Comparison of Puccinia Acroptili from Eurasia and the USA

465 Comparison of Puccinia acroptili from Eurasia and the USA

W.L. Bruckart III, F.M. Eskandari, D.K. Berner, and M.C. Aime

Abstract: A rust disease caused by Puccinia acroptili P. Syd. & Syd. occurs throughout the range of Russian knapweed (Rhaponticum (syn. Acroptilon) repens), including North America. Differences in teliospore dimensions had been observed previously, but not quantified, between a specimen of P. acroptili from Turkey and one from the United States of America (USA). Similar differences were found during a recent evaluation of P. acroptili for biological control of R. repens in the USA; e.g., teliospores from eight USA isolates were 2.4 µm shorter (P = 0.05) than four isolates from Eurasia (two from Turkey, and one each from Russia and Kazakhstan). This inspired the generation of additional biological and DNA sequence data to clarify the significance of these differences between USA and Eurasian isolates. Although the USA isolates were found to have shorter teliospores, as noted in the description by Savile, teliospores of the USA isolates were also significantly wider in diameter than isolates from Eurasia; the latter in contrast to Savile’s observation. Biologically, the isolates were the same; all were equally aggressive in causing disease under common greenhouse test conditions, and fertile crosses occurred between isolates regardless of source. DNA sequence analyses of the nuclear rDNA large subunit and internal transcribed spacer regions supported the notion that P. acroptili is a single species, but it also revealed that small levels of variation may exist within the species. For these reasons, it has been concluded that USA and Eurasian accessions can be considered a single species, i.e., P. acroptili. Also, Savile’s observation about differences in spore dimensions is substantiated. Key words: Asteraceae, biological control, plant pathogen, Pucciniales, Uredinales, weeds. Résumé : On retrouve une rouille pathogène causée par le Puccinia acroptili P. Syd. & Syd. sur l’ensemble de l’aire de la centaurée russe (Rhaponticum (syn. Acroptilon) repens), l’Amérique du Nord inclue. On a déjà observé des différences dans les dimensions des téliospores, mais sans les quantifier, entre des spécimens du P. acroptili de Turquie et un des États-Unis. On a également trouvé des différences similaires au cours d’une récente évaluation du P. acroptili, pour la lutte biologique du R. repens aux États-Unis; p. ex., les téliospores de 8 souches américains mesurent 2,4 µm de moins (P = 0,05) compara- tivement à 4 souches d’Eurasie (deux de Turquie, et une chacun pour la Russie et Kazakhstan). Ceci a motivé la recherche de nouvelles données biologiques et moléculaires pour clarifier la signification de ces différences entre les souches améri- For personal use only. cains et eurasiens. Bien qu’on retrouve des téliospores plus courtes chez les souches américains, tel que noté dans la description de Savile, les téliospores d’origine américaine sont également significativement plus larges en diamètre que les souches eurasiens – ces dernières étant différentes des observations de Savile. Biologiquement les souches sont semblables; toutes montrent la même agressivité à générer la maladie sous des conditions normales en serres, et on obtient des croise- ments fertiles entre les souches quelles que soient les sources. Les analyses de séquences ADN de la grande sous-unité du rADN nucléique et des régions des espaceurs internes transcrits, supportent la notion que le P. acroptili ne constitue qu’une seule espèce, mais on constate également que l’on peut retrouver une variation à petite échelle chez cette espèce. Pour ces raisons, on conclut que l’on peut considérer les accessions américaines et eurasiennes comme une seule espèce, c.-à-d. P. acroptili. On supporte de plus les différences observées par Savile.

Botany Downloaded from www.nrcresearchpress.com by USDA on 07/05/12 Mots‐clés : Asteraceae, la lutte biologique, phytopathogène, Pucciniales, Uredinales, mauvaises herbes. [Traduit par la Rédaction]

Received 31 January 2012. Accepted 12 February 2012. Published at www.nrcresearchpress.com/cjb on 31 May 2012. W.L. Bruckart III, F.M. Eskandari, and D.K. Berner. United States Department of Agriculture, Agricultural Research Service, Foreign Disease-Weed Science Research Unit, 1301 Ditto Avenue, Ft. Detrick, MD 21702, USA. M.C. Aime.* United States Department of Agriculture, Agricultural Research Service, Systematic Botany and Mycology Laboratory, 10300 Baltimore Avenue, Beltsville, MD 20705, USA. Corresponding author: William L. Bruckart III (e-mail: [email protected]). Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agri- culture and does not imply its approval to the exclusion of other products that also may be suitable. *Present address: Department of Plant Pathology and Crop Physiology, Louisiana State University Agricultural Center, Baton Rouge, LA 70803, USA.

Botany 90: 465–471 (2012) doi:10.1139/B2012-020 Published by NRC Research Press 466 Botany, Vol. 90, 2012

Introduction Measurements Urediniospores and teliospores were measured using a Ni- Puccinia acroptili P. Syd. & Syd. (Pucciniales, Basidiomy- kon Eclipse 80i microscope equipped with a DS-L1 camera cota) is the causal agent of a rust disease of Russian knap- and a flat screen monitor. Software enabled on-monitor weed, Rhaponticum repens (L.) Hidalgo (syn. Acroptilon measurement of objects in the field of view, calibrated for repens (L.) DC.), Asteraceae, tribe Cardueae. Russian knap- each level of magnification. Spores were mounted on slides weed, which is native to Mongolia, western Turkestan, Iran, with a drop of lactophenol containing 1% aniline blue, and Turkish Armenia, and Asia Minor (Moore and Frankton each slide was heated gently until spores were turgid (Savile 1974), is diseased by P. acroptili most likely wherever the 1970a). Urediniospores were measured for length and width. host occurs, including western North America (Bruckart et Teliospores were measured for length, width, and “waist”, al. 2006; Cummins 1978; Dugan and Carris 1992; Mortensen i.e., the diameter of the wall between the cells. and Molloy 1989; Palm and Vesper 1991; and Savile 1970b). Data were analyzed using Proc GLM in SAS (SAS Insti- Although diseased plants are common in North America, the tute, Cary, N.C., USA). Probabilities for differences between effect of P. acroptili on density of Russian knapweed has not least-squares means were generated using the PDIFF option, been tested. Considering the abundance of R. repens in North and means were considered significantly different only if P ≤ America, it is apparent that P. acroptili is not causing reduc- 0.05. Confidence intervals (CI, P = 0.05) were also gener- tion in densities of R. repens (Mortensen and Molloy 1989; ated for certain mean values. SAS Proc CORR was used to Mortensen et al. 1991). For this reason, evaluation of foreign determine if there was correlation between teliospore width acquisitions of P. acroptili on USA Russian knapweed was and length, and Proc GLM was used to generate a model to made to determine if more aggressive isolates might be available for biological control. describe that relationship. Variability within P. acroptili was noted earlier by Savile (1970b), who reported that teliospore dimensions of two Inoculations, biological measures, and crosses isolates used in his description were different, i.e., the sin- All inoculations involved Russian knapweed plants grown gle Turkish specimen had both longer and wider teliospores from seed collected near Galatea, Colorado, USA. Plants than those of the single specimen from California, USA (a were started in vermiculite and transplanted at the cotyledo- third specimen did not have teliospores). As a consequence, nary stage into an artificial greenhouse soil mix in 10-cm Savile postulated that the rust in the USA resulted from a clay pots. Seedlings and young plants were monitored for single introduction with a limited gene pool. During the fungus gnats and shore flies, in particular, to eliminate vec- present evaluation, differences in spore dimension similar tors of spermatia in crossing experiments. to those noted by Savile (1970b) were found between USA Plants free of insects were inoculated with teliospores of isolates and those from Eurasia, i.e., from Turkey, Russia, selected isolates 4–6 week after transplanting and placed into and Kazakhstan. Thus there was opportunity for a more an insect-exclusion cage during incubation, i.e., prior to de-

For personal use only. thorough investigation into the significance of differences velopment of fungal structures. Protocol for inoculation with reported by Savile, and additional data were collected to teliospores has been described (Bruckart et al. 2010). Both determine if isolates from Eurasia and the USA were the spermagonia and aecium-like pustules (ALP) developed fol- same or not. Fungal biology, aggressiveness, teliospore lowing foliar inoculations by teliospores (Bruckart et al. morphology, and DNA sequence data, were analyzed for 2010). Locations of spermagonia that developed were re- these comparisons. The objectives of this study were to corded on hand-drawn maps of leaves. Counts were made of compare USA and Eurasian isolates for similarity and dis- spermagonia and aecia, and data on the number of sori per tinction, and to determine if differences might be important leaf for each isolate were used in comparisons. Statistical in considerations concerning the Russian knapweed rust for analysis was made using PROC GLM, least squares means, biological control. and calculated probabilities, as described, in SAS.

Botany Downloaded from www.nrcresearchpress.com by USDA on 07/05/12 For crosses, spermatia from a single spermagonium were Materials and methods suspended in sterile distilled water, and a 1 µL drop of suspen- sion was applied to as many available spermatia as possible Isolates (Bruckart et al. 2010). The isolate that was the donor source Thirteen isolates of Eurasian and USA origin were in- of spermatia in each cross was classified as the male compo- cluded in this study (Table 1; Bruckart et al. 2010). Inoculum nent. Six isolates, representative of both USA and Eurasian ac- was received from the various locations as viable uredinio- cessions, were involved in attempted reciprocal crossings in this spores on dried leaf material, except for two, nonviable New study. Use of spermatia from a single spermagonium provided Mexican herbarium specimens that were used solely for DNA opportunity to evaluate efficiency or success rate by direction analysis. Material from original shipments was stored in an of the cross. Record was made of attempted crosses, and aecia ultra-cold (–80 °C) freezer, and the remainder of each acces- that developed were noted on hand-drawn leaf maps and sion was used for artificial inoculation to increase spores. counted as evidence of a successful cross. The proportion of Urediniospores from greenhouse inoculations were stored ei- successful crosses was calculated for each isolate, and data ther frozen (–80 °C) or under refrigeration (4 °C). All telio- were transformed into logits, i.e., y =ln(p /1– p), where y is spores were stored on dried leaf material either at room the natural log of the probability of a successful cross and p is temperature or under refrigeration. Voucher material for each the proportion of successful crosses in each replicate. The accession was deposited in the U.S. National Fungus Collec- transformed data were analyzed using Proc LOGISTIC in SAS tions, BPI (Table 1). and c2 tests for significance of individual events and contrasts.

Published by NRC Research Press Bruckart III et al. 467

DNA extraction and sequencing Methods for DNA extraction and polymerase chain reac- tions (PCR) follow those described by Aime (2006). Ap- proximately 1400 bp from the region of ribosomal repeat spanning the 5.8S subunit, the internal transcribed spacer region 2 (ITS-2), and the large subunit (28S), was amplified and sequenced with rust-specific primer Rust2inv (Aime This study 2006) and LR6 (Vilgalys and Hester 1990) as described in Aime (2006). Sequence data have been submitted to Gen- Bank (Table 1). * Automated DNA sequencing chromatograms were assembled and edited in Sequencher version 4.1.4 (Gene Codes Corp., Ann Arbor, Mich., USA). Newly generated sequences JN204185 This study DQ354517 Aime 2006 JN204186 JN204187 This study JN204194 This study JN204193 This study JN204190 This study JN204191 This study JN204192 This study JN204189 This study JN204196 This study JN204195JN204182JN204183 This study This study This study JN204184 This study AY787782 Deadman et al. 2005 were confirmed by BlastN analysis (http://www.ncbi.nlm.nih. gov/), and a data set of previously published (Aime 2006) and newly generated sequences was assembled from closely related Puccinia species from the Asteraceae (Table 1). Se- quences were aligned in MAFFT version 6.5 using Q-INS-I, which accounts for secondary structure of RNA (Katoh and Toh 2008), and assessed visually in Se-Al v2.0a11 (Andrew Rambaut, Dept. Zoology, University of Oxford, UK; http:// evolve.zoo.ox.ac.uk/). Maximum likelihood (ML) analyses were conducted in RAxML-HPC2 version 7.2.7 (Stamatakis Rhaponticum repens Sonchus oleraceus Rhaponticum repens Rhaponticum repens Rhaponticum repens Rhaponticum repens Rhaponticum repens Rhaponticum repens Rhaponticum repens Rhaponticum repens Rhaponticum repens Rhaponticum repens Balsamorhiza sagittata Centaurea calcitrapa Carduus nutans Carthamus tinctorius 2006) via the CIPRES portal (Miller et al. 2011) using de- fault parameters adjusting for 1000 bootstrapping replicates. Maximum parsimony (MP) analyses were conducted in Paup* version 4.0b10 (Swofford 2002) as heuristic searches with 1000 random addition replicates and TBR branch swap- ping. Support for branches was evaluated by bootstrap analysis derived from 1000 MP replicates with 10 random addition replicates each.

Results For personal use only. Teliospores from the USA, generated on the same plant accession and in the same greenhouse environment as the Eur- asian samples, were significantly shorter but wider than those from Eurasia (Table 2). Teliospores of USA accessions were shorter by a mean of 2.5 mm, i.e., 37.3 mm for six USA isolates, versus 39.9 mm for two Turkish isolates (01-595 and 02-048), 39.7 mm for the isolate from Southern Russia (05- m 05-085 Russia U-203 / BPI 86350901-595 (U-224) / BPI 863523 Turkey Turkey undetr. Asteraceae JN204188 This study 05-055 (U-735) USA: MT 05-070 (U-577) / BPI 1107952 USA: NM 05-051 (U-734) USA: CO 05-044 (U-692) USA: CO 05-045 (U-693) USA: CO 05-069 (U-578) / BPI 1110177 USA: NM 90-114 (U-733) Kazakhstan 05-056 (U-736) USA: MT 085), and 39.3 m for the isolate from Kazakhstan (90-114). Although shorter, teliospores of USA isolates were also

Botany Downloaded from www.nrcresearchpress.com by USDA on 07/05/12 slightly, and significantly, wider by a mean of 0.3 mm in diameter and 0.5 mm at the waist (Table 2). There was a significant correlation of –0.236 (P < 0.01) between teliospore length and width for the data set. Urediniospores from the USA and Eurasia also differed in dimension. Those from

(DC.) Cummins U-209 / BPI 863513Eurasia Turkey were significantly larger by 0.8 µm for both width and length (Table 2). Amount of disease, or aggressiveness, as measured by the number of sori per leaf that developed after inoculation by centaurea Peck U-354 / BPI 878003 USA: ID (Mont.) Jørst. U-63 / BPI 842230 USA: CA teliospores, was considerable in this study, i.e., nearly 10 var. Jackysori (>4.5 spermatia U-218 / BPI 863518 and >5.0 Turkey ALP) per leaf (Table 3). The Corda U-323 / BPI 863557 Oman P. Syd. & Syd. 02-048 (U-135) / BPI 877990 Turkey difference in disease caused by Eurasian and USA isolates was not statistically significant on this basis, whether for the number of spermatia or the number of ALP (Table 3). Nei- Collections used in phylogenetic analysis, locality, host, GenBank numbers, and origin.

Details about each isolate can be found in Bruckart et al. 2010. ther was there any difference in symptomatology associated with inoculations by isolates of either group. Note: *An additional sequence of the entire transcribed spacer region for this isolate has beenSuccessful deposited in GenBank as HQ700340. mating occurred for every combination of iso- Puccinia acroptili SpeciesMiyagia pseudosphaeria Collection/voucher No. Locality Host Genbank No. Origin Puccinia acroptili Puccinia acroptili Puccinia acroptili Puccinia acroptili Puccinia acroptili Puccinia acroptili Puccinia acroptili Puccinia acroptili Puccinia acroptili Puccinia acroptili Puccinia acroptili Puccinia balsamorrhizae Puccinia calcitrapae Puccinia carduorum Puccinia carthami Table 1. lates tested (Table 4), including selfs and reciprocal crosses

Published by NRC Research Press 468 Botany, Vol. 90, 2012

Table 2. Least-squares means of dimensions, in micrometers, for Puccinia acroptili teliospores and urediniospores examined in this study.

Teliospore Urediniospore Source*nLength Width Waist n Length Width Eurasia 400 39.7 22.9 21.6 300 23.6 21.4 United States 800 37.3 23.2 22.1 725 22.8 20.6 Note: Spore dimensions in each column are statistically significantly different (P ≤ 0.05). *Eurasian isolates (country): 90-114 (Kazakhstan), 01-595 (Turkey), 02-048 (Turkey), and 05-085 (Russia). United States isolates (state): 05-051, 05-052, 05-053, and 05-054 (Color- ado); 05-055 (Montana); 05-056 (Wyoming); and 05-069 and 05-070 (New Mexico).

Table 3. Least-squares means of the number of spermatia and aecium-like pustules (ALP) per leaf following teliospore inoculations with selected isolates of Puccinia acroptili from Eurasia or the United States of America (USA).

Isolate Number Source Location n Spermatia ALP 02-048 Turkey Eurasia 30 5.3 4.6 05-085 Russia Eurasia 59 4.8 7.2 90-114 Kazakhstan Eurasia 25 4.0 4.8 Mean, Eurasia* 4.8 5.9 05-055 Montana USA 57 4.5 4.3 05-056 Wyoming USA 31 4.4 4.7 06-037 Colorado USA 34 4.8 7.9 Mean, USA* 4.6 5.4 Eurasia vs. USA; P = 0.6370 0.6413 *Least squares means for all isolates from Eurasia or the USA, respectively.

Table 4. Percentage success for crosses within and between isolates of Puccinia acroptili from different source locations.

Isolate Female*

For personal use only. Source No. 02-048 05-085 90-114 05-055 05-056 06-037 Male Turkey 02-048 15 30 30 60 17 17 Russia 05-085 42 48 31 48 (2) 21 nt Kazakhstan 90-114 43 16 33 18 30 nt Montana 05-055 28 33 nt 30 50 nt Wyoming 05-056 20 11 38 22 26 20 Colorado 06-037 nt nt nt nt 21 42 Note: Success defined as spermagonia that developed aecia. nt, not tested (cross not made). *Isolates as females, i.e., recipients of spermatia.

Botany Downloaded from www.nrcresearchpress.com by USDA on 07/05/12 between USA and Eurasian isolates. The mean rate of suc- second, more variable subclade included all the USA isolates cessful crosses was 30.4% (±4.5; CI, P = 0.05). Analysis of and the isolate from Kazakhstan. Within the USA/Kazakh- crossing data indicated possible differences by individual iso- stan subclade there exist two single nucleotide polymor- late in suitability as a parent (Table 5), but there were no phisms (SNPs) and one indel that are not fixed. Between the clear effects associated with isolate and the direction of two P. acroptili subclades, there exist two SNPs in the ITS-2 crosses (data not shown). The near-zero logit value for the region, two SNPs within the LSU, and one 4-bp insert, also “exotic vs. domestic” contrast (Table 5) suggests that, in gen- in the ITS-2 region; all appear fixed. The two subclades were eral, USA and Eurasian isolates tested in this study are equiv- not strongly supported by bootstrapping value in MP analy- alent in terms of parental efficiency. ses, nor was reciprocal monophyly of the subclades sup- Comparison of P. acroptili DNA sequences, i.e., the 5.8S ported by ML analyses. subunit, ITS-2, and the 28S large subunit, resulted in clear separation of P. acroptili from species from related host Discussion plants in the tribe Cardueae (Asteraceae), represented by Puccinia carduorum and Puccinia carthami (Fig. 1; MP tree Rhaponticum rust is in the Puccinia centaureae – laschii length 26, CI = 0.962, RI = 0.963). Within P. acroptili were lineage, a “complex that embraces most of the brachycyclic two subclades of isolates (Fig. 1). The Russian and Turkish rusts on Cardueae” (Savile 1970a). These fungi are autoe- isolates, constituting one of these, were nearly identical, i.e., cious, in that they complete their life cycle (all five stages) there were no sequence differences detected among them. A on a single host (Bruckart et al. 2010), and they are distin-

Published by NRC Research Press Bruckart III et al. 469

Table 5. Logit values, differences in logit values, and probabilities some results from the present study did not fully support ob- 2 of greater c values, from logistic analyses of proportion of Rha- servations in the earlier report. ponticum repens plants developing aecia after crosses between iso- Although differences in spore dimensions between Eura- lates of Puccinia acroptili. sian and USA isolates were clear and consistent, they were also within expected limits reported for the species (Bruckart Success from crosses* et al. 2006; Cummins 1978; Dugan and Carris 1992; Morten- † ‡ Isolate Estimate Pr > c2 Proportion§ sen and Molloy 1989; Palm and Vesper 1991; Savile 1970b; 02-048; TU, Eurasian –1.21 <0.01 0.229 Sydow and Sydow 1904; Wei and Wang 1986). Considering 05-085; RU, Eurasian –0.63 <0.01 0.348 that there was a negative correlation between teliospore width 90-114; KZ, Eurasian –0.69 <0.01 0.334 and length, it is conceivable that teliospores from the various 05-055; MT, USA –0.76 <0.01 0.319 isolates could actually be the same volume, a metric not pos- 05-056; WY, USA –1.16 <0.01 0.239 sible to measure in this study. – 06-037; CO, USA 0.59 <0.01 0.357 The ability to achieve artificial infections from teliospore Contrast|| Estimate Pr > c2 inoculations and to make artificial crosses with P. acroptili 02-048 vs. others –2.22 <0.01 (Bruckart et al. 2010) enabled the study of isolate aggressive- 05-085 vs. others 1.27 0.11 ness and fecundity, including reciprocal crosses between 90-114 vs. others 0.89 0.48 USA and Eurasian accessions. Results of self-crosses have 05-055 vs. others 0.50 0.64 been described (Bruckart et al. 2010). Biological and taxo- 05-056 vs. others –1.93 0.04 nomic similarities between Eurasian and USA isolates were 06-037 vs. others 1.48 0.09 substantiated by success with crosses between isolates from Exotic vs. USA –0.02 0.97 each group. The rate of success for crosses between USA TU & RU vs. USA –0.08 0.55 and Eurasian isolates was not different than rates observed ei- TU & RU vs. KZ –0.23 0.38 ther for selfs or for reciprocal crosses between isolates from KZ vs. USA 0.14 0.58 the same geographic source; all were compatible with the *Formation of aecia after crossing attempt. others. †Foreign Disease-Weed Science Research Unit isolate numbers; Eurasian, Evidence from comparisons of DNA sequences, particu- isolates from Eurasia; USA, isolates from the United States of America; larly considering high values for ML and MP, is that P. acrop- TU, Turkish isolate; RU, Russian isolate; KZ, Kazakhstani isolate; MT, tili is a distinct species, clearly separated from closely related Montana isolate; WY, Wyoming isolate; CO, Colorado isolate. ‡ rusts of plant species in the same tribe as Rhaponticum. Logit value. §Ratio of success to attempts for each isolate as a parent. Within P. acroptili, even considering the limited number of ||Contrasts are the left, or first, element in the statement minus the latter, isolates and limited sequence size in the analysis, there were i.e., the estimate refers to the entry at the left side of the contrast in the two distinct groups. One included the Russian and Turkish comparison. isolates that were collected from relatively close geographical For personal use only. locations; they were identical in this molecular analysis. The guished by aecia that are morphologically similar to the ure- other group included the isolate from Kazakhstan and all the dinia (Kirk et al. 2001). Although P. acroptili is morpholog- USA isolates, and minor variations were found, not just when ically very similar to rust fungi from species of Carduus, compared with the other group, but also from within. The Carthamus, Centaurea,andCirsium (Savile 1970a,1970b), isolate from Kazakhstan was collected >2500 km to the east evidence has been published that it is distinct from these of the other Eurasian isolates, and although it had a molecu- other rust fungi on the basis of morphology (Savile 1970b), lar profile that was distinct and much more similar to that of host range (Mortensen et al. 1991), and polypeptides (Kim the USA isolates, teliospores were similar in size to those and Mortensen 1986). Analysis of DNA sequence data in from the other Eurasian isolates.

Botany Downloaded from www.nrcresearchpress.com by USDA on 07/05/12 the present study supports this. Despite differences reported in his description of P. acrop- However, differences in teliospore size were noted within tili, Savile (1970b) did not suggest that these isolates were P. acroptili (Savile 1970b), and data from this study confirm different species. Neither do results from this study, even this. Savile (1970b) observed in his description that, “… te- though some variation both in morphology and sequence liospores of the Turkish specimen have larger maximum was found. Therefore, we conclude that USA and Eurasian lengths and widths than that from California”. Although Eur- accessions can be considered a single species, i.e., P. acrop- asian teliospores were longer in the present study, as Savile tili. There are implications in terms of P. acroptili for biolog- observed, they were also significantly more narrow, both in ical control of Russian knapweed because of these findings. diameter and diameter at the waist. These differences are in Most importantly, the isolates did not differ in aggressive- contrast with Savile’s observation. ness; they caused the same amount of disease on the collec- Savile indicated also that urediniospore dimensions of the tion of Russian knapweed used in this study. For this reason, three specimens, “agree(d) closely”. Results from this study additional collecting for candidate biological control isolates differed from Savile’s findings in that urediniospores of USA is not justified for this organism. The importance of variation isolates were smaller, both in width and length, compared between isolates as candidates for biological control would with those from Eurasia. There is a clear difference between only become evident from tests involving additional acces- USA and Eurasian isolates based upon spore dimension, as sions of R. repens, the target plant, and only if differential re- Savile (1970b) reported, although with more samples and sponses occur among isolates on new host accessions. Such tests run under artificial, uniform greenhouse conditions, was not the focus of this investigation, and there has been

Published by NRC Research Press 470 Botany, Vol. 90, 2012

Fig. 1. Phylogenetic tree based on maximum likelihood analyses of ITS-2 and LSU sequences of Puccinia acroptili isolates and closely related Puccinia species showing species. Origins of sequences are provided in Table 1. Miyagia pseudosphaera was selected as the outgroup. Numbers above branches indicate bootstrapping support (1000 replicates) for each node as maximum likelihood/maximum parsimony. For personal use only. Botany Downloaded from www.nrcresearchpress.com by USDA on 07/05/12

no suggestion in the literature that differences within R. re- zakhstan), M.C. Becktell and D. Bean (Mesa State College, pens exist. Grand Junction, and Colorado Dept. Agr., Palisade, Colo., respectively) and Jeff Littlefield (Montana State University, Bo- Acknowledgements zeman) for USA isolates, and the curator of BPI for loan of The authors thank Jami Michael and Michael McMahon New Mexican herbarium specimens. Seeds of Russian knap- for DNA extractions and molecular characterizations at the weed were supplied by George Beck (Colorado State Univer- Foreign Disease-Weed Science Research Unit (FDWSRU), sity, Ft. Collins, Colo., USA). Appreciation is given to help William Schneider (USDA, ARS, FDWSRU) for advice on from Hye-young Yun (USDA, ARS, Systematic Mycology DNA sequence analysis, D.C. Sands (Montana State Univer- and Microbiology Laboratory, Beltsville, Md., USA) for sity, Bozeman, Mont., USA) for the isolate from Russia (Ka- translation of the Wei and Wang (1986) paper. Support for

Published by NRC Research Press Bruckart III et al. 471

the phylogenetic study was provided to M.C.A. by a USDA, Chan, L., Hoover, P., and Warnow, T. 2011. The CIPRES Portals. Animal and Plant Health Inspection Serivce cooperative Available from http://www.phylo.org/sub_sections/portal. [ac- agreement and the Louisiana Board of Regents. cessed 4 April 2011]. Moore, R.J. and Frankton, C. 1974. The thistles of Canada. Can. References Dept. Agric. Monogr. No. 10. Mortensen, K., and Molloy, M.M. 1989. Fungi detected on Aime, M.C. 2006. Toward resolving family-level relationships in rust Acroptilon repens (Russian knapweed) during surveys from 1981 fungi (Uredinales). Mycoscience, 47(3): 112–122. doi:10.1007/ to 1988. Can. Plant Dis. Surv. 69: 143–145. s10267-006-0281-0. Mortensen, K., Harris, P., and Kim, W.K. 1991. Host ranges of Bruckart, W.L., III, Eskandari, F.M., Becktell, M.C., Bean, D., Puccinia jaceae, P. centaureae, and P. acroptili, and P. carthami, Littlefield, J., Pilgeram, A.L., Sands, D.C., and Aime, M.C. 2006. and the potential value of P. jaceae as a biological control agent for Puccinia acroptili on Russian Knapweed in Colorado, Montana, diffuse knapweed (Centaurea diffusa) in North America. Can. J. and Wyoming. Plant Dis. 90(7): 971. doi:10.1094/PD-90-0971C. – Bruckart, W.L., III, Eskandari, F.M., Berner, D.K., and Aime, M.C. Plant Pathol. 13(1): 71 80. doi:10.1080/07060669109500968. 2010. Life cycle of Puccinia acroptili on Rhaponticum (= Palm, M.E., and Vesper, S.G. 1991. Russian knapweed rust caused by Acroptilon) repens. Mycologia, 102(1): 62–68. doi:10.3852/08- Puccinia acroptili in New Mexico. Plant Dis. 75(10): 1075. 215. PMID:20120230. doi:10.1094/PD-75-1075E. Cummins, G.B. 1978. Rust fungi on legumes and composites in Savile, D.B.O. 1970a. Some Eurasian Puccinia species attacking – North America. University of Arizona Press, Tucson. Cardueae. Can. J. Bot. 48(9): 1553 1566. doi:10.1139/b70-232. Deadman, M.L., Sadi, A.M.A., Jahdhami, S.A., and Aime, M.C. 2005. Savile,D.B.O.1970b. Autoecious Puccinia species attacking – First report of rust caused by Puccinia carthami on safflower in Cardueae in North America. Can. J. Bot. 48(9): 1567 1584. Oman. Plant Dis. 89(2): 208. doi:10.1094/PD-89-0208C. doi:10.1139/b70-233. Dugan, F.M., and Carris, L.M. 1992. Puccinia jaceae var. diffusa and Stamatakis, A. 2006. RAxML-VI-HPC: maximum likelihood-based P. acroptili on Knapweeds in Washington. Plant Dis. 76(9): 972. phylogenetic analyses with thousands of taxa and mixed models. doi:10.1094/PD-76-0972E. Bioinformatics, 22(21): 2688–2690. doi:10.1093/bioinformatics/ Katoh, K., and Toh, H. 2008. Improved accuracy of multiple ncRNA btl446. PMID:16928733. alignment by incorporating structural information into a MAFFT- Swofford, D.L. 2002. PAUP*: Phylogenetic Analysis Using Parsi- based framework. BMC Bioinformatics, 9(1): 212. doi:10.1186/ mony (*and other methods). Version 4. Sinauer, Sunderland, 1471-2105-9-212. PMID:18439255. Mass. Available from http://paup.csit.fsu.edu/about.html. Kim, W.K., and Mortensen, K. 1986. Differentiation of Puccinia Sydow, P., and Sydow, H. 1904. Monographia Uredinearum. Vol. 1. jaceae, P. centaureae, P. acroptili, and P. carthami by twodimen- Lipsiae: Fratres Borntraeger. sional polypeptide mapping. Can. J. Plant Pathol. 8(3): 233–240. Vilgalys, R., and Hester, M. 1990. Rapid genetic identification and doi:10.1080/07060668609501794. mapping of enzymatically amplified ribosomal DNA from several Kirk, P.M., Cannon, P.F., David, J.C., and Stalpers, J.A. 2001. Cryptococcus species. J. Bacteriol. 172(8): 4238–4246. PMID: Ainsworth and Bisby’s dictionary of the fungi. 9th ed. CAB 2376561.

For personal use only. International, Wallingford, UK. Wei, S.X., and Wang, Y.C. 1986. Taxonomic studies of Puccinia on Miller, M.A., Holder, M.T., Vos, R., Midford, P.E., Liebowitz, T., Compositae in China. Acta Mycol. Sin. Suppl. 1, 185–226. Botany Downloaded from www.nrcresearchpress.com by USDA on 07/05/12

Published by NRC Research Press