Research Article |
Corresponding author: Vladimir A. Lukhtanov ( lukhtanov@mail.ru ) Academic editor: Valentina G. Kuznetsova
© 2015 Vladimir A. Lukhtanov, Asya V. Novikova.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Lukhtanov VA, Novikova AV (2015) Interpretation of mitochondrial diversity in terms of taxonomy: a case study of Hyponephele lycaon species complex in Israel (Lepidoptera, Nymphalidae, Satyrinae). ZooKeys 538: 21-34. https://doi.org/10.3897/zookeys.538.6689
|
It is difficult to interpret mitochondrial diversity in terms of taxonomy even in cases in which a concordance exists between mitochondrial, ecological and morphological markers. Here we demonstrate this difficulty through a study of Israeli Hyponephele butterflies. We show that samples commonly identified as Hyponephele lycaon are represented on Mount Hermon in Israel by two sympatric groups of individuals distinct both in mitochondrial DNA-barcodes (uncorrected p-distance = 3.5%) and hindwing underside pattern. These two groups were collected in different biotopes. They also tended to be different in length of brachia in male genitalia, although the latter character is variable. We reject the hypothesis that the discovered COI haplogroups are selectively neutral intraspecific characters. We hypothesize that they represent: either (1) two different biological species, or (2) a consequence of a strong positive selection acting at intraspecific level and resulting in two intraspecific clusters adapted to low and to high elevations. If we accept the first hypothesis, then provisionally these two haplogroups can be attributed to transpalearctic H. lycaon sensu stricto and to H. lycaonoides, previously known from Iran and East Turkey.
adaptation to high/low elevation, biodiversity, COI , cryptic species, DNA barcoding, disruptive selection, habitat-related selection, molecular markers
Hyponephele Muschamp, 1915 is a large and taxonomically diverse genus of satyrine butterflies. The genus contains 39 species (
Within the genus, Hyponephele lycaon (Rottenburg, [1775]) is the best known and the most common species broadly distributed in the temperate zone of the Palearctic from Portugal in the west to Far East Russia in the east (
In our study we analysed mitochondrial DNA barcodes and morphological and ecological markers to show that butterflies commonly identified as Hyponephele lycaon are represented in Israel by two sympatric groups of individuals. We further discuss different possible evolutionary and taxonomic interpretations of the pattern discovered.
In the course of our DNA barcode survey of Israeli butterflies (2012–2015) we found butterflies similar to H. lycaon on Mount Hermon in northern Israel. They were collected in a small area situated between 33°17'12"N, 35°45'49"E, at 1440 m and 33°18'38"N, 35°47'07"E, at 2050 m. The distance between these extreme points of the collecting was 3460 m (measured using Google Earth map). Some of the butterflies were collected in the forest zone at 1450–1600 m above sea level, other were collected in the subalpine zone with predominance of xerophytous vegetation at 1800–2050 m above sea level (Table
Haplogroup or taxon | Country | Ecological zone | Pattern of the wing underside | BOLD Process ID | Field ID | GenBank accession # |
---|---|---|---|---|---|---|
I | Israel | forest | contrasting | BPAL2756-15 | CCDB-17969_A01 | KT864697 |
I | Israel | forest | contrasting | BPAL2757-15 | CCDB-17969_A02 | KT864698 |
I | Israel | forest | contrasting | BPAL2758-15 | CCDB-17969_A03 | KT864699 |
I | Israel | forest | contrasting | BPAL2760-15 | CCDB-17969_A05 | KT864700 |
I | Israel | forest | contrasting | BPAL2761-15 | CCDB-17969_A06 | KT864701 |
I | Israel | forest | contrasting | BPAL2765-15 | CCDB-17969_A10 | KT864702 |
II | Israel | forest | pale | BPAL2695-14 | CCDB-17968_C11 | KT864691 |
II | Israel | subalpine | pale | BPAL2705-14 | CCDB-17968_D09 | KT864692 |
II | Israel | subalpine | pale | BPAL2706-14 | CCDB-17968_D10 | KT864693 |
II | Israel | subalpine | pale | BPAL2733-14 | CCDB-17968_G01 | KT864690 |
II | Israel | subalpine | pale | BPAL2762-15 | CCDB-17969_A07 | KT864694 |
II | Israel | subalpine | pale | BPAL2763-15 | CCDB-17969_A08 | KT864695 |
II | Israel | subalpine | pale | BPAL2764-15 | CCDB-17969_A09 | KT864696 |
H. lupinus | Israel | n/a | n/a | BPAL2719-14 | CCDB-17968_E11 | KT864688 |
H. lupinus | Israel | n/a | n/a | BPAL2683-14 | CCDB-17968_B11 | KT864689 |
H. maroccana | Morocco | n/a | n/a | BPAL1378-12 | CCDB-03030_D12 | KT864703 |
H. maroccana | Morocco | n/a | n/a | BPAL1377-12 | CCDB-03030_D11 | KT864704 |
H. maroccana | Morocco | n/a | n/a | BPAL1376-12 | CCDB-03030_D10 | KT864705 |
DNA barcodes, 658 bp fragments within mitochondrial gene,
The procedure of phylogenetic inference was described previously (
Butterfly photographs were taken with Nikon D810 digital camera equipped with a Nikon AF-S Micro Nikkor 105 mm lens. Genitalia photographs were taken with Leica M205C binocular microscope equipped with Leica DFC495 digital camera, and processed using the Leica Application Suite, version 4.5.0 software.
During a 2012-2015 survey of Israeli fauna, H. lycaon-similar butterflies were found only on Mount Hermon in northern Israel. We never observed H. lycaon-similar butterflies in other parts of Israel, although the distantly related H. lupinus (Costa, 1836) was found not only in the northern, but also in central Israel. Thus, our observations support the finding that the geographic range of H. lycaon species complex is restricted in Israel to the northernmost part of the country (
Molecular analysis of H. lycaon-similar samples (Table
With a single exception (female sample BPAL2695-14|CCDB-17968_C11, Fig.
Standard χ2-test was used to distinguish between random vs. non-random distribution haplogroups I and II in the low (forest) and high (subalpine) zones. Empirical and expected frequencies of COI haplogroups I and II in low and high altitude belts were compared (Table
Primary data (number of samples) for χ2-analysis of random vs. non-random distribution of the COI I and II haplogroups in the low (forest) and high (subalpine) zones.
empirical values | expected values (in case of random distribution) | |||
---|---|---|---|---|
low altitude | high altitude | low altitude | high altitude | |
COI haplogroup I | 6 | 0 | 3.234 | 2.772 |
COI haplogroup II | 1 | 6 | 3.766 | 3.228 |
The representatives of these two clusters were also different in the pattern on the hindwing underside (Figs
Pattern of the wing underside in haplogroups I and II samples. The pictures were taken using a flash aCCDB-17969_A01 bCCDB-17969_A02 cCCDB-17969_A03 dCCDB-17969_A06 eCCDB-17969_A10 fCCDB-17969_A05 gCCDB-17968_C11 hCCDB-17968_D09 iCCDB-17968_G01 jCCDB-17969_A07 kCCDB-17969_A08 lCCDB-17969_A09 mCCDB-17968_C11.
A standard χ2-test was used to distinguish between random vs. non-random association between haplogroups I and II and hindwing underside pattern (Table
Primary data (number of samples) for χ2-analysis of random vs. non-random association between the haplogroup I and II and the hindwing underside pattern.
empirical values | expected values (in case of random distribution) | |||
---|---|---|---|---|
contrast pattern | pale | contrast pattern | pale | |
COI haplogroup I | 6 | 0 | 2.772 | 3.234 |
COI haplogroup II | 0 | 7 | 3.228 | 3.766 |
The representatives of these two COI haplogroups also tended to be different in the length of the brachia in male genitalia (Fig.
The COI genetic distance between haplogroups I and II (3.5 %) is higher than the ‘standard’ 2.7–3.0% DNA-barcoding threshold commonly used as a tentative indicator for species distinctness of the taxa compared (
Three alternative explanations can account for bimodal sympatric distribution of mitochondrial markers. First, the diverged COI sequences may be selectively neutral intraspecific characters. Both preservation of a variety of ancestral haplotypes and mitochondrial introgression due to complex phylogeographic history could be responsible for such a neutral polymorphism (
In our case the first hypothesis (neutral polymorphism) can be easily rejected. It predicts that the COI haplogroups I and II should be stochastically (i.e. randomly) distributed within high and low altitude belts. This prediction is not supported by χ2-test that demonstrated significantly non-random distribution of the COI haplogroups.
The second hypothesis (strong intraspecific positive selection) offers a more exotic, but not improbable, explanation. As COI sequence can be translated into a subunit of cytochrome c oxidase, a functional protein in mitochondria involved in energy metabolism (
The third hypothesis (two different species) seems to be a more likely explanation in the case of haplogroups I and II, especially if one takes into account the high level of genetic divergences between the haplogroups and concordance between molecular (Fig.
The presence of two sympatric, ecologically differentiated groups within H. lycaon complex in the Middle East is not a completely novel issue. A similar situation is known to exist in Iran and East Turkey (
If the species status of the discovered haplogroups will be confirmed in further studies, we suggest that, following
The financial support for this study was provided by the grant N 14-14-00541 from the Russian Science Foundation to the Zoological Institute of the Russian Academy of Sciences. We are grateful to Evgeny Zakharov (Canadian Centre for DNA Barcoding, Guelph, Canada) for sequencing the samples. We thank Dubi Benyamini for support of our DNA barcoding survey of Israeli butterflies and valuable comments.