An Australian Government Initiative
Centre for Australian National Biodiversity Research  
CPBR logo

Home > CANBR > Summer Scholarships > 2004

Phylogenetic relationships of the Australasian citrus: the Citreae tribe in a world context.

Sarah Rich (Summer Scholarship 2003-4)

Supervised by Randy Bayer

The agricultural production of citrus has flourished worldwide for centuries due to both its ornamental and culinary value. The citrus fruits and their wild relatives belong to the Rutaceae family, within the subfamily Aurantioideae. The subtribe Citrineae (tribe Citreae) is a group of these plants which are unique in bearing fruits containing pulp vesicles that fill all the space in the segments of the fruit not occupied by the seeds. They are almost exclusively native to China and the South Pacific region, including Australia. Their unique fruit are the basis for one of the most important fruit industry in warm countries of the world due to the great commercial importance of the juice extracted from species of the genus Citrus (Mabberley, 1998; Swingle and Reece, 1967).

Australia is the fourth largest citrus producing country in the Southern Hemisphere, with the citrus industry having an estimated gross production value of around $4.3 million (Anon, 2004). Wild members of Australasian Citreae are anticipated to possess genes which provide a number of beneficial traits for growth in the Australian environment and therefore it is predicted they will be useful in citrus breeding programs. Breeding programs are already investigating native members resistance to salt, boron and the cold but the breeding of new, high quality Citrus cultivars for the Australian environment is dependent on reliable information about the relationships of species in genera within the tribe Citreae (Birmingham, 1998).

Over the last 250 years a number of taxonomic classifications have been proposed for the plants contained within the tribes and subtribes of the Aurantioideae. The most recent is that of Swingle and Reece (1967) although Tanaka’s original treatment (1936) and more recent classifications are often referred to. The tribe Citreae contains three subtribes the Citrineae, Triphasiinae and the Balsamocitrinae. Within the Citrineae the genus Citrus itself has been variously described as consisting of from 1 to 162 species. Swingle’s widely accepted 1946 classification system recognises 16 species while Tanaka (1977) recognises 147 species (Moore, 2001). Closely related to the genus Citrus are several genera, which were recognized by Swingle, including the genera Clymenia, Eremocitrus, Fortunella, Microcitrus, and Poncirus. Recent work has questioned the distinctness of these genera and some botanists now include them within an enlarged concept of the genus Citrus (Mabberley, 1998).

Relationships among species and genera of the group are complicated by its long history of cultivation as well as a high frequency of bud mutation, a wide cross-compatibility among species as well as many species having some degree of apomixis (asexual reproduction in which plant embryos grow from egg cells without fertilisation by pollen), which tends to reduce variability within species leading to a proliferation of named species (Moore, 2001). CSIRO is aiming to provide a molecular-based evolutionary tree of relationships of members within the tribe Citreae based on chloroplast DNA sequences. The summer project concentrated on 16 species of citrus, mostly Australasian, sequences from these were to be combined with those already elucidated in earlier research.

Method

DNA was extracted from the leaves of fresh frozen samples from the US Department of Agriculture and dry samples from CSIRO’s Merbein Research Station. The polymerase chain reaction (PCR) was used to amplify three regions from the chloroplast genome; the trnL intron and trnL - trnF intergenic spacer (1300 base pairs); the rpsL16 intron (1460 base pairs) and the atpB-rbcL intergenic spacer region (1100 base pairs) (is putting the lengths of the regions weird??) of the chloroplast genome. The amplified PCR product was then cleaned and used to obtain sequences. These were manually aligned, with sequences downloaded from Genbank, for use in cladistic analysis. Phylogenetic analysis was carried out in PAUP* 4.0, employing a maximum parsimony, heuristic search with uninformative characters left out. Searches were conducted using a tree bisection-reconnection (TBR) branch swapping algorithm with all possible shortest trees saved to a maximum of 100000; from these, both strict and 50% majority consensus trees were computed. The time restraints of the summer scholarship program did not allow for a full analysis to be completed, however the preliminary findings were most encouraging.

Results

In essence all the trees produced support Swingle’s treatment by both the monophyletic nature of genera and by the presence of subtribal groups from his treatment, although with the limited number of taxa studied it is difficult to draw definitive conclusions. Figure one shows a strict tree of the combined rps16 intron and the trnL intron and spacer regions. This tree illustrates all of the trends found across the five consensus trees produced and highlights our most important finding; that all the Australasian members of the Citrinae subtribe appear to of monophyletic lineage, as suggested by Mabberley’s (1998) treatment of these taxa. Also falling within this clade in all trees were both Oxanthera neocaledonica and Clymenia polyandra, suggesting that they are both possible members of the Citrus genus. Given their origins (New Caledonia and Papua New Guinea, respectively) as well as their taxonomic histories this was not a particularly surprising find, but one that would definitely bear further investigation. Another surprising result concerned the genus Atalantia, which unlike all other genera studied does not appear to be monophyletic. In all trees, apart from that of the trnL intron and spacer region, Severinia buxifolia and Atalantia ceylanica appeared as a monophyletic clade. Originally part of the Atalantia genus, five of the six Severinia species were only segregated from Atalantia in 1938 by Swingle, but are morphologically very similar to Atalantia. With sequences of only four of the eleven Atalantia species and one Severenia, our results merely highlight the need for further work on these genera.

Overall the results produced at the end of the ten week summer program raised more questions than they solved, however as a preliminary study into the Australasian citrus, this project was successful in highlighting the areas of dissonance within the current classification system. Hopefully with additional work and a more in depth analysis of the data already collected it will be possible to further elucidate the origins and relationships found within the tribe Citreae.


Figure One: Strict consensus tree of combined rps16 intron and the trnL intron and spacer regions.n = 40; Informative characters = 201; Shortest tree length = 393; Rohlf's CI = 0.638; Outgroup = 2 (Ruta graveolens, Chloroxylon swietenia). * = sequenced by S.Rich

In essence all the trees produced support Swingle’s treatment by both the monophyletic nature of genera and by the presence of subtribal groups from his treatment, although with the limited number of taxa studied it is difficult to draw definitive conclusions. Figure one shows a strict tree of the combined rps16 intron and the trnL intron and spacer regions. This tree illustrates all of the trends found across the five consensus trees produced and highlights our most important finding; that all the Australasian members of the Citrinae subtribe appear to of monophyletic lineage, as suggested by Mabberley’s (1998) treatment of these taxa. Also falling within this clade in all trees were both Oxanthera neocaledonica and Clymenia polyandra, suggesting that they are both possible members of the Citrus genus. Given their origins (New Caledonia and Papua New Guinea, respectively) as well as their taxonomic histories this was not a particularly surprising find, but one that would definitely bear further investigation. Another surprising result concerned the genus Atalantia, which unlike all other genera studied does not appear to be monophyletic. In all trees, apart from that of the trnL intron and spacer region, Severinia buxifolia and Atalantia ceylanica appeared as a monophyletic clade. Originally part of the Atalantia genus, five of the six Severinia species were only segregated from Atalantia in 1938 by Swingle, although they are morphologically very similar to Atalantia. With sequences from only four of the eleven Atalantia species and one Severenia, our results merely highlight the need for further work on these genera.

Overall the results produced at the end of the ten week summer program raised more questions than they solved, however as a preliminary study into the Australasian citrus, this project was successful in highlighting the areas of dissonance within the current classification system. Hopefully with additional work and a more in depth analysis of the data already collected it will be possible to further elucidate the origins and relationships found within the tribe Citreae.

Acknowledgements

Thanks to Tthe Centre for Plant Biodiversity Research and CSIRO for providing me with the Summer Scholarship and the opportunity to work on this project. Special thanks to Randy Bayer for all the enthusiasm and encouragement during his supervision, along with the entire molecular systematics lab group for their patience and support

References

Anonymous. 2004 (last updated). Australian Citrus Growers Inc.
[ONLINE] http://www.austcitrus.org.au/internal.php?page_id=6
Accessed: 13 February 2004

Birmingham, E. 1998. ‘Australian native citrus: wild limes from the rainforest to the desert…’ The Australian New Crops Newsletter Issue 10.
[ONLINE] http://www.newcrops.uq.edu.au/newslett/ncn10211.htm
Accessed: 29 January 2004

Mabberley, D J. 1998. ‘Australian Citreae with notes on other Aurantioideae (Rutaceae).’ Telopea Vol 7(4): 333 – 344.

Moore, G A. 2001. ‘Oranges and Lemons: clues to the taxonomy of citrus from molecular markers.’ Trends in Genetics Vol 17(9): 536 – 540.

Swingle, W. T. 1938. A new taxonomic rearrangement of the orange subfamily, Aurantioideae. J. Wash. Acad. Sci. 28: 530-533.

Swingle, W. T. and Reece P. C. 1967. The botany of Citrus and its wild relatives. IN Reuther W., Webber H. and Batchelor L. [eds.] The Citrus Industry, Vol 1.p: 190 – 430. University of California, USA.

Tanaka, T. 1936. On the origin and the future of Japanese citrus flora. Stud. Citrol. 7: 155-175.

Tanaka, T. 1977. Fundamental discussion of Citrus classification. Stud. Citrol. 14: 1–6.

 



Updated 13 May, 2004 , webmaster, CANBR (canbr-info@anbg.gov.au)