Brief Summary of Minutes of Annual Meeting

Minutes Summary for the WERA-89 meeting San Diego Downtown Embassy Suites San Diego, CA, March 20-21, 2014 Chair: Alex Karasev, Univ. of Idaho, Moscow, ID Vice-chair: Mark Pavek, Washington State Univ., Pullman, WA (absent) Secretary: Carrie Wohleb, Washington State Univ., Ephrata, WA Minute taker: Raina Spence, Washington State Potato Comm., Ephrata WA State Certification Reports: Colorado – Kent presents. At the post-harvest test, they tested 516 lots in 2013 representing about 7,211 acres. About 3,000 acres were rejected in summer sessions. The rejections were largely due to mosaic. Kent notes their certified seed area is mixed with production areas as well, with Russet Norkotah being a prominent variety. Their tolerance is 1.5% of PVY in the summer, across the generations with some incremental levels for G1 and G2. Kent said the number of rejections in the summer and winter tests have been about equal. PVY NO/Wilga Recombinant (O serotype) and NTN were both found. 12 growers were found to have NTN, in 35 total lots. Kent noted that for export to Mexico (table potatoes), the PVY N testing is very important. With grade defect and generational rejection issues, the growers are very concerned with NTN positives and restrictions on anything positive at post-harvest test for plant back. Colorado has established a zero tolerance on NTN. This went into effect in 2012 with their seed act. Mexico is testing everything incoming. California – Pablo is expecting 840 acres this year to be submitted for testing. Last year they were a little over 700. Some of this seed is exported to Latin America and the Caribbean. The seed into CA comes mainly from Canada. They also get some seed from ID, OR, MT, NY and NE. They have inspections through November. Last year they rejected 15 acres, because they were above 2 percent disease. In the second inspection, his group saw suspicious symptoms, took 100 to 200 leaves per acre and sent it to the lab in ID. They found that Red Lasoda NY10 at 16% infection. Pablo thinks in this case that the weeds in the ditches of these fields were causing the problem, harboring aphids Idaho – Alan Westra. Alan presented winter test results. Going in to the winter test, they had 953 lots. 95% of acreage was represented in the winter test. Alan noted that some of the smaller lots are sprout tested. They picked leaves in Hawaii. Leaf roll was not found in the 2013 winter test. The winter PVY test results showed 51.3 lots were totally clean, which represents 37.7 percent of the acreage. 25.4 percent of the lots (40% of the acreage) had PVY at or above 2%. Phil Hamm said as a best practice, growers should know what percent of PVY is in their seed-lots before they buy them. Stewart said they have been advising growers to ask for this information. Jeff noted that some growers have to buy their seed before the results are back, or else there is not enough seed left. Group dialogue about tolerances continued. Alan noted that they have 100% leaf testing by ELISA of everything in the winter test. In 2013, 55% of the lots tested clean. Montana – Nina Zidack: MT acreage has been steady between 9,500 and 10,500 acres. Nina said this year there was a major hail storm in a seed production areas, defoliating some of the areas in August. At least 2,000 acres were severely hailed in the storm. Some of these fields were left to grow an extra few weeks. MT had uniform emergence in Hawaii grow-outs, and was planted under good conditions. In 2012 they had more virus than 2013. This year they saw later season infection, which may have been due to leaving the plants in the field longer. Nina said most of the Umatilla Russet looked very good this year. Russet Burbank also looked very good this year. The Alturas and Rangers are more problematic, because visually the virus infections are harder to see. Nebraska – Steve: They had 6,000 acres last year. 4,000 of that was entered into winter test. One lot failed. They had a few lots with herbicide damage. Anything that has virus in plot is put in a not-recommended category. The Nebraska winter grow-out is done in Florida. They test all the latent varieties, and anything new that comes in out of state. The rejected lot this year was a Red Lasoda. Nebraska, New York, Maine, and ND are all still doing winter tests in Florida. Oregon – Jeff: Their average acreage is around 3,000 in Oregon. PVY varied this year, depending on where they were. Some areas were good, and remained good. Some seed is near commercial acreage, and had high levels after planting mini-tubers based on management challenges. Oregon is a pass through state. Most of their acres are G1 and G2 seed from other states. They have challenges with Norkotah and Rangers. This year some Rangers had no symptoms in summer, and in the winter grow-out they had symptoms. Oregon has winter grow-outs in the greenhouse. Material was sent to Alex for testing. They had PVY NTN, O, and N. They plant at 4 foot by 4 foot grids in the greenhouse. It’s a visual program primarily. New York – Stewart. Steward will show a slide of the some of the data later in the meeting. Washington – PVY was a challenge last year, but the seed growers have been working hard to clean up material for this year. General Research/Extension Discussion: Stewart Gray, USDA-ARS Cornell, Ithaca, NY: 2013 post-harvest testing with respect to PVY strains and distribution Stewart presented a slide summarizing the surveillance of the US seed potato crop for PVY recombinants. The percentage of NTN in the overall lots has been variable. N-Wi is becoming much more dominant in the lots. Stewart says no state is free of NTN. Steward reiterated that NTN causes strong strains. Stewart reiterated that PVY management is a business issue and a scientific problem. Stewart encouraged industry to adopt new varieties, and the growers need to ask for information on varieties they are growing. NTN is more fit than other viruses, and in Europe they shifted from PVY O to NTN in about 20 years. We are still a low percentage of NTN infection and it can be controlled. Nina said that in her lab tests, ELISA has adequate functionality to pick up PVY in leaf tissue and is comparable to PCR in that regard. Nina Zidack, Montana State U, MT: Likelihood of making recertification tolerances given specific PVY levels detected during summer testing Nina reviewed historical post-harvest data and formulated percentages and summaries. All of Montana’s G1 and G2 summer lots are tested. In G1, they take ten leaves per each family unit. In G2, they take 200 leaves per sample. In G3, they don’t have a requirement that it be tested in summer, but many growers do it voluntarily. Nina clarified that they are composite samples for ELISA. Discussion was entertained about where the analysis formula came from. Phil Nolte says he has the paper that discusses the sampling formula. Alan Westra said Cheri in their lab validated the formula. Conclusions state there are differences between years and generations for PVY levels in G1 and G2. Nina offers this information to growers in her state. Alex Karasev, U of Idaho, ID: Changes in PVY strains circulating in the Pacific Northwest, 2011-2013 Alex worked with Phil Hamm and Jim Crosslin on this work. The project monitored PVY strains circulating in seed potato in the main potato producing states. Seed lots trails in Othello and Hermiston provide source material for the survey. The typing was done by ELISA and RT-PCR. Alex describes the challenges of seed procurement for Washington State growers, and emphasizes the significance of the seed lot trials. The percentage of PVY O has dropped significantly from 2011 to 2013 in the field trial. Conclusions show that NTN can be found in most states. N-Wi seems to be displacing PVY O. Alex believes breeding for resistance to PVY N-Wi is very important, since PVY O is reducing in prevalence. Jonathan Whitworth, USDA-ARS, Aberdeen, ID: Progress in developing PVY resistant cultivars in the US potato breeding programs Jonathan presented data and analysis for breeding lines. Data was shown for aphid versus mechanical transmission from the 2011 Idaho grow-outs. Some of the data shown suggests possible resistance, but the variety may or may not have markers. The data is based upon ELISA tests. Jonathan noted two promising varieties that are showing multi-strain PVY resistance, and one is in the processed trials. Phil Hamm asked if anything has shown powdery scab resistance. Jonathan said not that he is aware of. Kelsie Evans, U of Idaho, ID: Recombinants of PVY – origins and evolution Alex introduced Kelsie Evans. Kelsie works in bioinformatics and computational biology as a graduate student in the University of Idaho. She shows a gene graph of common recombinants. Her work involved whole-genome sequencing of a large number of field isolates of PVY from US potatoes. She notes challenges using Genbank when some of the submitted sequences may have been misidentified. She concludes that there are at least 30 unique PVY recombinant structures. Some breaking points are more conserved than others. Stewart Gray and Alex Karasev: Discussion – Directions for research on PVY and other tuber necrotic viruses Stewart opened a discussion on the SCRI project on PVY that has been ongoing for the last few years. Both the virus and zebra chip grants are up for renewal. The industry has decided to support both the ZC and virus proposal. So, we can move forward to develop a new virus proposal. Stewart noted they are looking for additional suggestions and participants Johnathan Whitworth gave a presentation on seed production and work in Kenya. The majority is farmer saved seed that they use. Kenya has many of the same viruses that are present in the U.S. They have bacterial wilt and late blight. Jonathan assisted with developing a lab manual for the local scientist and lab. Discussion was entertained about challenges with PVS and export shipments. Nina commented on the issue of tuber core vs sprout vs leaf data that Susie had collected. She presented a graph comparison of the different tissue analyzing data. She wanted to determine if tuber core samples were equivalent to a sprout test. The table showed results for ELISA, PCR and RT-PCR for PVY and PVA. The process of testing the tuber cores, sprouts, and leaves in this manner is labor intensive and will not be taking the place of a grow-out anytime soon. However, Nina notes that this is a good tool and excellent means of detection for growers to plan ahead before the results of postharvest tests. Alex gave a presentation entitled “An update on PVY strains in Idaho seed supply.” The data presented covered the years of 2009 to 2014. Alex’s lab uses RT-PCR and ELISA in these tests. The testing focuses on the strains known to cause tuber necrosis. All NE-11 samples came from a single producer. NE-11 is associated with tuber damage. Alex noted that tuber symptoms and tuber symptom development in storage is a challenging and complex problem. The 2014 data was from the Hawaii grow-out, and the stand was much better than prior years. PVY O types found are reducing in number. Discussion was entertained about how N-Wi is the dominant strain now in most states. Carrie Wohleb gave a presentation on her psyllid sampling network in the Columbia Basin. She noted that we have a look-alike problem purple top in potato that is difficult to differentiate from Zebra Chip. Carrie discussed the overwintering of psyllids on the PNW on bittersweet nightshade, which is common in the region. Challenges of identifying tuber and foliar symptoms and testing for the pathogen were discussed. Phil Hamm gave a brief presentation on zebra chip based upon his observations and work in the Pacific Northwest. Phil discussed the confusion regarding spray programs to control psyllids and their efforts to improve management. Impact statements for 2014: Alex calls for impact statement submissions Nina said all of the strain work would be relevant, and quantification of strain compositions important. Nina’s work comparing winter grow-out data with summer inspection data is important. Jonathan’s work on the varieties, showing the response of varieties to the different strains of PVY. Progress in developing PVY resistance. Evaluating new potential cultivars. Document success in virus management such as ZC. This group provides expertise to industry needs. Work Alex and Kelsie did on strain diversity and genetics – changing population. PVY Wi is the most diverse recombinant that occurs. Discussion was held about future locations. Alex moves to adjourn 11:34 Andy seconded the move to adjourn.

Potato Virus Y (PVY) Potato breeding programs are focusing more on developing Potato virus Y (PVY) resistance cultivars. In a nationwide trial, 13 advanced breeding lines/new cultivars were submitted from six U.S. potato breeding programs and screened for PVY resistance. The trials were conducted in Idaho, Wisconsin, and New York. Results showed that 6 lines had high levels of resistance (immunity in 4) to PVY. These 4 lines are resistant to 3 strains of PVY (PVY-O, PVY-NTN, PVY-N:O). One of these PVY immune clones is being used as a parent in Michigan and Maine breeding programs to develop new varieties. A second set of entries from U.S. breeding programs is currently being tested under the same protocol as the first set. An analysis of five years of PVY testing data revealed crucial relationships between the amount of virus detected at summer and postharvest testing (PHT). In Montana, historical data show that for generation 1 seed, if summer tests showed 0 PVY during summer testing there was an 87% chance that the seed lot would still be 0 PVY at PHT. Even with 0 PVY in the summer, 6% of the seed lots had >0-0.5% PVY and 7% actually exceeded the plant back standard of 0.5%. In this situation, the PVY inoculum is most likely coming in from potatoes planted around the seed plot. At low but detectable summer levels of PVY (>0-0.1%), there are distinct difference between the generations which are most likely due to the intensity of testing, inspection and roguing in the G1. At this level of summer infection, growers still have a 72% chance of getting G1’s back to 0. For G2 and G3, there is only a 45% and 34% chance of getting 0’s at PHT when you start out at that same >0-0.1 level. This is distinctly different than the situation where you start out with infection levels of >0.1-0.5%. At this higher level of infection, there is a 70% chance that the G1 seed lot will not make a plant back threshold of 0.5% which is equivalent to G2 at 70% and very close to G3 at 67%. Based on phylogenetic analysis of 119 newly sequenced whole genomes of PVY and 166 genomes deposited in the GenBank database, substantial heterogeneity was revealed in the PVY-O and PVY-N strains, producing several distinct lineages. Analysis of the recombinant strains of PVY suggested that the PVY-N-Wi strain was the most heterogeneous, and likely arose multiple times from different parental PVY-O and PVY-N genomes. Three year typing of PVY isolates collected in the Othello and Hermiston potato seed lot trials indicated a drastic shift in PVY strain composition. Between 2011 to 2013, the proportion of the PVY-O strain circulating in seed potato dropped from 60% to 20%, with the concomitant rise in the recombinant strains from 35% to 80%. Deep sequencing technology has enabled the analysis of small RNA profiles of virus-infected plants and could provide insights into virus-host interactions. Potato virus Y (PVY) is an economically important viral pathogen of potato worldwide. While much is known about this virus, little or no information is available regarding host response to PVY infection. In this study, Dr. Hanu Pappu and team investigated the nature and relative levels of virus-derived small interfering RNAs (vsiRNAs) in potato cv. Russet Burbank infected with three biologically distinct and economically important strains of PVY, the ordinary strain (PVY-O), tobacco veinal-necrotic strain (PVY-N) and tuber necrotic strain (PVY-NTN). The analysis showed vsiRNAs of 20-24 nt in PVY-infected plants. Considerable differences were present in the distribution of vsiRNAs as well as total small RNAs. The 21 nt class was the most prevalent in PVY-infected plants irrespective of the virus strain, whereas in healthy potato plants, the 24 nt class was the most dominant. vsiRNAs were derived from every position in the PVY genome, though certain hotspots were identified for each of the PVY strains. Among the three strains used, the population of vsiRNAs of different size classes was relatively different with PVY-NTN accumulating the highest level of vsiRNAs, whereas PVY-N infected plants had the least population of vsiRNAs. Unique vsiRNAs mapping to PVY genome in PVY-infected plants amounted to 3.13, 1.93 and 1.70% for NTN, N and O, respectively. There was a bias in the generation of vsiRNAs from the plus strand of the genome in comparison to the negative strand. The highest number of total vsiRNAs was from the cytoplasmic inclusion protein gene (CI) in PVY-O and PVY-NTN strains, whereas from PVY-N, the NIb gene produced maximum total vsiRNAs. In addition to previously reported conserved microRNAs, 258 non-conserved miRNAs as well as 6 novel miRNAs were identified in PVY-infected potato plants. The growers of Washington State were well served in the 2013 growing season by key members of the WERA-89 team. Several prominent growers and industry leaders unwittingly purchased seed with a high percentage of PVY infection, and planted over 200 acres of the infected lot under long-term storage processing contracts. To meet the terms of processing contracts, internal defects or damages exceeding a set threshold will lead to a complete rejection of the harvested field. Accurate and prompt typing of the PVY variants present in the lot was provided by Dr. Alex Karasev at the University of Idaho, and an impartial trial of all regional seed lots was conducted by Dr. Mark Pavek at Washington State University. Both of these gentleman provided information imperative for these two growers to make accurate management decisions. Information Distribution: The creation and updating of the Potatovirus.com web site has been very useful as a reference. Researchers can send vital information to growers quickly and efficiently regarding virus testing of specific varieties. http://www.potatovirus.com/ Potato Virus S (PVS): Potato virus S (PVS) is becoming increasingly important both in incidence and impact in the US and other parts of the world. Five Potato virus S (PVS) isolates from the USA and three isolates from Chile were characterized by Dr. Hanu Pappu based on biological and molecular properties to delineate these PVS isolates into either Ordinary (PVS-O) or Andean (PVS-A) strains. Five isolates, 41956, Cosimar, Galaxy, ND2492-2R, and Q1 were considered Ordinary strains as they incited local lesions on the inoculated leaves of Chenopodium quinoa, whereas the remaining three (FL206-1D, Q3, and Q5) failed to induce symptoms. Considerable variability on symptom expression and severity was observed among these isolates when tested on additional indicator plants and potato cv. Defender. Additionally, all eight isolates were characterized by determining the nucleotide sequences of the coat protein (CP) gene. Based on the biological and genetic properties, 41956, Cosimar, Galaxy, ND2492-2R, and Q1isolates were identified as PVSO. PVS-FL206-1D and the two Chilean isolates (PVS-Q3 and PVS-Q5) could not be identified based on phenotype alone, however, based on sequence comparisons, PVS-FL206-1D was identified as PVSO, while Q3 and Q5 clustered with known PVSA strains. C. quinoa may not be a reliable indicator for distinguishing PVS strains. Sequences of the CP gene should be used as an additional criterion for delineating PVS strains. A global genetic analysis of known PVS sequences from GenBank was carried out to estimate the nucleotide substitution, population selection, and genetic recombination to assess genetic diversity and evolution of PVS. The higher value of nucleotide diversity (?) of the CP gene compared to that of the 11K gene suggested greater variation in the CP gene. Between PVS-A and PVS-O strains, a higher ? value was found for PVS-A. Statistical tests of the neutrality hypothesis indicated a negative selection pressure on both CP and 11K proteins of PVSO, where a balancing selection pressure was found on PVS-A. Potato virus S (PVS) is widely prevalent in various potato-producing regions of the world. Late blight resistant (LBR) cv. Defender and a breeding line LBR4106 (A95056-61) were found to be highly susceptible to PVS infection by Dr. Hanu Pappu and his team. Two PVS isolates, PVS-WaDef and PVS-Id4106, isolated from Defender and LBR4106, respectively, were used to characterize the response of LBR cultivars and breeding material. PVS produced distinct and severe symptoms including severe foliar mosaic, necrotic lesions, leaf wilting, and early death of plants. Both isolates were identified as PVSO strain based on the Chenopodium spp. bioassay and phylogenetic analysis of the amino acid sequences of coat protein. Nicotiana occidentalis-37B was found to be a good indicator plant for identifying severe phenotypes of PVS. The response of selected potato cultivars, germplasm, breeding lines, and the pedigree of Defender were evaluated using these two isolates. PVS-Id4106 incited more severe symptoms than PVS-WaDef on selected cultivars and breeding lines. Late blight, caused by Phytophthora infestans, is a destructive disease of potato. Defender is the only cultivar in the U.S. with foliar and tuber resistance to this disease. However, this cultivar exhibits susceptibility to infection by Potato virus S (PVS) and severe symptoms appeared on leaves after infection with PVS. PVS is widespread in potato fields in the U.S. To investigate potential interactions between P. infestans and PVS, Dr. Hanu Pappu and team detached leaves of Defender and Ranger Russet (susceptible to late blight), that were either PVS-infected or non-infected, were inoculated with P. infestans BF-05. The amount of sporulation and the extent of lesion expansion on inoculated leaves were measured to estimate late blight severity. When inoculated with P. infestans only, as expected, Defender exhibited discrete, relatively small, dark purple to black hypersensitive reaction-like spots and on an average had twenty times fewer sporangia compared to Ranger Russet. However, in Defender plants infected with PVS, lesion expansion and sporulation increased significantly compared to PVS-free Defender. Information Distribution: Researchers, growers, and industry can interact more quickly and efficiently. Improved distribution of information allows growers to act quickly on events that could be the difference between profit and loss. Zebra Chip and Virus Detection: Candidatus Liberibacter solanacearum (Lso), the organism that causes ZC, can be difficult to detect from potato tubers, plant tissue, and the vector, the potato psyllids (Bactericera cockerelli Sulc) due to PCR inhibitors such as competitor DNA (host genomic DNA) or other inhibitors present in the sample. The OSU Hermiston Plant Pathology Lab has been developing a high-fidelity PCR protocol to increase detection of Lso from tubers, plants and potato psyllids. In initial studies, the high-fidelity PCR protocol improved detection of ZC from tubers, leaves, and psyllids by 30-60% over conventional PCR. Viruses that cause necrotic symptoms in potato tubers can be difficult to distinguish based on symptoms and frequently require multiple molecular tests to identify the causal agent. We have developed a reverse transcription high-fidelity multiplex PCR protocol that utilizes previously validated primers to detect six different potato viruses simultaneously: Alfalfa mosaic virus (AMV), Tobacco rattle virus (TRV), Tomato spotted wilt virus (TSWV), Potato mop top virus (PMTV), Potato virus Y (PVY), and Potato leafroll virus (PLRV).

Brown, C.R., K.G. Haynes, M. Moore, M.J. Pavek, D.C. Hane, S.L. Love, and R.G. Novy. 2013. Stability and Broad-sense Heritability of Mineral Content in Potato: Potassium and Phosphorous 90:516-523. Chikh Ali, M., Karasev, A.V., Furutani, N., Taniguchi, M., Kano, Y., Sato, M., Natsuaki, T., and Maoka, T. (2013) Occurrence of Potato virus Y strain PVYNTN in foundation seed potatoes in Japan, and screening for symptoms in Japanese potato cultivars. Plant Pathology 62: 1157-1165. Gray, S., Whitworth, J., Xu, H., Singh, R., and Karasev, A. (2013) ST 01: The current state (2012) of Potato virus Y (PVY) affecting potato grown in North America. NAPPO Science and Technology Documents, August 2013. NAPPO: North American Plant Protection Organization. Karasev, A.V. and Gray, S.M. (2013) Continuous and emerging challenges of Potato virus Y in potato. Annual Review of Phytopathology 51: 571-586. Lin, Y-H, J. Abad, C. J. Maroon-Lango, K.L. Perry, and H.R. Pappu. 2014. Molecular characterization of domestic and exotic Potato virus S isolates and a global analysis of genomic sequences. Archives of Virology. In press. Lin, Y-H., D. A. Johnson, and H.R. Pappu. 2014. Effect of Potato virus S infection on late blight (Phytophtora infestans) in potato (Solanum tuberosum). Amer. J. of Potato Research. In press. Naveed, K., N. Mitter, A. Dhingra, and H.R. Pappu. 2014. Comparative Analysis of virus-specific small RNA profiles of three biologically distinct strains of Potato virus Y (PVY) in PVY-infected potato (Solanum tuberosum) cv. Russet Burbank. Virus Research. In press. Novy R.G., J.L. Whitworth, J.C. Stark, B.A. Charlton, S. Yilma, N.R. Knowles, M.J. Pavek, R.R. Spear, T.L. Brandt, N. Olsen, M. Thornton, C.R. Brown, S.R. James, D.C. Hane. 2014. Teton Russet: an early-maturing, dual-purpose potato cultivar having high protein and vitamin C content, low asparagine, and resistances to common scab and Fusarium dry rot. Am J Potato Res. 10.1007/s12230-013-9362-8. Novy, R.G., J.L. Whitworth1, J.C. Stark, B.A. Charlton, S. Yilma, N.R. Knowles, M.J. Pavek, T.L. Brandt, N. Olsen, M. Thornton, C.R. Brown, H. Lozoya-Saldana, and M.I. Vales. 2013. Palisade Russet and Teton Russet: two new potato cultivars from the Northwest (Tri-State) Potato Variety Development Program. Abstract–The 96th Annual Meeting of the Potato Association of America. Am J Potato Res 90:143. Quintero-Ferrer, A., Robles-Hernandez, L., Gonzalez-Franco A.C., Kerlan, C., and Karasev, A.V. (2014) Molecular and biological characterization of a recombinant isolate of Potato virus Y from Mexico. Archives of Virology, published on-line January 9, 2014 (DOI 10.1007/s00705-013-1968-0). Whitworth J.L., R.G. Novy, J.C. Stark, S.L. Love, M.K. Thornton, B.Charlton, S. Yilma, N.R. Knowles, M.J. Pavek and X. Wang 2014. Huckleberry Gold: a high antioxidant purple-skin yellow-flesh specialty market cultivar with potato cyst nematode resistance (H1) and Potato virus X resistance (Nb and Rx1). (Accepted – In-Press) Swisher, K.D., Sengoda, V.G., Dixon, J., Munyaneza, J.E., Murphy, A.F., Rondon, S.I., Thompson, B., Karasev, A.V., Wenninger, E.J., Olsen, N., and Crosslin, J.M. (2014) Assessing potato psyllid haplotypes in potato crops in the Pacific Northwestern United States. American Journal of Potato Research, published on-line March 1, 2014 (DOI 10.1007/s12230-014-9378-8). Zommick, D.H, L.O. Knowles, M.J. Pavek and N.R. Knowles. 2014. In-Season heat stress compromises postharvest quality and low-temperature sweetening resistance in potato (Solanum tuberosum L.). Planta 10.1007/s00425-014-2048-8.