First Nuclear DNA C-values for 18 Eudicot Families

Abstract

• Background and Aims A key target set at the second Plant Genome Size Workshop, held at the Royal Botanic Gardens, Kew in 2003, was to produce first DNA C-value data for an additional 1 % of angiosperm species, and, within this, to achieve 75 % familial coverage overall (up from approx. 50 %) by 2009. The present study targeted eudicot families for which representation in 2003 (42·5 %) was much lower than monocot (72·8 %) and basal angiosperm (69·0 %) families.

• Methods Flow cytometry or Feulgen microdensitometry were used to estimate nuclear DNA C-values, and chromosome counts were obtained where possible.

• Key Results First nuclear DNA C-values are reported for 20 angiosperm families, including 18 eudicots. This substantially increases familial representation to 55·2 % for angiosperms and 48·5 % for eudicots.

• Conclusions The importance of targeting specific plant families to improve familial nuclear DNA C-value representation is reconfirmed. International collaboration will be increasingly essential to locate and obtain material of unsampled plant families, if the target set by the second Plant Genome Size Workshop is to be met.

INTRODUCTION

Swift (1950) introduced the term ‘C-value’ (C standing for ‘constant’) to refer to the amount of DNA in the unreplicated haploid or gametic nucleus of an individual (see also Greilhuber et al., 2005). Nuclear DNA amounts (genome sizes) for species are used in a wide variety of biological fields (Bennett and Leitch, 2000). Genome size is highly variable, differing approx. 1000-fold between angiosperm species.

Since 1950, C-values have been published for >4100 angiosperm species (Bennett and Leitch, 2005a). With C-values for gymnosperms, bryophytes, pteridophytes and algae, these have been pooled into an electronic database and placed on the Internet (Bennett and Leitch, 2004). Since its launch in 1997, this database has received >60 000 hits, demonstrating the utility of such C-value data.

In 1997, the first Angiosperm Genome Size Workshop and Discussion Meeting held at the Royal Botanic Gardens, Kew (RBG, Kew) identified major gaps in plant DNA C-value data and recommended targets and priorities to fill them by international collaboration. Two key goals were: (1) to estimate first C-values for a further 1 % of angiosperms (approx. 2500 species); and (2) to complete familial representation, by 2002.

A second Plant Genome Size Workshop and Discussion meeting (held at RBG, Kew in 2003) reviewed progress towards meeting the 1997 targets. Significant progress had been made, with at least 1700 first estimates for species being reported in 1998–2002. However, familial representation had only increased from approx. 30 % to approx. 50 % (Bennett and Leitch, 2005a). Moreover, within this, representation of eudicot families was low (43·1 %) compared with that of monocot (72·8 %) and basal (69·0 %) angiosperm families (Table 1). New and modified targets set for the next 5 years included to produce first C-values for an additional 1 % of angiosperm (2500) species and, within this, to achieve at least 75 % familial representation by 2009 (Bennett and Leitch, 2005b). With first data for 20 previously unrepresented families, including 18 eudicots, the present paper is a significant step towards achieving this aim, increasing angiosperm and eudicot familial representation to 55·2 and 48·5 %, respectively.

Table 1.

Cumulative proportion of angiosperm families with C-value data

Source of data	Cumulative proportion of 29 basal angiosperm families represented, % (n)	Cumulative proportion of 92 monocot families represented, % (n)	Cumulative proportion of 332 eudicot families represented, % (n)	Cumulative proportion of all 453 angiosperm families represented, % (n)
Angiosperm DNA C-values Database (release 3·0, September 2001)	48·28 (14)	46·74 (43)	26·20 (87)	31·79 (144)
Hanson et al. (2001a)	48·28 (14)	50·00 (46)	33·43 (111)	37·75 (171)
Hanson et al. (2001b)	48·28 (14)	55·43 (51)	39·76 (132)	43·49 (197)
Leitch and Hanson (2002)	65·52 (19)	55·43 (51)	39·76 (132)	44·59 (202)
Hanson et al. (2003)	65·52 (19)	72·83 (67)	42·47 (141)	50·11 (227)
Angiosperm DNA C-values Database (release 5·0, December 2004)	68·97 (20)	72·83 (67)	43·07 (143)	50·77 (230)
This study	68·97 (20)	75·00 (69)	48·49 (161)	55·19 (250)

The sample is diverse, comprising plants from a wide geographical range with a variety of uses. Hamamelis virginiana (witch hazel) is cultivated as an ornamental and for the medicinal properties of its bark and leaves. Another species known for its medicinal properties is Salvadora persica (toothbrush tree). The Bedouin use the bristly fibres and vessels from beaten twigs as chewing sticks, the antiseptic properties of which reduce tooth decay and heal gums. Seed of Strychnos nux-vomica is the commercial source of strychnine, used for rodent control in Europe since the early 1800s. Corynocarpus laevigatus also has poisonous seeds, whereas Salvadora persica and Dillenia indica both have edible fruits (all facts taken from Mabberley, 1997). Medusagyne oppositifolia—an endangered endemic from the Seychelles—is of high conservation importance. Once thought to be extinct, it was rediscovered in 1970 (Robertson et al., 1989). Other species in the sample inhabit highly specialized environments. Frankenia species can grow in saline environments, whereas Ecdeiocolea monostachya and Sarracenia flava grow on nutrient-poor soils. Leaves of the carnivorous species S. flava are modified to form pitchers that can trap insects. Thus, this highly diverse sample is interesting as a source of novel genome size data in previously unrepresented families, but also for providing useful information for several poorly represented groups, for example halophytes (as noted in Bennett and Leitch, 2005a).

MATERIALS AND METHODS

Plant material

Table 2 lists 20 perennial species from different families not previously represented in the Plant DNA C-values Database, together with their geographic distribution, the source of the material studied in the current work and their identification status.

Table 2.

Family, geographical region of origin, source of experimental material, RBG Kew identity number (ID no.), cytology number (cyt. no.) and identification status for 20 species studied in the present work

Taxon	Family	Geographical distribution of genus	Source of material§	ID no.	Cyt. no.	Identification status‡
MONOCOTS
COMMELINIDS
Commelinales
Conostylis candicans	Haemodoraceae	S.W. Australia	KPBG	–	03-3	b
Poales
Ecdeiocolea monostachya	Ecdeiocoleaceae	S.W. Australia	KPBG	–	02-160	a
EUDICOTS
Ranunculales
Euptelea pleiosperma	Eupteleaceae	Assam to S.W. and central China, Japan	MCN	–	02-161	b
CORE EUDICOTS
Dillenia indica	Dilleniaceae	Indian Ocean, India to New Guinea and Australia	CS	–	01-133	c
Caryophyllales
Basella alba	Basellaceae	Madagascar, E. Africa and Pantropics	RBG, K	2000-4205	01-138	b
Frankenia sp.	Frankeniaceae	Temperate Europe, Australia and America, and subtropical salty habitats	RBG, K	1992-1765	01-149	d
Saxifragales
Daphniphyllum pentandrum var. oldhamii	Daphniphyllaceae	China, Indonesia and Malaysia to tropical Australia	RBG, K	1994-1394	02-71	d
Hamamelis virginiana	Hamamelidaceae	Eastern N. America and E. Asia	RBG, K	2000-2327	01-141	c
Rosids
Stachyurus praecox	Stachyuraceae	Himalayas to Japan	RBG, K	2000-3874	01-142	c
Staphylea bumalda	Staphyleaceae	Northern temperate Europe	RBG, K	1997-7101	01-155	c
Eurosids I
Cucurbitales
Corynocarpus laevigatus	Corynocarpaceae	S.W. Pacific	RBG, K	1973-12186	03-63	b
Datisca cannabina	Datiscaceae	Asia, Malaysia to Australia, W.N. America	RBG, K	1938-79503	03-62	d
Fabales
Polygala calcarea ‘Lillet’	Polygalaceae	Widespread except New Zealand	RBG, K	1981-1444	01-120	d
Malpighiales
Medusagyne oppositifolia	Ochnaceae	Seychelles	RBG, K	1981-2059	05-1	b*
Oxalidales
Crinodendron patagua	Elaeocarpaceae	S. America	RBG, K	1955-32211	01-144	b
Eurosids II
Brassicales
Salvadora persica	Salvadoraceae	Tropical Africa to tropical Asia	RBG, K	1997-5264	02-16	c
Malvales
Gnidia polystachya	Thymelaeaceae	Tropical and S. Africa to Arabia, Madagascar to W. India and Sri Lanka	CS	–	00-60	b*
Asterids
Ericales
Sarracenia flava	Sarraceniaceae	Eastern N. America	RBG, K	1987-3807	02-158	c
Euasterids I
Gentianales
Strychnos nux-vomica	Loganiaceae	Tropical and warm Africa	RBG, K	1972-391	02-83	d
Lamiales
Verbena rigida	Verbenaceae	Tropical and temperate America, Europe	GGC	–	02-144	a

^§Plant material obtained from RBG, Kew (RBG, K), Chiltern Seeds, Cumbria (CS), Ginkgo Garden Centre, London (GGC), Mallet Court Nursery, Taunton (MCN) and Kings Park and Botanic Garden, Perth, Australia (KPBG).

^‡Identification information: a = taxonomically verified and herbarium voucher prepared for species; b = no herbarium voucher, but species has been taxonomically verified;

^b*= species taxonomically verified, and is currently being grown on at RBG, Kew to provide a herbarium voucher; c = species not taxonomically verified, but is currently being grown on at RBG, Kew to provide a herbarium voucher; d = species not taxonomically verified, but herbarium voucher prepared.

Growth of plants

Most of the plant material used in this study was taken from the Living Collections at RBG, Kew. Seeds of D. indica and Gnidia polystachya were germinated on 1 % agar in a Petri dish according to the temperatures and conditions outlined in the germination instructions provided. Fresh leaf material of E. monostachya was obtained from Kings Park and Botanic Garden (Perth, Australia). Seeds of Conostylis candicans were treated with Instant Smoke Plus Seed Primer (Kirstenbosch National Botanical Institute) prior to sowing on compost (Lloyd et al., 2000).

Chromosome counts

Chromosome counts were obtained using a standard root tip squash technique. Young healthy root tips were taken from either freshly germinated seeds or potted plants, pre-treated in 2 mm 8-hydroxyquinoline for 4·5 h at 18 °C then fixed in a freshly prepared solution of 3 : 1 ethanol : glacial acetic acid. Hydrolysis time in 1 m HCl at 60 °C varied between 5 and 10 min depending on the toughness of the material. Photographs of metaphase cells were taken on a Zeiss photomicroscope III using Pan F film and are retained for reference purposes. Slides, made permanent using liquid nitrogen, are stored at RBG, Kew.

Estimating nuclear DNA C-values

Nuclear DNA C-values were estimated using either flow cytometry or Feulgen microdensitometry. The method used for each taxon is shown in Table 3 together with the calibration standard used. Several different standards were used to cover the range of C-values encountered. The standard used against each test species was determined empirically. The 4C-values used to convert arbitrary units into absolute values were taken from Bennett and Leitch (2005a), except for Solanum lycopersicum ‘Gardener’s Delight' (Lycopersicon esculentum Gardener's Delight) which was determined by Obermayer et al. (2002). [Phylogenetic analysis (e.g. Spooner et al., 1993) has revealed that Lycopersicon is embedded in Solanum, and thus the preferred name for L. esculentum is now S. lycopersicum.]

Table 3.

Family, chromosome number (2n), nuclear DNA amounts, calibration standard and method used to estimate DNA C-values for 20 species studied in the present work

Taxon	Family	2n*	4C DNA amount ± s.d. (pg)	1C DNA amount (Mbp)§	Calibration standard species‡	Method¶
MONOCOTS
COMMELINIDS
Commelinales
Conostylis candicans	Haemodoraceae	16*	5·22 ± 0·02	1279	Solanum	FC
Poales
Ecdeiocolea monostachya	Ecdeiocoleaceae	Approx. 38	3·95 ± 0·02	968	Oryza	FC
EUDICOTS
Ranunculales
Euptelea pleiosperma	Eupteleaceae	28	2·67 ± 0·06	654	Solanum	FC
CORE EUDICOTS
Dillenia indica	Dilleniaceae	Approx. 52	2·05 ± 0·27	502	Vigna	Fe
Caryophyllales
Basella alba	Basellaceae	Approx. 48	7·17 ± 0·37	1757	Vigna	Fe
Frankenia sp.	Frankeniaceae	Approx. 28	1·28 ± 0·14	314	Vigna	Fe
Saxifragales
Daphniphyllum pentandrum var. oldhamii	Daphniphyllaceae	–	5·01 ± 0·02	1227	Solanum	FC
Hamamelis virginiana	Hamamelidaceae	24	2·34 ± 0·31	573	Vigna	Fe
Rosids
Stachyurus praecox	Stachyuraceae	24	1·42 ± 0·20	348	Vigna	Fe
Staphylea bumalda	Staphyleaceae	26	6·86 ± 0·29	1681	Vigna	Fe
Eurosids I
Cucurbitales
Corynocarpus laevigatus	Corynocarpaceae	44*	3·08 ± 0·014	755	Solanum	FC
Datisca cannabina	Datiscaceae	22*	4·59 ± 0·03	1125	Solanum	FC
Fabales
Polygala calcarea ‘Lillet’	Polygalaceae	Approx. 30	1·72 ± 0·03	421	Vigna	FC
Malpighiales
Medusagyne oppositifolia	Ochnaceae	–	1·33 ± 0·03	326	Solanum	FC
Oxalidales
Crinodendron patagua	Elaeocarpaceae	16	1·20 ± 0·14	294	Vigna	Fe
Eurosids II
Brassicales
Salvadora persica	Salvadoraceae	24*	1·68 ± 0·02	412	Solanum	FC
Malvales
Gnidia polystachya	Thymelaeaceae	18	3·37 ± 0·35	826	Vigna	Fe
Asterids
Ericales
Sarracenia flava	Sarraceniaceae	26	17·38 ± 0·20	426	Pisum	FC
Euasterids I
Gentianales
Strychnos nux-vomica	Loganiaceae	24* or 44*	1·71 ± 0·29	419	Solanum	FC
Lamiales
Verbena rigida	Verbenaceae	42	4·75 ± 0·44	1164	Oryza	FC

^*Chromosome numbers are taken from the literature. All others were determined in the present work.

^§1 pg = 980 Mbp.

^‡Calibration standard used: Pisum = Pisum sativum ‘Minerva Maple’, 4C = 19·46 pg, Oryza = Oryza sativa IR36, 4C = 2·02 pg; Vigna = Vigna radiata ‘Berken’, 4C = 2·12 pg; and Solanum = Solanum lycopersicum (Lycopersicon esculentum) ‘Gardener’s Delight', 4C = 4·00 pg.

^¶Fe = Feulgen microdensitometry; FC = flow cytometry.

Flow cytometry

Young healthy leaf tissue collected from the calibration standard and the test species was co-chopped in 1 mL of isolation buffer (0·1 m citric acid, 0·5 % Triton X-100 in distilled water). Samples were filtered through 30 µm nylon mesh, treated with 50 μL of RNase (3 mg mL⁻¹) and incubated at 37 °C for 30 min, after which each sample was briefly mixed before being divided equally into two tubes. The non-base-specific DNA stain propidium iodide (PI) was used for all samples. The staining solution comprised 11·36 g of Na₂HPO₄, 12 mL of PI stock (1 mg mL⁻¹) and 20 mL of 10× stock (100 mm sodium citrate, 250 mm sodium sulfate) made up to 200 mL with double-distilled water. A 2 mL aliquot of the staining solution was added to each tube to give a final PI concentration of 50 µg mL⁻¹ and incubated at 20–25 °C for 20 min. The addition of PI (pH approx. 9·5) to the isolation buffer (pH approx. 1·5) results in a final pH of approx. 7·5–8 (Obermayer and Greilhuber, 1999).

Samples were analysed on a Partec PA II flow cytometer with a 100 W high-pressure mercury lamp, a high-quality red-sensitive photomultiplier and a ×40 gel objective. The linearity of the machine was checked on a regular basis using chicken red blood cells. When the coefficients of variation (CVs) were <3 %, three samples were made and each one was run three times (5000 nuclei per run). When the CVs were >3 %, more samples were prepared and run. Absolute 4C DNA C-values were calculated using the formula: (mean of the test sample peaks÷mean of the calibration standard peaks) × known 4C-value of calibration standard used.

Feulgen microdensitometry

Young healthy root tips collected from the calibration standard and test species from either newly germinated seeds or potted plants were processed and measured using a Vickers M85a microdensitometer as described in Hanson et al. (2001a).

RESULTS AND DISCUSSION

Chromosome counts

Chromosome counts for the species are listed in Table 3. Original counts were obtained for 13 species. No count was obtained for Daphniphyllum pentandrum var. oldhamii or Medusagyne oppositifolia, and there are no previous counts published for these two species. A search through Fedorov (1969) and the Indexes to Plant Chromosome Numbers published by the Missouri Botanical Garden (e.g. see Goldblatt and Johnson, 2003) revealed that this paper is the first to report new specific counts for Stachyurus praecox (2n = 24), G. polystachya (2n = 18) and Crinodendron patagua (2n = 16), and an approximate generic count for E. monostachya (2n = approx. 38). Counts obtained in this study agreed with previously published counts for Euptelea pleiosperma, H. virginiana, Staphylea bumalda, S. flava and Verbena rigida. Previously published counts for D. indica were 2n = 24, 28 and 54. We found 2n = approx. 52, which is similar to the highest previous report. Published counts for Basella rubra were 2n = 41 or 44, whereas we found 2n = approx. 48. A published count of 2n = 34 in Polygala calcarea is similar to the 2n = approx. 30 reported here.

Nuclear DNA amounts

Table 3 lists 4C nuclear DNA amounts in the 20 taxa studied. These varied approx. 14-fold, from 1·20 pg in C. patagua to 17·38 pg in S. flava (mean 4C-value of the whole sample = 3·94 pg). The sample comprises 18 eudicots (including the two species just mentioned) with a mean 4C DNA amount of 3·87 pg. This is much lower than the 4C mean of all known eudicot values which is 12·56 pg, based on 2350 values in the Plant DNA C-values Database (Bennett and Leitch, 2004). Similarly, first values for two monocot families listed in Table 3 (mean 4·59 pg) are also low compared with the 4C mean for monocots as a whole (41·88 pg) based on 1690 values in the Plant DNA C-values Database (Bennett and Leitch, 2004).

Although these new values extend taxonomic representation usefully (see below), they all fall towards the lower end of the ranges of values previously reported for eudicots (4C = 0·4–317·2 pg) and monocots (4C = 0·8–509·6 pg) and hence do not extend the known variation for this aspect of biological diversity. Moreover, more detailed comparisons show that while the data are all first values for families, in no case do they extend the range of genome sizes reported for any higher order group within the eudicots or monocots. Although C-values for 3126 species have been published since 1982, none extended the approx. 1000-fold range for angiosperms (Bennett and Leitch, 2005c). This suggests that the overall picture of genome size variation at these course levels is nearing completion.

Genome size in carnivorous plants

Sarracenia flava has the largest amount of DNA in the sample (4C = 17·38 pg), but it is also of interest as it is a carnivorous plant. Carnivorous plants generally inhabit nutrient-poor environments—such as wetland areas—where minerals are leached out of the soil. In order to survive these conditions, carnivorous plants have adapted to acquire essential minerals from insects. Sarracenia flava is a pitcher plant with modified leaves that attract insects. The insects become trapped in the liquid in the pitcher and the nutrients from the insect are absorbed by the plant.

A search of the Plant DNA C-values Database revealed that data exist for only 12 carnivorous species from seven families (see Hanson et al., 2001a, b). Estimates for 11 of the 12 species range from 4C = 0·76 to 3·8 pg, and are all defined as ‘very small’ (4C ≤5·6 pg), but the twelfth estimate of Drosophyllum lusitanicum (4C = 60·00 pg) is defined as ‘large’ ( 4C = ≥56·0–<140·0 pg) according to Leitch et al. (1998).

Drosophyllum was once included in the family Droseraceae, but has been placed in a separate family Drosophyllaceae on molecular grounds (Albert et al., 1992). Drosophyllum also grows in dry areas and has a well developed root structure, which enables it to acquire water and minerals more easily than Drosera which prefers wet habitats and has a poorly developed root system (Juniper et al., 1989).

Hanson et al. (2001a) speculated that plants growing in nutrient-poor environments might have small genome sizes by necessity. With the exception of Drosophyllum (see above), the 4C-value of diploid S. flava (17·38 pg) is over four times larger than the other C-values reported for carnivorous plants so far. Whereas this still falls in the bottom 3·5 % of the known range of 4C-values in angiosperms (0–509 pg), it is above the mode for angiosperms (4C = 2·4 pg). This adds to our knowledge of carnivorous plants, showing that small or very small DNA C-values are not always an essential adaptation to allow the survival of some plant types that grow in nutrient-poor environments.

Progress towards completing familial representation targets

The Angiosperm DNA C-value Database (release 5·0) comprises 4119 prime estimates from 217 families. Since 2001, a project at RBG, Kew has been carefully targeting plant families for which no C-value data have previously been published (see Hanson et al., 2001a, b, 2003; Leitch and Hanson, 2002), and data for >70 families have been obtained by this work. At the 2003 workshop, a target to estimate C-values for an additional 1 % for angiosperms (approx. 2500 species) was set, and within this to achieve 75 % familial coverage (up from approx. 50 %) by 2009 (for the full report, see http://www.rbgkew.org.uk/cval/workshopreport.html). Table 1 shows the cumulative progress towards the familial representation goal since 2001.

As noted previously (e.g. Hanson et al., 2001a, b, 2003), authorities can differ as to how many angiosperm families they recognize, ranging from 200 to 533 depending on the classification system used (see Brummitt, 1992). Moreover, the number of families recognized can also vary with time, as new families are created and previously recognized families are split or sunk on the basis of new data. Such changes clearly complicate the endeavour of tracking how many and what proportion of families are represented in the database (e.g. Hanson et al., 2003). This study follows the narrow APG (2003) circumscription, recognizing 453 families.

By continued careful targeting, the present work has usefully increased total angiosperm familial representation by >4 %, from 50·8 to 55·2 % (Table 1). Moreover, it has significantly increased the representation of eudicots, which hitherto lagged considerably behind that of other groups, by >5 % from 43·1 to 48·5 %. Given the increasing difficulty of locating and measuring material for unsampled families, as noted previously (Hanson et al., 2003), the present work is a significant step towards meeting the 5-year targets for familial representation set in 2003. Such difficulties, which already limit progress, will intensify as familial representation approaches 100 % in the future. Consequently, international collaboration to locate materials and estimate genome sizes for still unsampled families will be essential, if the long-term goals of 75 % and then complete familial representation are to be achieved.