American Journal of Botany 99(9): 1501–1512. 2012. R IBOSOMAL DNA DISTRIBUTION AND A GENUS-WIDE PHYLOGENY REVEAL PATTERNS OF CHROMOSOMAL EVOLUTION 1 IN A LSTROEMERIA (ALSTROEMERIACEAE) J ULIANA C HACÓN 2,4 , A RETUZA S OUSA 2 , C ARLOS M. BAEZA 3 , AND S USANNE S. RENNER 2 2 Systematic Botany and Mycology, University of Munich, 80638 Munich, Germany; and 3 Departamento de Botánica, Facultad de Ciencias Naturales y Oceanográfi cas, Universidad de Concepción, Casilla 160-C, Concepción, Chile • Premise of the study: Understanding the fl exibility of monocot genomes requires a phylogenetic framework, which so far is available for few of the ca. 2800 genera. Here we use a molecular tree for the South American genus Alstroemeria to place karyological information, including fl uorescent in situ hybridization (FISH) signals, in an explicit evolutionary context. • Methods: From a phylogeny based on plastid, nuclear, and mitochondrial sequences for most species of Alstroemeria , we se- lected early-branching (Chilean) and derived (Brazilian) species for which we obtained 18S-25S and 5S rDNA FISH signals; we also analyzed chromosome numbers, 1C-values, and telomere FISH signals (in two species). • Key results: Chromosome counts for Alstroemeria cf. rupestris and A. pulchella confi rm 2 n = 16 as typical of the genus, which now has chromosomes counted for 29 of its 78 species. The rDNA sites are polymorphic both among and within species, and interstitial telomeric sites in Alstroemeria cf. rupestris suggest chromosome fusion. • Conclusions: In spite of a constant chromosome number, closely related species of Alstroemeria differ drastically in their rDNA, indicating rapid increase, decrease, or translocations of these genes. Previously proposed Brazilian and Chilean karyo- type groups are not natural, and the n = 8 chromosomes in Alstroemeria compared to n = 9 in its sister genus Bomarea may result from a Robertsonian fusion. Key words: Chilean Alstroemeria ; Alstroemeriaceae; FISH; 18S-25S rDNA; 5S rDNA; interstitial telomeric sequences; primary chromosomal rearrangements. Several genomic features are distinctive in monocots com- in Alstroemeria are asymmetric and bimodal (ca. 15 species pared to dicots, including greater genome size variation and have been investigated; Stephens et al., 1993 ; Buitendijk and greater fl exibility in how DNA is organized into chromosomes Ramanna, 1996 ; Kamstra et al., 1997 ; Sanso and Hunziker, ( Leitch et al., 2010 ). A review of monocot genome characteris- 1998 ; Sanso, 2002 ; Jara-Seguel et al., 2004 ; Baeza et al., 2006 ; tics based on data for 534 of the ca. 2800 genera revealed that Baeza et al., 2010 ). The karyotypes of the few species of Bo- Liliales have a wide range of ploidy levels (up to 22 x ) and that marea , Drymophila , and Luzuriaga that have been studied also they rarely have small chromosomes and small genomes ( Leitch are asymmetric and bimodal ( Jara-Seguel et al., 2005 , 2010 ; et al., 2010 ). Cytogenetic data for the Liliales, however, are Baeza et al., 2008 ). All nine Bomarea species counted have n = sparse and uneven, and very few clades have been analyzed in 9, while Luzuriaga and Drymophila species have n = 10 (Ap- a phylogenetic context (e.g., Leitch et al., 2007 : Liliaceae). pendix S1). A summary of the karyotype characteristics of the Among the Liliales families that have fascinated cytogeneti- four genera is shown in Fig. 1 . cists for a long time are the Alstroemeriaceae, which consist of In spite of the apparently invariable chromosome number, the neotropical genera Bomarea , with 120 species, and Alstro- studies using molecular-cytogenetic techniques suggest a dy- emeria with 78; the disjunctly distributed Luzuriaga , with three namic picture of chromosome restructuring in Alstroemeria . species in Chile and one in New Zealand; and Drymophila , with For example, fl uorescence in situ hybridization (FISH) analy- one species in Australia and one in Tasmania. Strasburger ses in seven Chilean and Brazilian species revealed high levels (1882) studied male meiosis in A. chilensis , with n = 8, a num- of polymorphism in the ribosomal DNA (rDNA) signals of pre- ber since reported for all 27 species of Alstroemeria whose sumed homologous chromosomes ( Kamstra et al., 1997 ; Kuipers chromosomes have been counted (Appendix S1, see Supple- et al., 2002 ; Baeza et al., 2007 ). Likewise, C-banding and mea- mental Data with the online version of this article). Karyotypes surements of nuclear DNA content (2C value), PI/DAPI indi- ces, and chromosome arm lengths in 12 Brazilian and Chilean species (fi ve of them the same as studied with FISH) showed 1 Manuscript received 8 March 2012; revision accepted 6 August 2012. large differences in these parameters ( Buitendijk and Ramanna, The authors thank L. Aagesen, Instituto de Botánica Darwinion, 1996 ; Buitendijk et al., 1997 ; Kuipers et al., 2002 ; the PI/DAPI Argentina, and D. Rougier, Universidad Andrés Bello, Chile, for samples index refl ects differences in the AT/GC ratio: Barow and of Alstroemeria ; F. Alzate, Universidad de Antioquia, Colombia, for a Meister, 2002 ). sample of Bomarea patinii ; M. Silber for assistance with the FISH The aim of the current study is to infer directions of chromo- experiments; and editor Mark Simmons and two anonymous reviewers for somal evolution in Alstroemeria by studying rDNA FISH data comments. This project was funded by a grant from the Deutsche Forschungsgemeinschaft (DFG RE 603/10-1). in the light of a phylogeny. Specifi cally, we wanted to test 4 Author for correspondence (e-mail: [email protected]) whether Chilean and Brazilian “karyotype species groups” dis- tinguished in earlier studies ( Buitendijk et al., 1997 ; Jara- doi:10.3732/ajb.1200104 Seguel et al., 2004 ) refl ect evolutionary homology or are the result American Journal of Botany 99(9): 1501–1512, 2012; http://www.amjbot.org/ © 2012 Botanical Society of America 1501 1502 AMERICAN JOURNAL OF BOTANY [Vol. 99 Fig. 1. Molecular phylogeny of the Alstroemeriaceae (simplifi ed from Chacón et al., 2012 ) showing cytogenetic characteristics, such as the haploid chromosome number ( n ), the total haploid length of all chromosomes (THL in µm), the level of karyotype asymmetry, and karyotype morphology (bimodal = karyotypes comprising two size classes). Information was taken from Conran (1987) ; Sanso and Hunziker (1998) ; Sanso (2002) ; Baeza et al. (2007 , 2008 ), Palma-Rojas et al. (2007) ; and Jara-Seguel et al. (2010) . of parallel evolution. A division into eastern and western karyo- root tips were washed in distilled water, digested with 1% (w/v) cellulase Ono- type groups might be inferred from the presence of Alstroeme- zuka-RS (Serva, Heidelberg, Germany), 0.4% (w/v) pectolyase (Sigma-Aldrich, ria on both sides of Andes—44 of its 78 species occur in Brazil, St. Louis, Missouri, USA), and 0.4% (w/v) cytohelicase (Sigma-Aldrich, Mis- souri, USA) in citric buffer (10 mmol/L, pH 4.8) for 50 min at 37 ° C. The mer- 34 in Chile. Starting with a family-wide phylogeny ( Chacón istems were dissected and squashed in a drop of 45% acetic acid. Coverslips et al., 2012 ), we selected a subset of early-branching and de- were removed after freezing in dry ice, and preparations were then air-dried at rived Alstroemeria species for which FISH data were available room temperature. The best slides were selected using phase-contrast micros- ( Baeza et al., 2007 ), and we then undertook additional FISH copy and stored at 20 ° C prior to fl uorescence in situ hybridization (FISH) studies to study ribosomal DNA changes across the genus. experiments. Changes in rDNA can serve to individually characterize chromosomes and to compare them between populations, spe- DNA probes and FISH — The following probes were used in the FISH cies, or clades, an approach widely used since the introduction experiments: The 18S-5.8S-25S rDNA unit from Arabidopsis thaliana in plasmid pBSK+, labeled with digoxigenin-11-dUTP (Roche Diagnostics, of fl uorescence in situ hybridization ( Pinkel et al., 1986 ). Varia- Basel, Switzerland) using a nick translation mix; and the 349-bp fragment of tion in the number and distribution of FISH signals indicates the 5S rRNA gene from Beta vulgaris was inserted into pBSK+ ( Schmidt et al., genome reorganization ( Hasterok et al., 2006 ; Heslop-Harrison 1994 ), labeled with biotin-16-dUTP (Roche Diagnostics, Basel, Switzerland) and Schwarzacher, 2011 ), and when rDNA variation is ana- using PCR. Additionally, an Arabidopsis -like telomeric probe was amplifi ed by PCR according to Ijdo et al. (1991) using the oligomer primers (5 ′ -TTTAGGG-3 ′ )5 lyzed in a phylogenetic context, the direction of karyotypic ′ ′ change can be inferred. Many studies on fl owering plants have and (5 -CCCTAAA-3 )5 and labeled with digoxigenin-11-dUTP using nick translation. established the power of the method ( Adams et al., 2000 : Aloe ; Chromosome and probe denaturation, posthybridization washes, and detec- Ran et al., 2001 : Clivia ; Shan et al., 2003 : Boronia ; Weiss- tion were performed using the methods of Sousa et al. (in press). The hybrid- Schneeweiss et al., 2008 : Hypochaeris ; Garcia et al., 2007 : ization mixtures consisted of 50% (w/v water) formamide, 2 × saline sodium Artemisia ; Martínez et al., 2010 : Iris ; Fukushima et al., 2011 : citrate (SSC), 10% (w/v) dextran sulfate, and 100–200 ng of labeled probe. The hybridization mix was denatured at 75 ° C for 10 min and cooled for 10 min on Byblis ; Lan and Albert, 2011 : Paphiopedilum ; Catalán et al., ° 2012 : Brachypodium distachyon ). ice. The slides and hybridization mix were denatured for 5 min at 75 C and hybridized for up to 20 h at 37 ° C. For digoxigenin and biotin detection, slides were incubated in blocking buffer (2% BSA in 2 × SSC) for 30 min at 37 ° C, followed by incubation (1 h, 37 ° C) with either antiDIG-FITC conjugate (Roche Diagnostics) to detect digoxigenin or streptavidin-Cy3 conjugate (Sigma- MATERIALS AND METHODS Aldrich) to detect biotin .