Comparison of rRNA genotype frequencies of Parmelia sulcata from long established and recolonizing sites following sulphur dioxide amelioration

Plant Systematics and Evolution

Plant Syst. Evol. 217:177-183 (1999)

© Springer-Verlag 1999 Printed in Austria

Comparison of rRNA genotype frequencies of Parmelia sulcata from long established and recolonizing sites following sulphur dioxide amelioration A. Crespo 1, P. D. Bridge 2, D. L. H a w k s w o r t h 3, M. G r u b e 4, and O. F. Cubero 1 ~Departamento de Biologfa Vegetal II, Universidad Complutense, Madrid, Spain 2Biosystematics and Molecular Biology, CABI Bioscience UK Centre, Bakeham Lane, Egham, Surrey UK 3MycoNova, London, UK 4Institut fiir Botanik, Karl-Franzens-Universit~it Graz, Graz, Austria Received July 24, 1997 Accepted July 20, 1998

Abstract. The variable internally transcribed spacer (ITS) regions of the rRNA gene cluster, and the termini of the large and small subunit genes, were amplified from 231 specimens of the lichen-forming fungus Parmelia sulcata from the UK and Spain. Amplification products comprised three distinct size groups of 580, 622 and 835 base pairs (bp). Analyses of the collections from England, Wales and central Spain established the 622bp genotype as the most widespread, and the 835bp genotype as the next most frequent. The size difference was due to a group I intron at the 3r end of the small subunit. The relative frequencies of the three genotypes were consistent between long established sites in Spain and the UK. The frequencies of the different size classes were examined in the vegetatively reproducing populations recolonizing following the amelioration of sulphur dioxide air pollution. Populations of P. sulcata from long established sites where sulphur dioxide levels are known to have risen and then fallen contained two genotypes. Populations from recolonising sites where P. sulcata had previously been lost consisted of a single genotype (622 bp). This technology provides a powerful tool for testing hypotheses relating to the individuality and population structure of lichens, and has implications for lichen conservation. Key words: Lichenized-fungi, recolonization, rRNA, Parmelia sulcata, SO2, genotypes, conservation, introns. Introduction

The decrease in lichen populations as a response to air pollutants, especially sulphur dioxide, is well-docu-

mented (Ferry et al. 1973, Crespo et al. 1977, Nash and Wirth 1988, Hawksworth 1990, Richardson 1992). Subsequent recolonisation has occurred in areas where sulphur dioxide concentrations have decreased, often due to a combination of restrictions on industry and domestic emissions, and the cleaning up or decommissioning of power stations (Rose and Hawksworth 1981, Seaward and Letrouit-Galinou 1991, Gilbert 1992). In London, species not seen for 200 years have returned, advancing markedly since Battersea power station ceased emissions in 1983 (Hawksworth and McManus 1989a, 1989b). The reinvading leafy and shrubby lichens characteristically lack apothecia, and most are of species in which these sexual structures are rare. Gilbert (1971) suggested that an anomalous population of Parmelia saxatiIis in close proximity to a power station in Nottinghamshire could represent an ecotype resistant to higher levels of sulphur dioxide than the species normally tolerates. Subsequently, the concept of episodic selection in relation to stress from air pollution and other sudden environmental disturbances in fungi has been introduced (Seaward and Letrouit-Galinou 1991, Brasier 1995). The recolonization of a site by a particular lichen species has been used as a bioindicator, with the assumption that previous biodiversity levels are restored. However, it is not generally known if recolonization represents a return of the original population, or a reinvasion from a particular genotype. Parmelia sulcata is a suitable

178

A. Crespo et al.: Parmelia sulcata genotypes

model that can be used to examine genetic variation in populations from diverse sites as it is especially widespread and one of the most actively recolonizing leafy species. Molecular methods have been used to investigate phylogenetic relationships amongst lichen-forming fungi, but there are few instances in which molecular tools have been applied to investigations of lichen population biology or in relation to environmental and geographic parameters (DePriest 1993a). "Genetic fingerprinting" techniques are not readily applicable to many lichen-forming fungi, due to difficulties of separating the fungal from the algal or cyanobacterial partner(s), but markers such as introns have the potential to determine variability in populations (Gargas et al. 1995a, Gargas et al. 1995b, DePriest 1993b). We selected the rRNA gene cluster for study as this has been widely used to determine phylogenetic and systematic relationships among lichenized and other fungi (Gargas et al. 1995a). The gene cluster is multiply repeated throughout the fungal genome, and consists of highly conserved coding regions and variable inter- and intra-genic spacers. Many introns

have been characterised from the large and small ribosomal subunit genes of this gene cluster (DePriest 1993b, 1994; Gargas et al. 1995a, 1995b; Beard and DePriest 1996). Within the lichen-forming fungi, the ribosomal subunit genes have been used to determine phylogenetic relationships, and DePriest (1994) suggested that introns may provide a model for rapid evolution. Materials and methods Collection of specimens. Two hundred and twenty six specimens of separate lichen thalli were collected from a range of sites in England, Wales and Central Spain during 1995-6, and these were supplemented with 5 herbarium specimens (see Table 1). Within the UK, 176 specimens were collected from 16 sites for which the historical occurrence and distribution of lichen species over recent decades was known, and where information was available on changes in air quality (see e.g. Rose and Hawksworth 1981, Gilbert 1992). Sites were selected to include those where P sulcata had: (1) died out during periods of high sulphur dioxide levels and had subsequently recolonised (sites 1-8); (2) persisted through periods where sulphur

Table 1. Collection sites of Parmelia sulcata in UK where the history of lichen presence and patterns of SO2 pollution were known Site number

Locality

Grid reference1

Size of PCR product: 622bp

835bp

Recolonising sites 1 Middlesex: Buckingham Palace 2 Middlesex: East Acton 3 Essex: Epping Forest 4 Middlesex: Ruislip, Gmbb Ground 5 Derbyshire: Kilburn Toll 6 West Yorkshire: Leeds 7 South Yorkshire: Sheffield 8 Nottinghamshire: Thrumpton power station

TQ289796 TQ185812 TQ420970 TQ115865 SK377464 SE267417 SK325825 SK515320

5 5 16 15 5 13 8 1

Long-established sites with known periods of elevated SO2 9 Surrey: Bookham Commons 10 Middlesex: Ruislip Local Nature Reserve

TQ127565, TQl15865

7 142

4 2

Long established sites with no periods of elevated SO2 11 Devon: Ramsley 12 Hampshire: New Forest 13 Kent: Hamstreet area 14 Norfolk: Sparham 15 Powys: Ty Manor area 16 Suffolk: Snape 17 Sussex: Henfield

SX645940 SU215030 TR0033 TG075181 SN996569 & 986571 TM386590 TQ210148

122 12 93 122 7 9 8

3 2 2

580bp

1 2

1Ordnance Survey grid references; 2Samples in which one or more specimens had apothecia; 3Samples included herbarium specimens.

179

A. Crespo et al.: Parmelia sulcata genotypes Table 2. Collection sites of Parmelia sulcata in Spain

Site number

Locality

Grid reference I

Size of PCR product: 622bp

Long-established sites with known periods of elevated S O 2 1 Madrid: Casa de Campo 2 Madrid: Carretera de Castilla

30TVK37 30TVK37

6 1

Long established sites with no periods of elevated SO2 3 CA Madrid: Valdemorillo 4 CA Madrid: E1 Pat-do 5 CA Madrid: E1 Escorial 6 CA Madrid: Montejo de la Sierra 7 CA Castilla y Ledn: Burgos 8 CA Castilla-La Mancha: Albacete

30TVK08 30TVK38 30TVK49 30TVL64 30TVM01 30SWH46

8 11 4 1 1

835bp

580bp

14 3 1 2 12

12 1

1UTH grid references; 2Samples in which one or more specimens had apothecia.

dioxide levels had historically risen, eliminating many other lichens, and had subsequently fallen (sites 9-10); and (3) thrived in and had never been exposed to levels of sulphur dioxide sufficient to affect much more sensitive lichen species (sites 11-17). At one extensive site (site 4) the local geography was such that both long-established and recolonizing populations had been recorded and these were sampled separately as sites 4 and 10. Fifty-five specimens were collected in Spain from eight sites (Table 2), all of which comprised long-established populations. Two sites (sites 1 and 2) were in areas where SO2 levels had historically risen and subsequently fallen, and six (sites 38) were in areas where atmospheric SO2 levels had always been low. Five or more samples of physically separated lichen thalli were collected from areas where P. sulcata was wellestablished; at least five specimens were taken from different trees when available, and in most cases five more in close proximity on a single tree were also gathered. In areas where P. suIcata was recolonising, up to sixteen specimens were collected, wherever sufficient material was present. Specimens were collected on Quercus robur, Q. petraea, or Salix fragilis wherever possible in order to provide some consistency to the samples, but other trees were sampled when these were unavailable. Representative specimens are preserved in the biosystematic reference collections at CABI Bioscience. Single specimens were used from the 5 herbarium samples tested. DNA extraction and amplification. Total fungal DNA was extracted from the rhizines of individual thalli according to the CTAB extraction method described by Crespo et al. (1997a). The region of the rRNA gene cluster containing the immediate 3' terminus of the small subunit gene, the two internally transcribed spacers, the 5.8s subunit gene and the immediate 5' terminus of the large subunit gene was amplified with the primers ITS1F (Gardes and Bruns 1993) and ITS4 (White et al. 1990). Polymerase chain

reaction conditions and electrophoresis were as described elsewhere (Crespo et al. 1997a). Five double stranded PCR products were sequenced by the PRISM Ready Reaction DyeDeoxy terminator cycle sequencing kit (Applied Biosystems) with detection on a 373A stretch automatic sequencing apparatus (Applied Biosystems). The total fragment was sequenced, comprising the Y end of the small subunit gene, ITS 1, the 5.8s gene, ITS 2 and the 5' terminus of the large subunit gene.

Results

P C R products, DNA was extracted from all specimens and the rDNA target sequence was amplified by PCR. Amplification of the rDNA target sequence from the 231 samples of Parmelia sulcata gave a product of one of three sizes: 622 bp in 82% of the specimens, 835bp in 16%, and about 580bp in 2% (Table 1). Representative 622bp and 835 bp PCR products from U K and Spanish specimens were sequenced and their length difference was found to be due to a group I intron of 213 bp that was inserted at the 31 end of the small subunit (Fig. 1). The intron was located in the highly conserved stem loop at position 1516 (Escherichia coli nomenclature); this insertion site is so far only known from lecanoralean lichenized ascomycetes. There were five differences at three points between the introns from UK specimens and Spanish specimens (see Fig. 1). The ITS1F and ITS4 primers used in this study are based on conserved regions at the 5' terminus of the small subunit gene and the 31 terminus of the large subunit gene. Amplification products therefore include both of these regions, and the intron is located at the end of the small subunit

A. Crespo et al.: Parmelia sulcata genotypes

180 uUc C--G U--A P5 C - - G C--G

ACU C UAA~ AA C 5" G, .__I~A -splice, u •

AC-%u

G G cuA GGA A U CG_ U U A--U U A C--G U C--G G P2G_C C--GP2,1

site P, agg-- cC @oG--Cu

C C A G AG_cA P4

c--GG'C--uGC

G-- C

U--A

u --A CG--C 5"--g ° U C G A--UA

P3 U° G G P6 G G--CAACGCUAUC A A U U U--AGCUCCAGA A II1" III "1111111 C GCGGUAG UUGAGGUCU C C P8 A C A A A 3" splice site IIUcGCCG ACUAG C IIII. I I I I I / L~GGCGGUUGAGAUAUGAUC G[GC--]C--GAL --GA U UA[G0"]G~ga a c c u --3" U--A P7

C ~G

U--A G . U P9 G--C C--G U U G U U U CcC

5"-- g u a g g u A G C A • 1111" I" GUGgaacc--3-

L..

P10

P9.0

IGS intmn

SSU

LSU

Fig. 1. Putative secondary structure of the group I intron at position 1516 in the SSU rDNA of Parmelia sulcata from London (substitutions and deletions in the intron from Spanish specimens are shown as circled nucleotides). UPPER CASE letters indicate the primary sequence of the intron, and lower case letters the SSU rDNA flanking regions. Canonical base pairing is indicated by dashes and non-canonical pairings by dots. The internal guide sequence is shown in the box to the left. The lower figure shows a schematic representation of the location of the intron in the SSU rDNA. The two arrows indicate the positions of the oligonucleotide primers ITS 1F and ITS4 used for PCR. The names of the stem loops are according to the conventions of Burke et al. (1987). Additional revision of the folding of the intron was provided by Luc Jaeger and Eric Westhof (Strasbourg)

gene and not in the ITS region. Similar group I introns have been reported in this position in other lichenforming species, including Cladonia subtenuis, and have been described as "optional" in a population of Cladonia chlorphaea (DePriest 1994, Gargas et al. 1995b, Beard and DePriest 1996). Long-established populations. The relative frequencies of occurrence of the three products in longestablished populations from areas of clean air quality were consistent between the UK and Spain. In these

sites in the UK the relative frequencies of the 835, 622 and 580 bp products were 12.5%, 85% and 2.5% ( n = 8 1 ) , and from similar sites in Spain the frequencies were 13%, 81% and 6% ( n = 31). Although the UK and Spanish sample sizes were very different, the similarities in the relative frequencies would suggest that these may be representative of natural populations. The relative frequencies of each PCR product differed between sites where P. sulcata had been long-

A. Crespo et al.: Parmelia sulcata genotypes established and air quality was known to have decreased. The 580bp fragment was absent from all specimens at these sites and the frequencies of the 835 and 622 bp fragments were 22% and 78% (n = 27) for UK sites, and 71% and 29% (n = 24) for sites in Spain. Reeolonizing sites. Specimens collected from sites in the UK that were known to have been recently recolonized consistently gave a single PCR product of 622 bp (n = 68). No other PCR products were found in any of the samples.

Discussion The results obtained here show that three different rRNA genotypes occur within populations of Parmelia sulcata. This is the first report of an intron within this species, and is the first demonstration of molecular variation within populations of a leafy lichen at individual sites. Previous reports of introns within lichen-forming fungi have identified a number of insertion sites within the ribosomal small subunit gene, and the location of the intron in this study at the extreme 3t terminus of the gene is in accordance with these (DePriest 1993a, 1993b, 1994; Gargas et al. 1995a). Previous studies with other lichen-forming species have also suggested the occurrence of more than one rRNA genotype within a single specimen (DePriest 1993b, Gargas et al. 1995a), although this was only observed rarely in this study (Crespo et al. 1997b). The different frequencies with which each rRNA type was recovered at each site raise a number of questions. The populations from sites where P. suIcata had been long-established, and where air quality has always remained high, may; give an insight into the naturally occurring variation within the species. The frequencies of the different genotypes at these sites were remarkably similar between both Spain and the UK, where all three genotypes were detected at similar frequencies, despite the differences in sample sizes (81 UK and 31 Spain). There were, however, minor differences in the sequence of the introns between the UK and Spain (see Fig. 1), which was consistent between representative samples, and may be suggestive of some past divergence of the populations between the two countries. Considering samples from all sites, the 580bp fragment was not detected in specimens from sites where atmospheric SO2 levels had risen and subsequently recovered. The significance of this is unclear, mainly due to the relatively small number of samples available. Only 27 samples were obtained from sites in this category in the UK, and as the 580bp frag-

181 ment was found in 2.5% of samples from clean air long-established sites in the UK, it would only be expected in one in every 55 samples. Similarly, the changes in frequency of the other two fragments must be considered subjectively, as the sample number is too small for the reliable interpretation of frequency data. A very noticeable difference in the populations from different sites concerns the sites in the UK where P. sulcata had been lost due to poor air quality, and then had subsequently recolonized as the air quality improved (see Ferry et al. 1973, Rose and Hawksworth 1981, Hawksworth and McManus 1989a). In all of these sites only the 622 bp genotype was recovered. Although limited to the UK, 68 specimens were obtained from the eight sites, and if the longestablished clean air sites are taken as indicative of the natural population, then more than eight specimens would have been expected to be of the intron containing genotype. Sites in both the UK and Spain where air quality had decreased and the lichen had remained established showed a similar trend in that the slightly smaller 580bp genotype was absent. However, the relative frequencies of the remaining two genotypes were markedly different, with the 835 bp intron containing genotype predominating in Spain, and the 622 genotype predominating in the UK. Although the geographic distance between these regions is relatively large, there are several clear distinctions between the sites in the UK and Spain, including climatic, precipitation and altitude factors. It is not possible at this time to determine a precise reason for the discrepancy between populations, but one possible hypothesis could be that higher frequencies of the intron in Spain may represent an older population, although this is not supported by the results from the clean air regions. Further sampling is required from a range of habitats in the UK and elsewhere in Europe to identify whether the discrepancies in frequencies in this category are indicative of different mechanisms, or whether the two sets of figures represent two extremes of the overall variation in the population. The results presented here show that there is greater diversity at the long-established sites as compared to those of recent recolonization. Within the UK, samples obtained from areas where air quality had decreased showed a reduction in the number of genotypes as compared to the postulated natural population from long established clean air regions. While the number and specific geography of the sites selected make overall comparisons difficult, two important observations are: (1) the sequence differences between the Spanish and UK populations; and

182 (2) the absence of the intron containing fragment from the areas of recolonization. The occurrence of the intron in the UK and Spain may have arisen through independent events in both populations, or through a single event prior to the spread of the Fungal partner. The small sequence differences between the two populations would suggest the latter history. The reduced levels of diversity within the recolonizing sites could perhaps be expected if recolonization occurs through "founder" organisms, as is the case in many other groups (Barbault and Sastrapradja 1995). This suggestion has significant implications for lichen conservation and monitoring as transplanted and recolonizing populations may be representative of only a part of the genotypic variation of the original population at the site. As a result bioindicators recolonizing after an ameliorating environmental change may not be fully representative of the original population, and so do not represent a full recolonization, or a return to the original situation. Why the intron containing fragment is lost from recolonizing populations, as opposed to the standard fragment is unclear. Similar introns have been identified occurring along possible phylogenetic lineages and it is possible that the possession of the intron is an original characteristic that is vulnerable to subsequent change. This would be in general agreement with Brasier's episodic selection suggestion for fungal microevolution (Brasier 1995). The marker that we have used in this study is an intron in a multicopy functional gene. While this demonstrates that the new population is different from the original, it does not provide any further information on change within the population. As the new populations are generally spread by vegetative propagation, the potential to gain or recover previous levels of variability is limited. These results provide the first suggestion of linkages between molecularly distinguishable genotypes related to environmental parameters within a single lichen species, and demonstrate that PCR technology provides a powerful tool for testing hypotheses relating to the individuality, population structure, and ecology and biogeography of lichens at a more refined level than has been possible by other methods.

Acknowledgements This work was carded out at CABI Bioscience with the support of projects ANT 94-0905. AMB94-1209, APC-0020 and PB 93-1129-C0201 from the Ministerio de Educacidn y Ciencia, Spain. Lichen specimens were kindly collected according to protocols supplied by B. Benfield (Devon), A.

A. Crespo et al.: Parmelia sulcata genotypes R. Burgaz (Madrid), R. Carballal (Santiago de Compostela), J.M. Egea (Murcia), A. Henderson (Leeds), C.J.B. Hitch (Suffolk), RW. Lambley (Norfolk), EL. Nimis (Trieste), R. Noya (Madrid) and R.G. Woods (Powys). Other material was collected by A.C. or D.L.H.

References Barbault R., Sastrapradja S. (1995) Generation, maintenance and loss of biodiversity. In: Heywood V. H. (ed.) Global Biodiversity Assessment. Cambridge University Press, Cambridge, pp. 196-274. Beard K. H., DePriest R T. (1996) Genetic variation within and among mats of the reindeer lichen, Cladina subtenuis (des Abb.) Hale and W. Culb. Lichenologist 28: 171-182. Brasier C. M. (1995) Episodic selection as a force in fungal microevolution, with special reference to clonal speciation and hybrid introgression. Canad. J. Bot. 73: S1213S1221. Burke J. M., Belfort M., Cech T. R., Davies R. W., Schweyen R. J., Shub D. A., Szostak J. W., Tabak H. E (1987) Structural conventions for group I introns. Nucl. Acids Res. 15: 7217-7221. Crespo A., Bridge R D., Hawksworth D. L. (1997a) Amplification of fungal rDNA-ITS regions from nonfertile specimens of the lichen-forming genus Parmelia. Lichenologist 29: 275-282. Crespo A., Bridge R D., Cubero O. E, Hawksworth D. L. (1997b) Determination of genotypic variability in the lichen-forming fungus Parmelia sulcata. Bibliotheca Lichenologica 68: 73-79. Crespo A., Manrique E., Barreno E., Serina E. (1977) Valoracion de la contaminacion atmosferica del area urbana de Madrid mediante bioindicadores (liquenes epifitos). Anales Instituto Botanico A. J. Cavanilles 34: 71-94. DePriest R T. (1993a) Molecular innovations in lichen systematics: the use of ribosomal and intron nucleotide sequences in the Cladonia chlorophaea complex. The Bryologist 96: 314-325. DePriest R T. (1993b) Small subunit rDNA variation in a population of lichen fungi due to optional group I introns. Gene 134: 67-74. DePriest R T. (1994) Variation in the Cladonia chlorophaea complex II: ribosomal DNA variation in Southern Appalachian populations. The Bryologist 97: 117-126. Ferry B. W., Baddeley M. S., Hawksworth D. L. (1973) Air pollution and lichens. The Athlone Press, London. Gardes M., Brnns T. D. (1993) ITS primers with enhanced specificity for basidiomycetes - application to the identification of mycorrhizae and rusts. Molec. Ecol. 2: 113-118. Gargas A., DePriest R T., Taylor J. W. (1995a) Positions of multiple insertions in SSU rDNA of lichen-forming fungi. Molec. Biol. Evol. 12: 208-218. Gargas A., DePriest R T., Grube M., Tehler A. (1995b) Multiple origins of lichen symbioses in fungi suggested by r SSU rDNA phylogeny. Science 268: 1492-1495.

A. Crespo et al.: Parmelia sulcata genotypes Gilbert O. L. (1971) Studies along the edge of a lichen desert. Lichenologist 5:11-17. Gilbert O. L. (1992) Lichen reinvasion with declining air pollution. In: Bates J. W., Farmer A. M., (eds.) Bryophytes and lichens in changing environment. Clarendon Press, Oxford, pp. 159-177. Hawksworth D. L. (1990) The long-term effects of air pollutants on lichen communities in Europe and North america. In: Woodwell G. (ed.) The Earth in transition: patterns and processes of biotic impoverishment. Cambridge University Press, Cambridge, pp. 45-64. Hawksworth D. L., McManus R M. (1989a) Lichen recolonization in London under conditions of rapidly falling sulphur dioxide and the concept of zone skipping. Botanical Journal of the Linnean Society 100: 99-109. Hawksworth D. L., McManus E M. (1989b) Lichens that tell a tale. Country Life 183: 144-145. Nash T. H., Wirth V. (1988) Bibliotheca Lichenologica Lichens, bryophytes and air quality 30. J Cramer (eds.) Berlin. Richardson D. H. S. (1992) Pollution-monitoring with Lichens. Slough: Richmond Publishing.

183 Rose C. I., Hawksworth D. L. (1981) Lichen recolonization in London's cleaner air. Nature 289: 289-292. Seaward M. R. D., Letrouit-Galinou M. A. (1991) Lichen recolonization of trees in the Jardin du Luxembourg, Paris. Lichenologist 23: 181-186. White T. J., Bruns T. D., Lee S., Taylor J. (1990) Amplification and direct sequencing of fungal ribosomal DNA genes for phylogenetics. In: Innis M. A., Gelfand D. H., Sninsky J. J., White T. J. (eds.) PCR Protocols. Academic Press, New York, pp. 315-322.

Addresses of the authors: A. Crespo and O. E Cubero, Departamento de Biologfa Vegetal II, Universidad Complutense, E-28040 Madrid, Spain. R D. Bridge, Biosystematics and Molecular Biology, CABI Bioscience UK Centre, Bakeham Lane, Egham, Surrey TW20 9TY, UK. D. L. Hawksworth, MycoNova, 114 Finchley Lane Hendon, London, NW4 1DG, UK. M. Grube, Institut ftir Botanik, Karl-Franzens-Universit~it Graz, Holteigasse 6, A-8010 Graz, Austria.