Xylella fastidiosa: interview with professor Alexander Purcell

For the italian translation, click here.

Today we are talking with Alexander Purcell, Professor Emeritus at the University of California-Berkeley for the Department of Environmental Science, Policy and Management. He is one of the most important scientists that study the bacterium known as Xylella fastidiosa. For forty years his research analyzed Xylella, insect vectors of plant pathogens and the global spread of Xylella. He also went here in Italy for observing the situation in Salento, where Xylella fastidiosa ssp. pauca is currently cited as the cause of the olive disease spread (CoDiRO).



Welcome to Italia Unita per la Scienza, dr. Purcell. As you know, in the italian region Salento, we are facing Xylella fastidiosa: there is a lot of confusion, everybody is worried for the agricultural impact (olives above all) and we really appreciate this interview where we can freely discuss about Xylella.

It was a pleasure to hear from you, and I would be delighted to talk to you about Xylella and the controversies that seem to be raging over the many aspects of how this bacterium is affecting olives in Italy. Its effects on olives and other plants is sure to expand throughout the Old World.

Let’s start with the recent paper about Xylella fastidiosa ssp. pauca in Brazil [1] [2]: the strong correlation (as quoted in the paper) between symptomatic olive tress and X. f. subsp. pauca in Argentina, Italy and now Brazil can be considered a definitve proof of the cause-effect relation between Xylella and the disease of olive trees?

Since coffee and orange strains of Xylella fastidiosa ssp. pauca are widespread and genetically variable in South America, it is not surprising that this ssp. has been found there in olive.

How large can be considered now the spreading of Xylella? How can we face this new outbreak? Could the plans proposed for Italy be suitable also for Brazil?

The spread of Xylella fastidiosa in Europe is a serious matter even if you disregard the damage to olive trees. Xylella fastidiosa is very variable genetically, as many studies have shown. Each subspecies and even isolates of the same subspecies can vary greatly as to what plants develop disease symptoms. Most plants are symptomless hosts, but the correlation of detecting Xylella and the occurrence of symptoms in olive quick decline (CoDiRo) is evidence that Xylella fastidiosa causes the disease. In addition, the evidence based on detection in other plants suggests that this “olive strain” of Xylella may attack other species of commercial or ecological importance. What appears to be the same strain of Xylella in olive is found in oleander, and the ssp. sandyi has been proven to cause oleander leaf scorch, with similar symptoms as for diseased oleanders in Salento. For what scientists traditionally require for definitive proof that Xylella causes CoDiRo disease we must await the completion of Koch’s postulates, the last steps of which are to show that experimental infection of olive plants cause CoDiRo symptoms and the same bacterium can be recovered from such experimental plants. I understand that these experiments have been started but must await definitive results. I have no role in these tests, so you need to ask the scientists who have conducted them to find exactly what has been done. Tree age and stress may affect the incubation period in the plant for symptoms to occur. This may take more than a year to complete. So what to do while the olive disease is expanding exponentially? It is well established that certain types of insects are vectors of Xylella fastidiosa, so we have to keep in mind that these vectors must be included in any control plans.

How do we deal this new disease? Two things have been repeatedly mentioned, the removal of diseased trees and the control of what seems to be the main insect vector, a spittlebug, Philaenus spumarius. If the bacterium moves mainly from olive tree to olive tree, tree removal may slow the spread of the disease. The spatial patterns and rate of disease spread of CoDiRo is consistent with exponential (logistic) spread. Diseased plant removal seems to be effective in Brazil for a xylella-caused disease (CVC) of orange trees. The spread of CoDiRo in Salento appears to be similar to the spread of CVC in Brazil but not similar to Pierce’s disease of grape in northern California. There are complicated reasons – mostly climatic – why this does not work for Pierce’s disease in northern California. For an explantion of this phenomenon (lack of effective vine-to-vine spread of Pierce’s disease) see 2013 Annual Review of Phytopathology 51:339-356.

For heavily diseased areas, eradication of a bacterium that has a rapid rate of spread and which infects many species of plants, tree removal, even in combination with vector control, is unlikely to save the few remaining symptomless trees. For areas where Xylella fastidiosa has as yet only affected a relatively few trees, there may be a chance to effectively halt or slow down the spread of the bacterium and the diseases it causes, but this will only work if can be done thoroughly and quickly. The longer the delays, the less chance the method can work. If only some affected orchards are treated and others ignored, everyone will lose. As I understand it, this diseased tree removal in conjunction with vector control is the method that has been proposed by scientists and disease control authorities as well as the relevant agencies of the European Union but this approach has been opposed by other groups that claim to act in the public interest. Is the proposed tree removal and vector control certain to be effective? We will not know unless it can be tested on a realistic scale. If this was a human or animal disease, there would be no serious opposition to the proposed methods and little concern for the environment. I think that what oppostion groups don’t consider is the great harm that their opposition poses to agriculture and the environment. The longer it takes to implement the proposed methods, the less likely they are to be effective. It may seem ironic, but the longer we wait to remove diseased trees and apply insecticides and other vector control measures, the faster that many more trees will be lost and the greater will be the need for vector control well into the future. It is true that there is a chance that the control methods may not work, but no serious alternatives have been offered except by those that deny any role of Xylella fastidiosa in killing olive trees and other plants – all without any evidence that their counter claims are true.

Recent publications tested the possibility of deploying bacteriophages against Xylella [3] [4] [5]. Lab tests showed promising results with the californian strain which attacks grapes [6], encouraging a possible usage of this tactic as a suitable weapon against Xylella. Anyway, tests were done only in greenhouse environment. How much time is required for definitive conclusions about this path? How difficult will be translating these results from greenhouse to open field? I imagine that in the latter case we would still face the problem of vectors constantly transferring Xylella from any host to another [7], but we could also use an integrated strategy.

The recent report of Gonzales et al. about therapy using bacteriophages is highly encouraging, but needs to be evaluated in a challenging field environment. Be careful to note the cautions the authors of this paper recommend in applying bacteriphage for control. One of the dangers of using another strain of Xylella to protect against the CoDiRo strain is that the introduced strain may cause damage in other plants. Another problem is the long times required to find and test suitable strains. Surely this would not be done in Europe, but doing these tests in Brazil, Argentina, or Costa Rica does not have to introduce strains of Xyella that are exotic to these countries.

What do you think about the considerations that italian scientists used when relating the olive disease (CoDiRO) to Xylella? Are their methods and their approach enough for an evidence?

The sampling and experimental methods that I am aware of are what I would have done. The scientists at Bari and the EU agency most involved (European Food Safety Agency) asked for suggestions and collaborations with scientists experienced in working with Xylella.

Italian magistrates blocked the Silletti plan and started investigations against the italian scientists in Bari. One of the reasons for their decisions is that they report that you said, during an EFSA workshop, this quote “don’t do our same mistake: against Xylella is completely useless to remove trees”. It is said here at 1:30 in italian: https://www.youtube.com/watch?v=pYXvgkmxDsA
Is this true?

According to translations to English from the video, they repeated my making comments claimed by Rosa D’Amato that I was opposed to tree removal. That is untrue.  See my comments below.

It is also reported by the europarlamentarian Rosa D’Amato, still in italian, here [8].
Is this true? Can you confirm or deny this statement? Any further consideration about the decisions of the magistrates (blocking the containment plan and the confiscation of trees)?

I can read Italian with help from Google. In this case I am confident enough to say that Ms. D’Amato’s quote of both myself and Prof. Joao Lopes is COMPLETELY NOT TRUE. I confirmed this by email with Prof. Lopes. After the workshop in Brussels I discussed this point for about an hour with Ms. D’Amato, as translated by her administrative assistant. I never indicated that I opposed tree removal, in fact I strongly endorsed it and also pointed out that it alone would not work – vector control had to be applied at the same time (actually vector control should be done just before tree removal).

More than my opinion or recollection, you can see for yourself just what I said at the Xylella workshop. I did not address tree removal in my talk. Prof. Lopes did mention tree removal. He pointed out that this was a major method used for control of the Xylella disease of orange trees in Brazil. A video of my talk and that of Prof. Lopes is on the Internet at


The talks begin in the center video (second of three videos) at 2 minutes (Purcell) and 37 minutes (Lopes). I did not mention diseased plant removal as a control method. As you can see, Lopes stated that tree removal was important for control of the CVC disease in young trees and pruning in cases where only the early CVC symptoms were present in older trees. Pruning was not effective for the Xylella-caused Pierce’s disease of graevines and may not work for the CoDiRo disease of olive.

Besides Italian magistrates currently reports claims from other researchers that in Italy we have 9 different strains of Xylella, and that they were introduced possibly decades ago, enough time for accumulating genetic mutations. Anyway, their work, upon which the magistrates are setting up their investigation, aren’t public. Only the magistrates have access to them. What do you think about?

Where is this report? Can you send me a copy?

I only know of 5 subspecies, with the pear leaf scorch bacterium possibly representing another Xylella species or subspecies. I cannot respond without seeing more details, preferably published scientifically.


That is the point: italian magistrates are the only ones with access to these reports. The scientific community can’t see and evaluate the alleged reports that Xylella is harmless or even endemic in Italy. It is also a strange situation because some of the experts appointed by the magistrates reported to news agency ANSA that the strain of Xylella in Salento belongs only to the ssp. pauca and is a particular strain. 
Here are two examples in italian (our readers were already informed of these interviews), by Ansa [9] and Corriere del Mezzogiorno [10]. Well, this position agrees with the evaluation of the researchers in Bari and the final report of the SELGE scientific network [11]. But the magistrates claim that their experts found very different conclusions! How should we react in your opinion to this confusion and lack of data? The perspective is certainly different between scientists, journalists, institutions and citizens, but the situation here seems to be a bit chaotic.

Without more details (data), it is not possible to comment further in addition to the observation that they confirm the main conclusions that Bari scientists published.

The second link repeats the previous Internet report, but adds comments that don’t make sense to me.  

Another claim is this. Franco Trinca of the NOGM Committee proposed this theory: taxonomic studies show that the common ancestor of all the current Xylella strains could be tracked up to 15.000 years ago (Nunney et al 2012; Schuenzel et al. 2005). This time corresponds to the birth of agriculture, thus it’s likely that the first populations of Xylella originated during this period in the Mediterranean basin. He believes that all the subspecies currently attacking oranges in Brazil and grapes in USA could have originated from an European strain. The original strain might be a direct progenitor of the ssp. pauca because the latter shows an highly-differentiated genetic polymorphism, and the current strain in Salento might be endemical.

A molecular clock [12] for Xf is currently not adequate because homologous recombination affects estimates. Clock is now anchored on assumptions that have never been tested for this bacterium. Diversity of Xf still poorly understood for such a statement anyway. Well known problems with molecular clocks are especially a problem for bacteria that undergo frequent homologous recombination. The evidence to date is very strong that that Xylella fastidiosa is indigenous to the Americas. 

Attempts to grow European grapes in the southern USA repeatedly failed, probably because of Pierce’s disease and other endemic diseases and pests.

I am not aware of any evidence that supports the hypothesis that Xylella in any form occurred in Europe before the movement of plant materials from the Americas. This theory is complete speculation.

Thank you for your explanation. Anyway the scientific magazine Le Scienze (Italian edition of Scientific American) published an open letter to the chief prosecutor Cataldo Motta, asking for free access to the reports done by the experts he hired [13]. Many scientists are signing this message. Do you think signing this open letter could contribute to transparency and respect of the scientific method in the trial?

I have no experience with or knowledge of Italian law, so I can’t comment on issues of law. Of course, relevant data is necessary to draw any conclusions about the validity of claims involving scientific conclusions or activity.

Thank you professor Purcell. It was a pleasure and we hope that our readers will find some explanations, since many people here in Italy are referring to your work in order to make different claims within different scenarios.

Thank you too and to all your readers!



[1] http://italiaxlascienza.it/main/2016/01/xylella-fastidiosa-negli-ulivi-in-brasile/ (italian)

[2] http://www.fupress.net/index.php/pm/article/view/17259 (english)

[3] http://italiaxlascienza.it/main/2015/10/virus-batteriofagi-per-la-xylella/ (italian)

[4] http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0128902#pone-0128902-g002

[5] http://www.apsnet.org/meetings/Documents/2014_meeting_abstracts/aps2014abS39.htm

[6] http://italiaxlascienza.it/main/2015/03/xylella-fastidiosa-e-i-vigneti-californiani/

[7] http://italiaxlascienza.it/main/2015/04/xylella-fastidiosa-insetti-vettore-trattamenti-per-la-sputacchina-e-contagio-in-francia/

[8] http://damatorosa.eu/xylella-damato-m5s-usa-e-brasile-confermano-che-abbattere-ulivi-non-serve-ma-ue-fa-orecchie-da-mercante-sulla-pelle-della-puglia/


[10] http://corrieredelmezzogiorno.corriere.it/lecce/cronaca/15_dicembre_29/xylella-consulenti-procura-specie-puglia-diversa-che-altrove-020d09dc-ae3e-11e5-bec0-c5dc32c2b9b8.shtml?refresh_ce-cp

[11] http://www.cespevi.it/servfito/pdf/Report_pauca_SELGE.pdf

[12] Molecular clock is a model used in biology to calculate the time when two species diverged from their last common ancestor. It’s based on the fact that the genentics mutations occur at a determined frequency. So, evalutating the differences in the DNA between two organisms is it possible to determinate the time when the divergence has occurred. More info here (Nature) and here (Berkeley University) in english and here in italian (from Alessandro Tavecchio, one of the members of our staff, talking about the first paper on Nature).

[13] http://www.lescienze.it/news/2016/01/15/news/xylella_lettera_aperta_procuratore_lecce_motta-2930682/

We also thank for the support:

  • Beatrice Mautino, biotechnologist and scientific journalist, Italian Committee for the Investigation of Claims of the Pseudosciences (CICAP)
  • Marco Cattaneo, physicist and director of Le Scienze magazine
  • Lisa Signorile, biologist and science communicator, National Geographic Italia
  • Paola Biasi, journalist, Siderlandia.it
  • Anna Rita Longo, journalist, Science Writers in Italy and CICAP
  • The Ninth, journalist and author for Bufale un tanto al chilo


%d blogger hanno fatto clic su Mi Piace per questo: