Nuclear DNA amounts in angiosperms: progress, problems and prospects

Abstract

Background: The nuclear DNA amount in an unreplicated haploid chromosome complement (1C-value) is a key diversity character with many uses. Angiosperm C-values have been listed for reference purposes since 1976, and pooled in an electronic database since 1997 (http://www.kew.org/cval/homepage). Such lists are cited frequently and provide data for many comparative studies. The last compilation was published in 2000, so a further supplementary list is timely to monitor progress against targets set at the first plant genome size workshop in 1997 and to facilitate new goal setting.

Scope: The present work lists DNA C-values for 804 species including first values for 628 species from 88 original sources, not included in any previous compilation, plus additional values for 176 species included in a previous compilation.

Conclusions: 1998-2002 saw striking progress in our knowledge of angiosperm C-values. At least 1700 first values for species were measured (the most in any five-year period) and familial representation rose from 30 % to 50 %. The loss of many densitometers used to measure DNA C-values proved less serious than feared, owing to the development of relatively inexpensive flow cytometers and computer-based image analysis systems. New uses of the term genome (e.g. in 'complete' genome sequencing) can cause confusion. The Arabidopsis Genome Initiative C-value for Arabidopsis thaliana (125 Mb) was a gross underestimate, and an exact C-value based on genome sequencing alone is unlikely to be obtained soon for any angiosperm. Lack of this expected benchmark poses a quandary as to what to use as the basal calibration standard for angiosperms. The next decade offers exciting prospects for angiosperm genome size research. The database (http://www.kew.org/cval/homepage) should become sufficiently representative of the global flora to answer most questions without needing new estimations. DNA amount variation will remain a key interest as an integrated strand of holistic genomics.

Fig. 1.

(A) Mean number per year of total (open symbols) and ‘first’ (closed symbols) DNA C-value estimates communicated in ten successive 5-year periods and the 3-year period 2000–2002, between 1950 and 2002. Based on analysis of data listed in the present Appendix table, and the Angiosperm DNA C-values database (release 4.0, January 2003). (B) Percentage of C-value estimates published or communicated during 1965–2002 that are first values for species listed in the present Appendix table and the Angiosperm DNA C-values database (release 4.0, January 2003).

Fig. 2.

Cumulative percentage of angiosperm families recognized by the Angiosperm Phylogeny Group (APG) (APGII, 2003) with a first C-value represented in the present Appendix table, the Angiosperm DNA C-values database (release 4.0, January 2003), plus eleven known to the present authors in September 2003.

Fig. 3.

(A) Expected error variation in a large population of DNA C-value estimates for one genotype as underestimates (in the lower tail) and overestimates (in the upper tail) surround more accurate, intermediate, genome size estimates. (B) Histogram showing frequency of C-values for the 85 smallest species in the database or Appendix.

Fig. 4.

The percentage of diploids (open bars) and polyploids (closed bars) among 3400 species of known DNA amount and ploidy level ranked in order of increasing DNA amount and divided into five groups with 680 species per group. Data taken from the Angiosperm DNA C-values database (release 4.0, January 2003) and the present Appendix.