structural organization of the control re- These segments include the ﬁrst 13 nu-
gion in S. longiceps and S. ﬁmbriata is dra- cleotides after the tRNAPro gene and the
matically different from that in S. albella stretch from the central conserved region Figure 4. (B) Neighbor-joining tree. A Kimura two-pa-
and S. tawilis, thus deﬁning two different to the tRNAPhe gene. The variable portion, rameter distance matrix of the control region sequenc-
major haplotypes, although homologous which encompasses sequences without es was used to construct a neighbor-joining tree using
the MEGA program and the pairwise deletion option
segments clearly identify all four species similarity between the two haplotypes, is for missing information. Numbers at the nodes indicate
as closely related to each other. mainly characterized by length variation the percent recovery in 500 bootstrap replications.
Samonte et al • Philippine Freshwater Sardine Phylogeny 251
to be present within 17 tested individuals cies cannot be dismissed. Second, the two FishBase, 1995. Fishbase: a biological database on ﬁsh,
version 1.2. Manila, Philippines: ICLARM.
of both populations of S. tawilis and 7 in- subpopulations of S. tawilis show indica-
Gonzalez Villasenor LI, Burkhoff AM, Corces V, and
dividuals of S. albella (data not shown). tions of genetic differentiation. The pres- Powers DA, 1990. Characterization of cloned mitochon-
Heteroplasmy has been noticed before in ence of a population-restricted substitu- drial DNA from the teleost Fundulus heteroclitus and its
various ﬁsh species, among them Atlantic tion in the control region is a ﬁrst hint for usefulness as an interspecies hybridization probe. Can
J Fish Aquat Sci 43:1866–1872.
cod (Arnason and Rand 1992), sturgeon genetic separation, possibly caused by re-
Gyllensten U and Wilson AC, 1987. Mitochondrial DNA
( Brown et al. 1992; Miracle and Campton duced ﬁsh migration across a shallow bar- of salmonids: inter- and intraspeciﬁc variability detect-
1995), and redﬁshes ( Bentzen et al. 1998). rier in the middle of the lake. A more ex- ed with restriction endonucleases. In: Population ge-
The occurrence of heteroplasmic mtDNA tended investigation with large sample netics and ﬁshery management. Seattle: University of
Washington Press; 302–317.
in only two of the sardine species and the sizes is needed to answer these questions.
Hargrove TR, 1991. The mysteries of Taal. Flagstaff, AZ:
virtual identity of the repeat sequences AGCOM International.
suggest a recent origin of the tandem re-
Harrison RG, 1989. Animal mitochondrial DNA as a ge-
peat array. References netic marker in population and evolutionary biology.
Alves-Gomes JA, Orti G, Haygood M, Heiligenberg W, Trends Ecol Evol 4:6–11.
Phylogenetics and Meyer A, 1995. Phylogenetic analysis of the South Hartley SE, Barlett SE, and Davidson WS, 1992. Mito-
American electric ﬁshes (Order Gymnotiformes) and chondrial DNA analysis of Scottish populations of Arc-
Because of lack of variation, cytb sequenc- the evolution of their electrogenic system: a synthesis tic charr, Salvelinus alpinus ( L). J Fish Biol 40:219–224.
es are of little use for the phylogenetic based on morphology, electrophysiology and mito-
chondrial sequence data. Mol Biol Evol 12:298–318. Irwin DM, Kocher TD, and Wilson AC, 1991. Evolution
analysis of the sardine species. However, of the cytochrome b gene of mammals. J Mol Evol 32:
a single nonsynonymous substitution was ´
Arnason E and Rand DM, 1992. Heteroplasmy of short 128–144.
tandem repeats in mitochondrial DNA of Atlantic cod,
found that may prove valuable in the rec- Gadus morhua. Genetics 132:211–220. Kimura M, 1980. A simple method for estimating evo-
ognition of two subpopulations of S. taw- lutionary rates of base substitutions through compar-
Avise JC, Reeb CA, and Saunders N, 1987. Geographic ative studies of nucleotide sequences. J Mol Evol 16:
ilis in Taal Lake. As expected the phylo- population structure and species differences in mito- 111–120.
genetic signal present in the control chondrial DNA of mouthbrooding marine catﬁshes (Ari-
idae) and demersal spawning toadﬁshes ( Batrachoidi- Kira T (ed), 1995. Data book of world lake environ-
region is more useful. Among 626 nucleo- dae). Evolution 41:991–1002. ments—a survey of the state of world lakes—Asia &
tides, excluding repeat-induced indels, Oceania. Otsu, Japan: Miyagawa Printing; 545.
Barlett SE and Davidson WS, 1991. Identiﬁcation of
eight to 22 substitutions were detected in Thunnus tuna species by the polymerase chain reaction Kocher TD, Thomas WK, Meyer A, et al., 1989. Dynam-
pairwise comparisons in S. tawilis and S. and direct analysis of their mitochondrial cytochrome ics of mitochondrial DNA evolution in animals: ampli-
b genes. Can J Fish Aquat Sci 48:309–317. ﬁcation and sequencing with conserved primers. Proc
albella. A similarly high rate variation was Natl Acad Sci USA 86:6196–6200.
observed by McMillan and Palumbi (1997) Bentzen P, Wright JM, Bryden LT, Sargent M, and Zwa-
nenburg KCT, 1998. Tandem repeat polymorphism and Kumar S, Tamura K, and Nei M, 1993. MEGA: molecular
in Paciﬁc butterﬂyﬁshes, where the con- heteroplasmy in the mitochondrial control region of evolutionary genetic analysis, version 1.0. University
trol region evolved 33–43 times faster than redﬁshes (Sebastes: Scorpaenidae). J Hered 89:1–7. Park, PA: Pennsylvania State University.
the cytb gene. Furthermore, the control re- Bernatchez L, Savard L, Dodson JJ, and Pallotta D, Lee W-J, Conroy J, Howell WH, and Kocher TD, 1995.
1988. Mitochondrial DNA sequence heterogeneity Structure and evolution of teleost mitochondrial con-
gion harbors a set of repeat-related indels trol regions. J Mol Evol 41:54–66.
among James–Hudson Bay anadromous coregonines.
that is useful as a cladistic marker. The Finn Fish Res 9:17–26. Martin AP, Naylor GJP, Palumbi SR, 1992. Rates of mi-
aligned sardine sequences were used to Billington N and Hebert PDN, 1991. Mitochondrial DNA tochondrial DNA evolution in sharks are slow com-
construct neighbor-joining and maximum diversity in ﬁshes and its implications for introduc- pared with mammals. Nature 357:153–155.
parsimony phylogenetic trees to deter- tions. Can J Fish Aquat Sci 48:80–94. McMillan WO and Palumbi SR, 1997. Rapid rate of con-
mine the closest relative of the freshwater Broughton RE and Dowling TE, 1994. Length variation trol-region evolution in Paciﬁc butterﬂyﬁshes (Chaeto-
in mitochondrial DNA of the Minnow Cyprinella spilop- dontidae). J Mol Evol 45:473–484.
sardine species S. tawilis ( Figure 4). Both tera. Genetics 138:179–190. McVeigh HP, Barlett SE, and Davidson WS, 1991. Poly-
trees are congruent and support the re- merase chain reaction/direct sequence analysis of the
Brown WM, 1983. Evolution of animal mitochondrial
cent reclassiﬁcation of S. tawilis into the DNA. In: Evolution of genes and proteins ( Nei M and cytochrome b gene in Salmo salar. Aquaculture 95:225–
genus Sardinella. Koehn RK, eds). city?: publisher?; 62–88. 233.
Two additional conclusions can be Brown JR, Beckenbach AT, and Smith MJ, 1992. Mito- Meyer A and Wilson AC, 1990. Origin of tetrapods in-
chondrial DNA length variation and heteroplasmy in ferred from their mitochondrial DNA afﬁliation to lung-
drawn. First, the closest marine relative to ﬁsh. J Mol Evol 31:359–64.
populations of white sturgeon (Acipenser transmontan-
S. tawilis is S. albella. Its grouping within us). Genetics 132:221–228.
Miracle AL and Campton DE, 1995. Tandem repeat se-
the tawilis branch and the small genetic Brown JR, Beckenbach AT, and Smith MJ, 1993. Intra- quence variation and length heteroplasmy in the mi-
distance of the tested individual to fresh- speciﬁc DNA sequence variation of the mitochondrial tochondrial DNA D-loop of the threatened gulf of Mex-
water sardines identiﬁes it as the most control region of white sturgeon (Acipenser transmon- ico sturgeon, Acipenser oxyrhynchus desotoi. J Hered 86:
tanus). Mol Biol Evol 10:326–341. 22–27.
likely ancestor to the Taal Lake sardines.
Carr SM and Marshall HD, 1991. Detection of intraspe- Moritz C, Dowling TE, and Brown WM, 1987. Evolution
Its control sequence differs by 1.3–3.5% ciﬁc DNA sequence variation in the mitochondrial cytb of animal mitochondrial DNA: relevance for population
from the S. tawilis sequences which show gene of Atlantic cod (Gadus morhua) by polymerase biology and systematics. Annu Rev Ecol Syst 18:269–
an intraspeciﬁc variation of 1.4–3.0%. An- chain reaction. Can J Fish Aquat Sci 48:48–52. 292.
cestral polymorphism and incomplete lin- Chapman RW and Powers DA, 1984. A method for rapid Munro ISR, 1995. The marine and freshwater ﬁshes of
isolation of mtDNA from ﬁshes. Technical report no. Ceylon. Canberra: Department of Environmental Af-
eage sorting, perhaps in combination with UM-SG-TS-84-05. College Park, MD: Maryland Sea Grant fairs.
an accelerated mutation rate in the con- Program.
Palumbi SR and Kessing BD, 1991. Population biology
trol region, may therefore account for the Clayton DA, 1982. Replication of animal DNA. Cell 28: of the trans-arctic exchange: mtDNA sequence similar-
observed DNA diversiﬁcation of S. tawilis 693–705. ity between Paciﬁc and Atlantic sea urchins. Evolution
from S. albella in the time span of some Conlu PV, 1986. Guide to Philippine ﬂora and fauna. Vol.
240 years since the formation of Taal Lake. IX. Manila, Philippines: National Resources Manage- Rau N and Rau A, 1980. Commercial marine ﬁshes of
ment Center, Ministry of Natural Resources, and Uni- the central Philippines ( bony ﬁshes). German Agency
The possibility, however, that S. tawilis versity of the Philippines. for Technical Cooperation; 25, 196, 203, 205.
and S. albella are ecomorphs of the same Felsenstein J, 1985. Conﬁdence limits on phylogenies: Russell PJ, 1994. Fundamentals of genetics. New York:
species and not truly differentiated spe- an approach using the bootstrap. Evolution 39:783–791. Harper Collins.
252 The Journal of Heredity 2000:91(3)
Saccone C, Attimonelli M, and Sbisa E, 1987. Structural parsimony, version 3.1. Champaign, IL: Illinois Natural Smith’s sea ﬁshes (Smith MM and Heemstra PC, eds).
elements highly preserved during the evolution of the History Survey. Berlin: Springer-Verlag; 199–204.
D-loop-containing region in vertebrate mitochondrial Zhu D, Jamieson BGM, Hugall A, and Moritz C, 1994.
DNA. J Mol Evol 26:205–211. Villanueva LS, Luistro AP, and Calabig CS, 1996. As-
Sequence evolution and phylogenetic signal in control-
sessment of Taal Lake capture ﬁsheries with emphasis
region and cytochrome b sequences of rainbow ﬁshes
Saiki RK, Gelfand DH, Stoffel, et al., 1988. Primer-direct- on the exploitation of Harengula tawilis. STIARC. (Melanotaeniidae). Mol Biol Evol 11:672–683.
ed enzymatic ampliﬁcation of DNA with a thermostable
DNA polymerase. Science 239:487–491. Whitehead PJP, 1985. FAO species catalogue. Vol. 7. Clu- ¨¨
Zischler H, Geisert H, von Haeseler A, and Paabo S,
peoid ﬁshes of the world. An annotated and illustrated 1995. A nuclear ‘‘fossil’’ of the mitochondrial D-loop
Saitou N and Nei M, 1987. The neighbor-joining meth- catalogue of the herrings, sardines, pilchards, sprats, and the origins of modern humans. Nature 378:489–492.
od: a new method for reconstructing phylogenetic shads, anchovies and wolf-herrings. Part 1—Chirocen- Received November 24, 1998
trees. Mol Biol Evol 4:406–425. tridae, Clupeidae and Pristigasteridae. FAO; 303. Accepted November 30, 1999
Swofford DL, 1993. PAUP: phylogenetic analysis using Whitehead PJP and Wongratana T, 1986. Clupeidae. In: Corresponding Editor: Stephen J. O’Brien
Samonte et al • Philippine Freshwater Sardine Phylogeny 253