CORE Metadata, citation and similar papers at core.ac.uk
Provided by Nature Precedings
Population genetic structure of the pelagic goby, Sufflogobius bibarbatus, in the Northern Benguela ecosystem, based on PCR- RFLP analysis of the mitochondrial control region and the ND 3 / 4 region
M.Phil Thesis
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 Pallagae Mangala Chathura Surendra De Silva
Department of Biology University of Bergen, Norway 2005
Population genetic structure of the pelagic goby, Sufflogobius bibarbatus, in the Northern Benguela ecosystem, based on PCR-RFLP analysis of the mitochondrial control region and the ND 3 ⁄ 4 region.
By
Pallagae Mangala Chathura Surendra De Silva
A thesis submitted in partial fulfilment of the requirements for the degree of Master of Philosophy In Fisheries Biology and Fisheries Management
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
Department of Biology University of Bergen Bergen, Norway 2005
Table of Contents
Acknowledgements…………………………………………………………………...... 2 Abstract…………………………………………………………………………………3 1. Introduction ……………………………………………………………………4
2. Materials & Methods …………………………………………………………..8 2.1. Sampling ……………………………………………………………………8 2.2. DNA Extraction……………………………………………………………10 2.3. PCR Methodology …………………………………………………………10 2.4. Restriction Fragment Length Polymorphism (RFLP) analysis …………12 2.5. Data Analysis ………………………………………………………………14
3. Results …………………………………………………………………………16 3.1. PCR products and Restriction Patterns ………………………………… 16 3.2. Haplotype numbers and Frequency Distribution………………………..16 3.3. Unique Haplotypes ……………………………………………………….19 3.4. Haplotype and Nucleotide Diversities ……………………………………19 3.5. Exact Test and Chi Square test …………………………………………..21
4. Discussion ……………………………………………………………………..23 4.1. General Overview …………………………………………………………23 4.2. Selection of Methods ………………………………………………………24 Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 4.3. Selection of Materials and Sampling Sites ……………………………….25 4.4. Variation within Areas ……………………………………………………26 4.5. Population Studies ………………………………………………………..27 4.6. Genetic Deviation through Haplotype Frequencies ……………………..28 4.7. Contributing Factors ……………………………………………………..29
5. Conclusions ……………………………………………………………………34 6. References ……………………………………………………………………..35 7. Appendix………………………………………………………………………..44
1
Acknowledgements
I am very much grateful to my supervisor, Professor Gunnar Naevdal at Department of Biology, University of Bergen, Norway and co supervisor Dr K.B Suneetha at Department of Zoology, University of Ruhuna, Sri Lanka for their guidance and encouragement through out my study. The immense help given by Professor Anne Gro Salvanes at Department of Biology is highly appreciated. I am really thankful for your inspiring supervision, indispensable guidance, sincerity and invaluable suggestions on my work.
It is very pleasant to work in the genetics lab in the Department of Biology and the technical assistance given by Solveig Thorkildsen during the laboratory work is very much appreciated. Special thanks for Dr Geir Dahle and Dr Torild Johansen for their enormous support in statistical analysis. I admire the support given by Berit Ogland and Kristin Paulsen through out my study period.
Also it is highly commendable all the support and assistance given by all my friends during my difficult times. I am grateful to Pham Mai Huong for your lovely smile, affection and kindness towards me. I am very much flattered by the invaluable help given by my best friends Janna, David, Srisuda, Hilka, Janya, Stutada, Ketut and Diep. All of you deserved the credit for this work otherwise I am sure that I could have ended up with nothing. Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
Finally but not at last, my special gratitude to my parents, for taking me this journey this much. Although you are far away from me, you never forgot me and in my heart I was always living with you. I am very much pleased for your love, trust and everlasting kindness.
2
Abstract
Polymorphism within the mitochondrial NADH-3,4 dehydrogenase(ND-3/4)) and control region (D-Loop) was studied by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) analysis of Sufflogobius bibarbatus (Pelagic Goby) population in the Northern Benguela ecosystem. Three major geographical areas representing the north-south axis (Ludritz, Walvis Bay and North of Ambrose Bay) were selected. Polymorphism was detected using three and two restriction enzymes respectively in ND- 3 / 4 and D-Loop regions, and a total number of 25 composite haplotypes were identified. The most common haplotype accounted for 36% of 238 individuals. Recorded haplotype diversities were quite high, ranging 0.70 to 0.86.The highest haplotype diversity was found in Walvis Bay followed by Ludritz and North of Ambrose Bay. Pairwise Fst values among regions indicated that significantly population differentiation existing. The Monte Carlo test revealed a significant difference among three geographical areas (P <0.05). Pairwise comparisons based on haplotype frequencies revealed significant differences between Ludritz and Walvis Bay (P < 0.05) and Ludritz and North of Ambrose Bay (P < 0.01) while no difference was found between Walvis Bay and North of Ambrose Bay (P >0.05).Similar observations were found by Exact test for haplotype frequencies. This study has shown existence of population subdivision of the pelagic goby in the Northern Benguela. Combination of factors like complex circulation pattern, physical barrier around Moeb Bay and adaptive selection can be Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 contributed to the observed divergence. Identified population sub structure will be vital in future decision and policy making of the proposed ecosystem based management in the Benguela region.
3 1. Introduction
The Benguela ecosystem is one of the four most productive ecosystems in the world (Cushing, 1971). It is a unique coastal upwelling system in the world of ocean, bounded at both equatorward and poleward ends by warm water regimes. The region, where cool upwelled coastal waters to be found includes coast of Namibia and the western coast of South Africa. The Northern limit is bounded by Angola system and southern by Agulhas current retroflection area (Shannon, 1985). Previous studies have divided the Benguela ecosystem into two major oceanographic areas, a Northern and Southern region. The Northern Benguela represents Cunene river mouth to 290 S. The conventional border which divides the Northern and Southern Benguela is around Orange River mouth (290 S). And from the border the Southern Benguela extends to East London (280 E) (Jarre- Teichmann et al, 1998). The ecosystem is extremely vital for the region due to its supportive role in commercially valuable fisheries.
Sufflogobius bibarbatus (Van Bonde, 1923), commonly known as pelagic goby, is endemic to the region and one of the key species in the Benguela ecosystem. Crawford (1987) describes the distribution of this species in coastal shelfwaters between 220S and 270S. More specifically the highest concentrations of pelagic goby can be found mainly between Walvis Bay and Ludertiz (Hewitson and Cruickshank, 1993). Also it is suggested that the distribution can be expanded as far north as Cunnene River (Cruickshank, 1980). The larval distribution has been recorded between 170 30`S and 330 30`S and over the 50–300m isobath (O`Toole, 1977; Shelton et al, 1986). The abundance Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 of the larvae could be seen through out the year, but peak abundance occurs during intense upwelling periods. Juveniles of this species are epipelagic and adults are found in deep waters. They prefer more open substrates and are particularly associated with soft substrate enrichd with benthic food (Gibbons et al, 2000). They spend much of their time in the water column during the night. It leads to the fact that pelagic goby use demersal habitats mainly for shelter (Gibbons et al, 2002). Early studies indicate that pelagic goby is mainly feeds on phytoplankton (Crawford, 1987). This is quite contradictory to recent studies where no phytoplankton has been recorded as their main food source (Gibbons et al, 2002).
4
Ecologically pelagic goby plays an important role in the Benguela ecosystem. It is one of the key prey species in the food chain. Shannon et al (1999) estimated that 1.5 million tons of gobies are prey for Benguela fish species such as hake (Merluccus capensis), penguin (Spheniscus demersus), Cape fur seal (Arctocephalus pusillus pusillus) and skate (Raja clarata). Their importance as a prey is clearly highlighted by reduction of hake and horse mackerel stocks, which reduce the predatory pressure on this species. During the same time Penguin and Cape fur seal shifted their prey items such as hake and horse mackerel to the pelagic goby (Hewitson and Cruickshank, 1993; Binachi, 1999). O`Toole (1977) reported that spawning of the pelagic goby and commercially important pilchard, Sardinops ocellatus, is occurring at the same time, hence they might be competing for same food resources. So in spite of few studies on pelagic goby, it is important to understand its role in the food web.
In spite of the ecological role, still this species has a very limited commercial value in major fisheries in the region. It is estimated out that annual catch of this species as 0.001 Tons/ Km2/ year. This amount seems negligible compared to estimated total biomass of all commercial species 600,000 tons in Northern Benguela (Shannon and Jarre- Teichmann, 1999). Recent collapse in the commercially important pelagic fishery might create a new dimension in these unexploited stocks of pelagic goby. It is expected that pelagic goby will play an important role in the commercial fish sector in near future. Also its unique ecological role in the Benguela ecosystem will be more highlighted when implementing the ecosystem based management in the Benguela. In spite of these Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 prospects the life history and population structure of this species is still not known. The population structure in relation to environmental variables of this species is extremely vital for future exploitation as well as to improve the understanding of the ecological role of pelagic goby in the Benguela ecosystem.
It is believed that identification of the population structure of marine organisms is rather difficult comparing to that of freshwater and terrestrial organisms (Ward et al. 1994). This is attributable to high dispersal ability of marine organisms due to less topographical barriers, which enhance the intermixing of the population. However oceanographic
5 features such as currents and tides further increase their ability of forming large panmictic units, which can be separated by oceanographic barriers (Palumbi et al, 2001). Nevertheless, identification of genetically meaningful units is extremely essential for the species that play an important role in the marine environment (Moran et al. 1999). A number of methods have been used in analyzing population structure of marine species. This includes morphometrics, growth rates, age composition, phenotypic and genetic analysis.
The use of molecular genetics in population studies goes as back as 1950. Genetic markers has been accepted as a valuable tool in studies of population structures (Utter, 1987), especially in marine fish (Nesbo et al, 2000; Ruzzante et al, 2000) and can be combined with phenotypic or ecological data. Genetic molecular markers such as allozymes, mtDNA, and microsatellites are commonly used to detect genetic diversity each with increasing temporal resolution. Genetic variation detected through these techniques is considered as neutral or nearly neutral to natural selection. Analysis of DNA has increasingly become more important in population studies (Bernatchez and Danzmann, 1993; Avise et al, 1994).
Apart from about 99% of the DNA which resides within the nucleus, mitochondria DNA (mtDNA) also plays an important role in population structure analysis due to its unique characters. The DNA found in mitochondria is circular in conformation, relatively small (approximately 16000 base pairs in size), maternally inherited and non-recombinatory. Also it exhibits a rapid rate of nucleotide change. The differentiation among populations Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 is thought to be approximately four times faster than that of nuclear genes (Wilson et al, 1985) and is therefore more likely to show differences between populations. Mitochondrial DNA variation has been examined for stocks of several species of marine fish (Bagley and Gall, 1998; Chubb et al, 1998; Arnaud et al, 1999).
This study will be the first of its kind on identification of population structure of pelagic goby in the Northern Benguela ecosystem. MtDNA variation of the population has been studied through Polymerase Chain Reaction (PCR) based Restriction Fragment Length Polymorphism (RFLP). PCR-RFLP is a straightforward and cost effective method in
6 analyzing mtDNA variation among populations as an alternative to sequencing. Nevertheless RFLP has been proved to be useful tool in population studies (Avise, 1994; Hall and Nawrocki, 1995). In the mt genome, protein coding genes are considered to be conserved while the control region is known to be highly variable. Many population studies have utilized the genetic variation in the control region to describe population structures (Szalanski et al, 2000).In the present study, partial NADH dehydrogenase 3,4 subunits and a partial control region have been subjected to PCR-RFLP.
Aims and Objectives
1. Generate genetic data from mtDNA for Sufflogobius bibarbatus over its main areas of geographical distribution using two regions of the mt genome.
2. Use mtDNA to assess the degree of spatial genetic differentiation (population structure) in Northern Benguela ecosystem in North-South axis.
3. To investigate the identifiable populations that can be tied to ecological, oceanographical and environmental variables.
4. Identification of population structure of local populations which are expected to play a vital role in future management strategies of this species and in the ecosystem. Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
7 2. Materials & Methods
2.1. Sampling
Fish samples of Sufflogobius bibarbatus were collected on board the research vessel Dr Fridtjof Nansen, during 11-27 February 2003. A total of 300 individual fish samples were collected from selected areas, alone the north to south axis in the Northern Benguela Ecosystem (Table 1). Three divisions of the population were considered namely, northern, southern and intermediate as represented by North Ambrose Bay, Ludritz, and the Walvis Bay respectively (Figure 1). The detailed information regarding the sampling is given in Table 1. An assumption to the hypothesis to be tested is that the sampling for this goby population is representative and unbiased. The chance that this problem will occur can be minimized by increasing the number of individuals tested from each hypothesized sub-populations from North and South. All the fish samples were dissected to get gill tissues individually which were preserved in 95% ethanol until processed in the laboratory.
Table 1. Account of the material sampled for genetic studies of Suflogobius bibarbatus. (PT: pelagic trawl, BT: bottom trawl)
Geographical Number Station Station coordinates Fishing gear Gear area of number depth samples 1226 S 270 23’ E 140 54’ PT 200 M Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
Ludritz 94 1230 S 260 41’ E 140 51’ BT 160 M
1234 S 250 43’ E140 25’ BT 178 M
1236 S 230 30’ E130 23’ BT 273 M Walvis Bay 81 1241 S 220 20’ E140 10’ BT 67 M
1244 S 210 05’ E130 09’ BT 125 M North of 64 Ambrose Bay 1245 S 200 01’ E120 34’ BT 143 M
8 Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
Figure 1. Map of the Nothern Benguela with bathymetry and sampling stations. • indicates sampling stations with bottom trawling and c with pelagic trawling.
9 2.2. DNA Extraction
Genomic DNA from gills of individual fish was extracted by using commercially available Quagin Dneasy Tissue Kit following the manufactures instructions. Approximately 20-25 mg of Gill tissue was digested overnight in a 1.5ml micro- centrifuge tube contained 180µl of ATL buffer and 20µl of Protinase K. Overnight incubation was followed by adding 4 µl of Rnase and mixed thoroughly by vortexing. The samples were incubated for 2 minutes in room temperature. Then 200 µl of buffer AL was added and mixed by vortexing. They were incubated at 700 C for 10 minutes. After the incubation 200µl of 100% ethanol was added to each sample and mixed well. The mixture was pipetted to DNeasy spin columns placed in 2 ml collection tubes and centrifuged at 8000 rpm. The follow-through and the collection tubes were discarded. The DNeasy mini columns were placed in new collection tubes, 500 µl of Buffer AW1 was added to the each samples followed by centrifugation at 8000 rpm for one minute. The spin columns were collected and 500 µl of Buffer AW2 was added. These were centrifuged for 3 minutes in full speed (15000rpm). Finally Spin columns were placed in 1.5 ml clean micro centrifuge tube and 200 µl of Buffer AE was added, incubated for one minute following a centrifugation at 6000 rpm for one minute. This last step was repeated. The integrity of the extracted DNA was visually inspected on 1 % Agarose (Gibcobrc, Life Technologies) gels. Isolated DNA was stored at 40 C until amplification.
2.3. PCR Methodology
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 Restriction fragment length polymorphism analysis was performed on two polymerase chain reaction (PCR) amplified segments encompassing the displacement loop (D-Loop) and the NADH-3/4 dehydrogenase (ND-3/4) regions of mitochondrial genome (Table 2).
10 Table 2. PCR amplified mtDNA segments and their primer sequences.
Mt DNA region Primer Sequence Reference D-Loop 5' CCA CTA GCT CCC AAA GCT A 3' Bernatchez et al 5' ACT TTC TAG GGT CCA TC 3' (1993)
ND – 3/4 5’TAACGCGTATAAGTGACTTCCAA 3’ Gharrett et al 5’TTTTGGTTCCTAAGACCAATGGAT 3’ (2001)
Displacement - loop (D-Loop)
Approximately 1.2 kbp of the control region of the mtDNA was amplified using primers designed from homologies observed among published fish sequences in the proline and phenylalanine t-RNA genes described by Bernatchez et al (1993). Each PCR consisted of 4 µl of Genomic DNA, 5 µl of 10Χ Buffer, 8 µl of dNTP mixture (1.25M), 1 µl of each primer, 0.25µl of Taq DNA polymerase (Amersham Biosciences) and deionized water to bring the final volume to 50 µl. The amplification cycle consisted of 950 C denaturation for 5 minutes, followed by 35 cycles of 940 C denaturing for 40 seconds, 520 C annealing for 1 minute, 720 C extension for 2 minutes and concluded with 10 minutes extension at 720C. Thermal cycling was performed in MJ DNA Engine PTC (MJ Research, Watertown, MA).
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 NADH-3/4 dehydrogenase (ND-3/4)
Approximately 2.3 kbp region of the NADH dehydrogenase gene was amplified using the primers designed by Gharret et al (2001). Each PCR consisted of 4 µl of Genomic DNA, 5 µl of 10Χ Buffer, 8 µl of dNTP mixture (1.25M), 1 µl of each primer, 0.25µl of Taq DNA polymerase (Amersham Biosciences) and deionized water to bring the final volume of 50 µl. Cycle parameters of 5 min at 950 C, followed by 35 cycles of 40 s at 940 C, 1 min at 540 C and 3 min at 720 C with a final extension time of 10 min at 720 C were used.
11 Thermal cycling was performed in MJ DNA Engine PTC (MJ Research, Watertown, MA).
PCR products were examined by electrophoresis through a 1% agarose gel (Gibcobrc, Life Technologies) in 0.5 MTBE buffer (0.045M Tris borate, 0.045M boric acid, 0.001M EDTA, pH 8.0) and stained by ethidium bromide. As size references, a 100bp ladder (PGEM DNA marker, Promega) was used on each gel.
2.4. Restriction Fragment Length Polymorphism (RFLP) analysis
Preliminary tests of restriction enzymes were performed on the amplified D-Loop and ND- 3/4 PCR products for 20 individuals from the three different geographical areas in Nothern Benguela ecosystem (Ludritz, Walvis Bay and North of Ambrose Bay). Initially 14 restriction enzymes (Acc ΙΙΙ, Alu Ι, BamH Ι, Cfo Ι, Dde Ι, Hae ΙΙΙ, Hind ΙΙΙ, Hpa ΙΙ, Hsp92 ΙΙ, Msp Ι, Rsa Ι, Sau3A Ι, Taq Ι, Tru9 Ι) were surveyed for D-loop and 10 restriction enzymes (Alu Ι, Dde Ι, Hae ΙΙΙ, Hind ΙΙΙ, Hpa ΙΙ, Hsp92 ΙΙ, Msp Ι, Rsa Ι, Sau3A Ι, Tru9 Ι) were surveyed for ND- 3/ 4. Eleven of them recognize tetra nucleotide sequence while other three recognize hexanucleotide sequences (Table 3).
Each PCR product (10 µl) was digested over 12 hrs with 0.5 µl of each restriction enzyme (10 units / µl) ,0.2µl of Acetylated BSA (10 µg / µl), 2µl of RE 10X Buffer and deionized water which make the final volume to 20 µl. Restriction fragments were
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 visualized under UV light on 2 % Metaphor® Agarose (Cambrex) gels stained with ethidium bromide. The molecular sizes of the fragment were calculated with the comparison with co-migrating 100bp base pair ladder (PGEM DNA marker, Promega).
Based on the initial phase of the restriction analysis, the five restriction enzymes (Hsp92 ΙΙ, DdeΙ and Hae ΙΙΙ for ND – 3/4 and Rsa Ι and Hsp92 ΙΙ, for D-Loop), which produced polymorphic patterns among the initial 60 individuals, were selected and used to screen all remaining individuals. The different restriction patterns were scored by eye from the Metaphor Agorose gels. The restriction pattern for each enzyme was given a letter code. The most common pattern for each enzyme was labeled as “A”, the second most frequent
12 pattern “B”, and so on. By assigning a letter code each composite mt DNA haplotype was defined by five letter codes. The individuals that had the most restriction pattern for all five enzymes were coded as AAAAA. The composite haplotype represented the sum of patterns from every enzyme used in two different segments in the following order; Hsp92 ΙΙ, DdeΙ , Hae ΙΙΙ , Rsa Ι and Hsp92 ΙΙ.
Table 3. Recognition sequences and type of end sequence of the used restriction endonucleases in the studied mtDNA segments. G= guanine, A= adenine, C= cytosine, T= thymine, N= any nucleotide Restriction Amplified Region Recognition Sequence End Sequence Endonuclease
Acc ΙΙΙ D-loop TCCGGA T▼CCGGA AGGCCT A GGCC▲T Alu Ι D-loop / ND- 3/4 AGCT AG▼CT TCGA TC▲GA BamH Ι D-loop GGATCC G▼GATC C CCTAGG C CTAG▲G Cfo Ι D-loop GCGC G CG▼C CGCG C▲GC G DdeΙ D-loop / ND- 3/4 CTNAG C▼TNA G GANTC G ANT▲C Hae ΙΙΙ D-loop / ND- 3/4 GGCC GG▼CC CCGG CC▲GG Hind ΙΙΙ D-loop / ND- 3/4 AAGCTT A▼AGCT T TTCGAA T TCGA▲A Hpa ΙΙ D-loop / ND- 3/4 CCGG C▼CG G G GCC G GC▲C Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 Hsp92 ΙΙ D-loop / ND- 3/4 CATG CATG▼ GTAC ▲GTAC Msp Ι D-loop / ND- 3/4 CCG G C▼CG G G GCC G GC▲C Rsa Ι D-loop / ND- 3/4 GTAC GT▼AC CATG CA▲TG Sau3A Ι D-loop / ND- 3/4 GATC ▼GATC CTAG CTAG▲ Taq Ι D-loop TCG A T▼CG A A GCT A GC▲T Tru9 Ι D-loop / ND- 3/4 TTA A T▼TA A A ATT A AT▲T
13 2.5. Data Analysis
Data analysis was initiated by preparing a matrix of data, which includes presence or absence of different restriction fragment patterns found with respect to each endonuclease. The presence of a given fragment was indicated by “1” while the absence was indicated by “0”.The frequencies of different composite haplotypes were calculated for the three geographic population units. (Table 4 & Appendix 1).
Table 4. Matrix of restriction fragments presence / absence for the different patterns found for each gene complex and endonuclease.
ND – 3/4 D- Loop Hsp92 ΙΙ, Dde Ι Hae ΙΙΙ Rsa Ι Hsp92 ΙΙ, A 11101111 A 111110 A 111111100 A 11110 A 011110 B 11101110 B 111111 B 111111001 B 11111 B 011111 C 11110100 C 111111110 C 101100
Genetic diversity in each locality was estimated by unbiased haplotype diversity (H), mean number of pairwise differences and nucleotide diversity (π) (Nei, 1987) using Arlequin version 2.0 (Schneider et al 2000). Genetic differentiation among and within areas was quantified by analysis of molecular variation (AMOVA) (Excoffier et al 1992)
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 using the Arlequin, version 2.0 package. The same genetic software was used to calculate
the population pairwise genetic distances. The pairwise FST values were also calculated as short-term genetic distances between populations according to Reynolds et al (1983) and Slatkin (1995).
Two different methods were used to test the population differences based on the haplotype frequencies. The uses of exact test of population differentiation test the hypothesis of a random distribution of haplotypes among populations described by Raymond and Rousset (1995). Arlequin ver 2.0 was used for exact test with 10000 steps of Markov chain procedure. Furthermore, a chi-square test of independence (Sokal &
14 Rohlf 1981) using the Monte Carlo randomization method (Roff and Bentzen 1989), in the program created with data analysis software (Microsoft Excel™, Washington, DC, USA) by Sugaya et al (2002), was used to test for significant differences in haplotype frequencies between pairs of localities. The use of Monte Carlo method ensures the problems associated with empty cells and infrequent haplotypes due to small sample sizes. The sequential Bonferroni procedure suggested by Rice (1989) was used to correct for multiple pairwise comparisons.
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
15 3. Results
3.1. PCR products and restriction patterns
Gel electrophoresis of the amplified PCR products of ND-3/ 4 gene region and D-Loop of the mitochondrial DNA did not reveal any detectable length differences among individuals. The approximate sizes of the PCR amplified product were ≈ 2.2 kbp for ND- 3 / 4 and ≈ 1.2 kbp for the D-loop. Polymorphisms, within ND-3/4 and D-Loop were revealed by three and two restriction enzymes respectively. The numbers of restriction morphs (patterns) were two or three in the enzymes that revealed polymorphism in this study. Three restriction patterns were observed by Hae ΙΙΙ and Hsp92 ΙΙ for the ND-3/4 region, while DdeΙ produced two restriction patterns. In the D-Loop Hsp92 ΙΙ produced three restriction patterns while Rsa Ι produced two patterns. The most common restriction pattern for each endonuclease accounted for 31% in area surrounding Ludritz and Walvis Bay and 52% in North of Ambrose Bay. Compositing the restriction patterns of the five endonucleases in two different segments of the mitochondrial genome have resulted a total of 25 haplotypes among the 238 individuals analyzed in the three geographical areas (Table 5).
3.2. Haplotype numbers and frequency distribution
The 25 composite haplotypes were found to be distributed among the three studied geographical areas; Ludritz, Walvis Bay and North Ambrose Bay. The number of Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 haplotypes found in each region varied and the total number of mitochondrial haplotypes in Ludritz, Walvis Bay and North Ambrose Bay was thirteen, seventeen and fourteen respectively (Table 6). The most common haplotypes in the Ludritz samples were H 1 (AAAAA) and H 6 (BAAAA). Haplotypes H 1 (AAAAA) and H 2 (AAABA) were recorded in highest frequencies in Walvis Bay and North Ambrose Bay. The AAAAA haplotype accounted for 36% of all recorded individual haplotypes. Totally eight composite haplotypes was shared among all three geographical areas. Haplotypes H 5 (BAABA) and H 8 (CAABA) were shared only among Ludritz and Walvis Bay. H 9 (ABAAA) was only shared between Ludritz and North Ambrose Bay.
16 Table 5. Twenty- five mt DNA haplotypes identified in pelagic goby population by RFLP analysis on ND-3/4 and D-Loop.
ND- 3/4 D-Loop Haplotype Composite Hsp92 ΙΙ Dde Ι Hae ΙΙΙ Rsa Ι Hsp92 ΙΙ Haplotype
H 1 A A A A A AAAAA
H 2 A A A B A AAABA
H 3 A A A A B AAAAB
H 4 A A A B B AAABB
H 5 B A A B A BAABA
H 6 B A A A A BAAAA
H 7 B A A A B BAAAB
H 8 C A A B A CAABA
H 9 A B A A A ABAAA
H 10 A A A A C AAAAC
H 11 C A A A A CAAAA
H 12 B A A B B BAABB
H 13 B B A A A BBAAA
H 14 B A C B A BACBA
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 H 15 A A B A A AABAA
H 16 B A B B A BABBA
H 17 B A B B B BABBB
H 18 B A B A B BABAB
H 19 C A B B A CABBA
H 20 A A B B A AABBA
H 21 B B A B B BBABB
H 22 C A A B B CAABB
H 23 C A A A B CAAAB
H 24 A B A A B ABAAB
H 25 A A C A A AACAA 17 Table 6. Haplotype numbers and frequencies (given by bold numbers) of the mitochondrial DNA (ND-3/4 and D-loop) in Goby from three geographical areas in the Bengula ecosystem.
Haplotype Ludritz Walvis Bay North of Ambrose N = 94 N = 81 Bay N = 64 n= H 1 29 ( 0.30) 25 (0.30) 33 (0.52)
H 2 11 (0.11) 10 (0.12) 9 (0.14)
H 3 2 (0.02) 8 (0.09) 6 (0.09)
H 4 5 (0.05) 5 (0.06) 3 (0.04)
H 5 13 (0.13) 7 (0.08) 0 (0)
H 6 17 (0.18) 5 (0.06) 3 (0.04)
H 7 1 (0.01) 3 (0.03) 1 (0.01)
H 8 3 (0.03) 2 (0.02) 0 (0)
H 9 1 (0.01) 0 (0) 1 (0.01)
H 10 1 (0.01) 0 (0) 0 (0)
H 11 4 (0.04) 1 (0.01) 1 (0.01)
H 12 6 (0.06) 2 (0.02) 1 (0.01)
H 13 1 (0.01) 0 (0) 0 (0)
H 14 0 (0) 1 (0.01) 0 (0)
H 15 0 (0) 2 (0.02) 0 (0)
H 16 0 (0) 4 (0.04) 0 (0)
H 17 0 (0) 1 (0.01) 0 (0) Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 H 18 0 (0) 2 (0.02) 0 (0)
H 19 0 (0) 1 (0.01) 0 (0)
H 20 0 (0) 2 (0.02) 0 (0)
H 21 0 (0) 0 (0) 1 (0.01)
H 22 0 (0) 0 (0) 1 (0.01)
H 23 0 (0) 0 (0) 1 (0.01)
H 24 0 (0) 0 (0) 1 (0.01)
H 25 0 (0) 0 (0) 1 (0.01)
18 3.3. Unique haplotypes
Unique haplotypes were detected in all three geographical locations. The largest number of unique haplotypes (n= 7) was observed in Walvis Bay. There were 5 composite haplotypes unique to North Ambrose Bay, but with relatively low frequencies. Only two haplotypes H 10 (AAAAC) and H 14 (BACBA) were restricted to Ludritz area and recorded only once in the study. But lager sample sizes would be necessary to determine whether they are truly geographically restricted or whether they occur at low frequencies in other areas as well.
3.4. Haplotype and nucleotide diversities
The software Arlequin 2.0(Schneider et al 2000) was used to measure the genetic variation with in the areas as determined by haplotype diversity, mean number of average pairwise differences and nucleotide diversities. These are presented in Table 7.
Table 7. Descriptions of the haplotype distribution, haplotype diversity, mean number of pairwise differences and nucleotide diversity within the three geographical areas of Sufflogobius bibarbatus.
Ludritz Walvis Bay North Ambrose Bay Number of haplotypes 13 17 14 Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 Number of shared haplotypes 8 8 8
Number of unique haplotypes 2 7 5
Haplotype diversity (h ± SE) 0.8376 ± 0.0022 0.8679 ± 0.0261 0.7005 ± 0.0575
Mean number of pairwise 1.7501 ± 1.0259 2.1722 ± 1.2166 1.3865 ± 0.8542 differences Nucleotide diversity ( π ± SE) 0.0514 ±0.0334 0.0638 ± 0.0396 0.0407 ± 0.0281
19
The haplotype diversities (h) with in each area ranged from 0.8679 to 0.7005 and nucleotide diversity (π) from 0.0638 to 0.0407. The highest haplotype diversity was observed in the Walvis Bay area ((h = 0.8679), followed by Ludritz (h = 0.8376) which indicates a similar value to the overall haplotype diversity (h= 0.8247). The lowest haplotype diversity was found in the northern area, North Ambrose Bay. The highest nucleotide diversity was found in Walvis Bay (π = 0.0638) behind Ludritz (π = 0.0514) and North Ambrose Bay (π = 0.0407) respectively. Similar pattern was observed for mean number of pairwise differences.
Analysis of molecular variance among and within areas is given in Table 8. The highest variation was recorded within the areas (95.17%). The variance among the three geographical areas was recorded as 4.83 per cent.
Table 8 . Analysis of molecular variance among and within geographical areas in the Bengula ecosystem.
Source of variation D.F Sum of Variance Percentage of squares components variation Among areas 2 8.946 0.0456 4.83
Within areas 235 211.256 0.8989 95.17
Total 237 220.202 .09446
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 The pairwise genetic distances show that all three geographical areas are significantly different from each other (P<0.05), also after application of sequencial Bonferroni
correction of significane level. The highest FST indicates the highest difference between Ludritz and North Ambrose Bay. Also it indicates that the unique haplotypes found in less numbers in Walvis Bay and North Ambrose Bay contribute significantly to the pairwise genetic distances which allow to differentiate the geographical areas (Table 9).
20 Table 9. Comparisons of pairs of populations in studied areas by distance method. Below
diagonal shows the population pairwise FST values and above diagonal are probability
values of the FST being significantly different from zero.
Geographical Area Ludritz Walvis Bay North Ambrose Bay
Ludritz _ 0.018 ∗ 0.000 ∗
Walvis Bay 0.02592 _ 0.000 ∗
North Ambrose Bay 0.07517 0.05417 _
3.5. Exact test and Chi square test
Despite the cases of haplotype sharing described, analysis of geographical heterogeneity in haplotype frequencies by using Monte Carlo simulation and exact test shown that the distribution of haplotypes significantly differed among the three geographical localities. The exact test population differentiation had shown an existence of global differentiation at P< 0.001 after the 6000 Markov steps. Also pairwise exact P values indicated that the populations in Ludritz and Walvis Bay (P~0.000) and populations in Ludritz and North Ambrose Bay (P = 0.01) were significantly different from each other. The significance level was set to 0.05 after 10000 steps of Markov chain length. But the difference between populations in Walvis Bay and North Ambrose Bay were not significant,
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 although quite high (P = 0.06).
Furthermore the use of Monte Carlo test has shown similar results to the test described above. By this method significant difference can be seen among the three geographical areas (χ2 =24.27, p= 0.02), showing geographical heterogeneity in the pooled sample. Furthermore significant differences could be identified in two of the three pairwise comparisons done. The haplotype frequencies between Ludritz and Walvis Bay (χ2 = 13.57, P= 0.04) as well as Ludritz and North Ambrose Bay (χ2 = 15.97, P = 0.01) were significantly different. But haplotype frequencies between Walvis Bay and North
21 Ambrose Bay (χ2 = 3.74, P = 0.79) did not show any statistical significances at the 0.05 level after the sequential Bonferronni correction (Table 10).
Table 10. Chi-square (above diagonal) and P – values (below diagonal) of the pair wise haplotype frequency comparisons based on χ2 statistics using the Monte Carlo randomization method. (∗) Denotes statistical significance at 0.05 and (ns) denotes non significance
Ludritz Walvis Bay North Ambrose Bay
Ludritz - 13.57 15.97
Walvis Bay P = 0.04 ∗ - 3.74
North Ambrose Bay P = 0.01 ∗ P = 0.79 ns -
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
22 4. Discussion
4.1. General overview
The population structure of the marine fish species is vital to understand. Their novel genetic, physiological, morphological and behavioural characters can promote distinguish differences in life history traits. The observed differences in life history traits such as growth rates, fecundity, abundance and spatial distribution contribute to their long term adaptability, existence, and resistance to anthropogenic and environmental perturbations. In fact the investigation of morphological, behavioural, ecological and biogeography of the marine fish species gives relevant information on population structure, and the more important factors such as gene flow and deviation from genetic equilibrium as well as natural selection should also be addressed. So the population genetics of a species is vital to make inferences about relationship between micro evolutionary forces and its interrelationship of the species. The genetic variation provides raw materials for natural selection. Also it indicates the potential for local adaptation or speciation with time (Bohonak, 1999).
Population genetics of most commercially important marine fish species has been performed (Bremer, 1998; Mork and Giaever 1999; Tudela et al, 1999; Nesbo et al, 2000; McPherson et al, 2001; Perrin and Borsa 2001; Appleyard et al, 2002; McPherson et al, 2004). This is mostly due to its validity in fisheries management and to a broader extent the conservation aspects. In a transition period, when the management tends to turn towards ecosystem based methods, it is vital to understand population genetics of
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 ecologically important species as well. Although Sufflogobius bibarbatus( pelagic goby) has a minor commercial value at the moment, its role in the Benguela ecosystem can not be neglected. Crawford (1987) has shown that pelagic gobies are consumed by most piscivorus fish species, marine mammals and sea birds. So it is evident that pelagic goby is a key prey species in the Benguela ecosystem. In spite of its role in the ecosystem, this species has been poorly studied. The present study is the first one which evaluates genetic variability and differentiation of pelagic goby by PCR- RFLP analysis containing mitochondrial DNA. This kind of study ultimately will provide important information on population at molecular level, and should in turn be highlighted by other biological studies.
23
4.2. Selection of methods
A wide range of molecular techniques are now available to answer the questions in population genetics. The long established allozymes still retain the choice of marker in case of cost associated with larger sample sizes (Ward and Grewe, 1995). But the more resolution of the DNA based methods has resulted in a replacement for allozymes. This is mainly due to the development of Polymerase Chain Reaction (PCR) (Saiki et al, 1988). With this development; the DNA-based methods have provided new tools for population genetic studies. DNA analysis provides better resolution of genetic variation, and Mitochondrial DNA (mtDNA) analysis has proved to be ideal for revealing DNA variation in natural populations. This is mostly due to its unique characters. MtDNA is being subjected to high mutation rates and it can produce intraspecific divergence in relatively short time (Avise et al, 1987). Maternal transmission and the absence of recombination can result in highly detectable geographical sub divisions of natural populations. The evolutionary rates as well as the genetic differentiation of mtDNA among populations are thought to be approximately four times higher than that exhibit by nuclear genes (Birky et al, 1983, 1989; Avise, 1994). Moreover it has been shown that mtDNA evolves quickly; at an average rate of 1-2 % nucleotide substitutions per million years. This is 5-10 times higher than that of nuclear genes (Brown et al, 1983).
These unique characters contribute to the fact that the majority of population comparisons in recent years have been mtDNA and restriction fragment length polymorphism (RFLP)
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 of mtDNA molecule, and this technique has been widely applied to marine species (Ovenden,1990). It has proved to be a useful technique in population analysis (Cronin et al, 1993, 1994; Bartlett and Davidson, 1995; Hall and Nawrocki, 1995; O’Connell et al, 1995). Also RFLP appear to be the most popular analytic marker technique among varieties of molecular techniques. According to Silva & Russo (2000), it has accounted for 44% of the scientific papers published in population biology, followed by DNA sequencing (18 %) and RAPD (14 %). In the present study this technique was chosen because it was believed that PCR- RFLP analysis of mtDNA was a very suitable molecular tool for studying population genetics of pelagic goby. Its high resolution and cost effectiveness are ideal for this kind of study. The PCR- RFLP analysis of mtDNA
24 was concentrated on the control region (D- Loop) and ND 3/ 4. Early attempts of mtDNA analysis were to examine the whole region. But now it is documented that PCR- RFLP analysis of separate regions may give better resolution and information (Cronin et al, 1993; Merker and Woodruff, 1996).
The D-Loop is a unique, non coding segment of mitochondrial DNA which is indicated to be the most variable portion (Mortiz et al, 1987). Detailed studies in mammalian mtDNA variation have confirmed this variability. This is due to the varying numbers of tandem repeat sequences present in the control region, and these are often targeted for frequent mutations; especially deletions, insertions and duplications (Brown, 1983). Zischler (2000) indicates that higher evolutionary rate of the region makes it a suitable marker for resolving recent divergences in populations. It is also proved that none coding D- Loop gives better information on differentiation than the other regions (Szalanski et al, 2000).
The NADH dehydrogenase (ND) complex gene also has proved to be more informative than other segments of mtDNA. In the studies carried out by McCusker et al (2000) and Nilsson et al (2001) on ND 1, Williams et al (1997) and Paragamian et al (1999) on ND 2 and Pigeon et al, (1998), Paragamian et al, (1999) on ND 5/6 has been detected considerable variation in this gene region. Merker and Woodruff (1996) as well as Nielsen et al (1998) have shown that significant variation could be found in the ND 3/4 gene region. This study also suggests that combination of the D- Loop and ND 3/4, both highly variable, gives a clear distinctive tool for population differentiation. Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
4.3. Selection of material and sampling sites
This study has shown that significant population differentiation occurs in pelagic goby population in Northern Benguela ecosystem. The northern Benguela extends from 150 S to 290 S (Shannon & Jarre-Teichmann, 1999), and the pelagic goby is mainly found between 220 S and 270 S (Crawford, 1987). But Cruickshank (1980) suggested that the distribution could be extended to more northern locations as well. In population study it is essential that sampling should represent the natural population as much as possible.
25 The existing information on distribution of pelagic goby is the basis for chose of the sampling locations. The most northern population is represented by North of Ambrose Bay (S 200 01’ - S 210 05’) and the southern distribution by the samples taken around Ludritz (S 250 43’- S 270 23’). The sampling locations around Walvis Bay(S 220 20’ - S 230 30’) represented the intermediate population between north and south. The critical assumption of the strategy of sampling is representative of the natural population. In population genetics, the genetic difference detected is often only slightly higher than that between individuals in that population. So the number of individuals representing the northern and southern regions should be relatively high. Silva & Russo (2000) noted that sample size over 30 individuals (>30) would be ideal to make a relatively accurate assessment. This study reveals that significant population differentiation exists along the north-south axis in the Benguela ecosystem. The relatively large size of sampling North of Ambrose Bay (n = 64), Walvis Bay (n = 81) and Ludritz (n = 94) clearly contributes to the significant results, shown in the study.
4.4. Variation within areas
Population genetic structure among and within geographical areas of pelagic goby could be suggested as observed differences in level of haplotype (h) and nucleotide diversities (π). Average haplotype diversity (0.82 ± 0.01) and average nucleotide diversity (0.05 ± 0.03) were quite high. This higher diversity might be due to the higher variability of the control region. This is further confirmed by analysis of molecular variance which has resulted in 95% variation among the whole sampled population in the region. The highest Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 haplotype diversity (0.86 ± 0.02) was found in the central region around Walvis Bay. This trend retains the same for highest number of haplotypes (17) in this region. Due to these facts nucleotide diversity was also recorded as the highest. It is interesting to look into these facts to try to explain the causes of the highest values. This might possibly be related to the main reproductive centre of the species. Brykov (2004) has shown that high haplotype diversity and nucleotide diversity occurred around the reproductive centres of walleye pollock, Theragra chdcograma. Similar observation might be relevant to the pelagic goby population in the Northern Benguela. O’Toole (1978) suggested that spawning of the pelagic goby take place along the entire coast of Namibia and the most
26 intensive spawning occurs in coastal waters around Walvis Bay. This is the same for the most of the fish species living in the Benguela ecosystem (O’Toole, 1977; Shannon and Pillar, 1986). So this study also suggests that high haplotype diversities and nucleotide diversities could be found in the reproductive centres.
The haplotype diversity in mtDNA found in fish species, for the whole molecule, has been well documented. This includes freshwater fish species like brook char (Danzmann and Ihseen, 1995), walleye (Ward et al, 1989) and anadromus species, brown trout (Hansen et al, 1995) and Atlantic salmon (O’connel et al, 1995). The haplotype diversity found in marine fish species has generally been reported to be comparatively lower, for example for species such as mummichog, Fundulus heteroclitus (Brown and Chapman, 1991) and for Atlantic cod (Carr et al, 1995). Higher diversities have been found in Atlantic herring (Kornfield and Bogdanowicz, 1987), walleye pollock (Mulligan et al, 1992) and yellowfin tuna (Scoles and Graves, 1993). But significant constraint exists when comparing these results with this study because here only part of mtDNA genome, D-loop and ND 3/ 4 region, is dealt with. Although the present study only dealt with part of the genome it is evident that relatively high haplotype diversity was found.
MtDNA haplotype diversity is considered as equivalent to heterozygosity and protein polymorphism (Nei, 1987). Some studies of population structure of the fish species has been done in the Benguela region. Early studies on Southern African Angler fish, Lophius vomerinus (Leslie and Grant, 1990) and on lantern fish, Lampanyctodes hectoris (Florence et al, 2002) did not reveal any genetic differentiation in the natural populations Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 in this region. This is in contrast to the higher haplotype diversities found in this study. Even a recent study on allozyme variation of pelagic goby has resulted in rather high heterozygosities which might indicate genetically diverse pelagic goby population in the Northern Benguela.
4.5. Population studies
This study revealed that the pelagic goby population in the Northern Benguela consisted of a few common haplotypes along with substantially larger number of rare haplotypes. Eight out of 25 haplotypes (H1, H2, H3, H4, H6, H7, H11 and H12) were shared among
27 the three geographical areas, and all the remaining haplotypes were only shared among two regions or unique for one particular region. Similar results have been obtained for most fish species studied so far. Billington and Hebert (1991) considered them as mutational derivatives of the common haplotype. High genetic similarities among haplotypes suggest a recent deviation from a common ancestral haplotype, and divergence of the haplotypes might indicate a deviation for longer time period (Billington, 2003). Similar haplotype diversities found in Ludritz and Walvis Bay might suggest a possible recent deviation. This observation is quite contradictory compared to the population farthest to the north, with haplotype diversities different from Ludritz and Walvis Bay. This observation might lead to the hypothesis that the Northern population has deviated from the southern population at an early time. Increased intensity of sampling of the genome may give better resolution of this hypothesis. The haplotype diversities found in the analysis are functions of both number of fish examined and the intensity of the sampling of the genome. Hence the number of haplotypes and haplotype diversities could be given only a relative assessment of the differentiation in the studied population. Further increased sample sizes and sampling genome may reveal variation that was not found in the present study.
4.6. Genetic deviation through haplotype frequencies
The scale of the marine fish dispersal is very much greater than terrestrial and freshwater organisms (Avise, 1987). It is very prominent that passive drift of larvae in ocean currents and active migration of the adults are the main reasons for higher dispersal Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 ability at some stage of their life cycle. This high dispersal ability has often found to be associated with small genetic differentiation over vast geographical areas (Ward et al, 1994; Graves, 1998; Waples, 1998). The small divergence is more characteristic in species with high fecundity and relatively large population size. Long distance dispersal potential of eggs, larvae, and adults also contributes significantly to this fact (Palumbi, 1994). In spite of dispersal ability, population structuring may occur, and genetic differentiation could be the result, also in marine species (Palumbi, 1994; Waples, 1998; Hoelzel, 1998).
28 The RFLP data produced by amplified mtDNA fragments, D-loop and ND 3/4 revealed clear evidence of significant differentiation between the studied geographical areas in Northern Benguela. The population pairwise genetic distances between populations indicate that all three geographical areas are significantly different from each other. It is a clear indication of existing heterogeneity in pelagic goby population in the Nothern
Benguela ecosystem. But pairwaise Fst only can be used as short term genetic distances between populations (Reynolds et al, 1983; Slatkin, 1995). Hence this study is more concentrated on divergence based on haplotype frequencies which is more illustrative of microgeographical variation.
This divergence is based on haplotype frequency distribution tested by exact test of population differentiation and more reliable Monte Carlo test. Each statistical test has its own statistical power of testing the hypothesis. Exact test of population differentiation is based on the random distribution of different haplotypes among various populations (Raymond and Rousset, 1995). Problems associated with small numbers in these test are avoided in the Monte Carlo test. Both tests show that the southern area (Ludritz) is significantly different from Northern region (North of Ambrose Bay). Also there is a significant difference in haplotype distribution between the Southern region and the central region (Walvis Bay). But no significant difference was observed between the central region (Walvis Bay) and the Northern region (North of Ambrose Bay). So this study shows that significant population subdivision might occur in the pelagic goby population in Nothern Benguela ecosystem. The characteristic feature of the structure is that, the southern population is differing from both the central and northern populations. Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
4.7. Contributing factors
It is interesting to see the possible factors that contribute to the observed two possibilities, which occurred in the pelagic goby population. There is no evidence for micro geographical heterogeneity in the northern and central regions, based on haplotype frequencies. Our data support the hypothesis of possible gene flow might occur between these two areas. High gene flow among the populations is evident for most of the fish populations studied in the marine environment. Gene flow from different locations could
29 result mixing of haplotypes seen in population that might result in a homogeneous population. In this circulation pattern and at any stage that subjected to dispersal could play a major role in this observation. The spawning of pelagic goby takes place in spring to early summer (September to January) in coastal waters around South of Walvis Bay. It has been reported that larvae, juveniles and adults occupy different layers in the water column (Le Clus and Melo, unpublished data). O’Toole (1978) reported that the icthyoplankton collections of Namibia mainly consist of larvae of pelagic goby and it account for 61 % of the total collection. The larval stages are distributed from Walvis Bay and northwards. In this dispersal, oceanic circulations play an important role.
The main oceanic current in the ecosystem is the Benguela current which has its primary source from the South Atlantic Current (Figure 2). Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
Figure 2. Overview of surface currents in the SE Atlantic Ocean and core positions. AC= Angola Current, ABF = Angola –Benguela Front, BOC = Benguela Oceanic Current, BCC = Benguela Coastal Current, SAC = South Atlantic Current, AGC = Agulhas Current, Striped area = coastal upwelling, Shaded area = filaments of upwelled water (Lutjeharms and Stockton, 1987; West et al, 2004)
30
The South Atlantic Current is an eastern boundary current flowing towards north and north-west. The Bengula current split into Benguela Coastal Current (BCC) and Benguela Oceanic Current (BOC) near south of Walvis ridge. Benguela coastal current is moving towards north (Peterson and Stramma, 1991; West et al, 2004). This circulation pattern and presence of pelagic larval stage of the goby might induce the possibilities of potential dispersal towards north. It is suggested that dispersal might occur from the central region around Walvis Bay to northwards of the studied locations around North of Ambrose Bay. This possibility of mixing might result in a genetically homogeneous population in Walvis Bay and North of Ambrose Bay. The northern population of this study is only represented by the locations around North of Ambrose Bay (20’01’ S). But the pelagic goby population is believed to be distributed also in more northern locations around Cunene River (170 40’S). So the additional sampling further north will be vital to make precise conclusions on the northern and central populations.
Palumbi (1994) noted mechanisms that result in population subdivision. In spite of the fact that potential dispersal appeared to be large in most marine species, the actual dispersal appears to be less. These mechanisms include invisible barriers such as current patterns, oceanic circulation and diffusion effects where density of dispersing in different life stages decrease with increasing distance. Furthermore behavioural characters and selection may also play an important role. These mechanisms might lead to varying spatial directional or temporal limits to dispersal resulting in populations that are partially isolated over time. The most interesting result of the study is significant differentiation Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 that was found between northern and southern populations. The Monte Carlo test based on haplotype frequencies revealed that considerable differences occur even after the Bonferroni correction.
The extent of geographical differentiation of marine fish species could be variable and negatively correlated with mobility and dispersal ability (Ward et al, 1994).As indicated earlier surface circulation pattern in the Benguela ecosystem and the incidence of larval stages subjected to drift, support the explanation of possible mixing of the population. This is clearly evident in the central and northern populations. In fact based on the
31 haplotype frequencies it is clear that southern population is significantly different from northern population. This might indicate that hypothesised gene flow that can occur in the population is not sufficient enough to make a genetically homogeneous population in spite of circulation pattern and dispersal larval stage. Also considering the fact that mtDNA can be considered as neutral, heterogeneity in haplotype distribution could possibly be linked to restrict drifting of larvae along the north – south axis. This might explain the existence of different populations in the northern and southern areas, which could be identified as genetically distinct groups. This might be due to the surface circulation pattern, which might not be strong enough to mix the two different populations. Especially the central region around Walvis Bay is believed to be reproductive centre for most fish species in the Benguela. In addition to pelagic goby (O’Toole, 1978), sardine, Sardinops sagax, around Walvis Bay (O’Toole, 1977), anchovy, Engraulis capensis, around North of Walvis Bay (Shannon and Pillar, 1986) and hake, Merluccius capensis (Kainge, 2002) has been reported. The important fact is that the area around Walvis Bay is supposed to be acting as a retention centre in the ecosystem. Stenevik (2003) has found that the highest retention of sardine eggs could be seen in Walvis Bay. This might suggest that current flow towards northern areas might be getting weaker in relation to geographical distance. If the Benguela current is strong enough to carry sardine eggs towards north, highest retention could not have been noticed in Walvis Bay. Also there is a possibility that density of dispersal stages decrease with increasing distance. This diffusion effect might result in unsuccessful transport over large distances which possibly affecting poor interchange of the population in north and sound. Surface current as an isolation factor is described by Kojima (1997). But in a system Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 where circulation pattern is so complex and changes due to seasonal and annual variations, it is not possible this study to provide strong explanation of the observed differentiation purely due to the current pattern.
The possible existence of hydrographical and environmental barrier might be the reason for separating the southern and northern populations. Previous studies by Shannon, (1985), Shannon and Pillar (1986) and Crawford et al (1987) have shown existence of environmental barrier around Ludritz, which separate the northern and southern parts of the system. This is defined by distribution of shoals, commercial catches, and larvae of
32 pelagic fish species in the Benguela ecosystem (Agenbag, 1980; Cruickshank, 1983; LeClus, 1985). This is further highlighted by Agenbag and Shannon (1988). They suggested that a possible biological discontinuity may be found at Meob Bay (240 30’S) in the Benguela ecosystem. It is believed that a combination of changes in circulation, turbulence and stratification might play important part in creating this discontinuity. The present study also suggests that the biological discontinuity that exists between the area that separates the southern area from the central and northern areas, might result in a population sub division in the pelagic goby population.
A study of age and growth of Lithognathus auretis in Namibian waters by Holtzhausen and Kirchner (2001) revealed geographical differences between north and south and suggested that differences in environmental conditions and biochemical genetic variation as the possible reasons for the observed differences. More supportive suggestions have been revealed by the same authors (Holtzhausen and Kirchner, 2004). A population study on west coast steenbars, Lithognathus aureti, has been done on the basis of similar locations defined in the present study; mainly concerned with northern, central and southern regions of the northern Benguela. This study has shown that distinct differences in growth rate, otolith structure, morphology, size at maturity, sex ratios and length at age between northern and southern populations. Also this has been supported by early genetic study based on electrophoretic analysis, where the two populations showed significant genotype differentiation at two loci, indicating that effective barriers could exist to isolate them (Van der Bank and Holtzhausen, 1998, 1999). The distribution of the haplotype frequencies in the present study also suggests that a population sub division exist due to Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 the physical barrier in the southern region, especially near Meob Bay where this act as a barrier to exchange of the pelagic goby population.
33
5. Conclusions
From the present study it can be concluded that considerable population sub division exist in the pelagic goby population in the Northern Benguela. No unambiguous differences between the populations in Walvis Bay and North of Ambrose Bay were found based upon the RFLP data from the studied mtDNA. The circulation pattern and dispersal capability of pelagic life stage might be the possible reasons. In contrast the Ludritz population was found to be significantly different from populations in Walvis Bay and North of Ambrose Bay, suggesting an adaptational barrier for possible isolation or partial isolation. However, it is not clear which factor contributes most for the observed divergence. But it is obvious that combination of factors like complex circulation pattern and well defined physical barrier at 240 30’, together with adaptive selection may play important roles. It is expected that current findings will strengthen the understanding of the pelagic goby population and will provide scientific insight in decision and policy making of the ecosystem based management in the Benguela ecosystem. Further extensive studies involving more contiguous population units of the species within its distribution range in the Benguela ecosystem as a whole may provide much thorough understanding of the population structure of pelagic goby.
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
34
6. References
Agenbag, J.J. 1980. General distribution of pelagic fish off South West Africa as deduced from Aerial fish spotting (1971-1974 and 1977) and as influenced by hydrology. Fish Bulletin South Africa 13: 55-67.
Agenbag, J.J. & L.V. Shannon. 1988. A Suggested Physical Explanation for the Existence of a Biological Boundary at 24-Degrees-30's in the Benguela System. South African Journal of Marine Science-Suid-Afrikaanse Tydskrif Vir Seewetenskap 6: 119-132.
Appleyard, S.A., R.D. Ward & P.M. Grewe. 2002. Genetic stock structure of bigeye tuna in the Indian Ocean using mitochondrial DNA and microsatellites. Journal of Fish Biology 60: 767-770.
Arnaud, S., F. Bonhomme & P. Borsa. 1999. Mitochondrial DNA analysis of the genetic relationships among populations of scad mackerel (Decapterus macarellus, D- macrosoma, and D-russelli) in South-East Asia. Marine Biology 135: 699-707.
Avise, J.C. 1994. Molecular Markers, Natural History and Evolution. Chapman & Hall, New York. 511-512 pp.
Avise, J.C., J. Arnold, R.M. Ball, E. Bermingham, T. Lamb, J.E. Neigel, C.A. Reeb & N.C. Saunders. 1987. Intraspecific Phylogeography - the Mitochondrial-DNA Bridge between Population-Genetics and Systematics. Annual Review of Ecology and Systematics 18: 489-522.
Avise, J.C., W.S. Nelson & C.G. Sibley. 1994. DNA-Sequence Support for a Close Phylogenetic Relationship between Some Storks and New-World Vultures. Proceedings of the National Academy of Sciences of the United States of America 91: 5173-5177.
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 Bagley, M.J. & G.A.E. Gall. 1998. Mitochondrial and nuclear DNA sequence variability among populations of rainbow trout (Oncorhynchus mykiss). Molecular Ecology 7: 945-961.
Bartlett, S.E. & W.S. Davidson. 1995. Genetic analysis of Salmo using mtDNA AMPFLPs. Aquaculture 137: 47-48.
Bernatchez, L. & R.G. Danzmann. 1993. Congruence in Control-Region Sequence and Restriction-Site Variation in Mitochondrial-DNA of Brook Charr (Salvelinus- Fontinalis-Mitchill). Molecular Biology and Evolution 10: 1002-1014.
35 Billington, N. 2003. Mitochondrial DNA. pp. 59 -87. In: E.M. Hallerman (ed.) Population genetics:principles and applications for fisheries scientists, American Fisheries Society, Bethesda, Maryland.
Billington, N. & P.D.N. Hebert. 1991. Mitochondrial-DNA Diversity in Fishes and Its Implications for Introductions. Canadian Journal of Fisheries and Aquatic Sciences 48: 80-94.
Binachi, G., K.E Carpenter, J.P Roux, F.J Molly and H. Boyer. 1999. Field guide to the living marine resources of Namibia,. FAO,Rome.
Birky, C.W., P. Fuerst & T. Maruyama. 1989. Organelle Gene Diversity under Migration, Mutation, and Drift - Equilibrium Expectations, Approach to Equilibrium, Effects of Heteroplasmic Cells, and Comparison to Nuclear Genes. Genetics 121: 613- 627.
Birky, C.W., T. Maruyama & P. Fuerst. 1983. An Approach to Population and Evolutionary Genetic Theory for Genes in Mitochondria and Chloroplasts, and Some Results. Genetics 103: 513-527.
Bohonak, A.J. 1999. Dispersal, gene flow, and population structure. Quarterly Review of Biology 74: 21-45.
Bremer JRA, S.B., Robertson NW, Ely B. 1998. Genetic evidence for inter-oceanic subdivision of bigeye tuna (Thunnus obesus) populations. Marine Biology 132: 547-557.
Brown, B.L. & R.W. Chapman. 1991. Gene Flow and Mitochondrial-DNA Variation in the Killifish, Fundulus-Heteroclitus. Evolution 45: 1147-1161.
Brown, W.M. 1983. Evolution of animal mitochondrial DNAs. pp. 62-88. In: M.N.a.R.K. Koehn (ed.) Evolution of genes and protiens, Sinauer, Sunderland, Massachusetts.
Brykov, V.A., N.E. Polyakova, T.F. Priima & O.N. Katugin. 2004. Mitochondrial DNA Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 variation in northwestern Bering Sea walleye pollock, Theragra chalcogramma (Pallas). Environmental Biology of Fishes 69: 167-175.
Carr, S.M., A.J. Snellen, K.A. Howse & J.S. Wroblewski. 1995. Mitochondrial-DNA Sequence Variation and Genetic Stock Structure of Atlantic Cod (Gadus-Morhua) from Bay and Offshore Locations on the Newfoundland Continental-Shelf. Molecular Ecology 4: 79-88.
Chubb, A.L., R.M. Zink & J.M. Fitzsimons. 1998. Patterns of mtDNA variation in Hawaiian freshwater fishes: The phylogeographic consequences of amphidromy. Journal of Heredity 89: 8-16.
36 Crawford, R.J.M., L.V. Shannon & D.E. Pollock. 1987. The Benguela Ecosystem .4. The Major Fish and Invertebrate Resources. Oceanography and Marine Biology 25: 353-505.
Cronin, M.A., S. Hills, E.W. Born & J.C. Patton. 1994. Mitochondrial-DNA Variation in Atlantic and Pacific Walruses. Canadian Journal of Zoology-Revue Canadienne De Zoologie 72: 1035-1043.
Cronin, M.A., W.J. Spearman, R.L. Wilmot, J.C. Patton & J.W. Bickham. 1993. Mitochondrial-DNA Variation in Chinook (Oncorhynchus-Tshawytscha) and Chum Salmon (O Keta) Detected by Restriction Enzyme Analysis of Polymerase Chain-Reaction (Pcr) Products. Canadian Journal of Fisheries and Aquatic Sciences 50: 708-715.
Cruickshank, R.A. 1980. Extension to the geographical distribution of pelagic goby Sufflogobius bibarbatus off South West Africa and some mensural and energetic information. Fish bullatin South Africa 13: 77-82.
Cruickshank, R.A. 1983. Ecology of Pilchard and Anchovy Shoals Off Namibia. South African Journal of Science 79: 147-149.
Cushing, D.H. 1971. Upwelling and Production on Fish. Advances in Marine Biology 9: 255.
Danzmann, R.G. & P.E. Ihssen. 1995. A phylogeographic survey of brook charr (Salvelinus fontinalis) in Algonquin Park, Ontario based upon mitochondrial DNA variation. Molecular Ecology 4: 681-697.
Excoffier, L., P.E. Smouse & J.M. Quattro. 1992. Analysis of Molecular Variance Inferred from Metric Distances among DNA Haplotypes - Application to Human Mitochondrial-DNA Restriction Data. Genetics 131: 479-491.
Florence, W.K., R.A. Hulley, B.A. Stewart & M.J. Gibbons. 2002. Genetic and morphological variation of the lanternfish Lampanyctodes hectoris Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 (Myctophiformes : Myctophidae) off southern Africa. South African Journal of Marine Science-Suid-Afrikaanse Tydskrif Vir Seewetenskap 24: 193-203.
Gharrett, A.J., A.K. Gray & J. Heifetz. 2001. Identification of rockfish (Sebastes spp.) by restriction site analysis of the mitochondrial ND-3/ND-4 and 12S/16S rRNA gene regions. Fishery Bulletin 99: 49-62.
Gibbons, M.J., A.J.J. Goosen & P.A. Wickens. 2002. Habitat use by demersal nekton on the continental shelf in the Benguela ecosystem, southern Africa. Fishery Bulletin 100: 475-490.
Gibbons, M.J., A. Sulaiman, K. Hissman, J. Schauer, I. McMillan & P.A. Wickens. 2000. Video observations on the habitat association of demersal nekton in the midshelf
37 benthic environment off the Orange River, Namibia. South African Journal of Marine Science-Suid-Afrikaanse Tydskrif Vir Seewetenskap 22: 1-7.
Grant, W.S., R.W. Leslie & I.I. Becker. 1987. Genetic Stock Structure of the Southern African Hakes Merluccius-Capensis and Merluccius-Paradoxus. Marine Ecology- Progress Series 41: 9-20.
Graves, J.E. 1998. Molecular insights into the population structures of cosmopolitan marine fishes. Journal of Heredity 89: 427-437.
Hall, H.J. & L.W. Nawrocki. 1995. A Rapid Method for Detecting Mitochondrial-DNA Variation in the Brown Trout, Salmo-Trutta. Journal of Fish Biology 46: 360-364.
Hansen, M.M., R.A. Hynes, V. Loeschcke & G. Rasmussen. 1995. Assessment of the Stocked or Wild Origin of Anadromous Brown Trout (Salmo-Trutta L) in a Danish River System, Using Mitochondrial-DNA Rflp Analysis. Molecular Ecology 4: 189-198.
Hewitson, J.D. & R.A. Cruickshank. 1993. Production and Consumption by Planktivorous Fish in the Northern Benguela Ecosystem in the 1980s. South African Journal of Marine Science-Suid-Afrikaanse Tydskrif Vir Seewetenskap 13: 15-24.
Hoelzel, A.R. 1998. Genetic structure of cetacean populations in sympatry, parapatry, and mixed assemblages: Implications for conservation policy. Journal of Heredity 89: 451-458.
Holtzhausen, J.A. & C.H. Kirchner. 2001. Age and growth of two populations of West Coast steenbras Lithognathus aureti in Namibian waters, based on otolith readings and mark-recapture data. South African Journal of Marine Science-Suid- Afrikaanse Tydskrif Vir Seewetenskap 23: 169-179.
Holtzhausen, J.A.A.K., C .H. 2004. Management regulations of Namibian angling fish species. In: U.R. Sumalia, Boyer, D., Skogen, M.D., Steinshamn, S.I (ed.) Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 Namibia's Fisheries, Ecological, economic and social aspects, Eburon Academic Publishers, CW Delft, The Netherlands.
Jarre-Teichmann, A., L.J. Shannon, C.L. Moloney & P.A. Wickens. 1998. Comparing trophic flows in the southern Benguela to those in other upwelling ecosystems. South African Journal of Marine Science-Suid-Afrikaanse Tydskrif Vir Seewetenskap 19: 391-414.
Kainge, P.I. 2002. Spawning time and reproductive investment of the hakes, Merluccius capensis and Merluccius paradoxus in the Namibian waters. M.Phill, University of Bergen, Bergen.
38 Kojima, S., R. Segawa & I. Hayashi. 1997. Genetic differentiation among populations of the Japanese turban shell Turbo (Batillus) cornutus corresponding to warm currents. Marine Ecology-Progress Series 150: 149-155.
Kornfield, I. & S.M. Bogdanowicz. 1987. Differentiation of Mitochondrial-DNA in Atlantic Herring, Clupea-Harengus. Fishery Bulletin 85: 561-568.
Le Clus, F. 1985. The spawning of anchovy, Engraulis capensis (Gilchrist) off South West Africa. M.Sc Thesis, University of Port Elizabeth. 202 pp.
Leslie, R.W. & W.S. Grant. 1990. Lack of Congruence between Genetic and Morphological Stock Structure of the Southern African Anglerfish Lophius- Vomerinus. South African Journal of Marine Science-Suid-Afrikaanse Tydskrif Vir Seewetenskap 9: 379-398.
Lutjeharms, J.R.E. & P.L. Stockton. 1987. Kinematics of the Upwelling Front Off Southern-Africa. South African Journal of Marine Science-Suid-Afrikaanse Tydskrif Vir Seewetenskap 5: 35-49.
McCusker, M.R., E. Parkinson & E.B. Taylor. 2000. Mitochondrial DNA variation in rainbow trout (Oncorhynchus mykiss) across its native range: testing biogeographical hypotheses and their relevance to conservation. Molecular Ecology 9: 2089-2108.
McPherson, A.A., P.T. O'Rielly & C.T. Taggart. 2004. Genetic differentiation, temporal stability, and the absence of isolation by distance among Atlantic herring populations. Transactions of the American Fisheries Society 133: 434-446.
McPherson, A.A., R.L. Stephenson, P.T. O'Reilly, M.W. Jones & C.T. Taggart. 2001. Genetic diversity of coastal Northwest Atlantic herring populations: implications for management. Journal of Fish Biology 59: 356-370.
Merker, R.J. & R.C. Woodruff. 1996. Molecular evidence for divergent breeding groups of walleye (Stizostedion vitreum) in tributaries to western Lake Erie. Journal of Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 Great Lakes Research 22: 280-288.
Moran, P., C. Lundy, C. Rico & G.M. Hewitt. 1999. Isolation and characterization of microsatellite loci in European hake, Merlucius merlucius (Merlucidae, Teleostei). Molecular Ecology 8: 1357-1358.
Moritz, C., T.E. Dowling & W.M. Brown. 1987. Evolution of Animal Mitochondrial- DNA - Relevance for Population Biology and Systematics. Annual Review of Ecology and Systematics 18: 269-292.
Mork, J. & M. Giaever. 1999. Genetic structure of cod along the coast of Norway: Results from isozyme studies. Sarsia 84: 157-168.
39 Mulligan, T.J., R.W. Chapman & B.L. Brown. 1992. Mitochondrial-DNA Analysis of Walleye Pollock, Theragra-Chalcogramma, from the Eastern Bering Sea and Shelikof Strait, Gulf of Alaska. Canadian Journal of Fisheries and Aquatic Sciences 49: 319-326.
Nei, M. 1987. Molecular Evolutionary Genetics. Columbia University Press,New York, NY, USA.
Nesbo, C.L., E.K. Rueness, S.A. Iversen, D.W. Skagen & K.S. Jakobsen. 2000. Phylogeography and population history of Atlantic mackerel (Scomber scombrus L.): a genealogical approach reveals genetic structuring among the eastern Atlantic stocks. Proceedings of the Royal Society of London Series B-Biological Sciences 267: 281-292.
Nielsen, E.E., M.M. Hansen & K.L.D. Mensberg. 1998. Improved primer sequences for the mitochondrial ND1, ND3/4 and ND5/6 segments in salmonid fishes: application to RFLP analysis of Atlantic salmon. Journal of Fish Biology 53: 216- 220.
Nilsson, J., R. Gross, T. Asplund, O. Dove, H. Jansson, J. Kelloniemi, K. Kohlmann, A. Loytynoja, E.E. Nielsen, T. Paaver, C.R. Primmer, S. Titov, A. Vasemagi, A. Veselov, T. Ost & J. Lumme. 2001. Matrilinear phylogeography of Atlantic salmon (Salmo salar L.) in Europe and postglacial colonization of the Baltic Sea area. Molecular Ecology 10: 89-102.
Oconnell, M., D.O.F. Skibinski & J.A. Beardmore. 1995. Mitochondrial-DNA and Allozyme Variation in Atlantic Salmon (Salmo-Salar) Populations in Wales. Canadian Journal of Fisheries and Aquatic Sciences 52: 171-178.
O'Toole, M.J. 1977. Investigations into some important fish larvae in the South East Atlantic in relation to the hydrological environment. PhD Thesis, University of Cape Town, South Africa.
O'Toole, M.J. 1978. Development,distribution and relative abundance of the larvae and Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 early juveniles of the pelagic goby Sufflogobius bibarbatus (von Bonde) off South West Africa. Investl Rep.Sea.Brch South Africa 116: 28.
Ovenden, J.R. & R.W.G. White. 1990. Mitochondrial and Allozyme Genetics of Incipient Speciation in a Landlocked Population of Galaxias-Truttaceus (Pisces, Galaxiidae). Genetics 124: 701-716.
Palumbi, S.R. 1994. Genetic-Divergence, Reproductive Isolation, and Marine Speciation. Annual Review of Ecology and Systematics 25: 547-572.
Palumbi, S.R. 2001. The ecology of the marine protected areas. In: S.G. M.Bertness, & M. Hay (ed.) Community Ecology, Sinauer Associates, Sunderland, MA.
40 Paragamian, V.L., M.S. Powell & J.C. Faler. 1999. Mitochondrial DNA analysis of burbot stocks in the Kootenai River Basin of British Columbia, Montana, and Idaho. Transactions of the American Fisheries Society 128: 868-874.
Perrin, C. & P. Borsa. 2001. Mitochondrial DNA analysis of the geographic structure of Indian scad mackerel in the Indo-Malay archipelago. Journal of Fish Biology 59: 1421-1426.
Peterson, R.G. & L. Stramma. 1991. Upper-Level Circulation in the South-Atlantic Ocean. Progress in Oceanography 26: 1-73.
Pigeon, D., J.J. Dodson & L. Bernatchez. 1998. A mtDNA analysis of spatiotemporal distribution of two sympatric larval populations of rainbow smelt (Osmerus mordax) in the St. Lawrence River estuary, Quebec, Canada. Canadian Journal of Fisheries and Aquatic Sciences 55: 1739-1747.
Raymond, M. & F. Rousset. 1995. An exact test for population differentiation. Evolution 49: 1280-1283.
Reynolds, J., B.S. Weir & C.C. Cockerham. 1983. Estimation of the Co-Ancestry Coefficient - Basis for a Short-Term Genetic-Distance. Genetics 105: 767-779.
Rice, W.R. 1989. Analyzing Tables of Statistical Tests. Evolution 43: 223-225.
Roff, D.A. & P. Bentzen. 1989. The Statistical-Analysis of Mitochondrial-DNA Polymorphisms - Chi-2 and the Problem of Small Samples. Molecular Biology and Evolution 6: 539-545.
Ruzzante, D.E., C.T. Taggart, S. Lang & D. Cook. 2000. Mixed-stock analysis of Atlantic cod near the Gulf of St. Lawrence based on microsatellite DNA. Ecological Applications 10: 1090-1109.
Saiki, R.K., D.H. Gelfand, S. Stoffel, S.J. Scharf, R. Higuchi, G.T. Horn, K.B. Mullis & Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 H.A. Erlich. 1988. Primer-Directed Enzymatic Amplification of DNA with a Thermostable DNA-Polymerase. Science 239: 487-491.
Scoles, D.R. & J.E. Graves. 1993. Genetic-Analysis of the Population-Structure of Yellowfin Tuna, Thunnus-Albacares, from the Pacific-Ocean. Fishery Bulletin 91: 690-698.
Schneider, S., D. Roessli & L.Excoffier. 2000. Arlequin ver. 2.000: A Software for population genetic data analysis. Genetics and Biometry Laboratory, University of Geneva, Switzerland.
Shannon, L.J. & A. Jarre-Teichmann. 1999. A model of trophic flows in the northern Benguela upwelling system during the 1980s. South African Journal of Marine Science-Suid-Afrikaanse Tydskrif Vir Seewetenskap 21: 349-366.
41
Shannon, L.V. 1985. The Benguela Ecosystem .1. Evolution of the Benguela, Physical Features and Processes. Oceanography and Marine Biology 23: 105.
Shannon, L.V. & S.C. Pillar. 1986. The Benguela Ecosystem .3. Plankton. Oceanography and Marine Biology 24: 65-170.
Shelton, P.A. 1986. Fish spawning strategies in the variable Southern Benguela current region. PhD, University of Capetown. 237 pp.
Silva, E.P. & C.A.M. Russo. 2000. Techniques and statistical data analysis in molecular population genetics. Hydrobiologia 420: 119-135.
Slatkin, M. 1995. A Measure of Population Subdivision Based on Microsatellite Allele Frequencies. Genetics 139: 457-462.
Sokal, R.R.&.Rohlf, F.J. 1981. Biometry. W.H Freeman and Co, San Francisco,CA.
Stenevik, E.K., M. Skogen, S. Sundby & D. Boyer. 2003. The effect of vertical and horizontal distribution on retention of sardine (Sardinops sagax) larvae in the Northern Benguela - observations and modelling. Fisheries Oceanography 12: 185-200.
Sugaya, T., M. Ikeda & N. Taniguchi. 2002. Relatedness structure estimated by microsatellites DNA and mitochondrial DNA polymerase chain reaction- restriction fragment length polymorphisms analyses in the wild population of kuruma prawn Penaeus japonicus. Fisheries Science 68: 793-802.
Szalanski, A.L., R. Bischof & G. Mestl. 2000. Population genetic structure of Nebraska paddlefish based on mitochondrial DNA variation. Transactions of the American Fisheries Society 129: 1060-1065.
Tudela, S., J.L. Garcia-Marin & C. Pla. 1999. Genetic structure of the European anchovy, Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 Engraulis encrasicolus l., in the north-west Mediterranean. Journal of Experimental Marine Biology and Ecology 234: 95-109.
Utter, F., P.Aebersold & G.Winas. 1987. Interpreting genetic variation detected by electrophoresis. In: N.R.F. Utter (ed.) Population genetics and Fishery management, University of Washinton press, Seattle.
Van Der Bank, F.H.a.H., J. A. 1998/1999. Apreliminary biochemical genetic study of two poulations of Lithgnathus aureti ( Perciformes:Sparidae). South African Journal of Aquatic Science 24: 47-56.
Waples, R.S. 1998. Separating the wheat from the chaff: Patterns of genetic differentiation in high gene flow species. Journal of Heredity 89: 438-450.
42 Ward, R.D., N. Billington & P.D.N. Hebert. 1989. Comparison of Allozyme and Mitochondrial-DNA Variation in Populations of Walleye, Stizostedion Vitreum. Canadian Journal of Fisheries and Aquatic Sciences 46: 2074-2084.
Ward, R.D. & P.M. Grewe. 1994. Appraisal of Molecular-Genetic Techniques in Fisheries. Reviews in Fish Biology and Fisheries 4: 300-325.
Ward, R.D., M. Woodwark & D.O.F. Skibinski. 1994. A Comparison of Genetic Diversity Levels in Marine, Fresh-Water, and Anadromous Fishes. Journal of Fish Biology 44: 213-232.
Ward, R.D.P.M.G. (ed.). 1995. Appraisal of molecular genetic techniques in fisheries. Chapman & Hall,London. 29-54 pp.
West, S., J.H.F. Jansen & J.B. Stuut. 2004. Surface water conditions in the Northern Benguela Region (SE Atlantic) during the last 450 ky reconstructed from assemblages of planktonic foraminifera. Marine Micropaleontology 51: 321-344.
Williams, R.N., R.P Evens., D.K Shiozawa. 1997. Mitochondrial DNA diversity patterns of bull trout in the upper Columbia River basin. pp. 283-297. In: M.K.B. W. C Mackay., M. Monita (ed.) Friends of the bull trout conference proceedings, The Bull TRout Task Force, Calgary, Alberta.
Wilson, A.C., R.L. Cann, S.M. Carr, M. George, U.B. Gyllensten, K.M. Helmbychowski, R.G. Higuchi, S.R. Palumbi, E.M. Prager, R.D. Sage & M. Stoneking. 1985. Mitochondrial-DNA and 2 Perspectives on Evolutionary Genetics. Biological Journal of the Linnean Society 26: 375-400.
Zischler, H. 2000. Nuclear integrations of mitochondrial DNA in primates: Inference of associated mutational events. Electrophoresis 21: 531-536.
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009
43 Appendix
Appendix 1. Matrix of restriction fragments presence/absence for the different patterns found for each endonuclease and for each species in three different regions (1= Ludritz, 2 = Walvis Bay, 3 = North of Ambrose Bay).
Population Individual No Hsp92II Dde I Hae III Rsa I Hsp92II
1 1 11101111 111110 111111100 11110 11110 1 2 11101111 111110 111111100 11111 11110 1 3 11101111 111110 111111100 11111 11110 1 4 11101111 111110 111111100 11110 11110 1 8 11101111 111110 111111100 11110 11110 1 10 11101111 111110 111111100 11110 11110 1 11 11101111 111110 111111100 11110 11111 1 12 11101111 111110 111111100 11111 11111 1 13 11101111 111110 111111100 11110 11110 1 14 11101111 111110 111111100 11111 11110 1 15 11101110 111110 111111100 11111 11110 1 16 11101111 111110 111111100 11111 11111 1 17 11101111 111110 111111100 11110 11110 1 18 11101111 111110 111111100 11110 11110 1 19 11101110 111110 111111100 11110 11110 1 20 11101110 111110 111111100 11110 11111 1 21 11101111 111110 111111100 11110 11111 1 22 11101111 111110 111111100 11110 11110 1 23 11101111 111110 111111100 11110 11110 1 24 11101111 111110 111111100 11110 11110 1 25 11101111 111110 111111100 11110 11110 1 26 11101111 111110 111111100 11110 11110 1 27 11101111 111110 111111100 11110 11110 1 31 11101111 111110 111111100 11111 11110 1 32 11101110 111110 111111100 11110 11110
Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 1 34 11101111 111110 111111100 11110 11110 1 35 11101111 111110 111111100 11111 11111 1 36 11101111 111110 111111100 11111 11110 1 37 11110100 111110 111111100 11111 11110 1 38 11101111 111110 111111100 11110 11110 1 40 11101111 111110 111111100 11111 11111 1 41 11101111 111110 111111100 11110 11110 1 42 11101111 111110 111111100 11110 11110 1 43 11110100 111110 111111100 11111 11110 1 44 11101111 111110 111111100 11111 11110 1 45 11101111 111110 111111100 11110 11110 1 46 11101111 111110 111111100 11111 11111 1 47 11101111 111110 111111100 11111 11110 1 48 11101111 111110 111111100 11110 11110 1 49 11101110 111110 111111100 11111 11110 1 50 11101111 111110 111111100 11110 11110
44 1 51 11101111 111110 111111100 11110 11110 1 52 11101111 111110 111111100 11110 11110 1 53 11101111 111110 111111100 11110 11110 1 54 11101111 111110 111111100 11110 11110 1 55 11101111 111110 111111100 11111 11110 1 56 11101111 111110 111111100 11110 11110 1 57 11101111 111110 111111100 11110 11110 1 58 11101111 111110 111111100 11111 11110 1 59 11101111 111111 111111100 11110 11110 1 60 11101111 111110 111111100 11111 11110 1 61 11101111 111110 111111100 11111 11110 1 62 11101111 111110 111111100 11111 11110 1 63 11101111 111110 111111100 11110 11100 1 64 11101111 111110 111111100 11110 11110 1 65 11101111 111110 111111100 11110 11110 1 66 11101110 111110 111111100 11110 11110 1 67 11101110 111110 111111100 11110 11110 1 68 11101110 111110 111111100 11110 11110 1 69 11101111 111110 111111100 11110 11110 1 70 11101110 111110 111111100 11110 11110 1 71 11110100 111110 111111100 11111 11110 1 72 11101110 111110 111111100 11110 11110 1 73 11101110 111110 111111100 11110 11110 1 74 11101110 111110 111111100 11110 11110 1 75 11101110 111110 111111100 11110 11110 1 76 11110100 111110 111111100 11110 11110 1 77 11101110 111110 111111100 11110 11110 1 78 11101110 111110 111111100 11110 11110 1 79 11101110 111110 111111100 11111 11111 1 80 11110100 111110 111111100 11110 11110 1 81 11101110 111110 111111100 11111 11110 1 82 11101110 111110 111111100 11111 11110 1 83 11101110 111110 111111100 11111 11110 1 84 11101110 111110 111111100 11111 11110 1 85 11101110 111110 111111100 11111 11110 1 86 11101110 111110 111111100 11111 11110 1 87 11101110 111110 111111100 11111 11110 Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 1 88 11101110 111110 111111100 11111 11110 1 89 11101110 111110 111111100 11111 11110 1 90 11101110 111110 111111100 11111 11111 1 91 11101110 111110 111111100 11111 11111 1 92 11101110 111110 111111100 11111 11111 1 93 11101110 111110 111111100 11110 11110 1 94 11101110 111110 111111100 11110 11110 1 95 11101110 111110 111111100 11111 11110 1 96 11101110 111110 111111100 11111 11110 1 97 11101110 111111 111111100 11110 11110 1 98 11101110 111110 111111100 11110 11110 1 99 11110100 111110 111111100 11110 11110 1 102 11101110 111110 111111100 11111 11111 1 103 11101110 111110 111111100 11111 11111 1 104 11101110 111110 111111100 11110 11110
45 1 105 11110100 111110 111111100 11110 11110 1 106 11101110 111110 111111100 11110 11110 2 111 11101111 111110 111111100 11110 11110 2 112 11110100 111110 111111100 11111 11110 2 113 11101110 111110 111111110 11111 11110 2 114 11101110 111110 111111100 11111 11110 2 115 11101110 111110 111111100 11110 11110 2 116 11101110 111110 111111100 11110 11110 2 117 11101110 111110 111111100 11110 11110 2 118 11101110 111110 111111100 11110 11110 2 119 11101110 111110 111111100 11111 11110 2 120 11110100 111110 111111100 11111 11110 2 121 11101110 111110 111111100 11111 11110 2 122 11101111 111110 111111001 11110 11110 2 123 11101110 111110 111111001 11111 11110 2 124 11101110 111110 111111001 11111 11111 2 125 11101110 111110 111111001 11110 11111 2 126 11101110 111110 111111001 11110 11111 2 127 11101111 111110 111111001 11110 11110 2 128 11101111 111110 111111100 11110 11110 2 129 11101111 111110 111111100 11111 11110 2 130 11101111 111110 111111100 11111 11110 2 131 11101111 111110 111111100 11111 11110 2 132 11101111 111110 111111100 11111 11110 2 133 11101110 111110 111111001 11111 11110 2 134 11101110 111110 111111001 11111 11110 2 135 11101111 111110 111111100 11111 11110 2 136 11101111 111110 111111100 11111 11110 2 137 11101110 111110 111111001 11111 11110 2 138 11101110 111110 111111100 11111 11110 2 139 11110100 111110 111111001 11111 11110 2 140 11101111 111110 111111100 11110 11110 2 141 11101110 111110 111111100 11111 11110 2 142 11101111 111110 111111100 11110 11110 2 143 11101111 111110 111111100 11110 11110 2 144 11101111 111110 111111100 11110 11110 2 145 11101111 111110 111111100 11110 11110 Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 2 147 11101111 111110 111111100 11110 11111 2 148 11101111 111110 111111100 11110 11110 2 149 11101110 111110 111111100 11111 11110 2 150 11101111 111110 111111100 11110 11111 2 151 11101111 111110 111111100 11111 11110 2 152 11101110 111110 111111100 11110 11110 2 154 11101111 111110 111111100 11111 11111 2 155 11101111 111110 111111100 11110 11111 2 156 11101111 111110 111111100 11110 11111 2 157 11101111 111110 111111100 11110 11110 2 158 11101111 111110 111111100 11111 11110 2 160 11101110 111110 111111100 11110 11111 2 161 11101111 111110 111111100 11110 11110 2 162 11101111 111110 111111100 11110 11110 2 163 11101111 111110 111111100 11110 11110
46 2 164 11101111 111110 111111100 11110 11110 2 165 11101111 111110 111111100 11110 11110 2 166 11101111 111110 111111100 11110 11110 2 167 11101111 111110 111111001 11111 11110 2 168 11101111 111110 111111100 11110 11110 2 169 11101111 111110 111111100 11110 11110 2 170 11101111 111110 111111100 11110 11110 2 171 11101111 111110 111111100 11110 11111 2 172 11110100 111110 111111100 11110 11110 2 173 11101111 111110 111111100 11110 11111 2 174 11101111 111110 111111100 11110 11110 2 175 11101111 111110 111111100 11110 11110 2 176 11101111 111110 111111100 11110 11110 2 177 11101111 111110 111111100 11110 11111 2 178 11101110 111110 111111100 11110 11111 2 179 11101111 111110 111111100 11110 11111 2 180 11101110 111110 111111100 11110 11111 2 181 11101111 111110 111111100 11111 11111 2 182 11101111 111110 111111100 11111 11111 2 183 11101111 111110 111111100 11111 11111 2 184 11101110 111110 111111100 11111 11111 2 185 11101111 111110 111111001 11111 11110 2 186 11101110 111110 111111100 11111 11111 2 187 11101111 111110 111111100 11111 11111 2 188 11101110 111110 111111100 11111 11110 2 189 11101111 111110 111111100 11111 11110 2 190 11101111 111110 111111100 11111 11110 2 191 11101111 111110 111111100 11110 11110 2 192 11101111 111110 111111100 11110 11110 2 193 11101111 111110 111111100 11110 11110 2 194 11101111 111110 111111100 11110 11110 3 195 11101111 111110 111111100 11111 11110 3 196 11101110 111110 111111100 11111 11111 3 198 11101110 111111 111111100 11111 11111 3 199 11101111 111110 111111100 11111 11111 3 200 11101111 111110 111111100 11110 11110 3 201 11101111 111110 111111100 11110 11111 Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 3 202 11101111 111110 111111100 11110 11110 3 203 11101111 111110 111111100 11111 11111 3 204 11101111 111110 111111100 11110 11110 3 205 11101111 111110 111111100 11110 11110 3 206 11101111 111110 111111100 11110 11110 3 207 11101111 111110 111111100 11110 11110 3 208 11101111 111110 111111100 11110 11110 3 209 11101111 111110 111111100 11110 11110 3 210 11101111 111110 111111100 11110 11110 3 211 11101111 111110 111111100 11110 11110 3 212 11101111 111110 111111100 11110 11110 3 213 11101111 111111 111111100 11110 11110 3 214 11101110 111110 111111100 11110 11110 3 215 11101111 111110 111111100 11110 11110 3 216 11110100 111110 111111100 11110 11110
47 3 217 11101111 111110 111111100 11110 11110 3 218 11101111 111110 111111100 11110 11110 3 219 11101111 111110 111111100 11110 11110 3 220 11101111 111110 111111100 11110 11110 3 221 11101111 111110 111111100 11110 11110 3 222 11101110 111110 111111100 11110 11110 3 223 11101111 111110 111111100 11110 11110 3 224 11101111 111110 111111100 11110 11110 3 225 11101111 111110 111111100 11110 11111 3 226 11101111 111110 111111100 11110 11110 3 227 11101111 111110 111111100 11110 11110 3 228 11101111 111110 111111100 11110 11110 3 229 11101111 111110 111111100 11111 11110 3 230 11101111 111110 111111100 11110 11110 3 231 11110100 111110 111111100 11111 11111 3 232 11101110 111110 111111100 11110 11111 3 233 11110100 111110 111111100 11110 11111 3 234 11101111 111110 111111100 11111 11110 3 235 11101111 111110 111111100 11111 11110 3 236 11101111 111110 111111100 11110 11110 3 237 11101111 111110 111111100 11110 11110 3 238 11101111 111110 111111100 11110 11110 3 239 11101111 111111 111111100 11110 11111 3 240 11101111 111110 111111100 11110 11110 3 241 11101111 111110 111111100 11110 11111 3 242 11101111 111110 111111100 11110 11111 3 243 11101111 111110 111111110 11110 11110 3 244 11101111 111110 111111100 11110 11110 3 245 11101110 111110 111111100 11110 11110 3 246 11101111 111110 111111100 11110 11110 3 247 11101111 111110 111111100 11110 11110 3 248 11101111 111110 111111100 11110 11110 3 249 11101111 111110 111111100 11110 11110 3 250 11101111 111110 111111100 11110 11110 3 251 11101111 111110 111111100 11111 11110 3 252 11101111 111110 111111100 11111 11110 3 253 11101111 111110 111111100 11110 11110 Nature Precedings : hdl:10101/npre.2009.3007.1 Posted 31 Mar 2009 3 254 11101111 111110 111111100 11110 11110 3 255 11101111 111110 111111100 11110 11111 3 256 11101111 111110 111111100 11110 11111 3 257 11101111 111110 111111100 11111 11111
48